4

COST ACTION 734 IMPACT OF CLIMATE CHANGE AND VARIABILITY ON EUROPEAN AGRICULTURE
www.cost734.eu cost734@disat.unifi.it
SURVEY OF AGROMETEOROLOGICAL PRACTICES AND APPLICATIONS IN EUROPE REGARDING CLIMATE CHANGE IMPACTS
Edited by: Pavol Nejedlik and Simone Orlandini
ESF provide the COST Office through an EC contract
COST is supported by the EU RTD Framework programme
ESSEM Earth System Science and Environmental Management
2008
ABOUT COST
COST is an intergovernmental framework for European Cooperation in Science and Technology, funded by its member countries through the EU Framework Programme. The objective of COST is to coordinate, integrate and synthesize results from ongoing national research within and between COST member countries to add value to research investment. COST Actions aim to deliver scientific syntheses and analyses of best available practice to aid problem identification, risk assessment, public utilities and policy development.
Acknowledgements: Francesca Natali and Valentina Di Stefano for the support in the editing of the book. Stefano Fallai for the photo in the cover and Maria Poli for the help in the editing of the logo.
INDEX
PREFACE Pavol Nejedlik and Simone Orlandini............................................................................1
1 CLIMATE CHANGES, IMPACTS ON AGRICULTURE, SUSTAINABILITY G. Maracchi, F. Rossi, L. Kaifez Bogataj 1.1 1.2 1.3 1.4 Expected impacts of climate change in Europe during this century ......................3 Increasing drought Risk with Global Warming in Europe.....................................6 Challenges, new approaches and research needs ...................................................7 References ............................................................................................................11 2. AGROCLIMATIC INDICES AND SIMULATION MODELS J. Eitzinger, S. Thaler, S. Orlandini, P. Nejedlik, V. Kazandjiev, V. Vucetic, T. H. Sivertsen, D.T. Mihailovic, B. Lalic, E. Tsiros, N. R. Dalezios A. Susnik, Christian K. C. Kersebaum, N. M. Holden, R. Matthews Abstract ...................................................................................................................15 2.1 Introduction..........................................................................................................15 2.2 Agroclimatic indices in use and research (Subgroup 1).......................................16 2.2.1 State of the Art...........................................................................................16 2.2.2 Present use of agrometeorological indices in Europe ................................20 2.2.3 Communicative value and limitations of the agrometeorological indices and simulation models ...............................................................................50 2.3 Application of process oriented models in Europe (Subgroup 2) .......................51 2.3.1 State of the Art...........................................................................................51 2.3.2 Present use of process oriented models in Europe.....................................53 2.4 Useful Outputs and main Limitations of Models for use in Climate Change Impact Research (Subgroup 3).............................................................................69 2.4.1 The problem of modelling natural phenomena ..........................................69 2.4.2 Present limitations on crop model applications Europe.............................72 2.5 Crop simulation models in combination with Remote Sensing and GIS (Subgroup 4).........................................................................................................75 2.5.1 State of the art ............................................................................................75 2.5.2 Spatial model applications in Europe.........................................................84 2.6 Conclusions ..........................................................................................................91 2.7 Acknowledgements ..............................................................................................92 2.8 References ............................................................................................................92
3. SUMMARIZING A QUESTIONNAIRE ON TRENDS OF AGROCLIMATIC INDICES AND SIMULATION MODEL OUTPUTS IN EUROPE V. Alexandrov, E. Mateescu, A. Mestre, M. Kepinska-Kasprzak, V. Di Stefano, N. Dalezios Abstract .................................................................................................................115 3.1 State of art ..........................................................................................................115 3.1.1 Observed climatic and agroclimatic trends..............................................115 3.1.2 Agroclimatic indices and crop models.....................................................117 3.1.3 Examples of previous case studies...........................................................121 3.2 Goal: a questionnaire..........................................................................................123 3.3 Summarizing the questionnaire..........................................................................124 3.3.1 Long-term meteorological and agrometeorological data .........................124 3.3.2 Numerical weather models, regional climate models, weather generators.................................................................................................131 3.3.3 Homogenization tests/procedures ............................................................139 3.3.4 Statistical methods for analyses of meteorological and simulation model output related time series ..............................................................144 3.3.5 Additional information listed within the questionnaire ...........................151 3.4 Concluding remarks ...........................................................................................152 3.5 Acknowledgments..............................................................................................153 3.6 References ..........................................................................................................153
4. SATELLITE SPECTRAL CLIMATIC AND BIOPHYSICAL DATA FOR WARNING PURPOSES FOR EUROPEAN AGRICULTURE L. Toulios, G. Stancalie, P. Struzik, M. Danson, J. Mika, D. Dunkel, E. Tsiros Abstract .................................................................................................................163 4.1 Introduction........................................................................................................164 4.1.1 The need of adaptation of agriculture to climate change .........................164 4.2 How the study on climate variability and change can benefit from space .........166 4.2.1 Making Sense of Satellite Data................................................................167 4.3 Satellite-based variables and models potential in monitoring of crop production ..........................................................................................................168 4.3.1 Climate and biophysical data records in responding to climate change impacts on agriculture .................................................................172 4.4 Satellite instruments for climate change management .......................................174 4.5 Status of satellite climatic and biophysical data for warning purpuses for agriculture in Europe..........................................................................................178
4.6 4.7 4.8 4.9
4.5.1 Questionnaire processed results...............................................................179 4.5.2 Type of data per country ..........................................................................182 4.5.3 Detailed analysis of the data per country.................................................187 Conclusions ........................................................................................................195 Acknowledgements ............................................................................................196 References ..........................................................................................................196 Annex 1 ..............................................................................................................200
5. SATELLITE REMOTE SENSING AS A TOOL FOR MONITORING CLIMATE AND ITS IMPACT ON THE ENVIRONMENT POSSIBILITIES AND LIMITATIONS P. Struzik, L. Toulios, G. Stancalie, M. Danson, J. Mika, C. Domenikiotis Abstract .................................................................................................................205 5.1 Introduction........................................................................................................205 5.2 Evolution of meteorological satellite system .....................................................207 5.3 Satellite climatology possibilities....................................................................213 5.3.1 Satellite observations of processes in atmosphere selected examples ..214 5.3.2 Satellite observations of processes at the Earth surface selected applications ..............................................................................................218 5.4 Difficulties in use of satellite data for climate observations ..............................223 5.5 Conclusions ........................................................................................................228 5.6 References ..........................................................................................................228 5.7 Appendix 1.........................................................................................................231
6. USE OF CLIMATE CHANGE SCENARIOS IN AGROMETEOROLOGICAL STUDIES: PAST EXPERIENCES AND FUTURE NEEDS P. Calanca, L. Kaifez Bogataj, T. Halenka, E. Cloppet, J. Mika Abstract ..............................................................................................................237 6.1 Introduction........................................................................................................237 6.2 Climate scenarios ...............................................................................................239 6.2.1 Types of scenarios used in the past in agroclimatological studies...........239 6.2.2 Tools used in the preparation of climate change scenarios for impact studies ......................................................................................................243 6.2.3 Dealing with uncertainties in climate change projections .......................244 6.2.4 Mapping of climate change scenarios to the small scale: downscaling ...246 6.2.5 Scenarios for extreme events ...................................................................248 6.3 Lessons learned and outlook ..............................................................................250 6.3.1 A summary of past experiences ...............................................................250 6.3.2 Completed and ongoing projects at the European level...........................251 6.3.3 Role of COST 734 ...................................................................................253
6.4 Conclusions ........................................................................................................255 6.5 References ..........................................................................................................255 6.6 Additional references .........................................................................................258 7. RISK ASSESSMENT AND FORESEEN IMPACTS ON AGRICULTURE J. E. Olesen, M. Trnka, C. Kersebaum, P. Peltonen-Sainio, A. O. Skejvg, F. Rossi, J. Kozyra, B. Seguin, F. Micale Abstract .................................................................................................................267 7.1 Introduction........................................................................................................268 7.2 Observed and projected climate in Europe ........................................................269 7.2.1 Observed climate change in Europe ........................................................269 7.2.2 Projections of climate change in Europe..................................................270 7.3 Current European cropping patterns ..................................................................273 7.4 Questionnaire .....................................................................................................278 7.5 Vulnerabilities and climate impacts on crop ......................................................280 7.5.1 Present climatic limitations and vulnerabilities .......................................282 7.5.2 Climate change impacts ...........................................................................284 7.6 Adaptation to climate variability and climate change ........................................292 7.6.1 Observed adaptation.................................................................................295 7.6.2 Future adaptation responses.....................................................................297 7.7 Improving awareness and adaptation to climate change ....................................300 7.7.1 National impact assessments, adaptation strategies and awareness.........300 7.7.2 Dissemination and warning systems ........................................................301 7.8 Implications and perspectives ............................................................................302 7.9 Conclusions ........................................................................................................304 7.10 Acknowledgements ............................................................................................304 7.11 References ..........................................................................................................304 7.12 Annex I...............................................................................................................311
ANNEX 1. LIST OF CONTRIBUTORS................................................................313 ANNEX 2. LIST OF QUESTIONNAIRES ............................................................317
PREFACE Pavol Nejedlik and Simone Orlandini
During 2006, COST Action 734 (CLIVAGRI-Impacts of Climate Change and Variability on European Agriculture) was launched thanks to the coordinated activity of 15 EU countries. The main objective of the Action is the evaluation of possible impacts from climate change and variability on agriculture and the assessment of critical thresholds for various European areas (COST 734 MoU. www.cost.esf.org). Secondary objectives are: the collection and review of existing agroclimatic indices and simulation models, to assess hazard impacts on various European agricultural areas relating hazards to climatic conditions; building climate scenarios for the next few decades; the definition of harmonised criteria to evaluate the impacts of climate change and variability on agriculture; the definition of warning systems guidelines. Four working groups, with the integration of Remote Sensing sub working group, were created to address these aims: WG1 Agroclimatic indices and simulation models WG2 Evaluation of the current trends of agroclimatic indices and simulation model outputs describing agricultural impacts and hazard levels WG3 Development and assessment of future regional and local scenarios of agroclimatic conditions WG4 Risk assessment and foreseen impacts on agriculture The activity of WGs has been structured like a matrix, presenting on the rows the methods of analysis and on the columns the phenomena and the hazards. Each intersection point describes the evaluation of past, present and future trends of climate and thus the impacts on agriculture. Based on these results, possible actions (specific recommendations, suggestions, warning systems) will be elaborated and proposed to the end-users, depending on their needs. At present 28 countries join the Action with the collaboration of Agricultural Meteorology Division - Word Meteorology Organization and Ispra- IPSCAGRIFISH UNIT - Joint Research Centre. Time schedule of activity includes three main phases: Planning, operational arrangements, establishment of WGs and inventory. Main scientific work to be conducted by each WG. WGs activities to be concluded with emphasis on disseminations, reports and final publications.
In this book the results of inventory phase are presented. For this aim, five questionnaires (titled: Agroclimatic Indices and Models; Trends in Agroclimatic Indices and Model Outputs; Satellite Data Records Survey; Climate Change Scenarios; Risk Assessment and Foreseen Impacts on Agriculture) were disseminated among COST 734 countries and about 20 answers for each questionnaire were collected. This book has to be considered as a general description of the activity performed in Europe in the field of climate change and variability impacts on agriculture. It has to be pointed out that the results of questionnaire analysis are not representative of the whole Europe, but they are limited by the quantity of answering COST countries and by the quality of the answers. However the wide range of presented information, including data, models, indices, methods, tools, can represent an useful support for the workers of agricultural sectors (farmers, technicians, decision makers, etc.) to better plan the analysis and the adaptation of European agriculture to climate change and variability impacts.
1. CLIMATE CHANGES, IMPACTS ON AGRICULTURE, SUSTAINABILITY Giampiero Maracchi, Federica Rossi, Lucka Kaifez Bogataj
1.1 Expected impacts of climate change in Europe during this century A wide ranging impacts of changes in current climate have been documented in Europe in the last decades. The observed changes are consistent with projections of impacts due to anthropogenic climate change. The warming trend and spatially variable changes in rainfall have already affected managed ecosystems (Easterling et al., 2007). For instance the European heat wave in 2003 had major impacts on agricultural systems and society by decreasing the quantity and quality of the harvests, particularly in Central and Southern Europe. The very high air temperature and solar radiation resulted in a notable increase in the crops' water consumption. This, together with the summer dry spell, resulted in an acute depletion of soil water and lowered crop yields. Over all of Europe, the main sectors hit by the extreme climate conditions were the green fodder supply, the arable sector, the livestock sector and forestry. Potato and wine production were also seriously affected. The fall in cereal production in EU reached more than 23 million tonnes as compared to 2002. More than 26.000 fires were recorded in Portugal, Spain, Italy, France, Finland, Austria, Denmark and Ireland. The estimate is that some 70.000 hectares of forest area (not including agricultural areas) were burned. Summer 2003 showed also the additional side effects, which were felt in the next year such as problems of soil erosion and flooding, effects on winter sowing, and the budding of trees (COPA COGECA, 2003). Over the last few years the EU has financed several large research projects on regional climate modelling and impact assessment. In particular, some projects (PRUDENCE, PESETA) have produced high-resolution maps representing the projected changes in climate variables, such as mean temperature and precipitation, and projected impacts, e.g. agricultural yields. These maps are very useful for policy-making and awareness-raising purposes (Fig. 1.1). They illustrate what can be expected in Europe by the end of the century, according to the IPCC scenario whereby no action is taken to reduce GHG emissions, so that the global mean temperature increases by about 3.4C.
Survey of agrometeorological practices and applications in Europe regarding climate change impacts
Figure 1.1: Simulated crop yield changes (%) by 2080s according to two different models (left) HadCM3, (right) ECHAM4 (European Commission, 2007)
These maps are all based on IPCC SRES scenario A2. Results are on impacts from the JRC-funded PESETA study (http://peseta.jrc.es). Maps with projections of future changes in temperature and precipitation are based on DMI/PRUDENCE data (http://prudence.dmi.dk), and processed by JRC within the PESETA study. Changes are projected for 2071-2100 relative to 19611990. Climate change in Europe is very likely to reduce ecosystem services such as soil fertility, water availability, climate regulation potential or biodiversity, among other. Therefore, existing pressures on the already burdened natural environment of Europe may increase (Table 1.1). Furthermore, impacts may be unevenly distributed between the North (some positive, some negative) and South (nearly all negative). For example: the extent of forests is expected to expand in the North and retreat in the South. Climate change will increase net primary productivity and total biomass in the North while reduced water availability is likely to decrease NPP and forest growth in Central Europe, and accelerate tree mortality in the South. On the other hand, the capacity of Europes social systems to cope with climate change is high and is expected to continue rising. Adaptive capacity will vary between countries because of their different socioeconomic levels.
1. Climate change, impacts on agriculture, sustainability
Table 1.1: Some of the main expected impacts of climate change in Europe during the 21st Century (adapted from Alcamo et al., 2007) Magnitude of impact: *, **, ***. Type of impact: P= positive; N= negative; * to *: change in character through time, na = not applicable. Sectors and systems Impact Floods Water Water availability resources Water stress Forest NPP Northward/inland shift of tree species Natural disturbances (e.g., Forest, grasslands fire, insects) and Change of stability of forest shrublands ecosystems Drying/ transformation of wetlands Wetlands and aquatic Disturbance of drained ecosystems peatlands Suitable cropping area Agricultural land area Summer crops (maize, sunflower) Agriculture Winter crops (winter wheat) and fisheries Irrigation needs Energy crops Livestock Marine fisheries Area North N** P** P** P*** P*** N* N** N** N*** P*** N** P*** P***
Atlantic N** P** P** P** P** N* N* N* na P** N**
Central N*** N* P** P** N* N* N* N** P* N** P* P*to* N N** P* N** na
Medit. N* N*** N*** P*to* N P*to* N N*** N** N** N* N** N** N*** N** N*** N** N** N*
East N*** N** N** P* N** N** N*** N*** N*** N* N** N** P* N* N* N** na
P** P** P*to* na N P*** P** P*to* N N* P** P*
Under a changing climate, drier conditions and rising temperatures in the Mediterranean region and parts of Eastern Europe may lead to lower yields. Bindi and Moriondo (2005) showed a general reduction in yield of agricultural crops in the Mediterranean region, under the IPCC SRES A2 and B2 scenarios by 2050 even when the fertilising effect of increased CO2 is taken into account (Table 1.2). Similar yield reductions have also been estimated for Eastern Europe, with increased variability in yield, especially in the steppe regions (Maracchi et al., 2004). Climate-related increases in crop yields are mainly expected in Northern Europe. For example wheat yield increase is projected to be +2 to +9% by year 2020, +8 to +25% by year 2050 and +10 to +30% by year 2080 (Olesen et al.,
2006; Audsley et al., 2006; Ewert et al., 2005). Another example is sugar beet yield increase of 14-20% until the 2050s in England and Wales (Richter and Semenov, 2005). Uncertainties in the projection of future precipitation complicate the estimates of future yield gains or losses. This is particularly true for Southern and SouthEastern Europe, where water will be a critical factor for agriculture in the future. In these areas, model results diverge to a great extent, depending on the scenarios in use and the model itself. For Central and Northern Europe, where water supply is less critical, projections are relatively robust.
Table 1.2: Changes of crop yields (%) for some Mediterranean regions by 2050 (modified from Bindi and Moriondo, 2005) A2 Without CO2 B2 A2 With CO2 B2
C4 summer N-W 0.2 5.8 4.2 N-E -4.4 -2.5 -0.6 Legumes N-W -24.9 -13.4 -14.4 N-E -18.6 -8.1 -7.2 C3 summer N-W -21.8 -10.4 12.4 N-E -15.6 -6.9 -5.4 Tubers N-W -10.4 -4.2 4.9 N-E -22.5 -6.8 -9.3 Cereals N-W -11.0 -3.5 -0.3 N-E -6.8 3.7 4.4 N-W = Portugal, Spain, France and Italy; N-E = Greece and Turkey;
8,8 0.2 4.9 1.0 2.9 1.0 7.5 4.4 4.7 12.5
1.2 Increasing drought Risk with Global Warming in Europe Increased drought risk associated with global warming and impacts on water resources are among the main concerns among agrometeorologists in Europe (Fig. 1.2). Several recent studies highlight the challenges that result from changes in water availability and water quality (EEA, 2004; IPCC, 2001&2007; Schrter et al., 2005; EEA, 2005). Under climate change conditions, it is expected that irrigation water demand will further increase, aggravating the competition with other sectors whose demand is also projected to increase. In addition, an expected lowering of the
groundwater table will make irrigation more expensive, which, in turn might have to be limited to cash crops. Extreme weather events such as heat waves will impact on peak irrigation requirements,. As the evaporative demand will increase due to higher temperatures, it is expected that capillary rise will increase the salinisation of soils, having a major impact on irrigation management. Overall European food and fiber production is not expected to be greatly altered by climate change. However, greater differences will arise between countries. To sum up, warmer in the North; drier in the South; intensification of rainfall; increased frequency of extremely hot days or seasons imply more benefits to the North; more disbenefits to the south of Europe. This may worsen current resource issues: e.g, more water shortage and heat stress in South; and more flooding in the Centre, North and mountains. Furthermore it may aggravate current environmental problems (eg desertification in South; soil leaching in North). Taken together, the above implies a South-to-North geographical shift of climate resources in Europe; increasing the difference in resource endowment between North and South of Europe. Might policies therefore be needed to support development in Southern Europe, if the challenge of climate change is greatest there? But in the global context, Europe faces less negative effects than most other parts of the world, implying: there may be an opportunity to increase Europes share of world food production; and, from the global viewpoint, it may be necessary to increase food production in Europe in order to maintain global food security.
1.3 Challenges, new approaches and research needs The trends in climate change and the likelihood of further changes occurring give urgency to determining potential impacts on farm activities and addressing adaptation measures. Adaptation action will be, however, efficient only if integration of climate change-related issues with other risk factors, such as climate variability and market risk, and with other policy domains, such as sustainable development will be accomplished. The societal progress of civilization and technology attained so far, generated from the XIV Century scientific revolution, has generated vulnerability. Finite resources are decreasing or are being less utilizable, and this process is still often underrated. Techno-industrial society is already fundamentally unsustainable (Rees, 2006), as the continuously
increasing use of fossil fuels for productive purposes and for commercial exchanges at the basis of our development model.
Figure 1.2: Ensemble mean soil moisture changes in Mediterrain between the periods 19611990 and 2070- 2099 in spring and summer under the IPCC SRES A2 and B2 scenarios (PRUDENCE, 2005)
The first, most visible and impacting signal of the human enterprise on the environment is represented by climate changes, strictly linked to the emissions of greenhouse gases. Fossil fuels consumption has changed the composition of the atmosphere and the natural cycle of Carbon. Together with CO2, nitrous oxides are emitted as a consequence of fertilization, and methane from rice paddocks, enteric fermentation of ruminant animals, underground extractions. Agriculture accounts for 52 and 84% of global anthropogenic methane and nitrous oxide emissions. Potentially secondary pollutants, which long-term effects are still unknown, are released into the atmosphere and may cause hazards to man, animals, eco and agro-systems. Moreover, the energy and the material continuously extracted are returned to the ecosphere in degraded form. Together with energy, land use is becoming a main problem: land is needed for food crops, for producing renewable biomass energy, for maintaining other ecosystem services, for urban-industrial uses, transport, material extraction, refuse deposition, but also for leisure, recreation, and nature conservation. Economic globalization, great flows of resources and commodities, open markets and liberalized trades are accelerating the rates of the human impacts, and agribusiness globalization is threatening the economic viability of traditional, local, peculiarities. Climate change issues contribute, as well, to pose additional challenges. Despite of the advantages brought to developed Countries from this knowledge-based, technological model, a drastic revision of our concepts of economic model is impellent. Although most adaptations are focused on reducing risk, there is a need to address local capacity to adapt, as well as the societal processes generating vulnerability. A systematic effort to organize future arrangements and decision should start from now, to allow progressive changes of the models so far adopted avoiding abrupt, brusque transformations. The reduction of pressures on land and resources, obtained through restorations of natural processes and local attitudes and tradition can be a very powerful mean. Multifunctional agriculture may produce economic and highvalue commodities, allowing to contain fast and large growing pressures, and appropriate land use may recover local vocationally and knowledge, as that linked to traditional no-food crops as, for example, fibres. Emergence of ecological agriculture, and precision production (information intensive) are likely to characterize twenty-first century. The management of production is expected to trend towards micromanagement of each specific production site, in order to lower impacts, reduce environment degradation, minimize costs, enhance crop quality. Precision farming will allow to use crop models and monitoring technologies (including GPS, remote sensing) to apply correct inputs, adjusted to the ambient environment, including meteorological variables.(Mc Bratney et al., 2005).
At the same time, awareness is increasing about the importance of ecological agriculture, that shifts management practices to apply low-input paradigms, as self-regulating pest management systems instead of pesticide applications, diverse crop or livestock instead of monocultures. The ecological management of agroecosystems that addresses energy flows, nutrient cycling, populationregulating mechanisms and system resilience can lead to the redesign of agriculture at a landscape scale. (Shennan, 2008). Strategies in agricultural activities to help farmers to better and less riskilyproduce in a climate change scenarios constitute a potentially great value for agriculture. At the same time, the adoption of environmentally sustainable practices in the field may positively reflect on climate, allowing agriculture itself to positively act to contrast climate change. The concept of Good Agricultural Practices has evolved in recent years as a result of the concerns and commitments of a wide range of stakeholders about food production and security, and the environmental sustainability of agriculture. Correct knowledge and interpretation of climate urgencies may address on-farm and post-production processes less sensitive and decrease vulnerability (ACCRETE project, 2007). Adequate land use and resource preservation policies must be coupled with appropriate energy saving actions. Biofuels are a potential low- carbon energy source, but whether biofuels offer carbon savings depends on how they are produced. Converting rainforests, savannas, or grasslands to produce food crop has been demonstrated to create a "biofuel carbon debt" by releasing 17 to 420 times more CO2 than the annual GHG reductions that these biofuels would provide by displacing fossil fuels. In contrast, biofuels made from waste biomass or from biomass grown on degraded and abandoned agricultural lands planted with perennials incur little or no carbon debt and can offer immediate and sustained GHG advantages (Fargione et al., 2008). The intensification of transports is one of the more preminent examples of deterimental effects of the globalization on the environment. Previsions are that, in 2030, 45% of the global GHG production will be due to tranports (Fig. 1.3). A reduction of related energy consumption has to be looked for, with the attemp to decrease the energy embedded into the food , and the impacts on climate changes. This would be possible if and when local and seasonal products consumption will be favoured and enhanced, to minimize the carbon foot-print that some imported and air-freighted fresh fruit and vegetables may be making to the planet. The transport, for example, of 1 kg of cherries from Argentina to Europe (12.000 Kms) has been weighted to contribute to the release in atmosphere of about 16 Kgs of CO2.
10
Figure 1.3: Sectorial emission trends and projections in the EU. Data from European Commission
New approaches are hence in a holistic perspective, that includes environment, food safety and other non-monetary benefits, as climate change mitigation (Pretty, 2008). This new vision minimizes the use of those non-renewable inputs that cause harm to the environment or to the health, makes productive use of the knowledge and skills of farmers, respect and revaluate traditions, so substituting human capital for costly external inputs. Outcomes will be positive for productivity, resource and energy, carbon balance, climate. 1.4 References
ACCRET-E, 2008. Code of attitudes to prevent impacts between agriculture and climate change. Printed in English, Italian, German, Greek, Slovenian, Romanian, Czeck versions, available on-line at http://www.accrete.eu/ and http://www.agrometeorology.org/ Alcamo J., J.M. Moreno, B. Novky, M. Bindi, R. Corobov, R.J.N. Devoy, C. Giannakopoulos, E. Martin, J.E. Olesen, A. Shvidenko, 2007. Europe. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the
11
Fourth Assessment Report of the IPCC, M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson, Eds., Cambridge University Press, Cambridge, UK, 541-580. Audsley E., K.R. Pearn, C. Simota, G. Cojocaru, E. Koutsidou, M.D.A. Rounsevell, M. Trnka and V. Alexandrov, 2006. What can scenario modelling tell us about future European scale agricultural land use, and what not? Environ. Sci. Pol., 9,148-162. Bindi M. and M. Moriondo, 2005. Impact of a 2 C global temperature rise on the Mediterranean region: Agriculture analysis assessment. Climate change impacts in the Mediterranean resulting from a 2C global temperature rise, C. Giannakopoulos, M. Bindi, M. Moriondo, and T. Tin, Eds., WWF, 5466. COPA COGECA, 2003. Factsheets: Assessment of the impact of the heat wave and drought of the summer 2003 on agriculture and forestry. http://www.copa-cogeca.be/en/dossiers.asp Easterling W.E., P.K. Aggarwal, P. Batima, K.M. Brander, L. Erda, S.M. Howden, A. Kirilenko, J. Morton, J.-F. Soussana, J. Schmidhuber and F.N. Tubiello, 2007. Food, fibre and forest products. Climate Change 2007. Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the IPCC, M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson, Eds., Cambridge University Press, Cambridge, UK, 273-313. EEA, 2004a. Impacts of Europe's changing climate: an indicator-based assessment, EEA Report No 2/2004, 107 pp. EEA, 2004b. EEA Signals 2004. A European Environmental Agency update on selected issues. 31 pp. EEA, 2005a. European Environmental Outlook. EEA Report No. 4/2005, 85 pp. EEA, 2005b. Vulnerability and adaptation to climate change in Europe. EEA Technical report No 7/2005, 79 pp. European Commission, 2007. Green Paper from the Commission to the Council, the European Parliament, the European Economic and Social Committee and the Committee of the Regions - Adapting to climate change in Europe options for EU action {SEC(2007) 849} Fargione J., J. Hill, D. Tilman, S. Polaski, P. Hawthorne, 2008. Land clearing and the biofuel carbon debt. Science, 319, issue 5867, 1235-1238. IPCC, 2001. Climate Change 2001, Impacts, Adaptation and Vulnerability, Section, Hydrology and Water Resources, Report of Working Group II of the Intergovernmental Panel on Climate Change. IPCC, 2007. Summary for Policymakers. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the IPCC [Solomon, S., D. Qin, M. Manning,Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, UK and New York, NY, USA. Maracchi G., O. Sirotenko, M. Bindi, 2004. Impacts of present and future climate variability on agriculture and forestry in the temperate regions: Europe. Climatic Change, 70(1), 117 135. Mc Bratney A., B. Whelan, T. Ancev, J. Bouma, 2005. Future directions of precision agriculture. Precision Agriculture, 6, (1), 7-23. Olesen J.E., T.R. Carter, C.H. Daz-Ambrona, S. Fronzek, T. Heidmann, T. Hickler, T. Holt, M.I. Mnguez, P. Morales, J. Palutikov, M. Quemada, M. Ruiz-Ramos, G. Rubk, F. Sau, B. Smith, M. Sykes, 2006. Uncertainties in projected impacts of climate change on European agriculture and terrestrial ecosystems based on scenarios from regional climate models. Climatic Change (in press). PESETA, 2006. (Projection of Economic impacts of climate change in Sectors of the European Union based on boTtom-up Analysis) http://peseta.jrc.es/docs/Agriculture.html
12
Pretty J., 2008. Agricultural sustainability: concepts, principles and evidence. Phyl. Trans. Royal Soc. B. Biol. Sci., 363, (1491), 447-465. PRUDENCE, 2005. Prediction of Regional scenarios and Uncertainties for Defining EuropeaN Climate change risks and Effects. Final Report (http://prudence.dmi.dk) Rees W.E., 2006. Globalization, trade and migration: undetermining sustainability. Ecological Economics, 59 (2), 220-225 Richter G. and M. Semenov, 2005. Re-Assessing Drought Risks for UK Crops using UKCIP02 Climate Change Scenarios. Final report of DEFRA Project CC0368. Schrter D., W. Cramer, R. Leemans, I.C. Prentice, M.B. Araujo, N.W. Arnell, A. Bondeau, H. Bugmann, T.R. Carter, C.A. Gracia, A.C. De La Vega-Leinert, M. Erhard, F. Ewert, M. Glendining, J.I. House, S. Kankaanp, R.J.T. Klein, S. Lavorel, M. Lindner, M.J. Metzger, J. Meyer, T.D. Mitchell, I. Reginster, M. Rounsevell, S. Sabate, S. Sitch, B. Smith, J. Smith, P. Smith, M.T. Sykes, K. Thonicke, W. Thuiller, G. Tuck, S. Zaehle, B. Zierl, 2005. Ecosystem service supply and vulnerability to global change in Europe. Science, 310(5752): 1333-1337. Shennan C., 2008. Biotic interactions, ecological knowledge and agriculture. Phyl. Trans. Royal Soc. B. Biol. Sci., 363, (1492), 717-739.
13
14
2. Agroclimatic indices and simulation models
2. AGROCLIMATIC INDICES AND SIMULATION MODELS Josef Eitzinger, Sabina Thaler, Simone Orlandini, Pavol Nejedlik, Valentin Kazandjiev, Visnja Vucetic, Tor H. Sivertsen, Dragutin T. Mihailovic, Branislava Lalic, Emmanouil Tsiros, Nicolas R. Dalezios, Andreja Susnik, K. Christian Kersebaum, Nicholas M. Holden, Robin Matthews Abstract During the past decades, in connection with the development of computers, many new software tools were developed to be used for agricultural research as well as for decision making. For example, crop and whole farm system modelling, pest and disease warning models/algorithms, models for irrigation scheduling or agroclimatic indices can help farmers significantly in decisionmaking for crop management options and related farm technologies. In research, models can be used to simulate and analyse the complex interactions in the soil-plant-atmosphere system for example in the important field of climate change impacts on agricultural production. All these modelled systems and their interactions include however many different kind of uncertainties and limitations, such as trends in technology and human activities, models representation of reality, lack of knowledge on system responses or lack of calibration data. Much research was done worldwide in the field of model development, model improvements or model comparisons. Also in Europe in many countries significant work was done in this area. The aims of Working Group 1 of COST734 are a review and assessment of agroclimatic indices and simulation models relevant for various European agricultural activities. The results of an europeanwide survey are presented in this study. It includes an overview of most used agrometeorological indices and process oriented models for operational and scientific applications, an analysis of the limitations for applications and an overview of spatial applications in combination with GIS and remote sensing in Europe.
2.1 Introduction Agricultural activity is strongly related to climatic variables that are the main causes of the year to year variation of quantity and quality of production. Physiological processes (respiration, photosynthesis, transpiration), as well as growth and development are influenced by the patterns of each climatic variable and their combination diurnally, monthly, seasonally and annually.
15
The direct consequence is differences in the success of cultivation, and profit for the growers. To determine the relationships between climatic conditions and agricultural systems, several indices and simulation models can be used. Simulation models and indices describe the effect of climate on a specific crop and a specific process (phenological development, growth, damage on production, disease development and water balance). In this way information can be obtained concerning the chemical, biological and physical relationships among the system components. At the same time this knowledge can be applied to manage the system with respect to irrigation, crop protection, harvesting, fertilisation and other management tasks. The aims of WG1 of COST734 are a review and assessment of agroclimatic indices and simulation models relevant for various European agricultural activities. This includes an actual overview of agroclimatic indices and simulation models applied in research or in practice. The relationships with specific crop responses as well as an description of important model outputs and index thresholds relevant to evaluate the responses of crops to climate change and variability are an important aspect to be investigated. Main problems relevant for operational applications in Europe should be identified as well. Concerning climatic impacts, the activity of Working Group 1 mainly addresses the following hazards, directly or indirectly arising from atmospheric conditions: rainfall, flood, frost, drought, hail, heat wave, seasonal shift (length of growing season, budbreak), change in pest and disease, fire, wind and snow. Their evaluation will be done according to the characteristics of selected crops in terms of seasonal development, characteristics of production, cultivation methods, etc. General descriptions can also be obtained, such as the number of frost days during the year, total precipitation, length of dry period and others. WG 1 tasks were devided into 4 subgroups, dedicated to analyse specific questions of a survey of the COST734 countries: Subgroup 1 Agroclimatic indices in use and research; Subgroup 2 Application of process oriented models in Europe; Subgroup 3 Present applications and limitations of crop models in Europe; Subgroup 4 Crop simulation models in combination with Remote Sensing and GIS.
2.2 Agroclimatic indices in use and research (Subgroup 1) 2.2.1 State of the Art At present, relative to accumulation of materials concerning interrelated plant and weather observation, taking the ever-increasing efficiency of computer
16
technology into consideration, multifactorial relationships are being developed which more fully expressed the complex impacts of climate conditions on plant productivity. Gain understanding and visualizing of the real present and possible future state of agroclimatic conditions could be chiefly useful to address the research and the political action in order to control the change of the natural balance of the soil-plant-atmosphere system that for the time being it seems to be completely tempering. It is deemed that the definition of climate indices is the point of departure of the study of the effect of the climate change, as matter of the fact that the climatic indexes are the synthesis of climate condition in a unit of time and space, we considered these are the underlying reason for we dedicated a working group to analyses this topic and to prepare progress report. Quantification by physical methods is the basis of researching and understanding of processes that explain phenomena determining drought, growth, development, yields, pest and disease of important plant and animal species in agriculture. When the extent of measurements is limited, but also to get useful information without the application of frequently too complex physical and biological simulation models, agrometeorological indices are a very powerful tool to (semi-) empirically relate studied phenomena to such environmental observations. Basic of agrometeorological indexes information is presented in this chapter, providing a look at the state of contemporary European research, as well as its application for the purpose of the more efficient use of climate resources and in order to control the impact of climate change on agriculture. In particular, questions 1 and 3 of the text are dedicated to the just exploration of the agroclimatic indices largely used at the moment in the COST 734 countries. The analysis has been finalized to assess which are the repercussions on the present and the futures climate because agroclimatic indexes are more and more used for the construction of climatic maps and classification. Considered processes are: drought, heat stress, excess rain, frost, snow cover, crop responses (incl. seasonal shifts, phenology, quality aspects and others aspects, such as forest fire). They can really provide exhaustive picture of the agroclimatic conditions of European agricultural areas, and the analysis of their temporal trend can provide indications about the current and future impact of climate variability and change. Their simple structure makes their application easier, so representing strong indicators for climate change studies. At the same time they are frequently used at operational level by extension services, providing the end-users with very useful information to manage and plan agricultural activities.
17
Agrometeorological indices The simple climatic variables, easy to be collected and analysed, are not able to fully describe the complex relations existing between climate and crops development and production. For this reason many agrometeorological indices, derived from the combination of meteorological data, were set up with different aims. Agrometeorological indices allow defining quantitatively the available resources for the different cultural needs. The obtainment of reliable information is obviously related to the quality of data and to the availability of a sufficient number of measurement points for the territory representation. For this reason a preliminary study of the geotopographycal characteristics of the interested area represents an important point. This information, in fact, is able to give indications about the best distribution of measurement points in order to represent the spatial and temporal variability of meteorological variables by using the lower number of stations. On the other hand, the quality of collected data needs to be controlled in order to avoid possible gaps in the series used for the calculations and to eliminate records affected by errors. For the calculation of the indices the following steps are required: -availability of the data: data have to be collected by specific meteorological station or by already existing network of stations. Usually hourly or daily data are required; -transmission of the data: depending on the indices, data need to be manually collected or automatically transmitted with a regular time step; -data control: because indices are the result of a combination of different meteorological data, it is particularly important that single data are controlled and eventually corrected before indices calculation in order to avoid the propagation of errors; -calculation of the indices: automatic procedures can be set up for the indices calculation; - elaboration of results: the single values, but also the annual trends and the interannual variability can be considered in order to obtain information concerning the management of the systems and they planning. The use of Agrometeorological indices The development of farming technologies and management brings quite complicated managerial systems which require different inputs in order to optimize the decision process. Agrometeorological indices provide required specific information about the stage and value of a particular parameter of weather, crops and/or reflect a certain relation in between and the impact of the weather on crop behaviour. General use of agrometeorological indices is in monitoring the development of the plants through their phenological stages.
18
They are either used as a concrete single value/tool giving the information about a specific parameter entering a certain information agricultural system or as an input into the models. The basic agrometeorological data are related to the environmental conditions and comprise atmospheric, soil and biological parameters together with the managerial practices taken at the particular time of the growing season. Computerized information systems require the agrometeorological data in a certain form in order to process it. Standard agrometeorological indices can be divided into categories according to their scope and character and the way of use. The way of use is defined by the purpose of the evaluation which differs both in time and space. The time scale is done by the character of the information and comprises following information: -evaluation of the past (pastcasting), -actual real time evaluation giving a very short prediction (nowcasting), - forecast. Some of the agrometeorological indices are set specifically to a certain time scale evaluation (f. ex. the indices describing snow cover concentrate on pastcastimg) but many of them are used in all time scales including pastcasting and forecasting (indices describing pest and diseases detection and prediction). Big part of the agrometeorological indices is a part and form agrometeorological forecast. Agrometeorological forecast is applied at both national and regional levels influencing market planning and agricultural policy and insurance policy as well as in farming practices. The main use of agrometeorological information is in the recent time in Europe concentrated on the farm level applications. The decision making process requires full information about the situation in the past, in the present and the final decision is usually based on the forecast. Agrometeorological indices calculation has to be based on in situ measurements and is site specific reflecting the conditions of the particular locality. Meteorological data form the basis for agrometeorological indices calculation. This data are of two categories: the data to calculate the indices and the data to calibrate and assess the models. Further to the farming practices the agrometeorological indices and models enter the evaluation of larger areas at the regional and country levels. The spatial scale at which the indices are used and at which the models operate is particularly relevant. In this cases weather data and further inputs over larger areas are required and remotely sensed data are widely employed as the input to calculate the indices. Increasing recognition of climate change and its impacts leads to the development of long-term agrometeorological prediction/forecast. Forecasting both the changes in the yields and in the quality production can support the planning of mitigation and adaptation activities. This brings a force for further
19
development of agrometeorological indices and simulation models as some new phenomena have to be investigated and observed mainly regarding the impact of higher level of carbon dioxide on C3 crops by increasing photosynthesis and decreasing water use. The changes of the climate also bring a significant impact on pests and diseases occurrence and further impact on risk management in crop growth. 2.2.2 Present use of agrometeorological indices in Europe A survey across Europe has been done in order to get the information on the present use of agrometeorological indices. Following countries have answered the questionnaire: Austria, Bulgaria, Croatia, Czech Republic, Germany, Greece, Finland, France, Hungary, Italy, Netherlands, Norway, Poland, Romania, Serbia, Slovakia, Slovenia, Spain and Switzerland. Not all answers were complete and some information was not available. Nevertheless, the summary gives good review about the use of agrometeorological indices in Europe in the recent years. Both agrometeorological monitoring and service is mostly operated by the national state bodies, in 12 cases national meteorological and hydrometeorological institutes provide these services. Agricultural services act in some countries at national level but they run the mostly advisory services at the regional level. Private agrometeorological services are scattered and usually concentrate on some specific points of service like extreme weather warning service /in AT and NO/ or advisory services in case of plant protection against pests and diseases /in SK/. In Slovenia (SI) beside agrometeorological network at Meteorological Office another agrometeorological network is applied under the authority of the Ministry of Agriculture, Forestry and Food (at present the Phytosanitary Administration of the Republic of Slovenia) which has been since 1998 used for public plant protection service and to minor extend also for irrigation. A complex information service for farmers including special weather forecast for farmers provided by a private company is organized in FI. In Greece, there is also a special weather forecast provided by the National Meteorological Service and broadcasted by the National Television Network. The forecast also provides advisory information for new practices, products and has special issues for plant protection according to phenological stage and plant species. Except from the National Meteorological Service and the National Observatory of Athens, Universities and private or public institutes provide weather and agrometeorological information. These informations are provided on users request or published as results of research programs. In some cases companies
20
selling chemicals or other materials and equipments to farmers include also some technical support and agrometeorological services and/or forecast as a part of their businesses. General agrometeorological information issued at the national level is mostly produced by national bodies e.g. meteorological services as they run the meteorological networks so that they are the owners of the databases. In many cases they cooperate with other national bodies providing them the data either free of charge or at the commercial base. Further description is focused on the use of the agrometeorological indices both in operational practices and in the research. However, some descriptions include also the use of some process oriented models the interim results of which are used as the agrometeorological indices mainly to evaluate the plant development and water balance. Water balance represents the top parameter regarding the plant growth. That`s why the indices describing various components of the water balance are most frequently used at various levels. Droughts and excess rain together with the pests and diseases occur as the most frequent detected phenomenon in operational agrometeorological practices (Fig. 2.1). The territorial distribution of the use of the indexes follows the territorial distribution of the natural phenomena while pest and diseases indexes are used equally through the whole Europe. Further to the natural phenomenos shown in Fig. 2.1 and Fig. 2.2 there are some agrometeorological indices describing mostly extreme weather events like hail and forest fire indices. The research activities regarding the development of the agrometeorological indices is clearly focused on the drought (Fig. 2.2) and the attention paid to the development of other indices does not correspond with their practical use in operational practices. Relatively little attention is paid to the operational monitoring of heat stress while the majority of responding countries notices the research activities in this field. The drought indexes are in many cases parameterized reflecting the impact of local climate and the activities being given doth to the research and operational use correspond. In many other cases the standard climatic and agrometeorological indices are applied. Generally, the agrometeorological indices are widely used in operational practices through Europe.
21
18 16 14 12 10 8 6 4 2 0
EXCESS RAIN CROP RESP. FROST P&D SNOW COVER DROUGHT HEAT STRESS OTHER OTHER
Nb. of countries
Event
Figure 2.1: Operational use of agrometeorological indices in European countries
18 16 14 12 10 8 6 4 2 0
DROUGHT FROST SNOW COVER HEST STRESS EXCESS RAIN CROP RESP. P&D
Nb. of countries
Event
Figure 2.2: Use of agrometeorological indices in research in European countries
Operational use of agrometeorological indices The use of agrometeorological indices and models varies from country to country and there is no information about the frequency of use of the indices and models inside the particular country. Many of agrometeorological indices used operationally are regionally based. The indices are selected and the models are parameterized towards the regional and environmental conditions
22
of the concrete region. However, in some countries such as in France the operational agrometeorological service is run by the regional bodies and it is difficult to collect full information about the use of the particular indices in the region. That`s why the further description of the use of agrometeorological indices will be rather topic oriented than country oriented. Following the questionnaire the indices and some partial model outputs were divided into the categories according to the phenomenon they are dealing with as in Fig. 2.1 and 2.2. Drought Drought indices are constructed to quantify the lack of water during plant growth and development cycle. The general problem of these indices is to include the physical and biological properties of the particular crop in order to reflect its sensitivity and limitations towards the lack of water supply during the vegetation period. Some indexes in use define the meteorological drought which does not in all cases describe the real shortage of water for the crops. Further problem of defining the drought is the time step used to calculate the particular indices. There was only one index bringing a certain forecast of drought listed. The major part of the indices is focused on pastcasting and some of them on nowcasting (Table 2.1). The description of the indexes enables to give an overview of what the services provide but only a few concrete indices were listed. There is a NPET- adapted crop drought model used in Austrian Weather service and Index of Hydrometeorological Drought /IHS/ used in the practice of Czech Hydrometeorological Service and. IHS is calculated out of water balance components as follows:
IHS = [VBi VBp ] ; where VBi is actual and VBp is multiyear mean of
i =1 n
water balance. VB represents a difference between precipitation and actual evapotranspiration and is parameterized by a coefficient depending on the season. These types of the indices are often regionally based as they have to use multiyear measured values of the particular parameters recorded or calculated for a certain locality. The major part of the indices in use are rather complex and deal with water balance components and precipitation measures. Water balance components are used in various modifications in almost all countries in the extent from national level to a farm level. A 10x10 km uniformed grid system is used for the production of all indices and parameters. The main users are listed farmers and extension services. Both the indices based on water balance components and on the precipitation amount for a given period are produced mainly by
23
national weather services as they run the meteorological networks at the regional and national level. In Greece, for example, several drought indices are used for the assessment and estimation of drought for research or operational (on users request) purposes. Apart from the standard indices (SPI, PDSI, PHDI, Z-index, RAI etc.) (e.g. Dalezios 1988; Loukas et al. 2002; Tsakiris et al., 2006), two other indices used in Greece are the Reconnaissance Drought Index (RDI) and satellite derived Vegetation Health Index, representing meteorological-hydrological drought and agricultural drought, respectively (e.g. Tsiros et al. 2004, 2008; Tsakiris et al., 2005,2006; Kanellou et al., 2008). Some institutes use the partial outputs of the models like WOFOST to define the days with the lack of water for the crops /in NL and SK/. In SI the irrigation model IRRFIB is used for daily calculation of crop water balance for different regions. It represents agricultural decision support tool which is running inside the Slovene Agrometeorological Information System (SAgMIS) package. From the standard indices SPI, PDSI, percent of normal precipitation and rainfall percentiles are in operational use. Some preliminary maps of the SPI (Fig.2.3), Percentiles and Precipitation for the region of southeastern Europe are products issued by Drought Management Center for South eastern Europe (DMCSEE) situated in Slovenia from 2006 and published on the web page http://www.dmcsee.org/. They are updated once per month. Final data maps with two months delay are available after 20th day of the current month. Firstguess maps are available after the 5th day of the next month.
Excess rain Excess rain as a water related phenomenon is observed in all countries by simple measurements of daily sums of precipitation. Further to this parameter rainfall intensity is measured either by pluviographs or by weight rain gages providing on line signal. Major part of listed rainfall parameters in Table 2.1 are issued in the standard forecast of each meteorological service mainly at the regional scale. Some of the services provide special rainfall maps in their pastcasting identifying the areas with high precipitation and/or anomalies. In Greece, for example, apart from high precipitation past-casting (P) maps and the examination time-series of rainfall data in different regions for the investigation of the variation of precipitation during the second half of the 20th century, an operational-research application of the non-hydrostatic model LMCOSMO of HNMS (Hellenic National Meteorological Service) has been used
24
for forecasting excess rain events. The model has been used for the simulation of severe thunderstorms (Avgoustopoulou et al. 2002). The data are collected from stations of the Hellenic National Meteorological Service and the Ministry of Agriculture. Generally excess rain represents a damaging weather event and its characteristics are usually issued for general use stressing the regional differences. However, it is difficult to use such information at the local/farm level.
Figure 2.3: SPI for southeastern Europe issued for February 2008
Heat stress Heat stress is a complex function of the height of temperature, duration and rate of increase of the temperature. The thresholds of the temperatures for the crops differ pretty much and they vary also according to the plant development stage. A threshold of the heat stress refers usually to the daily mean temperature over which a detectable reduction of growth or damages in plant begin. The heat stress prediction is naturally included in general weather forecast though there are very few services listed which provide special heat stress related indices (Table 2.2). Heat index forecast is provided by Hungarian Meteorological Service which includes the forecast of daily average temperature above 25C. In Greece, forecasts of surface temperature and wind speed over Attica and neighbouring areas are provided using the non-hydrostatic model MM5. This model has very high resolution (grid distance of 2 km) and the forecasts of the parameters are calculated every 18 hours (Kotroni and Lagouvardos 2002). The provider is the National Observatory of Athens, Institute of Environmental Research and Sustainable
25
Development and the aim was to provide accurate forecasts in order to improve the weather forecast for the Olympic Games 2004. Further to that three countries listed a pastcasting service at the national level providing different parameters of temperature. This information is usually issued for concrete farmers and consultants.
Frost The critical temperatures needed for damage to occur may vary depending on the duration that temperatures remain below freezing point. However, the detection and prediction on frost conditions considers it as the temperature below 0C. Frosts are frequently classified as either advective or radiative and this also defines their impact on the different type of crops. During radiative frosts occasions in many cases the frost line does not reach more than 1 2 m above ground so that only the crops close to the ground are affected by frost. Frost events are both forecasted and monitored by the national Meteorological services in all countries. A standard weather forecast includes the forecast of the frost or the possibility of ground frost occurrence. However, only a few special indices in operational use focusing on the nowcasting and pastcasting were listed. Frost forecast is usually issued at the national level for general purposes while the special indices listed in Table 2.2 are mainly used by farmers and consultants and insurance companies. Snow cover The snow cover brings a valuable protection of plants against hard frosts during the winter. On the other hand a long duration under unfavourable conditions can bring rotting out of the plants below the snow cover. The indices dealing with the snow cover are mostly focused on the postcasting which is done daily at different spatial scales from 10x10 km grids in Finland to the regional and national scale in the rest of countries (Table 2.2). In some cases the water content of the snow cover is reported which brings the possibility to estimate the amount of the water being stored in the snow cover as a water source in the spring. Specific events Further to the above listed indices some special agrometeorological indices in operational use are listed in Table 2.3. The use of forest fire/grass indices listed only three countries. However, these indices in various forms are in wide use through Europe, mainly in Mediterranean. Considering increasing occurrence of forest fire events more frequent use of these indices is expected. The German Weather Service (DWD) provides a daily risk index for forest fire which combines several indices: a Swedish index (Angstm), two German
26
indices (Baumgartner, M-68) and the Canadian forest fire warning system (FWI: Fire Weather Index, FFMC: Fine Fuel Moisture Code) (http://www.agrowetter.de/Agrarwetter/Waldbrand_en.html). Special weather forecast for farmers and complex growing season information is provided daily at the scale of 10x10 km by Finish Meteorological Service and a private company in Finland. This information includes probability of rain and frost, rain amount, temperature, relative humidity, wind speed and direction, index describing weather conditions for plant protection. The German Weather Service provides up to 4 times a day actualized 7-day forecasts concerning the drying of hay and grain moisture of cereals and maize. Additionally, recommendations are given for the sowing day of winter cereals, oats, potato, sugar beets and maize for the upcoming 6 days. Regarding to hail events, an operational project has been carried out in Greece, the Greek National Hail Suppression Project (NHSP) weather modification program. The objectives were to reduce hail damage and at the same time to examine and study the thermodynamic, dynamic and microphysical characteristics of the potential hail producing clouds. Also, Instability Indices are calculated for Operational Hail Forecasting in Greece. Some services provide information about the workability of the soil with the regards to the depth of the frozen soil considering also the impact of frost on lumps of clay during the winter.
Crop response and pests and diseases monitoring There are not many services monitoring the response of the crops to the weather regarding the growth and phenological development (Table 2.4). Operational phenological networks which comprise a sufficient number of stations work mainly in the region of Central Europe and in some countries in Balkan. These networks are run by the Meteorological Services and systematically monitor phenological development stages of selected plants and in several cases crop development including some pheno-metric parameters, pests and diseases and yields. The use of the data is mainly in pastcasting. In some cases some special parameters are monitored by remote sensing (greenness index). Remote sensing of phenological parameters is intensively used at the European scale by JRC Ispra within the MARS project. On the other hand crop parameters including yields and the level of pest and diseases occurrence are widely simulated either by using special simulation models or by using partial outputs of crop growth models. An exemplificative list of pest and diseases being monitored and forecasted by various indices and models is shown in Table 2.4. A special set of parameters regarding the plant condition close to the harvest is provided by German Meteorological Service. Further to that either standard (WOFOST) or specific (IPHEN) models are used to simulate the development of different plants.
27
Table 2.1:
Reported operational use of agroclimatic indices including statistical models estimating water availability Type of use Time step Spatial realization Aim of use Country of use
Used index/model
Soil water content for top 10 cm Index of hydrometeorological drought NPET
5 day
Drought Regional
Weekly
National
N, P
daily
National
Water balance components
N, P
Soil moisture content VHI
P, N
daily, weekly monthly, yearly daily, weekly, monthly Monthly
from site specific to national 10x10 km to national National
estimating drought affected regions estimating drought affected regions estimating drought affected crops and regions estimating drought affected crops and regions estimating drought affected areas estimating drought affected crops and regions estimating drought affected regions estimating drought affected regions water supply estimating drought affected regions estimating drought affected areas estimating monthly rainfall towards the normal precipitation deficit generating dry days
CH
CZ
AT
CZ, F, I, NO, PL, SI, SK, SR DE , FI, IT, RO, SI, SR GR
PDSI
on users request Monthly
National
HR, SR, GR I
Water table depth levels Usable water supply Palfai Aridity Index
Regional
P P
Weekly Year
National National
CZ HU, SI
Precipitation totals and anomalies Rainfall percentile
weekly, monthly, yearly monthly
Regional, National National
AT, I, PL, SI, SR HR
SPI WOFOST
P P
monthly daily
Regional Regional
ES, I, SI, SR, GR NL, SK
28
Excess rain Rainfall amount F hourly, daily national, regional, local nationaldistrict scale Regional National estimating affected regions estimating affected regions estimating affected regions estimating affected regions estimating affected regions? estimating affected regions All countries AT, DE, PL, SK CH GR
Rainfall intensity 5 days probability forecast of 1,0 mm+ Daily forecast of 25,0 mm+ precipitation Rainfall regime classes SPI maps
F F F
hourly daily
F P monthly, thirty, sixty and ninety days-daily calculation monthly decade
Regional Regional, national
RO I, SI, SR, HR
Precipitation totals anomalies Palmer`s Z index
P P
Regional national station network
estimating affected regions estimating affected regions
AT, I, SI SR
(F) forecasting, (N) nowcasting, (P) pastcasting
Table 2.2: Reported operational use of temperature related indices Used index/model Type of use Time step Spatial realization Heat stress Regional national station network Aim of use Country of use
Heat Index Maximum temp. of soil surface, upper canopy and under plastic cover Temperature sum Temperature percentile
F P
Daily monthly, growing period
days with heat stress heat stress indication for different crops and periods estimating affected regions estimating monthly air
HU AT, DE
P P
if needed monthly
National National
HR, SI HR
29
Heat Units Index /sum of daily maximum above 32C/ Number of days with maximum daily air temperatures 30 o C and 35 oC Deviation of mean maximum air temperature from long-term average
National
temperature towards normal days with heat stress
RO
monthly
national station network
Number of days with heat stress
AT, HR, SI, SR
daily, decade and monthly
national station network
Number of days with heat stress
AT, I, SR
Frost Frost forecast and ground frost forecast ALADIN regional model forecasts Winter cereals hardiness modelling Start and end of glazed frost risk Frost Units Index F Daily 10x10 km to national district scale district scale estimating affected regions temperature fields maps winter resistance All countries CZ, SI CZ
F, N N
hourly
P P
Daily Nov-March
National National
estimation of regional frost risk frost occurrence
DE RO
P Maximum snow cover Snow depth Snow cover maps Snow depth and water content P P P P
Daily Daily Daily
Snow cover National National 10x10 Regional
Daily
National
estimation of snow periods estimation affected areas Snow cover occurrence Snow cover occurrence estimation affected areas
DE, SI HR, SI A, FI, SI, SR IT SK
30
Table 2.3: Operational use of other agroclimatic indices including statistical models Used index/model Hail Type of use F Time step Spatial realization national hail protection network National Aim of use Country of use GR, SR
daily
hail appeareance
Instability Indices for hail forecasting Weather forecasts for farmers Growing season information Gras fire risk index Forest fire index FROSTGAR Depth of frozen soil Duration of global radiation, Potential evapotranspiration
On users request daily daily hourly daily daily daily Daily, decade
hail appeareance
GR
F F, P N P,N P P P
10x10 km 10x10 km National National National National national station network
various parameters various parameters grass fire risk forest fire risk soil workability soil workability calculation of potential evapotranspiration within the calculation of water balance
FI, GR FI DE DE, SI, SK, HR DE DE, SK AT, HR, SR
The use of agrometeorological indices and statistical models in research The set of indices used in research is relatively rich and reflects the needs of farming practice and advisory and monitoring services at various levels. They are mostly oriented towards monitoring regards of the impact of extreme factors and their forecasting. Research activities are done in many cases by the institutes which do not run operational services, and the transfer of the technology brings practical obstacles in missing operational database data transfer or dissemination tools. Many of the indices were designed in a research context for use at the scale of a field. They require considerable simplification to be usable under operational conditions. Further to that the number of these indices is much bigger than the number of those being used in operational.
31
Table 2.4: Operational monitoring of crop responses and pests and diseases detection Used Index/model Type of use N P P P, N Time step Spatial realization Observed National Regional Local to national National Aim of use/ notice Country of use
Greeness index Phenomaps Crop yields Phenological stages
unregula r Seasonal ly Daily, unregula r Daily
By RS
AT I, SI All AT, CH, CZ, DE, HR, RO, SI, SK All
Pest and diseases occurrence Crop damages
Local to national
When occurs Daily
Peronospora, Phytophthora infestans Apple scab, Delia brassicae, Psila rosae, Plasmospora viticila, Potato beatle - Leptinotarsa decemlineata Botrytis cinerea Etc. CLIMEX
F, N, P
local to national Simulated Regional
crop yields estimation estimation of phenological stages detecting pest and diseases occurrence estimating crop damages Pest and diseases detection and prediction
All
Many
10x10x km
STICS
regional?
FALLZAHL /Falling number in wheat and rye/ Cereals grain moisture Water content grass/hay
Daily
National
N, P, F N, P, F
Daily
Daily
National, >400 stations National, >400 stations
direct effect of abiotic factors on pest fruit tree flowering prediction estimation of optimum harvest time estimation of optimal harvest estimation of optimal harvest
FI
DE
DE
DE
32
Dry matter content of maize Sowing date winter cereals, oats, potato, sugar beet, maize Pollen information system IPHEN
N, P, F F
Daily
Daily
F F
daily
National, >400 stations National >400 stations Regional 2x2km
estimation of optimal harvest Estimation of optimal sowing date predicted pollen concentration phenological development of grapevine phenological development phenological development plasmospora, blue mould on grape
DE
DE
AT, HU, CH, SI I
WOFOST Computed phenological dates Galati vitis
N, P
Daily
local, regional
NL, SK RO SK
Weekly
Regional
The list of indices does not include any estimation of yields. Beyond their simplicity, their main advantage is the fact that calculations can be done easily, and in the fact that data requirements are limited. However yield estimation is done more and more often by various crop simulation models (see next chapter) which are rather complex and require real time data of a number meteorological parameters from in situ measurement. Core businesses focus on defining the agro-climatic characteristics of the particular regions and localities and estimating the extent and impact of various events as well as on the forecasting for farming management. More than one third of listed indices are dealing directly or indirectly with drought and its impacts (Fig. 2.4). Further to the drought the research was concentrated on the occurrence and forecast of temperature extremes. In the research of crop responses to the weather phenomenon dominate simulation methods over the observations those some research on making frequent camera pictures during the growing season of particular plant. Some more research activities on agrometeorological indices not being mentioned were taken in the recent past mainly in elaborating forest fire tools and their prevention. No research effort of any agrometeorological indices or models dealing with irrigation systems was mentioned. Further to above mentioned items big research effort has been done when implementing web based techniques for data collection and dissemination of the information and mapping techniques using GIS instruments. Many recent techniques of elaborating spatial information use neural network technology for the approximation and prediction of different
33
agrometeorological characteristics at state scales in different climate zones based on reported daily weather data. The list of indices in Table 2.5, 2.6 and 2.7 brings a certain review of the indices according to the event categories. Country of elaboration/research use is not mentioned in this case as there is an intensive transfer of technologies in research as well as for practical use and in many cases the research is focused on detail parameterization of the particular index towards the local environmental conditions.
Simulated crop response
Drought
Direct crop response
Frost
Excess rain Snow cover
Heat stress
Figure 2.4: The distribution of the numbers of agrometeorological indices used in research according to their purpose
Drought indices in Table 2.5 use a big set of standard parameters and characteristics dealing with water balance and soil water availability for the crops but some special indices towards concrete crops were produced /drought index for grape vine/. No remotely sensed characteristics and parameters entering the indices were mentioned as this is a subject of another chapter. Further to the indices directly estimating the extent of drought and the content of soil water being available for the crops some models were used to estimate the effect of drought period on the crops by using complex model WOFOST and Autoregressive model DARMA. Excess rain is monitored and studied mainly because of its destructive impacts leading frequently to the flash floods and bringing the soil erosion of high extent. Excess rain indices mostly use standard precipitation characteristics and no research of the excess rain towards a particular crop or the period of the development of some crops was mentioned. Extensive research bringing very detailed general statistical characteristics of excess rain at the regional level was done through the whole Europe. New radar techniques as well as almost real time monitoring of precipitation in automated precipitation networks brings further possibility in monitoring the excess rain. The ability of
34
probability forecast of excess rain by NWP models has increased considerably. However, there are only limited possibilities to protect the crops against the excess rain. Snow cover monitoring is done mostly by climatological services and no special indices associating snow cover with crop development were mentioned. The absence of the snow cover is an important issue from this point of view as it brings black frosts during the severe winters which can cause big damages on crops being sowed in autumn. Monitoring of the water content of the snow cover brings good possibilities to estimate the water content in the soil after the winter. Next set of indices in Table 2.6 deals with heat characteristics which bring the possible damage on the crops. Further to the simple temperature characteristics being subjected to the statistical processing Heat index and Actinometric index were mentioned as research tools. Heat index brings a combination of the ambient temperature and relative humidity while Actinometric index describes the capability of radiation to produce a photochemical reaction. Heat index is used also in agrometeorological practice for its simplicity while Actinometric index is not mentioned in practical use. All of these indices bring valuable information but further to the water regime management active protection against heat stress from the high temperatures is rather limited and concentrates to the genetics and breeding. The active protection against the frost is much wider that is why big attention is paid to increase the quality of the forecast of frost. Frost climatology, ground frost climatology at the mezzo meteorological scale plays a decisive role in frost protection planning. That is why many of the indices were investigated by statistical tools in order to detect the areas of frequent occurrence of frost. A special system for winter cereals hardiness modelling mentioned as a research tool is also used in agrometeorological practice.
Table 2.5: Water related agrometeorological indices including statistical models in research and development Used index/model Drought ET (Evapotranspiration) PET (Potential evapotranspiration) ET/PET, SWSI (Surface Water Supply Index), CMI (Crop Moisture Index), RAI (Rainfall Anomaly Index) SPI (Standardized Precipitation Index), PDSI (Palmer Drought Severity Index), RDI (Reclamation Drought Index), Aim of use
Climatic characteristics
35
RDI (Reconnaissanse Drought Index), BMDI (Bhalme-Mooley Drought Index) Blaney-Criddle, Agro-hydrologic potential Index of hydrometeorological drought Drought index for grape vine Palfai Aridity Index Soil moisture content Usable water supply Hydrotermic coefficient, de Martonne Index, Lange rain factor Precipitation totals Rainy spells of X days WOFOST DARMA (Discrete Autoregressive Moving Average model) Excess rain Rainfall amount Rainfall intensity Intensity (rain per rainday) Greatest 3, 5 and 10-day total rainfall Daily totals over a certain threshold Generalized Extreme Value (GEV) distribution Intensity Duration Frequency (IDF) curves Snow cover Snow cover duration Snow depth Water content of snow cover Delineation of snow covered surface Estimating drought affected crops
Estimating drought affected crops and regions and crop management at farm level Climatic characteristics Estimating affected regions
Estimating the frequency of extreme events Estimation of snow periods Amount of snow Volume of water in snow cover Snow cover detection (Remotely sensed)
Table 2.6: Water related agroclimatic indices including statistical models in research and development Heat stress Tmax 90th percentile, Heat Wave Duration, 90th Percentile Heat Wave Duration Temperature sum Days with Tmax > 35 Actinothermic index (10cm, 50 cm height) Frequency distributions of the heat wave events Heat index Intensity Duration Frequency (IDF) curves
Climatic characteristics Heat stress affected regions Estimating the frequency intensity of extreme events and
36
Frost Start and end of glazed frost risk ALADIN regional model forecasts Winter cereals hardiness modelling Duration of the frost free period Number of frost days - Tmin < 0 degC Frost frequency in the Nordic/Baltic region First and last frost day -> frost free period Degree Days Frost days Estimation of regional frost risk Temperature fields maps Winter resistance Climatic characteristics Climatic characteristics Climatic characteristics
Table 2.7: Monitoring of crop responses and pests and diseases detection in research and development Used index/model Directly monitored crop responses Phenological stages cereals and other crops Start of the vegetation (thresholds of 5, 10 and 15oC) Continental-scale estimation of climate change impacts Phenological responses to climate variability Simulated responses for assessing the impact of climate warming Infolding Index (beginning of growing season) WOFOST /temp. sums/ Duration of farming season Simulated crop responses Thermal Growing period (average daily temperature > 5 C) Huglin-Index: temperature sum for site suitability for grape wine varieties Maturation index for vine Heat unit, Photothermal unit, Vapor pressure deficit, Phototemperature, Nyctotemperature and Cumulative precipitation Agricultural weather index / yields of 5 main crops and on the records of 10 meteorological stations/ Dynamics of soil water contents at various depths Heat Units Index Frost Units Index Relative humidity-air temperature relations Aim of use
Estimation of phenological stages Climatic characteristics
Climatic characteristics
Production estimation
37
Table 2.8 and 2.9 summarize reported country related literature and web links on the use and applications of agrometeorological indices in Europe.
Table 2.8. Reported country-related literature on operational used agroclimatic indices, pestdisease warning (see chapter 2.3) and agrometeorological forecasts in Europe Austria Auer I., E. Korus, 2005. The Variability of Heat Waves and Dry Spells in the flat and mountainous Regions of Austria, ICAM/MAP 2005, Croatian Meteorological Journal, 604-607. Auer I., 2002. KLIM der digitale Klimaatlas sterreichs (The digital climatic atlas of Austria). 7. Klimakolloquium 20. Mrz 2002, Universitt fr Bodenkultur, Wien. Auer I., R. Bhm, R. Potzmann, M. Ungersbck, 2003. nderung der Frosthufigkeit in sterreich (Change of frost frequency in Austria). Terra Nostra. Schriften der Alfred Wegener Stiftung 2003/6, 25-29, Selbstverlag der Alfred Wegener Stiftung, Berlin Soja G., A. Soja, J. Eitzinger, G. Gruszcynski, M. Trnka, G. Kubu, H. Formayer, W. Schneider, F. Suppan, T. Koukal, 2005. Analyse der Auswirkungen der Trockenheit 2003 in der Landwirtschaft sterreichs - Vergleich verschiedener Methoden. Endbericht von StartClim2004.C; (Analysis of drought impacts on agriculture in Austria in 2003) in StartClim2004: Analysen von Hitze und Trockenheit und deren Auswirkungen in sterreich Endbericht, Auftraggeber: BMLFUW, BMBWK, BMWA, sterreichische Hagelversicherung, sterreichische Nationalbank, Umweltbundesamt, Verbund AHP Harlfinger O., 2000. Die klimatischen Bedingungen fr den Qualittsweinbau in sterreich (Climatic conditions for wine production in Austria), Der Frderungsdienst 48, Heft 9, S 77-80. Harlfinger O., E. Koch, H. Scheifinger, 2002. Klimahandbuch der sterreichischen Bodenschtzung (Climatic handbook of the Austrian soil taxation), Klimatographie, Teil II, Universittsverlag Wagner, Innsbruck, ISBN 3-7030-0376-6. Koch E., E. Dittmann, W. Lipa, A. Menzel, J. Nekovar, A. v. Vliet, 2007. COST725 establishing a European phenologial database for climatological applications: overview and first results, Tagungsbericht EMS/ECAM Moberg A., P.D. Jones, D. Lister, A. Walther, M. Brunet, J. Jacobeit, L.V. Alexander, P.M. Della-Marta, J. Luterbacher, P. Yiou, C. Deliang, A.M.G. Klein Tank, O. Saladie, J. Sigro, E. Aguilar, H. Alexandersson, C. Almarza, I. Auer, M. Barriendos, M. Begert, H. Bergstrm, R. BHM, C.J. Butler, J. Caesar, A. Drebs, D. Founda, F.W. Gerstengarbe, G. Micela, M. Maugeri, H. sterle, K. Pandic, M. Petrakis, Srnec, R. Tolasz, H. Tueomenvirta, P.C. Werner, H. Liderholm, A. Philipp, H. Wanner, E. Xoplaki, 2006. Indices for daily temperature and precipitation extremes in Europe analysed for the period 1901-2000. J.Geophys.Res., 111, D22106, doi:10.1029/2006JD007103. Mller W., 1993. Agroklimatische Kennzeichnung des zentralen Marchfeldes (Agroclimatic characteristics of the central Marchfeld region).-Beihefte zu den Jahrbchern der ZAMG - Klimatologie, H.3, Wien. Scheifinger H., A. Menzel, E. Koch, Ch. Peter, 2002. Trends of Spring Time Frost Events and Phenological Dates in Central Europe. Fourth European Conference on Applied Climatology, Brussels, 12-15 November 2002. Bulgaria Alexandrov V., 1998. The Effect of Climate Variability and Change on Agroclimatic
38
Resources and Crop Productivity in Bulgaria. Proceeding on the 2nd European Conference on Applied Climatology, October, 1998, Vienna, Austria, (CD) 6 pp. Alexandrov V., 2000. Duration of the Frostfree period during the Warm-half of the year in Bulgaria. Proceedings of the 3rd European Conference on Applied Climatology "Tools for the environment and man of the year 2000" Pisa, Italy (CD), 6 pp. Alexandrov V., B. Dubuisson, J.M. Moisselin, E. Koleva, 2006. A case study on Utilization of Precipitation Indices in Bulgaria. Proceedings of the International conference on Water Observation and Information System for Decision Support (BALWOIS), Ohrid, Macedonia, 23-26 May 2006, (CD) 18 pp. Alexandrov V., M. Schneider, E. Koleva, J.M. Moisselin, 2004. Climate Variability and Change in Bulgaria during the 20th Century. Theoretical and Applied Climatology 79(3-4): 133-149. Kazandjiev V., 2004. Agrometeorological Aspects of Drought in Bulgaria During Last Decade (1994-2003), Proceedings of the EWRA Symposium on Water Resources Management- Risks and Challenges for the 21st Century, Izmir, Turkey 2-4 September, 2004, p. 331-342. Kazandjiev V., N. Slavov, 2006. Phenological Development s Indicator f Meteorological Conditions, Proceedings of BALWOIS Conference-CD version, Ohrid 24-26 May 2006. Koleva E., N. Slavov, V. Alexandrov, 2004. Drought during the 20 Century. In: Knight, C. G., I. Raev and M. Staneva (eds.). 2004. Drought in Bulgaria: A Contemporary Analog for Climate Change. Aldershot, UK: Ashgate Publishing Limited, pp. 53-66. Slavov N., Ek. Koleva, V. Alexandrov, 2000. Climatic Characteristics of Drought in Bulgaria. Bulgarian Journal of Meteorology and Hydrology 11 (3-4): 100-113 Slavov N., Kazandjiev 2003. Phenological Development Investigation of Common Locust (Robinia pseudoacacia L.) Trees in Bulgaria, J. Silva Balcanica v. 5(2), p. 61-70. Petkova N., E. Koleva, V. Alexandrov, 2004. Snow Cover Variability and Change in Mountainous Regions of Bulgaria, 1931-2000. Meteorologische Zeitschrift 13(1): 19-23. Croatia Gaji-apka M., 1996. The snow regime in the Northern part of mount Velebit. Hrvatski meteoroloki asopisCroatian meteorological journal, 31; 1521. Gaji-apka M., 1991. Short-term precipitation maxima in different precipitation climate zones of Croatia, International journal of Climatology, 11, 6; 677687 Mihajlovi D., 2006. Monitoring the 2003-2004 meteorological drought over Pannonian part of Croatia, Int. J. Climatol., Vol. 26, Issue 15, 22132225. (www.interscience.wiley.com, DOI: 10.1002/joc.1366) Pandi K., I. imuni, F. Tomi, S. Husnjak, T. Likso, D. Petoi, 2006. Comparison of three mathematical models for estimation of 10-day drain discharge. Theoretical and Applied Climatology. 85; 107-115 Penzar B., 1976. Drought severity index for Zagreb and its statistical forecast (in Croatian), RaspravePapers, 13, 158. Vueti V., M. Vueti, 1995. Degree days in the Croatian lowlands, International Conference Climate Dynamics and the Global Change Persperctive, Krakov, Poland, 17-20 October 1995, 359364. Vueti V., M. Vueti, 2005. Temperature sums and climatic variations in the Parg region, Croatian Meteorological Journal, 40, 673676. Vueti M., V. Vueti, 1997. Weather influence on weak crop of olives in Dalmatia during 1996, Extraordinary meteorological and hydrological events in Croatia in 1996, 20, 7179. (in Croatian)
39
Vueti M, V. Vueti, . panjol, D. Bari, R. Rosavec, A. Mandi, 2006. Secular variations of monthly severity rating on the Croatian Adriatic coast during the forest fire season, Forest Ecology and Management, 234, supplement 1; 251261 (CD-rom). Zaninovi, K., M. Gaji-apka, 2000. Changes in components of the water balance in the Croatian lowlands. Theoretical and Applied Climatology. 65; 111-117 Czech Republic Bauer Z., 2006. The phenological trends of the components of flood plain of South Moravia in the case of Ulmi-Fraxineta Caripin in the period from 1961-200. Part I. Wild plants phenology., Meteorologick zprvy, 59: 80-86 Dubrovsk M., M. Hayes, M. Trnka, M. Svoboda, D.A. Wilhite, Z. alud, D. Semerdov, 2006. Projection of Future Drought Conditions Using Drought Indices Applied to GCM-simulated Weather Series, The Geological Society of America, Managing Drought and Water Scarcity in Vulnerable Environments: Creating a Roadmap for Change in the United States (1820 September 2006), 2006. Huth R., J. Kysel, L. Pokorn, 2000. A GCM simulation of heatwaves, dry spells and their relationship to circulation. Climatic Change, 46: 29-60. Kysel J., 2006. Spatial variability of heat waves in the Czech Republic and summer temperature pculiarity in southwest Bohemia, Meteorologick zprvy: 59: 183-189 Mozny M.,2006. Monitoring of usable water supply in the surface soil layer under grassland. Meteorolog.Zprvy, 59, 4, s.118-121. Spitz P., J. Filip, 2001. Poteba zvlah pi predikovan klimatick zmn v esk republice (Need for irrigation under the predicted climate change), Seminar: Drought evaluation and forecasting. In proceedings from seminar held on 19.Nov. 2001, Brno, http://www.chmi.cz/meteo/CBKS/sucho01 Litschamnn T., J. Ronovsk, 2001. Palmerv index zvanosti sucha a jeho aplikace pro lokalitu abice, (Palmer drought severity index and its application of the abice station), In proceedings from seminar held on 19.Nov. 2001, Brno, http://www.chmi.cz/meteo/CBKS/sucho01 Trnka M., M. Dubrovsk, D. Semerdov, Z. alud, M. Svoboda, M. Hayes, D. Wilhite, 2006. New Method for Assessment of the Drought Climatology - Czech Republic as a Case Study, European Geosciences Union 2006, April 2-7 2006 Vienna, Austria Trnka M., P. Hlavinka, D. Semerdov, M. Dubrovsk, Z. alud, M. Mon, 2007. Agricultural drought and spring barley yields in the Czech Republic, Plant, soil & Environment (in print) Tolasz R., T. Mkov, A. Valerinov, V. Voenlek, 2007. Atlas podneb eska (Climatic Atlas of Czechia) Finland Carter T.R. 1998. Changes in the thermal growing season in Nordic countries during the past century and prospects for the future. Agricultural and Food Science in Finland 7: 161-179. Carter T.R. and M. Granskog, 2001. Climatic conditions and constraints in the Nordic/Baltic region. In: J. Salonen, B. Bromand, L.N. Jrgensen (Eds). Crop Production Conditions in the Northern European Region with Special Reference to Crop Protection. DIAS report Plant Production no. 59, Danish Institute of Agricultural Sciences, Foulum, Denmark, pp. 16-47. Carter T. and R. Saarikko, 1995. Effects on spring wheat and spring barley in Finland. In: P.A. Harrison, R.E. Butterfield and T.E. Downing (eds.). Climate Change and Agriculture in Europe: Assessment of Impacts and Adaptation. Research Report No. 9, Environmental
40
Change Unit, University of Oxford, pp. 319-330. Carter T.R. and R.A. Saarikko, 1996. Estimating regional crop potential in Finland under a changing climate. Agric. For. Meteorol. 79: 301-313. Fronzek S. and T.R. Carter, 2007. Assessing uncertainties in climate change impacts on resource potential for Europe based on projections from RCMs and GCMs. Climatic Change, 81, 357371. Jansson P.E. and L. Karlberg, 2001. Coupled heat and mass transfer model for soil-plantatmosphere systems. Royal Institute of Technology, Dept of Civil and Environmental Engineering, Stockholm, p. 321. Saarikko R.A. and T.R. Carter, 1996a. Phenological development in spring cereals - response to temperature and photoperiod under northern conditions. Eur. J. Agron. 5: 59-70. Saarikko R.A. and T.R. Carter, 1996b. Estimating regional spring wheat development and suitability in Finland under climatic warming. Climate Research 7: 243-252. Venlinen S., T. Salo, C. Fortelius, 2005. The use of numerical weather forecast model predictions as a source of data for irrigation modelling. Meteorol. Appl., 12(4), 307318. Venlinen A. and M. Heikinheimo, 2002. Meteorological data for agricultural applications. Physics and Chemistry of the Earth. Vol. 27/23-24, 1045 - 1050. France Boone A., F. Habets, J. Noilhan, D. Clark, P. Dirmeyer, S. Fox, Y. Gusev, I. Haddeland, R. Koster, D. Lohmann, S. Mahanama, K. Mitchell, O. Nasonova, G.Y. Niu, A. Pitman, J. Polcher, A.B. Shmakin, K. Tanaka, B. Van den Hurk, S. Verant, D. Verseghy, P. Viterbo, 2004. The Rhone-Aggregation Land Surface Scheme Intercomparison Project: An Overview. J. Climate, 17, 187-208. Habets F., P. LeMoigne, J. Noilhan, 2003. On the utility operational precipitation forecasts to served as input for streamflow forecasting. J.Hydrology, 293 270-288. Huglin P., 1978. Nouveau mode d'valuation des possibilits hliothermique d'un milieu viticole. C.R. Acad. Agric., 1117-1126 Huglin P., 1986. Biologe et ecologie de la vigne. Ed. Payot Lausanne, Pars, Jones, G.V. and Davis, R.E. Riou C., 1998. Facteurs explicatifs des critres de qualit de la recolte. Application au rseau modles qualit en viticulture. Porgrs Agricole et Viticole (Montpellier), 115, 11, 247-252. Rousset F., F. Habets, E. Gomez, P. Le Moigne, S. Morel, J. Noilhan, E. Ledoux, 2004. Hydrometeorological modeling of the Seine basin using the SAFRAN-ISBAMODCOU system. J. Geophys. Res., 109, D14105 Germany Chmielewski F.M., K. Blmel, Y. Henniges, A. Mller, 2007. Klimawandel und Obstbau in Deutschland (KliO) Projektspezifischer Zwischenbericht der Humboldt Universitt zu Berlin. Available as pdf at: http://www.agrar.hu-berlin.de/struktur/institute/pfb/ struktur/agrarmet/forschung/fp/KliO_html Chmielewski F.M., Th. Rtzer, 2000. Annual and spatial variability of the beginning of growing season in Europe in relation to air temperature changes. Agrarmeteorologische Schriften 8, Humboldt University Berlin. Available as pdf at: http://www.agrar.huberlin.de/struktur/institute/pfb/struktur/agrarmet/service/sr Chmielewski F.M., A. Mller, E. Bruns, 2002. Climate changes and trends in phenology of fruit trees and field crops in germany, 1961-2000. Agrarmeteorologische Schriften 10 Humboldt University Berlin. Available as pdf at: http://www.agrar.hu-
41
berlin.de/struktur/institute/pfb/struktur/agrarmet/service/sr Chmielewski F.M., K. Blmel, Y. Henniges, A. Mller, 2007. Klimawandel und Obstbau in Deutschland (KliO) Projektspezifischer Zwischenbericht der Humboldt Universitt zu Berlin. Available as pdf at: http://www.agrar.hu-berlin.de/struktur/institute/pfb /struktur/agrarmet/forschung/fp/KliO_html Chmielewski, F.M., 2003. Phenology and Agriculture. Agrarmeteorologische Schriften 12, Humboldt University Berlin. Available as pdf at: http://www.agrar.huberlin.de/struktur/institute/pfb/struktur/agrarmet/service/sr Flrke M., J. Alcamo, 2007. Assessment of global scale water stress indicators In: Lozn J., H. Gral, P. Hupfer, L. Menzel, C.D. Schnwiese,. (Eds.) 2007. Global change: Enough water for all? Wissenschaftliche Auswertungen und GEO-Verlag, Hamburg, 200-203 Friesland & Lpmeier, 2006. The activities of the German Weather Service (DWD) in the field of agroclimatology. Meteorological applications 13, Supplement 1, 61-67. Kropp J.P., 2007. Extreme Temperatures over Europe: Expectations and Challenges for the 21st Century. In: Integrating Natural Disasters and Risk Management, EU-MEDIN, Springer, Berlin, accepted. Rachner M., G. Schneider, H. Matthus, 2000. Die mittleren klimatologischen Bedingungen in Deutschland (Teil IV). Das Andauerverhalten der Schneedecke. DWD Klimastatusbericht 2000. http://www.dwd.de/de/FundE/Klima/KLIS/prod/KSB/ksb00/klimbed.pdf Stock M. (ed.) 2005. KLARA: Klimawandel Auswirkungen Risiken Anpassung. PIK Report 99, Potsdam. Greece Dalezios N.R., Z.G. Papazafiriou, D.M. Papamichail, T.S. Karacostas, 1991. Drought assessment for the potential of precipitation enhancement in Northern Greece. Theor. Appl. Climatol., Vol. 44, pp. 75-88. Dalezios N.R. and N.K. Papamanolis, 1991. Objective Assessment of Instability Indices for Operational Hail Forecasting in Greece. "Meteorology and Atmospheric Physics", 45, 87-100. Dalezios N.R. and E.E. Lavrediadou, 1995. Features of frost- affected areas from digital METEOSAT IR images. Advances in Space Research, 15: 123-126. Dalezios N.R., A. Loukas, L. Vasiliades, E. Liakopoulos, 2000. Severity-duration-frequency analysis of droughts and wet periods in Greece. Hydrological Sciences, 45(5), pp. 751769. Dalezios N.R., A. Loukas, L. Vasiliades, E. Liakopoulos, 2000. Severity-duration-frequency analysis of droughts and wet periods in Greece. Hydrological Sciences, 45(5), pp. 751769. Dalezios N.R., A. Loukas, D. Bampzelis, 2002. Assessment of NDVI and Agrometeorological Indices for Major Crops in Central Greece. Physics and Chemistry of the Earth.Vol. 27, No. 23-24, 1025-1029. Domenikiotis C., M. Spiliotopoulos, E. Kanellou, N.R. Dalezios, 2004. Classification of NOAA/AVHRR images for mapping of frost affected areas in Thessaly, central Greece. Proceedings of the International Symposium in Geographical Information Systems and Remote Sensing: Environmental Applications, Volos, Greece, 7-9 November 2003, pp. 25-31. Gouvas, M. and N. Sakellariou, 2006. On the Estimation of the Monthly, Annual Average Number of Days with Snow and Snowcover in Greece. 8th Conference on Meteorology Climatology and Atmospheric Physics, Athens, 24-26 May 2006, Volume of Abstracts, 19p.
42
Katsiabani K., A. Mavrakis, K. Kartalis, G. Theocharatos, 2006. Estimation of the Temporal Evolution of Drought Episodes in Greece with the use of the SPI Meteorological Drought Index, 8th Conference on Meteorology Climatology and Atmospheric Physics, Athens, 24-26 May 2006, Volume of Abstracts, 42p. Karakostas T.S.2002. The Evaluation of the Greek National Hail Suppression Project. Proceedings of 6th Hellenic conference in Meteorology Climatology and Atmospheric Physics, Vol 2, Ioannina, 25-28 September 2002, pp. 614-620. Hungary Bihari Z. and S. Szalai, 2007. A parlagf pollen lgkri koncentrcijnak becslse meteorolgiai tnyezkbl, (Estimation of atmospheric concentration of ragweed (Ambrosia elatior) Erd s klma V. ktet, pp. 75-82 Nmeth A., 2007. DEM based method for the determination of potential frost-risk territories. In: M. Kappas, Ch. Kleinn, B. Sloboda [eds.]: Global Change Issues in Developing and Emerging Countries. Universittsdrucke Gttingen; pp. 393-400. (ISBN-13: 978-3938616-93-2) Plfai I., 2004. Belvizek s aszlyok Magyarorszgon, (Inland inundation and drought in Hungary) Hidrolgiai tanulmnyok. Budapest Pk Z. and L. Helyes, 2004. The effect of daily temperature on truss flowering rate of tomato.Journal of the Science Food and Agriculture 84: (13) pp. 1671-1674.IF: 0.871 Szalai S. and S. Bella, 2004. A 2003. vi Aszly, (Drought in 2003), Vzgyi Kzlemnyek,(Hidraulic Engineering) LXXXVI. vfolyam, 2004. vi 1-2 fzet, szerk. (editor) Szlvik Lajos, Budapest, pp.197-216 Szalai S., S. Bella, A. Nemeth, 2007. Drought sensitivity research in Hungary and influence of climate change on drought sensitivity, pp.179-188, Spatial Interpolation for climate data - the use of GIS in Climatology and meteorology, Edited by Hartwig Dobesch, Pierre Dumolard and Izabela Dyras, 2007, ISTE ltd., London, UK, 284pp, ISBN 978-1905209-70-5 Italy Ferrari R., M. Pasqui, L. Bottai, S. Esposito, E. Di Giuseppe, 2005. Assessment of soil erosion estimate based on a high temporal resolution rainfall dataset", EMS05-A00285, European Meteorological Society 2005 Marletto V., F. Zinoni, L. Botarelli, C. Alessandrini, 2005. Il monitoraggio della siccit in Emilia-Romagna (extended abstract). RIAM 9: 116-117. Quaderno dei riassunti Convegno AIAM Vasto Caramanico 3-5/5/2005. (Drought monitoring in EmiliaRomagna). Marletto V., F. Zinoni, L. Botarelli, C. Alessandrini, 2005. Studio dei fenomeni siccitosi in emilia-romagna con il modello di bilancio idrico criteria (extended abstract). RIAM 9: 32-33. Quaderno dei riassunti Convegno AIAM Vasto/Caramanico 3-5/5/2005. (Drought studies with the Criteria water balance model). Tomozeiu R., C. Cacciamani, V. Pavan, A. Morgillo, A. Busuioc, 2007. Climate change scenarios for surface temperature in Emilia-Romagna (Italy) obtained using statistical downscaling models. Theoretical and Applied Climatology. DOI 10.1007/s00704-0060275-z Tomozeiu R., V. Pavan, C. Cacciamani, M. Amici, 2006. Observed temperature changes in Emilia-Romagna: mean values and extremes. Climate Research, 31, 217-225. Pavan V., R. Tomozeiu, C. Cacciamani, M. Di Lorenzo, 2007. Daily precipitation observations over Emilia-Romagna: mean values and extremes, Submitted to Int.J.of Clim.
43
Cacciamani C., R. Tomozeiu, V. Pavan, S. Tibaldi, 2007. Evidenze di cambiamenti climatici dalla scala globale alla scala regionale. Possibili scenari futuri sul territorio della regione Emilia-Romagna, ARPA-Rivista ( climate change evidences from global to regional scale). Perini L. et al., 2004. Atlante agroclimatico agroclimatologia, pedologia, fenologia del territorio italiano. UCEA - Roma. ISBN 88-901472-0-2 (Agroclimatic atlas). Perini L., 2006. Caratteristiche agrometeorologiche del territorio italiano e possibili scenari di cambiamento climatico. CRA-UCEA Progetto di ricerca Climagri, Risultati Conclusivi, 59-78, ISBN 88-901472-6-1. Zinoni F., G. Antolini, T. Campisi, V. Marletto, F. Rossi, 2002. Characterisation of EmiliaRomagna region in relation with late frost risk. Physics and Chemistry of the Earth, 27:1091-1101 Norway rsvoll K., 1978. Studies on factors causing winter damage in Norwegian grasslands, with special reference to snow mould fungi, Doctorial Thesis( Dr. Agric), NLH, s, Norway (Compillation of eight papers in scientific journals). Baadshaug,O.E., and A.O. Skjelvg, 1986. Eng og beitedyrking III. Overvintring hos eng- og beitevekster (Cultivation of meadows and pastures. III. Overwinering of pasture species). Landbruksbokhandelen s ( written in Norwegian) Harstad M., 1997. Fenologisk utvikling hos timotei (Phenological development of timothy), Thesis of Master Degree, Institutt for plantefag, Agricultural University of Norway, 57 pp. Appendices Harstad M., 1997. Fenologisk utvikling hos timotei (Phenological development for timothy). Mgne Hrstad, Thesis of Master Degree, Institutt for plantefag, NLH, 57 pp, Appendices. Skjelvg A.O., 1987. Frste bermarksdag om vren ( First day of snowless ground in spring in Aust-Agder), Norsk landbruksforsknng 1 215 223, ( written in Norwegian) Skjelvg A.O., 1986. Utrekning av frste sdag ved vrobservasjonar.(Calculation of first day of sowing using weather observations) Forskning og forsk i landbruket 37: 295-301. Poland Demidowicz G., A. Doroszewski, T. Grski, 1996. Wpyw niedoboru opadw na straty w produkcji ziemniaka i buraka. (Effects of precipitation deficit on losses in the production of potatoes and sugar beets) Zesz. Post. Nauk Rol. 438, 43-52. Demidowicz G., A. Doroszewski, T. Grski, 1997. Metodyka szacunku strat w produkcji rolinnej powodowanych deficytem opadw. (A method for evaluation of crop yield losses due to the precipitation shortage) Rocz. AR Pozn. CCXCI, Melior. In. rod. 17, 233-243. (impacts on yield). Doroszewski A., G. Demidowicz, T. Grski, 1997. Wpyw niedoboru opadw na straty w produkcji zb jarych w Polsce. (Effect of precipitation deficit on losses in the production of spring cereals in Poland) Rocz. AR Pozn. CCXCI, Melior. In. rod. 17, 223-231. (impacts on yield) Grski T., G. Demidowicz, T. Deputat, K. Grska, I. Marcinkowaska, W. Spoz-Pa, A. Krakowiak, 1994. Agrometeorological quantification of agricultural year in Poland. Zesz. Post. Nauk Rol. 405, 81-87. Grski T., 2006. Zmiany warunkw agroklimatycznych i dugo okresu wegetacyjnego w ostatnim stuleciu. Changes in agroclimatic conditions and in the length of vegetative period in the last century). (w:) Dugotrwae przemiany krajobrazu Polski w wyniku
44
zmian klimatu i uytkowania ziemi. 65-77. IGBP-Global Change, Pozna (climatic water balance variability from April to October). Kalbarczyk E. and Kalbarczyk R., 2005. Identyfikacja okresw suszy atmosferycznej w okolicy Szczecina w latach 1963-2002. (Identification of atmospheric drought periods in the vicinity of Szczecin). Woda rod. Obsz. Wiej. 5 (14), 171-183. (for assessing drought frequency). Komiski C. and Michalska B., (eds.) 2001. Atlas of climatic risk to crop cultivation in Poland. Szczecin. Agricultural University of Szczecin. Komiski C., T. Grski, B. Michalska, 1990. Climatic atlas of elements and phenomena hazards to agriculture in Poland. IUNG, Puawy. abcki L. and B. Bk, 2004. Drought mapping in Poland using SPI. Proc. 21st ICID Conf. 15-19 May 2005, Frankfurt-Subice, Germany-Poland, CD-ROM. (spatial resolution and realization) Mager P., M. Kunicka, M. Kepiska-Kasprzak, R. Farad, 2000. Changes in the intensity and frequency of occurrence of droughts in Poland (1891-1995). Geographia Polonica, 73; 41-47. (for assessing drought frequency). Romania Mateescu E., I. Poiana, I.V. Pescaru, O. Oprea, D. Alexandru, 2004. Agrometeorological monitoring system decision-making instrument to prevent and mitigate the agricultural drought in Romania, Technical workshop on drought preparedness in the Balkans within the context of the UNCCD 25-26 October 2004, Poiana Brasov, Romania; Mateescu E., A. Marica, R. Opriescu, 2003. The frequency and areal of the agricultural drought in the south and south-eastern area of Romania, Workshop Drought Measures to mitigate the effects upon crops, dedicated to the celebration of 75 years since the setting up of the Romanian Institute of Agronomical Researches (ICAR) 1927-2002, AGRIS Publishing House Agricultural Journals editorial office, Bucharest, 2003, ISBN 973-8115-27-2, pg. 145-156; Mateescu E., N. Tanislav, V.V. Vtmanu, 2004. Impact of the drought conditions upon the wheat and maize crops in the Caracalului Plain, SITECH Publishing House, Craiova, ISBN 973-657-535-7, pg. 1 163; Mateescu E., R. Povar, R. Opriescu, 1999. Study of the agrometeorological stress parameters and their impact upon the winter wheat during the earing-flowering-beans filling period Scientific papers USAMV Bucharest, Series: A, Vol. XLI, 1999, ISSN 1221-5339, Bucharest, pg. 133-149. Palade A.R., E. Mateescu, R. Opriescu, 2002. The hydric potential available for the winter wheat and maize crop in the Moldova Plateau, Annual Scientific Papers Delivery Session NIMH, Bucharest, Papers compendium, 3-6 June 2002, ISBN 973-002675-9; Povar R., E. Mateescu, R. Opriescu, 2003. Assessment of the influence of the precipitation deficit in the autumn months upon the vegetation condition and the wheat crop in Vlcea district, Geography of Vlcea District Theory and practice, Volume III, Offsetcolor Publishing House, Rmnicu Vlcea, ISBN 973-86352-7-6, pg. 90-98. Tuinea P., V. Turcu, V. Adamiade, 2000. Drought phenomenon risk zonality on Romanian agricultural territory and its impact upon agricultural yields. Central and Eastern European Workshop on Drought Mitigation, Budapest-Felsogod, Hungary, 12-15 April, 2000.
45
Serbia Arseni I., D.T. Mihailovi, B. Lali, 2005. An interactive software for climatological characterisation of Vojvodina province, Proceedings, International Conference on Sustainable Agriculture and European Integration Processes, 19-24 September 2004, Contemporary Agriculture, 54, 1-2, 335-339 GPAB, 2007. Guidelines for Preparing of regular Agrometeorological Bulletins, reviews and analyses, Republic Hydrometeorological Service of Serbia, internal technical paper, 6th edition, pp. 50, 2007, Belgrade, Serbia. Kostadinov S. and Spasov P., 2006. Indicators of desertification in Serbia, BALWOIS 2006, Conference on Water Observation and Information System for Decision Support, Ohrid, Macedonia, May 23-26. Lali B., D.T. Mihailovi, I. Arseni, 2002. Analisys of Lange's rain factor for climatic characterisation of Vojvodina province, Proceedings,Eco-conference 2002, Novi Sad, 25-28 September 2002, 69-75. Lali B., I. Koi, D.T. Mihailovi, 2005. Agrometeorological modelling powerful tool of modern agriculture, Proceedings, International Conference on Sustainable Agriculture and European Integration Processes, 19-24 September 2004, Contemporary Agriculture, 54, 1-2, 312-317. Lalic B., L. Pankovic, D.T. Mihailovic, 2005. Prediction of malting barley phenology based on air and soil temperature accumulation, Abstracts of Fifth Annual Meeting of the European Meteorological Society (EMS), September 12-16, Utrecht (The Netherlands), ISSN 1812-7053. Lali B., D.T. Mihailovi, S. Malinovi, 2004. Extreme temperatures in Vojvodina during 1948-2003. Period., Proceedings, Eco-conference 2004, 22-25 September 2004, Novi Sad, Serbia, 49-55. Mihailovi D.T., I. Arseni, B. Lali, D. Radlovi, I. Koi, 2001. Heat index for extremely high temperatures during july 2000 in Novi Sad. Scientific meeting organised by Matica Srpska Health of people in Vojvodina, Novi Sad, 17-18 January, Proceedings, 85-9l. Mihailovi D.T., 1997. Crop: Between soil and climate, Periodical of Institut for Field and Vegetable Crops, Novi Sad, 29, 63-71. Mihailovi D.T., and Djori E., 2005. Estimating the quantitative parameters of chaos from filtered and non-filtered air temperature. Proceedings, International Conference on Sustainable Agriculture and European Integration Processes, 19-24 September 2004, Novi Sad (Serbia), Contemporary Agriculture, 54, 1-2, 329-334. Slovakia Nejedlik P., 1999. Agroclimatic condition of the Cereals growing within East Slovak Lowland Region. In: Zbornik prac, zv. 42, SHMU. Bratislava 1999. ( In Slovak). Nejedlik P., 1999. Agroclimatic condition of the Cereals growing within East Slovak Lowland Region. In: Zbornik prac, zv. 42, SHMU. Bratislava 1999. ( In Slovak). Slovenia Kajfe-Bogataj L. and Sunik, A., 2002. Operativni agrometeoroloki modeli za izraun vodne bilance kmetijskih tal (Comparison of three operational irrigation models). Novi izzivi v poljedelstvu 2002. Str. 164-169. Knapi V., K. Beber, G. Seljak, V. kerlavaj, S. Tome, 1999. Using automatic meteorological stations for early pest warning and irrigation prediction. Lectures and Papers presented at the 4 th Slovenian Conference on Plant protection in Portoro, March 4-5, 1999,67-72.
46
Kurnik, B., Navodila za uporabo modela IRRFIB, model IRRFIB, podatki za uporabo v modelu. Sunik A. and Kurnik B., 2004. On-line irrigation support system in Slovenia. On-line agrometeorological application with decision support at the farm level. COST 718: Meteorological application to agriculture. DINA Research report no. 109: 113-121 Sunik A., I. Matajc, I. Kodri, 2006. Agrometeorological support of fruit production. Application in SW Slovenia. WMO CAgM/ET. Meteorological Appl. (Supplement):8186(2006). Sunik A., 2006a. Vodni primanjkljaj v Sloveniji in moni vplivi podnebnih sprememb. Magistrsko delo, Univerza v Ljubljani, Biotehnika fakulteta, Oddelek za agronomijo: 147 str (in slovene). Switzerland Bartelt P.B. and M. Lehning, 2002. A physical SNOWPACK model for Avalanche Warning Services. Part I: Numerical Model, /Cold Reg. Sci. Technol./, 35/3, 123-145. Calanca P , 2006. Climate change and drought occurrence in the Alpine region: how severe are becoming the extremes? Global Planetary Change, available online, doi:10.1016/j.gloplacha.2006.11.001. Calanca P., 2004. Interannual variability of summer mean soil moisture conditions in Switzerland during the twentieth century: A look using a stochastic soil moisture model. Water Resources Research, VOL. 40, W12502, doi:10.1029/2004WR003254. Calanca P., A. Roesch, K. Jasper, M. Wild, 2006. Global warming and the summertime evapotranspiration regime of the Alpine region. Climatic Change 79, 6578, DOI 10.1007/s10584-006-9103-9. Jasper K., P. Calanca, J. Fuhrer, 2006. Changes in summertime soil water patterns in complex terrain due to climatic change. Journal of Hydrology 327, 550 563 Jasper K., P. Calanca, D. Gyalistras, J. Fuhrer, 2004. Differential impacts of climate change on the hydrology of two alpine river basins. Climate Research, 26, 113-129. Lehning M., P.B. Bartelt, R.L. Brown, C. Fierz, P. Satyawali, 2002. A physical SNOWPACK model for the Swiss Avalanche Warning Services. Part III: Meteorological Boundary Conditions, Thin Layer Formation and Evaluation, /Cold Reg. Sci. Technol./, 35/3, 169-184. Lehning M., I. Vlksch, D. Gustafsson, T.A. Nguyen, M. Sthli, M. Zappa, 2006. ALPINE3D: A detailed model of mountain surface processes and its application to snow hydrology, /Hydrol. Processes,/ 20, 2111-2128. Riedo M. A., M. Grub Rosset, J. Fuhrer, 1998. A pasture simulation model for dry matter production, and fluxes of carbon, nitrogen, water and energy. Ecological Modelling 105: 14118 Torriani D., P. Calanca, S. Schmid, M. Beniston, J. Fuhrer, 2007. Potential effects of changes in mean climate and climate variability on the yield of winter and spring crops in Switzerland. Climate Research. In press.
47
Table 2.9. Reported country-related web links on operational used agroclimatic indices, pestdisease warning (see chapter 2.3) and agrometeorological forecasts in Europe Austria zamg.ac.at zamg.ac.at/phaenologie agrarwetter.at hagel.at landwirt.com/Wetter/Agrarwetter.html adcon.at proplantexpert.com/expert/index.jsp Croatia meteo.hr Czech Republic chmi.cz/meteo/CBKS/sbornik04/prispevky/MOZNY.pdf chmi.cz/meteo/CBKS/sucho01 chmi.cz/meteo/ok/dpp.html chmi.cz/meteo/ok/oba/obs/o.html chmi.cz/meteo/om/inform/se.html chmi.cz/meteo/ov/aladin/results/index.php srs.cz Denmark eucablight.org/EucaBlight.asp planteinfo.dk Finland climate.fmi.fi/data/halla.gif climate.fmi.fi/data/lsumma.gif climate.fmi.fi/data/lumi.gif climate.fmi.fi/data/sade5.gif ilmatieteenlaitos.fi/products/weather.html portal.mtt.fi/portal/page/portal/kasperit France acta.asso.fr/meteopro/meteopro.htm eaufrance.fr ecologie.gouv.fr/Situation-Hydrologique.html pv.agriculture.gouv.fr/srpv.htm Germany agrar.hu-berlin.de/struktur/institute/pfb/struktur/agrarmet/forschung/fp/KliO_htm agrar.hu-berlin.de/struktur/institute/pfb/struktur/agrarmet/service/sr agrarmet.de/Forschungsstelle/index.htm agrowetter.de/Agrarwetter/fbidx.htm agrowetter.de/Agrarwetter/GFI_en.htm
48
agrowetter.de/Agrarwetter/phyto.htm agrowetter.de/Agrarwetter/PNP_en.htm agrowetter.de/produkte/doku/Prognose/heutrocknung_doku.htm agrowetter.de/produkte/doku/Prognose/kornfeu_doku.htm agrowetter.de/produkte/doku/Prognose/mais_modell.html am.rlp.de/internet/global/startpage.nsf/start/Home_Am?OpenDocument dwd.de/de/FundE/Klima/KLIS/prod/KSB/ksb00/klimbed.pdf dwd.de/de/WundK/Warnungen/index.htm eumetsat.de lfl.bayern.de/agm/ ISIP.de unwetterzentrale.de/uwz/regen.html p7115.typo3server.info/index.php?475 Italy agrifish.jrc.it/marsstat/Crop_Yield_Forecasting/cgms.htm agrifish.jrc.it/marsstat/Crop_Yield_Forecasting/METAMP/04000002.HTM agrifish.jrc.it/Public/CGMS/doc/GridWeather.doc agrometeorologia.it arpa.emr.it/ia_siccita/index.htm?idlivello=120 arpa.emr.it/sim/?telerilevamento/innevamento ermesagricoltura.it/wcm/ermesagricoltura/fitosanitario/home_sezioni/home_fitosanitario.htm sias.regione.sicilia.it ucea.it/iphen Netherland lei.dlo.nl/nl/content/agri-monitor/pdf/Febr2007Oogstschadeverzekering.pdf topshare.wur.nl/cost725 Norway bioforsk.no Poland agrometeo.pl imgw.pl/wl/internet/zz/oddzialy/poznan/prognoza/prognoza.html imgw.pl/wl/internet/zz/pogoda/opady.html imgw.pl/wl/internet/zz/pogoda/snieg.html Romania agriplus.ro inmh.ro Slovakia galati.sk/vitis shmu.sk Slovenia dmcsee.org
49
Spain hispagua.cedex.es/documentacion/especiales/sequia/indicadores_sequia.htm mma.es mma.es/portal/secciones/acm/aguas_continent_zonas_asoc/ons/ Switzerland agrometeo.ch blw.admin.ch/agroscope fusaprog.ch meteodat.ch meteoschweiz.ch/web/en/services/agriculture/the_agricultural_industry_professional.html meteoschweiz.ch/web/en/services/product_overview/5dayforecasts.html meteoswiss.ch meteotest.ch phytopre.ch slf.ch slf.ch/avalanche/avalanche-en.html slf.ch/swiss-snow/snowinfo.html sopra.admin.ch
2.2.3
Communicative value and limitations of the agrometeorological indices and simulation models
During the eighties many agrometeorological simulation models and indices were set up and applied with the idea to automatically solve problems affecting important sectors of agriculture, from the irrigation to the pest and diseases control to the optimization of the yields. This approach was obviously fully unsuccessful and led to a lack of trust in these methods. Anyway, the experience was useful and in the last years, after a period of study and scientific efforts, the use of indices and models can be reconsidered as an important tool for the support to decisions in agriculture. Agrometeorological indices, in fact, are not a precise description of a specific environment but represent a good indication of the territory characteristics. On the other hand, models can supply the end-users with different information depending on their structure. In fact, empirical models usually have qualitative outputs that give an idea of what is happening in the field, such as the potential risk for a specific disease. Conversely, mechanistic models are able to give quantitative information concerning, for example, the actual severity of that disease. In any case, simulation models, as their name suggest, are a representation and a simplification of the different phenomena. Moreover, agrometeorological indices and simulation models require meteorological data as input for calculation so the quality of these data result to be a crucial point. In fact, before any further process, data need to be
50
checked both for the presence of possible gaps in the series and for the consistence of single data. As a consequence their output, both qualitative and quantitative, needs to be analysed, understood, interpreted and eventually corrected before making any decision. For these reasons indices and models should be used by trained users and the personal scientific knowledge and technical experience have to be taken in great consideration. In this way agrometeorological tools represent a valid support and a fundamental base for a correct management of a modern and efficient agricultural activity.
2.3 Application of process oriented models in Europe (Subgroup 2) 2.3.1 State of the Art Process oriented crop simulation models have been studied for more than 50 years. The three most important schools of development from Australia, the Netherlands and the United States include APSIM models (Asseng et al., 2000), models from the School of De Wit (Van Ittersum et al., 2003; Penning de Vries, 1989) and the DSSAT family of crop models (Jones et al. 2001, 2003). However, a mostly still unknown group of crop models in the western research community are the models developed in the former Sovjet Union (e.g. Sirotenko and Abashina, 1994), which were especially applied in the new EU-members countries in the past. These models are not described in this report, but an overview is given by Sirotenko (1983) and Baker (2001). Crop simulation models are based on physical plant processes. Here, daily effects of change in growing environment on plant growth and development are simulated. Phenology, photosynthesis, dry matter production, dry matter partitioning, with the aim to simulate the potential production, are the core processes of the model. Those aiming at crop-specific behaviour contain modules for phyllochron, branching pattern as well as potential flowers/grain filling sites. The water and nutrition limited environment is added by models of soil water balance with transpiration by crop, and nitrogen transformations in soil with remobilization within plant. Models of effects of weeds and pests are being developed and are available in the new generation of crop simulation models (Dadhwal 2003). Three categories of variables can be recognized in dynamic crop simulation models: state, rate and driving variables (Fig. 2.5). The state variables are quantities like biomass, amount of nitrogen in soil, plant, soil water content and can be measured at specific times. Driving variables, or forcing functions, describe the effect of the environment on the system at its limits, and their
51
values must be monitored continuously, for example meteorological variables. Each state variable is linked with rate variables and describe their rate of change at a certain instant as a result of specific processes. These variables represent for example flow of material or biomass between state variables. Their values depend on the state and driving variables according to rules, which are based on knowledge of the physical, chemical and biological processes during crop growth (Dadhwal, 2003).
Figure 2.5: Simplified scheme of a crop simulation model (Dorigo et al., 2007, adapted from Delcolle et al., 1992)
Figure 2.5 presents a crop simulation model in computational iterations and the time step of the model. At each iteration, vegetation state variables are updated and based on the input driving variables as well as the values of the state variables at the previous time step (Delcolle et al. 1992, Dorigo et al. 2007). Crop simulation models, depending on their complexity, need more or less various input data. The model DSSAT (Decision Support System for Agrotechnology Transfer) for example, needs information about weather, soil, management and genetics. Output is flowering and maturity data, yield, nitrogen and water balance, leaf area and harvest index etc. on local scale (Fig 2.6). As crop simulation models produce a point output, spatial data analysis techniques and geographical information system (GIS) can help to increase information of these outputs to a larger area (Gurif and Duke 2000).
52
Figure 2.6: Input- and output data of the crop simulation model DSSAT (Decision Support System for Agrotechnology Transfer)
2.3.2 Present use of process oriented models in Europe As shown in Table 2.10 and 2.11 and Figure 2.7 as a result of the COST734 and literature survey (e.g. Alexandrov, 2000) the most frequently used process oriented crop models for research or operational applications in Europe are DSSAT (CERES models) (Jones et al., 2003), WOFOST/SUCROS and STICS, however, with distinct differences between countries. WOFOST is the only model which is operationally integrated at the European level for the European crop yield prediction system, covering all countries (see also chapter of subgroup 4).
Figure 2.7: Reported crop model applications (operational and research, one count per crop and country)
53
Table 2.10: Reported operational applications of process-oriented models in Europe Summary
54
Table 2.11: Reported scientific applications of process oriented models in Europe Summary.
55
Clearly it can be seen that research applications dominate and that only few models are already applied operationally at the beginning of the 21st century. Most operational applications are reported in the group of pest and disease models (see country reports below). Often the number of national or European applications of the relevant models is related to established research institutions working on model developments, especially for operational applications. Examples are the STICS model developed by researcher on INRA in France, or WOFOST by Research Institutions in the Netherlands. Model applications, especially in research, are often related to a relatively small group of researchers or single persons in the different European countries, so the number of applications is often related to personal curricula and therefore often temporary. The main application of the crop models is clearly in climate change impact research on agriculture (e.g. Harrison et al. 1995; Downing et al., 2000), whereas the operational applications have the focus on crop yield forecasting. The applications however, in specific often include the assessment of the dependence of growth, development and yields of crops on limitations of soilwater regime; the assessments of crops development and related timing for crop management tasks; the assessment of yield response of crops by changes in production techniques such as fertilizing, cultivation, irrigation, plant protection etc. and early warnings or mitigation of damages from extreme meteorological phenomena and processes. Most crop simulation models in Europe (Fig. 2.8) are applied for annual crops, especially cereals and maize, reflecting the economically most important crops in Europe. Regionally, however, also permanent grassland, potatoes, sugar beet, oilseeds and others play an important role which results in specific model applications.
Figure 2.8: Reported crop model applications (operational and research, one count per model and country)
56
In the following all reported model applications (with the possibility to be incomplete) in European countries are described, including also statisticalempirical approaches and pest-disease models . Dynamic crop yield models are not yet used operationally in Austria. However, currently the AMBETI agrometeorological model (for crop water demand and canopy climate) is applied for operational daily (and hourly) forecasts through the Austrian Weather Service (carried out by German Weather Service). The main users are farmers and agricultural extension services. The FAO-GRAM grassland yield model (Kromp-Kolb et al., 2005) is a proecess oriented model for crop water balance combined with a statistical approach for grassland yield for detection of drought damages on a daily base and on field level scale and still in a testing phase, provided and used by HBLFA Gumpenstein. In research, dynamic crop yield models such as from the DSSAT model group were used in Austria for assessing impacts of climate change and variability on crop yields (especially winter wheat and spring barley, maize and soybean) and crop field water balance (Cajic et al., 2000; Alexandrov et al., 2001; Alexandrov and Eitzinger, 2001; Eitzinger et al., 2001; Alexandrov et al., 2002; Eitzinger et al., 2003, 2004). Also SIMWASER (Stenitzer, 2004; Stenitzer and Murer, 2003; Stenitzer et al., 2007), WOFOST (Eitzinger et al., 2000), PERUN were used for similar tasks. EPIC was especially used for determination of nitrate leaching and crop management assessments for various crops and crop rotations (Cepuder et al., 2002; Mller et al., 2006; Schmid et al., 2006). Several pest and disease models are used in Austria operationally, many of them within the decision support system PROPLANT (http://www.proplantexpert.com/expert/index.jsp) applied e.g. by the Austrian agricultural extension service (http://www.agrar-net.at/). An overview of most important operationally applied pest and disease models used in Austria (and other countries) can be found at www.zepp.info, including 26 models for cereals, orchards, vegetables and other crops (for more details see under Germany). In Bulgaria WOFOST has been used in many research studies and was adapted for three types of crops winter wheat, maize and soybean for operational use (Kazandiev and Georgieva, 2007; 2005a, 2005b). It has been applied as a tool for the analysis of yield risk and interannual yield variability, of yield variability over soil types, or over a range of agrohydrological conditions, of relative importance of growth determining factors, of sowing strategies, effects of climate change and critical periods for use of agricultural machinery. The model has also been used for predictive purposes, in quantitative land evaluation, such as regional assessments of crop yield
57
potential in the form of maximum yield levels, estimation of benefits from irrigation and from fertilizer use, detection of adverse growing conditions by simulation-monitoring the agricultural season, and regional yield forecasts. For scientific applications CERES-Wheat (Alexandrov, 1997, 1998, 1999, 2000, 2003; Alexandrov and Hoogenboom, 2000; Alexandrov and Eitzinger, 2005) and CERES-Maize (Alexandrov and Slavov, 1998; Alexandrov, 2001; Alexandrov and Eitzinger, 2000) were validated for the assessment of climate change impacts (especially with focus on drought) on yield, water balance, adaptation options, and regional and field based investigations ). Other models which were used for scientific applications are WEATHER-YIELD (Russian model) for simulation of yield and water balance at the regional and field based level and ROIMPEL (Audsley at al., 2006) for regional and national assessments on yield under changing climate. Also multifactor statistical models were used for similar purposes. In Croatia CERES-Maize is used to assess climate change impact on maize yield at the regional scale for scientific investigations in the frame of EUproject AGRIDEMA (Vucetic, 2006a, 2006b). To determinate major changes in maize growth, development and yield in the important agricultural area over the period 1949-2004 and projection at the end of the 21st century, CERESMaize has been applied as a tool for the assessment of maize phenological stages, biomass, maximum leaf area index (LAI) and grain yield for the present and future climate. The linear trend anallysis and Mann-Kendall and Spearman rank statistics have been applied on the results representing the present climate. The input wheather series representing the change climate for the CERES-Maize model was created by the stochastic weather generator Met&Roll for different climate change scenarios during the 21st century. Crop desease models are appleid follows: Operational use of process oriented models is done for for Tomato late blight (Phytophthora infestans mont. de bary) where a mathematical forecasting model (Cvjetkovi et al., 2006) is used which is based on "negative prognosis" (combination of PhytProg, NegFry and Pro. Phy models). Other applications are for Potato late blight (Phytophthora infestans mont. de bary) (ubi and Cvjetkovi, 2001) and Gray mould (Botrytis cinerea) where the prognostic model BOTMAN (Milievi et al., 2006) is used for Strawberry Plantations. Fire blight (Erwiania amylovora) is predicted by a forecasting system using Internet (Cvjetkovi et al., 1999; Hitrec et al., 2000). The Croatian Agricultural Extension Institute (CAEI) uses two types of agroclimatic stations: Agra and Methos for forecasting the occurrence and spreading of causative agents of plant diseases (Biak, 2006): Fire blight (Erwinia amylovora) is used Billing's Integrated System (BIS, Lecomte et al.,
58
1998); Apple scab (Venturia inaequalis) is used Mills-LaPlante model by temperature sums for predicting the beginning of spores flight (Creemers et al., 1996); Downy mildew (Plasmopara viticola) is used Mller forecasting system; Potato blight (Phytophthora infestans) is used mathematical forecasting model which is based on "negative prognosis" (Ullrich and Schrdter, 1966 - combination of PhytProg and Smitht models). In the Czech Republic several research applications which crop models were carried out, which includes for wheat the CERES-Wheat model (effect on yield, water balance, adaptation options (Trnka et al., 2004a, Eitzinger et al., 2004; Eitzinger et al., 2003)), WOFOST (effect on yield, water balance (Dubrovsk et al., 2004), STICS (effect on yield, water balance and phenology (Trnka et al., 2006). For Maize CERES-Maize was applied (effect on yield, water balance, adaptation options (alud and Dubrovsk, 2002)) and for spring barley CERES-Barley (effect on yield, water balance and adaptation options (Trnka et al., 2004b, Trnka et al., 2006). The model SHOOTGRO was tested for wheat (Zalud et al., 2003). AGRICLIM is a new developed software for the assessment of agricultural production potential, including several agrometeorological algorithms and a phenological model (Trnka et al., 2008). Crop pest models as research tools were applied for the European Corn Borer, namely ECAMON (effect of climate change on development of ECB, likelyhood of the shift from monovoltine to bivoltine populations in the Czech Republic (e.g. Trnka et al., 2007)) and CLIMEX (effect of climate on the European corn borer climate niche (Trnka et al., 2006). For the Colorado Potato Beetle CLIMEX was applied to assess the effect of climate on the CPB climate niche (Trnka et al., 2006). The Newhall simulation model (NSM) was used to determine the soil climate regimes (both hydric and thermic ones) for the present and changed climate. New version was developed and tested by Kapler et al. (2006) and Trnka et al. (2006). Since 2006 a new operational pest and disease monitoring system (http://www.srs.cz/srsmapa/) covers over 30 agricultural and horticulture species and almost 100 pests and dieseases. It is based on the set of degree day models and combination of the statistical and process based pest/disease models in combination with observed weather data. In Denmark a simlified grassland yield model on a daily basis (Segaard et al., 2005) and an irrigation scheduling programm (Plauborg and Olesen, 1991; Thysen and Detlefsen, 2006)via internet (www.planteinfo.dk) are applied operationally.
59
Models used for research studies are CLIMCROP for wheat, barley, grass, potato on effects on yield and water balance and to assess adaptation options in terms of timing of field operations (Olesen et al., 2000; Olesen, 2005). DAISY was used for wheat, barley, oilseed rape simulating effects on yield, water balance and nitrogen cycling in order to catch crops to reduce N leaching (Olesen et al., 2004, 2007). Planned operational use of process-oriented models in Finland is reported for the development of fiber content in silage grass, for potato and barley yield and quality monitoring and for certain pest and diseases (https://portal.mtt.fi/portal/page/portal/kasperit). Pest and disease models and warning systems have been published during the NJF congress 2007 in Denmark, (please see the NJF- home page http://www.njf2007.dk/). Potato late blight model is based on Nordic-Baltic collaboration and use of a special blight model published (http://www.eucablight.org/EucaBlight.asp). Delia brassicae and Psila rosae development have been forecasted based on simple models using effective heat accumulation as the input. CLIMEX has been and will be used for predicting the direct effect of abiotic factors on establishment, survival and development of pest populations. For research applications CERES-Wheat is used for for yield estimation under changing climate and was tested for several sites and applied over a 10 km grid nationally in Finland (Carter et al., 2000; Saarikko, 2000). Other models used for research are AFRC-Wheat (calibration and performance testing for Finland (Laurila, 1995)); CropWatN for yield estimation over a 10 km Finnish grid for changing climatic conditions (Karvonen and Kleemola, 1995; Carter et al., 1996). Research studies with pest and diseases models were carried out for potato blight (Phytophthora infestans model; Finland (Kaukoranta, 1996; Carter et al., 1996)), potato cyst nematode (Globodera rostochiensis model; Finland (Carter et al., 1996)), columbia root-knot nematode (Meloidogyne chitwoodi model; Finland (Tiilikkala et al., 1995; Carter et al., 1996)) and European corn borer (Ostrinia nubilalis model; Europe (Porter et al., 1991)). Climex is the main tool to be used for studies on climate change and predicting pest risks for 2020, 2050, 2080. A combination of CLIMEX and habitat models will be developed by Sini Olander at the University of Helsinki. In France an experimental operational monitoring is proposed from 2003 by the INRA unit Agroclim in Avignon, for the monitoring of 7 crops (wheat, maize, irrigated maize, rapeseed, pea, sunflower, permanent pasture) for 10 representative sites, using the STICS model for a set of given soil and crop system parameters (http://www.avignon.inra.fr/veille_agroclimatique/).
60
For grassland the ISOP monitoring system is operated by Meteo-France for the Ministry of agriculture and fisheries (AGRESTE system, driven by the statistical service SCEES) in cooperation with INRA, using the STICS model (Ruget, 2006; Ruget et al., 2006) adapted for pastures on a daily time step and a 20km grid for assessment of general conditions for forage resources during the growing season and estimation of production losses at the end. The main providers are Meteo-France, SCEES, INRA and the main user is MAP (ministry of agriculture and fisheries). More information can be obtained at http://www.agreste.agriculture.gouv.fr/. Pest and disease algorithms and models for operational use have been built-up by the combined efforts of services from the Ministry of Agriculture (PV, for the Service de la Protection des Vgtaux) and of professional institutes (Arvalis, CTIFL, ITV), with the collaboration of research (INRA). With this services several pest and disease models for several crops are applied including winter wheat (Spirouil : Puccinia triticina; Presept : Septoria; Top: Pseudocercosporella; Yello: Puccinia striiformis), maize (pyraausta nubilalis, sesamia), sunflower (phomopsis), rapeseed (sclerotinia), potato (phytophtora infestans), vine (EPI 89, Milvit, Milstop: Plasmapora viticola, other for : eudemis, conchylis, oidium), apple (cydia pomonella, Clean Arbo: venturia inequalis) and plums (grapholihta funebrana) and the service is available online (http://pv.agriculture.gouv.fr/srpv.htm). The advices are operationally diffused by the regional services of PV (SRPV), corresponding to the 21 administrative regions. Some agrometeorologica regional institutions are also disseminating this information, together with irrigation bulletins and spring frost warnings. Additionnally, a few privates firms also try to start a market with these forecasts, and the Meteo-Pro software as an output of automated stations. In research, the STICS model is widely used at INRA and other organizations for scientific investigations (Brisson, 2004; Brisson et al., 2003; Brisson et al., 2004, 2006). More information is available at www.avignon.inra.fr/stics. STICS is currently further improved for permanent grassland and vine (Garca de Cortzar Atauri, 2006; Seguin and Garca de Cortzar, 2004) and was also adapted for intercropping (Brisson et al., 2004, 2006). CROPSYST (Stockle et al., 1997) and SUCROS were used in some studies as well (e.g. Drr et al., 1999). In Germany several process oriented models are used already operationally. Pest and disease models for several crops are used, for example, within the decision support system PROPLANT by e.g. extension services (http://www.proplantexpert.com/expert/index.jsp). Available models can be found for example at www.zepp.info. For Wheat: the AMBETI soil-plant-
61
atmosphere model (Braden, 1995; Friesland and Lpmeier, 2006) is applied by the agrometeorological department of the German Weather Service (DWD; www.agrarmet.de/Forschungsstelle/index.htm), for services of farmers and advisors. However, it does not contain a dynamic crop growth module. Pest and disease models for wheat are SIMCERC (www.ISIP.de; http://www.zepp.info/) for prediction of Pseudocercosporella herpotrichoides in winter wheat and winter rye and SIMONTO (www.ISIP.de; http://www.zepp.info/) for phenological development of winter cereals. PUCREC (www.ISIP.de; http://www.zepp.info/) is used for Puccinia recondite prediction of first occurrence and infection pressure and SIMSEPT (www.ISIP.de; http://www.zepp.info/) for prediction of Septoria tritici and Septoria nodorum pressure. LAUS (Friesland and Lpmeier, 2006; www.agrarmet.de/Forschungsstelle/index.htm), a population model (number of aphids per ear and flag leaf) is used for cereal aphids in wheat (Sitobion avenae). For Maize: AMBETI is used for hourly and national wide prediction of water demand and canopy climate, which is also used as input for other models. AMBETI is also in use for grasslands, oilseed (Rape), potato and sugar beet and many other crops (34 crops). SKLEROPRO, a decision support model for Sklerotinia sclerotiorum infection pressure for rape is provided by (www.ISIP.de; http://www.zepp.info/). For potato: SYMPHYT1 and -3, SIMBLIGHT for estimation of Phytophthora infestans first occurrence and infection and SIMLEP a Population dynamic of potato beetle (Leptinotarsa decemlineata) is provided as well by (www.ISIP.de; http://www.zepp.info/). CERCBET1 and -3 (www.ISIP.de; http://www.zepp.info/) is used for sugar beet to estimate begin of infection, infection density of Cercosporella beticola. For scientific purposes following process oriented models were used and applied for : Wheat: AGROSIM (Mirschel et al., 2001, 2005): effect on yield and water balance, field based; APSIM (Wessolek and Asseng, 2006): effect on yield and water balance, regional; DAYCENT (Schaldach and Alcamo, 2006; Stehfest et al., 2007; WZU, 2005)): effect on yield and carbon balance, regional; FASSET (Berntsen et al.(2006): effect on yield and water balance, field based; HERMES (Kersebaum et al., 2005; Kersebaum 2007): effect on yield, nitrogen leaching, adaptation, fertilization, (sub)field based and regional; SWIM (EPIC) (Krysanova et al., 1999; Gerstengabe et al., 2003; Stock, 2005): effect on yield and water balance, regional assessment. Maize: HERMES (Herrmann et al., 2004, 2005): effect on yield, N-uptake, Nleaching, field based; SWIM (EPIC), (Stock, 2005): effect on yield and water
62
balance, regional; DAYCENT (Stehfest et al., 2007; WZU, 2005): effect on yield and carbon balance, regional. Grassland: DAYCENT (Stehfest et al., 2007): effect on yield and carbon balance, regional. Oilseed (Rape): DAYCENT (Stehfest et al., 2007): effect on yield and carbon balance, regional. Sugar beet: AGROSIM (Mirschel et al., 2001, 2002): effect on yield and water balance, field based; DAYCENT (Stehfest et al., 2007): effect on yield and carbon balance, regional; HERMES (Kersebaum, 2007): effect on yield, nitrogen leaching, adaptation, (sub) field based. Apple: ASCHORF (Wittich, 1998; used in COST 718) was applied for apple scab (days with infection, infection severity). Grape: N-VINO (Nendel and Kersebaum, 2004a,b; 2005) was applied for nitrogen dynamics in vineyards (crop growth, water and N-dynamics). The phenological model of Hoppmann & Berkelmann-Lhnertz (2000) was applied to estimate the phenology of wine in Germany (FAG, 2005). The model system EXPERT-N is a modular simulation tool used for nutrient, organic and water balance analysis of various crops, incorporating basically the three crop models CERES, SPASS and SUCROS and several methodological approaches (Priesack et al. 2007). In Greece there is still no operational use of crop models. For scientific and research purposes the following models are used such as the crop-growth simulation model for calculating biomass production potentials of sorghum within the beaver (Biomass Economic Appraisal & eValuation ExpeRt BEAVER) expert system environment (Danalatos, 1997), Agrometeorological Simulation Using PERO Model for Grape Vine Downy Mildew (Dalezios et al., 2000), WOFOST model for maize, cotton and wheat (Danalatos and Sgouras, 1987), Pedotransfer functions in predicting ground water recharge at regional scale (Kosmas et al., 1996), CERES-wheat and CERES-maize (Danalatos et al., 1998). Lastly, Danalatos et al. (1994) investigated the change in the specific leaf area of maize grown under Mediterranean conditions. In Hungary there are still no crop models in operational use. CERES-Wheat and CERES-Maize are however often used for scientific investigations. Recently a new software tool (4 M) was developed for Hungary, incorporating DSSAT models (Fodor et al., 2003; Fodor and Kovacs, 2003; Fodor, 2006). In Ireland models have been used primarily for scientific investigation of the impact of climate change. There is only one model used for routine
63
management and that is the Johnstown Castle Grass Model (JTC grass model) (Brereton et al., 1996). This is used to estimate current and forecast grass growth and the results are published in the Irish Farmers Journal. Arable, horticulture and vegetable crops represent such a small proportion of agricultural area and production in Ireland (e.g. cereals account for only 4.8% of gross output, Anon, 2008), that little effort has been directed towards using models for operational purposes. The Irish Environmental Protection Agency has funded work on quantifying the impact of climate change on Irish agriculture. Initial efforts focused on assessing impacts on yield for key indicator crops: grass as the main agricultural crop using JTC grass model; barely as a key feed and food grain using CERES-Barley (in DSSAT); maize as a marginal crop with potential for introduction into Ireland using CERESMaize (in DSSAT); potato because of its traditional cultural position in Ireland using SUBSTOR (in DSSAT) and soybean as a very marginal exotic which could act as an indicator for the future using CROPGRO (in DSSAT). Results from this work can be found in Holden and Brereton (2002; 2003a; 2003b), Holden et al (2003) and Holden et al (2004). More recently work has focused on impact and adaptation at the system level, with the focus on nutrient and water use for potato and barley (using DSSAT, Holden and Brereton, 2006) and on dairy production systems (using the dynamic dairy system model Dairy_sim, which was specifically developed for the purpose, Fitzgerald et al., 2005). Results on dairy system impact and adaptation are currently in press (Fitzgerald et al., 2008a; 2008b; Holden et al., 2008). Current work being conducted at NUI-Maynooth is starting to use GIS analysis to investigate spatial aspects of climate change and pest / disease and development of energy crops, and the Community Climate Change Consortium for Ireland (C4I) project will also be developing further applications for agricultural modeling (www.C4I.ie). In Italy many research activities have been carried out in the last years on several aspects on the main crops, based both on direct and simulated responses. Important crop models applied in research are DSSAT models (CERES) (e.g. Mastrorilli et al., 2003; Castrignano et al., 2002) and CROPSYST (Donatelli et al., 1997; Tubiello et al., 2000; Stockle et al., 2003; Confalonieri and Bocchi, 2005; Confalonieri and Bechini, 2004). ORYZA1 was tested for rice (Casanov et al., 2000). CRITERIA and WOFOST integrated water balance and crop growth model is used as an operational tool (Marletto et al., 2001, 2005, 2007). CRITERIA is a model embedded within a geographical interface allowing to work with ArcGis shape files for soil maps and crop maps. P.RADA is an example of an operational application of
64
grapevine downy mildew model integrated with radar data for the estimation of rainfall and leaf wetness. In the Netherlands the operational use is focused on SUCROS and WOFOST (Supit et al., 1994) for winter wheat, spring barley, rape seed, grain maize, sugar beet, potato, sunflower, durum wheat, winter barley, rye, oats, field beans, field peas, soybean and rice. In addition to the mainstream of WOFOST versions several models have been elaborated on the basis of WOFOST (Van Ittersum et al, 2003). A typical example is the SWACROP2 model formed by linking the WOFOST crop module to the SWATRE (soil water and transpirationrate model), (Huygen, 1992). Groot (1987) simulated the nitrogen dynamics in crops and soil. Poels and Bijker (1993) developed the model TROPFOR to simulate growth and water use of tropical rainforest by adapting WOFOST and De Ruijter et al. (1993) adapted WOFOST for simulating tulip growth. Applications of WOFOST ver the last ten years, the successive WOFOST versions and their derivates have been used in many studies (e.g. Wolf and van Diepen, 1994, 1995; Wolf, 1993; Wolf and Erickson, 1993; Nonhebel, 1993). WOFOST has been applied as a tool for the analysis of yield risk and interannual yield variability, of yield variability over soil types, or over a range of agrohydrological conditions, of differences among cultivars, of relative importance of growth determining factors, of sowing strategies, effects of climate change and critical periods for use of agricultural machinery. The model has also been used for predictive purposes, in quantitative land evaluation, such as regional assessments of crop yield potential in the form of maximum yield levels, estimation of maximum benefits from irrigation or from fertilizer use, detection of adverse growing conditions by simulationmonitoring the agricultural season, and regional yield forecasts. Other models which originate from the C.T. de Wit Wageningen School of Production Ecology are LINGRA (Bouman et al., 1996) and LINTUS. SWAP (Kroes and van Dam, 2003) is another well known soil-water-atmosphereplant model focused on sophisticated soil water balance dynamics and integrating the WOFOST crop growth module (often used for irrigation studies) and ANIMO (Kroes and Roelsma, 2007) for soil carbon and nitrogen dynamics. The Australian model APSIM was tested for conditions in Netherlands (Asseng et al., 2000). In Norway process-oriented models are used for the simulation of cereals growth and yield as a research tool (Hutchinson et al., 1992). The dynamic crop models KONOR (Bleken, 2001) for wheat and ENGNOR (Baadshaug and Lantinga, 2002) for grass and clover grass leys are used for operational
65
applications. WOFOST is in use for studies of the potential production of barley in South-Eastern Norway (Hofmeister et al., 2001). In Poland the prognostic statistical-empirical yield model IPO (regional level), developed at the IUNG-PIB in Pulawy, is used operationally (Grski and Spoz-Pa, 1989; Grski et al., 1997). Based on decadal meteorological data from synoptic stations there are provided prognosis for: winter wheat, spring wheat, barley, oats, rape, sugar beet, potato. The prognosis are prepared for the Ministry of Agriculture and Rural Development. In the field of scientific investigations the model WOFOST was used for wheat (Faber et al., 1996) and potato (Bloch et al., 1997a,b) crops. An example of an applied disease model is NegFry, a Phytophtora model for potatoe (Kapsa, 2007). In Portugal CERES-Maize was adapted for a decision support system for maize (Braga, 2007); (information based only on a web survey). In Romania crop models are operationally used for Wheat and Maize (Marica and Busuioc, 2004; Marica, 2005; Marica et al., 2005a, 2005b). CROPWAT for assessment of water use and evapotranspiration was operationally used during 2002-2006 (Stancali and Marica, 2005), the FAO method for daily ETP is operationally used from 2006-present. The CERES Wheat implementation is in progress. Process-oriented models for scientific use are CERES-Wheat and Maize and the CROPWAT model for the assessment of evapotranspiration, the water demands of crops, in-soil humidity deficits and for maize yield. ROIMPEL (DSSAT based) was used at the national level for climate change impact studies for several annual crops (Audsley et al., 2006). In Serbia CROPSYST is in phase of experimental operational exploitation with data for six chosen locations in the most important agricultural areas, simulation and analysis of the growth and development of three maize hydrides from various maturation groups, forecasting yield and development. Scientific applications are reported for Wheat with SIRIUS (effect on yield, adaptation options, regional, field based), (Lalic et al., 2006), CROPSYST (for maize) and PERUN (effect on yield, water balance, regional, field based). For pest and desease modelling the BAHUS model system (Mihailovic et al. 2000, 2001, 2002; Lalic et al. 2003, 2007) was developed and tested for Potatoes (duration of incubation period, time and intensity of infection); for apple (duration of incubation period, time and intensity of infection) and for Grape (duration of incubation period, time and intensity of infection). Also a Cereal Leaf Beetle - Computing model (appearance and intensity of attack) was applied and tested (Pankovii et al., 2006).
66
The model WOFOST is in operational use for yield estimation and water balance monitoring in different governmental structures in Slovakia, such as in the Soil Science and Crop Research Institute, Ministry of soil management (www.vupu.sk). This model is used for wheat, maize, sunflower, potatoes and oilseed (rape) crops. Scientific use of process oriented models are reported for wheat and maize with WOFOST (effect on yield and water balance, regional), (Nejedlik and van Diepen, 1999), DAISY (effect on yield and water balance, regional) and CERES-Maize (effect on yield plant development, regional, field based applications), (Siska and Samhuel, 2007). The numerical model GLOBAL (Majercak and Novak, 1995) was applied for several crops for assessement of soil water dynamics. In Slovenia some research applications of plant disease models were used. For the purpose of better grapevine downy mildew protection the simulation of pathogen life cycle based on agrometeorological variables the model PLASMO and PERO have been realised and validated in the area of Gorika (Orlandini et al., 2005). Leaf wetness was calculated by the physical model DROPBEN shared in the frame of COST 718 activities. In the poject PRADA PRADA (SI/IT) project leaf wetness model upgrade in regard to radiation data has been introduced by SI (Sunik et al., 2007). Some recent research activities have been performed on water shortage impacts on agricultural plants Model IRRFIB was used for water balance scheduling of different crops and fruit trees (Sunik et al. 2006, KajfeBogataj et al., 2002). Model was incorporated into geographical information system of the Ministry of Agriculture for drought damage estimation maps elaboration. The model CERES Maize was used in the study of climate change impacts on maize yield in Slovenia (Kajfe-Bogataj, 1996). For Spain no application of an process crop model is reported, however, there is available a system of agroclimatic information for irrigation scheduling (SIAR, http://www.mapa.es/siar/) which operates on a daily time step, providing national (in total 361 agroclimatic weather stations covering the agricultural areas all over Spain) wide information (the main provider is Ministry of Agriculture). The aim is to provide representative agroclimatic data for Eto assessment and irrigation scheduling. These data and information are also used for research purposes in agroclimatology and for pest and diseases control. CERES-Wheat and CERES-Maize were used for climate change impact assessment or irrigation studies in scientific studies (e.g. Farre et al., 2000; Guerena et al., 2001; Iglesias and Mnguez, 1995; Iglesias et al., 2000; Iglesias, 2006). SWAP was introduced for studies on water balance on
67
crops under field irrigation in the frame of EU-project AGRIDEMA (Utset et al., 2007). An agrometeorological model for the prediction of grape yield was developed in the framework of the project OLIWlN aimed at the prediction of crop productivity for grape and olive tree (Marin, 1999). An example of disease model application in Spain is the evaluation of the different Alternaria prediction models on a potato crop (Iglesias et al., 2007). In Switzerland scientific applications of crop models are reported for wheat, maize and potato with the crop model CROPSYST (Stockle et al., 2003; Torriani et al., 2007) and PaSim for grassland (Riedo et al., 1998). For pest and disease warning operational seasonal forecasts for the appearance of apple aphid, apple sawfly, fruit moth, apple moth, cherry fruit fly, and pear sucker are available (http://www.sopra.admin.ch/). Other applications are the PhytoPRE model for a 2 days forecast for late blight (www.phytopre.ch), (Steenblock and Forrer, 2002; Steenblock et al., 2002; Musa-Steenblock and Forrer, 2005), 14-days forecast for risk of infection from Fusarium graminearum (www.fusaprog.ch), forecasts for downy mildew, fire blight and scab (www.agrometeo.ch). In United Kingdom the SIRIUS model, developed originally for New Zealand conditions (Semenov and Porter, 1995; Jamieson et al., 1998a,b; Lawless and Semenov, 2005), has been used to assess the effect of changing climate on maximum soil moisture deficit, drought-related reduction of potential yield, and wheat yields (Richter and Semenov, 2005). At Wye College, the CERES-Wheat model was used to predict crop responses under six climate change scenarios for the years 2025 and 2050 (Ghaffari et al., 2002). Similarly, at the University of Nottingham, both the SUCROS model and the CERES-Wheat model were evaluated for their ability to forecast final grain yield and crop biomass for four sites in the United Kingdom (Bannayan & Crout, 1999; Bannayan et al., 2003). Simpler models were also developed at the University of Nottingham to investigate dry matter accumulation and radiation use efficiency (Gillett et al., 2001). The same group also developed a number of models for tropical crops (e.g. Azamali et al., 1994; Bradley et al., 1994; Clifford et al., 2000), including the PARCH model to simulate the growth and development of sorghum, millet, and to a lesser extent, maize, in semi-arid environments (Bradley and Crout, 1994). The PARCH model was developed further to investigate aspects of rainwater harvesting in Tanzania, resulting in the PARCHED-THIRST model (Young and Gowing, 1996). At the University of Reading, the General Large-Area Model for Annual Crops (GLAM) has been developed to simulate growth of tropical crops
68
(Challinor et al., 2004), and has been incorporated into the Met Office Surface Exchange Scheme (MOSES) of the HadAM3 General Circulation model (Osborne et al., 2007). At the Macaulay Institute in Scotland, the LADSS (Land Allocation Decision Support System) farm-scale integrated modelling framework consists of a core of biophysical simulation models overlaid by financial, social and environmental accounting modules (Matthews et al., 1999). The crop module is a version of the CROPSYST model developed at Washington State University (Stockle et al., 1994). Also at Macaulay Institute, the People and Landscapes Model (PALM) is being used to investigate the interrelationships between socio-economic and biophysical processes (Matthews and Pilbeam, 2005; Matthews et al., 2005; Matthews, 2006). PALM is an agent-based model operating at the level of a catchment, which contains a number of decisionmaking entities (e.g. farm households) located on a landscape made up of a number of heterogeneous land units, each of which contains routines to simulate its water balance and carbon and nitrogen dynamics over time. Organic matter decomposition is simulated by a version of the CENTURY model (Parton et al., 1988), while water and nitrogen dynamics are simulated by versions of the routines in the DSSAT crop models (Tsuji et al., 1998).
2.4
Useful Outputs and main Limitations of Models for use in Climate Change Impact Research (Subgroup 3)
2.4.1 The problem of modelling natural phenomena Numerical meteorological prognoses; and also the different quantitative scenarios of the future climate of the earth are composed of very complicated scientifically based models based on the scientific principle. Generally modelling in agro meteorology (and in modelling in related fields of biology) is scientifically based predictions to be used for tactical and strategic decisins in crop production. The discussion in this paper is connected to possible strategic decisions for future crop production in Europe. In order to understand and discuss the temporal and spatial scope of modelling in meteorology and agro meteorology we ought to look at the scientific principle, see (Sivertsen, 2005, 2006) and (Sivertsen and Gailis, 2007). In these papers is presented an interpretation of the scientific principle containing four levels, classification of phenomena of nature, the basic hypotheses (laws of nature), the hypotheses derived from the laws of nature, and the initial values of the parameters. The outcome of the discussion is a documentation system for parameters (interpreted as quantitative entities attached to the
69
phenomena and the sub-phenomena). Two types of parameters are presented, those connected to the different systems for making measurements, and those being elements of the models of the phenomena of nature. In Fig. 2.9 this interpretation is shown graphically. The testing is not interpreted as merely a testing of the hypotheses of physics; but the idea is that the temporal and spatial scope of the system should be possible to discuss by using this interpretation. Just by looking at the definitions and the units of the input parameters, the parameters used in the sub-processes and the output parameters of crop growth models one may discuss several elements of the spatial and temporal scope of the models. The crop growth models discussed in this paper are all considered processoriented models. Eliens (1995) describes process-oriented models in this manner: With the process-oriented approach the components of the model consist of entities, which represent the existence of some object in the system such as a philosopher. An entity receives a user-defined phase that determines the behavior of the entity. We first identify the entities (or the types) in the model. The events are represented as methods of an entity. The function operator calls these events based on the phase the entity is in. This description is in fact based on object oriented thinking connected to modelling natural phenomena quantitatively. A process-oriented model of nature consists of the three levels, classification of phenomena, the laws of nature and the equational interpretation of these laws. The parameters used may be meteorological, physical and biological parameters, but the parameter of length of time should be contained in the model. In addition, several of the parameters and equations may be derived by empirical or half-empirical methods. In the contribution from Subgroup 2 in this paper, the processes contained in process oriented crop growth models is discussed more specifically. The general ideas connected to the scope of scientific modelling presented above are relevant for discussing the scenarios of future climate on global and regional European scale, for discussing systems for making weather predictions, and for discussing the crop growth models as well. All the models mentioned are very complicated constructions, but the each model and sub-model may be analysed according to the scheme presented in Fig. 2.9. Much information on the scope of any sub-system may be compiled by getting a review of the name, definition and unit of each parameter; and then find out how initialization of the parameters are made (name, definition, unit and measuring procedure for the measured parameters).
70
Figure 2.9: An interpretation of the scientific principle as this is used in meteorology and agrometeorology
71
2.4.2 Present limitations on crop model applications Europe The basis of the analysis was two questions of the survey to COST734 members: (a) Which model output of the named models are most useful in order detect the impact on climate change and variability regarding drought, excess rain, frost, snow cover and crop responses. (b) Please indicate the main limitations in order to apply process-oriented models for operational use in your country. For input data, please name which inputs. To help analysing the answers of the questionnaires two concepts were introduced, an interpretation of the scientific principle (as shown in Fig. 2.9), and the concept of a process oriented model is discussed connected to this interpretation. In Table 2.12 the answers on question (a) are summarized.
Table 2.12: Number of recommendations for the use of process-oriented models to determine consequences of drought, excess rain, excess frost, snow cover and crop responses (seasonal shifts etc.) Model CERES STICS WOFOST CropSyst PaSim AMBETI SWIM HERMES DAYCENT AGROSIM Weather Yield ROIMPEL APSIM Drought 5 2 4 1 1 1 1 1 1 1 1 1 Excess rain 8 Excess frost 1 1 1 Snow cover Seasonal shifts 8 1 1 Time step Daily Daily Daily Daily Daily and hourly Daily Daily Daily Daily Daily Daily Daily Daily
1 1
1 1 1 1 1 1 1
Recommended applications of crop models for assessing climate impacts as shown in Table 2.12 are the result of many years experience in different countries. Climate change is a very serious challenge for the agricultural crop production of Europe. Therefore the context of the actual application of the models is important to provide advice to societies in Europe.
72
The crops considered in Table 2.12 are wheat and maize connected to the model CERES, and wheat and grassland connected to the French model STICS, see Brisson et. al (2003). Specific crops are not explicitly mentioned for the other models. Most useful model outputs named for assessing drought impacts are water stress, phenology, yield (CERES, SWIM, HERMES and STICS), biomass (STICS, HERMES), soil water content (DAYCENT, Weather Yield, Roimpel, WOFOST, HERMES), and CARBON and NO-emissions (DAYCENT). For the models WOFOST and CERES see Faber et. al (1996) and Eitzinger et. al (2004). For the study of final biomass of the model STICS, see Brisson et al. (2003) and Brisson (2004). For drought response and performance of HERMES see Kersebaum et al. (2008). For assessing impacts of excess rain named most useful outputs are water water stress, phenology and yield (CERES, WOFOST, HERMES), oxygen shortage (WOFOST, HERMES) and nitrogen leaching (HERMES). For frost it was named biomass, where the parameter of interest is minimum air temperature. The parameters characterizing snow cover are snow depth, the duration of the snow cover, and water equivalent of snow which are in same cases of interest for simulated soil water content, soil temperature and phenology. Main named outputs for detecting seasonal shifts are phenology, water uptake, biomass and yield. Limitations of crop model applications are often related to availability and quality of model input data, which is also reflected by the survey (Table 2.13). A weak quality of input data are a main source of uncertainty in simulated outputs (beyond the representation of significant natural processes in the model) such as the spatial representation of the weather and soil model input data or quality problems of data caused by measurement methods. Especially soil input data need to be considered critically, because of its importance for soil water storage and availability for crops. Often, for example, spatial data on field capacity or wilting point are not directly available and need to be assessed by e.g. pedotransfer functions. In the paper of Gijsman et al. (2002) it is shown that even the use of different pedotransfer functions can lead to significant deviations in simulated crop yields. The weather input data for crop models and their spatial representation are very important in climate change impact studies (Tsvetsinskaya et al., 2003). Also the representation of historical temperature, precipitation and solar radiation is important in order to be applied for climate change scenarios. In several studies crop model sensitivities on these factors are compared and estimated (Nonhebel, 1993;
73
Dubrovsk et al., 2000). For a better spatial representation, several authors apply interpolation methods on these weather parameters which should be chosen carefully (Trnka et al., 2005). A significant source of uncertainty remains from the applied methods and models, as different crop models can have various sensitivities and levels of representation of certain soil-crop-atmosphere processes (Janssen, 1994; Jamieson et al., 1998). For climate change impact studies, for example, a problem is the accurate simulation of the direct CO2-effect on biomass accumulation which is in many cases considered (if at all) as a fixed (positive) value in crop models, often only distinguishing between C3 and C4 crops. FACE (Free Air Carbon Experiments) experiments show a much more complex picture and strong variability in the magnitude of this effect between cultivars and environments, which is difficult to simulate (Wolf et al., 2002; Kartschall et al., 1995). Also effects on crop yield quality are normally not considered in crop models, such as experimentally measured reduced content of protein in several crops. Another similar problem is the still highly empirical simulated process of root growth, which has a strong feedback to soil water availability and use. It is shown that the representation of root growth can have a strong feedback on simulated soil water contents with soil depth (Eitzinger et al., 2004). The models should therefore be calibrated well in this aspect, which is of course often not possible or carried out under different soil conditions. In climate change impact studies for large areas e.g. the European scale often simplified models or empirical procedures within complex models are used, where these aspects are not considered at all due to lack of data and resources.
Table 2.13: Named limitations connected to the use of crop growth models in the survey Type of limitation Lack of availability of biological data for validation of models (long term yield of specific varieties etc.) Low quality of soil data like parameter value describing texture, organic content, soil physical indicators (too low spatial resolution for field applications etc.) Daily weather data not available in time Lack of quality of weather data Number 12
11
5 1
The answers in Table 2.13 reflect also practical problems connected to socioeconomic conditions and local administration of data (data policy of data holding institutions) in Europe. Especially the weak availability of relevant biological or crop data for model calibration and the low quality and/or
74
availabilty of physical soil data for modelling purposes is a main problem. Only 5 countries reported problems for the availability of daily weather data, which is a specific countrywise problem of data policy often related to the costs of data. Only one comment was connected to the low quality of weather data, which is good news as the quality of the data (historical or of climate scenarios) is crucial for the use of crop models in order to get reliable results or to make strategic decisions based on simulation results.
2.5
Crop simulation models in combination with Remote Sensing and GIS (Subgroup 4)
2.5.1 State of the art Interfacing crop simulation models with a GIS Interfacing crop simulation models with geographic information system (GIS) increases the possibilities of applying these models for regional planning and policy analysis (Hartkamp et al., 1999). GIS is a computer-assisted system that acquires, storages, analyses and displays geographic data. Due to the increasing pressure on land and water resources, land use management and forecast (crop, weather, fire, etc.) at spatial levels, information systems have become essential. GIS is developed to a powerful tool at the disposition of policy and decision makers (Maracchi et al., 2000; Heinemann et al., 2002). Interfacing crop simulation models with a GIS helps to accomplish spatial and temporal analysis at the same time: regional-scale crop behaviour has a spatial dimension and models produce a temporal output. GIS visualize the results in spatial and hence their study by spatial analysis of model results (Dadhwal, 2003). In this context following terms should be explained: Linking: Linking (Fig 2.10a) includes a GIS for spatially displaying model outputs, where simple linkage strategies are used. A simple approach, for example, is an interpolation of model outputs. An advanced strategy is to use GIS functions like interpolation, overlay, slope, etc. to produce a database, which contains model inputs as well as outputs. Communication between GIS and model is with definite grid cells or polygons in input and output files, which are transferred in ASCII or biary format between GIS and model. Linking is not able to utilize all possibilities of the system and suffers from limitation due to (a) dependence on formats of GIS and model, (b) incompatibility of
75
operating environments and (c) not fully utilizing the capability of GIS (Dadhwal, 2003). Combining: Combining (Fig. 2.10b) involves processing data in a GIS and displaying model results: the model is configured with GIS and data are exchanged automatically. This can be realized with facilities in GIS package of macro language, interface programmes, libraries of user callable. It requires more complex programming and data management than simple linking (Dadhwal, 2003). An example of combining is AEGIS (Agricultural and Environmental GIS) with ArcView (Engel et al., 1997). Integrating: One system is incorporated in another one: a model is embedded in GIS or a GIS system is integrated in a modelling system. This accords automatic use of relational database and statistical packages (Fig 2.10c). An expertise, effort and understanding of the two tools are necessary (Dadhwal, 2003).
Figure 2.10: Organizational structure for (a) linking, (b) combining and (c) integrating GIS and crop models (Hartkamp et al., 1999)
76
One example of application is the CGMS (Crop Growth Monitoring System) of MARS (Monitoring Agriculture with Remote Sensing) (http://mars.jrc.it/). The main intention of this project is to provide information on weather indicators and crop status during the growing season and to provide objective forecasts of crop yield for the EU member states early in the crop growth season (de Witt et al., 2005). The project used two crop models (WOFOST and LINGRA) and Arc/Info for operational yield forecasting of main crops in the European Union (Supit, 1997). Databases of meteorological data, soil characteristics from the European soil map (King et al., 1995) and crop specific parameters (BoonsPrins et al., 1993) are available for the whole EU (Russell et al., 1999). The system-analytical part includes three main components: Interpolation of meteorological data to a square grid (50x50 km) in real time simulation of daily crop growth and statistical evaluation of the results (historical yield statistics thought regression analysis in combination with a time-trend) The resulting regression equations per crop per region can be used to make actual yield forecast. CGMS produces on a 10 day and monthly basis three types of output on current cropping season: maps of accumulated daily weather variables on 50x50 km grid to uncover any anomalies, for example drought, frost maps of agricultural quality indicators based on comparison of simulated crop indicators with their long-term means maps and tables of yield forecasts (Dadhwal, 2003) In precision farming, the link between GIS and crop simulation models is very vital. Precision farming is a way of agricultural production that takes into account the within-field variability. It is a technology where the application seeding, nutrient replacement, spraying, etc. takes place to act on the local conditions of a given field (Nemnyi et al., 2003). Examples of the combination of GIS and crop models in the field precision farming are given by Han et al. (1995) with interface between Arc/Info GIS and SIMPOTATO or Satti and Jacobs (2004) with GWRAPPS, which tightly couples ArcGIS with the Agricultural Field Scale Irrigation Requirements Simulation (AFSIRS) model. Another area, where GIS and crop models are used, is in agro-ecological zoning. Aggarwal (1993) utilized WTGROWS to simulate potential and water-limited wheat yields for 219 weather stations spread all over India. The district areas as polygons and model input parameters of soil, weather stations
77
as well as agro-ecological regions were saved in ARC/INFO GIS. The model outputs of potential and rainfed productivity were stored in GIS as polygon attribute data. Based on potential and rainfed productivity, the districts were classified into ten yield zones represented as map (Dadhwal, 2003). A further example gives the study of Caldiz et al. (2001), where agro-ecological zones for potato production in Argentina were analyzed. It was demonstrated how GIS for land evaluation and the crop simulation model LINTUL-POTATO can be used together to assess possibilities for increasing crop production at regional or national scales. To evaluate agricultural land use options for the state of Haryana, India, Aggarwal et al. (1998) used simulation modelling, GIS as well as optimization techniques. The models for specific crops were linked to GIS layers of administrative boundaries, physiographic features, climate, soil and agro climatic zones. The crop model outputs studied effects of future climatic changes on crop potential/productivity (Bacsi et al., 1991; Carter and Saariko, 1996; Dadhwal, 2003).
Linking crop simulation models to Remote Sensing and GIS
The main benefit of using remote sensed information is that it provides a quantification of the actual state of crop for large area (Dadhwal, 2003), while crop models give a continuous estimate of growth over time. Already Wiegand et al. (1979) and Richardson et al. (1982) proposed the use of remotely sensed information to improve crop model outputs. Spectrally derived LAI (leaf area index) can be used as direct input to physiological crop model or as an independent check to model calculation for its re-initialization. In this way, less labour and material intensive methods are necessary (Dadhwal, 2003). The different ways to combine a crop model with remote sensing observations (radiometric or satellite data) were at first described by Maas (1988) and further revised by Delcolle et al. (1992) as well as by Moulin et al. (1998). Remote sensing data integration into the models can be classified into 5 methods: the direct use of a driving variable estimated from remote sensing data in the model; the updating of a state variable of the model (for example LAI) from remote sensing data (forcing strategy); the adjustment of an initial condition to obtain a simulation in agreement with the remote sensing observations (re-initialization strategy);
78
the adjustment of model parameters to obtain a simulation in agreement with the remote sensing observations (reparameterization/re-calibration strategy); the corrective method; a relationship between error in some intermediate variable as estimated from remotely sensed measurement and error in final yield. This relationship may be applied to a case in which final yield is not known (Dadhwal, 2003).
Direct use of driving variable
Driving variables of crop simulation models are weather input data, such as maximum and minimum temperature, solar radiation, precipitation, wind speed etc.. Thornton et al. (1997) for example used METEOSAT based rainfall estimates to incorporate the current seasons rainfall data as input to CERES-Millet in Burkina Faso. The rainfall based on cold cloud duration as measured by thermal infrared radiometers on the METEOSAT satellite and was calculated every 10 days for a 7.6 km grid. With this new input provincial yields were simulated halfway the growing season and were within 15% of their final end-of-season values. Maas (1988) calculated the ratio of daily absorbed PAR (Q) to integrated daily PAR (R) from radiometric NDVI and generated daily values of Q/R by linear interpolation between NDVI measurements. The result was used as driving variable in a simplified maize growth model. The model showed an over assessment of + 6.2% ground biomass at anthesis. De Wit and van Diepen 2007 used an Ensemble Kalman filter (EnKF) for assimilating coarse resolution soil moisture estimates in the WOFOST crop model in order to compensate the effect of uncertainty in the rainfall. Investigation area was Spain, France, Italy and Germany for the period 19922000, crop model simulations with the EnFK were used for winter wheat and grain maize. The results were evaluated with regard to the EnFK filter innovations and the relationship with yield statistics on administrative regions. The assimilation of soil moisture estimates has clearly improved the relationship with crop yield statistics for winter wheat for the majority of regions (66%) where a relationship could be established. For grain maize the improvement is less evident because improved relationships could only be found for 56% of the regions. One reason is that these areas were irrigated and the model does not include irrigation.
Forcing strategy
The forcing strategy is to update at least one state variable in the model using remote sensing data. As the crop model runs at a constant time step, remote
79
sensing imagery is normally not produced at such high-resolution temporal steps. The common way is to fit an empirical curve to the estimated values from remote sensing observations and then carry out the interpolation according to the model time step. LAI is the most commonly updated state variable in the forcing strategy. The concepts of a simple crop simulation model and its modification for remote sensing derived LAI forcing is demonstrated in Fig. 2.11 (Delcolle and Gurif, 1988). a) b)
Fig. 2.11: (a) Simple schematic of a crop simulation model. (b) Modified structure of crop simulation model with remote sensing - based LAI forcing (Delcolle et al., 1988)
Some examples of forcing spectrally derived LAI in crop simulation models are summarized in Table 2.14 (Dadhwal, 2003).
Table 2.14: Selected case studies on use of remote simulation models (Dadhwal, 2003) Crop Model LAI estimation & interfacing Maize Ground NDVI-LAI on obs. Dates Wheat AFRCWHEAT SPOT/HRV LAIWDVI relation: daily interpolated LAI LAI-WDVI sensing derived LAI for forcing crop Evaluation of performance AGDM estimation improved Yield RMSE decreased Biomass at harvest Reference Maas 1988
Delcolle et al. 1988
Wheat
SUCROS
Bouman 1992, 1995
Re-initialization strategy
The re-initialization method involves adjustment of initial condition of the state variables. The main objective is to minimize the difference between a
80
derived state variable or the radiometric signal and its simulation (Dadhwal, 2003). Maas (1988) modified the initial value of LAI (L0) in a simplified maize model at emergence based on the minimization of an error function between remotely sensed LAI values and simulated LAI values during the course of simulation. When re-initialization is used for only one observation produced results, then it is quite similar to updating (forcing). The stability of model estimates obtained through re-initialization increased as more observations were used. The observation at 51 days after emergence, which caused a 42% error using updating, resulted in less than a 3% error using reinitialization (Maas, 1988).
Re-calibration/re-parameterization strategy
The model is formally adequate but requires re-calibration. This can be reached by minimizing the error between the remote sensing derived state variable (usually LAI) and its simulation by the model (Dadhwal, 2003). Delcolle et al. (1992) studied a re-calibration for rice crop with the GRAMI model. Values of one to four parameters were re-calibrated between the simulated LAI profile and observed LAI values. The results showed that improvement in simulated LAI profile by re-calibration depends largely on the number and timing of LAI observations. Gurif and Duke (1998) calibrated the SUCROS model on local scale with the SAIL reflectance model (Verhoef, 1984) on emergence and early growth parameters. The resulting model was calibrated under standard conditions and afterwards it was evaluated under test conditions to which the studied parameters of the SUCROS model were adjusted. The test conditions seedbed structure was coarser and the sowing depth was greater than expected. Therefore, emergence occurred later and the initial leaf area was smaller. The SUCROS simulation using standard values for emergence and early growth parameters did not accurately predict crop growth under these test conditions. But the inversion of the combined model using a set of canopy reflectance measurements during crop establishment provided new parameter values that can be used to accurately estimate crop yield. Jongschaap (2006) tested various run-time calibration scenarios for replacing simulated values by remotely observed values to improve simulation results. The Rotast 1.0 model was used as it simulated daily interaction between climate, soils and crops. For a number of times in the growing season, simulated values of LAI and canopy nitrogen contents were replaced with values estimated form remote sensing. This was taken by CropScan equipment, which covers the electromagnetic spectrum between 460-810 nm in eight spectral bands. Better simulation results of variables could be accomplished with this method.
81
Corrective approach
Sehgal et al. (2001) used the corrective strategy for generating the wheat yield maps for farmers fields during 1998-99 in Alipur block (Delhi). The remote sensing inputs (LAI estimates) were linked to the crop simulation model WTGROWS for yield mapping. The results were validated with yield observations. Biometric relation of grain yield and LAI was predicted from simulations by running the model for a combination of input resources, management practices and soil types occurring in the area. Afterwards this biometric relationship was applied to all the crop fields of the study area for which the LAI was computed from remote sensing data. The WTGROWS simulated grain yield for the combination of inputs showed a yield variability of 1.1 to 4.9 tha-1. The matching range of simulated LAI on 27th Julian day was 0.6 to 4.2. The regression equation fitted on that day between simulated LAI and simulated grain yield showed saturating logarithmic nature with a R value of 0.81. The empirical biometric relation was applied to the LAI map of the wheat pixels and grain yield map of Alipur block was generated. The predicted yields ranged between 2.1 and 4.8 tha-1. The relationship of predicted grain yield and observed yield for the 22 farmers fields showed high correlation coefficient of 0.8 and a root mean square error of 597 kg ha-1 which was 17 per cent of the observed mean yield (Dadhwal, 2003). Doraiswamy et al. (2005) modified the crop yield model to incorporate biophysical parameters derived from MODIS 8-day imagery to access crop yields for McLean Country, Illinois, USA, during the 2000 growing season. Crop model parameters were adjusted by minimizing the differences between MODIS derived LAI and crop model simulated LAI for the entire growing season. Corn and soybean on a 71 x 52 km area were tested. The crop simulations were conducted on a 1.6 x 1.6 km spatial resolution grid and the results integrated to the county level. The results were within 10 % of observed county yields. Added value by remote sensing data Crop simulation models are used to describe the impact of climatic conditions and management strategies at field scale and can be applied in a distributed model at regional scale. The main problem with models is often an oversimplified description of the natural system, inaccurate parameterization and uncertainty, and therefore a low prediction performance (Dorigo et al., 2007). These problems are mainly evident at regional scales where model input parameters have to be gathered from scattered point locations such as weather stations (de Wit et al., 2005). Data on boundary conditions, like soil, management, are often not available or of low quality (see also chapter above)
82
and model parameters have to be estimated from limited experimental data (Dorigo et al., 2007). The most promising method to estimate crop yield is combining crop growth models and remote sensing data. The different ways for integrating crop models with remote sensing data can be generally classified into following groups: re-initializing or re-calibrating crop simulation models; the model outputs of LAI match with remote sensing observations (Fig. 2.12a), and direct use of remote sensing inputs as forcing variable (Fig. 2.12b), using simulation model to estimate impact of variation in a state variable (e.g. LAI) and final yield; using crop simulation models remote sensing differences to modelling yield predictions (Fig. 2.12c) (Dadhwal, 2003).
Figure 2.12: Schematic representation of different methods for the assimilation of remotely sensed model state variables in crop models: (a) calibration, (b) forcing, and (c) updating (Dorigo et al. 2007, adapted from Delcolle et al., 1992)
The linkage between crop simulation models and remote sensing has a number of applications in regional crop forecasting, agro-ecological zoning, crop suitability, yield gap analysis as well as in precision agriculture. In future this linkage will be more important and improvements in sensor capabilities (spatial resolution, hyper-spectral data) as well as retrieval of additional crop
83
parameters like chlorophyll, leaf N and canopy water status can be expected. Thermal remote sensing can provide canopy temperatures and microwave data the soil moisture. The improved characterization of crop and its growing environment would offer additional ways to modulate crop simulation towards capturing the spatial and temporal dimensions of crop growth variability (Dadhwal, 2003).
2.5.2 Spatial model applications in Europe The following report contains an overview of the state of art and results of an European survey by questionnaires on following aspects of crop model applications in Europe: Crop and other agrometeorological model applications (operational or research) which were done with combination of GIS and/or Remote Sensing. Crop and other agrometeorological model applications (operational or research) which were done beyond the national scale. A number of agrometeorological models are applied on a larger spatial scale than just on single sites at field level. These applications mostly include the use of GIS for visualization of model inputs and outputs. Spatial applications further often include the use of remote sensing data, combined for example in so called crop growth monitoring systems (CGMS), for an improved assessment of spatial variable model input data or control parameters for verification of model outputs (such as Leaf area index). Spatial applications are further focused on the assessment of regional or larger area crop growth and yield information for monitoring issues or climate change impact assessments. In Europe the number of spatial applications of agrometeorological models vary from country to country, often limited by the availability and costs of spatial variable model input data such as soil data (see chapter 2.4) or actual data such as weather data for operational applications. Several model applications were done not only for regions within nations but also on a larger scale including national scale and whole Europe, often for climate change impact studies. Also the spatial resolution of these applications varies, often limited by the availability of specific model input data. Table 2.15 (in 2 parts) gives an overview on the results of the COST734 survey for all European countries (including information from literature sources).
84
Table 2.15: Survey results on European spatial agrometeorological model applications including GIS, Remote Sensing data (bold: models in operational use or planned for operational use)
85
86
Spatial agrometeorological model applications in Europe for operational use Only few applications of crop growth monitoring systems are already operational in Europe. The general item of remote sensing data assimilation in crop models has been the subject of mainly methodological research work these last years: they have allowed to elaborate practical solutions, but the operational application is still limited by the large amount of data to be processed with a high technical level for a final result only slightly improved. The best known example of an operational application is the MARS Crop Yield Forecasting System (MCYFS) for food security for Europe and other parts of the world (http://agrifish.jrc.it/marsstat/), which is providing quantitative crop statistics at EU (for a 50x50km grid for NUTS units) and national levels, in near real time (Rembold et al., 2006). MCYFS was adapted also for national CGMS at a finer grid scale of 1x1km to 10x10km (for defined zones below NUTS level) for Belgium (B-CGMS; http://b-cgms.cra.wallonie.be/en/). B-CGMS is based on the existing European harvest forecasting system but the databanks are supplemented and refined by Belgian physical (soil data) and technical (temperature sums, crop management) parameters. Satellite data are used as an aid to arrive a quantitative estimate of production in B-CGMS wheres at the European CGMS uses it in for qualitative interpretation. According to Tychon et al. (2001), the B-CGMS crop yield estimates are the integrated result of three independent procedures : a spatial agrometeorological model, a trend function which copes with the long-term increases due to technological development, and the information provided by the 1 km resolution imagery of NOAAAVHRR and SPOT-Vegetation. All three approaches provide an independent estimation of the yield. The three results are introduced in a multi regression statistical model and the weight of each component is obtained by statistical fitting for each decade. During all the growing period, an optimization of the weights given to the three system components is searched in order to minimize the prediction error. A national example of spatial agroclimatic monitoring is SIGA (Servicio de Informacin Geogrfico Agrario-Service of Agrarian Geographics Information) is an application running at the Ministry of Agriculture (Deputy Direction of annual crops) in Spain (Sanchez et al., 2005). The application (SIGCH- GIS related to the management of annual crops) offers cartographic and alfanumerical information, thematic maps on agroclimatic variables and information about the plan of productive regionalisation of Spain for the application of the EC rules (EC-1251/1999) of the European Commission. There are also regional projects with similar characteristics like the SITNA, that is a territorial information system developed by the regional government of Navarra region. SAgMIS is internet
87
based GIS information system managed by the Environmental Agency of the Republic of Slovenia which includes in situ information on crop water balance and irrigation forecast. Maps of water balance for different areas in Slovenia could be obtained for different time scales upon request (Sunik and Kurnik, 2004). Similar tools have been developed for yield (or yield risk and crop status) prediction on the global scale during the past years. Two examples of global crop monitoring system should be mentioned which is the Global Water Satisfaction Index (GWSI) as a near real time monitoring system (Nieuwenhuis et al., 2006) operationally managed with MCYFS at JRC and the FAO-AEZ method (Agro Ecological Zoning Methodology; http://www.iiasa.ac.at/Research/LUC/index.html or http://www.iiasa.ac.at/Research/LUC/index.html) for estimating of potential and actual yields (focused on climate change impacts on crop production), (Fischer et al., 1998). The term AEZ refers to the Agro-Ecological Zones system, developed by the Food and Agriculture Organization of the United Nations (FAO) in collaboration with the International Institute for Applied Systems Analysis (IIASA). The approach enables rational land-use planning on the basis of an inventory of land resources and an evaluation of their biophysical limitations and potentials for crop production. The AEZ programs utilize the land resources inventory to assess all feasible agricultural land-use options and to quantify expected production of cropping activities relevant in a particular agro-ecological context, for specified management conditions and levels of inputs. The characterization of land resources includes all relevant components of climate, soils and landform, which are basic for the supply of water, energy, nutrients and physical support to plants. Recent availability of digital global databases of climatic parameters, topography, soil and terrain, and land cover has allowed for revisions and improvements in calculation procedures of AEZ crop suitability and land productivity potentials, and for expanding the geographical scope to temperate and boreal environments. This effectively enabled global coverage for AEZ assessments of agricultural potentials, and it has led to this Global AEZ study. Spatial agrometeorological model applications in Europe for research Most of spatial crop model applications in combination with GIS and remote sensing were carried out for research aims, mostly for climate change impact research studies. Further, the above described already operational systems are based on prior research studies. Almost all european countries are covered by relevant studies, however, not all countries were involved in relevant studies.
88
An extensive overview of the scaling problem in crop model application (spatial and temporal) is given by Faivre et al. (2004). They conclude that different techniques have been used but it appears that there is a lack of analysis of the methods and strategies and of the requirements. Although there exist a number of applications of crop model spatialisation, there is a crucial lack of operational and transferable tools adapted to this problem. Efforts exist but are not coordinated. In most examples, the interfaces are partial and not automated, which does not make the tool easily transferable to other researchers. These studies are often specific to one application; that is, they consider one particular model, they use one particular set of data from one study site, they address one specific question. Several research studies on spatial model applications including GIS and/or remote sensing data were carried out with the SUCROS model in the Netherlands and France (Bouman, 1992,1995; Guerif and Duke, 1998, 2000; Launay and Guerif, 2005). Guerif and Duke for example coupled the radiative transfer model SAIL with SUCROS in order to derive certain sensitive parameters of the crop model to yield, especially emergence and early crop growth, by simulated canopy reflectance. Similar examples were done by other studies using different crop models, eg. with the STICS crop model (Prevot et al., 2003). Very often NDVI data as a measure of the Leaf area index is used in combination with crop model applications to improve spatial yield estimates (see also prior chapter), e.g. by Moriondo et al. (2007) in combination with the crop model CROPSYST or Clevers et al. (2002) by using SPOT satellite data and the ROTASK crop model. Di Bella et al. (2004b, 2004c) used SPOT satellite for grassland biomass simulation by the STICS model. Jongschaap and Schouten (2005) used an approach combining optical and radar emote sensing data with point-based crop growth modelling by the crop model ROTASK. Flowering dates for wheat crops can be estimated from timeseries of C-band radar data, as radar signals are attenuated maximally at the flowering stage. This enables scaling up point models to regional applications without an increase in (phenological) field observations on the ground. This approach gave satisfactory results that are in agreement with regional yield statistics. Radar data (ENVISAT ASAR) were also used for multi temporal information of above ground biomass and LAI in combination with CERES Wheat (Dente et al., 2006). In another example de Wit and Van Diepen (2006) investigated the possibilities of deriving basic meteorological inputs (temperature, radiation, evapotranspiration) of a crop model (WOFOST) from observations of MeteoSat. The results demonstrate that the MeteoSat derived temperature and radiation products can used be as input in a mechanistic crop model. For
89
water-limited production levels they concluded that the MeteoSat based simulations are unable to reproduce the drought stress which usually occurs under Mediteranean conditions. This is a result of the fact that the MeteoSat based reference evapotranspiration is on average 30 percent smaller compared to the standard Penman reference evapotranspiration. Therefore a considerable recalibration of the evapotranspiration related components of the WOFOST model would be necessary before using the MeteoSat based reference evapotranspiration in the model. Upscaling methods of crop modelling without Remote Sensing data use were presented e.g. by Harrison et al. (2000) for phenological development to the European scale by using ARCWHEAT2, where weather generators are used for simulation of daily temperature values from monthly weather data. The CERES models were used in upscaling to national levels for climate change impact studies by aggregation of soil and climatic units for example in Finland (Saarikko, 2000) and in Czech Republic (Trnka et al., 2007). Upscaling methods in general often were applied within several EU-projects detecting climate change impacts on growth and yield of several crops, for example in ACELLERATES using the CERES based crop model ROIMPEL (Audsley et al., 2006). Methods of spatial input data aggregation were investigated in several papers (e.g. Lagacherie et al., 2000). An example for rye yield modelling for a specific catchment area in Northern Germany is presented by Richter (1999), testing aggregation of soil types. Generally, the temporal variation of climatic inputs affected simulated crop growth more than spatial variation of soils. For a known distribution of soil types in a catchment, the area-weighted average of yield simulations based on mean soil properties equals the weighted mean of the individual simulated values. However, in another study was shown that spatial soil variation is often a main problem (Lagacherie et al., 2000). This case study demonstrated that a significant effort must be made for increasing the precision of yield estimates over vast areas. Several possible improvements of the existing soil databases can be envisaged such as extending the range of systematic soil characterisation (e.g. bulk density), deriving more precise pedotransfer functions, providing more accurate descriptions of STUs (e.g. by means of range of values by soil horizons), and enlarging the mapping scale in view to locate the STU. Finally, an example for smaller spatial scales (e.g. catching within-field variations for precision agriculture applications) for maize, wheat and soybean was demonstrated by Basso et al. (2007).
90
2.6 Conclusions From the survey of Working Group 1 in the COST 734 participating countries in order to assess applications of agrometeorological indices and models in Europe it is clear that the relevant approaches and methods are of immense importance for research as well as for practical applications in European agriculture. Due to their simple application, agroclimatological indices can be considered as most important tools both for research and operational applications. Particularly the possibility of using a wide temporal time step (daily, weekly, monthly) makes the indices suitable for the application with historical climatic series, avoiding the use weather generators to create the hourly data frequently required by complex mechanistic models. The results of the questionnaires elaboration pointed out their large use at European level for many purposes, spatial (regional, national) and temporal (nowcasting, pastcasting, etc.) scales. However it seems to be clear the needs of a standardisation of their structure, in order to obtain correspondence tables, needed for the inter-comparison of the results and to improve the understanding of the results. There are few cases (drought index, grapevine quality index) where the different levels of indices are analysed taking into account thresholds describing the consequences of obtained values and the expected intervention needed to manage and to protect the agricultural systems from the main impacts. The more complex approaches, namely process oriented models are still very limited in operational applications (especially crop yield models), except the more simple models focused on irrigation scheduling or the already widely applied models for pest and disease management. In research, however, process oriented crop models play a very important role, especially for the assessment of global and climate change impacts on agriculture. However, a majority of these studies were carried out on a larger scale, neglecting the necessarily finer spatial resolution to be of relevance for local practical recommendations for farmers. One of the main difficulties for the application of process oriented models in a high spatial resolution at the research level is often the lack of model input data (not available, high costs, expensive data management, etc.), for example mostly named crop and management data for model validation and soil data. On the other hand new methods are being developed to overcome these problems partly such as by using GIS and integration of remote sensing data. Only very few examples exist for operational crop yield forecasting intergrating all these available tools, however, only to be used at the expert level. Beside the effects of climate change on crop productivity, which are the dominating studies till now, it is recommended that the modelling community should also have a
91
closer look on other aspects such as soil fertility and environmental issues like groundwater recharge and water quality, soil carbon stocks, erosion, trace gas emissions, etc. in the future. This is not only required from the environmental view, but also because some of these issues cause feedback effects on crop productivity, e.g. soil organic matter content and fertility or climate change itself trough trace gas emissions. Therefore, integrated modelling approaches are required which include the most relevant interactions in the soil-cropatmosphere system. We therefore also should try to combine our modelling of climate change impacts with ideas and experience of sustainable production.
2.7 Acknowledgements The authors thank all who contributed to this chapter, especially to all who responded to the questionnaires. Thanks to all other COST734 members for their support, especially to the leader and co-leader of the action. Part of the work on the study report of WG1 was directly supported by the research project DROSMON (Crop Drought Stress Monitoring by Remote Sensing) financed by the Austrian Science Fund (FWF).
2.8 References
Aggarwal P.K., N. Kalra, S.K. Bandhyopadhyay, H. Pathak, V.K. Sehgal, R. Kaur, T.B.S. Rajput, H.C. Joshi, R. Choudhary, R. Roetter, 1998. Exploring agricultural land use options for the State of Haryana: Biophysical modelling, pages 59-65. In: Roetter R., Hoanh C.T., Luat N.V., van lttersum M.K., van Laar H.H. (Eds.): Exchange of methodologies in land use planning. SysNet Research Paper Series No. 1, IRRI, Los Banos, Philippines. Aggarwal P.K., 1993. Agro-ecological zoning using crop growth simulation models: characterization of wheat environments of India, pages 97-109. In: Penning de Vries F.W.T., Teng P., Metselaar K. (Eds.): Systems Approaches for Agricultural Development, Vol. 2. Kluwer Academic Publishers, Dordrecht, The Netherlands. Alexandrov V and Hoogenboom G., 2000. The Impact of Climate Variability and Change on Crop Yield in Bulgaria. Agricultural and Forest Meteorology, 104(4): 315-327. Alexandrov V and Eitzinger J., 2000. An Approach for Irrigation Planning and Management of Maize Crop in North Bulgaria. Proceedings of the XXth Conference of the Danube Countries on Hydrological Forecasting and Hydrological Bases of Water Management, Bratislava, The Slovak Republic, (CD) 8 pp. Alexandrov V. and Eitzinger J., 2001. Potential Climate Change Impact on Winter Wheat and Spring Barley in Austria. Proceedings of the international conference on 150 Years of Meteorological Service in Central Europe, Stara Lesna, Slovakia, (CD) 15 pp. Alexandrov V. and Eitzinger J., 2005. The Potential Effect of Climate Change and Elevated Air Carbon Dioxide on Agricultural Crop Production in Central and Southeastern Europe. Journal of Crop Improvement 13(1-2): 291-331.
92
Alexandrov V. and Slavov N., 1998. Variations of Maize Yield Depending on Meteorological Conditions. Plant Sciences 35(1): 11-17 (in Bulgarian). Alexandrov V., 1997. Vulnerability of Agronomic Systems in Bulgaria. Climatic Change 36: 135-149 Alexandrov V., 1998. The Effect of Climate Variability and Change on Agroclimatic Resources and Crop Productivity in Bulgaria. Proceeding on the 2nd European Conference on Applied Climatology, October, 1998, Vienna, Austria, (CD) 6 pp. Alexandrov V., 1999. Vulnerability and Adaptation of Agronomic Systems in Bulgaria. Climate Research 12(2-3): 161-173. Alexandrov V., 2000. Bioproductivity, Agriculture and Climate Change. In: Staneva, M., G. Knight, T. Hristov and D. Mishev (eds.). Global Change and Bulgaria, University Park, Pennsylvania, USA and Sofia, pp. 177-198. Alexandrov V., 2001. Climate Change Impact on Water use of Maize in Bulgaria. Proceedings of the international conference on 150 Years of Meteorological Service in Central Europe, Stara Lesna, Slovakia, (CD) 13 pp. Alexandrov V., 2002. Summarizing Crop Growth Simulation Models in Europe with Potential for Operational Assessment of Crop Status and Yield Prognosis. In: Dunkel, Z., V.Alexandrov, Z. Gat, R. Guerreiro, A. Kleschenko and Y. Ozalp, 2002. Report of the RA VI Working Group on Agricultural Meteorology. CAgM Report No.89, WMO/TD No.1113, Geneva, Switzerland, pp. 119-214. Alexandrov V., J. Eitzinger, V. Cajic, M. Oberforster, 2002. Potential impact of climate change on selected agricultural crops in north-eastern Austria. Global Change Biology 8 (4), 372-389. Alexandrov V., 2003. The Impacts of Climate Variability and Change on Agriculture in Bulgaria. Proceedings of the International Symposium on Climate Change (ISCC), Beijing, China, WMO/TD-No.1172, pp. 232-237. Alexandrov V., J. Eitzinger, M. Oberforster, 2001. Climate Change Impacts on Agroecosystems in Selected Agricultural Regions in Austria. Proceedings of the 2nd International Symposium on Modelling Cropping Systems, Florence, Italy, pp. 95-96. Anon, 2008. Factsheet on Irish Agriculture. Economics and Planning Division, Department of Agriculture and Food, Government of Ireland. Asseng S., H. van Keulen, W. Stol, 2000. Performance and application of the APSIM Nwheat model in the Netherlands, European Journal of Agronomy, Volume 12, P: 37- 54, 2000. Audsley E., K.R. Pearn, C. Simota, G. Cojocaru, E. Koutsidou, M.D.A. Rounsevell, M. Trnka, V. Alexandrov, 2006. What can scenario modelling tell us about future European scale land use, and what not? Environmental Science and Policy 9(2): 148-162 Avgoustoglou, E. et al., 2002. Operational-Research Investigation of the model (LMCOSMO) at the Hellenic National Meteorological Service Proceedings of 6th Hellenic conference in Meteorology Climatology and Atmospheric Physics, Vol 1, Ioannina, 2528 September 2002, pp. 35-42. Azamali S.N., N.M.J. Crout, R.G. Bradley, 1994. Perspectives in modelling resource capture by crops. Resource Capture by Crops:125-148. Baadshaug O.H and Lantinga,E.A., 2002. ENGNOR. A Grassland Crop Growth model for High Latitudes Documentation, Report no, 2/2002 NLH Bacsi Z., P.K. Thorton, J.B. Dent, 1991. Impacts of future climate change on Hungarian crop production: An application of crop growth simulation models. Agric. Sys. 37, 435-450. Baker J.M., 2001. Introduction to Crop Modeling Papers from Eastern Europe and the Former Soviet Union, Agronomy Journal, Volume 93, P: 650.
93
Bannayan, M. and Crout N.M.J., 1999. A stochastic modelling approach for real-time forecasting of winter wheat yield. Field Crops Research 62(1):85-95. Bannayan M., N.M.J. Crout, G. Hoogenboom, 2003. Application of the CERES-Wheat model for within-season prediction of winter wheat yield in the United Kingdom. Agronomy Journal 95(1):114-125. Basso B., J.T. Ritchie, F.J. Pierce, R.P. Braga, J.W. Jones, Spatial validation of crop models for precision agriculture, Agricultural Systems, Volume 68, Number 2, pp. 97-112, May 2001. Basso B., M. Bertocco, L. Sartori, E.C. Martin, 2007. Analyzing the effects of climate variability on spatial pattern of yield in a maizewheatsoybean rotation, European Journal of Agronomy, Volume 26, P: 82- 91, 2007. Beaujouan V., P. Durand, L. Ruiz, 2001. Modelling the effect of the spatial distribution of agricultural practices on nitrogen fluxes in rural catchments. Ecol. Model, 137: 93-105. Bella C.D., R. Faivre, F. Ruget, B. Seguin, 2005. Using VEGETATION satellite data and the crop model STICS-Prairie to estimate pasture production at the national level in France, Physics and Chemistry of the Earth, Volume 30, P: 39, 2005. Berntsen, J., B.M. Petersen, Olesen, J.E., 2006. Simulating trends in crop yield and soil carbon in a long term experiment effects of rising CO2, N deposition and improved cultivation. Plant and Soil 287, 235-245. Biak L., 2006. Primjena modela za prognozu vanih biljnih bolesti u Republici Hrvatskoj, (Model application for forecasting the occurrence of more important plant diseases in Croatia), magistarski rad, Agronomski fakultet, Zagreb,103pp. Bleken, M.A., 2001. KONOR. A model for simulation of cereal growth. Documentation. Agricultural University of Norway. Report No. 2/2201. 33 pp. Boch Z., A. Faber, G. Demidowicz, J. Kamasa, 1997a. Zastosowanie modelu WOFOST dosymulacji wzrostu i plonowania ziemniakw uprawianych w Polsce. I. Kalibracja modelu. (Growth and yield simulations of potato cultivated in Poland using WOFOTS model calibbration). Frag. Agron., 2 (54), 88-95. Boch Z., A. Faber, G. Demidowicz, J. Kamasa, 1997b. Zastosowanie modelu WOFOST do symulacji wzrostu i plonowania ziemniakw uprawianych w Polsce. I. Weryfikacja modelu. (Growth and yield simulations of potato cultivated in Poland using WOFOST model verification). Frag. Agron., 2 (54), 96-101. Bochenek Z., K. Dbrowska-Zieliska, A. Ciokosz, S. Drupka, K. Vijendra, Boken, 2005. Monitoring Agricultural Drought in Poland. W: Monitoring Agricultural Drought - A Global Study. Oxford University Press. Boons-Prins E.R, G.H.J.de Koning, C.A. van Diepen, F.W.T. Penning de Vries, 1993. Crop Specific Simulation Parameters for Yield Forecasting across the European Community. Simulation Reports No 32. CABO-DLO, The Netherlands. Bouman B.A.M., 1995. Crop modelling and remote sensing for yield prediction. Netherlands J. Agric. Sci. 43, 143-161. Bouman B.A.M., 1992. Linking physical remote sensing models with crop growth simulation models applied for sugar beet, International Journal of Remote Sensing, Volume 13, P: 2565 2581, 1992 Bouman, B.A.M.; A.H.C.M. Schapendonk, 1996. Stol, W.; van Kraalingen, D.W.G., Description of the growth model LINGRA as implemented in CGMS. In Quantitative Approaches in Systems Analysis, Wageningen: European Commission, Production Ecology and AB-DLO, Wageningen University and Research Centre , 7, 11-22 Braden H., 1995. The model AMBETI A detailed description of a soil-plant-atmosphere model. Berichte des Deutschen Wetterdienstes 195, Offenbach.
94
Bradley R.G. and Crout N.M.J., 1994. The PARCH Model for Predicting Arable Resource Capture in Hostile Environments: Users' Guide. Tropical Crops Research Unit, University of Nottingham, Sutton Bonington, Leicestershire, UK. 122 pp. Bradley R.G., N.M.J. Crout, S.N. Azam-Ali, 1994. Modelling crop growth in semi-arid environments. Journal of Agricultural Science 122(1):161-162. Braga R.P, M.J. Cardoso, J.P. Coelho, 2007. Crop model based decision support for maize (Zea mays L.) silage production in Portugal, European Journal of Agronomy(Article in press), 2007. Brereton A.J., S.A. Danielov, D. Scott, 1996. Agrometeorology of Grass and Grasslands for Middle Latitudes. Technical note no.197. Geneva, Switzerland: World Meteorological Organization Brisson N., 2004. Contribution of the STICS model outputs to the agronomic analysis of the 2003 drought in France. ESA symposium Copenhagen Brisson N., F. Bussire, H. Ozier-Lafontaine, H.R. Sinoquet Tournebize, 2004. Adaptation of the crop model STICS to intercropping. Theoretical basis and parameterisation. Agronomie, 24: 409-421 Brisson N., G. Corre-Hellou, A. Dibet, M. Launay, Y. Crozat, 2006. Evaluation of the STICS crop model within the EU INTERCROP project. Grain Legumes, 45: 10-12 Brisson N., C. Gary, E. Justes, R. Roche, B. Mary, D. Ripoche, D. Zimmer, J. Sierra, P. Bertuzzi, P. Burger, F. Bussiere, Y.M. Cabidoche, P. Cellier, P. Debake, J.P. Gaudillere, C. Henault, F. Maraux, B. Seguin, H. Sinoquet, 2003. An overview of the crop model STICS. Eur. J. Agron.,18, 309-332. Buffet D., 2000. Monitoring Crop Growth at National or Regional Scale with an AgroMeteorological Model and Remote Sensing Data through a GIS Web- Based Mapping Application, 6th EC- GI&GIS Workshop, France, 28- 30 June 2000. Buffet D., R. Oger, D. Dehemd, B. Tychon, H. EerensS., 1999a. Implementation of the Belgian Crop Growth Monitoring System with a Web-based mapping application linking a Crop Modelling System and GIS database, B- CGMS Project: ESA International Symposium at Lleida, Spain, 1999. Buffet D., R. Oger, D. Dehem, B. Tychon, F. Veroustraete, K. Wouters, 1999b. A Web-Based Mapping Application To Link A Crop Modelling System (B- CGMS) And A GIS Database, B- CGMS Project: ESA International Symposium at Lleida, Spain, 1999. Cajic V., J. Eitzinger, V. Alexandrov, 2000. Simulated Response of Soybean to Incremental Climate Change Scenarios at a Specified Location in Austria using CROPGRO Model. Proceedings of the 3rd European Conference on Applied Climatology "Tools for the environment and man of the year 2000" Pisa, Italy, (CD) 6 pp. Caldiz D.O., F.J. Gaspari, A.J. Haverkort, P.C. Struik, 2001. Agro-ecological zoning and potential yield of single or double cropping of potato in Argentina, Agricultural and Forest Meteorology 109 (4), 311-320. Carter T.R. and Saariko R.A., 1996. Estimating regional crop potential in Finland under a changing climate. Agric. For. Meteorol. 79, 301-313. Carter T.R., R.A. Saarikko, S.K.H. Joukainen, 2000. Modelling climate change impacts on wheat and potato in Finland. Chapter 16 in: T.E. Downing, P.A. Harrison, R.E. Butterfield and K.G. Lonsdale (Eds). Climate Change, Climatic Variability and Agriculture in Europe: An Integrated Assessment. Research Report No. 21, Environmental Change Unit, University of Oxford, Oxford, UK, pp. 289-312. Casanova D., J. Goudriaan, A.D. Bosch, 2000. Testing the performance of ORYZA1, an explanatory model for rice growth simulation, for Mediterranean conditions, European Journal of Agronomy, Volume 12, P: 175189.
95
Castrignano A., N. Katerji, F. Karam, M. Mastrorilli, A. Hamdy, 2002. A modified version of ceres-maize model for predicting crop response to salinity stress, Mediterranean crop responses to water and soil salinity: Eco-physiological and agronomic analyses . Bari : CIHEAM-IAMB, p. P: 161-181 : 5 ill. 6 tables. 24 ref. Summary (En). (Options Mditerranennes : Srie B. Etudes et Recherches ; n. 36), 2002. Cepuder P., V. Aus der Schmitten, T. Alamirew, 2002. EPIC and GIS - a tool to reduce nitrogen leaching. In: Srimuang Printing Co., Ltd. (Ed.): 17th World Congress of Soil Science, 14.-21.8.2002, Bangkok, Thailand; Proceedings, Vol. V, 1621, Bangkok, Thailand. Challinor A.J., T.R. Wheeler, P.Q. Craufurd, J.M. Slingo, D.I.F. Grimes, 2004. Design and optimisation of a large-area process-based model for annual crops. Agric. For. Meteorol. 124:99-120. Clevers J.G.P.W., C. Bker, H.J.C. van Leeuwen, B.A.M. Bouman, 1994. A Framework For Monitoring Crop Growth By Combining Directional And Spectral Remote Sensing Information, Remote Sensing of Environment, Volume 50, Issue 2, P: 161-170, November 1994. Clevers J.G.P.W., O.W. Vonder, R.E.E. Jongschaap, J.F. Desprats, C. King, L. Prvot ,N. Bruguier, 2002. Using SPOT data for calibrating a wheat growth model under Mediterranean conditions, Agronomie, Volume 22, P: 687694. Clifford S.C., I.M. Stronach, C.R. Black, P.R. Singleton-Jones, S.N. Azam-Ali, N.M.J. Crout, 2000. Effects of elevated CO2, drought and temperature on the water relations and gas exchange of groundnut (Arachis hypogaea) stands grown in controlled environment glasshouses. Physiologia Plantarum 110(1):78-88. Confalonieri R., L. Bechini, 2004. A preliminary evaluation of the simulation model CropSyst for alfalfa, European Journal of Agronomy, Volume 21, P: 223237, 2004. Confalonieri R., S. Bocchi, 2005. Evaluation of CropSyst for simulating the yield of flooded rice in northern Italy, European Journal of Agronomy, Volume 23, P: 315- 326, 2005. Courault D., F. Ruget 2001. Impact of local climate variability on crop model estimates in the south-east of France. Climate research, 18: 195-204. Cvjetkovi B., E. Halupecki, J. poljari 1999. The occurence and control of fire blight in Croatia, Acta Horticulturae 489, 71-77. Cvjetkovi B, M. ubi, Z. Matotan, 2000. Modeli prognoze za suzbijanje plamenjae (Phytophthora infestans mont. de bary) na rajici (Forecasting methods for controlling late blight (Phytophthora infestans mont. de bary) on tomato), Zbornik saetaka Postignnua i perspektive hrvatskog poljodjelstva, Osijek, 163. Dabrowska-Zielinska K.; F. Kogan, A. Ciolkosz,; Gruszczynska, M.; Kowalik, W., 2002. Modelling of crop growth conditions and crop yield in Poland using AVHRR-based indices. International Journal of Remote Sensing, Vol. 23, Issue 6, 1109 1123. Dadhwal V.K. 2003. Crop growth and productivity monitoring and simulation using remote sensing and GIS. In: Sivakumar M.V.K, Roy P.S., Harmsen, K., Saha S.K. (Eds.): Satellite Remote Sensing and GIS Applications in Agricultural Meteorology. Proceedings of the Training Workshop 7 11 July, 2003, Dehra Dun, India, 263-290. Dalezios N. R., 1988. The Palmer Drought Severity Index (PDSI) for rain enhancement in northern Greece. International Symposium/ Workshop on Weather Modification, OGA, 1-4 Aug., Thessaloniki, Greece. Dalezios N.R. and Spanos S.I., 1995. Operational Efficiency Assessment of Hail Suppression for Agriculture in Greece. J. of Wea. Mod., Vol.27 (1), 21-35. Dalezios N.R., D. Bampzelis, G. Daoularis, 2000. Agrometeorological Simulation Using PERO Model for Grape Vine Downy Mildew in Greece.
96
Danalatos N.G. and Sgouras I., 1987. WOFOST: A methodology for quantifying production potentials and its applicability in Greece. Proceedings of 2nd Panhellenic Conference in Soil Sciences, Greek Publications, pp. 497-512. Danalatos N.G., 1997. A crop-growth simulation model for calculating biomass production potentials of sorghum within the beaver expert system environment. Final Partial Report of the E.C. Project AIR3-CT93-0985: Biomass Economic Appraisal & eValuation ExpeRt (BEAVER). Agricultural University of Athens, Dept. of Agric. Engineering. Athens, 16 pp. Danalatos N.G., C. Kosmas, P.M. Driessen, N. Yassoglou, 1994. The change in the specific leaf area of maize grown under Mediterranean conditions. Agronomie 14(1994): 1-8. Danalatos N.G., C. Kosmas, St. Gerontides, S. Tzortzios, 1998. A simplified cropwaterbalance simulation model and its applicability for estimating the water-limited production of wheat under semi-arid conditions. 7th ICCTA International Congress for Computer Technology in Agriculture, 15-18 November 1998, Florence, Italy. (Proceedings in press). De Noblet-Ducoudr N., S. Gervois, P. Ciais, N. Viovy, N. Brisson, B. Seguin, A. Perrier 2006. Coupling the Soil-Vegetation-Atmosphere-Transfer Scheme ORCHIDEE to the agronomy model STICS to study the influence of croplands on the European carbon and water budgets. Agronomie, 24: 397-407. De Wit A.J.W. and van Diepen C.A., 2007. Crop model data assimilation with the Ensemble Kalman filter for improving regional crop yield forecasts. Agricultural and Forest Meteorology 14, 38-56. De Wit A.J.W., H.L. Boogaard, C.A. van Diepen, 2005. Spatial resolution of precipitation and radiation: The effect on regional crop yield forecasts. Agricultural and Forest Meteorology 135, 156-168. De Wit A., K. van Diepen, 2006. Crop Growth Modelling And Crop Yield Forecasting Using Satellite- Derived Meteorological Inputs, ISPRS Commission VII Mid-term Symposium Remote Sensing: From Pixels to Processes, Netherlands, P: 667- 671, 811 May 2006. Delcolle R., and Gurif M., 1988. Introducing spectral data into a plant process model for improving its prediction ability. Preceedings of the 4th International Colloquium Signatures Spectrales dObjects en Teledetection, 18-22 Jan, 1988, Aussols, France, 125-127. Delcolle R., S.J. Maas, M. Gurif, F. Baret, 1992. Remote sensing and crop production models: present trends. ISPRS J. Photogramm. Remote Sens. 47, 145-161. Dente L., F. Mattia, M. Rinaldi, G. Satalino, 2006. Synergistic Use of ASAR AP Data and Crop Growth Model for Yield Map Estimates, Geophysical Research Abstracts, Volume 8. Di Bella C., R. Faivre, F. Ruget, B. Seguin. 2004b. Using VEGETATION satellite data and the crop model STICS- Prairie to estimate pasture production at the national level in France. Physics and Chemistry of the earth,30,3-9. Di Bella C., R. Faivre, F. Ruget, B. Seguin, M. Guerif, B. Combal, M. Weiss, C. Rebella, 2004c. Use of VEGETATION satellite data to improve pasture production simulated by STICS-Prairie in France. Agronomie,24,437-444. Di Bella C., R. Faivre, F. Ruget, B. Seguin, M. Guerif, B. Combal, M. Weiss, C. Rebella 2004a. Remote sensing capabilities to estimate pasture production in France. International Journal of Remote Sensing, 25, 5359-5372.
97
Donatelli M., C. Stockle, E. Ceotto, M. Rinaldi, 1997. Evaluation of CropSyst for Cropping Systems at two location of northern and southern Italy, European Journal of Agronomy, Volume 6, P: 35-45,1997. Doraiswamy P.C., T.R. Sinclair, S. Hollinger, B. Akhmedov, A. Stern, J. Prueger, 2005. Application of MODIS derived parameters for regional crop yield assessment. Remote Sensing of Environment 97, 192-202. Dorigo W.A., R. Zurita-Milla, A.J.W. de Witt, J. Brazile, R. Singh, M.E. Schaepman, 2007. A review on reflective remote sensing and data assimilation techniques for enhanced agroecosystem modelling. International Journal of Applied Earth Observation and Geoinformation 9, 165-193. Downing T.E, P.A. Harrison, R.E. Butterfield, K.G. Lonsdale, (eds) 2000. Climate Change, Climatic Variability and Agriculture in Europe. An Integrated Assessment, Research Report No. 21, Brussels, Belgium: Commission of the European Union, Contract ENV4-CT95-0154, 445 pp. Dubrovsky M., Z. Zalud, M. tastn, 2000. Sensitivity of CERES-Maize yields to statistical structure of daily weather series. Climatic Change 46, 447- 472. Drr C., M. Gurif, F. Brochery, F. Ferr, 1999. Study of crop establishment effects on subsequent growth using a crop growth model (SUCROS). /ESA International Symposium "Modelling cropping systems", June 1999, Lleida (SP),171-172/. Eitzinger J., M. tastn, Z. alud, M. Dubrovsk, 2003. A simulation study of the effect of soil water balance and water stress on winter wheat production under different climate change scenarios. Agricultural Water Management, 61, 3, 163-234. Eitzinger J., M. Trnka, J. Hsch, Z. alud, M. Dubrovsk, 2004. Comparison of CERES, WOFOST and SWAP models in simulating soil water content during growing season under different soil conditions. Ecological Modelling 171 (3), 223-246. Eitzinger J., Z. Zalud, V. Alexandrov, C.A. Diepen, M. Trnka, M. Dubrovsky, D. Semeradova, M. Oberfoster, 2001. A local simulation study on the impact of climate change on winter wheat production in north-eastern Austria, Austrian Journal of Agricultural Research 52(4): 199-212. Eitzinger J., Z. alud, C.A. van Diepen, M. Trnka, D. Semerdov, M. Dubrovsk, M. Oberforster, 2000. Calibration and evaluation of the WOFOST model for winter wheat. 8th International Poster day ' Transport of Water, Chemicals and Energy in the System Soil-Crop Canopy-Atmosphere' 16.11.2000, Bratislava. (CD version, ISBN 80-9684800-3, Institute of Hydrology, Slovak Academy of Sciences). Engel T., G. Hoogenboom, J.W. Jones, P.W. Wilkens 1997. AEGIS/WIN - a program for the application of crop simulation models across geographic areas. Agronomy J. 89, 919928. Faber A., 2002. rodowiskowe uwarunkowania produkcji rolinnej w Polsce i Europie wedug symulacji CGSM. (Natural conditions for crop production in Poland and Europe according to CGMS simulation). Pam. Pu. 130, 137-151 Faber A., Z. Boch, A. Nierbca, G. Demidowicz, L. Kaczyski, 1996. Symulacja wzrostu i plonowania pszenicy ozimej uprawianej w Polsce przy uyciu modelu wzrostu WOFOST. II Weryfikacja Modelu. (Growth and yield simulations of winter wheat cultivated in Poland using WOFOST). Fragm. Agron. 4, 51-58. FAG-Forschungsanstalt Geisenheim, 2005. Der Einfluss klimatischer Vernderungen auf die phnologische Entwicklung der Rebe, die Sorteneignung sowie Mostgewicht und Surestruktur der Trauben. INKLIM Final Report. http://www.hlug.de/medien/luft/inklim/dokumente/endberichte/weinbau.pdf
98
Faivre R., D. Leenhardt, M. Voltz, M. Benot, F. Papy, G. Dedieu, D.Wallach, 2004. Spatialising crop models, Agronomie, Volume 24, P: 205217, 2004. Farre I., M. van Oijen, P.A. Leffelaar, J.M. Faci, 2000. Analysis of maize growth for different irrigation strategies in northeastern Spain. European Journal of Agronomy, 12: 225 238. Fischer G., G. Granat, M. Makowski, 1998. AEZWIN. An interactive multiple-criteria analysis tool for land resources appraisal. Report IR-98-051. International Institute for Applied Systems Analysis, Laxenburg, Austria. Fitzgerald J.B., A.J. Brereton, N.M. Holden, 2005. Dairy system simulation for assessing regional climate variation effects on management Grass and Forage Science 60: 283296 Fitzgerald J.B., A.J. Brereton, N.M. Holden. 2008a. Simulation of the influence of poor soil drainage on low-cost grass-based dairy production. Grass and Forage Science (reviewed) Fitzgerald J.B., A.J. Brereton, N.M. Holden. 2008b. Assessment of the adaptation potential of low-cost grass-based dairy systems to climate change in Ireland the maximised production scenario. Agricultural and Forest Meteorology (reviewed) Fodor N., 2006. 4M Software For Modelling And Analysing Cropping Systems, Journal of Universal Computer Science, Volume 12, No. 9, P: 1196-1207, 2006. Fodor N., G. Kovacs, 2003. Sensitivity of crop models to the inaccuracy of meteorological observations, Geophysical Research Abstracts, Volume 5, 09186, 2003. Fodor N., G.M. Gaspar, K. Pokovai, G.J. Kovacs, 2003. 4M- Software Package For Modeling Cropping Systems, European Journal of Agronomy 18, P: 389-393, 2003. Friesland and Lpmeier, 2006. The activities of the German Weather Service (DWD) in the field of agroclimatology. Meteorological applications 13, Supplement 1, 61-67. Garca de Cortzar Atauri I., 2006. Adaptation du modle STICS la vigne (Vitis vinifera L.). Utilisation dans le cadre dune tude du changement climatique lchelle de la France. Thse de doctorat de lENSA. 349pp (Adaptation of the STICS model to grapevine; Use within a study of climate change at the scale of France, PhD thesis) Gerstengarbe F.W., F. Badeck, F. Hattermann, V. Krysanova, W. Lahmer, P. Lasch, M. Stock, F. Suckow, F. Wechsung, P.C. Werner, 2003. Studie zur klimatischen Entwicklung im Land Brandenburg bis 2055 und deren Auswirkungen auf den Wasserhaushalt, die Forst- und Landwirtschaft sowie die Ableitung erster Perspektiven. PIK-Report, Nr. 83, ISSN 1436-0179 Ghaffari A., H.F. Cook, H.C. Lee, 2002. Climate change and winter wheat management: A modelling scenario for south-eastern England. Climatic Change 55(4):509-533. Gillett A.G., N.M.J. Crout, D.T. Stokes, R. Sylvester-Bradley, R.K. Scott, 2001. An approach to modelling the effect of environmental and physiological factors upon biomass accumulation in winter wheat. Journal of Agricultural Science 136:369-381. Gijsman A.J., S.S. Jagtap, J.W. Jones, 2002. Wading through a swamp of complete confusion: how to choose a method for estimating soil water retention parameters for crop models. European Journal of Agronomy 18 (1-2): 75-105. Grski T., G. Demidowicz, T. Deputat, K. Grska, I.Marcinkowaska, W. Spoz-Pa, 1997. Empiryczny model plonowania pszenicy ozimej w funkcji czynnikw meteorologicznych. (Empirical model of winter wheat yield in the function of meteorological parameters ). Zesz. Nauk. AR Wrocaw, 313, 99-144. Grski T., W. Spoz-Pa, 1989. A method for forecasting sugar beet yield in Poland. Zesz. Probl. Post. Nauk. Rol. 369, 31-43.
99
Groot J.R.R., 1987. Simulation of nitrogen balance in a system of winter wheatand soil. Simulation Reports CABO-TT 13. CABO-DLO, WAU-TPE, Wageningen. Guerea A., M. Ruiz-Ramos, C.H. Diaz-Ambrona, J.R. Conde, M.I. Minguez, 2001. Assessment of climate change and agriculture in Spain using climate models, Agronomy Journal 93:237-249. Gurif M. and Duke C.L., 1998. Calibration of the SUCROS emergence and early growth module for sugar beet using optical remote sensing data assimilation. European Journal of Agronomy 9, 127-136. Gurif M. and Duke C.L. 2000. Adjustment procedures of a crop model to the site specific characteristics of soil and crop using remote sensing data assimilation. Agriculture, Ecosystems and Environment 81, 57-69. Hadria R., B. Duchemin, A. Lahrouni, S. Khabba, S. Er-Raki, G. Dedieu, A. Chehbouni 2006. Monitoring of irrigated wheat in a semi-arid climate using crop modelling and remote sensing data: Impact of satellite revisit time frequency, International Journal of Remote Sensing, 27: 1093-1117. Han S., J.W. Hummel, C.E. Goering, M.D. Cahn 1995. Cell size selection for site-specific crop management. Trans. ASAE 37 (1), 1926. Harrison P., R. Butterfield, T. Downing (eds.), 1995. Climate Change and Agriculture in Europe - Assessment of Impacts and Adaptation. University of Oxford, UK, 411 pp. Harrison P.A., J.R. Porter, T.E. Downing, 2000. Scaling-up the AFRCWHEAT2 model to assess phenological development for wheat in Europe, Agricultural and Forest Meteorology, Volume 101, P:167186, 2000. Hartkamp D.A., J.W. White, G. Hoogenboom, 1999. Interfacing geographic information systems with agronomic modeling: A review. Agron. J. 91, 761-772. Heinemann A.B., G. Hoogenboom, R.T. de Faria, 2002. Determination of spatial water requirements at county and regional levels using crop models and GIS An example for the State of Parana, Brazil. Agricultural Water Management 52, 177-196. Herrmann A., K.C. Kersebaum,. F. Taube, 2005. Nitrogen fluxes in silage maize production: relationship between nitrogen content at silage maturity and nitrate concentration in soil leachate. - Nutrient Cycling in Agroecosystems 73 (1): 59-74. Herrmann A., K.C. Kersebaum, M. Wachendorf, F. Taube, 2004. Simulation of nitrogen dynamics in forage maize using the HERMES model. - In: Land Use Systems in Grassland Dominated Regions : Proceedings of the 20th General Meeting of the European Grassland Federation, Luzern, Switzerland, 21 - 24 June 2004:. vdf Hochschulverlag, Zrich. 735-737. Hitrec I., E. Halupecki, L. Biak, B. Cvjetkovi, 2000. Internet based Fire blight forecasting system, Proceedings of the 2nd Croatian Academic and Research Network (CARNet) Users Conference CUC 2000 (CD edition), Zagreb, Croatia, September 24-26, 2000. Hofmeister F., A.H. Arnoldussen, O. Klakegg, 2001. Production of Land Resources maps for Barley for South East Norway with the plant production modell WOFOST, NIJOS report 09/ 2001 8 in norwgian), ( Produksjonsgrunnlag for bygg p stlandet basert p simuleringer med plantevekstmodellen WOFOST. Holden N.M, A.J. Brereton, J.B. Fitzgerlad, 2008. Impact of Climate Change on Irish Agricultural Production Systems. In J. Sweeney (editor). Climate Change: Refining the Impacts (provisional title). ERTDI Report. Environmental Protection Agency, Johnstown Castle, Wexford. P 33-79. (in press) Holden N.M, A.J. Brereton, R. Fealy, J. Sweeney 2003. Possible change in Irish climate and its impact on barley and potato yields. Agricultural and Forest Meteorology 116: 181196.
100
Holden N.M, J. Sweeney, A.J. Brereton, R. Fealy 2004. Climate change and Irish agriculture. In T. Keane and J. F. Collins (editors). Climate Weather and Irish Agriculture (2nd Edition). Agmet, Dublin. Holden N.M. and Brereton A.J., 2002. An assessment of the potential impact of climate change on grass yield in Ireland over the next 100 years. Irish Journal of Agricultural and Food Research 41: 213-226. Holden N.M. and Brereton A. J., 2003a. Climate change and the introduction of maize and soybean to Ireland. Irish Journal of Agricultural and Food Research 42: 1-15 Holden N.M. and Brereton A.J. 2003b. The Impact of Climate Change on Irish Agriculture In: J. Sweeney (editor): Climate Change Scenarios and Impacts (2000-LS-5.2.1-M1) Final Report. ERTDI Report Series No. 15. Environmental Protection Agency, Johnstown Castle, Wexford. P 33-79. Holden N. M., A. J. Brereton 2006. Adaptation of water and nitrogen management of spring barley and potato as a response to possible climate change in Ireland. Agricultural Water Management 82: 297-317. Hoppmann D., B. Berkelmann Lhnertz, 2000. Prognosis of phenological stages of Vitis vinifera (cv. Riesling) for optimizing pest management., Bulletin OEPP/EPPO (30) 121-126. Hutchinson M.F., H.A. Nix, J.P. McMahon, 1992. Climate constraints on cropping systems. pp 37-58 in: C.J. Pearson (ed.) Field Crop Systems. Vol. 18 of D.W. Goodall (ed.) Ecosystems of the World. Elsevier, Amsterdam. Huygen J., 1992. SWACROP2, a quasi-two-dimensional crop growth & soil water flow simulation model. User's guide. WAU, Department of Water Resources, DLO Winand Staring Centre, Wageningen. Iglesias A. and Mnguez M.I., 1995. Perspectives for maize production in Spain under climate change, in: Rosenzweig C., Allen Jr., L.H., Harper, L.A., Hollinger, S.E., Jones, J.W. (ed) Climate Change and Agriculture: Analysis of Potential International Impacts, Vol 13. American Society of Agronomy, Madison, Wisconsin, p 259-273. Iglesias A., C. Rosenzweig, D. Pereira, 2000. Agricultural impacts of climate change in Spain: developing tools for a spatial analysis, Global Environmental Change-Human and Policy Dimensions 10:69-80. Iglesias I., F.J. Rodriguez-Rajo, J. Mendez, 2007. Evaluation of the different Alternaria prediction models on a potato crop in A Limia (NW of Spain), AEROBIOLOGIA, 23 (1): 27-34. Iglesias A., 2006. Use of DSSAT models for Climate Change impact ans assessment: calibration and validation. CGE Hands-on Training Workshop on V&A Assessment. Jakarta 20-24 March 2006. Janssen P.H.M., 1994. Assessing sensitivities and uncertainties in models: a critical evaluation. In : Grasman, J. & G. van Straten (Eds.). Predictability and nonlinear modeling in natural sciences and economics.- Kluwer, Dordrecht, 344-361. Jamieson P.D., J.R. Porter, J. Goudrian, J.T. Ritchie, H. van Keulen, W. Stol, 1998b. A comparison of the models AFRCWHEAT2, CERES-Wheat, Sirius, SUCROS2 and SWHEAT with measurements from wheat grown under drought. Field Crops Research, 55: 23-44. Jamieson, P.D., M.A. Semenov, I.R. Brooking and G.S. Francis, (1998a): Sirius: a mechanistic model of wheat response to environmental variation. European Journal of Agronomy, 8: 161-179. Jones J.W., B.A. Keating, C.H. Porter 2001. Approaches to modular model development. Agricultural System 70, 421-443.
101
Jones J.W., G. Hoogenboom, C.H. Porter, K.J. Boote, W.D. Batchelor, L.A. Hunt, P.W. Wilkens, U. Singh, A.J. Gijsman, J.T. Ritchie, 2003. The DSSAT cropping system model, European Journal Of Agronomy, Volume 18 (3-4), P: 235-265. Jongschaap R.E.E., 2006. Run-time calibration of simulation models by integrating remote sensing estimates of leaf area index and canopy nitrogen. Europ. J. Agronomy 24, 316324. Jongschaap R.E.E. and Schouten L.S.M., 2005. Predicting wheat production at regional scale by integration of remote sensing data with a simulation model, Agronomy Sustainable Development., Volume 25, P: 481- 489, 2005. Kajfe-Bogataj L., 1996. Effects of climate warning on ceres-maize yield in Slovenia: sensitivity study. Zbornik biotehnike fakultete universe v Ljubljani. Kmetijstvo. Agricultural issue. t. 67, str. 11-18. Kajfe-Bogataj L. and Sunik A., 2002. Operativni agrometeoroloki modeli za izraun vodne balance kmetijskih tal (Comparison of three operational irrigation models). Novi izzivi v poljedelstvu 2002. Str. 164-169. Kanellou E., E. Tsiros, C. Domenikiotis, N.R. Dalezios, 2008. Meteorological and Agrohydrological drought monitoring based on conventional and remotely sensed data. European Geosciences Union General Assembly 2008, Vienna, Austria, 13 18 April 2008 (accepted). Kantylis P.I., P.T. Nastos, D.I. Nikolakakis, 2002. Contribution to the Study of the Frost Days in Greece. Proceedings of 6th Hellenic conference in Meteorology Climatology and Atmospheric Physics, Vol 1, Ioannina, 25-28 September 2002, pp. 91-99. Kapler P., M. Trnka, D. Semeradova, M. Dubrovsky, M. Zalud, M. Svoboda, J. Eitzinger, J. Hsch, M. Mozny, 2006. Newhall Model for Assessment of Agricultural Drought Event Probability under Present and Changed Climatic Conditions. In: EGU 2006, Vienna, April 2-7 2006. Geophysical Research Abstract Volume 8, ISSN 1029-7006. Kapsa J., 2007. Wpyw warunkw klimatycznych na zmienn presj infekcyjn Phytophora Infestans i jej szodliwo dla plantacji ziemniaka. (The impact of climatic conditions on varied Phytophora Infestans pressure and its harmfulness at potato crops). XLVII Sesja Naukowa Instytutu Ochrony Rolin , Pozna 15-16 2007. 97-98. Karakostas T.S., P. Pennas, G. Zerva, 2002. Studying the Effects of Extreme Climatic Events on Wheet. Proceedings of 6th Hellenic conference in Meteorology Climatology and Atmospheric Physics, Vol 2, Ioannina, 25-28 September 2002, pp. 621-629. Kartschall T.H., S. Grossman, P.J. Pinter, J.R. Garcia, R. L., Kimball, B. A., Wall, G. W., Hunsaker, D. J., R.L. LaMorte, 1995. A Simulation of Phenology, Growth, Carbon Dioxide Exchange and Yields under Ambient Atmosphere and Free-Air Carbon Dioxide Enrichment (FACE) Maricopa, AZ, for Wheat, Blackwell Science Ltd, Journal of Biogeography, 611-622. Karvonen T. and Kleemola, J., 1995. CROPWATN: Production of water and nitrogen limited crop production. In: B. Kabat et al. (eds.). Modelling and Parameterization of the SoilPlant-Atmosphere System: A Comparison of Potato Growth Models. Wageningen pers, Wageningen, The Netherlands, pp. 335-369. Kaukoranta T., 1996. Impact of global warming on potato late blight: risk, yield loss and control. Agricultural and Food Science in Finland 5: 311-327. Kazadjiev V.,M. Moteva,V. Georgieva, 2007. Verification of WOFOST Model for Soil and Meteorological Conditions of Bulgaria, EMS7/ECAM8 Abstracts, Volume 4, EMS 2007-A-00107, 7th EMS Annual Meeting / 8th ECAM, 2007. Kazandjiev V., V. Georgieva 2005a. Extremal meteorological conditions and production of autumn cereal crops in Bulgaria, Proceedings from First Scietific and Practical
102
Conference Management in extraordinary conditions and defence population, Sofia BAS, p. 106-115. Kazandjiev V. and Georgieva V., 2005b. WOFOST Model calibration and Forecast of soybean (Glycine hispida (Moench) yields in Bulgaria, Proceedings of Scientific Conference 80 Years Soybean Institute, Pavlikeni, p. 244-253. Kersebaum K.C., A. Wurbs, R. de Jong, C.A. Campbell, J. Yang, R.P. Zentner, 2008. Longterm simulation of soil-crop interactions in semiarid southwestern Saskatchewan, Canada. Eur. J. Agronomy 29: 1-12Kersebaum K.C., 2000. Model based evaluation of land use and management strategies in a nitrate polluted drinking water catchment in North-Germany. In: R. Lal (ed.): Integrated Watershed Management in the Global Environment. CRC Press, Boca Raton. 223-238. Kersebaum K. C., H. I. Reuter, K. Lorenz, O. Wendroth 2002. Modelling crop growth and nitrogen dynamics for advisory purposes regarding spatial variability. In: L. Ahuja, L. Ma & T. Howell (eds.): Agricultural system models in field research and technology transfer. Lewis Publishers, Boca Raton, 229 - 252. Kersebaum K.C., 2007. Modelling nitrogen dynamics in soilcrop systems with HERMES. Nutrient Cycling in Agroecosystems 77 (1), 39-52. Kersebaum K.C., K. Lorenz, H. I. Reuter, J. Schwarz, M. Wegehenkel, O. Wendroth, 2005. Operational use of agro-meteorological data and GIS to derive site specific nitrogen fertilizer recommendations based on the simulation of soil and crop growth processes. Physics and Chemistry of the Earth 30 (1-3): 59-67 Kersebaum K.C., H.I. Reuter, K. Lorenz, O. Wendroth, 2005. Long term simulation of soil/crop interactions to estimate management zones and consequences for site specific nitrogen management considering water protection. - In: Stafford, J. V. [Hrsg.]: Precision agriculture 05: [peer-reviewed papers presented at the 5th European Conference on Precision Agriculture]: 795-802; Wageningen (Academic Publishers). King D., C. Le Bas, A.J. Daroussin, A.J. Thomasson, R.J.A. Jones 1995. The EU map of soil water available for pants. In King D., Jones R.J.A., Thomasson A.J. (Eds.): European Land Information System for Agro-environmental Monitoring. European Commission, Luxemburg, 131-141. Knapi V., K. Beber, G. Seljak, V. kerlavaj, S. Tome, 1999. Using automatic meteorological stations for early pest warning and irrigation prediction. Lectures and Papers presented at the 4 th Slovenian Conference on Plant protection in Portoro, March 4-5, 1999,67-72. Koleva E. and Alexandrov V., 2007. Drought in Bulgarian low regions during the 20th century. Theoretical and Applied Climatology Kroes J.G. and Roelsma J., 2007. Simulation of water and nitrogen flows on field scale: application of the SWAP-ANIMO model for the Mncheberg data set. In: K.C. Kersebaum, J.M. Hecker, W. Mirschel, M. Wegehenkel (eds.): Modelling water and nutrient dynamics in soil-crop systems. Springer, Dordrecht. 111 128. Kroes J.G., J.C. van Dam, (eds) 2003. Reference Manual SWAP version 3.0.4. Wageningen, Alterra, Green World Research.. Alterra-Report 773. 211 pp, Wageningen, The Netherlands Kosmas C., N. Danalatos, E. Ntzanis, N. Yassoglou, 1996. The application of pedotransfer functions in predicting ground water recharge at regional scale. In: The Use of Pedotransfer in Soil Hydrology Research in Europe. Proceedings of the Second Workshop, INRA Orleans (France), 10-12/10/1996, p. 111-119. Kotroni V. and Lagouvardos K., 2002. Fune Grid Weather Forecasts over Attica, Greece: Results of one year Operational Use. Proceedings of 6th Hellenic conference in
103
Meteorology Climatology and Atmospheric Physics, Ioannina, 25-28 September 2002, Vol 1, pp. 132-137. Kromp-Kolb H., J. Eitzinger, H. Formayer, M. Trnka, W. Laube, G. Gruszczynski, 2005. Bestimmung der Auswirkungen von Trockenperioden im Grnland mittels Wachstumsmodellen und klimatologische Analyse sterreichischer Daten (Estimation of impacts of drought periods on grassland by growth models and climatological analysis of Austrian data). Endbericht. Im Auftrag d. Bundesministeriums fr Landund Forstwirtschaft, Umwelt und Wasserwirtschaft, Forschungsprojekt Nr. 1282. Krysanova V., F. Wechsung, A. Becker, W. Poschenrieder, J. Grfe, 1999. Mesoscale ecohydrological modeling to analyze regional effects of climate change. Environmental Modeling & Assessment, 4, 259-271. Lagacherie P., D.R. Cazemier, R. Martin-Clouaire, T. Wassenaar, 2000. A spatial approach using imprecise soil data for modelling crop yields over vast areas Source. Agric. Ecosys. and Env., Vol. 81/1, 5-16. Lali B., D.T. Mihailovic, J. Balaz, I. Koci, 2003. Prediction of occurrence of plant diseases using the coupled atmosphere-land surface models, Sixt European Conference on Applications of Meteorology, September 15-19, Rome (Italy), Abstracts, p.13. Lalic B., L. Pankovic, D.T. Mihailovic, 2006. Calibration of SIRIUS Model for Different NSsorts of Small Grains, Abstracts of Sixth Annual Meeting of the European Meteorological Society (EMS), September 4-8, Ljubljana (Slovenija). Lalic B., D.T. Mihailovic, S. Radovanovic, J. Balaz, , A. Cirisan, 2007. Input Dara representativeness Problem in Plant Disease Forecasting Models, Idojaras (In press). Launay M. and Guerif M., 2005. Assimilating remote sensing data into a crop model to improve predictive performance for spatial applications, Journal of Agriculture, Ecosystems and Environment, Volume 111,P: 321339, 2005. Laurila H., 1995. Modelling the effects of elevated CO2 and temperature on Swedish and German spring wheat varieties with CERES-Wheat and AFRC-Wheat crop models. Journal of Biogeography 22: 591595. Lawless C., M.A. Semenov, 2005. Assessing lead-time for predicting wheat growth using a crop simulation model, Journal of Agricultural and Forest Meteorology, Volume 135, P: 302313. Maas S.J., 1988. Use of remotely sensed information in agricultural crop growth models. Ecol. Modelling 41, 247-268. Majercak J. and Novak V., 1995. GLOBAL - a numerical model for water movement in the soil root zone.- Research Report, Institute of Hydrology, Slovak Academy of Sciences, Bratislava. Maracchi G., V. Perarnaoud, A.D. Kleschenko, 2000. Applications of geographic information systems and remote sensing in agrometeorology. Agric.For.Meteorol. 103, 119-136. Marica A., A. Busuioc, R. Bojariu, C. Boroneant, 2005b. Application of climatic predictions and simulation models for the benefit of agriculture in Romania, WMO Expert Team Meeting on Impact of Climate Change/Variability on Medium- to Long-Range Predictions for Agriculture, 15-18 February 2005, Brisbane, Australia. Marica A., 2005. Applications of the CROPWAT model for irrigation planning and management in Romania, In Irrigation and Pest and Disease Models: Evaluation in Different Environments and Web-Based Applications, Edited by G. Maracchi, L. Kajfez Bogataj, S. Orlandini, F. Rossi, M Barazutti, European Commission, Directorate General XII, Science, Research and Development, Environment research programmeCOST Action 718, pp 73-79;
104
Marica A., A. Busuioc, 2004. The potential impacts of climate change on the main components of water balance relating to maize crop, Romanian Journal of Meteorology, Vol.6, No.1-2, Bucharest, Romania. Marica A., G. Stancalie, D. Alexandru, A. Nertan, C. Flueraru, 2005a. Application of GIbased procedures for soil moisture mapping and crop vegetation status monitoring in Romania, WMO, FAO & COST 718 Workshop on Climatic Analysis and Mapping for Agriculture, 14-17 June 2005, Bologna, Italy. Marn C., 1999. Modelos agrometeorolgicos de prediccin del rendimiento de cosechas (la vid). Tesis Doctoral. E.T.S.I. Agrnomos. (Agrometeorological models for the prediction of crop yield (grape)). Marletto V., and Zinoni F., 1998. The Criteria project: integration of satellite, radar, and traditional agroclimatic data in a GIS-supported water balance modelling environment. In: EUR 18328, Dalezios N.R. (ed.), 1998. Proc. COST 77, 79, 711 Int. Symp. on Applied Agrometeorology and Agroclimatology, Volos, Grecia, 24-26 april 1996, ISBN 92-828-4137-5, 173-178. Marletto V., F. Zinoni, L. Criscuolo, G. Fontana, S. Marchesi, A. Morgillo, M. Van Soetendael, E. Ceotto, U. Andersen, 2005. Evaluation of downscaled DEMETER multimodel ensemble seasonal hindcasts in a northern Italy location by means of a model of wheat growth and soil water balance. Tellus A, 57(3), 488-497 Marletto V., L. Criscuolo, M. Van Soetendael, 2001. Implementation of Wofost in the framework of the Criteria geographical tool. Proc. of the 2nd Int. Symp. Modelling Cropping Systems, European Society for Agronomy, Florence, 16-18 July 2001, 219220. Marletto V., F. Ventura, F. Tomei, G. Fontana, 2007. Wheat growth simulation and yield prediction with seasonal forecasts and a numerical model, Agricultural and Forest Meteorology, Volume 147, P: 7179. Mastrorilli M., N. Katerji, B.B. Nouna, 2003. Using the CERES-Maize Model in a Semi-arid Mediterranean Environment. Validation of Three Revised Versions, European Journal of Agronomy, Volume 19, P: 125- 134, 2003. Matthews K.B., A.R. Sibbald, S. Craw, 1999. Implementation of a spatial decision support system for rural land use planning: integrating geographic information system and environmental models with search and optimisation algorithms. Comput. Electron. Agric. 23(1):9-26. Matthews R.B., 2006. The People and Landscape Model (PALM): towards full integration of human decision-making and biophysical simulation models. Ecol. Modelling 194(4):329-343. Matthews R.B. and Pilbeam C.J., 2005. Modelling the long-term sustainability of maize/millet cropping systems in the mid-hills of Nepal. Agriculture, Ecosystems & Environment 111(1-4):119-139. Matthews R.B., J.G. Polhill, N. Gilbert, A. Roach, 2005. Integrating agent-based social models and biophysical models. In: A. Zerger & R.M. Argent (Editors), Proceedings of MODSIM 2005 International Congress on Modelling and Simulation, Dec 9-12, 2005. Modelling and Simulation Society of Australia and New Zealand, Melbourne, Australia, pp. 1617-1623. Migdakos E., and Gemtos T., 1998. Relationship between Growing Degree Days and phenological stages of cotton growth in Greece-A case study of Karditsa Perfecture. Proceedings of International Symposium on Applied Agrometeorology and Agroclimatology, 24-26 April, Volos, 425-430.
105
Mihailovi D.T, J. Eitzinger, I. Koi, B. Lali, I. D. Arseni, D. Radlovi, V. Mircov, 2000. Predicting the occurrence of Erwinia amylovora using the biometeorological system BAHUS. Proceedings of Annual Meeting of Review Scientific Works of Banatian University of Agricultural Sciences and Veterinary Medicine, 12-13 December, Timisoara (Romania), 45-51. Mihailovi D.T., I. Koi, B. Lali, I. Arseni, D. Radlovi, J. Bala, 2001. The main features of BAHUS - biometeorological system for messages on the occurrence of diseases in fruits and vines. Environmental Modelling and Software., 16, 691-696. Mihailovi D.T., J. Eitzinger, I. Koi, B. Lali, J. I. Arseni, J. Bala, 2002. Biometeorological system BAHUS for predicting the occurrence of plant diseases and ensuring their efficient control. International Workshop on Environmental Risk Assessment of Pesticides and Integrated Pesticide Management in Developing Countries, Kathmandu, Nepal, 6-9 November, Landschaftsokologie und Umweltforschung, 38, 120-129. Mijukovi D., Delevi S., Perovi T., 2002: Spring frosts in the period of fruit flowering, 18th International Conference on Carpathian Meteorology: Mountain influence on weather, 7-11 October, Belgrade, Serbia. Milievi T., D. Ivi, B. Cvjetkovi, 2006. Application of BOTMAN Forecast Model in Integrated Botrytis Managment in Croatian Strawberry Plantations, Proceedings of the 12th Congress of the Mediterranean Phytopathological Union, Rhodes Island, Greece, 330-332. Mirschel W., H. Frkel, U. Franko, 2002. Modulares dynamisches Wachstumsmodell fr Zuckerrben als integrativer Bestandteil von komplexen agrarkologischen Simulationsmodellen. In: GNAUCK, A. (Hrsg.): Systemtheorie und Modellierung von kosystemen. Umweltwissenschaften, (UmweltWissenschaften), Physika-Verlag (Unternehmen des Springer-Verlages) Heidelberg, 2002, 136 156. Mirschel W., K.O. Wenkel, A. Schultz, J. Pommerening, G. Verch, 2005. Dynamic phenological model for winter rye and winter barley, European Journal of Agronomy, Volume 23, P: 123- 135. Mirschel W., A. Schultz, , K.O.Wenkel, 2001. Assesing the Impact of Land Use Intensity and Climate Change on Ontogenesis, Biomass Production, and Yield of Northeast German Agrolandscapes. In: TENHUNEN, J.D.; LENZ, R.; HANTSCHEL (Eds.): Ecosystem Approaches toLandscape Management in Central Europe. Ecological Studies, Vol. 147, Springer-Verlag Berlin Heidelberg, New York, 299-313. Moriondo M., F. Maselli, M. Bindi, 2007. A simple model of regional wheat yield based on NDVI data, European Journal of Agronomy, Volume 26, Issue 3, P: 266-274, April 2007. Moulin S., A. Bondeau, R. Delcolle, 1998. Combining agricultural crop models and satellite observations: from field to regional scales. Int. J. Remote Sens. 19 (6), 1021-1036. Mller B., E. Schmid, P. Liebhard, K. Eschelbck, 2006. Evaluation of alternative management measures in vegetable production systems by field measurements and EPIC simulations. In: A. elkov, F. Matejka (Eds.), 14th International Poster Day, Transport of Water, Chemicals and Energy in the Soil-Plant-Atmosphere System, 329334, 14th International Poster Day, Transport of Water, Chemicals and Energy in the Soil-Plant-Atmosphere System, 25.11.2006, Bratislava; ISBN: 8085754150. Musa-Steenblock T.and Forrer H.R., 2005. Bio-PhytoPRE ein Warn- und Prognosesystem zur Bekmpfung der Kraut- und Knollenfule im kologischen Kartoffelanbau in der Schweiz. In: Ende der Nische, Beitrge zur 8. Wissenschaftstagung kologischer Landbau, Kassel 1.- 4. Mrz 2005, Hrsg.: J. Hess und G. Rahmann, S. 133-136
106
Nejedlik P. and Diepen C.A. van, 1999. Effects of Data Resolution on Estimation of Regional Crop Yields in Western Slovakia with the WOFOST Crop Simulation Model, Poster Session. In: Geophysical Research Abstracts Volume1, Number 2, 1999. European Geophysical Society, ISSN 129-7006, Aberystwyth SY23. Nejedlik P. and Nieplova E., 1998. Preparation of MERA Database and the Test of the WOFOST Crop Growth Simulation Model. In: Proceedings from 2nd Slovak - Swiss Workshop, Zurich/Payerne, March 1998. Nemnyi M., P.A. Mesterhazi, Z. Pecze, Z. Stepan 2003. The role of GIS and GPS in precision farming. Computers and Electronics in Agriculture 40 (1), 45-55. Nendel C., K.C. Kersebaum, 2004a. A simple model approach to simulate nitrogen dynamics in vineyard soils. - Ecological Modelling 177 (1-2): 1-15 Nendel C., S. Reuter, K.C. Kersebaum, R. Kubiak, R. Nieder, 2005. Nitrogen mineralization from mature bio-waste compost in vineyard soils: II. Test of N-mineralization parameters in a long-term in situ incubation experiment. - Journal of Plant Nutrition and Soil Science 168 (2): 219-227. Nendel C.; S. Reuter, K.C. Kersebaum, R. Nieder, R. Kubiak, 2004b. Modelling nitrogen mineralisation from mature bio-waste compost applied to vineyard soils. - In: Controlling nitrogen flows and losses: 207-208; Wageningen (Wageningen Acad. Publ.). Nieuwenhuis G.J.A., A.J.W de Wit, D.W.G. van Kraalingen, C.A. van Diepen, H.L. Boogaard, 2006. Monitoring crop growth conditions using the global water satisfaction index and remote sensing, ISPRS Commission VII Mid-term Symposium Remote Sensing: From Pixels to Processes, Netherlands, P: 684- 687, 8-11 May 2006. Nonhebel S., 1993. The importance of weather data in crop growth simulation models and assessment of climate change effects, Ph.D. Thesis, Wageningen Agricultural University. Olesen J.E., P.K. Bcher, T. Jensen, 2000. Comparison of scales of climate and soil data for aggregating simulated yields of winter wheat. Agriculture, Ecosystems and Environment 82, 213-228. Olesen J.E., T.Jensen, J. Petersen, 2000. Sensitivity of field-scale winter wheat production in Denmark to climate variability and climate change. Climate Research 15, 221-238. Olesen J.E., T. Jensen, J. Petersen, 2000. Sensitivity of field-scale winter wheat production in Denmark to climate variability and climate change. Climate Research 15, 221-238. Olesen J.E. 2005. Climate change and CO2 effects on productivity of Danish agricultural systems. Journal of Crop Improvement 13, 257-274. Olesen J.E., G. Rubk, T. Heidmann, S. Hansen, C.D. Brgesen, 2004. Effect of climate change on greenhouse gas emission from arable crop rotations. Nutrient Cycling in Agroecosystems 70, 147-160. Olesen J.E., T.R. Carter, C.H. Diaz-Ambrona, S. Fronzek, T. Heidmann, T. Hickler, T. Holt, M.I. Minguez, P. Morales, J. Palutikof, M. Quemada, , M. Ruiz-Ramos, G. Rubk, F. Sau, Smith, B. & Sykes, M. 2007. Uncertainties in projected impacts of climate change on European agriculture and ecosystems based on scenarios from regional climate models. Climatic Change 81, 123-143. Orlandini S., A. Dalla Marta, H. Fiesland, A. Sunik, 2005. Description and testing of the model PERO for grapevine downy mildew predictions. Irrigation and pest and disease models: evaluation in different environments and web-based applications. Maracchi G., Kajfe-Bogataj L. (ur). COST Action 718 Meteorological Applications for Agriculture: 127 139.
107
Orlandini S., A. Dalla Marta, H. Fiesland, A. Sunik, 2005. Description and testing of the PLASMO model for the simulation of grapevine downy mildew. Irrigation and pest and disease models: evaluation in different environments and web-based applications. Maracchi G., Kajfe-Bogataj L. (ur). COST Action 718 meteorological Applications for agriculture: 140-159. Osborne T.O., D.M. Lawrence, A.J. Challinor, J.M. Slingo, T.R. Wheeler, 2007. Development and assessment of a coupled crop-climate model. Global Change Biology 13:169-183. Ottl P.C., J.C Calvet,. N. Brisson 2004. Future directions for advanced evapotranspiration modeling: assimilation of remote sensing data into crop simulation models and SVAT models. Irrigation and Drainage Systems. Pakalidou N., Th. Karakostas, T. Makrogiannis, Th. Mavrommatis, 2006. Time Projection of Heatwave Events in Thessaloniki. 8th Conference on Meteorology Climatology and Atmospheric Physics, Athens, 24-26 May 2006, Volume of Abstracts,74p. Pankovic L., V. Pankovic, B. Lalic, 2006. Computing Model for Cereal Leaf Beetle, Abstracts of Sixth Annual Meeting of the European Meteorological Society (EMS), September 48, Ljubljana (Slovenija). Parton W.J., J.W.B. Stewart, C.V. Cole, 1988. Dynamics of C, N, P and S in grassland soils: a model. Biogeochemistry 5:109-131. Penning de Vries F.W.T, D.M. Jansen, H.F.M ten Berge, A.Bakema, 1989. Simulation of ecophysiological processes of growth in several annual crops. Pudoc, Wageningen, 271 pp. Plauborg, F., Olesen, J.E., 1991. Udvikling og validering af modellen MARKVAND til vandingsstyring i landbruget (Development and validation of the model MARKVAND for irrigation scheduling in agriculture). Tidsskrift for Planteavls Specialserie S2113. Statens Planteavlsforsg. Poels R.L.H. and Bijker W., 1993. TROPFOR, a computer program to simulate growth and water use of tropical rain forests developed from the "WOFOST program. WAU, Department of Soil Science and Geology. Porter J.H., M.L. Parry, T.R. Carter, 1991. The potential effects of climatic change on agricultural insect pests. Agric. For. Meteorol. 57: 221-240. Prvot L., H. Chauki, D. Troufleau, M. Weiss, F. Baret, N. Brisson, 2003. Assimilating optical and radar data into the STICS crop model for wheat, Journal of Agronomie, Volume 23, P: 297303, 2003. Priesack E., S. Gayler, H.P. Hartmann, 2007. The impact of crop growth sub-model choice on simulated water and nitrogen balances.- In : Modelling water and nutrient dynamics in soil-crop systems, Proceedings, ed.: Kersebaum, K.H., Hecher J-M., Mirschel, W., Wegehenkel, M.; Springer, p. 27-36. Rembold F., J. Delinc, H. Boogard, A. Burger, 2006. Spatial Information Systems in Crop Monitoring Developing New Global Models and Sharing the Data, GSDI-9 Conference Proceedings, Chile, 6-10 November 2006. Richardson A.J., C.L. Wiegand, G.F. Arkin, P.R. Nixon, A.H. Gerbermann 1982. Remotely sensed spectral indicators of sorghum development and their use in growth modelling. Agric. Meteorol. 26, 11-23. Richter GM., 1999. Variability of winter rye grain yield in a glacial plain catchment modelling and observation. Europ. J. of Agronomy. Vol.11 /3-4, 239-253. Richter G.M. and Semenov M.A., 2005. Modelling impacts of climate change on wheat yields in England and Wales: assessing.drought risks. Agricultural Systems 84(1):77-97.
108
Riedo M., A. Grub, M. Rosset, J. Fuhrer, 1998. A pasture simulation model for dry matter production, and fluxes of carbon, nitrogen, water and energy. Ecological Modelling 105: 141183 Rinker A., F. Deunert, W. Schrder, 2007. Changing P and N turnover in soils due to increasing air temperatures in Germany? A pilot study coupling the GIS-based modelling system WASMOD with REMO modelling IPCC scenarios. In: Ecological complexity and sustainability. Abstracts of EcoSummit 2007, Beijing, China, p. 7 Ruget F., S. Novak, S. Granger, 2006. Du modle STICS au systme ISOP pour estimer la productivit fourragre. Adaptation la prairie, application spatialise (Use of the ISOP system, based on the STICS model, fort he assessment of forage production. Adaptation to grassland and spatialized application), Fourrages, 186, 241-256 Ruget F., 2006. Introduction to the crop model STICS, use for estimating crop production, consumption and losses, in the proceedings of the Summer university on IT in agriculture and rural developement , Deabrecen (Hungary), 21-22 august 2006, 7-13 Ruijter F.J. de, W.A.H. Rossing, J. Schans, 1993. Simulatie van opbrengstvorming bij tulp met WOFOST. Simulation Reports CABO-TT 33. CABO-DLO, WAU-TPE, Wageningen. Saarikko R.A. 2000. Applying a site based crop model to estimate regional yields under current and changed climates, Journal of Ecological Modelling, Volume 131, P: 191206, 2000. Snchez J.A., F. Cnovas, A. Lacasa, 2005. Un sistema de informacin geogrfica (SIG) para el manejo de enemigos naturales, plagas y enfermedades en los cultivo de pimiento del Campo de Cartagena (Murcia). Agroinformacin.com (in spanish) Schaldach R., J. Alcamo, 2006. Coupled simulation of regional land use change and soil carbon sequestration: A case study for the state of Hesse in Germany. Environ. Model. and Software 21, 1430-1446. Schaumberger A., M. Trnka, J. Eitzinger, H. Formayer, 2005. Implementierung eines Bewertungsmodells fr Trockenschden in einem Geographischen Informationssystem (GIS), (Implementation of a model for drought damage estimation in a GIS). Mitteilungen der Gesellschaft fr Pflanzenbauwissenschaften, 17, Wasser und Pflanzenbau Herausforderungen fr zuknftige Produktionssysteme, 277-278, Kurzfassungen d. 48.Jahrestagg., 27-29 Sept., Wien, ISBN 3-935380-16-X; (http://dnb.ddb.de/). Schmid E., B. Stuermer, B. Mueller, J. Balkovic, 2006. Food and non-food crop management systems for EPIC modeling in EU25. Deliverable for the European Non-food Agriculture (ENFA) project. University of Hamburg, Germany. Europische Kommission Brussels Europische Union, pp 23. Schneider K. and Mauser W. 2000. "Using Remote Sensing Data to Model Water, Carbon and Nitrogen Fluxes with PROMET-V", Remote Sensing for Agriculture, Ecosystems and Hydrology, SPIE Vol. 4171, pp.12-23. Seguin B. and Garca de Cortzar Atauri I., 2004. Climate warning: consequences for viticulture and the notion of terroirs in Europe. Proceedings of The Seventh International Symposium of Vineyard Physiology and Biotechnology. 21-25 juin Davis (USA), Acta Horticulturae, 689: 61-70. Semenov M.A. and Porter J.R., 1995. Climatic variability and the modelling of crop yields.Agricultural and Forest Meteorology 73, Nos. 3-4, 265-284. Sirotenko O. and Abashina E., 1994. Influence of global warming on agroclimatic resources and productivity of agriculture in Russia. Meteorology and Hydrology, No. 4, Gidrometeoizdat, pp. 101 - 112. (In Russian).
109
Sirotenko O., 1983. Development and applications of dynamic simulation models in agrometeorology. CAgM Reprt No.13, WMO, Geneva. ika B. and Samuhel P., 2007. Climate Change Impact on maize Yields in Conditions of Danubian Lowland. Meteorological Journal 1, Bratislava, 2007. Siverten T.H., 2006. Quality considerations on meteorological parameters to be used for modelling UV-radiation, Proceedings of SPIE 11-14 September 2006, Remote Sensing of Clouds in the Atmosphere XI, Volume 6362, The Internation Society of Optical Engineering. Sivertsen T.H, 2005. Reflections on the Theme of Classifying, Documenting and Exchanging Meteorological Data, Atmospheric Science Letters, Vol.6 , Issue 3, p 171-175, John Wiley & Sons. Sivertsen T.H. and Gailis J., 2007, Discussing a web-based system for administration, evaluation, and correction of meteorological and biological datain perspective of actornetwork theory, Idojaras, Quarterly journal of the Hungarian Metterological Service, vol. III, No.2-3 April-september 2007. Segaard K., J. Berntsen, K.A. Nielsen, I. Thysen, 2005. Forecast of herbage production under continuous grazing. In: Utilization of grazed grass in temperate animal systems. Proceedings of satellite workshop of the XXth International Grassland Conference, p. 218. Stnclie G. and Marica A., 2005. Comparison of actual crop evapotranspiration simulated by the CROPWAT model with the estimated data from satellite images in Irrigation and Pest and disease models: Evaluation in Different Environments and web-Based applications, edited by: G. Maracchi, L. Kajfez-Bogataj, S. Orlandini, F. rossi, m. barazutti, European Commission, DG XII, science, research and Development, Environment Research Programme, COST Action 718, p. 80-95, 2005. Steenblock T., H.R. Forrer, P.M. Fried, 2002. www.phytopre.ch: the internet based decision support system (DSS) PhytoPRE+2000 to control late blight on potatoes in Switzerland. In: Proceedings of the sixth Workshop of an European network for development of an integrated control strategy of potato late blight, C.E. Westerdijk and H. Schepers (eds), Edinburgh, Scotland, 26-30. September 2001, PPO-Special Report No. 8, 183-192. Steenblock T., H.R. Forrer, 2002. Krautfuleberatung via Internet. Agrarforschung (9)05: 207214. Stehfest E., M. Heistermann, J.A. Priess, D. Ojima, J. Alcamo 2007. Simulation of global crop yields with the Ecosystem model DayCent. Global Biogeochemical Cycles (submitted). Stenitzer E., 2004. SIMWASER - a numerical model on soil water balance and plant growth.IKT-Reprt 5/2004. Institut fr Kulturtechnik und Bodenwasserhaushalt, A-3252 Petzenkirchen, Austria. Stenitzer E., H. Diestel, U. Franko, R. Schwartengrber, T. Zenker, 2007. Performance of the model SIMWASER in two contrasting case studies on soil water movement.- In : Modelling water and nutrient dynamics in soil-crop systems, Proceedings, ed.: Kersebaum, K.H., Hecher J-M., Mirschel, W., Wegehenkel, M.; Springer, p. 27-36. Stenitzer E., E. Murer, 2003. Impact of soil compaction upon soil water balance and maize yield estimated by the SIMWASER model. Soil Till Res 73: 43-56. Stock M., 2005. KLARA: Klimawandel Auswirkungen Risiken Anpassung. PIK Report 99, Potsdam. Stockle C.O., M. Donatelli, R. Nelson, 2003. CropSyst, a cropping system simulation model. Europ. J. Agron. 18: 289-307.
110
Stockle C.O., M. Cabelguenne, P. Debaeke, 1997. Comparison of CropSyst performance for water management in southwestern France using submodels of different levels of complexity, European Journal of Agronomy, Volume 7, P: 89- 98. Stockle C.O., M. Donatelli, R. Nelson, 2003. CropSyst, a cropping systems simulation model, European Journal of Agronomy, Volume 18, P: 289- 307, 2003. ubi M; B. Cvjetkovi, 2001. Prognoza krumpirove plijesni (Phytophthora infestans mont. de bary) (Forcasting potato late blight (Phytophthora infestans mont. de bary)). Glasilo biljne zatite. 1 2001, 2; 69-72. Supit I., 2000. An exploratory study to improve predictive capacity of the crop growth monitoring system as applied by the European commission, Tree Book 4, ISBN 90804443-5-9, Treemail Publishers, www.treemail.nl, Heelsum, The Netherlands, January 2000. Supit I., A.A. Hooijer, C.A. Van Diepen, (Eds.), 1994. System description of the WOFOST 6.0 crop simulation model implemented in CGMS. Volume 1: Theory and Algorithms. Joint Research Centre of the European Commission, EUR 15956 EN, Luxembourg. Sunik A. and Kurnik B., 2004. On-line irrigation support system in Slovenia. On-line agrometeorological application with decision support at the farm level. COST 718: Meteorological application to agriculture. DINA Research report no. 109: 113-121 Sunik A., I. Matajc, I. Kodri, 2006. Agrometeorological support of fruit production. Application in SW Slovenia. WMO CAgM/ET. Meteorological Appl. (Supplement):8186(2006). Sunik A., 2006a. Vodni primanjkljaj v Sloveniji in moni vplivi podnebnih sprememb. Magistrsko delo, Univerza v Ljubljani, Biotehnika fakulteta, Oddelek za agronomijo: 147 str (in slovene). Sunik A.and Ceglar, A., 2007. tudija teoretinih osnov modela za izraun trajanja omoenosti lista. Project INTERREG IIIA Italija Slovenija. Izvedba sistema za meritev okube Plasmopara viticola na teritorialnem nivoju (PRADA). Projektna faza 2:modul 2.2. Omoenost lista. Agencija RS za okolje. Urad za meteorologijo. Oddelek za agrometeorologijo. Ljubljana 2007. 21 str. (to be published) (in slovene). Stockle C.O., S.A. Martin, G.S. Campbell, 1994. CropSyst, a cropping systems simulation model: water/nitrogen budgets and crop yield. Agric. Syst. 46(3):335-359. Thornton PK., WT. Bowen, AC. Ravelo, PW. Wilkens, G. Farmer, J. Brock, JE. Brink, 1997. Estimating millet production for famine early warning: An application of crop simulation modelling using satellite and ground-based data in Burkina Faso. Agric. For. Met., Vol 83/1-2, 95-112. Thysen I. and Detlefsen N., 2006. Online decision support for irrigation for farmers. Agric. Water Manage. 86, 269-276. Tiilikkala K., T. Carter, M. Heikinheimo, A. Venlinen, 1995. Pest risk analysis of Meloidogyne chitwoodi for Finland. EPPO Bulletin 25: 419-435. Torriani D., P. Calanca, S. Schmid, M. Beniston, J. Fuhrer, 2007. Potential effects of changes in mean climate and climate variability on the yield of winter and spring crops in Switzerland. Climate Research. In press. Trnka M., J. Eitzinger, J. Semeradova, Hlavinka P., Balek J., Dubrovsky M., Kubu G., Stepanek P., Thaler S., Zalud Z., 2008. Expected changes of agroclimatic conditions in Central Europe, Agriculture and Forest Meteorology (submitted). Trnka M., Muka F., Semerdov, M. Dubrovsk, E. Kocmnkov, Z. alud, 2008. European Corn Borer Life Stage Model: Regional Estimates of Pest Development and Spatial Distribution under Present and Future Climate., Ecological Modeling (submitted)
111
Trnka M., F. Muska, D. Semeradova, M. Dubrovsky, E. Kocmankova, Z. Zalud, 2007. European Corn Borer life stage model: Regional estimates of pest development and spatial distribution under present and future climate, Journal of ecological modeling, volume 207, P: 6184, 2007. Trnka M., M. Dubrovsk, P. Hlavinka, D. Semerdov, Z. alud, 2006a. Spring barley production in the climate change hot spot Czech Republic as a case study. In 6th Annual Meeting of the EMS / 6th ECAC, Ljubljana, 2006. Trnka M., P. Kapler, D. Semeradova, M. Dubrovsky, Z. Zalud, M. Svoboda, J. Eitzinger, M. Mozny, 2006b. How to assess impact of climate change on soil climate by Newhall model , Central Europe as a case study area. In.: Managing Drought and Water Scarcity. In.: Vulnerable Environments: Creating a Roadmap for Change in the United States, Longmont, Colorado, 1820 September 2006. Trnka M., F. Ruget, P. Hlavinka, M. Launay, J. Eitzinger, Z. alud, E. Steinitzer, 2006c. Application of crop model STICs in the Central European conditions evaluation and climate change impact estimates. In 6th Annual Meeting of the EMS / 6th ECAC, Ljubljana, 2006. Trnka M., Z. alud, J. Eitzinger, M. Dubrovsk 2005. Quantification of uncertainties introduced by selected methods for daily global solar radiation estimation. Agricultural and Forest Meteorology, Volume 131, Issues 1-2: 54-76. Trnka M., M. Dubrovsk, D. Semerdov, Z. alud, 2004a. Projections of uncertainties in climate change scenarios into expected winter wheat yields, Theoretical and Applied Climatology , Vol. 77, 229-249 Trnka M., M. Dubrovsk, Z. alud, 2004b. Climate change impacts and adapation strategies in spring barley production in the Czech Republic, Climatic Change Vol. 64, 227-255 Tsakiris G. and Vangelis H., 2005. Establishing a drought index incorporating evapotranspiration. Ewra, European Water 9/10, E.W. Publications, pp. 3-11. Tsakiris G., D. Pangalou, H. Vangelis 2006. Regional drought assessment based on the Reconnaissance Drought Index (RDI). Water Resour. Manage, (in press), (available online: www.springer.com). Tsiros E., C. Domenikiotis, M. Spiliotopoulos, N.R. Dalezios, 2004. Use of NOAA/AVHRRbased vegetation condition index (VCI) and temperature condition index (TCI) for drought monitoring in Thessaly, Greece, EWRA Symposium on water resources management: risks and challenges for the 21st century, Izmir, Turkey, 2-4 September 2004, pp. 769-782. Tsiros E., E. Kanellou, C. Domenikiotis, N.R. Dalezios, 2008. Identification of Water Limited Growth Environment Zones Using NOAA/AVHRR Data. 4th International Conference on Information and Communication Technologies in Bio and Earth Sciences, HAICTA 2008, 18-20 September 2008, Athens, Greece (submitted). Tsuji G.Y., G. Hoogenboom, P.K. Thornton, 1998. Understanding Options for Agricultural Production. Kluwer Academic Publishers, Dordrecht. Tubiello F.N., M. Donatelli, C. Rosenzweig, C.O. Stockle, 2000. Effects of climate change and elevated CO2 on cropping systems: model predictions at two Italian locations, European Journal of Agronomy, Volume 13, P: 179- 189. Tveito O.E., I. Bjrdal, A.O. Skjelvg, B. Aune, 2005. A GIS-based agro-ecological decision system based on gridded climatology. Meteorological Applications 12: 57-68. Tsvetsinskaya E.A., L.O. Mearns, T. Mavromatis, W. Gao, L. McDaniel, M.W. Downton 2003. The effect of spatial scale of climatic change scenarios on simulated maize, winter wheat, and rice production in the southeastern United States. Climate Change 60 (1-2): 37-71.
112
Tychon B., D. Buffet, D. Dehem, H. Eerens, R. Oger, 2001. The Belgian Crop Growth Monitoring System, 2nd International Symposium: Modeling Cropping Systems Florence, Italy, July 16-18, 2001. Utset J.,J.Eitzinger, V.Alexandrov.2007..AGRIDEMA: An EU-Funded Effort to Promote the Use of Climate and Crop Simulation Models in Agricultural Decision-Making in Climate Prediction and Agriculture :Advances and Challenges. Edited by Mannava V. K. Sivakumar and James Hansen Van der Keur P., S. Hansen, K. Schelde, A. Thomsen, 2001. Modification of DAISY SVAT model for potential use of remotely sensed data. Agricultural and Forest Meteorology, 106: 215231. Van Ittersum M.K., P.A. Leffelaar, H. vanKeulen, M.J. Kropff , L.Bastiaans, J. Goudriaan, 2003. On approaches and applications of the Wageningen crop models, European Journal of Agronomy, Volume 18, P: 201- 234, 2003. Verhoef W., 1984. Light scattering by leaf layers with application to canopy reflectance modelling, The SAIL model, Remote Sens. Environ., 16, 125-141. Verhoef W., H. Bach, 2002. Remote Sensing Data Assimilation by Model Inversion of Landsat TM Reflectance Images and Coupling to a Crop Growth Model, EGS XXVII General Assembly, Nice, France, April 2002. Vueti V., 2006a. Impact of climate change on the maize productivity in Croatia, 6EMS/6ECAC, Ljubljana, Slovenia, 48 September 2006, (CD-Rom). Vueti V., 2006b. Report of the Pilot Assessment Modelling of maize production and the impact of climate change on maize yields in Croatia, Specific Support Action Introducing tools for agricultural decision-making under climate change conditions by connecting users and tool-providers (FP6-2003-Global-2 Proposal 003944, AGRIDEMA), 32 pp http://www.agridema.org/opencms/export/sites/Agridema/Documentos/croatia_vucetic. pdf Wegehenkel M., H. Jochheim, K.C. Kersebaum, 2005a. The application of simple methods using remote sensing data for the regional validation of a semidistributed hydrological catchment model. - Physics and Chemistry of the Earth 30 (8-10): 575-587. Wegehenkel M., H. Jochheim, K.C. Kersebaum, 2005b. A case study for the regional validation of a hydrological catchment model using remote sensing data. - In: Remote sensing & GIS for environmental studies: applications in geography: 189-196; Gttingen (Goltze). Wegehenkel M., K.C. Kersebaum, 2003. A Modelling System For Calculating Water Balance And Crop Growthat Different Spatial Scales Using Remote Sensing And GIS, Geophysical Research Abstracts, Volume 5, 2003. Weiss M., D. Troufleau, F. Baret, H. Chauki, L. Prevot, A. Olioso, N. Bruguier, N. Brisson 2001. Coupling canopy functioning and radiative transfer models for remote sensing data assimilation. Agric. Forest Meteorol., 108: 113-128 Wessolek G and Asseng S. 2006. Trade-off between wheat yield and drainage under current and climate change conditions in Northeast Germany. Eur. Journal Agronomy 24, 333342. Wittich K.P., 1998. Apple scab potentials and limitations of operational infection forecasts in Germany. In: N.Dalezios (ed.): COST 77, 79, 711 Internat. Symposium on Applied Agrometeorology and Agroclimatology, Volos, Greece, 24-26 April 1996, Eur. Comm., EUR 18328 EN, Luxembourg 1998, p.229-234. Wolf D. and Erickson J., 1993. Carbon Dioxide Effects on Plants: Uncertainties and Implications for Modeling Crop Response to Climate Change. In: Agricultural
113
Dimensions of Global Climate Change / edited by H. M. Kaiser and T. E. Drennen, St. Luice Press, USA, pp. 153 - 178. Wolf J. and Van Diepen C.A., 1994. Effects of climate change on silage maize production potential in the European Community.-Agricultural and Forest Meteorology 71, Nos.12, 33-60. Wolf J. and van Diepen C.A. 1995. Effects of climate change on grain maize yield potential in the European Community. Climatic Change 29: 299-331. Wolf J., 1993. Effects of climate change on wheat production potential in the European Community.- Eur. J. Agron.2, 281-292. Wolf J., M. van Oijen, C. Kempenaar, 2002. Analysis of the experimental variability in wheat responses to elevated CO2 and temperature. Agriculture Ecosystems & Environment 93 (1-3): 227-247. WZU - Wissenschaftliches Zentrum fr Umweltsystemforschung, Universitt Kassel, 2005. Klimawandel und Landwirtschaft in Hessen: Mgliche Auswirkungen des Klimawandels auf landwirtschaftliche Ertrge. INKLIM Final Report. http://www.hlug.de/medien/luft/inklim/dokumente/endberichte/landwirtschaft.pdf Young M.D.B. and Gowing J.W., 1996. The PARCHED-THIRST Model User Guide (v1.0). (User Guide). Department of Agricultural and Environmental Sciences, University of Newcastle upon Tyne, Newcastle-upon-Tyne, UK. alud Z., and Dubrovsk M., 2002. Modeling climate change impacts on maize growth and development. Theoretical Applied Climatology, 72: 85-102. Zalud Z., G.S. McMaster, W.W. Wilhelm, 2003. Evaluating SHOOTGRO 4.0 as a potential winter wheat management tool in the Czech Republic, European Journal of Agronomy, Volume 19, P: 495- 507, 2003.
114
3. Summarizing a questionnaire on trends of agroclimatic indices and simulation model outputs in Europe .
3. SUMMARIZING A QUESTIONNAIRE ON TRENDS OF AGROCLIMATIC INDICES AND SIMULATION MODEL OUTPUTS IN EUROPE Vesselin Alexandrov, Elena Mateescu, Antonio Mestre, Malgorzata KepinskaKasprzak, Valentina Di Stefano, Nicolas Dalezios Abstract Some of the European systems and sectors have shown particular sensitivity to recent trends in temperature and precipitation. The major goal of this work is to summarize a questionnaire on trends in agroclimatic indices and crop model outputs in Europe. This questionnaire was developed and disseminated by Working Group 2 of the COST 734 Action Impacts of Climate Change and Variability on European Agriculture CLIVAGRI. The first part of the survey is related to the availability of long-term historical meteorological and agrometeorological data, its temporal and spatial resolution, area coverage, etc. The second part is dedicated on the various meteorological models applied in selected European countries numerical weather models, global and regional climate models, weather generators. A special attention in this survey is paid on data homogenization tests, techniques and software. Finally, the answers based on the statistical methods for analyses of meteorological and simulation model output related time series, for respected European countries, are listed and summarized.
3.1 State of art 3.1.1 Observed climatic and agroclimatic trends The Fourth Assessment IPCC report (2007) concludes that the worlds average surface temperature has increased by around 0.74C over the past 100 years (Fig. 3.1). This figure is higher than the 2001 reports 100-year estimate of 0.6C due to the recent series of extremely warm years, with 11 of the last 12 years ranking among the 12 warmest years since modern records began around 1850. The warming trend throughout Europe is well established (+0.90C for 1901 to 2005; updated from Jones and Moberg, 2003). However, the recent period shows a trend considerably higher than the mean trend (+0.41C/decade for the period 1979 to 2005; Jones and Moberg, 2003). For the 1977 to 2000 period, trends are higher in central and north-eastern Europe and in mountainous regions, while lower trends are found in the Mediterranean region (Bhm et al., 2001; Klein Tank, 2004).
115
Survey of agrometeorological practices and applications in Europe regarding climate change impacts .
Figure 3.1: Annual anomalies of global land-surface air temperature, 1850 to 2005, relative to 1961-1990 mean (IPCC 4AR, 2007)
Temperatures are increasing more in winter than summer (Jones and Moberg, 2003). An increase of daily temperature variability is observed during the period 1977 to 2000 due to an increase in warm extremes, rather than a decrease of cold extremes (Klein Tank et al., 2002; Klein Tank and Knnen, 2003). Precipitation trends are more spatially variable. Mean winter precipitation is increasing in most of Atlantic and northern Europe (Klein Tank et al., 2002). In the Mediterranean area, yearly precipitation trends are negative in the east, while they are non-significant in the west (Norrant and Dougudroit, 2006). An increase in mean precipitation per wet day is observed in most parts of the continent, even in some areas which are becoming drier (Frich et al., 2002; Klein Tank et al., 2002; Alexander et al., 2006). Some of the European systems and sectors have shown particular sensitivity to recent trends in temperature and (to a lesser extent) precipitation: Upward shift of the tree line (Kullman, 2002; Camarero and Gutirrez, 2004; Shiyatov et al., 2005; Walther et al., 2005). Phenological changes (earlier onset of spring events and lengthening of the growing season); increasing productivity and carbon sink during 1950 to 1999 of forests (in 30 countries) (Menzel et al., 2006; Nabuurs et al., 2003, Shvidenko and Nilsson, 2003; Boisvenue and Running, 2006).
116
Change in high mountain vegetation types and new occurrence of alpine vegetation on high summits (Grabherr et al., 2001; Kullman, 2001; Pauli et al., 2001; Klanderud and Birks, 2003; Peuelas and Boada, 2003; Petriccione, 2003; Sanz Elorza and Dana, 2003; Walther et al., 2005). Northern Europe: increased crop stress during hotter, drier summers; increased risk to crops from hail (Viner et al., 2006). Germany: Advance in the beginning of growing season for fruit trees (Menzel, 2003; Chmielewski et al., 2004). Britain, southern Scandinavia: increased area of silage maize - more favorable conditions due to warmer summer temperatures (Olesen and Bindi, 2004). France: Increases in growing season of grapevine; changes in wine quality (Jones and Davis, 2000; Duchene and Schneider, 2005).
3.1.2 Agroclimatic indices and crop models Climate plays a fundamental role in agriculture because of its direct and indirect influence on production. Each physical, chemical and biological process determining agricultural activity is regulated by specific climatic requirements, and any deviation from these patterns may exert a negative influence. European agriculture, mainly oriented to production of high quality food, may be more susceptible to meteorological hazard impacts because it is based on highly developed farming techniques (COST 734 MoU, 2006) To define agricultural responses to climate, studies can be based on the application of agroclimatic indices and simulation models. They can be used to describe the effect of climatic conditions on key agricultural aspects, including production, protection, fertilization, site selection, watering, etc. Agroclimatic indices Some of the above results were based on application of agroclimatic indices and crop models. An agroclimatic index is a measure or indicator of an aspect of the climate that has specific agricultural significance. A similar definition of an agroclimatic index is the following: an index relating some particular agricultural aspect or operation with one or more elements of the local climate. The agro-climatic indices are based on simple relationships of crop suitability or potential to climate. This type of empirically-derived coefficients is especially useful for broad-scale mapping of areas of potential impact. The indices are derived variables that are defined either by manipulating values of a meteorological variable into a different form or by combining variables with empirically-defined coefficients into a composite term.
117
The most common derived variable to describe the thermal agro-climate is the Effective Temperature Sum (ETS), usually measured in growing degree days. It is calculated as the excess of temperature above a fixed datum (base temperature) over a period required for a specific phase of crop development. Growing degree-days above 5C (GDD): are usually computed by calculating the amount by which average daily mean temperature (Tmean) exceeds 5.0C and summing these values from the time when Tmean first exceeds 5.0C in spring until the last date of Tmean exceeded 5.0C in fall Chapman and Brown 1978). Indices frequently used to measure moisture include Thornthwaites Precipitation Effectiveness Index, the Palmer Drought Index (Fig. 3.2), and the Relative Dryness Index (Palmer, 1965).
Figure 3.2: Long-term variations of PDSI in 3 locations in Bulgaria (Kercheva, 2004)
Other examples of an agroclimatic index are average length of growing season (period between average last and first freezing temperature dates), average total chill hours or chill units, average evapotranspiration, Agroclimatic Resource Index (ACRI), etc. Simple agro-climatic indices combined with geographical information systems have been used to provide an initial evaluation of the global agricultural climate change impacts (Leemans and Solomon, 1993; Fischer and van Velthuizen, 1996). When combined with a spatially comprehensive data base and a geographic information system (GIS), simple agro-climatic indices enable the mapping of altered crop potential for quite large areas at relatively low cost. Regional scenarios of seasonal
118
temperature and precipitation change for 32 world regions analyzed in the Fourth IPCC Assessment Report (2007) show the current variability of climate and the range of changes predicted by GCMs for 30-year time periods centered on 2025, 2055, and 2085. This background information is essential to interpret the potential impacts of climate change on crops and livestock production. Equally important background information is provided by agroclimatic indices. Agroclimatic indices are useful in conveying climate variability and change in terms that are meaningful to agriculture. They give a first approximation of the potential effects of climate change on agricultural production and should continue to be used (Sirotenko et al., 1995; Sirotenko and Abashina, 1998; Menzhulin, 1998). (See also chapter 2). Crop models Models, in general, are a mathematical representation of a real-world system (e.g. Mize and Cox, 1968; Hoogenboom, 2000). During the last decades the application of simulation and system analysis in agricultural research has increased considerably (e.g. Tsuji et al., 1998). Crop models, in general, integrate current knowledge from various disciplines, including meteorology, soil physics, soil chemistry, crop physiology, plant breeding, and agronomy, into a set of mathematical equations to predict growth, development and yield (e.g. Hoogenboom, 2000). Crop simulation models are increasingly being used in agriculture to estimate production potentials, design plant ideotypes, transfer agrotechnologies, assist strategic and tactical decisions (Fig. 3.3), forecast real time yields and establish research priorities (e.g. Bannayan and Crout, 1999; Penning de Vries and Teng, 1993; Uehera and Tsuji, 1993). Numerous crop growth and yield models have been developed for a wide range of purposes in the last decades (e.g. Casanova et al., 2000; Hoogenboom, 2000). These models range in complexity from the most sophisticated simulators of plant growth, primarily intended for research into plant physiological interactions, to multiple regression models using only a few monthly weather variables to forecast regional crop yields. One use of the crop models developed in recent years is to simulate the effects of cultural practices and climatic scenarios on crop growth and yield. However, their use for predicting yields over large areas is limited by the difficulty in obtaining information about local conditions or crop characteristics at any given point. Some crop or soil features may be considered to be constant for a group of genotypes in a given region, but others depend on changes in local conditions (e.g. Guerif and Duke, 1998). Testing over a range of environmental conditions is required to establish confidence in applying models (e.g. Goudriaan and Van Laar, 1994). In the future, models may be useful for improving the efficiency of agricultural systems and could be a tool for
119
farmers trying to improve the profitability of their farms (e.g. Jacobson et al., 1995). Nevertheless, before this is possible, models must be calibrated and evaluated for each climatic region where they are intended for use in decision making (e.g. Sau et al. 1999). Crop simulation models permit the summary of scientific knowledge on the biological processes that regulate plant growth. They are generally built with an analytical purpose. Yet, these models are sometimes used as a predictive tool (e.g. Trousland-Kerdiles and Grondona, 1997).
650 Evapotranspiration [mm] 600 550 500 450 400 1920 1930 1940 1950 1960 1970 1980 1990 2000 b)
400
a)
Irrigation [mm]
300
200
100
0 1920 1930 1940 1950 1960 1970 1980 1990 2000
Figure 3.3: Variations of simulated irrigation and evapotranspiration during the crop-growing season of maize in Tifton, Georgia; CERES model (Alexandrov and Hoogenboom, 2001)
Large area yield forecasting prior to harvest is of interest to government agencies, commodity firms and producers. Early information on yield and production volume may support these institutions in planning transport activities, marketing of agricultural products or planning food imports. Moreover, at world scale, agricultural market prices are affected by information on the supply or consumption of foodstuffs. Market price adjustments or change in agricultural supplies in one area of the world often causes price adjustments in other areas far distant (Supit and van der Goot, 2002). It is no longer necessary nowadays to demonstrate the usefulness of simulation models to explain and predict crop yields or changes in the environment at various scales of agricultural production (e.g. Boote et al., 1996). The value of exploring agronomic situations not tried experimentally is all the greater when the model can simulate several crops arranged in succession, and when as many cropping techniques and environmental limiting factors as possible are included (e.g. Cabelguenne et al., 1999). Crop models can also be used to generate input data for models for technical/economic optimisation, notably in the context of the analysis of European or national policies for competitiveness and environmental protection (e.g. Flichman, 1995; van
120
Ittersum and Rabbinge, 1997). In an economic context in which techniques and regulations are rapidly evolving, or where the objectives and limitations applied to cropping systems are also very diverse, long-term experiments cannot provide answers quickly enough for action to be taken. Models are called upon more and more to contribute to the formulation of innovative cropping systems. Clearly, the credibility of the conclusions from long-term exploratory simulations rests heavily on the reliability of the models, and especially on a good prediction of the yields of crops subjected to various water and thermal stresses (e.g. Cabelguenne et al., 1999). 3.1.3 Examples of previous case studies Wilby and Perry (2006) concluded there is significant evidence that regional variation in climate, particularly the rise of temperature, have already affected agricultural systems in Europe, increasing hazard impacts. Examples of observed changes include the lengthening of the growing season, latitudinal and altitudinal shifts of plant range, earlier flowering, outbreak of plant diseases, acceleration in breakdown of organic matter in soils, and emergence of insects. With respect to the latter for instance, between 1964 and 2004 in England, a 1C increase in temperatures is associated with a 16-day shift earlier in the first appearance of the peach-potato aphid and a 6-day advance in peak flight time of the orange tip butterfly. More frequent precipitation and more humid conditions favour the spread of diseases. The highest intensity of rainfall reduces the infiltration of water in the soil, decreasing the net available soil water content. Finger (2007) analyzed trends in yield growth and yield variability of barley, maize, oats, rye, triticale and wheat in Switzerland from 1961 to 2006. In contrast to linear trends in crop yield growth for most European countries he found significant trends of slowing yield growth for cereal yields in Switzerland. This is caused by the introduction of direct payment schemes that foster environmentally friendly crop farming practices in general and extensive cereal farming in particular. The recently introduced reform of common agricultural policy in the European Union will foster higher shares of reduced input, i.e. extensive, farming. Thus, the saturation of cereal yield growth in Switzerland might indicate future development of crop yields in the European Union. Horvth et al. (2005) conducted a study with the aim to give a modern climatographical analysis on the varied hydrometeorological relations of the region, based on reliable observations of meteorological stations. The analysis includes statistical characteristics of the inter-annual variability, spatial and temporal correlation of the available soil moisture content and long-range changes, as well as their possible relation with climatic trends for greater
121
regions. The above-mentioned aims are intended to be realised on the basis of the Palmers Drought Severity Index (PDSI) data series. Monthly PDSI data series of five stations (Miskolc, Nyregyhza, Debrecen, Kecskemt and Szeged) were determined for the 20th century, in three versions. To study the spatial correlations short PDSI sets of 17 stations were calculated for the period between 1951 and 1992, with Thornthwaites plant-independent method and without homogenisation. The objective and results of our work is to help further the inter-disciplinary study of agro-ecological problems, influenced by the soil moisture content, by performing objective regionalization based on soil moisture anomalies in the region; by publishing the characteristics and the calendar of the objective yeartypes; and also by determining climatically representative, long dry and wet periods. Shen et al. (2005) analyze the long-term (1901-2002) temporal trends in the agroclimate of Alberta, Canada, and explore the spatial variations of the agroclimatic resources and the potential crop-growing area in Alberta. Nine agroclimatic parameters are investigated: May-August precipitation (PCPN), the start of growing season (SGS), the end of the growing season (EGS), the length of the growing season (LGS), the date of the last spring frost (LSF), the date of the first fall frost (FFF), the length of the frost-free period (FFP), growing degree-days (GDDs), and corn heat units (CHUs). The temporal trends in the agroclimatic parameters are analyzed by using linear regression. The significance tests of the trends are made by using Kendall's tau method. The results support the following conclusions. 1) The Alberta PCPN has increased 14% from 1901 to 2002, and the increment is the largest in the north and the northwest of Alberta, then diminishes (or even becomes negative over two small areas) in central and southern Alberta, and finally becomes large again in the southeast corner of the province. 2) No significant long-term trends are found for the SGS, EGS, and LGS. 3) An earlier LSF, a later FFF, and a longer FFP are obvious all over the province. 4) The area with sufficient CHU for corn production, calculated according to the 1973-2002 normal, has extended to the north by about 200-300 km, when compared with the 1913-32 normal, and by about 50-100 km, when compared with the 1943-72 normal; this expansion implies that the potential exists to grow crops and raise livestock in more regions of Alberta than was possible in the past. The annual total precipitation follows a similar increasing trend to that of the May-August precipitation, and the percentile analysis of precipitation attributes the increase to low-intensity events.
122
3.2 Goal: a questionnaire The major goal of this work was to summarize a questionnaire on trends in agroclimatic indices and crop model outputs. This questionnaire was developed and disseminated by Working Group 2 of the COST 734 Action Impacts of Climate Change and Variability on European Agriculture CLIVAGRI(See Annex 2). In order to help the achievement of the above key WG2 deliverables the following questionnaire was developed and disseminated among the national delegates of the COST 734 Action (Table 3.1). The following important notes were also disseminated together with the questionnaire: Please consider ONLY data, models, methods, information, etc. which might be useful for implementation of the WG2 tasks and achievement of the respective WG2 deliverables (detailed within the COST 734 MoU) Please skip a question/point if you are not able to provide information or try to obtain it from other colleagues in your country The example attached is mainly to assist you it should not be assumed as a mandatory one. Hence, fill free to use your own style The European countries who have submitted the questionnaire are presented in Figure 3.4: Bulgaria, Croatia, Czech Republic, France, Germany, Greece, Italy, Norway, Poland, Romania. Serbia, Slovakia, Spain, Switzerland, Slovenia, Finland
Figure 3.4: European countries (in dark grey) submitted the questionnaire
123
3.3 Summarizing the questionnaire 3.3.1 Long-term meteorological and agrometeorological data Long-term meteorological data The evaluation of climate impacts can be performed both on past and future data. There are different ways to analyze past climate, such as the use of historical long-term meteorological data, remote sensing and re-analysis. Historical long-term meteorological data are available at the meteorological services and some other weather related institutions. Typical meteorological elements with long-term records in each European country are air temperature and precipitation as well as wind speed and direction, solar radiation, air humidity, etc. The spatial resolution of the meteorological data varies depending on the density of the respective weather networks and/or the interpolation techniques applied. The data temporal resolution covers different time slices mainly during the 20th century. Both monthly and daily values of the relevant meteorological elements. Most of the gridded weather data sets as well as data series from countries such as Germany, Finland are generally available for the COST 734 applications. However countries such as Bulgaria, France, Romania could provide mainly secondary data, including maps, various indices, etc. (Table 3.1)
An example from Norway
Data set 1 - elements: air temperature ((2m, average, maximum, minimum), precipitation ( winter precipitation only available for 11 station), global radiation(2m), wind velocity ( 2m), leaf wetness duration ( 2m), relative humidity of the air) - temporal resolution: hourly, daily and monthly averages - spatial resolution: site specific measurements of about 70 automated stations in summer and 50 stations in winter. in agricultural districts - area/country/region: agricultural districts in Norway - availability for the WG2 tasks implementation: available on the internet - additional information: The oldest series are from 1987. Very few series are older than 1991.
124
Table 3.1: Information on long-term meteorological data applied in each country
125
126
Data set 2 - elements: Air temperature (2m, average, deviation from normal, average maximum, average minimum, absolute maximum, absolute minimum, number of days with minim temperature less than 0, degree-days base 5, degree-days base 17), relative humidity of the air (2m, average), precipitation (2m, total precipitation, % of normal precipitation, maximum daily precipitation, number of days with precipitation >=0.1 mm), cloud cover (average clod cover, number of days covered by clouds), number of days with sunshine. - temporal resolution: monthly values - spatial resolution: site specific measurements at 71 meteorological stations in Norway mainland and islands in the Arctic - area/country/region: site specific measurements at 71 meteorological stations in Norway mainland and islands in the Arctic - availability for the WG2 tasks implementation: ( most of the data since 1996 are available on the internet on the web-page of Norwegian Meteorological institute) Data on paper available since about 1960 . - additional information: It exists yearly reports of measurements since about 1870 from this Norwegian Meteorological institute. Data set 3 - elements: Long term precipitation variations (% change), Long term temperature variations ( % change) - temporal resolution: annual and seasonal - spatial resolution: map of the whole country - area/country/region: country - references: Nasjonalatlas for Norge Data set 4 - elements: gridded CRU data set - air temperature 1931-60, snow cover - temporal resolution: Temperature for the months April and July of the normal period 1931-60, and snow thaw for the years 1957- 83. - spatial resolution: Temperature in a 10 km grid, snow thaw to agricultural communities - area/country/region: Temperature for Norway, snow thaw for Aust-Agder County in southernmost Norway - availability for the WG2 tasks implementation: Temperature on map and data file, snow thaw on map in publication - references: Skjelvg, 1987a,b.
127
Long-term agrometeorological data In recent years, the availability and use of agrometeorological data from weather stations or networks have become more and more important for the management and planning of agricultural activities and for the decision support system (Hoogenboom 2000). In fact, agrometeorological variables such as temperature, relative humidity, solar radiation, wind and precipitation heavily affect the characteristics of a given territory. These variables are then often used as input in many simulation models to investigate issues such as genotype improvement, environmental characterisation, evaluation of potential production, prediction of the effect of climatic change and variability on crop growth and yield and definition of farm management techniques (Maracchi 2003). The trend of these variables is not constant over the territory, but it changes depending on the geomorphologic and topographical characteristics of the landscape, which can induce really wide differences over a short distance. In particular, geographic coordinates, altitude above sea level, distance from bottom of valleys, slope and aspect affect the spatial distribution of weather variables. In most European countries, phenological data are collected or have been collected in the past over several decades. Monitoring phenological phases is carried out in many European countries like Slovenia, Slovakia, Croatia, Bulgaria, Czech Republic, Switzerland, Germany, Austria, and others. Each country has its own database, in some cases still on paper, mostly on databanksystems, going back in many cases to the 1950s and a few very long timeseries from single locations like bud burst of horse chestnut in Geneva since 1808 (Defila C., 2001). Besides the scientific research in phenology that is now focused on climate warming and its impact on vegetation, pheno-data are used for crop modeling, pollen forecast and general information to the public via media and in schools. The observations should have been made following the same, similar, or at least comparable, rules and quality checks. Phenological phases reflect among other things the environmental characteristics of the climate in the region where they occur. Consequently, long series of phenological observations may be used for the detection of climate variability or climate change. The information on agrometeorological data available in selected COST 734 countries is presented in Table 3.2
128
Table 3.2: Information on long-term agrometeorological data
129
130
3.3.2 Numerical weather models, regional climate models, weather generators Numerical weather models According the COST 734 Memorandum of Understanding numerical weather model outputs can be also used to obtain trends in agro-climatic indices and crop model outputs. An important element in the application of the numerical weather models is reanalysis data sets. In respect to reanalysis reproducing past weather synoptic weather charts are based on meteorological elements including the wind velocity, atmospheric temperature, and humidity, at regular grid points covering the whole of the earth (globe). The grid point data is prepared by analytical processing of meteorological data collected by a variety of observational means using advanced computational techniques. The data is used as initial values for computer-based weather forecasts. The accuracy of weather forecasts largely depends upon the accuracy of initial values, and to obtain this accuracy the analytical processing technique used in meteorological observation has undergone constant improvements. The reanalysis repeals the analytical processing of meteorological observations carried out in the past by application of the most advanced data processing technology available. Although the grid point data accumulated in the weather forecasts is very valuable information, its quality, such as its accuracy and characteristics, has varied with changes in data processing technology. For this reason, long-term reanalysis data covering more than 10 years is required for an accurate understanding of a variety of climatological performed in Europe and the U.S since the 1990s and their results are utilized internationally as basic data for weather forecasts and studies on climate variability. According to the information in Table 3.3 as well as the other answers in the questionnaire ECMWF, LM COSMO and ALADIN are among the most applied numerical weather models in Europe. The numerical weather models and /or their related outputs used in Greece can be summarised as follow: - ECMWF and LMCOSMO. In a study implemented by Anadranistakis et al. (2006) the prognostic values of the wind force (up to 48 hours with time step of 6 hours) of the two numerical models are compared to the measurements of the wind force at 2 meters above sea level recorded from the National Marine Centres 5 buoys (for the time interval of 1 year), which cover the Aegean Sea. In all statistical parameters the ECMWF model (resolution 0.5) has better behaviour than the LM-COSMO model (resolution 0.0625). The reason for this differentiation is that the analysis and the initial conditions of LMCOSMO are based on the global model of German Meteorological Service.
131
- BOLAM. Modern computers have enhanced computational capabilities at low cost allowing thus the detailed description of topography in weather forecasting models through the use of high-resolution grids (Argiriou et al. 2006). However even forecasts based even on 2-km horizontal resolution grid, present non-systematic discrepancies, when compared with measured weather parameters. In order to cope with this problem, several empirical methods have been proposed. In this paper presented the potential of applying such corrections to the near surface air temperature and wind speed predicted by the high resolution model BOLAM, based on surface measurements, through the use of artificial neural networks (ANNs). - LM. The need for very high resolution numerical weather prediction products either for direct operational use or to support meteorological products of broader meteorological significance, by solving the model equations on a very fine horizontal grid, motivated the integration of the nesting technique into the LM code. In the resulting version of the model (LM-nest) the numerical results from a coarse grid are used to provide boundary conditions for a fine grid embedded into the coarse grid for every integration time step. The algorithm of LM-nest is presented by Avgoustopoulou and Papageorgiou (2006) as well as results for characteristic weather situations over selected areas of Greece. The following list of numerical local area models is adopted in Italy: LAMI (by Servizio Meteorologico dell'Aeronautica) LAMBO (by ARPA Emilia Romagna SIM http://www.arpa.emr.it/sim/) BOLAM by CNR Isac, managed by Servizio Agrometeorologico Regionale della Sardegna (http://www.sar.sardegna.it/) and Universit di Genova (http://www.fisica.unige.it/ atmosfera/bolam_avn.htm) DALAM (by Ucea www.ucea.it) LILAM (by Meteoliguria http://www.meteoliguria.it) RAMS (by LAMMA - http://www.lamma.rete.toscana.it/) A review on present state and perspectives of limited area models for agriculture in Italy was carried out by Buzzi (2002). In Poland presently, two mesoscale weather prediction models ALADIN and COSMO are used. Regional climate models are not used in Poland. Depending on research needs, the results of global models are adapted for estimation of expected regional conditions within the area of Poland.
132
Table 3.3: Numerical weather models in selected countries
The forecasters in Romania use 4 numerical models: ALADIN, integrated twice a day (00 and 12 TU) for an anticipation of 78 hours (horizontal spatial resolution of 10 km); HRM, integrated twice a day (00 and 12 TU) for an anticipation of 72 hours (horizontal spatial resolution of 20 km); MM5, integrated 4 times a day (00, 06, 12, and 18 TU) for an anticipation of 24 hours (horizontal resolution of 15 km); LM, non-hydrostatic Lokal Model, elaborated at DWD and developed within COSMO (Consortium for Small Scale Modeling).
133
The verification of NWFC-issued weather forecasts is made daily both for the whole country and for Bucharest. All the verification results are transmitted to mass-media and presented on-line in graphs on the NMA Internet site http://brutus.inmh.ro. Climate models Italy applies the General Circulation Model PUMA (Portable University Model of Atmosphere) and Planet Simulator. Both were developed in Germany, University of Hamburg: - temporal resolution: monthly values - spatial resolution: 3.5 deg - area/country/region: globe - availability for the WG2 tasks implementation: free from owner - additional information: other free weather generators could be provided upon request - references (incl. web pages): www.mi.uni-hamburg.de/plasim; Hoskins and Simmons (1975); James (1994); Fraderick et al. (2005a,b,c) The regional climate models RegCM3, PRECIS and MM5 are used in Europe. The regional model RegCM3 is used in study, the dynamic core of which is based on the hydrostatic version of the mesoscale model MM5 while recently added a tracer model with six tracers for the study of aerosol transport and their feedbacks on climate. Simulations performed with a grid resolution of 60*60 km over a greater European area as well as with a finer grid resolution 20*20 km over Greece using the NCEP meteorological fields as boundary conditions (Zanis et al. 2006). In order to investigate climate change and impacts in Greece as well as in the Eastern Mediterranean area, the regional climate model PRECIS, has been implemented in the National Observatory of Athens (NOA). For the application of the PRECIS model at NOA a horizontal analysis of 25 km was selected, which is the finest resolution used so far in the area as well as the complex land-sea distribution (Kotroni et al. 2006). A recent study (Oikonomou et al. 2006) examines the potential to improve rain forecasting over the Greek Peninsula by making use of the atmospheric mesoscale model MM5, which runs operationally at NOA. For this purpose, initially the 3DVAR Data Assimilation system was setup, and then 16 study cases of severe rainfall were selected within the period 2002-2003. The performance of the results is examined by applying statistical methods and in comparison with independent precipitation measurements. The Czech COST 734 partner provided information on IPCC DDC data and PRUDENCE outputs which are available on the internet. Data Distribution Centre (DDC) of the Intergovernmental Panel on Climate Change (IPCC). The DDC offers access to baseline and scenario data for representing the evolution
134
of climatic, socio-economic, and other environmental conditions. The data are provided by co-operating modelling and analysis centres. It also provides technical guidelines on the selection and use of different types of data and scenarios in research and assessment. The DDC is designed primarily for climate change researchers, but materials contained on the site may also be of interest to educators, governmental and non-governmental organizations, and the general public. Analysis of climate impacts, adaptation, and vulnerability involves a set of activities designed to identify the effects of climate variability and change, to evaluate and communicate uncertainties, and to examine possible adaptive responses. Methods for analysis of impacts, adaptation, and vulnerability have evolved over the past decade, and a large array of methods and tools are now available for use in specific sectors, at different scales of analysis, and in contrasting environmental and socio-economic contexts. Most assessments of the impacts of future climate change are based on the results of impact models that rely on quantitative climatic and non-climatic data and scenarios. The identification, selection, and application of baseline and scenario data is a crucial step in the analytical process. The great diversity of the data required and the need to maintain consistency between different scenario elements can pose substantial challenges to researchers. The IPCC DDC seeks to provide access to such data and scenarios and to offer guidance on their application. PRUDENCE is a European-scale investigation with the following objectives: to address and reduce the above-mentioned deficiencies in projections; to quantify our confidence and the uncertainties in predictions of future climate and its impacts, using an array of climate models and impact models and expert judgment on their performance; to interpret these results in relation to European policies for adapting to or mitigating climate change. Climate change is expected to affect the frequency and magnitude of extreme weather events, due to higher temperatures, an intensified hydrological cycle or more vigorous atmospheric motions. A major limitation in previous studies of extremes has been the lack of: appropriate computational resolution obscures or precludes analysis of the events; long-term climate model integrations - drastically reduces their statistical significance; co-ordination between modelling groups - limits the ability to compare different studies. These three issues are all thoroughly addressed in PRUDENCE, by using state-of-the-art high resolution climate models, by co-ordinating the project goals to address critical aspects of uncertainty, and by applying impact models and impact assessment methodologies to provide the link between the provision of climate information and its likely application to serve the needs of European society and economy.
135
Weather generators Long-term series of daily weather data are often required for the analysis of weather-impacted systems (e.g., cropping management systems, hydrologic studies, environmental studies, and others). Weather generators are computer programs that use existing weather records to produce long series of synthetic daily climatic data. The statistical properties of the generated data are expected to be similar to those of the actual data. Weather variables required by many applications include precipitation, maximum and minimum temperature, rainfall, solar radiation, wind speed and some measurement of air water vapor (Acock and Acock, 1991). In some cases, records of such variables may be not available, incomplete, insufficient in length, or only summarized in monthly archives. Weather generators are practical tools to bypass those problems (Johnson et al., 1996). Several computer programs have been developed that are capable of producing stochastically generated weather data from existing daily data. Examples include WGEN (Richardson and Wright, 1984), WXGEN (Sharpley and Williams, 1990), CLIGEN (Arnold and Elliot, 1996), USCLIMATE (Johnson et al., 1996), CLIMAK (Danuso et al., 1997), and ClimGen (Stckle et al., 1998). The following weather generators were reported in the COST 734 questionnaire: WGEN (applied in Bulgaria, Spain) CLIMGEN (applied in Germany) LARS-WG (applied in Slovenia, Switzerland) Met&Roll (applied in Czech Republic, Croatia, Serbia, etc.) One model for weather generation that has been applied extensively in the United States is WGEN by Richarson and Wright (1984). This model generates estimates of daily precipitation, maximum and minimum temperatures, and solar radiation, and it is designed to preserve interdependence between variables as well as persistence and seasonal characteristics of each variable. In WGEN, four parameters are required for precipitation generation, which are held constant within each month but are varied from month to month. This approach can introduce inaccuracies in generating precipitation data. The values of these parameters were determined from long records of data in 139 location in the United States. The procedure used for generating solar radiation and temperature is based on the assumption that these are weakly stationary processes. A one-term Fourier series is used to model the seasonal variation in both temperature and solar radiation. The coefficients of the Fourier term were determined throughout the locations tested and it was found that some of the coefficients were strongly location dependent (Richardson and Wright, 1984), thus limiting the application of this
136
model to areas where these coefficients are available. WGEN requires long records of daily weather data to estimate parameters, limiting its use to regions of the world where sufficient data are available. Monthly summaries of weather data cannot be used to generate daily data. Parameters required for WGEN are currently available only for the continental U.S. Richarson and Wright (1984) also reported that, for some locations, there were differences in mean monthly precipitation and temperatures between actual and generated data. In some cases the cause of the differences was attributed to temporal and spatial smoothing that are inherent in WGEN. ClimGen, is a weather generator that uses similar general principles than WGEN, the first and most widely used weather generator in the US, but with significant modifications and additions. ClimGen generates precipitation, daily maximum and minimum temperature, solar radiation, air humidity, and wind speed. It uses a Weibull distribution to generate precipitation amounts instead of the Gamma distribution used by WGEN. The Weibull distribution is easier to parameterize, describes well the distribution of precipitation amounts, and can be simplified for applications to conditions with minimum data. In ClimGen, all generation parameters are calculated for each site of interest while WGEN used fixed coefficients optimized from a large US weather data base. The advantage is that ClimGen can be applied to any world location with enough information to parameterize the program. WGEN uses truncated Fourier series fits to produce daily values for monthly-calculated quantities of mean weather variables. This arbitrarily chosen functional form can lead to relatively poor fit to the data. ClimGen uses quadratic spline functions chosen to ensure that the average of the daily values are continuous across month boundaries, and that the first derivative of the function is continuous across month boundaries. Other features of ClimGen that are not available in WGEN include the generation of vapor pressure deficit (VPD) and wind speed. In addition, alternative approaches allow users to estimate VPD and solar radiation from existing temperature records. LARS-WG is a model to simulate time-series of a suite of climate variables at a single site, which can be used: to generate long weather time-series suitable for the assessment of agricultural and hydrological risk; to provide the means of extending the simulation of weather to unobserved locations; to serve as a computationally inexpensive tool to produce high resolution climate change scenarios incorporating changes in climate variability. LARS-WG has been used recently to develop high temporal (daily) and spatial (site) resolution climate change scenarios based on UKCIP02 and HadRM3 projections. Scenarios incorporate changes in climatic variability such as duration of dry and wet spells or temperature variability, derived from daily output from regional HadRM3 (Racsko et al., 1991; Semenov and Barrow, 1997; Semenov
137
et al., 1998; Semenov and Brooks, 1999; Lawless and Semenov, 2005; Semenov, 2007) Met&Roll (http://www.ufa.cas.cz/dub/impacts/met&roll.htm ) is a Czech weather generator which is designed to produce synthetic series of four daily climatic characteristics: SRAD = solar radiation, TMAX = temperature maximum, TMIN = temperature minimum, RAIN = precipitation amount. Martin Dubrovsky started his work in spring 1994 and it was inspired partly by weather generators included in the DSSAT software package (WGEN and SIMMETEO weather generators) and partly by paper of Wilks (1992). In fact, the model is the same as in WGEN and SIMMETEO: 1st order Markov chain for precipitation occurrence, gamma distribution for the precipitation amount, trivariate 1st order autoregressive model for SRAD, TMAX and TMIN). Met&Roll differs from WGEN and SIMMETEO by number of parameters estimated from the "learning" sample (observed series of the four weather characteristics): Met&Roll estimates means and standard deviations of SRAD, TMAX, TMIN for both wet and dry days separately (WGEN included in the DSSAT does not distinguish wet and dry versions for TMIN; SIMMETEO does not calculate any standard deviation and does not distinguish wet and dry versions for SRAD, TMAX, TMIN - it employs parametrisations); Met&Roll saves the annual courses of all above statistics in a day-by-day file. WGEN and SIMMETEO saves only monthly statistics and the daily ones are estimated by Fourier analysis just before the generation process. The detailed quality analysis includes comparison of Met&Roll with WGEN and SIMMETEO and it was found that the more detailed representation of the parameters of the generator's model (which is the case of the Met&Roll) improves reproduction of the structure of the weather series. Main Met&Roll features are: runs in a graphical mode (mouse-operated, user-defined color palette, ...) runs either in an interactive mode (managed by Menu bars and Option windows, mouse-operated) or non-interactive mode called from a batch file using `initialisation files') all actions are monitored in the report file (this file contains records of times and important parameters of all processes - ANALYZE, MODIFY, GENERATE, SMOOTH). Thus, after quitting the Met&Roll, you may reconstruct all processes after viewing the report file. Optionally, the whole report file or its blocks (if separated), may be used as a regular initialization file. the contents of the option panels (concerns only: ANALYZE, GENER, MODIFY), may be saved/loaded on/from the disk. This feature may be also used to prepare the initialization file and run the Met&Roll in a non-
138
interactive mode (the structure of the panel files is a same as a structure of the blocks in the initialization file). 3.3.3 Homogenization tests/procedures For long-term climate analyses - particularly climate variability and change analyses - to be accurate, the climate data used must be homogeneous (e.g. Peterson et al., 1998). A homogeneous climate time series is defined as one where variations are caused only by variations in weather and climate (e.g. Conrad and Pollak, 1950). However, many climatologists have noticed that many factors are likely to introduce homogeneity breaks into long-term climatological series (e.g. Auer and Boehm, 1994; Easterling et al., 1996; Vincent, 1998). Inhomogeneous data can be the reason of false conclusions that do not correspond to the real climate changes (e.g. breaks could be higher than trends). Unfortunately, most long-term climatological time series have been affected by a number of non-climatic factors that make these data unrepresentative of the actual climate variation occurring over time (e.g. Scholefield, 1997). These factors include changes in: instruments, observing practices, station locations, formulae used to calculate means and station environment (e.g. Peterson et al., 1998; Sneyers, 1999). Some changes cause sharp discontinuities while other changes, particularly change in the environment around the station, can cause gradual biases in the data. This effect can be different for different weather parameters. Nowadays the weather measurements are well defined by the World Meteorological Organization (WMO). Some time ago, however, the conditions for weather measurements were not standardized. All these inhomogeneities can bias a time series and lead to misinterpretations of the studied climate (e.g. Moisselin and Mestre, 2002; Tuomenvirta and Heino, 1996). It is important to remove the inhomogeneities or at least determine the possible error they may cause (e.g. Torok and Nicholls, 1996) There are several direct and indirect methodologies for homogeneity testing. The direct methodologies include, for example, use of metadata, side by side comparisons of instruments, statistical studies of instrument changes. The indirect methodologies consider use of single station data, development of reference time series, subjective and objective methods. The available objective methods include: Potter's method; Standard normal homogeneity test; two-phase regression; rank order change point test; Craddock test; Caussinus-Mestre technique; multiple analysis of series for homogenization (e.g. Alexandersson and Moberg, 1997; Peterson et al., 1998; Szentimrey, 1999). According to the answers from the questionnaire the following homogenization tests/techniques/software are applied in European countries:
139
Standard Normal Homogeneity Test (e.g. in Croatia, Slovenia, Spain, etc.): Craddock test (e.g. in Austria, Slovenia) Caussinus-Mestre technique (e.g. in France, Bulgaria) AnClim software (e.g. in the Czech Republic, Slovakia, Italy, Bulgaria, etc.) ClimDex software ( Italy) MASH (Multiple Analysis of Series for Homogenisation), (eg. in Hungary) Both, Craddock and SNHT tests (e.g. Alexandersson 1986, Craddock 1979) can be applied on yearly, seasonal or monthly time series of mean, minimum and maximum temperature, precipitation and sunshine duration. For both methods homogenous reference time series is needed. Craddock test is based on the analysis of von Neumann ratio, which is closely related to the firstorder series correlation coefficient. If the series is homogenous with constant mean, von Neumann ratio is close to 2, otherwise it is smaller than 2. SMHT method uses T-test to analyse the shifts in mean and standard deviation of time series. The detected shifts should be approved in the history of the station and than they are adjusted. For temperature also daily data are homogenized using harmonic functions on monthly corrections. The Caussinus-Mestre method, simultaneously accounts for the detection of unknown number of multiple breaks and generating reference series. It is based on the premise that between two breaks, a time series is homogeneous and these homogeneous sections can be used as reference series. Each single series is compared to others with the same climatic area by making series of ratio (e.g. for precipitation) and differences (e.g. for air temperature). These ratios or difference series are tested for discontinuities. When a detected break remains constant throughout the set of comparisons of a candidate station with its neighbours, the break is attributed to the candidate station time series (e.g. Caussinus and Mestre, 1997; Mestre, 1999, 2000; Peterson et al., 1998). For detection purposes, the formulation described by Caussinus and Lyazrhi (1997) is used which allows the determination of a normal linear model with an unknown number of breaks and outliers. They formulated it as a problem of testing multiple hypotheses. At each step, one or two more breaks are added to the previous selected hypothesis. Analytical studies (e.g. Mestre, 1999) show that this double step procedure gives better detection results than the single step procedure for up-and-down breaks. Furthermore, a triple step procedure, much more greedy in terms of computation time, leads to small improvements (e.g. Mestre, 1999, 2000). The Causinus-Mestre method, with a double step procedure, is now the standard detection part of the homogenization method used in Mto-France (e.g. Moisselin and Mestre, 2002; Moisselin et al., 2002). The knowledge of break positions can be a very interesting aspect for
140
some users. For many applications (such as climate change studies) it is the first half-part of the problem. The other one, described below, is the break correction. A two factors linear model is proposed for correction purposes (e.g. Mestre, 2000). The series within the same climatic area are considered as affected by the same climatic signal factor at each time, while the station factor remains constant between two breaks. The model is applied after break detection. It provides the correction coefficient of a set of inhomogeneous series, through weighted least-squares estimation of the parameters. The weighted least squares allow correction of series with missing data. It also allows the data weighting, according to their supposed quality, which can be estimated, for example, with the correlation between the stations. The above formulation is equivalent to an exact modelling of the relative homogeneity principle. Given a set of inhomogeneous instrumental series, it allows unbiased estimations of the breaks affecting these series. This method does not require computation of regional reference series, and is currently the standard correction part of the homogenization method used at Mto-France (e.g. Moisselin and Mestre, 2002; Moisselin et al., 2002) Software packages such as AnClim-Pro, CLimDB, LoadData are frequently used nowadays, e.g. in Italy. The AnClim software: Input format: TXT files, working with one station at a time Menu is ordered in a sequence (steps) to be taken during data processing: viewing data, adjusting (transformation), testing distribution, finding outliers, homogeneity testing (both absolute and relative homogeneity tests), analysis, filtering. managing the software: Series Controler (form in the right bottom corner: info about period, length of series, number of missing values, using monthly data or seasonal and annual averages), settings for the software: use menu Options / Settings, documentation can be found using menu Help / Documentation. Working with one series: graphs, outliers, statistical characteristics, testing Working with two series: merging two series (using differences, ratios), graphs, outliers, statistical characteristics, testing Time series analysis (mainly cyclicity) Filtering (smoothing) output data Other tools: filling missing values, creating reference series, automatization LoadData application for loading data from database (e.g. Oracle) Creating connection (using ODBC) Specification what to download (stations, elements, periods) Adjusting output (cross tables)
141
Output to TXT files, Excel, AnClim Running LoadData application from the AnClim Download wizard guides through all the steps Transformation to files suitable for AnClim automatically ProClimDB software: Used for processing whole datasets (all stations at a time) There are two main input files in the software: Data file (dbf file with all stations data) and Data_info file (list of stations with their geography etc.) How to proceed: select a function, specify files, set options, run the function for a whole dataset User has full control over the processing all the time, a lot of auxiliary output is created First step of processing is getting information about all available stations, their period etc., the other step is importing geography, then we can calculate statistical characteristics , correlations, reference series, outliers, etc. A lot of tools for managing dataset is available as well. Homogeneity testing in AnClim: after we export candidate and its reference series to TXT files (using ProClimDB), we can use automatization in AnClim running homogeneity tests for differences (ratios) between candidate and its reference series for a whole dataset. It is recommended to use several tests: e.g. t-test (on differences), ManWhitney-Pettit test (non-parametric test), SNHT (several modifications), Bivariate test, Vincent test (two-phase linear regression). Further it is useful to run the tests for several types of reference series: based on correlations, distances, regional average (good for temperature but in case of precipitations we can get only one meaningful type of reference series) Testing monthly, seasonal and monthly averages Results from homogeneity testing are put back to ProClimDB (imported to DBF file) and further processed. Numbers of inhomogeneities detections per individual years or groups of years are calculated (summed). Where the inhomogeneity detection using various tests, various reference series etc. coincides, we can regard such cases as very probable to be inhomogeneous, then to go to metadata and verify them etc. After we decide inhomogeneites, we can adjust them, and in the end to fill missing values ClimDex is a Microsoft Excel program designed to assist researchers in the analysis of climate change and detection. More specifically, ClimDex guides a user through a four-step analysis process, using a graphical user interface. This
142
process consists of the following steps: 1. Quality Control; 2. Homogeneity Testing; 3. Calculate Indices; 4. Region Analysis Weather data in the Czech Republic that are certified for the use by various research teams are provided by the Czech Hydrometeorological Institute. These data are homogenized before handover to the user based on the metadata and information from surrounding stations. All certified data originate from a CLIDATA (http://www.ataco.cz/clidataweb/introduction/introduction.jsp) which is used for data quality control and is used for the administration of climatology stations and station observations. The access into the primary database is severely restricted and only homogenized and complete so called technical series are allowed to be used by researchers outside the CHMI. Experts of CHMI (e.g. Dr. Stepanek) conduct extensive research in the field of data homogenization and provide freely available software and know how (e.g. AnClim software package: http://www.klimahom.com/software/ AnClim.html). In Spain for precipitation: the SNHT and Wald-Wolfowitz tests are applied to the monthly precipitation values in stations with more than 20 years of data from 1960 (approximately 1200 series). For temperature: Mann test (applied to the monthly temperature values in stations with more than 20 years of data from 1960) as well as Petit test (applied to the monthly temperature values in stations with long series). In Poland large variety of homogenization methods are generally accepted (e.g. Kouchowski, 1990; Pruchnicki, 1987). In some cases, several methods are applied simultaneously for more accurate estimation of a given situation. For procedures of detection non-homogenous series we use the following methods: difference - quotient method,; parametric tests, i.e.: T-Student test, F-Snedecor test, etc.; non-parametric tests, i.e.: Smirnow test, WaldWolfowitz test (series test), Wilcoxon test, etc. In case of non-homogeneity detected in observation series, the following methods are applied to remove it, i.e.: difference - quotient method, double-mass method, isomer method, correlation method, Standard Normal Homogeneity Test developed by Alexandersson. Presently, there is no coordinated program of climate data homogenization. This process is conducted in different research centers, but the volume of published homogenous measurement series is small. Dalezios et al. (2004) deal with tests on homogeneity of temperature and time series in Greece. The emphasis is placed on the identification of inhomogeneities in temperature and precipitation time series as well as on the specification of certain years, in which inhomogeneities occur. Moreover in this study correction factors are identified to artificially homogenize the time series. This accomplished by employing various homogeneity tests to monthly data over 37 years (1951-1987) at 31 stations over Greece which has been
143
classified in 5 regions using Factor analysis. Pnevmatikos et al. (2006) worked on homogeneity and quality control of rainfall time series. They used different methods and techniques that have been developed in the past, for homogeneity adjustments of annual rainfall data of the 20th century in 36 Greek stations. Several methods have been used in Switzerland (e.g. Auer et al., 2006; Begert et al., 2005; Della-Marta, 2006). National weather data has been homogenized in the framework of the CLIMATE90 programme. The methods are described in Aschwanden et al. (1996). 3.3.4 Statistical methods for analyses of meteorological and simulation model output related time series Country examples For trend calculation the following statistics are applied in Bulgaria, Croatia, Czech Republic, Slovenia, Poland, Spain, Italy, etc.: least squares; minimum absolute deviation; significance testing: confidence intervals for least squares, the Mann-Kendall and Spearman rank statistics. Least squares linear regression is a maximum likelihood estimate i.e. given a linear model, what is the likelihood that this data set could have occurred? The method attempts to find the linear model that maximises this likelihood. Least squares linear regression, like many statistical techniques, assumes that the departures from the linear model (errors) are normally distributed. Techniques that do not rely on such assumptions are termed robust. Non parametric correlation statistics are an attempt to overcome the limited resistance and robustness of the linear correlation coefficient, as well as the uncertainty in determining its significance. Serbia applies time series analysis using quantitative parameters of chaos. This method includes deriving low attractors in atmospheric data time series and calculations corresponding quantities as the Lyapunov exponent, Kolmogorov entropy and Kaplan-Yorke dimension. It is also combining with filtering techniques for time series, particularly with the 4253H filter. In the Czech Republic the statistical methods include standard statistical packages, ANCLIM (see above), neural networks, wavelet analysis packages, cluster analysis; various techniques for assessment links with the agrometeorologically relevant events and e.g. regional circulation patterns (e.g. GWL). Slovenia uses correlation analysis, multiple regression analysis, trend calculation: seasonal or monthly decomposition (Census 1 or 2); STATISTICA software which needs a license. Switzerland: Fourier and spectral analysis, and others, mostly using available FORTRAN routines (e.g. Visual Numerics, 2001; Press et al., 1992). Recently,
144
an increasing number of investigations have been carried out using the R language (R Development Core Team 2006.), which is becoming a standard. Of particular interest could be the studies on precipitation and the detection of extreme events in observed time series (Frei and Schr, 2001; Scherrer et al., 2006; Schmidli et al. 2002; Schmidli and Frei, 2006) Spain: Empirical Orthogonal Function and Principal Component. Composites. Statistical regression models. (e.g. Rodrguez-Puebla et al., 2007; Libiseller and Grimvall, 2002). Slovakia: GIS analysis and standard statistical methods are used. Poland: the basis for time series is generally adapted statistical methods. Calculations are performed by computer software of general use such as Excel, and more specialized such as STATISTICA, or original software written in specific research centers. Most applied methods (Kouchowski, 1990) are: presentation of measurement data with empirical formulae and estimation of their compatibility with tests such as 2 or KolmogorowSmirnow test; spectral analysis of time-series data, harmonic analysis of timeseries data, filters, tests of statistical significance in phenomena changeability research, empirical orthogonal functions (EOF) In Italy, beyond the common statistical methods, indices for extremes (Klein Tank and Konnen, 2003) as in ECA&D (http://eca.knmi.nl) are applied. The most used statistical software is MATLAB. Some specific software for extremes is available from ECA&D (ClimDex). Greece: According to Anagnostopoulou et al. (2006) analysis of changes in extreme rainafall have used both parametric and non parametric methods. The parametric methods usually involve fitting a suitable distribution to the data then analyzing changes in the distributions parameter. Non-parametric methods have utilized a large number of extreme rainfall indices. In the present study the Generalized Extreme Value (GEV) Distribution and the Pareto Distribution are applied and their results are analyzed. Daily precipitation data derived from 22 Greek stations evenly distributed in the Greek region have been used for the time period 1958-2000. The results derived from the analyses concern the threshold selection as well as the return period of extreme rainfall events in the study area. Skourkeas et al. (2006) focus on the estimation of mean maximum and minimum winter temperature over the Greek region, by applying a statistical downscaling method based on the CCA technique. Several test-hypothesis for the variance and the mean value were done between the observed and the estimated temperatures. The CCA approach is turned out to be a very useful multivariate technique for the construction of reliable linear models in downscaling methods.
145
Trend calculation and significant testing The following text is an overview of linear regression methods for reference by members of the STARDEX project (Haylock, 2004; STARDEX, 2004).
Least squares
Least squares linear regression is a maximum likelihood estimate i.e. given a linear model, what is the likelihood that this data set could have occurred? The method attempts to find the linear model that maximises this likelihood. Suppose each data point yi has a measurement error that is independently random and normally distributed around the linear model with a standard deviation i. The probability that our data occurred is the product of the probabilities at each point:
For further information, see Wilks (1995) or Press et al. (1986). Maximizing this is equivalent to minimizing:
If the standard deviation i at each point is the same, then this is equivalent to minimizing:
Solving this by finding a and b for which partial derivatives with respect to a and b are zero, gives the best fit parameters for the regression constant and coefficient ( and ):
146
Minimum absolute deviation
Least squares linear regression, like many statistical techniques, assumes that the departures from the linear model (errors) are normally distributed. Techniques that do not rely on such assumptions are termed robust. Least squares regression is also sensitive to outliers. Although most of the errors may be normally distributed, a few points with large errors can have a large affect on the estimated parameters. Techniques that are not so sensitive to outliers are termed resistant. A more resistant method for linear trend analysis is to assume that the errors are distributed as a two-sided exponential. This distribution, with its larger tails, allows a higher probability of outliers:
The solution to this needs to be found numerically. Example code can be found in Press et al. (1986).
Three-group resistant line
This method derives its resistance from the fact that one of the simplest resistant measures of a sample is the median. Data are divided into three groups depending on the rank of the x values. The left group contains the points with the lowest third of x values. In a time series this is equivalent to the first third of the series. Similarly, the middle and right groups contain points with the middle and highest third of ranked x values respectively. Next the x and y median values are determined for the three groups to give the points
The slope of the line is taken as the gradient of the line through the medians of the left and right groups:
The intercept of the line is calculated by finding the three lines with slope b0, then averaging their intercept:
The three-group resistant line method usually requires iteration. After the first pass to find a0 and b0, the process can be repeated on the residuals to find a1
147
and \ . The iterations are continued until the adjustment to the slope is sufficiently small in magnitude (at most 1%). The final slope and intercept is the sum of those from each iteration. Further information can be found in Hoaglin et al. (1983).
Logistic Regression
Linear regression has been generalized under the field of generalized linear modelling, of which logistic regression is a special case. This method utilizes the binomial distribution and can therefore be used to model counts of extreme events. Often in a series, the variance of the residuals (from the linear model) varies with the magnitude of the data. This goes against the assumptions of least squares regression, which assumes residuals to have constant variance, but is a natural element of the binomial distribution and logistic regression. Therefore data do not need to be normalized. The logistic regression model expresses the probability of a success (e.g. an event above a particular threshold) as a function of time:
Since the probability of a success is in the range [0,1], it needs to be transformed to the range (-, +) using a link function:
Model fitting can be done using a maximum likelihood method. Further information about logistic regression, together with an example using extreme precipitation in Switzerland, can be found in Frei and Schr (2001).
Confidence intervals for least squares
Often the standard deviations . for the observations are not known. If we assume that the linear model does fit well and that all observations have the same standard deviation , the assumption that the residuals are normally distributed around the linear model implies that:
148
with N-2 appearing in the denominator because two parameters are estimated. From the above, it can be shown that the regression coefficient b will be normally distributed with variance:
Since the variance of b is estimated, Student's t-distribution is used to define the multiplier t for the confidence limits for the regression coefficient:
The assumption that the residuals are normally distributed can be tested with a quantile-quantile (Q-Q) plot of the residuals against the quantiles from a Gaussian distribution. For further information, see Wilks (1995) or Press et al. (1986).
Linear Correlation
The linear correlation coefficient (Pearson product-moment coefficient of linear correlation) is used widely to assess relationships between variables and has a close relationship to least squares regression. The correlation coefficient is defined by: the ratio of the covariance of x and y to the product of their standard deviations. In a least squares linear model, the variance of the predictand can be proportioned into the variance of the regression line and the variance of the predictand around the line: SST=SSR+SSE Sum of Squares Total = Sum of Squares Regression + Sum of Squares Error In a good linear relationship between the predictor and predictand, SSE will be much smaller than SSR i.e. the spread of points around the line will be much smaller than the variance of the line. This goodness of fit can be described by the coefficient of determination:
149
= variance of predictand explained by the predictor It can be shown that the coefficient of determination is the same as the square of the correlation coefficient. The correlation coefficient can therefore be used to assess how well the linear model fits the data. Assessing the significance of a sample correlation is difficult, however, as there is no way to calculate its distribution for the null hypothesis (that the variables are not correlated). Most tables of significance use the approximation that, for a small number of points and normally distributed data, the following statistic is distributed for the null hypothesis like Student's t-distribution:
The common basis of the correlation coefficient and least squares linear regression means that they share the same shortcomings such as limited resistance to outliers. See Wilks (1995) or Press et al. (1986) for further information.
Spearman rank-order correlation coefficient
Non parametric correlation statistics are an attempt to overcome the limited resistance and robustness of the linear correlation coefficient, as well as the uncertainty in determining its significance. If x and y data values are replaced by their rank, we are left with the set of points (i,j), i,j=1,N which are drawn from an accurately known distribution. Although we are ignoring some information in the data, this is far outweighed by the benefits of greater robustness and resistance. The Spearman rank-order correlation coefficient is just the correlation coefficient of these ranked data. Significance is tested as for the linear correlation coefficient using the last equation, but in this case the approximation does not depend on the distributions of the data. See Press et al. (1986) for further information.
Kendall-Tau
Kendall's Tau differs from the Spearman rank-order correlation in that it only uses the relative ordering of ranks when comparing points. It is calculated over all possible pairs of data points using the following:
150
where concordant is the number of pairs where the relative ordering of x and y are the same, discordant where they are the opposite, sameX where the x values are the same and sameY where the y values are the same. T is approximately normally distributed with zero mean and variance:
One advantage of Kendall's tau over the Spearman coefficient is the problem of assigning ranks when data are tied. Kendall's tau is only concerned whether a rank is higher or lower than another, and can therefore be calculated by comparing the data themselves rather than their rank. When data are limited to only a few discrete values, Kendall's tau is a more suitable statistic. See Press et al. (1986) for further information.
Resampling
Resampling procedures are used extensively by climatologists and could be used to assess the significance of a linear trend. The bootstrap method involves randomly resampling data (with replacement) to create new samples, from which the distribution of the null hypothesis can be estimated. Therefore no assumption needs be made about the sample distribution. If enough random samples are generated, the significance of an observed linear trend can be assessed by where it appears in the distribution of trends from the random samples. A problem, however, is that the maximum likelihood derivation of the least squares estimate for the linear trend assumed that data residuals about the line were normally distributed. Therefore if the distribution of the residuals is not Gaussian, then the least squares estimate is not valid. Still, bootstrapping could be used to test the significance of a least squares linear trend, given that this may not be the best trend estimate. An important assumption in resampling is that observations are independent. Zwiers (1990) showed that, for the case of assessing the significance of the difference in two sample means, the presence of serial correlation greatly affected the results. A method has been proposed by Ebisuzaki (1997) whereby random samples are taken in the frequency domain (with random phase) to retain the serial correlation of the data in each sample. 3.3.5 Additional information listed within the questionnaire There were just a few answers on the last question (for additional information if any) into the submitted questionnaires. They were mainly related to the data problems. For example a representative from Spain pointed out lack of data or poor quality data when the need is for data of high temporal and spatial
151
resolution. In many cases it could be difficult to provide good quality gridded data of daily meteorological values for grids smaller than 40 km2. In a similar way the respond from Slovakia was: pure quality of some agrometeorological data (soil moisture and agrotechnical data); some data are still stored on the paper sheets (part of soil moisture and phenological data)
3.4 Concluding remarks Obviously all European countries have historical long-term meteorological records. Differences in the field, however, are observed in the terms of their temporal and spatial resolution, area coverage, especially availability for end users. Some of the European institutions/countries are ready to provide primary weather data, while others at this stage are open to present mainly secondary data, including various indices, maps, etc. A possible solution of this problem is the application of gridded meteorological data sets, which are freely available on the internet. The same conclusion could be directed to the agrometeorological data, although its measurement is not so spread across the continent as the collection of weather data. As future steps, main research agrometeorological activities, for example in Romania, should include: Calibration and testing the performance of models to simulate the soil moisture dynamic in different climate conditions and soil depths; Development and improving agrometeorological applications through using remote sensing, GIS and modeling techniques coupled with decision support systems for agriculture; Quantification the effects of climatic variability on the main components of water balance (evapotranspiration, crop water requirement, soil moisture deficit, effective rainfall) using e.g. CROPWAT model; Analyze the spatial and temporal evolution of soil moisture dynamic over a long-term period in order to identify the, trends, intervals and zones with high risk at the occurrence of extreme climatic events (droughts, excess moisture, etc.). Models such as numerical weather models, global and regional climate models as well as weather generators are more and more applied in Europe. However, numerical weather models still represent short reanalysis data set, hence the attention on the long term variations and trends of given agroclimatic indices should be focused on application of global and especially regional climate models as well as weather generators. There is no doubt that the long-term climate series should be homogeneous. Inhomogeneous data sometimes can introduce wrong results and especially conclusions related to the issue of climate change. The homogeneous series do not replace uncorrected (raw) data. The homogeneous series can be changed in the future. The raw data can be used for completing missing data or to improve
152
the break detection or correction. There are even some analyses, where unadjusted data are preferred (e.g. Peterson et al., 1998). These are often station specific studies that do not involve long-term trend analysis. Therefore, it is important to preserve the original data as well as the homogeneityadjusted versions. Also, the original data need to be preserved because new and better or improved approaches to homogeneity adjustments will probably be developed in the near future. During the last two decades, considerable work has been done on homogeneity testing and data adjustments and research will continue in this field. It seems that various statistical methods and software are applied on time series in Europe. All this shows the existing of well organized capacity building as well as the opportunity to apply and develop know how in the field. However, a strategy for development of a common statistical approach analyzing time series from different European regions should be considered. In this sense, the achievement of consistent and/or coherent results in this sphere in the European Union and beyond would be more real. Finally, those readers who consider the above information as insufficient, they may wish to refer to the filled in questionnaires, edited and posted on the COST 734 web page.
3.5 Acknowledgments Special thanks are addressed to all colleagues, who submitted filled in questionnaires on trends of agroclimatic indices and simulation model outputs in Europe. The authors wish also to thank to the COST 734 chairman, Prof. S.Orlandini, for his help and useful suggestions, as well as to any colleagues, who replied to the e-mail requests for assistance and provided additional information in the field. Finally, warm thanks are directed to the COST office in Brussels for its financial support related to this survey.
3.6 References
Acock B. and Acock M.C., 1991. Potential for using long-term field research data to develop and validate crop simulators. Agron. J. 83:56-61. Alexander L.V., X. Zhang, T.C. Peterson, J. Caesar, B. Gleason, A.M.G. Klein Tank, M. Haylock, D. Collins, B. Trewin, F. Rahimzadeh, A. Tagipour, K. Rupa Kumar, J. Revadekar, G. Griffiths,. L. Vincent, D.B. Stephenson, J. Burn, E. Aguilar, M. Brunet, M. Taylor, M. New, P. Zhai, M. Rusticucci, J.L. Vazquez Aguirre, 2006. Global observed changes in daily climate extremes of temperature and precipitation. J. Geophys. Res., 111, D05109, doi:10.1029/2005JD006290. Alexandersson A., 1986. A homogeneity test applied to precipitation data. J. Climatol., 6(6): 661-675.
153
Alexandersson H., 1986. A homogeneity test applied to precipitation data. Int J Climatol 6:661-675. Alexandersson H. and Moberg A., 1997. Homogenization of Swedish temperature data. Part: A homogeneity test for linear trends. Int J Climatol 17:25-34 Alexandrov V. and Hoogenboom G., 2001. Climate Variations and Agricultural Crop Production in Georgia, USA. Climate Research , 17(1): 33-43. Anadranistakis M., A. Mamara, P. Fragkouli, T. Andreadis, 2006. Verification of prognostic fields of surface wind of numerical weather prediction models LM-COSMO & ECMWF. Proc. 8th Conference on Meteorology - Climatology - Atmospheric Physics, Athens, May 24-26, 2006. Anagnostopoulou C., K. Tolika, C. Michailidou, P. Maheras, 2006. Extreme rainfall events over the Greek area: theory and application of the generalized extreme value (GEV) distribution and pareto distribution. Proc. 8th Conference on Meteorology - Climatology Atmospheric Physics, Athens, May 24-26, 2006. Argiriou A.A., N. Ioannou, D. Skourellos, K. Lagouvardos, V. Kotroni, 2006. Correction of surface temperature and wind speed predictions of the BOLAM weather forecasting model using artificial neural networks. Proc. 8th Conference on Meteorology Climatology - Atmospheric Physics, Athens, May 24-26, 2006. Arnold C.D. and Elliot W.J., 1996. CLIGEN Weather Generator Predictions of Seasonal Wet and Dry Spells in Uganda. Trans. of ASAE 39(3):969-972. Aschwanden A. et al., 1996. Bereinigte Zeitreihen: Ergebnisse des Projekts Klima90. Klimatologie der Schweiz. 4 Volumes available from the Swiss Federal Office of Meteorology and Climatology. Auer I, R. Bhm, A. Jurkovic, W.Lipa, A. Orlik, R. Potzmann, W. Schner, M. Ungersbck, C. Matulla, K. Briffa, P.D. Jones, D. Efthymiadis, M. Brunetti, T. Nanni, M. Maugeri, L. Mercalli, O. Mestre, J.M. Moisselin, M. Begert, G. Mller-Westermeier, V. Kveton, O. Bochnicek, P. Stastny, M. Lapin, S. Szalai, T. Szentimrey, T. Cegnar, M. Dolinar, M. Gajic-Capka, K. Zaninovic, Z. Majstorovic, E. Nieplova, 2006. HISTALP-historical instrumental climatological surface time series of the Greater Alpine Region. Int. J. Climatology,27:17-46. Auer I. and Boehm R., 1997. Data, metadata and the question of homogeneity for the Austrian climatological network. Proceedings of the first seminar for homogenization of surface climatological data. Budapest: Hungarian Meteorological Service, pp 83-96. Auer I. and Boehm R., 1994. Combined temperature-precipitation variations in Austria during the instrumental period. Theor Appl Climatol 49:161-174. Avgoustopoulou E. and Papageorgiou I., 2006. Integration of the nesting technique into the LM model Proc. 8th Conference on Meteorology - Climatology - Atmospheric Physics, Athens, May 24-26, 2006. Bannayan M. and Crout N.M.J., 1999. A stochastic modelling approach for real time forecasting of winter wheat yield. Field Crops Research, 62: 85-95. Begert M., T. Schlegel, W. Kirchhofer, 2005. Homogeneous temperature and precipitation series of Switzerland from 1864 to 2000. Int. J. Climatol.,25:65-80. Boehm R., I. Auer, M. Brunetti, M. Maugeri, T. Nanni, W. Schoener, 2001. Regional temperature variability in the European Alps: 1760-1998 from homogenized instrumental time series. Int. J. Climatol. 21: 1779-1801. Boisvenue C. and Running S.W., 2006. Impacts of climate change on natural forest productivity-evidence since the middle of the 20th century. Glob. Change Biol., 12: 862882.
154
Boote K.J., J.W. Jones, N.B. Pickering, 1996. Potential uses and limitations of crop models. Agronomy Journal 88: 704-716. Cabelguenne M., P. Debaeke, A. Bouniols, 1999. EPIC phase, a version of the EPIC model simulating the effects of water and nitrogen stress on biomass and yield, taking account of developmental stages: validation on maize, sunflower, sorghum, soybean and winter wheat. Agricultural Systems 60: 175-196. Camarero J.J. and Gutierrez E., 2004. Pace and pattern of recent treeline dynamics response of ecotones to climatic variability in the Spanish Pyrenees. Climatic Change, 63: 181-200. Casanova D., J. Goudriaan, A.D. Bosch, 2000. Testing the performance of ORYZA1, an explanatory model for rice growth simulation, for Mediterranean conditions. European Journal of Agronomy 12: 175189. Caussinus H. and Lyazrhi F., 1997. Choosing a linear model with a random number of change-points and outliers. Ann Inst Statist Math 49:761-775 Caussinus H. and Mestre O., 1997. New mathematical tools and methodologies for relative homogeneity testing. Proceedings of the first seminar for homogenization of surface climatological data. Budapest: Hungarian Meteorological Service, pp 63-82 Chapman L.J. and Brown D.M., 1978. The Climates of Canada for Agriculture. Canada Land Inventory Report No. 3. Revised 1978. Environment Canada, Lands Directorate, 24 pp. Chmielewski F.M., A. Mueller, E. Bruns, 2004. Climate changes and trends in phenology of fruit trees and field crops in Germany, 1961-2000. Agric. For. Meteorol., 121: 69-78. Conrad V. and Pollak C., 1950. Methods in climatology. Cambridge: Harvard University Press, 459 pp Craddock J.M.,1979. Methods of comparing annual rainfall records for climatic purposes. Weather. 34(9): 332-346. Dalezios N., L. Tsidarakis, S. Spanos, 1994. Testing homogeneity of temperature and precipitation series in Greece. Proc. International symposium / workshop on climatic variability and impact to agriculture, Volos, April 21, 1994. Danuso F. et al., 1997. CLIMAK reference manual. DPVTA, University of Udine, Italy, 36 p. Defila C. and Clot B., 2001. Phytophenolgical trends in Switzerland. Int J. Biometeorol 45: 203-207. Della-Marta P.M., 2006. A method of homogenizing the extremes and mean of daily temperature measurements. J. of Climate,19(17): 4179-4197 Duchene E. and Schneider C., 2005. Grapevine and climatic changes: a glance at the situation in Alsace. Agron. Sust. Dev., 25: 93-99. Easterling D.R., T.C. Peterson, T.R. Karl, 1996. On the development and use of homogenized climate data sets. J Climate 9:1429-1434. Ebisuzaki W., 1997. A method to estimate the statistical significance of a correlation when the data are serially correlated. J. Clim., 10: 2147-2153. European Climate Assessment and Dataset (ECA&D). 2003. http://www.knmi.nl/samenw/eca Finger R., 2007. Evidence of slowing yield growth - the example of Swiss cereal yields. Unpublished. Fischer G., H.T. Van Velthuizen, 1996. Climate change and global agricultural potential project: A case study of Kenya. International Institute for Applied Systems Analysis, Laxenburg, Austria, 96 pp. Flichman G., 1995. Analysis of the Socio-Economic Impacts of Agricultural Reform in Certain European Regions: Competitiveness and Environmental Protection (Final Report for EU contract No. 4706 A). Commission of European Communities, DG VI Agriculture. IAM, Montpellier, France.
155
Frland E.J. and Hanssen-Bauer I., 1994. Homogenizing long Norwegian precipitation series. Journal of Climate 7:1001-1013. Fraderick K., H. Jansen, E. Kirk, F. Lunkeit, 2005a. The Planet Simulator: Green planet and desert world. Meteorol. Z. 14: 305314. Fraderick K., H. Jansen, E. Kirk, U. Lunksch, F. Lunkeit, 2005b. The Portable University Model of the Atmosphere (PUMA): Storm track dynamics and low frequency variability. Meteorol. Z., submitted. Fraderick K., H. Jansen, E. Kirk, U. Lunksch, F. Lunkeit, 2005c. The Planet Simulator: Towards a user friendly model - Meteorol. Z. 14: 299304. Frei C. and Schr C., 2001. Detection probability of trends in rare events: theory and application to heavy precipitation in the alpine region. J. Clim., 14: 1568-1584. Frei C. and Schr C., 2001. Detection probability of trends in rare events: theory and application of heavy precipitation in the Alpine region. Journal of Climate 14: 1568 1584. Frich P., L.V.Alexander, P. Della-Marta, B. Gleason, M. Haylock, A.M.G.K. Tank, T. Peterson, 2002. Observed coherent changes in climatic extremes during the second half of the twentieth century. Clim. Res., 19: 193-212. Goudriaan J. and Van Laar H.H., 1994. Modeling Potential Crop Growth Processes: Textbook with Exercises. Kluwer Academic Publishers, Dordrecht. Grabherr G., M. Gottfried, H. Pauli, 2001. Long term monitoring of mountain peaks in the Alps. Tasks Veg. Sci., 35: 153-177. Guerif . and Duke C., 1998. Calibration of the SUCROS emergence and early growth module for sugar beet using optical remote sensing data assimilation. European Journal of Agronomy, 9: 127136. Hawkins D.M., 2001. Fitting multiple change-points to data. Comput Statist Data Anal 37:323-341. Haylock M., 2004. Linear Regression Analysis for STARDEX. Climatic Research Unit. Heino R., 1997. Metadata and their role in homogenization. Proceedings of the first seminar for homogenization of surface climatological data. Budapest: Hungarian Meteorological Service, pp 5-8. Hoaglin D.C., F. Mosteller, J.W. Tukey, 1983. Understanding robust and exploratory data analysis. Wiley. 129-165. Hoogenboom G., 2000. Contribution of agrometeorology to the simulation of crop production and its application. Agricultural and Forest Meteorology, 103: 137-157. Hornyi A., S. Kertsz, L. Kullmann, G. Radnti, 2006. The ARPEGE/ALADIN mesoscale numerical modeling system and its application at the Hungarian Meteorological Service. Idjrs, vol. 110 3-4, 203-227. Horvth ., I. Geresdi, K. Csirmaz, 2006. Numerical simulation of a tornado producing thunderstorm: A case study Idjrs, vol. 110 3-4, 279-297. Horvth Sz., L. Makra, J. Mika, 2005. 20th century variations of the soil moisture content in east Hungary acta climatologica et chorologica. universitatis szegediensis 38-39: 85-95. Hoskins B.J. and Simmons A.J., 1975. A multi-layer spectral model and the semi-implicit method. Quarterly Journal of the Royal Meteorological Society, 101: 637655. Jacobson B.M., J.W. Jones, S.W. Welch, 1995. Decision support system to assess agronomic, economic, and environmental impacts of soybean and corn management, ASAE Paper No. 952696, Am. Soc. Agric. Engineering, St. Joseph, MI, p. 17 James I.N., 1994. Introduction to circulating atmospheres. Cambridge University Press.
156
Johnson G.L., C.L. Hanson, S.P. Hardegree, E.B. Ballard, 1996. Stochastic Weather Simulation: Overview and analysis of two commonly used models. Journal of Applied Meteorology 35:1878-1896. Jones G.V. and Davis R., 2000. Climate influences on Grapewine phenology, grape composition, and wine production and quality for Bordeaux, France. Am. J. Enol.Vitic., 51: 249-261. Jones P.D. and Moberg A., 2003. Hemispheric and large scale surface air temperature variations: an extensive revision and an update to 2001. J. Clim., 16: 206-223. Kercheva M., 2004. Information base for modeling components of soil water balance and assessment of agroecological risks. PhD thesis (in Bulgarian) Klanderud K. and Birks H.J.B., 2003. Recent increases in species richness and shifts in altitudinal distributions of Norwegian mountain plants. The Holocene, 13: 1-6. Klein Tank A.M.G. and Konnen G.P., 2003. Trends in indices of daily temperature and precipitation extremes in Europe, 194699. Journal of Climate 16: 36653680. Kljuev N.N., 2001. Russia and its Regions. Nauka, Moscow, 214 pp. (in Russian). Kotroni V., K. Lagouvardos, S. Lykoudis, D. Lalas, 2006. Climatic projections in the eastern Mediterranean using the regional climatic model PRECIS. Proc. 8th Conference on Meteorology - Climatology - Atmospheric Physics, Athens, May 24-26, 2006. Kouchowski K. (ed.)., 1990. Materials for climate history in period of instrumental observation. Wydawnictwo Uniwersytetu dzkiego. d, pp. 116-143. Kullman L., 2001. 20th century climate warming and tree-limit rise in the southern Scandes of Sweden. Ambio, 30: 72-80. Kullman L., 2002. Rapid recent range-margin rise of tree and shrub species in the Swedish Scandes. J. Ecol., 90: 68-77. Kundzewicz Z.W., M. Parry, W. Cramer,. J.I Holten, S.G. Shiyatov, M.M. Terentev, V.V. Fomin, 2005. Spatiotemporal dynamics of forest-tundra communities in the polar Urals. Russian J. Ecol., 36: 69-75. Lavielle M., 1998. Optimal segmentation of random processes. IEEE Transactions on Signal Processing 46(5):365-1373. Lawless C. and Semenov M.A., 2005 Assessing lead-time for predicting wheat growth using a crop simulation model Agric Forest Meteorology 135:302313. Leemans R. and Solomon A.M., 1993. The potential response and redistribution of crops under a doubled CO2 climate. Clim. Res. 3: 79-96. Libiseller C. and Grimvall A., 2002 .Performance of Partial Mann Kendall Tests for Trend Detection in the Presence of Covariates, Environmetrics 13: 71-84 Maracchi G., 2003 Meteorologia e climatologia applicate.Florence, Italy: Editrice LUniverso. Matalas N.C., 1967. Mathematical assessment of synthetic hydrology. Water Resources Research 3(4):937-945. Menzel A., 2003. Plant phenological anomalies in Germany and their relation to air temperature and NAO. Climatic Change, 57: 243. Menzel A., T.H. Sparks, N. Estrella, E. Koch, A. Aasa, R. Ahas, K. Alm-Koeler, P. Bissoli, O. Braslavska, A. Briede, F.M. Chmielewski, Z. Crepinsek, Y. Curnel, . Dalh, C. Defila, A. Donnelly, Y. Filella, K. Jatczak, F. Mge, A. Mestre, . Nordli, J. Peuelas, P. Pirinen, V. Remiova, H. Scheifinger, M. Striz, A. Susnik, A. VanVliet, F.E. Wielgolaski, S. Zach, A. Zust, 2006. European phenological response to climate change matches the warming pattern. Glob. Change Biol., 12: 1969-1976. Menzhulin G.V., 1998. Impact of global warming of climate on agriculture in Russia. In: Influence of Global Change in the Environment and Climate on Functioning of Economy of Russia. Moscow, Russia, pp. 48-73 (in Russian).
157
Mestre O., 1999. Step-by-step procedures for choosing a model with change points. Proceedings of the second seminar for homogenization of surface climatological data, WCDMP 41/WMO-TD 962. Geneva: WMO, pp 15-26. Mestre O., 2000. Mthodes statistiques pour l'homognisation de longues sries climatiques. Thse de doctorat de l'Universit Paul Sabatier. Toulouse: Universit Paul Sabatier. Mika J., Sz. Horvth, L. Makra, Z. Dunkel, 2005. The Palmer Drought Severity Index (PDSI) as an indicator of soil moisture. Physics and Chemistry of the Earth vol. 30, 223-230. Mize J.H. and Cox J.G., 1968. Essentials of simulation. Prentice Hall, Englewood Cliffs, NJ. Moisselin J.M. and Mestre O., 2002. Research, digitization and homogenization of long-term data series. Actes du Colloque "La Prennisation et la Valorisation des donnes scientifiques et techniques" (Ensuring Long-Term Preservation and adding Value to scientific and technical data) organis par le CNES du 5 au 7/11/2002. Toulouse. Moisselin J.M., M. Schneider, C. Canellas, O. Mestre, 2002. Les changements climatiques en France au XXe siecle. Etude des longues sries homognises de donnes de temprature et de prcipitations. La Mtorologie 38:45-56. Norrant C. and Douguoudroit A., 2006. Monthly and daily precipitation trends in theMediterranean. Theor. Appl. Climatol., 83: 89-106. Oikonomou E., V. Kotroni, K. Lagouvardos, 2006. Forecasting extreme events over the eastern Mediterranean by use of 3DVAR assimilation of synoptic and satellite data. Proc. 8th Conference on Meteorology - Climatology - Atmospheric Physics, Athens, May 24-26, 2006. Olesen J.E. and Bindi M., 2004. Agricultural impacts and adaptations to climate change in Europe. Farm Policy Journal, 1: 36-46. Plfai I., 1991. The drought of year 1990. Vizugyi Kozlemenyek 1991.2.fuzet (in Hungarian). Pauli H., M. Gottfried, G. Grabherr, 2001. High summits of the Alps in a changing climate. The oldest observation series on high mountain plant diversity in Europe.Fingerprints of Climate Change - Adapted Behaviour and Shifting Species Ranges, Walther, G.-R., Burga, C.A., Edwards, P.J., Eds., Kluwer Academic Publisher, Norwell, Massachusetts, 139-149. Penning de Vries F.W.T. and Teng P.S. (eds.), 1993. Systems approach to agricultural development. Kluwer Academic Publishers, the Netherlands. Peterson D.C. and Easterling D.R., 1994. Creation of homogeneous composite climatological reference series. Int J Climatol 14:671-679. Peterson T.C., D.R. Easterling, T.R. Karl, P. Groisman, N. Nicholls, N. Plummer, S. Torok, I. Auer, R. Boehm, D. Gullet, L. Vincent, R. Heino, H. Tuomenvirta, O. Mestre, S. Szentimrey, J. Salinger, E.J. Forland, I. Hanssen-Bauer, H. Alexandersson, P. Jones, D. Parker, 1998. Homogeneity adjustments of In Situ atmospheric climate data: a review. Int J Climatol 18:1493-1517. Petriccione B., 2003. Short-term changes in key plant communities of Central Apennines (Italy). Acta Bot. Gall., 150: 545-562. Peuelas J. and Boada M., 2003. A global change-induced biome shift in the Montseny mountains (NE Spain). Glob. Change Biol., 9: 131-140. Pnevmatikos J.D., M. Th. Tzanakou, G.A. Kornaros, 2006. Homogeneity and quality control of rainfall time series. Proc. 8th Conference on Meteorology - Climatology Atmospheric Physics, Athens, May 24-26, 2006. Press W.H. et al., 1992. Numerical Recipes in Fortran. The Art of Scientific Computing, 2nd Ed., Cambridge University Press, Cambridge Press W.H., B.P. Flannery, S.A. Teukolsky, W.T. Vetterling, 1986. Numerical recipes: The art of scientific computing. Cambridge Univ. Press, 488-493.
158
Pruchnicki J., 1987. Metody opracowa klimatologicznych. PWN, Warszawa, pp. 42-61. R Development Core Team, 2006. R Foundation for statistical computing, Vienna. http://www.r-project.org. Racsko P., L. Szeidl, M.A. Semenov, 1991. A serial approach to local stochastic weather models. Ecological Modelling, 57 :27-41. Richardson C.W., 1982. Dependence structure of daily temperature and solar radiation. Trans. of ASAE 25:735-739. Richardson C.W., and Wright D.A., 1984. WGEN: A model for generating daily weather variables. USDA-ARS, 235 p. Robert L., L. Wilby, L.W. George Perry, 2006. Climate change, biodiversity and the urban environment: a critical review based on London, UK Progress in Physical Geography, 30(1): 73-98. Rodrguez-Puebla S., M. Ayuso, M.D. Fras, L.A. Garca-Casado, 2007. The effects of climate variations on winter cereal production in Spain. Climate Research. Sanz Elorza M. and Dana E.D., 2003. Changes in the high mountain vegetation of the central Iberian Peninsula as a probable sign of global warming. Ann. Bot., 92: 273-280. Sau F., K. Boote, R. Ruiz-Nogueira, 1999. Evaluation and improvement of CROPGROsoybean model for a cool environment in Galicia, northwest Spain. Field Crops Research, 61: 273-291. Scherrer S., C. Appenzeller, M.A. Liniger, 2006. Temperature trends in Switzerland and Europe: implications for climate normals. Int. J. Climatol. 26: 565580. Schmidli J. and Frei C., 2006. Trends of heavy precipitation and wet and dry spells in Switzerland during the 20th century. Int. J. Climatol. 25: 753771. Schmidli J., C. Schmutz, C. Frei, H. Wanner, C. Schr 2002. Mesoscale precipitation variability in the region of the European Alps during the 20th century. Int. J. Climatol. 22: 10491074. Scholefield P., 1999. The importance of the homogenization of climate data time series in the context of the world climate data and monitoring programme. Proceedings of the first seminar for homogenization of surface climatological Data. Budapest: Hungarian Meteorological Service, pp 1-3. Semenov M.A., 2007. Development of high-resolution UKCIP02-based climate change scenarios in the UK Agric Forest Meteorology 144:127138. Semenov M.A. and Barrow E.M., 1997. Use of a stochastic weather generator in the development of climate change scenarios. Climatic Change, 35:397-414. Semenov M.A. and Brooks R.J., 1999. Spatial interpolation of the LARS-WG stochastic weather generator in Great Britain. Climate Research, 11:137-148. Semenov M.A., R.J. Brooks, E.M. Barrow, C.W. Richardson, 1998. Comparison of the WGEN and LARS-WG stochastic weather generators for diverse climates. Climate Research 10:95-107. Sharpley A.N. and Williams J.R., 1990. EPIC-Erosion/Productivity Impact Calculator: 1. Model Documentation. US Department of Agriculture Technical Bulletin N 1768, 235 p. Shein K.A., 1999. The role of metadata in climate data homogeneity. Proceedings of the second seminar for homogenization of surface climatological data, WCDMP 41/WMOTD 962. Geneva: WMO, pp 195-202. Shen S.S.P., H. Yin, K. Cannon, A. Howard, S. Chetner, T.R. Karl, 2005. Temporal and spatial changes of the agroclimate in Alberta, Canada, from 1901 to 2002. Journal of Applied Meteorology 44(7):1090-1105. Shvidenko A. and Nilsson S., 2003. A synthesis of the impact of Russian forests on the global carbon budget for 1961-1998. Tellus, 55B:391-415.
159
Sirotenko O.D. and Abashina E.V., 1998. Agroclimatic resources and physical-geographic zones of Russia under global warming. Meteorology and Hydrology, 3, 92-103 (in Russian). Sirotenko O.D., E.V. Abashina, V.N. Pavlova, 1995. Sensitivity of Russian agriculture to changes in climate, chemical composition of the atmosphere, and soil fertility. Meteorology and Hydrology, 4:107-114 (in Russian). Skourkeas F., Kolyva-Machera, C. Anagnostopoulou, K. Tolika, P. Maheras, 2006. Model estimation of mean maximum and minimum temperatures in Greece. Proc. 8th Conference on Meteorology - Climatology Atmospheric Physics, Athens, May 24-26, 2006. Sneyers R., 1999. Homogenizing time series of climatological observations: the search and adjustment for inhomogeneities principles of methodology and example of results. Proceedings of the second seminar for homogenization of surface climatological data, WCDMP 41/WMO-TD 962. Geneva: WMO, pp 5-14. STARDEX, 2004. Diagnostic Extremes Indices Software. User Information Version 3.3.1, June 2004. Stckle C.O., Campbell, G.S. and Nelson. R. 1999. ClimGen manual. Biological Systems Engineering Department, Washington State University, Pullman, WA, 28 p. Stockle C.O., P. Steduto, R.G. Allen, 1998. Estimating daily and daytime mean VPD from daily maximum VPD. 5th Congress of the European Society of Agronomy, Nitra, The Slovak Republic. Supit I. and van der Goot (eds.) E., 2002. Updated System description of the WOFOST crop growth simulation model as implemented in the Crop Growth Monitoring System (CGMS) applied by the European Commission http://www.iwan-supit.cistron.nl/~iwansupit/contents/ Szentimrey T., 2004. Multiple Analysis of Series for Homogenization (MASH); Verification procedure for homogenized time series, Proceedings of the Fourth Seminar for Homogenization and Quality Control in Climatological Databases, Budapest, Hungary, 6-10 October 2003; WMO, WCDMP-No. 56, pp. 193-201. Szentimrey T., 1999. Multiple analysis of series for homogenization (MASH). Proceedings of the second seminar for homogenization of surface climatological data, WCDMP 41/WMO-TD 962. Geneva: WMO, pp 27-46. Szinell Cs., A. Bussay, T. Szentimrey, 1998. Drought Tendencies in Hungary. International Journal of Climatology 18: 1479-1491. Szintai B. and Ihsz I., 2006. The dynamical downscaling of ECMWF EPS products with the ALADIN mesoscale limited area model:preliminary evaluation 253-277. Tank K.A.M.G., J.B. Wijngaard, G.P. Konnen, R. Bohm, G. Demaree, A.M. Gocheva, S. Mileta, L. Pashiardis, C. Hejkrlik, R. Kern-Hansen, P. Heino-Bessemoulin, G. MullerWestermeier, M. Tzanakou, S. Szalai, T. Palsdottir, D. Fitzgerald, S. Rubin, M. Capaldo, M. Maugeri, A. Leitass, A. Bukantis, R. Aberfeld, A.F.V. VanEngelen, E. Forland, Mietus, M., Coelho, F., Mares, C., Razuvaev, V., Nieplova, E., Cegnar, T., Lpez, J.A., Dahlstrom, B., Moberg, A., Kirchhofer, W., Ceylan, A., Pachaliuk, O., Alexander, L.V. and Petrovic, P. 2002. Daily dataset of 20th-century surface air temperature and precipitation series for the European Climate Assessment. Int. J. Climatol., 22: 14411453. Tank K.A.M.G. and Koennen G.P., 2003. Trends in indices of daily temperature and precipitation extremes in Europe. J. Clim., 16, 3665-3680. Torok, S. and Nicholls, N. 1996. A historical annual temperature dataset for Australia. Aust Meteor Mag 45:251-260.
160
Trouslard-Kerdiles V. and Grondona M.O., 1997. A case of combined use of crop simulation models and general linear models. Ecological Modelling, 99: 71-85. Tsuji G., G. Hoogenboom, P. Thornton, 1998. Understanding Options for Agricultural Production. Kluwer Acad. Publ., 399 pp. Tuomenvirta H. and Heino R., 1996. Climatic changes in Finland recent findings. Geophysica 32:61-75. Tuomenvirta H. and Alexandersson H., 1997. Review of the methodology of the Standard Normal Homogeneity Test (SNHT). Proceedings of the first seminar for homogenization of surface climatological data. Budapest: Hungarian Meteorological Service, pp 35-46 Uehara G. and Tsuji G.Y., 1993. The IBSNAT project. In Systems approach to agricultural development, p. 505-514. F.W.T. Penning de Vries and P.S. Teng, eds. Kluwer Academic Publishers, the Netherlands. van Ittersum M.K. and Rabbinge R., 1997. Explorative land use studies and their role in supporting regional decision making. In: ten Berge, H.F.M., Stein, A. (Eds.), Modelbased Decision Support in Agriculture, Proceedings of the INRA-KCW Workshop on Decision Support Systems, Laon, France, October 1997, Quantitative Approaches in Systems Analysis 15, pp. 75-82. Vincent L., 1998. A technique for the identification of inhomogeneities in Canadian temperature series. J Climate 11:1094-1104. Viner D., 2006. Tourism and its interactions with climate change. J. of Sustainable Tourism, 14: 317-322. Visual Numerics, 2001. IMSL, Fortran Subroutines for Mathematical Applications, http://www.vni.com/products/imsl/; Walther G.R., S. Beissner, C.A. Burga 2005. Trends in upward shift of alpine plants. J. Veg. Sci., 16: 541-548. Wilks D.S., 1995. Statistical Methods in the Atmospheric Sciences. Academic Press. 160-176. Williams G.D.V., 1975. An Agroclimatic Resource Index for Canada and its use in Describing Agricultural Land Losses. Unpub. Int. Rpt., Agriculture Canada, Ottawa Willmott C.J., 1982. Some comments on the evaluation of model performance. Bulletin of American Meteorological Society 63:1309-1313. World Meteorological Organization (WMO), 1990. On the statistical analysis of series of observations. Technical Note 143/WMO 415. Geneva: WMO. Zanis P.C., K. Repapis, D. Philandras, M. Melas, Kanakidou, 2006. Regional climate model simulations for Greece with RegCM3 Proc. 8th Conference on Meteorology - Climatology - Atmospheric Physics, Athens, May 24-26, 2006. Zwiers F.W., 1990. The effect of serial correlation on statistical inferences made with resampling procedures. J. Clim., 3: 1452-1461.
161
162
4. Satellite spectral climatic and biophysical data for warning purposes for european agriculture
4. SATELLITE SPECTRAL CLIMATIC AND BIOPHYSICAL DATA FOR WARNING PURPOSES FOR EUROPEAN AGRICULTURE Leonidas Toulios, Gheorghe Stancalie, Piotr Struzik, Mark Danson, Janos Mika, Zoltan Dunkel, Emmanouil Tsiros
Abstract In this study, the contribution of Earth Observation (EO) data (satellite-derived data) for warning purposes in agriculture due to climate variability and change is discussed. Climate variability and change is a global issue, which must be addressed with global models and global data are needed as input to these models. EO from space has a unique capacity to provide such global data sets continuously and consistently not only on this level, but also on the national and local levels and the use of alert and warning systems must be based on such data. Evaluation of the agricultural impacts represents an important contribution for the assessment of vulnerability of agricultural systems to climate variability and change. Some of the climate and biophysical variables essential for understanding and monitoring the climate system and the impact on agriculture can be efficiently observed from space since this technology enables their systematic, global and homogeneous measurement. Climate and agriculture research is generally based on data collected for other purposes, primarily for weather prediction. To make these data useful for climate impact and warning studies, it is usually necessary to analyze and process the basic observational raw data and integrate into models. In the frame of COST 734, satellite data records, e.g. series of observations over time that measures variables believed to be associated with climate variation and change, were surveyed and collected among European countries, based on a specific questionnaire. The analysis and the presentation in Tables of the data records which have been developed from operational satellite observations, presents the status of satellite climatic and biophysical data for warning purposes for agriculture, in Europe. Among European countries there is a great unhomogeneity concerning climatic and biophysical data received from satellite sensors or collected as satellite-derived ready products. Some of them are currently collecting satellite data for years and these data records could be useful for models for climate change impact studies. The main variables that are collected in operational or experimental way are land surface temperature and NDVI. Some examples of satellite images and satellite-derived products as referred in the questionnaire answers are also presented.
163
4.1 Introduction 4.1.1 The need of adaptation of agriculture to climate change European agriculture will face many challenges over the coming years such as international competition and population decline with long-range economic impacts. (Sivakumar et al., 2000). Climate variability and change will add to these pressures and will make the challenges more difficult and costly. The general consensus is that changes in temperature and precipitation normals will lead to adjustments in land and water regimes that will affect agricultural productivity. The recent droughts in Europe caused billions of Euros in crop damage (Kogan, 2000; Ciais et al., 2005). On the other hand exceptionally wet years engender similar devastating effects on production. Reliable seasonal forecasts of precipitation and temperature can have enormous positive economic impact for the global agricultural industry. Agriculture is responsible for an estimated one third of global warming and climate change. It is generally agreed that about 25% of the main greenhouse gas, carbon dioxide, is produced by agricultural sources, mainly deforestation and the burning of biomass. Most of the methane in the atmosphere comes from domestic ruminants, forest fires, wetland rice cultivation and waste products, while conventional tillage and fertilizer use account for 70% of the nitrous oxides (FAO, 2001). Climate change will impact agriculture by causing damage and gain at scales ranging from individual plants or animals to global trade networks. At the plant or field scale, climate change is likely to interact with rising CO2 concentrations and other environmental changes to affect crop and animal physiology. Impacts and adaptation (agronomic and economic) are likely to extend to the farm and surrounding regional scales. Climate change over the long-term, in particular global warming, could affect agriculture in a number of ways - the majority of which would threaten food security for the world's most vulnerable people: The overall predictability of weather and climate would decrease, making planning of farm operations more difficult. Climate variability might increase, putting additional stress on fragile farming systems. Climate extremes - which are almost impossible to plan for - might become more frequent. The sea-level would rise, threatening valuable coastal agricultural land, particularly in low-lying small islands.
164
Biological diversity would be reduced in some of the world's most fragile environments, such as mangroves and tropical forests. Climatic and agro-ecological zones would shift, forcing farmers to adapt, as well as threatening natural vegetation and fauna. The current imbalance of food production between cool and temperate regions and tropical and subtropical regions could worsen. Distribution and quantities of fish and seafoods could change dramatically, wreaking havoc in established national fishery activities. Pests and vector-borne diseases would spread into areas where they were previously unknown (FAO, 2005). The natural variability of rainfall, temperature and other conditions is the main factor behind variability in agricultural production, which in turn is one of the main factors behind food insecurity. Climate extremes - violent and unusual events such as floods, drought, and storms - though by nature more apparently dramatic, have less overall effect on agricultural production than chronic climate deficiencies. Both climate variability and climate extremes may increase as a result of global warming. Under a changing climate, the role of agriculture as provider of environmental and ecosystem services will further gain importance. The recent reforms of the Common Agricultural Policy (CAP) have been a first step towards a framework for the sustainable development of EU agriculture. Future adjustments of the CAP could provide opportunities to examine how to better integrate adaptation to climate change in agriculture support programmes and to promote good farming practices which are compatible with the new climate conditions and which contribute proactively to warning purposes in agriculture and to preserving and protecting the environment. The EU's research strategy places a strong emphasis on climate change, both in terms of predictive capacity, modelling and adaptation strategies. Long-term comprehensive and Europe-wide high-resolution datasets and models are needed. Coordination between data centres, information systems and networks should be improved. Access to existing climate, meteorological, satellite and integrate data relevant for adaptation should also be improved. The use of existing Community-supported information systems, e.g. alert and warning systems must be encouraged and brought to full potential. Updated synthesis reports on climate impacts, adaptation and vulnerabilities should be produced based on the results from the Framework Programmes and national research. Information and Communication Technologies (ICT) and their further developments will be a key instrument to support this adaptation process,
165
enabling relevant, flexible and speedy responses to the adaptation requirements for example for monitoring environmental changes, anticipating and assessing risks, managing crisis situations and the assessment of possible impacts on agriculture.
4.2 How the study on climate variability and change can benefit from space Substantial research is needed to achieve a better understanding of the dynamics of climate change affecting on farming conditions. EU scientists, in the frame of research actions, are collaborating to better understand seasonal variability of climate and apply that understanding to agricultural issues. Research methods, among others, include quantification and monitoring the state of the biosphere from satellites using remote sensing techniques. Space provides an ideal vantage point for the measurement of critical parameters for agricultural production, such as water availability, radiation and vegetation health, over large areas and at low cost (Fisher and Mustard, 2007). Research organisations are collaborating to incorporate space-based measurements into models and systems used to monitor and forecast global and domestic agricultural production. Earth observation from space provides satellite data that are necessary for the scientific community. Climate change is a global issue, which must be addressed with global models and global data are needed as input to these models. Earth observation has a unique capacity to provide such global data sets continuously and consistently not only on this level, but also on the national and local levels. The most important contributions that users of remote sensing data in impact studies of climatic change on agriculture, can make to breaking down the barriers to the use of satellite derived products is to provide very clear statements of their information requirements so that technology can develop to meet these requirements. Remote sensing role in characterizing, understanding and predicting climate variability and change is centred around providing the global scale observational data sets on the components of the climate system, their forcing, and the interactions with the entire Earth system. Understanding these interactions goes beyond observations, but includes developing and maintaining a modelling capability that allows for the effective use, interpretation and application of the data. The ultimate objective is to enable predictions of change in climate on time scales ranging from seasonal to multi-decadal. As new satellite measurements
166
enable this capability, the research community works to transition the observational capabilities to operational capabilities. The European Space Agency (ESA) develops and operates EO satellites, which are monitoring the environment with many different instruments. ESA thus contributes to the global efforts for providing needed data for checking the state of our Earth, in collaboration with other space agencies and EO operators. However, ESA does not only contribute to the provision of needed data, but also takes an active part in ensuring that these data are effectively being used by institutional users (www.esa.int/eo). 4.2.1 Making Sense of Satellite Data Space technology facilitates humanity and science with a global revolutionary view of the Earth through the acquisition of EO satellite data. Satellites capture information over different spatial and temporal scales and assist in understanding natural climate processes and in detecting and explaining climate change. Accurate EO data is needed to describe climate and biophysical processes by improving the parameterizations of different elements. One area of study is clouds and aerosols, and their role and effects on radiation. Vegetation status, soil moisture and other elements that play key roles in climate and agriculture are also studied. EO data should be able to detect changes of climate and biophysical elements that may be indicators of climate change. The Committee on EO satellites (CEOS) by the spatial agencies operates satellite that collect data from three domains, atmospheric, oceanic and terrestrial, necessary to establish key Earth system parameters, the so-called Essential Climate Variables (www.ceos.org/pages/CEOSResponse_1010A.pdf). Although almost all EO satellite systems were not specifically designed for climate monitoring, space agencies efforts have initiated a comprehensive climate data record that is forming the basis for better understanding the Earth climate system (GCOS, 2004; GCOS, 2006) Algorithms to produce climate and biophysical parameters from raw satellite observations should go through selection processes or participate in intercomparison programmes to ensure performance reliability. Many of these satellite-derived products, particularly those on medium and large scales, are occasionally or not at all used by potential beneficiaries due to a poor dissemination, difficulties in interpreting them and especially due to the lack of validations in operational activity. Within the quality management and
167
systematic product evaluation context, their calibration and validation is a major problem (Congalton and Green, 1999; Barnsley et al., 2002). Parameter datasets, obtained from satellite observations, should pass a quality control before they are accepted in global databases for impact, diagnostic or sensitivity studies. Calibration and validation are activities that endeavor to ensure that remote sensing products are highly consistent and reproducible. This is an evolving discipline that is becoming increasingly important as more long-term studies on global change are undertaken, and new satellite missions are launched. Calibration is the process of quantitatively defining the system responses to known, controlled signal inputs. Validation is the process of assessing, by independent means, the quality of the data products derived from the system outputs. Well-instrumented test sites and data sets for calibration should be supported, particularly for land applications, to provide calibration information to supplement/substitute for on-board calibration, ensuring continuity and reliability to access their data with minimal delay (www.esa.int/eo). The Committee on Earth Observation Satellites (CEOS) Program, through the Working Group for Calibration and Validation established a strategy on calibrating/validating data provided by global observation satellites, as well as a white charter on the Data quality guide for satellite observations relevant to the GEOSS Programs calibration and validation aspects (www.ceos.org).
4.3
Satellite-based variables and models potential in monitoring of crop production
The importance of systematic global observation for understanding climate change has been recognized by the global scientific community. Some of the variables essential for understanding and monitoring the climate system can be efficiently observed from space since this technology enables their systematic, global and homogeneous measurement. ESA initiated several global-scale projects in order to transform satellite data into meaningful parameters that provide insight into climate change issues. A synergetic use of data from the satellite instruments may provide information on the two following categories of variables: variables controlling the radiative fluxes between the surface and the overlying atmosphere, namely surface albedo (SA), land surface temperature (LST) and soil moisture (SM), and state variables of the radiative transfer problem, namely the snow cover (SC) and biophysical parameters such as the leaf area index (LAI) and the fractional vegetation coverage (FVC).
168
The most relevant variables that can be measured over land are: solar radiation, daily global albedo (the fraction of sunlight reflected back from the Earth), vegetation indices, LAI, land surface temperature, rainfall, cloud cover, fires and burnt areas, snow cover of both hemispheres, digital elevation maps of the ice sheet surfaces, glaciers evolution and land cover (Huang, 2005; Baret et al., 2007; Houborg and Boegh, 2008; Quattrochi and Luvall, 2004; Yao et al., 2008). Some of these variables are required as inputs to models designed to better understand and give an immediate view of climate change impact. Some example of vegetation products derived of satellite data are: Vegetation Indices (VI). A VI is a quantitative measure used to measure biomass or vegetative vigor, usually formed from combinations of several spectral bands, whose values are added, divided, or multiplied in order to yield a single value that indicates the amount or vigor of vegetation. VI are the most simple approach to characterize vegetation parameters and for evidencing their spatial and temporal variation for phenologic and change detection studies. Several vegetation indices were defined starting from the first simple ratio between infrared and red spectral channels (NDVI), Enhanced Vegetation Index (EVI), Soil-Adjusted Vegetation Index (SAVI), up to the more recent Soil and Atmosphere Resistant Vegetation Index (SARVI). Each of them was developed in order to evaluating vegetation characteristics and, in some cases, correcting radiometric aberrations (e.g. atmospheric distortion) and reducing the background effects in non-dense vegetated areas (Gitelson, 2004). Vegetation indices have also been used for agricultural drought quantification and mapping. In particular Vegetation Condition Index (VCI) (an extension of NDVI), Temperature Condition Index (TCI) (derived from CH4 data) and Vegetation Health Index (VHI) (a combination of VCI and TCI), have been developed and tested for drought monitoring in several areas of the world with different environmental and economic resource (Kogan 1990, 1994, 1995, 1997, 2000, 2001; Domenikiotis et al. 2002, 2004b; Bhuiyan et al., 2006; Tsiros et al., 2004). Lastly, VHI derived from a time series of NOAA/AVHRR ten-day images, for 20 successive hydrological years (1981-82 to 2000-01) has been recently used for agroclimatic classification (Tsiros et al., 2008). Maximum greenness during the growing season. This product represents the maximum value of the normalized difference vegetation index (NDVI) during the growing season, as determined from the seasonal trajectory of the NDVI curve
169
(www.ccrs.nrcan.gc.ca/optic/veg/index_e.php). Total greenness during the growing season. This product represents the area under the vegetation index (NDVI) curve for the growing season period, i.e. for the time during which surface temperature (obtained from satellite data) exceeded a temperature threshold (e.g. +10C, correspond approximately to the average daily air temperature of +5C). This product is prepared using the NDVI and surface temperature data, obtained from satellite measurements and corrected for atmospheric effects. The units are 'NDVI days'. The images are able to show the amount and duration of chlorophyll in the two growing seasons and the differences, both positive and negative, between the two years in various regions (Ji and Peters, 2004). Fraction of Photosynthetically Active Radiation (FPAR). FPAR is defined as the fraction of photosynthetically active radiation absorbed by a plant canopy. It excludes the fraction of incident PAR reflected from the canopy and the fraction absorbed by the soil surface or the combination of forest floor and understory, but includes the portion of PAR which is reflected by the soil/understory and absorbed by the canopy on the way back to space. Green FPAR refers to the fraction absorbed by green leaves only after the removal of the contribution of the supporting woody material to the PAR absorption. The instantaneous green FPAR is integrated over the day with a weight equal to the cosine of the solar zenith angle to obtain the daily green FPAR presented in the map. The daily green FPAR can be used as a parameter to convert the daily-absorbed PAR to daily total incident PAR (Knyazikhin et al., 1998; Yang et al., 2006). Absorbed Photosynthetically Active Radiation (APAR). APAR is the solar energy (400-700 nm) consumed by green canopy in the photosynthetic process. Remote sensing of APAR has been achieved through estimation of downwelling PAR at the surface, PARd, and the fraction of PAR intercepted by the canopy, FPAR. A new approach is proposed which defines APAR as the product of APART and RPAR. APART is the total PAR absorbed by all surface materials including canopy, soil, litter, etc., while RPAR is the ratio of the PAR absorbed by the green canopy only, to APART. Leaf area index (LAI). LAI is defined as half the total leaf area per unit ground surface area. In the calculation of LAI from NDVI, different algorithms are used for different vegetation types (Holben et al., 1980; Gardner and Blad, 1986; Aase et al., 1986; Wiegand et al., 1992; Li et al., 1996; Buermann et al., 2001). Errors in LAI estimation
170
are due to different factors such as the effect of vegetation canopy architecture. Crop monitoring and early yield assessment are important for agriculture planning and policy making at regional and national scales. Numerous crop growth simulation models are generated using crop state variables and climate variables at the crop/soil/atmosphere interfaces to get the pre-harvest information on crop yields (Monteith, 1977; Bouman et al., 1996). However, most of these models are limited to specific regions/periods due to significant spatial and temporal variations of those variables. Furthermore, the limited network of stations and incomplete climate data make crop monitoring and yield assessment a daunting task (Kogan, 1997). In addition, the meteorological data may miss important variability in vegetation production, which highlights the need for quantification of vegetation changes directly when monitoring climate impacts upon vegetation (Zhang et al., 2004a, 2004b, 2005). In this sense, remotely sensed metrics of vegetation activity have the following advantages: a unique vantage point, synoptic view, cost effectiveness, and a regular, repetitive view of nearly the entire Earths surface, thereby making them potentially better suited for crop monitoring and yield estimation than conventional weather data. Empirical relationships between the remotely sensed data and crop evolution and production estimates have been developed for monitoring and forecasting purposes since the early 1980s. Vegetation indices derived from data from the Advanced Very High Resolution Radiometer (AVHRR) were also used for crop prediction, environmental monitoring, and drought monitoring/assessment (Kogan, 1990, 1995; White et al., 1997; DabrowskaZielinska et al., 2002; Domenikiotis et al., 2002, 2004a, 2004b, 2005a, 2005b; Tsiros et al., 2004; Bhuiyan et al., 2006). Vegetation indices derived from satellite data are widely applied in real-time heat and drought monitoring and is shown to provide quantitative estimation of duration and effect on vegetation and provide an indication of the final yield (Kogan 1990, 1995; Dabrowska-Zielinska et al., 2002; Domenikiotis et al., 2004a, 2004b, 2005a, 2005b; Tsiros et al., 2004; Bhuiyan et al., 2006). Domenikiotis et al. (2005b) used VCI for zoning areas of cotton production and developed empirical relationships between the VCI values and cotton production in Greece, reaching 5% percentage deviation between the predicted and the actual production, at regional and country scale. Unfortunately, the satellite data are not always ideally suited for vegetation monitoring applications because of the lack of precise calibration, poor quality of geometric registration, and difficulties in cloud screening (Zhang et al., 2003).
171
4.3.1
Climate and biophysical data records in responding to climate change impacts on agriculture
The importance of understanding and predicting climate variation and change and avoid the impact on agriculture, has escalated significantly in the last decade. To integrate space research on global scale climate change, or in European or regional level, it is needed to develop data bases with Climate and Biophysical Data Records from operational satellites, to provide a framework for the use of data from existing and new instruments aboard satellites. The goal is to ensure that satellite data are processed, archived, and distributed to users in a manner that is scientifically defensible for monitoring, diagnosing, understanding, predicting, modeling, and assessing climate variation and change and possible impact on agriculture. Climate research is generally based on data collected for other purposes, primarily for weather prediction. To make these data useful for climate studies, it is usually necessary to analyze and process the basic observational raw data and integrate into models. Satellite data records are series of observations over time that measures variables believed to be associated with climate variation and change. These changes may be small and occur over long time periods (seasonal, interannual, and decadal to centennial) compared to the short-term changes that are monitored for weather forecasting. Thus, it is usually necessary to construct data records from data that span long time scales and sometimes from multiple data sources. Scientists must characterize and quantify the sensor, spatial and temporal errors of these diverse and frequently large data sets in order to produce a sufficiently accurate time series for studying trends in climate variability and change. Examples of climate data records which have been developed from operational satellite observations in USA are radiation budget (using POES and AVHRR /NOAA data), climate products like rainfall, rain frequency, snow cover, sea ice cover, clouds, water vapor (using DMSP SSM/I /NOAA data), vegetation indices NDVI and drought index (using POES/AVHRR data), atmospheric temperature (using POES/MSU /NOAA data), snow cover (using POES/AVHRR, GOES, Meteosat, GMS Visible imagery, DMSP/SSM/I data), clouds products (using POES/AVHRR, GOES, Meteosat, GMS Visible IR imagery/NASA and NOAA data), and precipitation ( using POES /AVHRR, GOES, Meteosat and GMS Visible IR imagery, DMSP SSM/I/ NOAAdata) (www.nesdis.noaa.gov, www.orbit.nesdis.noaa.gov, www.ncdc.noaa.gov).
172
4.4 Satellite instruments for climate change management METEOSAT is the European contribution to the global observing system required for both meteorology and climatology. The data and services are mainly focused on the requirements of operational meteorology, with the emphasis on support to operational weather forecasting. However, the data are of use for all areas of this discipline, including marine, agricultural and aviation meteorology, as well as, for example, climatology and the monitoring of planet Earth. The main service provided by the METEOSAT system is the generation of images of the Earth, showing its cloud systems both by day and by night, and the transmission of these images to the users in the shortest practical time. METEOSAT Second Generation (MSG) is the enhanced follow-on system to the previous generation of METEOSAT and includes a series of four geostationary meteorological satellites, along with ground-based infrastructure, that will operate consecutively untill 2018. The first MSG satellite to be launched was METEOSAT-8, in 2002. The second satellite followed up in December 2005. MSG serves the needs of Nowcasting applications and Numerical Weather Prediction in addition to provision of important data for climate monitoring and research. The main mission of MSG is the Imaging Mission and provides operational image data from the satellite's field of view. It provides particular support to now-casting applications through the following particular characteristics of the imaging mission: observation of air mass properties, thermodynamic and cloud physical parameters through dedicated spectral channels inherited from GOES and HIRS sounder instruments; dedicated high resolution imagery to monitor convective cloud evolution; high repeat rate of image data to observe rapid changes of clouds; near real-time processing of SEVIRI raw imagery to geometrically and radiometrically corrected data. The MSG system has brought major improvements in these services through the 12 spectral bands of its radiometer, the Spinning Enhanced Visible and InfraRed Imager (SEVIRI). SEVIRI delivers daylight images of the weather patterns with a resolution of 3 km, plus atmospheric pseudo-sounding and thermal information. The High Resolution Visible (HRV) channel has a resolution of 1 km. The full disc view allows frequent sampling, every 15 minutes, enabling monitoring of rapidly evolving events, which is useful for the weather forecaster in the swift recognition and prediction of dangerous weather phenomena such as thunderstorms, heavy rain, fog and explosive development of small but intense depressions, which can lead to devastating windstorms (Lacaze and Bergs, 2005)
173
EUMETSAT and the Satellite Application Facilities (SAFs) extract information from the processed SEVIRI data and turn it into products of particular use to meteorologists and climatologists, such as wind field diagrams, sea surface temperature, precipitation estimates and analyses of cloud coverage, height and temperature. In addition to the SEVIRI, the MSG satellites carry the Geostationary Earth Radiation Budget (GERB) instrument, which provides valuable data on reflected solar radiation and thermal radiation emitted by the Earth and atmosphere. Besides these two instruments, MSG satellites carry a comprehensive communications payload serving the needs of satellite operation, data communication and user data dissemination. This package includes also the Search and Rescue transponder and relays distress signals from ships, aircraft and others in peril (www.eumetsat.int). NOAA (www.nesdis.noaa.gov) has statutory responsibility for long-term archiving of the environmental data and has recently integrated several data management functions. These functions include careful monitoring of observing system performance for long-term applications, the generation of authoritative long-term records from multiple observing platforms, and the proper archival of and timely access to data and metadata. NOAA maintains a comprehensive archive of climate-related data and information spanning the ice age to the space age. NOAA also operates the US's operational satellite observing system. Data and information from NOAA space-based and ground-based observing systems are used along with other climate-related NOAA and non-NOAA observing system data to construct long-term records regarding local, regional, national, and global climate variability and change (NOAA, 2003). The NOAA operational satellite data program, currently collects, receives, produces, distributes, and archives data about climate, including the climate satellite products. Many of these data are processed in response to specific requests from the scientific community who need long-time series climate records. In some cases, the raw data and metadata are provided to external investigators such as those in academia or those involved in international projects, which produce the climate data records. NOAA scientists also produce a number of climate products, either in-house or in collaboration with NASA (Shimabukuro et al., 1997). Among other satellite systems used for climate studies are the Tropical Rainfall Measuring Mission (TRMM) that provides comprehensive data on precipitation at tropical and subtropical latitudes, and the Altimeters on board the TOPEX/Poseidon and Jason-1 satellites that can be used to monitor water levels in reservoirs around the globe.
174
AMSR-E is the Advanced Microwave Scanning Radiometer for the Earth Observing System developed by the National Association of State Departments of Agriculture. The AMSR-E is onboard the NASA AQUA satellite as a joint mission between Japan and the United States. Aqua was launched in May 2002 and is in a circular orbit of altitude 805km. AMSR-E is a conically scanning total power passive microwave radiometer sensing microwave radiation (brightness temperatures) at 12 channels and 6 frequencies ranging from 6.9 to 89.0 GHz (6.9, 10.65, 18.7, 23.8, 36.5 and 89.0 GHz). Horizontally and vertically polarized radiations are measured separately at each frequency. The sensor has a Sun-synchronous track with equator crossings at 1:30 p.m. (ascending) and 1:30 a.m. (descending) local solar time. During a period of 1.5 seconds, the spacecraft sub-satellite point travels 10 km. Even though the instantaneous field-of-view for each channel is different, active scene measurements are recorded at equal intervals of 10 km (5 km for the 89 GHz channels) along the scan. The half cone angle at which the reflector is fixed is 47.4 which results in an Earth incidence angle (u) of 55.0. TERRA (formerly EOS AM-1) is the flagship polar-orbiting satellite of NASA's Earth observing systems. Terra is the first EOS (Earth Observing System) platform and provides global data on the state of the atmosphere, land and oceans, as well as their interactions with solar radiation and with one another. Because Terra emphasizes observations of terrestrial surface features, its orbit is designed to cross the equator at this time when cloud cover, which obscures the land surface, is at its daily minimum. The orbit is been adjusted so that it covers the complete Earth every 16 days. The five instruments onboard Terra include the Clouds and the Earth's Radiant Energy System (CERES), the Multi-angle Imaging SpectroRadiometer (MISR), the Moderate-Resolution Imaging Spectroradiometer (MODIS), the Measurements of Pollution in The Troposphere (MOPITT) and the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) instrument. - The CERES instruments perform measurements of the Earth's "radiation budget," the process that maintains a balance between the energy that reaches the Earth from the sun and the energy that goes from Earth back out to space. The critical components that affect the Earth's energy balance are the planet's surface, atmosphere, aerosols and clouds. - MISR will measure the variation of surface and cloud properties, and particles in the atmosphere, with cameras pointed in nine simultaneous different viewing directions. MISR will monitor monthly, seasonal and longterm interactions between sunlight and these components of Earth's environment. Over a seven-minute period, points on the Earth within a 224
175
mile (360 kilometer) wide swath will be observed successively at all nine angles. - The sophisticated Moderate Resolution Imaging Spectroradiometer (MODIS), flown on both the Aqua and Terra satellites, collects daily global data on vegetation condition, surface temperature, snow cover and evapotranspiration. MODIS measures the atmosphere, land and ocean processes (including surface temperature of both the land and ocean), ocean color, global vegetation, cloud characteristics, temperature and moisture profiles, and snow cover. It views the entire surface (land, oceans, clouds, aerosols, etc) of the Earth every 1-2 days at a "moderate resolution" of onequarter to one kilometer. The capabilities of MODIS present some exciting possibilities for improved and timely monitoring of crop production. The Climate-Variability Impact Index (CVII), defined as the monthly contribution to anomalies in annual growth, quantifies the percentage of the climatological production either gained or lost due to climatic variability during a given month. By examining the integrated CVII over the growing season, this LAIbased index can provide both fine-scale and aggregated information on vegetation productivity for various crop types. The radiometric and geometric properties of MODIS provide a significantly improved basis for vegetation monitoring and yield predictions with remotely sensed data (Zhang et al., 2003). Also, given the improved atmospheric correction and cloud screening and the high temporal and spatial resolution of the various MODIS vegetation products (e.g. leaf area index), this sensor seems well suited for near real-time crop monitoring (Doraiswamy et al., 2006; Chen et al., 2006; Muchoney et al., 2000). In general, the daily MODIS products are released 12 days after the image is taken. For the 8-day or monthly composite products, a 12 weeks processing period is required. Once the data have been archived, they can be downloaded from Earth Observing System data gateway (www.cr.usgs.gov/pub/imswelcome/). A valuable metric of vegetation production derived from this sensor are the satellite-based estimates of leaf area index (LAI). Leaf area responds rapidly to abiotic and biotic influences and the variability of LAI can integrate various conditions affecting plant growth and development. - The MOPITT instrument, provided by the Canadian Space Agency, map carbon monoxide and methane concentrations at altitudes between 10 miles and the ground. MOPITT is an infrared gas correlation radiometer and produces maps over the entire globe every 4-16 days. From these measurements the sources, motions and sinks of these gases can be determined. - The ASTER instrument, provided by Japan's Ministry of International Trade and Industry, measures cloud properties, vegetation index, surface mineralogy,
176
soil properties, surface temperature, and surface topography for selected regions of the earth. Every 1 to 2 days, Terra uses the unique perspective from space to observe the Earth's continents, oceans and atmosphere with measurement accuracy and capability never before flown. Hundreds of scientists are taken full advantage of Terra observations to address key scientific issues and their environmental policy impacts. The LANDSAT satellite program provides operational data on agricultural production at the regional or local scale. Landsat systematically provides wellcalibrated, multispectral, moderate resolution, substantially cloud-free, digital images of the Earth's continental and coastal areas with global coverage on a seasonal basis. The Landsat project is a joint initiative of the U.S. Geological Survey and the NASA to gather Earth resource data using a series of satellites. NASA was responsible for developing and launching the spacecrafts, while the USGS is responsible for flight operations, maintenance, and management of all ground data reception, processing, archiving, product generation, and distribution. Landsat is the longest-running project for acquisition of moderate resolution imagery of the Earth from space. The Landsat 1 satellite was launched in 1972; the most recent, Landsat 7, was launched in 1999. Landsat 7 is a satellite designed for an Earth mapping orbit with a 16-day repeat cycle. Its payload is a single nadir-pointing instrument, the Enhanced Thematic Mapper Plus (ETM+). The ETM+ provides for an eight-band multispectral scanning radiometer capable of providing high-resolution image information of the Earth's surface. The ETM+ is designed to collect, filter and detect radiation from the Earth in a swath 185 km wide as it passes overhead. The instruments on the Landsat satellites have acquired millions of images. These images form a unique resource for applications in agriculture, geology, forestry, regional planning, education, mapping, and global change research (Doraiswamy et al., 2004). The VEGETATION Programme of the SPOT-4 and SPOT-5 Earth observation satellite is conceived to allow daily monitoring of terrestrial vegetation cover through remote sensing, at regional to global scales. The instrument and associated ground services for data archival, processing and distribution are operational since April 1998. The first VEGETATION instrument is part of the SPOT 4 satellite and a second payload, VEGETATION 2, is now operationally operated onboard SPOT 5 (http://vegetation.cnes.fr). The VEGETATION instrument is an imaging system in 4 spectral bands : blue (0,43-0,47 microns), red (0,61-0,68 microns), near infrared (0,78-0,89 microns), and SW infrared (1,58-1,75 microns). The red and near infrared are
177
particularly well adapted to describe the vegetation photosynthesis activity, while the SW infrared is a good detector for the ground and vegetation humidity. The blue is design in this case to make atmospheric corrections. VEGETATION uses telemetric optics giving a quasi constant spatial resolution through the field of view, particularly wide (2 200 km on the ground) : this resolution is 1,15 km at nadir, and still 1,7 km on the sides of the field of view (101). Thanks to its sensitivity to plant photosynthetic activity as well as to the vegetation and soils' moisture content, its daily monitoring of the entire earth surface with an excellent geometric precision (local distortion better than 0.3 pixel), the systematic generation of 'ready-to-interpret' already corrected standard synthesis products, quick availability through internet, the VEGETATION system allows operational & near real-time applications, at global, continental and regional scales. The range of agriculture applications include: crop monitoring, crop production quantification, yields forecasting, agricultural pest prediction and monitoring, early warning systems to prevent food shortage, improve decision support tools and optimise the actions of governments, food aid agencies, international financing &/or cooperation agencies towards food security. These NASA and ESA missions are helping the science community better understand the forces that drive global climate and weather. Data from these systems feed complex climate models developed to improve climate forecasts. Research indicates that many extreme year-to-year changes in local weather conditions are associated with changes in global-scale climatic phenomena. Measurements of surface temperatures and biology are fed into models for predicting the timing, duration, and strength of these climatic events. The ongoing efforts to improve these models will lead to more accurate local weather and precipitation forecasts tailored to the needs of agricultural planning.
4.5
Status of satellite climatic and biophysical data for warning purpuses for agriculture, in Europe
The main objective of the Action 734 Impacts of Climate Change and Variability on European Agriculture CLIVAGRI is the evaluation of possible impacts from climate change and variability on agriculture and the assessment of critical thresholds for various European areas. The contribution of the Working Group 2.1 - Remote Sensing, to the accomplishment of this main goal is the elaboration of a comprehensive study on the benefits of satellite remote sensing on climate change and variability
178
impact on agriculture. The research effort will be directed to the analysis of the role of satellite data in the suitable models and indices for assessing the impact of climate change and variability on European agriculture The key deliverables will be represented by: - Collection of spectral climate and biophysical data for several European regions; - Assessment of required spatial, spectral and temporal resolution of satellite data for the climate change and variability impact study in agriculture; - Determination of current trends of agroclimatic indices based on spectral data; - Determination of interannual variability of agroclimatic conditions based on spectral data. In order to describe the status of satellite climate and biophysical data that are used for warning purposes for agriculture, in Europe, an initiative was started within COST 734 project, for the registration of the satellite data records existing in European countries. This inventory was created through a questionnaire disseminated to the national delegates of COST 734 member countries. Among 25 countries signed the MoU of COST action 734, 18 countries answered to the questionnaire covering almost all parts of Europe. These countries are (in alphabetical order): AUSTRIA (A), BULGARIA (BG), CROATIA (CR), CZECH REPUBLIC (CZ), FRANCE (F), GERMANY (DE), GREECE (GR), HUNGARY (HU), ITALY (I), NORWAY (NR), POLAND (PL), ROMANIA (RO), SERBIA (SI), SLOVAKIA (SK), SLOVENIA (SL), SPAIN (SP), SWITZERLAND (CH) and UK. In ANNEX 1 a list of sample products of satellite images is presented. 4.5.1 Questionnaire processed results In Table 4.1 the type of climatic and biophysical variable surveyed in each country is presented. In parentheses the satellite systems and the instruments used for the data achievement are also presented. More and more useful spectral information has been supplied to geoscientists from sophisticated space sensors with increasing detecting ability. Because the quality of spectral information depends on its sensor characteristics, sensor development is the key to remote sensing techniques. In this context the Table 4.1 revels that the new generation of satellite sensors (e.g. MODIS on TERRA platform, VEGETATION on SPOT, SEVIRI on MSG, AVHRR-3 on EPS/NOAA) have brought an upgraded level of remote-sensed information to the user community thanks to a much better spatial, temporal, spectral and
179
angular sampling of the radiative fields emerging from the surface of the Earth. The time resolution and global coverage provided by the new instruments, together with the extensive sampling in both the spectral and angular domains, paved the way for a broad spectrum of novel applications, namely within the scope of land surface processes and land-atmosphere interactions. For example SEVIRI/METEOSAT data are extensively used to obtained climate and biophysical variable.
Table 4.1: Type of climatic or biophysical data variables surveyed by country Variable Soil Moisture (METOP / EUMETSAT) Vegetation cover (TM / LANDSAT) NDVI (MODIS / TERRA) NDVI (MODIS / TERRA-AQUA) Surface temperature (MODIS / TERRA) Land cover (TM / LANDSAT) Snow cover (MODIS / TERRA) NDVI (AVHRR / NOAA) Surface temperature (AVHRR / NOAA) NDVI (VEGETATION / SPOT) LAI (MODIS / TERRA) NDVI (MODIS / TERRA) Land cover (TM / LANDSAT) Surface temperature (TM / LANDSAT) Evapotranspiration (TM / LANDSAT) NDVI (TM / LANDSAT) Land cover (ASTER / TERRA) Surface temperature (ASTER / TERRA) Evapotranspiration (ASTER / TERRA) NDVI (SEVIRI / METEOSAT) SAF products (METEO 5 / EUMETSAT) Land cover (SEVIRI / METEOSAT) NDVI (AVHRR / NOAA) Surface temperature (MODIS / TERRA) Veget. condition index (VCI) (AVHRR/NOAA) Temp. condition index (TCI) (AVHRR/NOAA) Veget. Health Index (VHI) (AVHRR/NOAA) Cold Cloud Duration (CCD) (METEOSAT) Temperature Vegetation Dryness Index (AVHRR/NOAA) Surface temperature (AVHRR / NOAA) Evapotranspiration (AVHRR / NOAA) Degree days (AVHRR / NOAA) Cloud (snow) (AVHRR / NOAA) NDVI (AVHRR / NOAA) Surface Temperature (AVHRR / NOAA) Radiation budget (METEOSAT) Country A
BG
CZ FR
DE
GR
(TVDI)
HU
180
NDVI (AVHRR / NOAA) NDVI (VEGETATION / SPOT) NDVI (MODIS / TERRA) Surface temperature (AVHRR / NOAA) Rainfall (MSG + MW) Rainfall (MSG + MW) Rainfall (GEO IR and LEO MW) Cloud products (METEOSAT) Cloud products (SEVIRI / METEOSAT Cloud products (AVHRR / NOAA) Precipitation (TOVS / NOAA) Air temperature (AVHRR, TOVS / NOAA) Albedo (AVHRR, TOVS / NOAA) Sea ice (AVHRR / NOAA) Sea wind (METOP) NDVI (AVHRR / NOAA) Cloud products (SEVIRI / METEOSAT) Rainfall (SEVIRI / METEOSAT) Solar radiation (MTP-HPI/ METEOSAT) Air stability (SEVIRI / METEOSAT) Storm detection (SEVIRI / METEOSAT) Temperature (TOVS / NOAA) Precipitation (TOVS / NOAA) Ozone content (TOVS / NOAA) SAF products (precipitation, soil moisture, snow) (NOAA, METEOSAT, AQUA) Cloud products (SEVIRI / METEOSAT) Precipitation (SEVIRI / METEOSAT) NDVI (SEVIRI / METEOSAT) Solar radiation (SEVIRI / METEOSAT) NDVI (VEGETATION / SPOT) MSAVI (VEGETATION / SPOT) Surface temperature (MODIS /AQUA-TERRA) Snow (MODIS / AQUA - TERRA) Clouds (SEVIRI / METEOSAT) NDVI (MODIS / TERRA) Vegetation cover (MODIS / TERRA) Surface temperature (METEOSAT) Radiation (SEVIRI /METEOSAT) Albedo (SEVIRI / METEOSAT) Snow cover (SEVIRI / METEOSAT) Vegetation cover (SEVIRI / METEOSAT) NDVI (AVHRR / NOAA) Surface temp. (SEVIRI / METEOSAT) Albedo (SEVIRI / METEOSAT) Solar radiation (SEVIRI / METEOSAT) Snow cover (SEVIRI / METEOSAT) NDVI (AVHRR / NOAA)
NO
PL
RO
SK SL
SP
CH
181
MSG provides an image repeat cycle of 15 minutes offering new opportunities to detect short-term evolution of vegetation resources. Such feature is particularly relevant over areas characterized by a high cloud occurrence as well as for semi-arid ecosystems having short vegetation cycles. On the other hand, the diurnal and sub-diurnal sampling of thermal signatures by MSG, afford solving the land surface temperature cycles. Availability of high temporal resolution from MSG are optimally suited to the measurement of environmental parameters that change rapidly in time as well as to those parameters where the signal change over time contains information about the parameter or the process of interest. However, the spatial characteristics of sensors mainly relate to events at the regional to continental scales. Presently remote sensing is limited to mapping single crops to slightly higher than 90% accuracy when multi-date, multi-sensor, or GIS data is also used. Remote sensing alone, however, is more limited at discrimination of multiple crops to this level of accuracy. Remote sensing is also often unable to detect direct sources of crop damage. These limitations may decrease in the future as spatial and spectral resolutions, and repeat cycles increase. In Table 4.2 the climatic and biophysical variables recorded at least in one country are presented.
Table 4.2: Climatic and biophysical variables surveyed Climatic variables Surface temperature Precipitation Snow cover Solar radiation Albedo Cloud cover and other cloud products SAF products (precipitation, soil moisture, snow) Air stability Storm detection Ozone content Sea ice, sea wind Biophysical variables NDVI MSAVI LAI VCI TCI TVDI Soil moisture Vegetation cover Land cover Evapotranspiration Degree days CCD
4.5.2 Type of data per country In Table 4.3 the type of variable in order of the number of countries surveyed is presented. It is obvious that NDVI is the main biophysical variable that is recorded and used by most countries (12 countries). Among climatic variables the mostly used is surface temperature (11 countries). In a second series of the climatic variables are cloud products (6 countries), snow cover and radiation
182
(5 countries) land cover and precipitation (4 countries), SAF products (3 countries). Evapotranspiration and albedo follows, recorded for 2 countries and all the rest (Air-stability, Storm detection, Ozone content, VCI, TCI, VHI, TVDI, CCD, Soil moisture, MSAVI, LAI, Degree days, sea ice and sea wind) are recorded only in 1 country. The mentioned differences between different countries regarding the use of climate and biophysical variables can be explained by the fact that the high level products (like evaporation, soil moisture, storm detection, etc) require quite complex algorithms or schemes. The SAF products are not extensively used by many countries. In this respect, the 'Satellite Application Facility on Climate Monitoring' (CM-SAF), which started its operational activities in March 2007, will provide most valuable complex products in the near future. It has also to be mentioned that in many countries the assimilation of satellite data into crop growth simulation models is still in an experimental stage.
Table 4.3: Type of variable in order of the number of countries surveyed VARIABLE NDVI Surface temperature Cloud products Snow cover Radiation Vegetation cover, land cover Precipitation SAF products Albedo Evapotranspiration Air- stability Storm detection Ozone content VCI, TCI, VHI, TVDI, CCD Soil moisture MSAVI LAI Degree days Sea wind and ice SUM OF COUNTRIES 12 (A, BG, F, DE, GR, HU, I, PL, RO, SL, SP, CH) 11 (BG, FR, DE, GR, HU, I, NO, PL, RO, SL, SP) 6 (I, NO, PL, RO, SK, SL) 5 (CZ, HU, RO, SL, SP) 5 (HU, PL, RO, SL, SP) 4 (A, BG, DE, SL) 4 (I, NO, PL, RO) 3 (DE, PL, SP) 2 (NO, SP) 2 (DE, GR) 1 (PL) 1 (PL) 1 (PL) 1 (GR) 1 (A) 1 (RO) 1 (DE) 1 (GR) 1 (NO)
This table reveals that the most used satellite derived - biophysical variable is the NDVI. This vegetation index is still considered one of the most successful of many attempts to simply and quickly identify vegetated areas and their "condition" and it remains the most well-known and used index to detect live
183
green plant canopies in multispectral remote sensing data. In addition to the simplicity of the algorithm and its capacity to broadly distinguish vegetated areas from other surface types, the NDVI also has the advantage of compressing the size of the data to be manipulated by a factor 2 (or more), since it replaces the two spectral bands by a single new field. Nevertheless, it must be noticed that the NDVI has tended to be over-used in applications for which it was never designed. For example using the NDVI for quantitative assessments raises a number of issues that may seriously limit the actual usefulness of this index if they are not properly addressed. The NDVI should be used with great caution in any quantitative application that necessitates a given level of accuracy. All the perturbing factors (atmospheric soil effects, anisotropic effects and spectral effects) that could result in errors or uncertainties of that order of magnitude should be explicitly taken into account; this may require extensive processing based on ancillary data and other sources of information. More recent versions of NDVI datasets have attempted to account for these complicating factors through processing. The satellite-derived surface temperature (for land and sea), is also a broad used climate variable among the surveyed countries. Surface temperature is used in various agro-meteorological applications like: surface heat energy balance study, characterization of local climate in relation with topography and land use; mapping of low temperature for frost conditions or winter cold episodes, derivation of thermal sums (using surface temperature instead of air temperature) for monitoring crop growth and development conditions. Polar orbiting satellites in low orbit can provide much better spatial resolution and hence potentially more useful estimates of surface temperature than can other measurement methods. Table 4.3 shows that some variables like albedo, evapotranspiration, airstability, storm detection, ozone content, soil moisture, sea wind and ice are used by a much reduced number of countries. This can be explain by the fact that the procedures used to retrieve such variables are still in experimental phase and do not satisfy the users requirements related with accuracy, spatial or temporal scales etc. For example soil moisture is an important parameter for weather and climate prediction as well as for crop monitoring. Many efforts have been made for soil moisture estimation with space-borne sensors and insitu measurements. These approaches measure soil moisture at different spatial scales and each of them have certain advantages and limitations. Microwave remote sensing measurements can provide physical retrieval of soil moisture in low vegetation areas, but have poor spatial resolution. Optical/IR measurements can be used to retrieve soil moisture at high spatial resolution statistically, but limited to clear days. In spite of these results, presently soil moisture retrieval with satellites is still not operationally available. The new
184
generation of microwave remote sensing satellites (e.g. Terra SAR X) will provide soil moisture products in the near future. In Table 4.4 for each variable the type of the satellite instrument used is presented. This table reveals that SEVIRI/METEOSAT and AVHRR/NOAA are the most popular satellite sensors which provide climate and biophysical variables, among the surveyed countries. These two satellite systems are widely used by the European Meteorological Services, most countries having their own satellite reception systems.
Table 4.4: Type of satellite / instrument per climate and biophysical product recorded Climate and biophysical product NDVI Surface temperature LAI MSAVI Cloud products Snow cover Radiation Vegetation / land cover Precipitation SAF products Air-stability Storm detection Ozone Evapotranspiration Soil moisture VCI TCI VHI TVDI Degree days CCD Albedo Sea ice Sea wind Type of satellite / instrument (MODIS / TERRA-AQUA, AVHRR / NOAA, VEGETATION / SPOT, TM / LANDSAT, SEVIRI / METEOSAT) (AVHRR / NOAA, TM / LANDSAT, ASTER / TERRA, MODIS / TERRA-AQUA, SEVIRI / METEOSAT) (MODIS / TERRA) (VEGETATION / SPOT) (SEVIRI / METEOSAT, NOAA / AVHRR) (MODIS / TERRA, SEVIRI / METEOSAT) (SEVIRI / METEOSAT) (TM-ETM / LANDSAT, ASTER /TERRA, SEVIRI / METEOSAT) (SEVIRI / METEOSAT, GEO / LEO satellites, TOVS/NOAA) (METEOSAT, NOAA, AQUA) SEVERI / METEOSAT) (SEVERI / METEOSAT) (TOVS / NOAA) (TM/LANDSAT,ASTER/TERRA, AVHRR/NOAA) (ASCAT / METOP) (AVHRR / NOAA) (AVHRR / NOAA) (AVHRR / NOAA) (AVHRR / NOAA) AVHRR / NOAA) (METEOSAT) (SEVIRI / METEOSAT, AVHRR / NOAA) (AVHRR / NOAA) (METOP)
MODIS and ASTER onboard TERRA or AQUA platforms are preferred by a lot of the European countries due to easy accessibility via internet and because their improved spatial, temporal and spectral characteristics are appropriate for many agricultural applications.
185
In Table 4.5 the variables used operationally or/and experimentally are presented. NDVI, Surface temperature, Snow cover and Cloud products are used in both ways. The rest are used either operationally or experimentally. It is very interested that in countries with aerospace development there are no data records used operationally.
Table 4.5: Operational or experimental use of data Operational NDVI (GR, I, PL, RO, SP) Surface temperature (GR, I, SP) Snow cover (SP) Cloud products (PL, RO, SK) Soil moisture (A) Precipitation (I, PL, RO) Solar radiation (PL, RO, SP) Air stability (PL) Storm detection (PL) Ozone content (PL) MSAVI (RO) Albedo (SP) Experimental NDVI (A, BG, FR, DE, GR, CH) Surface temperature (BG, FR, DE, GR, RO) Snow cover (CZ, RO) Cloud products (I) Vegetation cover (A) LAI (DE) Land cover (BG, DE) Evapotranspiration (DE, GR) SAF products (DE) VCI, VHI, TVDI (GR) Degree days (GR) CCD (GR)
In Table 4.6 the variables that are used in models in each country are also presented. From this table it is obvious that only a few countries use spectral data in models. For example in Chech Republic MODIS/TERRA data are used for snow cover verification. In France NDVI and surface temperature calculated by AVHRR or SPOT/VEGETATION data are used experimentally in models for biomass, ETP and frost estimation. In Italy a few experimental data are used for rainfall and clouds estimation, and in Romania for LST evapotranspiration and snow cover. In Greece also all experimental data received are used in models.
Table 4.6: The variables that are used in models in each country Variable / satellite system / country Snow cover NDVI Surface temperature NDVI NDVI, VCI, TCI, VHI, TVDI LST, ETP, DD CCD Precipitation Cloud products Surface temperature MODIS / TERRA AVHRR / NOAA AVHRR / NOAA VEGETATION / SPOT AVHRR/NOAA AVHRR/NOAA METEOSAT METEOSAT and GEO/LEO METEOSAT MODIS/TERRA (CZ) (FR) (FR) (FR) (GR) (GR) (GR) (I) (I) (RO)
186
A preliminary analysis shows that the countries involved in Cost 734 use very limited satellite data or derived products in models.
4.5.3 Detailed analysis of the data per country AUSTRIA : Biophysical and climate satellite data parameters are collected from three types of satellite systems: METOP-ASCAT satellite system, produced by EUMETSAT (for soil moisture, start 2008). LANDSAT satellite system, produced by NASA (for vegetation cover). MODIS satellite system, produced by NASA (for NDVI). Both in experimental use since 2005. Not yet assimilated to models. BULGARIA : Surface temperature and NDVI are collected from MODIS/TERRA and AQUA satellites experimentally since 2005. Land cover maps are produced in scale 1:100.000 occasionally from Landsat data. CROATIA : Satellite data records are not used in the Meteorological and Hydrological Service of Croatia for the agrometeorological purpose yet. Satellite data records are used only for the weather forecast with the satellite/instrument Meteosat 8/SEVIRI. CZECH REPUBLIC : Biophysical satellite data parameters are collected from one type of satellite system (MODIS Satellite system, produced by TERRA). The climate parameter is snow and it is covered experimentally since 2006, using a verification model. FINLAND : Climate satellite data parameters are collected from two types of satellite systems. AVHRR satellite system, produced by NOAA, and SEVIRI satellite system, produced by METEOSAT. From NOAA/AVHRR and SEVIRI/METEOSAT only cloud cover is collected operationally since 2000.
187
FRANCE : Climate and biophysical parameters are collected from two types of satellite systems. AVHRR satellite System, produced by NOAA and VEGETATION /SPOT satellite system, produced by VITO-Anvers. The collected data are NDVI and surface temperature, from AVHRR, and NDVI from SPOT/VEGETATION. AVHRR data were used experimentally during 1983-1998. Since 1998 0nly VEGETATION data are used as NDVI ready product for biomass estimation. The use of satellite data for agrometerological purposes has been an major item of research in France, especially at INRA with the support of CNES and the collaboration with CNRS, during about 20 years (from 1978 to 1998 approximately, mainly using Meteosat and NOAA-AVHRR). It has been proved that satellites were able to provide significant information for drought and frost mapping and to contribute to yield forecasting. The way for operational application has appeared however to be better in accordance with the European scale, and the expertise gained by the research team has been transferred to the team of MARS project in Ispra rather than at the national level. The mentioned satellite data (AVHRR, Meteosat, Envisat, GOES, Vegetation, MODIS), are received and processed in France by various institutions and laboratories, but are used for studies concerning the biosphere, and not strictly agriculture. Nothing appears to exist on an operational basis, and the research area in this field is still pursued in INRA, but on the more methodological aspect of assimilation into crop models, which is still limited to experimental studies. GERMANY : In Germany climate and biophysical satellite data parameters are collected from five types of satellite systems: MODIS satellite system, produced by TERRA, TM5 satellite System, produced by Landsat, ASTER satellite System, produced by TERRA, SEVIRI satellite system, produced by Meteosat, And Meteo 5 produced by Eumetsat. From MODIS, the parameters LAI and NDVI are collected experimentally since 2001. From LANDSAT, the parameters land cover, surface temperature, evapotranspiration and NDVI, are collected since 1992.
188
Using the ASTER instrument the parameters land cover, surface temperature and evapotranspiration are collected since 2006. From SEVIRI, NDVI and land cover are collected since 2006. Finally SAF products are also collected from METEO 5/EUMETSAT since 2006. GREECE : Satellite data parameters are collected from three types of satellite systems: AVHRR / NOAA, MODIS / TERRA and METEOSAT. From AVHRR the parameters NDVI, Vegetation Condition Index (VCI), Temperature Condition Index (TCI), VHI, Land Surface temperature and evapotranspiration are collected and calculated since 1981. Two time series with different spatial resolution are used for the indices calculations. The first consists of 20 hydrological years (1981-2001) of ten-day composite raw images (for all channels) with 8x8 km spatial resolution and the second of NDVI and LST thematic maps (1x1 km) from 1998 until today. Degree Days were collected only for 2004 and 2005. From MODIS, Surface temperature is used since 2004. From METEOSAT satellite images are used for several products since 1995. Specific comments for models: Abbreviations: (F)=Forecasting Model, (N)=Now casting Model, (P)=Past casting Model. NDVI has been extensively used for (P) and (N) monitoring and modeling. VCI and VHI have excellent ability to detect drought and to measure the time of its onset and its intensity, duration and impact on vegetation. The droughtmonitoring VCI and VHI algorithm was developed for the 1981-2001 time series (8x8 km) and used for (P) and (N) modeling. Developing empirical relationships between the VCI values of the critical ten-day periods of the growing season and the yield, a good indication of the yearly production can be obtained ((F) model). Further development for production estimation is based on the VCI data which are proceeded with the Bhalme and Mooley Drought Index (BMDI) methodology. BMVCI was calculated for the 1981-2001 time series and used in (F) models. Potential Evapotranspiration (ETp) is used for (P) and (N) monitoring. LST has been used in order to calculate ETp. Before using LST as an input in ETp computation equation, LST has to be converted to air temperature using area specific models. The potential evapotraspiration is calculated with the use of Blaney-Criddle method. CCD is experimentally used for precipitation (F) model and TVDI in (P) and (N) models for drought monitoring.
189
HUNGARY : From locally received NOAA/AVHRR data a subimage of 1024x1024 pixels containing Hungary is processed with programs developed in Hungary. Cloud (snow) mask is computed using all channel data. Atmospherically corrected NDVI and albedo are calculated (for cloud free area) using NOAA/AVHRR 1 and 2 channel data. Land surface temperature (LST) is calculated (for cloud free area) using NOAA/AVHRR 4 and 5 channel data. NDVI and albedo have been calculated since 1996 and 1999 respectively, in full resolution, from one operational NOAA satellite data, from NOAA 14, 16 and presently from 18 satellite data, only daytime, typically once a day in the afternoon. LST has been calculated since 1996 from one operational NOAA satellite (NOAA14, 16, 18) data, typically twice a day: one at night and on during the day, in full resolution. From the daily data, 10-day composite NDVI, albedo, as well as daily LST maps are calculated. Data are transferred on a stereographic map, and choose the maximum value of 10-day periods. An example of a 10-day composite image is shown in Fig 1. for the Charpathian Basin region with a spatial resolution of 2 km. he daily data at original AVHRR resolution and the 10-day LST, NDVI and albedo maps are archived. adiation budget components were calculated from the METEOSAT first generation (using NOAA/AVHRR data to estimate the albedo) at 10 km resolution. Using these data the following variables were calculated from 2004 until July 2006: Net radiation, Shortwave and longwave net radiation, and Shortwave and longwave incoming radiations at the surface. o process the METEOSAT-8 (METEOSAT Second Generation) data, the SAFNWC/MSG program package for a big Central European region is used. The images are processed every 15 minutes and archive the results (for example: the cloud mask, cloud type, cloud top height, pressure, temperature, probability of precipitation, convective rain rate etc). Fig 2 is an example. The results are archived in full MSG resolution (3x3km in sub satellite point). The corresponding quality flags are also archived. The SAFNWC/MSG program package is being developed by an international working group. ITALY : Climate and biophysical parameters are collected from several types of satellite systems: NDVI, from AVHRR/NOAA, VEGETATION/SPOT and TERRA/MODIS. Surface temperature, from AVHRR/NOAA.
190
Rainfall, from MSG+MW satellites (produced by Italian National Met Service and LaMMA) and GEO IR and LEO MW satellites (produced by NASAGSFC) through the Global Precipitation Climatology Project (GPCP). And cloud products from MSG satellites (produced by ISAC-CNR). NORWAY : No satellite data is at the moment used for agro meteorological purposes by the Norwegian Institute for Agricultural and Environmental Research. At the Norwegian Meteorological Institute it is developed software using satellite data for calculating the surface temperature of the earth. Calculation of the surface temperature of the tops of the clouds are provided for operational use every day. The use of these data is mainly connected to weather forecasts of short temporal scale. These data the Norwegian Meteorological Institute get from different satellites. Some of the data they get on the Internet, but also some data they get directly from satellites. At the NORUT institution satellite data are used for getting information connected to the cover of natural vegetation. Agricultural crops are not included in the operational registrations. Cloud products, sea ice, sea wind, air temperature and albedo are calculated locally using METEOSAT/SEVIRI, NOAA/AVHRR, TOVS data, operationally. POLAND : The main climate and biophysical parameters that are collected from satellite systems are: NDVI with NOAA/AVHRR data. METEOSAT/SEVIRI data are used for cloud type and mask, convective rainfall rate, precipitating clouds, air stability, total precipitation and storm detection. olar radiation at the surface is calculated with METEOSAT (MTP) HRI data. arameters like distribution of temperature, dew point temperature, geopotential height, wind speed and direction, total precipitation and total ozone content are calculated with NOAA/TOVS data. Other products related to climate and agro are received but final product is still in experimental phase: From Land SAF: - DSSF Downwelling Shortwave Radiation, - DSLF Downwelling Longwave Radiation, - LST Land Surface Temperature, From Ocean and Sea Ice SAF Sea Surfach Temperature (SST), Solar Shortwave Incoming Radiation at the ground (SSI),
191
Downwelling Long Wave Radiation at the ground (DLI), Sea Ice extend, concentration, type, Wind Speed and Direction over sea. From EUMETSAT MPEF (Central processing): - Multisensor Precipitation Estimates, In the frame of H-SAF activities (Satellite Application Facility in support to Operational Hydrology and Water Management) new satellite products for hydrology are at the moment at experimental status. Development phase of HSAF will be finished in 2010 and at this time or earlier presented products will be operational. There is also interest for precipitation products, soil Moisture products and snow products. ROMANIA : Biophysical and climate data for agrometeorological use are collected from three types of satellite systems: From MSG-1/SEVIRI satellite system (produced by EUMETSAT) the parameters: Cloud mask, cloud type, cloud top height & temperature, precipitating clouds, convective rainfall rate, total precipitable water, NDVI and solar direct radiation, are calculated operationally. From SPOT 5/VEGETATION satellite system, decadal NDVI synthesis, color composites and MSAVI synthesis are calculated operationally. From AQUA-TERRA/MODIS system, the parameters: Land surface temperature and daily snow cover are experimentally calculated. In Romania, the National Meteorological Administration - Satellite Department benefits by the direct reception of digital High Resolution Imagery Transmissions data from METEOSAT SECOND GENERATION station which has been provided by the VCS-Engineering. It actually receives stores and process all the HRIT and LRIT data dissemination formats. Since 2004 Romania has become Cooperating States of EUMETSAT. The National Meteorological Administration operationally receives MSG data and SAF products through EUMETCAST. MSG HRIT raw data is processed, all 12 channels being available. The NWCSAF products (Cloud Mask, Cloud Type, Cloud Top Height & Temperature, Precipitating Clouds, Convective Rainfall Rate, Total Precipitable Water ) and NDVI product are locally obtained from the MSG real-time data, running the NWCSAF software on a SUN system, respective a developed internally software. The LandSAF products (Downwelling Short-wave radiation flux, Downwelling Long-wave radiation flux) are ready products received through EUMETCAST. he data referred above are not yet assimilated into models.
192
The time intervals of the data archives for MSG - derived products covers the period 2005 to present, and for SPOT VEGETATION - derived products from 2004 to present. SERBIA : The country is not member of EUMETSAT. NHMI is or will be focal point for this kind of data. SLOVAKIA : Only clouds are calculated operationally from SEVIRI/METEOSAT, using a local receiver. SLOVENIA : NDVI and vegetation cover are the biophysical parameters that are collected from MODIS instrument on board TERRA/AQUA, since 2006 experimentally. LandSAF in the frame of METEOSAT-MSG data are used for land Surface Temperature, the down-welling surface short-wave radiation flux., downwelling Surface Longwave Radiation Flux, surface albendo, snow Cover, and fractional Vegetation Cover calculation, since 2006, experimentally. SPAIN : NDVI is used operationally as a ready product, from AVHRR/NOAA data, since 1993. Among climate variables, land temperature, albedo, downwelling short-wave radiation flux and snow cover are used recently as ready products from MSGSEVIRI satellite system, produced by EUMETSAT (land-SAF). SWITZERLAND : Only NDVI is used experimentally. Data are collected from AVHRR/NOAA by a local receiver. UK : Researching the use of satellite data records for agrometeorology in the UK, the main conclusion is that there is really very little evidence of operational use of satellite data for this purpose. I think that the main reason for this is that the UK has a very well developed and extensive network of meteorological stations, a highly advanced ground based weather radar system and a long history of meteorological measurements through the UK Meteorological Office. These data, along with data from MSG are used for weather forcasting of course and weather data are highly accessible for agrometeological
193
operational applications or for research. A second reason is the high frequency of cloud cover in the UK which prevents routine surface observations being obtained. The use of MSG, AATSR or MODIS for land surface temperature measurements does not appear to have been a major research area for applications in the UK, and is certainly not operational. AVHRR data are received operationally at the Dundee Satellite Station and some operational data processing for coastal and marine applications is carried out. There are no examples of operational processing of AVHRR data for agricultural applications in the UK. (www.sat.dundee.ac.uk). MODIS data are only used experimentally in UK based applications to date. The application of medium spatial resolution satellite data is more widespread for experimental agrometeorology, but again the cloud cover problem is a major limitation for multi-temporal studies or for operational applications. Landsat TM and ETM, and SPOT HRVIR data are used for land cover mapping but this is only done overey 5-10 years and is of little relevance to agrometeorology.An example of an operational systems for agriculture is Farmstar which is a crop management tool from Infoterra. This system delivers field maps to farmers, based on medium spatial reslution satellite data (primarly SPOT). These data are used by farmers to aid in-field crop management (http://www.infoterra.co.uk/applications_land_farmstar.php).This is a narrow range of examples but I believe that it does reflect the current picture of satellite applications for agrometeorology in the UK. JRC : In the questionnaire table are presented RS data that are currently available or processed in the framework of JRC activities. The data comprise the satellites NOAA-AVHRR, SPOT-Vegetation and TERRA-MODIS (as indicated in the table) as well as LandSAF products of MSG-SEVIRI, which are already contained in the table and therefore did not added again.The variables NDVI, SAVI, fAPAR, DMP and TS are calculated from NOAA/AVHRR data and used operationally in EU scale since 1981. The variables NDVI, SAVI, fAPAR and DMP are calculated also from SPOT/VEGETATION and used operationally in european scale since 1998. The same variables are calculated also from MODIS/TERRA and used operationally in EU since 2002. All data are provided by VITO. The link to JRC online portal (ImageServer), which gives an overview of the data that are available is: http://cid.jrc.it/idp/thematic-portals/mars-statimageserver/.
194
4.6 Conclusions Increased availability of satellite sensor data is resulting in an increase in satellite sensor data and products use for warning purposes in agriculture due to climate variability and change. The analysis, based on the specific questionnaires received from the COST 734 National delegates, shows a general interest in using satellite climate and biophysical data and products to better understand how climate affects crop growth and yield, and warning purposes for agriculture. Climatic and biophysical variables are very relevant for monitoring vegetation status and predict the possible impact on crops. Some efforts are currently made by the remote sensing community to provide such products from current medium spatial resolution satellite observations operationally available, including VEGETATION, MODIS, and AVHRR. Among European countries there is a great unhomogeneity concerning climatic and biophysical data received from satellite sensors or collected as ready products. Some of them are currently collecting satellite data for years and these data records could be useful for models for climate change impact studies. The main variables that are collected in operational or experimental way are land surface temperature and NDVI. However, the products are currently poorly validated because of the lack of ground measurements. This is even worth with regards to a given type of vegetation such as crops, over a given continent such as Europe. In addition, most of the validation activities correspond to one shot ground measurements, preventing from accessing the seasonality and associated phenology of the surfaces, although most of the information on vegetation functioning lies in its dynamics. It is needed to develop a network of sites across Europe to contribute to the validation of the satellite products specifically for agricultural areas. This direct validation should be based on continuous ground measurements in order to access the phenology of the crops. Additional information on crop stages will allow calibrating relationships with remote sensing seasonal characteristics as derived from satellite products time course. The validation should be based on the development of a network of ground measurements sites. It should be designed to sample the climatic, edaphic and cultural practices observed over Europe. The main crops should be sampled and could include: wheat, maize, sunflower, sugar-beet, sunflower, rapeseed, cotton, and others. A meteorological station should be close to each site, to possibly link development stages with phenological models. This could also allow further exploitation of the measurements with regards to climate change impact and its simulation/estimation via crop models.
195
Although remote sensing from satellites has high potential for providing spatially variable data needed for precision agriculture, the limitations of fixed spectral bands, too coarse spatial resolutions, inadequate repeat cycles, and long delivery times make it non operational at this time. In the next phase of our action (COST 734) it will be investigated how it is possible, using existing satellite data, to produce long-satellite records and adapt observations designed for weather prediction to climate issues in an ad hoc way, and if it is sufficient to produce reliable findings and to draw reasonable conclusions about climate change impacts on agriculture. The development of quality satellite data records is key and that a program that focuses on the development, retention, and distribution of satellite data records in Europe will be necessary to meet the needs of the science community.
4.7 Acknowledgements COST 734 WG2.1. members acknowledge all COST 734 colleagues and all country associates for their help in collecting the data and answering the questionnaires.
4.8 References
Aase J.K., J.P. Millard, B.S. Brown, 1986. Spectral Radiance Estimates of Leaf Area and Leaf Phytomass of Small Grains and Native Vegetation. IEEE Transactions on Geoscience and Remote Sensing, 24, pp 685-692. Baret F., O. Hagolle, B. Geiger, P. Bicheron, B. Miras, M. Huc, B. Berthelot, F. Fernando Nio, M. Weiss, O. Samain, J.L. Roujean, M. Leroy, 2007. LAI, fAPAR and fCoverCYCLOPES global products derived from VEGETATION, Remote Sensing of Environment, 110 (3), pp 275-286. Barnsley M., P. Lewis, M. Disney, G. Thackrah, T. Quaife1, J. Shaw, L. Rebelo, G. Roberts, P. Hobson, J.P. Muller, 2002. MOD43 Validation: UK Activities. CEOS/WGCV Land Product Validation Workshop on Albedo, Boston University, October 2002. Available at www-modis.bu.edu/brdf/barnsley.pdf (accessed 2007-02-14). Bhuiyan C., R.P. Singh, F.N. Kogan, 2006. Monitoring drought dynamics in the Aravalli region (India) using different indices based on ground and remote sensing data. International Journal of Applied Earth Observation and Geoinformation, 8, pp 289-302. Bouman B.A.M, H. van Keulen, H.H. van Laar, R. Rabbinge, 1996. The 'School of de Wit' Crop Growth Simulation Models: A pedigree and Historical Overview, Agricultural Systems, 52 (2), pp 171-198. Buermann W., J. Dong, X. Zeng, R.B. Myneni, R.E. Dickinson, 2001. Evaluation of the Utility of Satellite-Based Vegetation Leaf Area Index Data for Climate Simulations, Journal of Climate, 14 (17), pp 3536-3550. Chen P.Y., G. Fedosejevs, M. Tiscareo-LPez, J.G. Arnold, 2006. Assessment of MODISEVI, MODIS-NDVI and VEGETATION-NDVI Composite Data Using Agricultural
196
Measurements: An Example at Corn Fields in Western Mexico. Environmental Monitoring and Assessment, Volume 119, Numbers 1-3 , 69 82, DOI 10.1007/s10661 005 9006 7. Ciais P. et al., 2005. Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature, 437, pp 529-533. Congalton R.G. and Green K., 1999. Assessing and accuracy of remote sensed data: principles and practices. Lewis Publishers, Washington D.C. Dabrowska-Zielinska K., F.N. Kogan, A. Ciolkosz, M. Gruszczynska, W. Kowalik, 2002. Modelling of crop growth conditions and crop yield in Poland using AVHRR-based indices. International Journal of Remote Sensing 23, pp 1109-1123. Domenikiotis C., M. Spiliotopoulos, E. Tsiros, N.R. Dalezios, 2002. Application of NOAA/AVHRR VCI for drought monitoring in Thessaly, Greece, 6th International Conference of Protection and Restoration of the Environment, 1-5 July 2002, Skiathos, Greece, pp 1663-1670. Domenikiotis C., M. Spiliotopoulos, E. Tsiros, N.R. Dalezios, 2004a. Early Cotton Yield Assessment by The Use Of The NOAA/AVHRR Derived Drought Vegetation Condition Index In Greece, International Journal of Remote Sensing, 25, pp 28072819. Domenikiotis C., M. Spiliotopoulos, E. Tsiros, N.R. Dalezios, 2004b. Early cotton production assessment in Greece based on the combination of the drought vegetation condition index (VCI) and Bhalme and Mooley drought index (BMDI), International Journal of Remote Sensing, 25 (23), pp 5373-5388. Domenikiotis C., M. Spiliotopoulos, E. Tsiros, N.R. Dalezios 2005a. Remotely Sensed Estimation f Annual Cotton Production Under Different Environmental Conditions in Central Greece, Physics and Chemistry of the Earth, 30, pp 45-52. Domenikiotis C., E. Tsiros, M. Spiliotopoulos, N.R. Dalezios 2005b. Zoning of cotton production areas based on NOAA/AVHRR images, International Symposium in GIS and Remote Sensing: Environmental Applications, Volos, Greece, 7-9 November 2003, pp 119-132. Doraiswamy P.C., T.R. Sinclair, S. Hollinger, B. Akhmedov, A. Stern, J. Prueger, 2006. Application of MODIS derived parameters for regional crop yield assessment, Remote Sensing of Environment, 97, pp 192202. Doraiswamy P.C., J.L. Hatfield, T.J. Jackson, B. Akhmedov, J. Prueger, A. Stern, 2004. Crop condition and yield simulations using Landsat and MODIS, Remote Sensing of Environment Volume 92, Issue 4, 30 September 2004, pp 548-559. FAO, 2001. Committee on Agriculture report "Climate variability and change", Rome. FAO, 2005. Impact of Climate Change, Pests and Diseases on Food Security and Poverty Reduction. Special event background document for the 31st Session of the Committee on World Food Security. Rome, 23-26 May 2005. Fisher J.I. and Mustard J.F., 2007. Cross-scalar satellite phenology from ground, Landsat, and MODIS data. Remote Sensing of Environment, 109 (3), pp 261-273. Gardner B.R. and Blad B.L., 1986. Evaluation of Spectral Reflectance Models To Estimate Corn Leaf Area While Minimizing the Influence of Soil Background Effects. Remote Sensing of Environment, 20, pp 183-193. GCOS, 2004. Implementation plan for the Global Observing System for Climate in support of the UNFCCC. Report GCOS 92 (WMO/TD No. 1219), 136 p. GCOS, 2006. Satellite-based products for climate. Supplemental details to the satellite- based component of the Implementation Plan for the Global Observing System for Climate in Support of the UNFCCC, GCOS-107 (WMO/TD No. 1338), 90 p.
197
Gitelson A.A., 2004. Wide Dynamic Range Vegetation Index for Remote Quantification of Biophysical Characteristics of Vegetation, Journal of Plant Physiology, 161 (2), pp 165-173. Holben B.N., C.J. Tucker, C.J. Fan, 1980. Spectral Assessment of Soybean Leaf Area and Leaf Biomass. Photogrammetric Engineering and Remote Sensing, 46, pp 651-656. Houborg R. and Boegh E., 2008. Mapping leaf chlorophyll and leaf area index using inverse and forward canopy reflectance modeling and SPOT reflectance. Remote Sensing of Environment, 112 (1), pp 186-202. Huang S., 2005. The potential of multi-sensor satellite data for applications in environmental monitoring with special emphasis on land cover mapping, desertification monitoring and fire detection. Dissertation, LMU Munchen: Faculty of Biology (http://edoc.ub.unimuenchen.de/3483). Ji L. and Peters A.J., 2004. Forecasting vegetation greenness with satellite and climate data, Geoscience and Remote Sensing Letters, IEEE, 1, pp 3-6. Knyazikhin Y., J.V. Martonchik, R.B. Myneni, D.J. Diner, S. Runing, 1998. Synergistic algorithm for estimating vegetation canopy leaf area index and fraction of absorbed photosynthetically active radiation from MODIS and MISR data. J. Geophys. Res., 103, pp 32257-32275. Kogan F.N., 1990. Remote sensing of weather impacts on vegetation. Int. J. Remote Sens. 6, pp 1417-1434. Kogan F.N., 1994. NOAA plays leadership role in developing satellite technology for drought watch. Earth Observation Magazine, September a, pp 18-21. Kogan F.N., 1995. Droughts of the late 1980s in the United States as derived from NOAA polar-orbiting satellite data. Bull. Am. Meteorol. Soc., 76, pp 655-668. Kogan F.N., 1997. Global drought watch from space. Bull. Am. Meteorol. Soc., 78, pp 621636. Kogan F.N., 2000. Contribution of Remote Sensing to Drought Early Warning. In Early Warning Systems for Drought Preparedness and Drought Management, Proceedings of an Expert Group Meeting held September 57, 2000, Lisbon, Portugal. WMO/Technical Document N 1037, pp. 86100. World Meteorological Organization, Geneva, Switzerland (http://www.drought.unl.edu/monitor/EWS/EWS_WMO.html). Kogan F.N., 2001. Operational space technology for global vegetation assessment. Bulletin of the American Meteorological Society, 82 (9), pp 1949-1964. Lacaze B., and Bergs J.C., 2005. Contribution of Meteosat Second Generation (MSG) to drought early warning. Proceedings of the International Conference: Remote Sensing and Geoinformation Processing in the Assessment and Monitoring of Land Degradation and Desertification: State of the Art and Operational Perspectives, September 7th to 9th, Trier, Germany. Li Z., J. Cihlar, X. Zheng, L. Moreau, H. Ly, 1996. The bidirectional effects of AVHRR measurements over northern regions. IEEE Transactions on Geoscience and Remote Sensing. Monteith J.L., 1977. Climate and the efficiency of crop production in Britain. Philos. Trans. R. Soc. Lond. Ser. B, 281, pp 277-294. Muchoney D., J. Borak, H. Chi, M. Friedl, S. Gopal, J. Hodges, N. Morrow, A. Strahler, 2000. Application of the MODIS global supervised classification model to vegetation and land cover mapping of Central America, International Journal of Remote Sensing, 21, (6-7), pp 1115-1138(24).
198
NOAA, 2003. New Priorities for the 21st Century, NOAA's Strategic Plan for FY2003-2008 and Beyond March, 2003, specifically outlines NOAA's roles and responsibilities for providing quality climate data. Quattrochi D.A. and Luvall J.C., 2004. Thermal remote sensing in land surface processing 440 p, Lavoisier, Paris. Shimabukuro Y.E., V.C. Carvalho, B.F.T. Rudorff, 1997. NOAA-AVHRR data processing for the mapping of vegetation cover, International Journal of Remote Sensing, http://www.informaworld.com/smpp/title~content=t713722504~db=all~tab=issueslist~ branches=18 - v1818 (3), pp 671-677. Sivakumar M.V.K., R. Gommesb, W. Baier, 2000. Agrometeorology and sustainable agriculture, Agricultural and Forest Meteorology, 103 (2000), pp 11-26. Tsiros E., C. Domenikiotis, M. Spiliotopoulos, N.R. Dalezios, 2004. Use of NOAA/AVHRRbased vegetation condition index (VCI) and temperature condition index (TCI) for drought monitoring in Thessaly, Greece, EWRA Symposium on water resources management: risks and challenges for the 21st century, Izmir, Turkey, 2-4 September 2004, pp 769-782. Tsiros E., E. Kanellou, C. Domenikiotis, N.R. Dalezios, 2008. Identification of Water Limited Growth Environment Zones Using NOAA/AVHRR Data. 4th International Conference on Information and Communication Technologies in Bio and Earth Sciences HAICTA 2008, 18-20 September 2008, Athens, Greece (submitted). White M.A., P.E. Thornton, S.W. Running, 1997. A continental phenology model for monitoring vegetation responses to inter-annual climatic variability. Global Biogeochem. Cycles, 11, pp 217-234. Wiegand C.L., S.J. Maas, J.K. Aase, J.L. Hatfield, P.J. Pinter Jr., R.D. Jackson, E.T. Kanemasu, R.L. Lapitan, 1992. Multisite Analyses of Spectral-Biophysical Data for Wheat. Remote Sensing of Environment 42, pp 1-21. Yang W., J.C. Stroeve, N.V. Shabanov, D. Huang, B. Tan, P. Zhang, 2006. Analysis of Leaf Area Index and Fraction of PAR Absorbed by Vegetation Products from the Terra MODIS Sensor: 20002005, IEEE Transactions on Geoscience and Remote Sensing, 44 (7), pp 1829-1842. Yao Y., Q. Liu, X. Li, 2008. LAI retrieval and uncertainty evaluations for typical row-planted crops at different growth stages. Remote Sensing of Environment, 112 (1), pp 94-106. Zhang P., B. Anderson, M. Barlow, B. Tan, R. Myneni, 2004a. Climate related vegetation characteristics derived from MODIS LAI and NDVI. J. Geophys. Res. 109, D20105 doi:10.1029/2004JD004720. Zhang P., B. Anderson, B. Tan, D. Huang, R. Myneni, 2005. Potential monitoring of crop production using a satellite-based Climate-Variability Impact Index, Agricultural and Forest Meteorology, 132 (3-4), pp 344-358. Zhang X., M. Friedl, H. Yu, W. Wu, M. Shaikh, 2004b. Comparison of seasonal and spatial variations of leaf area index and fraction of absorbed photosynthetically active radiation from Moderate Resolution Imaging Spectroradiometer (MODIS) and Common Land Model, Journal of Geophysical Research, vol. 109, D01103. Zhang X., M.A. Friedl, C.B. Schaaf, A.H. Strahler, J.C.F. Hodges, F. Gao, B.C. Reed, A. Huete, 2003. Monitoring vegetation phenology using MODIS. Remote Sens. Environ. 84, pp 471475. www.esa.int/eo www.ceos.org www.ceos.org/pages/CEOSResponse_1010A.pdf www.eumetsat.int
199
www.nesdis.noaa.gov, www.orbit.nesdis.noaa.gov, www.ncdc.noaa.gov www.cr.usgs.gov/pub/imswelcome/ http://vegetation.cnes.fr
4.9 Annex 1
Fig. 4.1: Summer droughts of 2003, 2005 and 2006 in France. NDVI deviations (1 August/mean 2002-2004) based on VEGETATION/SPOT 5 data (by CNES, processed by O. Hagolle)
Fig. 4.2: The 2003 Heatwave at the European scale (July 2003/July 2002). Temperature map based on MODIS data (by NASA arth Observatory - VISIBLE EARTH)
Fig. 4.3: Atmospherically corrected NDVI map of Hungary. NOAA/AVHRR 10-day composite image of 1-10 September 1997
200
Fig. 4.4: Cloud Type of central Europe. METEOSAT-8 data of 24-5-2006, 15:30 UTC, processed with SAFNWC.
Fig. 4.5: Cloud cover image of NW Europe (NOAA, 4-7-2007, at 4:27 UTC: AVHRR ch 124 (combined)), available for subscribers at web-site serving farmers (FMI), provided only for operational use in agriculture
Fig. 4.6: Cloud cover of Europe (Meteosat, 4-7-2007, at 4:15 UTC: SEVIRI ch9 (IR 10,8m)), available for subscribers at website serving farmers (FMI), provided only for operational use in agriculture
201
Fig. 4.7: NDVI map of Tuscany (/AVHRR data).
Fig. 4.8: NDVI map of Italy. 16 days composite (Terra/MODIS data)
Fig. 4.9: NDVI map (SPOT/VEGETATION data)
of
Italy
202
Summer
NDVI Anomly 19822004
Summer
Fig. 4.10: Drought Effects on the Mid-Latitude Carbon Sinks, based on NDVI. A number of major droughts in mid-latitudes have contributed to the weakening of the growth rate of terrestrial carbon sinks in these regions (Angert et al. 2005, PNAS; Buermann et al. 2007, PNAS; Ciais et al. 2005, Science)
203
204
5. Satellite remote sensing as a tool for monitoring climate and its impact on the environmentpossibilities and limitations
5. SATELLITE REMOTE SENSING AS A TOOL FOR MONITORING CLIMATE AND ITS IMPACT ON THE ENVIRONMENT POSSIBILITIES AND LIMITATIONS
Piotr Struzik, Leonidas Toulios, Gheorghe Stancalie, Mark Danson, Janos Mika, Christos Domenikiotis
Abstract Application of satellite data available during last 47 years for detection of climate change and resulted changes to biosphere is presented. For a better understanding of satellite sensor history, the evolution of Earth remote sensing and meteorological satellite systems are described. Essential climate variables, which can be determined from satellite observations are listed. Necessary data stability and accuracy are discussed. Examples of possibilities and results from long satellite data series are presented focusing on: atmospheric sounding, precipitation, radiation budget, cloud cover, snow/ice cover, ozone. Sensitivity of temperature retrievals by hyperspectral satellite sensors to changes of atmosphere and surface properties due to climate change are analysed. Changes in the biosphere are presented as examples of the evolution of vegetation indices and ocean colour over long term periods. Difficulties in the use of satellite data for climate observations are analysed, taking into account: satellite orbit drift, sensors degradation, different methods of post-launch sensors calibration, variability of sensor spectral response in consecutive satellite missions. For biosphere observations, the limiting factor is cloudiness. Availability of cloud free scenes in mid latitudes is discussed. Problems in proper calibration of data and their representativeness are presented as an example of surface temperature changes over period 19782002 determined from ground and space observations. Benefits, limitations and challenges existing when satellite data are used for climate change analysis are discussed and summarized in the conclusions.
5.1 Introduction The main objective of COST 734 Action is the evaluation of the possible impacts arising from climate change and variability on agriculture. Evaluation of the current trend of agroclimatic indices requires information about changes
205
of climate and biosphere related variables. Taking into account rapid changes of mentioned variables during last 20-40 years, use of satellite information is possible and must be more deeply analysed. Remote sensing, due to spatial and temporal resolution of the sensors, can make possible analysis of large area phenomena including uninhabited areas, seas and oceans. Climate analysis and monitoring of changes are routinely based on classical meteorological observations performed for more than two centuries. Typical observations are located only on land and do not completely cover areas of oceans and seas which in fact cover the majority of Earths surface. The longest series of observations belong to stations which are located in the cities, where anthropogenic influence due to urbanization and increasing population cannot be neglected and significantly perturbs the temporal course of temperature or other climate related variables. Ground observation networks are dense in developed countries (Europe, North America), much more sparse in poor or sparsely populated areas (Sahara, Amazon jungle, Antarctica etc.). Additional meteorological observations are provided from ship reports extending land network to the ocean area. Typical daily coverage of observations made at meteorological stations and from ship reports are presented on Fig. 5.1.
Figure 5.1: Daily coverage of meteorological observations ground stations and ship reports (Obasi, 2003)
The need for extension of classical meteorological observations to the large areas resulted in inclusion of meteorological observations and measurements (manned and automatic) performed on ships and aircrafts. Unfortunately typical ship and aircraft routes still do not cover the whole globe. Most frequent observations are available from northern hemisphere where ship traffic is heavier. Still, large areas of our globe are not covered with meteorological data. Figure 5.2 presents example of aircraft weather reports and ship reports for selected 2 week periods.
206
Figure 5.2: Example of aircraft weather reports 1-15 Oct. 2001 (left) and ship weather reports 1-15 Apr. 2002 (right) coverage (Obasi, 2003)
Development of satellite remote sensing allowed for completely new possibilities in weather observations and collection of long time-series of measured variables. Data coverage, both spatial and temporal was significantly increased. Sensors installed on meteorological satellites were designed for measurement and observation of typical meteorological variables, which are also a base for climate monitoring. Additionally, satellite sensors make possible observations of land and sea surface and their features related to actual state of the surface vegetation, temperature, moisture, chlorophyll concentration, suspended matter, wind etc. The era of remote observations of the Earth started in fact in the 19th century from the first photographs taken from balloon. Space observations are dated from 1940s, followed by regular meteorological observations since the 1960s. Of course more precise instruments are available in space starting from 1980s. Such a long period of observations allows for climate studies based on Earth observations from space.
5.2 Evolution of meteorological satellite system To look at the Earth from above was a dream of mankind since for centuries. To capture a view and share it with others was the next dream. Realisation of this dream was started in the 19 century. The first aerial photo was taken by Gaspard-Flix Tournachon, better known as Nadar in 1858, using a tethered balloon over the Bievre Valley, France. Nadars aerial photos were lost - so the oldest surviving aerial photo, shown below (Fig. 5.3), was that of Boston USA in 1860, taken by James Wallace Black, also using a balloon. Following technology development, the first photograph from an airplane was taken by L. P. Bonvillain while flying in the Wright Brothers' craft in 1908 (http://www.aerialarts.com/History/history.htm).
207
Figure 5.3: Worlds oldest surviving aerial photo
On October 24, 1946, years before the Sputnik satellite opened the space age, a group of soldiers and scientists in the New Mexico desert saw something new and wonderful - the first pictures of Earth as seen from space (Fig. 5.4). The grainy, black-and-white photos were taken from an altitude of 65 miles by a 35-millimeter motion picture camera riding on a V-2 missile launched from the White Sands Missile Range. Snapping a new frame every second and a half, the rocket-borne camera climbed straight up, then fell back to Earth minutes later, slamming into the ground at 500 feet per second. The camera itself was smashed, but the film, protected in a steel cassette, was unharmed (http://www.airspacemag.com)
Figure 5.4: The first rocket-borne look at Earth from beyond the atmosphere (1946)
208
More than 1,000 Earth pictures were returned from V-2s between 1946 and 1950, from altitudes as high as 100 miles. The purpose of making such photos was navigation of space flights. Very quickly was found, that on the images mainly clouds are seen. The weather related applications born in 1950s and finalised with launch of the first meteorological satellite on April 1st, 1960 (Fig. 5.4). Since that time regular observations of weather phenomena were started. Since that time, rapid development of meteorological satellites has taken place. Television images useless during night time, were replaced by scanning radiometers operating also in infrared part of spectrum. First prototype satellite with such an instrument onboard was ITOS-1 launched in 1970 (Fig. 5.5). The need for frequent observations for monitoring of meteorological processes in the atmosphere resulted with launch of the first meteorological geostationary satellite in 1966, ATS-1 constructed by USA (Fig. 5.6 and 5.7). The first European geostationary meteorological satellite METEOSAT-1 was placed on 0 deg position in 1977. Full Earth coverage from geostationary orbit was achieved in 1990 after the launch of a Russian satellite positioned over Indian Ocean.
Figure 5.5: TIROS, USA, 01.04.1960, first picture at all taken by a meteorological satellite (left). ITOS, USA, 14.02.1970, prototype spacecraft for the second generation of operational sun-synchronous meteorological spacecraft, first IR (right)
209
Figure 5.6: First years of meteorological satellites development
Figure 5.7: Continuous development of global observing system for meteorology, focus European perspective (picture - courtesy W. Benesch)
The actual configuration of meteorological satellites is presented below (Fig. 5.8 and 5.9). Currently about 30 satellites are available for forecasters covering whole Earth including polar regions well observed by fleet of polar orbiting satellites of NOAA, METEOR and Feng Yun series with first European METOP-A.
210
Figure 5.8: Actual status of meteorological satellite system
Figure 5.9: Geostationary orbit coverage examples of images
211
The development of satellite from the user requirements to the launch, takes 10-20 years. Thats a reason of long term planning. At the moment perspective of 70 years of meteorological space observations is known (Fig. 5.10).
Figure 5.10: Future evolution of meteorological satellites system
Processes which take place in the atmosphere are four dimensional. They have determined geographical location, altitude and time. Monitoring of them require frequent observations, in certain cases in range of minutes (e.g. convection). Below is presented evolution of repetition rate of meteorological satellites with an outlook for near future (Fig. 5.11).
Figure 5.11: Evolution of repetition rate of meteorological satellites past, present and near future
212
5.3 Satellite climatology possibilities Meteorological satellites have been present in space since 47 years. Such a long period allows for climatological studies based on remotely sensed observations of the Earth and atmosphere. The longest series concern cloud observations but continuous improvement of instruments made possible also: air sounding, land surface properties observations, oceans and seas monitoring, snow and polar ice caps observations and many other application. Essential climate variables in three domains atmosphere, ocean and land, which can be observed/measured by satellite instruments are listed below in Table 5.1.
Table 5.1: Essential Climate Variables which can be observed by satellite sensors
Monitoring of climate require observations which are related to processes which are driving forces for possible changes. Space observations provide valuable information in global, continental and regional scale which help in better understand processes which are hardly detected by point ground measurements. Especially important are satellite observations, which help in key areas of uncertainty in understanding climate and Global Change, such as: Earths radiation balance and the influence of clouds on radiation and the hydrologic cycle, Oceanic productivity, circulation and air-sea exchange, Transformation of greenhouse gases in the lower atmosphere, with emphasis on the carbon cycle, Changes in land use, land cover and primary productivity, including deforestation, Sea level variability and impacts of ice sheet volume, Chemistry of the middle and upper stratosphere, including sources and sinks of stratospheric ozone, Volcanic eruptions and their role in climate change.
213
For analysis of long term processes related to climate, tools with high stability and low uncertainty are required. The question is whether instruments placed onboard of meteorological satellites are well characterised in terms of those parameters (Fig. 5.12). This paper is an attempt to answer this question.
Figure 5.12: Where satellite sensors can be placed on this diagram?
5.3.1 Satellite observations of processes in atmosphere selected examples. Satellite observations of the processes which take place in the atmosphere have the longest history. Since the first image from space taken in 40s, clouds are continuously observed and analysed providing valuable climatological datasets. But processes taking place in atmosphere concern not only cloudiness. Satellite sensors measure: water wapour content and distribution in time and space, airmass circulation, wind speed and direction, ozone content, other trace gases. Such measurements are possible with use of sounding instruments using both infrared channels and microwave measurements located at absorption bands of H2O, CO, CO2, O3 and other gases of Earth atmosphere. Brief history presenting evolution of sounding instruments is listed below: ATMOSPHERE TEMPERATURE SOUNDING: In 1969 Nimbus 3 carried the first of a new class of remote-sounding sensors, the Space Infra-Red Sounder (SIRS A), First operational sounder system in 1972, the Vertical Temperature Profile Radiometer (VTPR), aboard the NOAA 2 In 1978, Next generation (TIROS N) was launched with an improved 20channel High Resolution Infrared Sounder (HIRS) accompanied by the
214
Microwave Sounder Unit (MSU) and Stratospheric Sounder Unit (SSU) forming the TIROS Operation Vertical Sounder (TOVS), In 2002 NASA launched Aqua satellite carrying the first hyperspectral spectrometer Atmospheric Infrared Sounder (AIRS) with 2378 channels in the infrared (3.7 15.4 m) and 4 channels in the visible part of spectrum, 2006 first METOP satellite with hyperspectral interferometer IASI (8546 channels in 3.6- 15.5 m spectrum).
The importance of hyperspectral sounding for climate applications is described in Fig. 5.13, where in the infrared part of spectrum is depicted sensitivity of remote sensed brightness temperature, to the changes of typical parameters describing our environment (emissivity, surface temperature, low level tropospheric moisture, entire troposphere moisture and ozone content). Changes in our environment related to global warming are well detected by hyperspectral sensors in different parts of spectrum depending on observed parameters. Simulation of spectral response as a result of climate change is presented taking AQUA satellite AIRS sensor as an example.
Figure 5.13: Brightness temperatures changes of AIRS hyperspectral sensor due to atmospheric and surface anomalies related to climate change
215
Ground measurements of temperature, especially for really long time-series suffers problem of urbanisation. Meteorological stations located outside of the cities are becoming inside the cities due to they grow. The amount of heat released is continuously growing due to industry, infrastructure, traffic etc. As a result, part of temperature increase is anthropogenic. Due to availability of satellite missions with temperature sounding measurements, studies on temporal behaviour of temperature were done. Figure 5.14 presents comparison of mean temperature of our globe in period 1978-2002 retrieved from ground observations and in the frame of two satellite projects. The surface temperature trend is +0.20 0.06C decade1. The linear trend through 2002 for the UAH T2 product is 0.03 0.09C decade1 (University of Alabama Huntsville UAH, Christy et al., 2003). The linear trend through 2002 for the RSS is 0.11 0.09C decade1 (Remote Sensing System RSS, Mears et al., 2003). Analysis based on satellite data suggest much lower (if really exist) trend of temperature increase.
Figure 5.14: Annual mean anomalies of global average temperature (1979-2002) for the lower troposphere from satellites (T2) and for the surface (SFC T)
One of the most interesting problems related to human influence to the atmosphere is ozone depletion in stratosphere, causing increased UV radiation, especially in polar region. Ozone plays different role depending of altitude where exist (Committee on Earth Studies, 2000): In the stratosphere, where 90 percent of atmospheric ozone resides, ozone plays a critical role in absorbing ultraviolet (UV) radiation and preventing it from reaching Earth's surface.
216
In the upper and middle troposphere, ozone is a major greenhouse gas, causing inhomogeneous radiative forcing. In the lower and middle troposphere, ozone maintains the oxidizing power of the atmosphere by providing a source of the hydroxyl radical (OH) in the presence of water vapor. Oxidation by OH is the main sink for a number of environmentally important gases, including methane (CH4), carbon monoxide (CO), hydrofluoro-carbons, and methyl bromide. In surface air, ozone is a pernicious pollutant, toxic to humans and vegetation. It is the principal contributor to smog over the United States. Satellite sounding instruments measure total ozone content and ozone profile since the first instrument onboard NOAA satellites. Merge ozone dataset created with use of different space borne instruments presents temporal changes in most populated region 50 deg. S to 50 deg N (Fig. 5.15). Ozone depletion trend, which was evident in 80-ties and 90-ties were recently stopped according to results presented on Figure 5.16.
Figure 5.15: Instrumets used for creation of Merged Ozone Dataset and total ozone temporal changes in period 1979- 2006 (Gleason and Butler, 2007)
217
Figure 5.16: Total ozone trend in last 25 years as measured by satellite instruments (Yang et al., JGR-Atm 111, D17309, 2006)
5.3.2 Satellite observations of processes at the Earth surface selected applications Meteorological satellites were designed for observation of phenomena taking place in atmosphere but they provide also information on actual state of the surface allowing continuous monitoring of biosphere spatial and temporal changes. Parameters which can be obtained are: vegetation status, snow/ice cover, radiation balance and many others. The type and distribution of vegetation native to a geographic region are diagnostics of the areas climate. This is because vegetation integrates the effects of precipitation and temperature over all time frames longer than a few days. In addition the vegetation Leeds back into climate because of the plant species contribution to the surface energy and moisture balance and its impact on surface roughness and albedo. For these reasons, observing vegetation changes in the seasonal to inter-annual time frame and over long term is important to climate monitoring. Examples of long series of vegetation indices anomalies (1981-2001) based on AVHRR/NOAA data, for different climatic zones are presented below. Two factors influencing vegetation measurements from space are clearly seen: volcanic eruption and satellite data degradation for selected satellites. Vegetation index monthly anomaly time series for period July 1981 to December 2000 is presented on Figure 5.17. The impact of satellite drift is clearly noticeable, especially in the case of NOAA 11 and 14. Likewise, the impact of the Mt. Pinatubo eruption in June 1991 and El Chichon in March 1982 is also discernable.
218
Figure 5.17: Vegetation anomalies in period 1981-2001, in different latitude zones. Arrows indicate time of greatest volcano eruptions during last 20 years
Long term monitoring of vegetation status make possible to detect temporal changes of beginning and finish of vegetation season. Example of such studies for latitudes above 45 deg is presented on Figure 5.18. During the period 1981-1994, both spring and autumn dates were shifted, resulting with longer vegetation season. Also maximal values of NDVI (Normalised Difference Vegetation Index) were grown during mentioned 13 years.
Figure 5.18: Temporal changes of beginning and finish of vegetation season based on satellite measured NDVI (Ohring et al., 2004)
219
AVHRR observations suggest that the growing season increased between 1981 and 1994 by 10%, but questions remain, mainly with respect to calibration and inter-calibration of sequential satellite instruments. The main question is whether is it greening trend? But other possibilities are: orbital drift, intersensor variation or simply noise in the channel data. Time series of NDVI, derived from NOAA/AVHRR have also been extensively used in vegetation monitoring, crop yield assessment/forecasting, and hazard detection and mapping (Tucker and Choudhury, 1987; Benedetti and Rossini 1993, Hayes et al., 1996). However, the variability of NDVI values is related to the contribution of geographic resources to the amount of vegetation. This contribution fluctuates considerably depending mainly on climate, soils, vegetation type, and topography of an area. For the purposes of estimating the impact of the weather on vegetation, the nonweather effects have to be filtered out. Studies have been carried out in several parts of the world (Subbiah, 1993; White et al., 1993; Kogan, 1997) showing further potential of satellite-based data in climate impact assessment. Examination of the temporal changes along the vegetation growth period as a function of the time-varying photosynthetic activity and according to the background environmental conditions (rainfall, air temperature), allows derivation of the potential vegetation cover (according to soil conditions and environmental factors). These temporal differences could be an indicator of expected productivity which varies with the ecosystem potential and the specific conditions prevailing in a given site (for a given year or seasonal cycle) (Domenikiotis et al., 2004a). A vegetation index that seems to be appropriate for incorporating these conditions is the Vegetation Condition Index (VCI). VCI separates the short-term weather signal in the NDVI data from the long-term ecological signal (Kogan, 1997). VCI is a vegetation index adjusted for land climate, ecology and weather conditions. It has excellent ability to detect the existing anomaly conditions and to measure the time of their onset and intensity, duration and impact on vegetation. The VCI algorithm was developed and tested in several areas of the world with different environmental and economic resource (Kogan, 1995; Domenikiotis et al., 2002; Tsiros et al., 2004). In many parts of the world empirical relationships between VCI and yield or production have been developed for early production/yield assessment (Dabrowska-Zielinska et al., 2002; Domenikiotis et al., 2004a). Higher accuracy assessment has been obtained based on the application of the Bhalme and Mooley Drought Index (BMDI) methodology on the Vegetation Condition Index (VCI) extracted by NOAA/AVHRR data (Domenikiotis et al., 2004b). Domenikiotis et al. (2005) developed empirical relationships between the VCI values derived from a time series of NDVI ten-day Maximum Value Composite (MVC) images, for 18
220
successive years (19821999) and cotton production. The predicted values showed 5% percentage deviation from the actual production, at regional and country scale. For the identification of vegetative stresses and their impact in vegetation health Kogan (2001) proposed the Vegetation Health Index (VHI) which represents overall vegetation health and used it for agricultural drought mapping. VHI have been used for the detection of agricultural drought (Tsiros et al., 2004; Bhuiyan et al., 2006). VHI derived from a time series of NOAA/AVHRR ten-day images, for 20 successive hydrological years has been recently used for agroclimatic classification (Tsiros et al., 2008) and for monitoring agricultural drought (Kanellou et al., 2008). Changes in snow and ice cover represent potential changes in climate forcing due to the snow-albedo feedback mechanism. Satellite sensors used for snow and ice monitoring use visible, infrared and microwave parts of spectrum. Data records for long term studies begin in 60-ties and are continuous until present time. The example of Northern Hemisphere snow cover anomalies from November 1996 to October 2003 calculated from NOAA snow maps and colour coded by season are presented on Fig. 5.19. The record provides the longest, most consistent snow cover product available for documentating the state of the environment.
Figure 5.19: 12-month running anomalies of hemispheric snow extent (left), monthly anomalies are colour coded by season: fall: orange; winter: blue; spring: green; summer: red (Climate Data Records from Environmental Satellites: Interim Report, National Academies P)
Oceans cover 70% of the earth surface, playing a major role in the spatial and temporal distribution of weather patterns, the production of natural resources and the large-scale transport and storage of greenhouse gases. For example, marine physical, chemical, and biological processes are responsible for the
221
removal of carbon dioxide (CO2) from the atmosphere in certain regions and its release in other. The global rates of marine CO2 sequestration (i.e. the balance between global atmospheric removal and release) and the rates of transport are a function of both ocean circulation and biological activity. To this end, the use of satellite-derived oceanographic observations provides the means to characterize large-scale seasonal, inter-annual, and decadal spatial changes in physical and biological sea surface properties. These changes can then be statistically coupled to long term changes in atmospheric patterns to assess the sensitivity of oceanic processes to climate change. Satellite-based oceanographic observations provide unique regional and basin-scale datasets that can be used to test and improve our climate models and constrain error estimates in our forecasts. Observed from satellites parameters are: sea surface temperature, surface circulation patterns, surface wind speed and direction (scatterometer), surface roughness (synthetic aperture radar), ocean colour. In many respects ocean colour is the marine analogue to vegetation dynamics and land cover, and it is motivated by similar concerns. Marine phytoplankton accounts for approximately half of the global annual primary production, and thus plays a significant role in the global carbon cycle. The most widely used product derived from ocean color measurement is chlorophyll concentration (mg m3), a measure of phytoplankton biomass (Fig. 5.20 and 5.21). Ocean colour remote sensing has been a successful area of technological development since the Coastal Zone Color Scanner (CZCS), a proof-ofconcept sensor launched in 1978 onboard Nimbus 7.
Figure 5.20: Chlorophyll concentration on global and regional scales based on MODIS (left) and SEAWifs satellite data (right)
222
Figure 5.21: Chronology of Ocean Colour satellite Missions (Letelier et al., 2005)
5.4 Difficulties in use of satellite data for climate observations Most of the operational satellites were created as weather rather than climate platforms. As a result, long term absolute accuracy of satellite measurements was not a crucial issue. In the measurement of the climate variable it is vital for understanding climate processes and changes. However, it is not as necessary for determining long-term changes or trends as long as the data set has the required stability. And, when it comes to building satellite instruments, stability appears to be less difficult to achieve than accuracy. The difficulty arises because of the many known and unknown systematic uncertainties that are to be accounted for in the calibration of the instruments. Although excellent absolute accuracy is not critical for trend detection, it is crucial for understanding climate processes and changes. During creation of satellite based Climate Data Records, unique challenges appear: the need to manage extremely large volumes of data; restrictions of spatial sampling and resolution; accounting for orbit drift and sensor degradation over time; temporal sampling; difficulty of calibrating after launch (e.g., vicarious or onboard calibration); the need for significant computational resources for reprocessing. A chronic difficulty in creating a continuous, consistent climate record from satellite observations alone is that satellites and instruments have a finite lifetime of a few years and have to be replaced, and their orbits are not stable. Most important is proper calibration of satellite sensors during their entire time. This can be done by: pre-launch calibration, post-launch vicarious calibration, intercalibration.
223
Nominal calibration involves determining the calibration of a single sensor on a single platform, and while this is considered standard prelaunch practice, it is important to calibrate the sensor in orbit as well (Guenther et al, 1997; Goody and Haskins, 1998; Byerly et al., 2002). Vicarious calibration monitoring involves measuring a known target or comparing the satellite signal with simultaneous in situ, balloon, radiosonde, or aircraft measurements (Fig. 5.22).
Figure 5.22: Comparison of albedo measurements with and without vicarious calibration (Rao and Chen, 1995)
These instruments should undergo vicarious calibration monitoring at regular intervals, regardless of on-board nominal calibration, to prevent drifting of the data over time due to orbital drift and drift in the observation time, which aliases the diurnal cycle onto the record. Satellite-to-satellite cross-calibration involves adjusting several same-generation instruments to a common baseline, and this is particularly important for long term studies, as each sensor will have slightly different baselines even if they are built to the same specifications (Ohring et al., 2004). Without proper post launch calibration, spurious trends in the data can occur. The problem is with selection of the objects with stable properties, used for vicarious calibration. Different objects with stable albedo are used for calibration of visible sensors, like deserts, ice caps, dense tropical vegetation or even Moon used for SeaWifs postlaunch calibration. Differences between sensors of consecutive satellites require intercalibration of satellite data. Figure 5.23 presents spectral response functions of first 2 channels of AVHRR instrument used for long term studies of vegetation anomalies (NOAA 6-16, MODIS/TERRA, SPOT and ADEOS satellites).
224
Figure 5.23: Differences in spectral characteristics of NOAA AVHRR sensors (Latifovic et al., 2004)
Required accuracies and stabilities for climate variable data sets were listed in Appendix 1. Taking values from Appendix 1 (Table 5.1), requirements for satellite instruments were computed. Finally assessment which parameter measured by satellite data fit requirements is presented in Table 3 (state for 2002) (Source: Satellite Instrument Calibration for Measuring Global Climate Change (Report of a Workshop at the University of Maryland Inn and Conference Center, College Park, MD, November 12-14, 2002) NISTIR 7047, edited by George Ohring NOAA/NESDIS (Consultant), Bruce Wielicki NASA Langley Research Center, Roy Spencer NASA Marshall Space Flight Center, Bill Emery University of Colorado, Raju Datla NIST). Stability requirements are being met, or appear to be close to being met for solar irradiance, cloud cover, cloud temperature, cloud height, atmospheric temperature, total column water vapour, ozone, ocean colour, snow cover, and sea ice measurements. Long term data sets have been assembled for many of these variables by stitching together observations from successive satellites and exploiting satellite overlap periods to account for systematic differences between successive instruments. Time series of climate variables have been constructed from those series. Among the problems which occur using different satellite platforms is satellite drift causing a change in the local time
225
of the observations during each satellites lifetime, especially for the NOAA satellites (Fig. 5.24).
Figure 5.24: Drift of orbital parameters of NOAA
Unfortunately, for many climate variables, current-observing systems cannot meet both accuracies and stabilities (Appendix 1). In some cases, we dont know whether current systems are adequate, and studies are needed to answer the question (question marks in the Appendix 1 Table 5.3). Cloud cover is a major constraint on optical remote sensing, whether it is space borne, airborne or ground-based observation, particularly in cloudy regions such as the United Kingdom presented as example (Armitage et al.,
226
2007; Ramirez et al., 2007). The impact of cloud cover on operational applications that require a time series of images can be significant (Kontoes and Stakenborg, 1990; Fuller et al., 1994; Hollingsworth et al., 2001; Intrieri et al., 2002). Clouds provide a major impediment to passive remote sensing at visible and infrared wavelengths. The presence of scattered clouds can cause objects of interest to be obscured, or can cast shadows which causes problems when processing images. Total cloud cover prevents any observation of the ground from space borne sensors and can severely limit data collection from airborne sensors. Cloud cover frequencies have a major effect on climatological applications of remote sensing, particularly when regular repeat data collection is required. The example (Fig. 5.25) clearly shows that some regions are continuously obscured by clouds (only a few cloud free days in year). Determination of any surface parameters or features with use of remote sensing is highly difficult.
Figure 5.25: Spatial distribution of cloud-free imagery frequencies for whole year and selected individual months, derived from the MODIS Cloud Mask SDS product at the 95% certainly level, across the UK in 2005 (Armitage et al., 2007)
227
5.5 Conclusions Satellite data offer an unprecedented potential for climate research provided that separate sensor/satellite data are integrated into high-quality, globallyintegrated climate products. Also climate change influence on biosphere can be monitored with use of satellite data. Presence of meteorological and environmental satellites in space since the 1960s allows for real climatological studies. Examples of successful use of long time satellite data series were presented in this study together with description of limitations of this technology. Main issues are accuracy and stability of satellite measurements. Actually, not all climatic related variables can be traced with use of satellite sensors due to their not sufficient accuracy (Appendix 1). Much improved post-launch calibration of satellite instruments, and intercalibration of similar instruments flying on different satellites is highly required to achieve continuity of observations. This requires overlapping periods of consecutive satellite missions. Other problems concern data management (processing and reprocessing). Rapid development of Earth observations resulted extremely huge volume of satellite data. Regarding future missions, new and more accurate sensors are envisaged.
5.6 References
Armitage R.P., F.A. Ramirez, E.Y. Ogunbadewa, F.M. Danson, 2007. Comparison of AVHRR and MODIS Cloud Products for Estimating Cloud Cover Probabilities for the United Kingdom, Proceedings of Remote Sensing and Photogrammetry Society, Annual Conference 2007, Newcastle, UK. Benedetti R. and Rossini P., 1993. On the use of NDVI profiles as a tool for Agricultural Statistics. The case study of wheat yield estimate and forecast in Emilia Romagna. Remote Sensing of Environment, 45, pp 311-326. Bhuiyan C., R.P. Singh, F.N. Kogan, 2006. Monitoring drought dynamics in the Aravalli region (India) using different indices based on ground and remote sensing data. International Journal of Applied Earth Observation and Geoinformation, 8, pp 289-302. Byerly W.P, J.E. Clement, T. Dorman, J. Engel, R. Julian, S.W. Miller, J.B. Young, J.A. Walker, 2002. Radiometric calibration, Visible/Infrared Imager/Radiometer Suite, Algorithm Theoretical Basis Document, Version 5, Rev. 1, Raytheon ITSS. Christy J.R., R.W. Spencer, W.B. Norris, W.D. Braswell, D.E. Parker, 2003. Error estimates of version 5.0 of MSU-AMSU bulk atmospheric temperatures. J. Atmos. Oceanic Technol., 20, pp 613-629. Dabrowska-Zielinska K., F.N. Kogan, A. Ciolkosz, M. Gruszczynska, W. Kowalik, 2002. Modelling of crop growth conditions and crop yield in Poland using AVHRR-based indices. Int. Journal of Remote Sensing. 23, pp 1109-1123. Domenikiotis C., M. Spiliotopoulos, E. Tsiros, N.R. Dalezios, 2004a. Early Cotton Yield Assessment by The Use Of The NOAA/AVHRR Derived Drought Vegetation Condition Index In Greece, Int. Journal of Remote Sensing, 25, pp 2807-2819.
228
Domenikiotis C., M. Spiliotopoulos, E. Tsiros, N.R. Dalezios, 2002. Application of NOAA/AVHRR VCI for drought monitoring in Thessaly, Greece. 6th Int. Conference of Protection and Restoration of the Environment, 1-5 July 2002, Skiathos, Greece, pp 1663-1670. Domenikiotis C., M. Spiliotopoulos, E. Tsiros, N.R. Dalezios, 2004b. Early cotton production assessment in Greece based on the combination of the drought vegetation condition index (VCI) and Bhalme and Mooley drought index (BMDI). Int. Journal of Remote Sensing, 25, pp 5373-5388. Domenikiotis C., E. Tsiros, M. Spiliotopoulos, N.R. Dalezios, 2005. Zoning of cotton production areas based on NOAA/AVHRR images. Int. Symposium in GIS and Remote Sensing: Environmental Applications, Volos, Greece, 7-9 November 2003, pp 119-132. Fuller R.M., G.B. Groom, S.M. Wallis, 1994. The availability of Landsat TM images of Great Britain. Int. Journal of Remote Sensing, 15(6), pp 1357-1362. Gleason J. and Butler J., 2007. From EOS, through NPP, to NPOESS: The Satellite Climate Data Record, presentation at AMS 2007 Conference. Goody R.M. and Haskins R., 1998. Calibration of radiances from space, J. Climate 11, pp 754-758. Guenther B., J. Butler, P. Ardanuy, 1997. Workshop on Strategies for Calibration and Validation of Global Change Measurements. NASA Reference Publication 1397, 125 p. Hayes M.J. and Decker W.L., 1996. Using NOAA AVHRR data to estimate maize production in the United States corn belt. Int. Journal of Remote Sensing, 17, pp 3189-3200. Hollingsworth B., L. Chen, S.E. Reichienbach, R. Irish, 2001. Cloud cover in Landsat observations of the Brazilian Amazon. Int. Journal of Remote Sensing, 22, pp 38553862. Intrieri J.M., M.D. Shupe, T. Uttal, B.J. McCarty, 2002. An annual cycle of Arctic cloud characteristics observed by radar and Lidar at SHEBA. Journal of Geophysical Research, 107, pp 1-15. Issues in the Integration of Research and Operational Satellite Systems for Climate Research: Part I. Science and Design, Committee on Earth Studies, Space Studies Board, National Research Council, The National Academies Press, 2000. Kanellou E., E. Tsiros, C. Domenikiotis, N.R. Dalezios, 2008. Meteorological and Agrohydrological drought monitoring based on conventional and remotely sensed data. European Geosciences Union General Assembly 2008, Vienna, Austria, 13-18 April 2008 (accepted). Kogan F.N., 1995. Droughts of the late 1980s in the United States as derived from NOAA polar orbiting satellite data. Bulletin of the American Meteorological Society, 76, pp 655-668. Kogan F.N., 1997. Global drought watch from space. Bulletin of the American Meteorological Society, 78, pp 621-636. Kogan F.N., 2001. Operational space technology for global vegetation assessment. Bulletin of the American Meteorological Society, 82, pp 1949-1964. Kontoes C. and Stakenborg J., 1990. Availability of cloudfree Landsat images for operational projects. The analysis of cloud-cover figures over the countries of the European Community. Int. Journal of Remote Sensing, 11(9), pp 1599-1608. Latifovic R., A. Trishchenko, J. Chen, B. Park, R. Fernandes, D. Pouliot, K. Khlopenkov, J. Cihlar, C. Ungureanu, 2004. Generating Satellite Climate Data Record Over Canadian
229
Land Mass. Satellite Direct Readout Conference, Miami, Florida, December 6-10, 2004. Letelier R.M., M.R. Abbott, P. Strutton, 2005. Ocean Color Climate Data Records Workshop Report, August 2005, Oregon State University. Mears C.A., M.C. Schabel, F.J. Wentz, 2003. A re-analysis of the MSU channel 2 tropospheric temperature record. J. Climate, 16 (22), pp 3650-3664. National Research Council (NRC), 2004. Climate Data Records from Environmental Satellites. National Academies Press, Washington, DC, 116 p. Obasi G.O.P., 2003. The Role and Importance of WMO and the National Meteorological and Hydrological Services for Sustainable Development, WMO, Geneva. Ohring G., B. Wielcki, R. Spencer, B. Emery, R. Datla, 2004. Satellite instrument calibration for measuring global climate change, Rep. NISTIR 7047, 101 pp., Natl. Inst. of Stand. and Technol., Gaithersburg, Md. Ramirez A., R.P. Armitage, F.M. Danson, C.G. Marston, E.Y. Ogunbadewa, M. Yebra, 2007. Web-based model for analysis of time series remotely sensed data. 10th Int. Symposium on Physical Measurements and Signatures in Remote Sensing, Davos, 1214th March 2007. Rao C.R.N. and Chen J., 1995. Inter-satellite calibration linkages for the visible and near infrared channels of the AVHRR on the NOAA-7, -9 and -11 spacecrafts. Int. Journal of Remote Sensing, 16, pp 1931-1942. Satellite Instrument Calibration for Measuring Global Climate Change (Report of a Workshop at the University of Maryland Inn and Conference Center, College Park, MD, November 12-14, 2002), edited by G. Ohring, B. Wielicki, R. Spencer, B. Emery, R. Datla, 2004. Subbiah A.R., 1993. Indian drought management from vulnerability to resilience. In :Wilhite, D. A. (Eds), Drought Assessment, Management and Planning Theory and Case, Kluwer Academic, Boston, pp 157-181. Tsiros E., C. Domenikiotis, M. Spiliotopoulos, N.R. Dalezios, 2004. Use of NOAA/AVHRRbased vegetation condition index (VCI) and temperature condition index (TCI) for drought monitoring in Thessaly, Greece, EWRA Symposium on water resources management: risks and challenges for the 21st century, Izmir, Turkey, 2-4 September 2004, pp 769-782. Tsiros E., E. Kanellou, C. Domenikiotis, N.R. Dalezios, 2008. Identification of Water Limited Growth Environment Zones Using NOAA/AVHRR Data. 4th International Conference on Information and Communication Technologies in Bio and Earth Sciences, HAICTA 2008, 18-20 September 2008, Athens, Greece (submitted). Tucker C.J. and Choudhury B.J., 1987. Satellite remote sensing of drought conditions. Remote Sensing of Environment, 23, pp 243-251. White D., D. Collins, M. Howden, 1993. Drought in Australia, prediction, monitoring, management and policy. In :Wilhite, D. A. (Eds), Drought Assessment, Management and Planning Theory and Case, Kluwer Academic, Boston, pp. 213-237. Yang E.S., D.M. Cunnold, R.J. Salawitch, M.P. McCormick, J. Russell, J.M Zawodny, S. Oltmans, M.J. Newchurch, 2006. Attribution of recovery 1 in lower-stratospheric ozone. JGR-Atm, 111, March 2006, 57p. http://www.aerialarts.com/History/history.htm http://www.airspacemag.com
230
5.7 Appendix 1 Required accuracies and stabilities for climate variable data sets were listed in a table below. Taking values from Table 5.2, requirements for satellite instruments were computed. Finally assessment which parameter measured by satellite data fit requirements is presented in Table 5.4 (state for 2002) (Source: Satellite Instrument Calibration for Measuring Global Climate Change (Report of a Workshop at the University of Maryland Inn and Conference Center, College Park, MD, November 12-14, 2002) NISTIR 7047, edited by George Ohring NOAA/NESDIS (Consultant), Bruce Wielicki NASA Langley Research Center, Roy Spencer NASA Marshall Space Flight Center, Bill Emery University of Colorado, Raju Datla NIST).
Table 5.2: Required accuracies and stabilities for climate variable data sets. Column labeled signal indicates the type of climate signal used to determine the measurement requirements
231
232
Table 5.3: Required accuracies and stabilities of satellite instruments to meet requirements of Table 5.2. The instrument column indicates the type of instrument used to make the make the measurement
233
Table 5.4: Ability of current observing systems to meet accuracy and stability requirements
234
235
236
6. Use of climate change scenarios in agrometeorological studies: past experiences and future needs
6. USE OF CLIMATE CHANGE SCENARIOS IN AGROMETEOROLOGICAL STUDIES: PAST EXPERIENCES AND FUTURE NEEDS Pierluigi Calanca, Lucka Kaifez Bogataj, Tom Halenka, Emmanuel Cloppet, Jnos Mika
Abstract As a contribution to COST 734, a survey was conducted among the signatory countries concerning the use of climate change scenarios in agrometeorological studies. The questionnaire was structured in such a way as to characterize the past experiences and identify the future needs. The survey clearly indicated that coordinated efforts are needed to: (i) provide a coherent view of climate change across all European countries; (ii) systematically quantify uncertainties in climate change projections; (iii) develop common tools for downscaling; and, (iv) better account for the risks related to extreme events.
6.1 Introduction Climate change scenarios are the backbones of impact studies, providing the basis for assessing the impact of climate change on human activities, including agriculture (e.g. Parry et al., 2007). Scenarios are neither predictions nor forecasts in a traditional sense; rather they are images of the future, or alternative futures that are meant to assist in climate change analysis (Nakicenovic and Swart, 2000). Scenarios are inherently uncertain, a point often neglected in impact studies. This does not necessarily imply that climate change scenarios are unreliable: considerable efforts have indeed been produced during the last decade to improve our knowledge of the climate system and our capacity to model its dynamics. This evolution is well documented in the reports issued by the Intergovernmental Panel on Climate Change (IPCC), in particular the second (SAR, Houghton et al., 1995) third (TAR, Houghton et al., 2001) and fourth assessment reports ( AR4, Solomon et al., 2007). Advances are also evident in
237
the outcomes of research projects promoted within the 5th and 6th framework programmes of the European Union, such as PRUDENCE1 or ENSEMBLES2. Impact studies with a focus on agriculture have not exclusively relied on climate change scenarios simulated with global (GCM) or regional climate models (RCM), sometimes taking arbitrary, synthetic (usually obtained by specifying ad-hoc incremental annual, seasonal or monthly anomalies for the long-term mean temperature and precipitation) or analogue scenarios as a starting point for the analysis. In earlier days the choice of scenarios was often dictated by practical considerations (e.g. technical difficulties, limited amount of resources available for the studies) rather than by the objectives of the study. In agrometeorological studies, while working with scenarios, difficulties have often been encountered in relation to the disparity of scales between global and regional climate models, on the one hand, and application models, on the other hand. For this reason, GCM or RCM results are usually downscaled with the help of high resolution atmospheric models (dynamical downscaling) and/or statistical models and stochastic weather generators (statistical downscaling). From the point of view of the physical consistency of the results, dynamical downscaling is certainly the method of choice. However, this approach is much more demanding in terms of computational resources. Statistical downscaling has the advantage of being computationally fast, but tends to introduce additional uncertainties in the scenarios. Moreover, while statistical downscaling of temperature and precipitation has received much attention in the past, the same cannot be said concerning radiation, air humidity and wind speed, and in relation to extreme events. Apart from the technical issues mentioned in the previous paragraphs, a coherent use of climate scenario to study the possible impacts of climate change agriculture was, at times, hindered by other kinds of difficulties, including institutional problems (e.g. lack of sufficient human resources). Although COST 734 can not solve all of these problems, by the fact that it has a clear focus it can provide a platform for coordinating national initiative, exchanging valuable information and experiences, setting the focus for new projects and pin point remaining weaknesses in the attempt to better understand climate change and its effects on European agriculture.
1 2
http://prudence.dmi.dk, accessed 20/01/2008 http://ensembles-eu.metoffice.com/, accessed 20/01/2008
238
To better appreciate the current state of knowledge and research in the different countries participating in COST 734, a questionnaire was distributed to the COST 734 members in spring 2007 to collect detailed information about the use of climate scenarios in agrometeorological studies conducted during the past 10 years. A total of 16 answers were returned to WG3, and this report provides an overview of the collected information. However, rather than just going through each of questions, the text is developed in such a way as to give a more general perspective of the main themes. This is also reflected in the selection of references. Those cited in the text have been chosen to provide an adequate coverage of the main topics. The full list of the references included in the questionnaire returned is provided separately.
6.2 Climate scenarios 6.2.1 Types of scenarios used in the past in agroclimatological studies In first instance, the members were asked to report about the types of scenarios use in the past in agrometeorological studies. The results of the survey are summarized in Table 6.1.
Table 6.1: Summary of the scenarios used in impact studies during the last 10 years Country Austria Scenarios Scenarios from 2 GCMs (ECHAM4 and ECHAM5), dynamically downscaled to 10 km with RCMs, and further statistically downscaled to 1 km. GCM scenarios from the IPCC DDC (ECHAM4, HadCM2, CGCM1, CSIRO-Mk2b, GFDL-R15), with IS92a emission pathway. Downscaling of ECHAM4 and HadCM2 achieved with the MAGICC and SCENGEN. Time windows: 2010-2039, 2040-2069, 2070-2099. arbitrary & analogue scenarios Simulations with ECHAM5-MPI-OM and A2 emission pathway. Time window: 2041-2070. GCM scenarios from the IPCC-DDC (CCSR/NIES, CGCM1, CSIRO-Mk2, ECHAM4/OPYC3, GFDL-R15, HadCM2, NCAR DOE-PCM) with IS92a emission pathway GCM scenarios from the IPCC-DDC (HadCM3, ECHAM4/POYC3, CSIRO-Mk2, NCAR DOE-PCM, CGCM2, GFDL-R30) RCM scenarios from PRUDENCE Transient GCM simulations with ARPEGE and A2 and B2 emission pathways Simulations with ECHAM4 and ECHAM5, and A1B, A2 and B1 emission pathway
Bulgaria
Croatia Czech Republic
Finland
France Germany
239
Greece
Hungary
Italy
Norway Poland
Romania Slovenia
Spain
Switzerland
United Kingdom
Simulations with HadCM3 (provided through STARDEX), RegCM3 driven with NCEP/ERA-40, PRECIS (A2 and B2 emission pathways), and CGCM2 (A2 and B2 emission pathways) PRUDENCE A2 and B2 scenarios GCM outputs, used directly or scaled by MAGICC/SCENGEN Empirica/statistical regression local vs. global variables GCM scenarios from the CMIP database RCM scenarios from the STARDEX, on the background of the HadAM3P global runs, and with reference to the A2 and B2 emissions pathways GCM scenarios from HADAM3H and ECHAM4, dynamically downscaled with HIRHAM Earlier studies based on GCM scenarios from GISS and GFDL (2 x CO2). Later GFDL-R15 and ECHAM4 for time horizon 2025. Analogue scenarios used for time horizons 2010 and 2025. GCM scenarios from GISS and CCCM (2 x CO2). GCM scenarios from the IPCC-DDC (CSIRO-Mk2, HadCM3, DOE-NCAR/PCM, ECHAM4-OPYC3), statistically downscaled. RCM scenarios from the PRUDENCE project and from simulations carried out at the University of Castilla-La Mancha (UCM) GCM scenarios from the IPCC-DDC (CCSR-NIES1, CGCM1, CGCM2, ECHAM4-OPYC3, DOE-PCM, CSIRO, HadCM3), with statistical downscaling in the postprocessing. Time windows: 2071-2100. RCM scenarios from the PRUDENCE project (in particular HIRHAM4) arbitrary scenarios Scenarios provided directly by the Hadley Centre or the IPCC-DDC
The majority of studies carried out in Europe considered scenarios simulated with GCMs or RCMs as a source of information. The use of analogue or arbitrary scenarios appears to be restricted to a few countries, although this may reflect a bias in the coverage of the studies considered for the survey (e.g. only the most recent publications). A common source of scenarios was the database maintained by IPCC Data Distribution Centre (IPCC-DDC)3. The
http://www.ipcc-data.org/, accessed 20/01/2008
240
purpose of the IPCC-DDC is to make available to the impacts community a set of recent GCM outputs that both reflect the state-of-the-art of model experiments and provide a representative range of results from different GCMs. For this reason criteria were defined to identify a small number of GCM experiments whose results could be deposited at the IPCC DDC. The full list of scenarios can be found on the web page of the IPCC-DDC. We include here some information concerning the Scenarios prepared for the SAR and TAR, as these are those accessed in the past. All of the SAR scenarios are based on the IS92a emission scenario, and were obtained from simulations by the following centres (name of the GCMs used in brackets): UK Hadley Centre for Climate Prediction and Research (HadCM2); German Climate Research Centre (ECHAM4); Canadian Centre for Climate Modelling and Analysis (CGCM1); US Geophysical Fluid Dynamics Laboratory (GFDL-R15); Australian Commonwealth Scientific and Industrial Research Organisation (CSIRO-Mk2); National Centre for Atmospheric Research (NCAR-DOE); Japanese Centre for Climate System Research (CCSR). More scenarios were adopted in preparing the TAR. Denoting them through their acronyms, the centre providing the data, the model used for the simulations and the SRES emission scenarios considered, these were: MPIfM by Max-Planck Institute for Meteorology (Germany) based on ECHAM4/OPYC3 simulations for the A2 and B2 emissions scenarios; HCCPR by the Hadley Centre for Climate Prediction and Research (UK) based on the HadCM3 simulations for the A1FI, A2(b and c), B1 and B2(b) emission scenarios; CSIRO by the Australian Commonwealth Scientific and Industrial Research Organization (Australia) based on CSIRO-Mk2b simulations for the A1, A2, B1 and B2 emission scenarios; NCAR by the National Center for Atmospheric Research (USA) based on NCAR-CSM, respectively NCAR-PCM simulations for the A2, respectively A1B, A2 and B2 emission scenarios; GFDL by the Geophysical Fluid Dynamics Laboratory (USA) base on R30 simulations for the A2 and B2 emission scenarions; CCCma by the Canadian Center for Climate Modeling and Analysis (Canada) based on CGCM2 simulations for the A2(b and c) and B2(b and c) emission scenarios; and, CCSR/NIES by the Center for Climate System Research, National Institute for Environmental Studies (Japan) based on CCSR/NIES AGCM and CCSR OGCM simulations for the A1, A1FI, A1T, A2, B1 and B2
241
emission scenarios. Data are available for four different time windows: 19611990, 2010-2039, 2040-2069, and 2070-2099 (the latter three periods are referred to as 2020s, 2050s, and 2080s). Another source of GCM scenarios considered for agrometeorological impact studies has been the Coupled Model Intercomparison Project (CMIP)4. CMIP is the analogue of the Atmospheric Model Intercomparison Project (AMIP) for global coupled ocean-atmosphere general circulation models. CMIP began in 1995. The project is one of the outcomes of CLIVAR (Climate Variability and Predictability, the main focus if the World Climate Research Programme of the WMO), and was initiated by the Joint Scientific Committee for the World Climate Research Programme5. The relevant scenarios were generated during Phase II of CMIP, which started in 1997. As seen in Table 1 some investigations relied on scenarios obtained from RCM simulations, in particular results of the RCM experiments carried out in the framework of PRUDENCE (Christensen et al., 2007). As stated by Christensen et al. (2007), the main objective of the PRUDENCE project was to provide high resolution climate change scenarios for Europe at the end of the twenty-first century by means of dynamical downscaling (regional climate modelling) of global climate simulations. Among the specific objectives were the evaluation and intercomparison of the models in representing 1961-1990 observed climate in Europe, the intercomparison of climate projections for 2071-2100, the characterization of the uncertainties in projections attributable to model formulation and natural/internal climate variability, and the assessment of changes in the occurrence and incidence of extreme events (Christensen et al., 2007). A total of 11 RCMs were used to simulate climate scenarios at a spatial resolution of roughly 50 km x 50 km for the time windows 1961-1990 and 2071-2100 (Christensen and Christensen, 2007). The models were: HIRHAM, CHRM, CLM, HadRM3H, HadRM3P, RegCM, RACMO, REMO, RCAO, PROMES and a stretched version of ARPEGE. More than 30 experiments were conducted with respect to the A2 and B2 SRES greenhouse gases emission scenarios. Further details concerning the experimental setup are given in Christensen and Christensen (2007), and reviews of the results of the projects can be found in a series of papers published in a special issue of
4 5
http://www-pcmdi.llnl.gov/projects/cmip/index.php, accessed 20/01/2008 http://wcrp.wmo.int/, accessed 20/01/2008
242
Climatic Change, (Vol. 81, Supplement 1, May 2007). All of the results are distributed through the project home page6. Two other European projects, STARDEX7 and MICE8 were run in a cluster with PRUDENCE. STARDEX (Statistical and Regional dynamical Downscaling of Extremes for European regions) was concerned with improving statistical downscaling methods for constructing scenarios of changes in the frequency and intensity of extreme events, while MICE (Modelling the Impact of Climate Extremes) made use of information from both dynamically and statistically downscaled methods to explore the potential impacts of extreme events in Europe. Scenarios from both STARDEX as well as MICE have found their way into agrometeorological impact studies, though not at the same rate as scenarios from PRUDENCE. Finally it is to be noted that the use of analogue or arbitrary scenarios was quite common in the 1990s, but the situation has changed in recent years due to the availability of other types of scenarios. There is still a justification in using analogue of arbitrary scenarios when the focus of the investigation is on the sensitivity of a target system or agrometeorological indices with respect to shifts in the climatic boundary conditions. Arbitrary scenarios may also be helpful to put the results of more detailed analyses into a broader perspective (see e.g. Jasper et al., 2004).
6.2.2
Tools used in the preparation of climate change scenarios for impact studies
At least in one case, the derivation of regional climate scenarios was implemented with tools such as MAGICC9 (Model for the Assessment of Greenhouse-gas Induced Climate Change, see below) and SCENGEN10 (SCENario GENerator). Thus in Alexandrov (2006) climate change scenarios for the Balkan Peninsula - for the 2025s (2011-2040), 2050s (2036-2065) and 2100s (2086-2115) were inferred from databases prepared with the help of MAGICC and SCENGEN. In view of their infrequent use, it is in order to briefly describe their main features.
6 7
http://prudence.dmi.dk/, accessed 20/01/2008 http://www.cru.uea.ac.uk/projects/stardex/, accessed 20/01/2008 8 http://www.cru.uea.ac.uk/projects/mice/, accessed 20/01/2008 9 http://www.cgd.ucar.edu/cas/wigley/magicc/index.html, accessed 24/01/2008 10 http://www.cgd.ucar.edu/cas/wigley/magicc/index.html, accessed 24/01/2008
243
According to the information provided on the home page, MAGICC and SCENGEN are coupled, user-friendly interactive software suites that allow users to investigate future climate change and its uncertainties at both the global-mean and regional levels. MAGICC carries through calculations at the global-mean level using the same upwelling-diffusion climate model that has been and is employed by the IPCC. The latest version gives the same globalmean warming and sea-level rise results as published in the IPCC Third Assessment Report (TAR). SCENGEN uses these results, together with results from a set of coupled Atmosphere/Ocean General Circulation Models (AOGCMs) and a detailed baseline climatology, to produce spatially-detailed information regarding future changes in temperature and precipitation, changes in their variability, and a range of other statistics. Concerning MAGICC it is further stated that MAGICC consists of a suite of coupled gas-cycle, climate and ice-melt models integrated into a single software package. This software allows the user to determine changes in greenhouse-gas concentrations, global-mean surface air temperature and sealevel resulting from anthropogenic emissions of carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), reactive gases (CO, NOx, VOCs), the halocarbons (e.g. HCFCs, HFCs, PFCs) and sulfur dioxide (SO2). As for SCENGEN it is noted that scenarios for the world [are constructed] by exploiting the results from MAGICC and a set of AOGCM experiments, and combining these with observed global and regional climate data sets. SCENGEN contains a set of greenhouse gas-induced patterns of regional climate change obtained from different AOGCM experiments and also sulfate aerosol-induced patterns of regional climate change obtained from a series of sulfate aerosol experiments performed with the University of Illinois at Urbana-Champaign GCM. Since the GCM experiments report results on different spatial grids, all GCM data have been interpolated onto a common 5 latitude/longitude grid. A geographically-explicit climate change scenario is constructed by selecting a future time interval, a month or season, a variable (temperature or precipitation), and one or more of the AOGCMs in SCENGEN's library of model results. Pattern-scaling methods are employed to create the climate change fields at 5 resolution which can then be added to an observed 1961-90 baseline climate data set to obtain actual climate scenario values for the future time period in question. 6.2.3 Dealing with uncertainties in climate change projections Despite advances in our understanding of the processes governing the climate system, climate change scenarios remains highly uncertain. Uncertainties step in at several stages of development, including among others the specification of future emissions of greenhouse gases, the capability to model the climate
244
sensitivity, the carbon cycle, the ocean mixing and the aerosol forcing. As noted by Stainforth et al. (2005), the range of possibilities for future climate evolution needs to be taken into account when planning climate change mitigation and adaptation strategies. The assessment and representation of the uncertainties in global climate change projections have been systematically addressed in preparing the AR4 (Meehl et al., 2007; Christensen et al., 2007) and in the context of the PRUDENCE project (Dqu et al., 2007). Concerning the latter project, four sources of uncertainty were considered: sampling uncertainty in relation to the finite number of years simulated; model uncertainty due to the fact that the models use different techniques to discretize the equations and to represent sub-grid effects; radiative uncertainty introduced by the fact that only but one (two) of the IPCC SRES emission scenarios was taken into account; and, boundary uncertainty as a result of the fact that the regional models have been run under the constraint of the same global model. Dqu et al. (2007) pointed out that the role of boundary forcing was generally greater than the role of the model formulation, in particular for temperature, but estimated that the signal from the PRUDENCE ensemble was nevertheless significant. A second point of concern when dealing with projections is the misrepresentation of the current climate by GCMs and RCMs. This is true for all of the relevant atmospheric state variables, but while in some cases tracing and solving the problem is straightforward, as for the temperature, in other cases, as for precipitation or for extreme events, the task is less obvious. It is also evident that the representation of atmospheric fields in mountain regions remains a challenge (Frei et al., 2003). Systematic errors in representing the current climate are not limited to the mean conditions, but also appear in relation to the seasonal and inter-annual variability as well as in relation to the extreme events. A systematic study of the differences among RCMs in representing climate variability has been conducted by Vidale et al. (2007) in the context of the PRUDENCE project. The result of their analysis was that the RCMs used in PRUDENCE were able to represent some of the observed west-east gradient in climate variability observed across the European continent. However, using the climatology prepared by the Climate Research Unit of the University of East Anglia as a reference (New et al., 2000), Vidale et al. (2007) noticed that the majority of the RCMs exhibited a clear tendency to overestimate the inter-annual variability of the extra-tropical summer climate. An intercomparison of precipitation extremes as simulated by six different European regional climate models was undertaken by Frei et al. (2006). All RCMs had comparable model settings and were driven with boundary data from the same global climate model. An evaluation of the simulations for
245
present climate in the region of the European Alps indicated that RCMs are capable of representing spatial patterns in precipitation extremes not resolved by GCMs. However, Frei et al. (2006) found that model biases were large in some cases, in particular concerning the summer season. In the context of model uncertainties it is also worth pointing out that the performance of GCMs and RCMs for other state variables of relevance in agrometeorology, such as solar radiation or the vapour pressure or the wind field, have received considerably less attention than in the case of temperature and precipitation. Currently it is therefore difficult to provide a measure of the reliability of climate projections in relation to these variables. Finally, one of the evident results of the survey was that many impact studies suffer from the mismatch between the spatial and temporal scales of the scenarios and the scales implied by the application. This is true irrespective of whether the climate scenarios were issued from GCM or RCM simulations. Facing this situation, further downscaling of GCM and RCM scenarios has been a constant necessity in agrometeorological studies during the past 10 years. 6.2.4 Mapping of climate change scenarios to the small scale: downscaling Scenarios with a spatial resolution of roughly 50 km x 50 km, for instance those issued by PRUDENCE, are clearly inadequate to address the possible effects of climate change on scales of the order of 1 to 10 km. In countries where mountains are a dominant component of the landscape, even a resolution of 10 km is insufficient to adequately address the topographic control of the regional climate. For this reason, so-called downscaling techniques have been developed and applied to obtain scenarios at the desired resolution. Techniques used in agrometeorological studies conducted in European countries during the past 10 years are summarized in Table 6.2. According to the survey, four options have been considered across the studies reported: (i) combination of climate anomalies from GCM or RCM simulations with historical observations; (ii) statistical downscaling (including pattern scaling techniques); (iii) generation of weather patterns from basic climate data by means of statistical techniques stochastic weather generator; and, (iv) dynamical downscaling by atmospheric models. In one case, use of a GIS was also mentioned (Zaliwski and Grski 1998, Zaliwski et al 1999).
246
With respect to dynamical downscaling, the choice of the model has been dictated by the specific needs. Thus, the application of RCMs driven with boundary conditions from GCM simulations has been considered to obtain scenarios at scales between 20 and 50 km. In other studies, non-hydrostatic, mesoscale models (such as RAMS11 or MM512) have been applied to obtain scenarios with a spatial resolution of the order of 1 to 10 km. One study made use of a diagnostic, mass preserving atmospheric model (CALMET13) to generate scenarios with a spatial resolution of less than 1 km. Concerning statistical downscaling, the survey indicates a variety of approaches, relying on different statistical methods canonical correlation analysis (CCA), singular value decomposition (SVD), multiple linear regression (MLR), or pattern scaling techniques and different combinations of predictors and predictands. As pointed out by Huth (2002), it is imperative that the different methods are evaluated.
Table 6.2: Summary of the techniques used in the past for downscaling Country Austria Downscaling techniques used in the past Nested downscaling. Dynamical downscaling of GCM to 30 km and 10 km resolution using the non-hydrostatic model MM5 (PSU/NCAR), further downscaling to 200 m resolution using the diagnostic model CALMET Mainly GCM output combined with historical observations. Dynamical downscaling with RCMs. GCM output combined with historical data Use of a stochastic weather generator Dynamical downscaling with a RCM (ALADIN) Statistical downscaling based on CCA, SVD or MLR Statistical downscaling based on circulation patterns recognition. GCM output combined with historical data Use of a stochastic weather generator Dynamical downscaling with RCMs. Dynamical downscaling using ARPEGE. GCM output combined with historical data Use of a stochastic weather generator Dynamical downscaling with RCMs (REMO and CLM) Statistical downscaling using WETTREG
Bulgaria Croatia Czech Republic
Finland
France Germany
11 12
http://rams.atmos.colostate.edu/, accessed 24/01/2008 http://www.mmm.ucar.edu/mm5/, accessed 24/01/2008 13 http://www.src.com/calpuff/calpuff1.htm, accessed 24/01/2007
247
Greece Hungary
Italy Norway
Poland Romania Slovenia Spain
Switzerland United Kingdom
Statistical downscaling based on circulation patterns recognition. Dynamical downscaling using RCMs (COSMO-LM, BOLAM, RAMS, POSEIDON) Statistical downscaling from diurnal pressure patterns (objective macro-synoptic types and conditional autocorrelation) regression between local vs. hemispherical observations physical downscaling (since 2005, first projections with 25 km resolution in April, 2008): ALADIN, REMO, PRECIS, RegClim LARS WG and other (built-into impact models) weather generators geoagraphycal and historical analogy to obtain WG imput parameters Statistical downscaling by CCA Nested downscaling using a RCM (HIRHAM) and statistical techniques Statistical downscaling applied directly to GCM output Downscaling of GCM output exploiting GIS technology and time series analysis. Statistical downscaling of GCM output based on CCA Dynamical downscaling using RegCM-50 and RegCM-25 Statistical downscaling GCM output combined with historical data & use of stochastic weather generator Dynamical downscaling with a RCM (PROMES) Statistical downscaling Use of a stochastic weather generator Different kinds of downscaling
While a method can be adequate to approximate the temporal structure, another may be needed to represent the spatial structure. Clearly, statistical downscaling still needs to be explored systematically in terms of the range of predictands that are of potential interest for agrometeorology. Thus, while temperature and precipitation have been considered in the majority of studies, far less has been done concerning for instance air humidity or solar radiation. There are nevertheless notable exceptions, as for instance the analysis of humidity variables by Huth (2005). Additional efforts are also needed to improve the downscaling of extreme events, although advances have been achieved in the recent past (e.g. Kysely, 2002; Busuioc et al., 2008; Hundecha and Brdossy, 2008) Another tool used in agrometeorological studies for downscaling is stochastic weather generation. Weather generators examine the statistical structure of the observed weather and simulate synthetic sequences of weather data consistent with this structure. Information obtained from climate scenarios can be used to selectively modify the statistical structure and generate synthetic series consistent with the changes imposed in this way (e.g. Dubrovsky, 1998; Semenov, 2007). In doing so, care should be taken to make sure that the model adequately represents scales of variability (Dubrovsky et al., 2004).
248
One of the first studies to systematically examine the stochastic simulation of daily data is that reported by Richardson (1981). Probably at the exception of LARS-WG14 (Rackso et al., 1991), Richardsons (1981) model (WGEN) has been adopted as a basis for developing most of the follow-up products (e.g. Met&Roll15 by Dubrovsky, 1995 and 1996). A comparison of LARS-WG and WGEN can be found in Semenov et al. (1998). Weather generators are usually run at the local scale. In recent years, however, the attention has also moved toward finding way to use weather generators to create regional scenarios. The spatial interpolation of synthetic data has been examined by Semenov and Brooks (1999). Wilks (1999), on the other hand, has considered the simultaneous generation of daily data, based on an extended version of WGEN. Both approaches can give valuable support to studies where complex topography is a major constraint. 6.2.5 Scenarios for extreme events In agriculture, extreme events can have catastrophic effects on production. In this sense, it is not surprising that the question of how climate change might affect the occurrence of extreme events has been tackled in many investigations. A summary of extreme events mentioned in the survey is given in Table 6.3. Extreme precipitation events, heat waves and droughts are among the events more frequently addressed in a European context. This is also reflected in the analysis of the RCM simulations carried out in the framework of PRUDENCE (Beniston et al., 2007), and the results of that survey provide a good overview of what found in the majority of studies. In practice, the PRUDENCE scenarios suggest that by the end of the twenty first century countries in central Europe will experience the same number of hot days as they are currently experienced in southern Europe. The intensity of extreme temperatures was found to increases more rapidly than the intensity of more moderate temperatures over the continental interior, a fact that has been ascribed to increases in temperature variability. Further, it has been found that heavy winter precipitation increases in central and northern Europe but decreases in the south, while heavy summer precipitation increases in northeastern Europe and decreases in the south.
14 15
http://www.rothamsted.bbsrc.ac.uk/mas-models/larswg.php, accessed 24/01/2008 http://www.ufa.cas.cz/dub/dub.htm#met&roll, accessed 24/01/2008
249
Table 6.3: Types of extreme events addressed in past studies on climate change Country Austria Bulgaria Croatia Czech Republic Finland France Germany Greece Italy Norway Poland Romania Slovenia Spain Switzerland United Kingdom Type of extreme events considered in the past Hydrological extremes, wind storm and hail NA Wind storms and heavy precipitation Extreme temperatures and precipitation, more specifically heat waves and droughts Hot days Not specified, but reference made to projects PRUDENCE, STARDEX and ENSEMBLES Heat waves, heavy rain, wind storms and frost Droughts and dry spells, heavy rainfall and floods, hail NA (but analysis of extreme temperatures and precipitation events mentioned in the answer to Question 1) Heavy precipitation and floods, wind storm, sea storm surges Not specified, but reference made to projects PRUDENCE, STARDEX and ENSEMBLES Extreme precipitation events NA Extreme temperatures and rainfall Heat waves and droughts, extreme precipitation events NA
Risk of droughts has been suggested to increase in central Europe; for the Mediterranean area, the PRUDENCE simulations indicate that droughts start earlier in the year and last longer. A point sometimes overseen is impact studies, is that the precise definition of extreme events is central for understanding the results. For instance, little information was provided in the questionnaires. It appears, however, that in the majority of cases, studies have adhered to the definitions adopted for STARDEX16 project.
6.3 Lessons learned and outlook 6.3.1 A summary of past experiences In the past, the development and application of climate change scenarios have not been devoid of difficulties. Problems identified by the participants were of variegated nature. As already mentioned, a major difficulty in the application of climate scenarios was found in relation to the spatial representativeness of
16
http://www.cru.uea.ac.uk/projects/stardex/deis/Core_Indices.pdf, accessed 24/01/2008
250
GCM output and the lack of generic but reliable procedures for downscaling the results. This holds particularly true with respect to mountain regions, where the complexity of the topography on spatial scale often of the order of 1 km or less implies pronounced variability in the climatic fields. In many studies, statistical downscaling was the only choice available for inferring climate scenarios at the regional to local scale. Statistical downscaling was not criticized per se. Rather it was pointed out that it may be difficult to find the appropriate combination of predictors and methods and that even for the optimal choice of predictors and procedures the results can be quite uncertain. It was also mentioned that the procedure adopted to temporally downscale the output of monthly or seasonal scenarios should be chosen with care. Most of the participants emphasized, that the use of ensemble scenarios is to recommend, even though this may pose troubles in handling the data volume. In this way, a systematic analysis of uncertainties is possible at all stages of the analysis. In some countries, data needed for running the models or verifying the predictions were difficult to access, either for technical or institutional reasons A few countries complained about the lack of human and technical resources. In the past, this often implied the impossibility to conduct specific studies and the necessity to resort to open access data and supplementary information. Although nowadays storage capacity is not anymore an issue, it was noted that small or private institutions can still face problems in handling very large amounts of data. Finally, one questionnaire mentioned the fact that impact studies may suffer from more fundamental problems, related for instance to the lack of clarity in the definition of variables, understanding of the questions, documentation of datasets, and the like (Sivertsen, 2005a and 2005b) 6.3.2 Completed and ongoing projects at the European level The need for more reliable scenarios of future climatic conditions in mitigation and adaptation studies implies continuous efforts to improve assumptions and methods adopted to infer scenarios. At the European level, several initiatives have already been completed or are currently in action to provide a better access to climate scenarios for the 21st century. The six key projects reported
251
in the survey are: ACCELERATES17, PRUDENCE18, STARDEX19, ENSEMBLES20, CECILIA21, and CLAVIER22. As seen in Table 6.4, most of the signatory countries of COST 734 were or are participating in one or more of these projects. Of the three projects completed, PRUDENCE and STARDEX have already been described in Section 2.1 and are not further treated in what follows. ACCELERATES (Assessing Climate Change Effects on Land use and Ecosystems; from Regional Analysis to The European Scale) differed from these other two projects, because its main goal was to examine the relationship between agricultural land use responses to environmental change drivers and environmental protection, as reflected specifically in the management of biological resources. Specific objectives were, among others: the analysis of the impacts of future climate and socio-economic change on agroecosystems at the European and regional scales using the integrated models; and, the investigation of adaptive responses to climate change of agroecosystems using the integrated models.
The aim of the three ongoing projects, ENSEMBLES, CECILIA and CLAVIER, is clearly to bridge the gap between the climate information provided by GCMs and that needed in impact studies. Two aspects stand in the foreground: the need to better quantify uncertainties in climate projections by providing probabilistic projections; and, the need to capture the effects of topographical and associated land-use features on the local climate in regions characterized by complex terrain.
To address uncertainties in future climate at the seasonal, decadal and longer timescales, the EC FP6 project ENSEMBLES has been initiated with two goals in mind: (i) develop an ensemble prediction system based on global and regional Earth System models, validated against observations and analyses; (ii) quantify and reduce uncertainty in the representation of physical, chemical, biological and human-related feedbacks in the Earth System. Eventually, achieving these objectives will allow a more intensive use of scenarios in application studies, including agrometeorological investigations.
17 18
http://www.geo.ucl.ac.be/accelerates/, accessed 24/01/2008 http://prudence.dmi.dk/, accessed 24/01/2008 19 http://www.cru.uea.ac.uk/projects/stardex/, accessed 24/01/2008 20 http://ensembles-eu.metoffice.com/, accessed 24/01/2008 21 http://www.cecilia-eu.org/, accessed 24/01/2008 22 http://www.clavier-eu.org/clavier/, accessed 24/01/2008
252
Table 6.4: Participation of signatory countries of COST734 in the following European projects with a focus on climate change: AC(CCELERATES), PR(UDENCE), ST(ARDEX), EN(SEMBLES), CE(CILIA), and CL(AVIER) Country Austria Bulgaria Croatia Cyprus Czech Republic Denmark Finland France Germany Greece Hungary Ireland Italy Luxembourg (The) Netherlands Norway Poland Portugal Romania Serbia Slovak Republic Slovenia Spain Switzerland United Kingdom AC PR ST EN CE CL
Relying on dynamical downscaling, i.e. nesting of a RCM or a fine-scale limited area model within a GCM, ENSEMBLES will provide access to a set of transient runs covering the whole period of 1950-2050 (2100) from selected combinations of 14 RCMs and 6 GCMs. The spatial domain will be the whole of the European continent, and the spatial resolution of the scenarios will be of 25 km (as compared to 50 km in the PRUDENCE scenarios). In view of the amount of resources necessary to carry out the simulations, only the emission scenario A1B will be considered for the experiments.
253
Based on key runs from ENSEMBLES, regional climate simulations for selected areas in Central and Eastern Europe are run in the context of CECILIA. The objective of CECILIA is to assess the local impacts of the A1B emission scenario in two time slices, the mid (2021-2050) and end (20712100) of the 21st century. A similar aim applies to CLAVIER, but here the focus is clearly on a small region encompassing Hungary, Romania, and Bulgaria. CLAVIER will also differ from CECILIA in that both the A2 and B1 emissions scenarios will be adopted, either in time slices or transient run experiments. Noting that even a resolution of 25 km, as prescribed in the simulations carried out for the ENSEMBLES project, is not sufficient to properly resolve the spatial scales dominating the northern flanks of the Alps, the Pyrenees, the Carpathians, and smaller mountain chains and highlands in the Czech Republic, Slovakia, Romania and Bulgaria, a spatial resolution of 10 km is adopted in CECILIA and CLAVIER for running the RCM simulations. This will open opportunities for investigating the consequences of climate change for the occurrence of weather extremes in the regions under study. In addition, statistical downscaling methods for verification of the regional model results will be developed and applied within CECILIA, and assessments of their use in localization of model output for impact studies will be performed. 6.3.3 Role of COST 734 Besides advancing our understanding in the field of climate change, projects at the European level, such as ENSEMBLES, CECILIA and CLAVIER, have the merit of creating a network of partner institutions that can be exploited for the coordination of other activities. Asked for how they see the role of COST 734, the participants in the survey offered quite a wide spectrum of possibilities (Table 6.5), ranging from the coordination of research activities to the promotion of collaboration between scientists and stakeholders, touching such aspects as the necessity to provide common tools for the application of climate scenarios but also the possibility to overcome some of the financial and technical needs and the needs for human resources in some countries.
Table 6.5: Needs for future studies Country Austria Requirements for future studies Foster collaboration between climate modelling and agricultural research communities; make uncertainty analysis a key research issue Needs for financial, human and technical resources
Bulgaria
254
Croatia Czech Republic
Finland France Germany Greece
Hungary
Italy Norway
Poland Romania Slovenia
Spain
Foster education and promote use of scenarios among stakeholders Overcome human and technical restrictions through concentration or coordination of the various efforts; promote the use of and provide probabilistic scenarios NA Standardize the scenarios (create a database of reference scenarios) and the downscaling techniques Assess the suitability of particular climate scenarios; promote the use of probabilistic scenarios Uncertainties in the emissions scenarios; differences among GCM or RCMs; performance of statistical downscaling; observational data availability Availability of RCM model outputs (statistical parameters, at least); Adequate modelling or additional simulation of real autocorrelation in the anomalies in subsequent months/seasons, which lead e.g. to drought; comparison of the scenarios derived by different methods to assess the reliability of previously performed impact studies Foster the operational use and the quality assessment of climate scenarios across European countries Broaden the field of research, including topics such as sustainability of the agricultural production and, eventually the whole food chain. Need for general support Need for common set of scenarios and downscaling tools Need for cooperation between neighbour countries; need for common downscaling techniques; need to more carefully address the issue of the extreme events. Improve climate models with respect to, or devise techniques to better account for atmosphere/biosphere feedbacks. Need to promote the use of probabilistic scenarios; need to provide common downscaling tools. NA
6.4 Conclusions In Europe crop yields increased significantly during the second half of the 20th century. However, the pace at which productivity was raised started to decline in the 1990s, and appears to have already reached its maximum. Moreover, extreme events such as the heat wave that struck large portions of the continent during the summer of 2003 have clearly shown that the agricultural sector remains vulnerable to climate. In view of the changes in climatic conditions projected by global and regional climate models for the
255
coming decades, it is likely that adaptation to climate change, if not already, will soon become a key aspect of the agricultural policy in Europe. Defining and implementing effective measures of adaptations will require quantitative knowledge of the possible evolution of the climatic conditions at the regional scale, including projections for changes in climate variability and occurrence and severity of extreme events. Advancing our understanding of the climate system and our ability to simulate the relevant processes, thus reducing the uncertainties in the projections, is necessary to support policy makers and stakeholders in their effort to face climate change. The survey carried out among the signatory states of COST 734 had not the pretention of being exhaustive, but nevertheless provided a useful overview of the present state of knowledge in a representative sample of European countries. It helped identify problems in relation to the development and application of climate scenarios and some of the needs for the future work.
6.5 References
Alexandrov V., 2006. Climate variability and change and their influence on ecosystems in Central and Southeastern Europe as well as Southeastern USA. DrSc thesis, 189 pp. Beniston M., D.B. Stephenson, O.B. Christensen, C.A.T. Ferro, C. Frei, S. Goyette, K. Halsnaes, T. Holt, K. Jylh, B. Koffi, J. Palutikof, R. Schll, T. Semmler, K. Woth, 2007. Future extreme events in European climate: an exploration of regional climate model projections. Climatic Change 81:7195. Busuioc A., R. Tomozeiu, C. Cacciamani, 2008. Statistical downscaling model based on canonical correlation analysis for winter estreme precipitation events in the EmiliaRomagna region. Int. J. Climatol., in press. Christensen J.H. and Christensen O.B., 2007. A summary of the PRUDENCE model projections of changes in European climate by the end of this century. Climatic Change 81: 730. Christensen J.H., T.R. Carter, M. Rummukainen, G. Amanatidis, 2007. Evaluating the performance and utility of regional climate models: the PRUDENCE project. Climatic Change 81: 1-6. Christensen J.H., B. Hewitson, A. Busuioc, A. Chen, X. Gao, I. Held, R. Jones, R.K. Kolli, W.-T. Kwon, R. Laprise, V. Magaa Rueda, L. Mearns, C.G. Menndez, J. Risnen, A. Rinke, A. Sarr, P. Whetton, 2007. Regional Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Dqu M., D.P. Rowell, D. Lthi, F. Giorgi, J.H. Christensen, B. Rockel, D. Jacob, , E. Kjellstrmde, M. Castro, B. van den Hurk, 2007. An intercomparison of regional climate simulations for Europe: assessing uncertainties in model projections. Climatic Change 81: 5370 Dubrovsky M., 1995. Met&Roll: The weather generator for the crop growth model. In: Regional Workshop on Climate variability and climate change vulnerability and
256
adaptation (proceedings), Institute of Atmospheric Physics, Praha - U.S. Country Studies Program, Washington, D.C. Dubrovsky M., 1996. Met&Roll: the stochastic generator of daily weather series for the crop growth model. Meteorological Bulletin, 49, 97-105. Dubrovsky M., 1998. Estimating climate change impacts on crop yields with use of crop growth model and weather generator. In: Proc. 14th Conf. Prob. Stat., AMS. Dubrovsky M., J. Buchtele, Z. Zalud, 2004. High-Frequency and Low-Frequency Variability in Stochastic Daily Weather Generator and Its Effect on Agricultural and Hydrologic Modelling. Climatic Change 63 (No.1-2), 145-179. Frei C., J.H. Christensen, M. Dqu, D. Jacob, R.G. Jones, P.L. Vidale, 2003. Daily precipitation statistics in regional climate models: Evaluation and intercomparison for the European Alps. JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 108, NO. D3, 4124, doi:10.1029/2002JD002287 Frei C., R. Schll, S. Fukutome, J. Schmidli, P.L. Vidale, 2006. Future change of precipitation extremes in Europe: Intercomparison of scenarios from regional climate models. JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 111, D06105, doi:10.1029/2005JD005965 Houghton J.T, L.G. Meira Filho, B.A. Callender, N. Harris, A. Kattenbergand, K. Maskell, (Eds), 1995. Climate Change 1995: The Science of Climate Change. Contribution of Working Group I to the Second Assessment of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Houghton J.T., Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, K. Maskell, C.A. Johnson, (Eds.), 2001. Climate change 2001: The Scientific Basis. Contribution of the Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Hundecha Y. and Brdossy A., 2008. Statistical downscaling of extremes of daily precipitation and temperature and construction of their future scenarios. Int. J. Climatol., in press. Huth R., 2002. Statistical downscaling of daily temperature in central Europe. J. Climate, 15, 1731-1742. Huth R., 2005. Downscaling of humidity variables: A search for suitable predictors and predictands. Int. J. Climatol., 25, 243-250. Jasper K., P. Calanca, D. Gyalistras, J. Fuhrer, 2004. Differential impacts of climate change on the hydrology of two alpine river basins. Clim. Res. 26: 113129. Meehl G.A., T.F. Stocker, W.D. Collins, P. Friedlingstein, A.T. Gaye, J.M. Gregory, A. Kitoh, R. Knutti, J.M. Murphy, A. Noda, S.C.B. Raper, I.G. Watterson, A.J. Weaver, Z.C. Zhao, 2007. Global Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Nakicenovic N. and Swart R., 2000. Emissions Scenarios. IPCC Special Report, Cambridge University Press, Cambridge. New M, M. Hulme, P. Jones, 2000. Representing twentieth-century space-time climate variability. Part II: Development of 190196 monthly grids of terrestrial surface climate. J Climate 13: 22172238. Parry M., O. Canziani, J. Palutikof, P. van der Linden, C. Hanson, (Eds.), 2007. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II
257
to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Racsko P., L. Szeidl, M. Semenov, 1991. A serial approach to local stochastic weather models. Ecological Modelling 57, 27-41. Richardson C.W., 1981. Stochastic simulation of daily precipitation, temperature and solar radiation. Water Resourc. Res. 17: 182-190. Semenov M.A., 2007. Development of high-resolution UKCIP02-based climate change scenarios in the UK Agric Forest Meteorology 144:127138. Semenov M.A., R.J. Brooks, E.M. Barrow, C.W. Richardson, 1998. Comparison of the WGEN and LARS-WG stochastic weather generators for diverse climates. Climate Research 10:95-107. Semenov M.A. and Brooks R.J., 1999. Spatial interpolation of the LARS-WG stochastic weather generator in Great Britain. Climate Research, 11:137-148. Siversten T.H. 2005a. Discussing the scientific method and a documentation system of meteorological and biological parameters. Physics and Chemistry of the Earth 30: 35-43. Siversten, T.H. 2005b. Implementation of a general documentation system for web-based administration and use of historical series of meteorological and biological data. Physics and Chemistry of the Earth 30: 217222. Solomon S., D. Qin, M. Manning, M. Marquis, K. Averty, M.M.B. Tignor, H.M. Jr. LeRoy, Z. Chen, (Eds) 2007. Climate Change 2007: The Scientific Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Stainforth D.A., T. Aina, C. Christensen, M. Collins, N. Faull, D.J. Frame, J.A. Kettleborough, S. Knight, A. Martin, J.M. Murphy, C. Piani, D. Sexton, L.A. Smith, R.A. Spicer, A.J. Thorpe, M.R. Allen, 2005. Uncertainty in predictions of the climate response to rising levels of greenhouse gases. Nature 433: 403-406. Vidale P.L., D. Lthi, R. Wegmann, C. Schr, 2007. European summer climate variability in a heterogeneous multi-model ensemble. Climatic Change 81: 209-232. Wigley T.M.L. and Raper S.C.B., 2001. Interpretation of High Projections for Global-Mean Warming. Science 293: 451-454. Wilks D.S., 1999. Simultaneous stochastic simulation of daily precipitation, temperature and solar radiation a multiple sites in complex terrain. Agric. Forest Meteor. 96: 85-101. Zaliwski A. and Grski T., 1998. Numerical Maps of Temperature Sums in Poland. 2nd European Conference on Applied Climatology ECAC98, Vienna, Austria, 19-23 Oct. 1998. CD ROM, ZAMG. Zaliwski A., T. Grski, S. Lipski, R. Winiarski, E. Wrblewska, 1999. Numerical Maps of Profit Probability for Maize Production in Poland. EFITA/99 Conf. Bonn, vol. A, 217224.
6.6 Additional references

R1. References cited in the answers to the questionnaire with respect to the type of scenarios Alexandrov V., 1997. GCM Climate Change Scenarios for Bulgaria. Bulgarian Journal of Meteorology and Hydrology 4(4): 205-211. Alexandrov V., 1999. Vulnerability and Adaptation of Agronomic Systems in Bulgaria. Climate Research 12(2-3): 161-173.
258
Alexandrov V., 2000. Bioproductivity, Agriculture and Climate Change. In: Staneva, M., G. Knight, T. Hristov and D. Mishev (eds.). Global Change and Bulgaria, University Park, Pennsylvania, USA and Sofia, pp. 177-198. Alexandrov V., 2003. The Impacts of Climate Variability and Change on Agriculture in Bulgaria. Proceedings of the International Symposium on Climate Change (ISCC), Beijing, China, WMO/TD-No.1172, pp. 232-237. Alexandrov V., 2006. Climate variability and change and their influence on ecosystems in Central and Southeastern Europe as well as Southeastern USA. DrSc thesis, 189 pp. Alexandrov V. and Eitzinger J., 2001. Potential Climate Change Impact on Winter Wheat and Spring Barley in Austria. Proceedings of the international conference on 150 Years of Meteorological Service in Central Europe, Stara Lesna, Slovakia, (CD) 15 pp. Alexandrov V. and J. Eitzinger, 2005. The Potential Effect of Climate Change and Elevated Air Carbon Dioxide on Agricultural Crop Production in Central and Southeastern Europe. Journal of Crop Improvement 13(1-2): 291-331. Alexandrov V. and G. Hoogenboom, 2000. The Impact of Climate Variability and Change on Crop Yield in Bulgaria. Agricultural and Forest Meteorology, 104(4): 315-327. Alexandrov V., J. Eitzinger, V. Caijc, M. Oberfoster, 2002. Potential Climate Change Impacts on Selected Agricultural Crops in Northeast Austria. Global Change Biology 8(4): 372389. Bartholy J., R. Pongrcz, G. Gelyb 2007. Regional climate change expected in Hungary for 2071-2100. Applied Ecology and Environmental Research, 5.1-17. Bengtsson L., 2001. ECHAM4_OPYC_T42_22670_Scenario: Greenhouse Gas Experiment (GHG), available from: http://cera-www.dkrz.de. Bergant K., S. Trdan, D. nidari, Z. repinek, L. Kajfe Bogataj, 2005. Impact of climate change Developmental Dynamics of Thrips Tabaci (Thysanoptera: Thripidae): Can it be quantified? Environmental Entomology, 34(4), 755-766. Bergant K., S. Trdan, L. Kajfe Bogataj, 2006. Uncertainties in modelling of climate change impact in future: an example of onion thrips (Thrips tabaci Lindeman) in Slovenia. Ecological modelling, 194, 1-3, 244-255. Bergant K., M. Belda, T. Halenka, 2007. Systematic errors in the simulation of European climate (1961-2000) with RegCM3 driven by NCEP/NCAR reanalysis. International Journal of Climatology 27(4), 455-472. Bergant K., 2007. Projections of climate change for Slovenia. Symposium: Climate changes: impact on forest and forestry (ed. Jurc M.), 19-20 April 2007, Biotechnical Faculty, Department of Forestry and Renewable Forest Resources Slovenia, University of Ljubljana, Studia forestalia Slovenica, 130, 67-86 (only in Slovenian). Bis K., G. Demidowicz, T. Deputat, T. Grski, A. Harasim, S. Krasowicz, 1993. Ekonomiczne konsekwencje zmian klimatu ocean wstpna. (Economic consequences of climate changes in agriculture preliminary appraisal). Probl. Agrofizyki, 68, Ossolineum Wrocaw. Busuioc A., R. Tomozeiu, C. Cacciamani 2007. Statistical downscaling model based on Canonical Correlation Analysis for winter extreme precipitation events in the EmiliaRomagna region , Int. J.of Clim., in press. Busuioc A., V. Cuculeanu, P. Tuinea, A. Geicu, C. Simota, Adriana Marica, A. Alexandrescu, N. Patrascanu, V.Al. Stanescu, P. Serban, I. Tecuci, Marinela Simota, C. Corbus, 2003. Impactul Potential al Schimbarii Climei in Romania, Ed. ARS DOCENDI, Academia Romana-Comitetul National pentru Modificarile Globale ale Mediului, ISBN 973-558125-6, Bucuresti, pg 1-23.
259
Cacciamani C., R Tomozeiu, V. Pavan, S. Ribaldi, 2007. Evidenze di cambiamenti climatici dalla scala globale alla scala regionale. Possibili scenari futuri sul territorio della regione Emilia-Romagna, ARPA-Rivista. Calanca P., 2007. Climate change and drought occurrence in the Alpine region: How severe are becoming the extremes? Glob. Planet. Change, 57: 151-160. Calanca P., A. Roesch, K. Jasper, M. Wild, 2006. Global warming and the summertime evapotranspiration regime of the Alpine region. Climatic Change 79, 6578. Cuculeanu V., A. Marica, C. Simota, 2000. Climate change impacts on agricultural crops in Romania and adaptation options, In: Nobuo Mimura (eds.), January 2000, Proceedings of the Conference on National Assessment Results of Climate Change, held in San Jose, Costa Rica, March 25-28, 1998 (Printed by the Japan Environmental Agency and Overseas Environmental Cooperation Center), pp. 647-666. Cuculeanu V., A. Marica, C. Simota, 1999. Climate change impacts on agricultural crops in Romania and adaptation options, Climate Research (CR SPECIAL 6, Book Version), vol. 12, Inter-Research ISSN 0936-577X, Germany, pp. 153-160. Dqu M., C. Dreveton, A. Braun, D. Cariolle, 1994. The ARPEGE-IFS atmosphere model: a contribution to the French community climate modelling. Clim Dyn 10:249-266. Dqu M. and Piedelievre J.P., 1995. High-Resolution climate simulation over Europe. Clim Dyn 11:321-339. Dqu M., P. Marquet, R.G. Jones, 1998. Simulation of climate change over Europe using a global variable resolution general circulation model. Clim Dyn 14:173-189. Douville H., S. Planton, J.F. Royer, D.B. Stephenson, S. Tyteca, L. Kergoat, S. Lafont, R.A. Betts, 2000. The importance of vegetation feedbacks in doubled-CO2 time-slice experiments. J Geophys Res 105:14841-14861. Dubrovsky M., I. Nemesova, J. Kalvova, 2005. Uncertainties in climate change scenarios for the Czech Republic. Climate Research 29, 139-156 Dudhia J., 1993. A nonhydrostatic version of the Penn State NCAR mesoscale model: Validation and simulation of an atlantic cyclone and cold front, Mon. Wea. Rev. 121, 1493-1513. Frei C., J.H. Christensen, M. Dqu, D. Jacob, R.G. Jones, P.L. Vidale, 2003. Daily precipitation statistics in regional climate models: evaluation and intercomparison for the European Alps. J. Geophys. Res. 108: ACL 91-19 Gerstengarbe F.W., F. Badeck, F. Hattermann, V. Krysanova, W. Lahmer, P. Lasch, M. Stock, F. Suckow, F. Wechsung, P.C. Werner, 2003. Studie zur klimatischen Entwicklung im Land Brandenburg bis 2055 und deren Auswirkungen auf den Wasserhaushalt, die Forstund Landwirtschaft sowie die Ableitung erster Perspektiven. PIK-Report, Nr. 83, ISSN 1436-0179 Gibelin A.L. and Dqu M., 2003. Anthropogenic climate change over the Mediterranean region simulated by a global variable resolution model. Climate Dynamics 20:327-339 Grski T., T. Deputat, K. Grska, I. Marcinkowska, W. Spoz-Pa, 1997. Rozkady statystyczne plonw gwnych rolin uprawnych dla stanu aktualnego i dwch scenariuszy klimatycznych. Raport IUNG. (Statistical distribution of the yields main cultivated crops in Poland according to actual climatic condition and two climate change scenarios), IUNG Report. Grski T.and Ku J., 2002. The impact of climate change on agricultural production in Poland. Papers on Global Change; 9: 79-90. Jasper K., P. Calanca, D. Gyalistras, J. Fuhrer, 2004. Differential impacts of climate change on the hydrology of two alpine river basins. Clim Res 26: 113129.
260
Jasper K., P. Calanca, J. Fuhrer 2006. Changes in summertime soil water patterns in complex terrain due to climatic change. Journal of Hydrology 327: 550 563. Johns T.C., J.M. Gregory, W.J. Ingram, C.E. Johnson, A. Jones, J.A. Lowe, J.F.B. Mitchell, D.L. Roberts, D.M.H. Sexton, D.S. Stevenson, S.F.B. Tett, M.J. Woodage, 2003. Anthropogenic climate change for 1860 to 2100 simulated with the HadCM3 model under updated emission scenarios. Climate Dyn 20: 583-612. Jylh K., H. Tuomenvirta, K. Ruosteenoja, 2004. Climate change projections for Finland during the 21st century. Boreal Environment Research, 9, 127-152. Kajfe Bogataj L., Bergant K. 2005. What might the climate of Slovenia look like in this centrury? Ujma, 19, 218-223 (only in Slovenian language). Kozyra J., 2004. Climatic conditions for millet cultivation in Poland. CAgM Report No. 94, WMO, Geneva, Switzerland, 34-35. Kozyra J. and Grski T., 2004. Wpyw zmian klimatu na upraw rolin w Polsce (Impact of climate change on crop cultivation in Poland). w Klimat rodowisko Czowiek. Polski Klub Ekologiczny, Okrg Dolnolski, Wrocaw, 41-50. Loibl W., H. Formayer, W. Schner, B. Ahrens, M. Dorninger, A. Gobiet, 2004. Reclip:more Research for Climate Protection: Model Run Evaluation, 1. Jahresbericht, Austrian Research Centers systems research, Seibersdorf. Marica A., A.M. Roman, V. Cuculeanu, G. Roman, 1998. Modelling maize responses to carbon dioxide doubling and climate changes, Proceedings of the First Asian Conference for Information Technology in Agriculture (AFITA Conference), 24-26 January, Wakayama City, Japan, Ed. by The Asian Federation for Information Technology in Agriculture, pg.173-181. Marica A. and A. Busuioc, 2004. The potential impacts of climate change on the main components of water balance relating to maize crop, Romanian Journal of Meteorology, Vol.6, No.1-2, ISSN 1223-1118, Bucharest, Romania. Meehl G.A., G.J. Boer, C. Covey, M. Latif, R.J. Stouffer, 2000. The coupled model intercomparison project (CMIP). Bull Am Met Soc 81:313318. Miglietta F., M. Tanasescu, M. Adriana, 1995. The expected effects of climate change on wheat development (IATA-Wheat phenological model), Global Change Biology, Volume1(Number 6), ISSN 1354-1013, pp 407-415. Molnr K. and Mika J., 1997. Climate as a changing component of landscape: Recent evidences and projections for Hungary. Zeitschrift fr Geomorphologie 110,185-195. Pavan V., R. Tomozeiu, C. Cacciamani, M. Di Lorenzo, 2007. Daily precipitation observations over Emilia-Romagna: mean values and extremes, Submitted to Int.J.of Clim. Pope V.D., M.L. Gallani, P.R. Rowntree, R.A. Stratton 2000. The impact of new parameterizations in the Hadley Centre model: HadAM3. Climate Dyn 16: 123-146. Stephenson D.B. and Pavan V., 2003. The North Atlantic Oscillation in coupled climate models: a CMIP1 evaluation. Clim Dyn, 20, 381-399. Reynolds R.W. and Smith T.M. 1994. Improved global sea surface temperature analyses using Optimum Interpolation. J. Climate 7: 929-948. Riedo M., D. Gyalistras, J. Fuhrer, 2000. Net primary production and carbon stocks in differently managed grasslands: simulation of site-specific sensitivity to an increase in atmospheric CO2 and to climate change. Ecological Modelling 134 (2000) 207227. Riedo M., D. Gyalistras, J. Fuhrer, 2001. Pasture responses to elevated temperature and doubled CO2 concentration: assessing the spatial pattern across an alpine landscape. Clim Res 17: 1931.
261
Ricard J.L. and J.F. Royer, 1993. A statistical cloud scheme for use in an AGCM. Ann. Geophysicae 11:1095-1115. Roeckner E., G. Baeuml, L. Bonaventura, R. Brokopf, M. Esch, M. Giorgetta, S. Hagemann, I. Kirchner, L. Kornblueh, E. Manzini, A. Rhodin, U. Schlese, U. Schulzweida, A. Tompkins 2003. The Atmospheric General Circulation Model ECHAM5. Part 1: Model Description, Report 349, Max Planck Institute for Meteorology (MPI), Hamburg. Royer J.F., D. Cariolle, F. Chauvin, M. Dqu, H. Douville, S. Planton, A. Rascol, J.L. Ricard, D. Salas y Melia, F. Sevault, P. Simon, S. Somot, S. Tyteca, L. Terray, S. Valcke 2002. Simulation of climate changes during the 21-st century including stratospheric ozone. Gosciences 334 (3):147-154. Simota C., M. Adriana, V. Cuculeanu, 2003. Impactul schimbarii climei asupra ecosistemelor agricole; evaluarea vulnerabilitatii si a masurilor de adaptare, Ed. ARS DOCENDI, Academia Romana-Comitetul National pentru Modificarile Globale ale Mediului, ISBN 973-558-125-6, pg 37-99; In Impactul Potential al Schimbarii Climei in Romania, Aristita Busuioc, V. Cuculeanu, P. Tuinea, A. Geicu, C. Simota, Adriana Marica, A. Alexandrescu, N. Patrascanu, V.Al. Stanescu, P. Serban, I. Tecuci, Marinela Simota, C. Corbus, Coordonator: V. Cuculeanu, Bucuresti 2003, pg. 1-230. Simota C. and Adriana M., 1997. Assessment of the potential effects of climate change on agriculture ecosystem and evaluation adaptation measures (by using DSSATv.3.0), pp. 188 In V. Cuculeanu & colab.: National Country Study on Climate Change Impact in Romania. Element 2: Vulnerability Assessment and Adaptation Options, Final Raport, US-Country Study Program, E.P.A. (Environmental Protection Agency), Washington D.C., USA, August 1997, 345 pp. Stephenson D.B.; V. Pavan, M. Collins, M.M. Junge, R. Quadrelli, 2006. North Atlantic Oscillation response to transient greenhouse gas forcing and the impact on European winter climate: a CMIP2 multi-model assessment. Clim Dyn 27: 401420. Stock, M., (ed.) 2005. KLARA: Klimawandel Auswirkungen Risiken Anpassung. PIK Report 99, Potsdam. Tomozeiu R., V. Pavan, C. Cacciamani, M. Amici, 2006. Observed temperature changes in Emilia-Romagna: mean values and extremes. Climate Research, 31, 217-225. Tomozeiu R., C. Cacciamani, V. Pavan, A. Morgillo, A. Busuioc 2007. Climate change scenarios for surface temperature in Emilia-Romagna (Italy) obtained using statistical downscaling models. Theoretical and Applied Climatology. DOI 10.1007/s00704-0060275-z. Torriani D., P. Calanca, S. Schmid, M. Beniston, J Fuhrer, 2007. Potential effects of changes in mean climate and climate variability on the yield of winter and spring crops in Switzerland. Climate Research. In press. Torriani D., P. Calanca, M. Lips, H. Amman, M. Beniston, J. Fuhrer, 2007. Regional assessment of climate change impacts on maize productivity and associated production risk in Swtzerland. Regional Environmental Change, in press, DOI:10.1007/s10113-0070039-z. Trnka M., M. Dubrovsk, D. Semerdov, Z. alud, 2004a. Projections of uncertainties in climate change scenarios into expected winter wheat yields, Theoretical and Applied Climatology , Vol. 77, 229-249. Trnka M., M. Dubrovsk, , Z. alud, 2004b. Climate change impacts and adapation strategies in spring barley production in the Czech Republic, Climatic Change Vol. 64, 227-255. Trnka M., F. Muka Semerdov, M. Dubrovsk, E. Kocmnkov, Z. alud. 2007. . European Corn Borer Life Stage Model: Regional Estimates of Pest Development and Spatial
262
Distribution under Present and Future Climate., Ecological Modeling (will be published in July 07). alud Z. and Dubrovsk M., 2002. Modeling climate change impacts on maize growth and development. Theoretical Applied Climatology, 72: 85-102. Wilby R.L., T.M.L. Wigley, D. Conway, P.D. Jones, B.C. Hewitson, J. Main, D.S. Wilks, 1998. Statistical downscaling of general circulation model output: A comparison of methods. Water resources research 34:2995-3008. R2. References cited in the answers to the questionnaire with respect to the type of downscaling techniques Alexandrov V., 2006. Climate variability and change and their influence on ecosystems in Central and Southeastern Europe as well as Southeastern USA. DrSc thesis, 189 pp. Alexandrov V. and Hoogenboom G., 2000. The Impact of Climate Variability and Change on Crop Yield in Bulgaria. Agricultural and Forest Meteorology, 104(4): 315-327. Alexandrov V. and Eitzinger J., 2005. The Potential Effect of Climate Change and Elevated Air Carbon Dioxide on Agricultural Crop Production in Central and Southeastern Europe. Journal of Crop Improvement 13(1-2): 291-331. Bardossy A., I. Bogardi, I. Matyasovszky, 2005. Fuzzy-rule based downscaling of precipitation. Theoretical and Applied Climatology, 82, 119-129. Bartholy J., I. Bogrdi, I. Matyasovszky, 1995. Effect of climate change on regional precipitation in lake Balaton watershed. Theor. Appl. Climatol. 237-250. Benestad R.E., 2004. Tentative probabilistic temperature scenarios for northern Europe. Tellus, 56A, 89-101. Benestad R.E., 2004. Clim.pact: Empirical-Statistical Downscaling in Climate Modelling. EOS. Vol. 85, Number 42, October 19. Benestad R.E., 2005. Climate change scenarios for northern Europe from multi-model IPCC AR4 climate simulations. Geophys. Res. Lett., 32, doi:10.1029/2005GL023401, 2005 Benestad R.E., E.J. Frland, I. Hanssen-Bauer, 2006. On Models for empirical-statistical downscaling of precipitation Int. J. Clim, 10.1002/joc.1421. Bergant K., M. Belda, T. Halenka, 2007. Systematic errors in the simulation of european climate (1961-2000) with RegCM3 driven by NCEP/NCAR reanalysis. Int. J. Climatol., 27, 455-472. Busuioc A., F. Giorgi, X. Bi, M. Ionita, 2006. Comparison of regional climate model and statistical downscaling simulations of different winter precipitation change scenarios over Romania. Theor. Appl. Climatol., 86. Busuioc A., M. Ionita, F. Giorgi, X. Bi, 2004. Analysis of dynamical and statistical approaches for generating winter precipitation change scenarios over Romania. Romanian Journal of Meteorology vol.6 no 1-2. Busuioc A. and von Storch H., 2003. Conditional stochastic model for generating daily precipitation time series, Climate Research, 24. Busuioc A., D.L. Chen, C. Hellstrom, 2001. Performance of statistical downscaling models in GCM validation and regional climate change estimates: Application for Swedish precipitation, INTERNATIONAL JOURNAL OF CLIMATOLOGY 2001, Vol 21, Iss 5, pp 557-578. Busuioc A., D. Chen, C. Hellstrom, 2001. Temporal and spatial variability of precipitation in Sweden and its link with large-scale circulation, Tellus 53A, no 3, 348-367. Busuioc A., H. Von Storch, R. Schnur, 1999. Verification of GCM-generated regional seasonal precipitation for current climate and of statistical downscaling estimates under changing climate conditions, JOURNAL OF CLIMATE 1999, Vol 12, Iss 1, pp 258-272.
263
Busuioc A. and Von Storch H., 1996. Changes in the winter precipitation in Romania and its relation to the large-scale circulation, TELLUS SERIES A-DYNAMIC METEOROLOGY AND OCEANOGRAPHY 1996, Vol 48, Iss 4, pp 538-552. Dobi-Wantuch I., J. Mika, L. Szeidl, 2000. Modelling wet and dry spells with mixture distributions. Meteorology and Atmos. Phys. Vol. 73, 245-256. Dubrovsky M., J. Buchtele, Z. Zalud, 2004. High-Frequency and Low-Frequency Variability in Stochastic Daily Weather Generator and Its Effect on Agricultural and Hydrologic Modelling. Climatic Change 63 (No.1-2), 145-179. Dudhia J., 1993. A nonhydrostatic version of the Penn State NCAR mesoscale model: Validation and simulation of an atlantic cyclone and cold front, Mon. Wea. Rev. 121, 1493-1513. Engen-Skaugen T, 2007. Refinement of dynamically downscaled precipitation and temperature scenarios. (Accepted, Climatic Change) Engen-Skaugen T., J.E. Haugen, O.E.Tveito. Temperature scenarios for Norway: from regional to local scale (Accepted, Climate Dynamics) Enke W., F. Th. Schneider, 2005. Deutschlnder, A novel scheme to derive optimized circulation pattern classifications for downscaling and forecast purposes. In: Theor. Appl. Climatol. 82 (2005), S. 5163. Gaertner M.A., O.B. Christensen, J.A. Prego, J. Polcher, C. Gallardo, M. y Castro, 2001. The impact of deforestation on the hydrological cycle in the western Mediterranean: an ensemble study with two regional climate models. Climate Dynamics, 17, 857-873. Gallardo C., A. Arribas, J.A. Prego, M.A. Gaertner, M. y Castro 2001. Multi-year simulations with a high resolution regional climate model over the Iberian Peninsula: Current climate and 2xCO2 scenario, Quart J Roy Meteor Soc., 127, 1659-1682. Gerstengarbe F.W. and Werner P.C., 2005. Simulationsergebnisse des regionalen Klimamodells STAR. In: Wechsung, F., Becker, A., Grfe, P., GLOWA-ELBE I Integrierte Analyse der Auswirkungen des globalen Wandels auf Wasser, Umwelt und Gesellschaft im Elbegebiet. PIK-Report, No. 95, ISSN 1436-0179, 100-106. Gobiet A., H. Truhetz, A. Riegler, 2006. A climate scenario for the Alpine region, reclip:more project year 3 - WegCenter progress report, Wegener Center, Univ. of Graz, Austria. Grski T. and Zaliwski A., 2002. Model Agroklimatu Polski (A model of Polands agroclimate). Pam. Pu., 130, 242-250. Gyalistras D., 2000. Klimaszenarien fr den Alpenraum und die Schweiz: Neuester Stand und Vergleich (Climate scenarios for the Alps and Switzerland: state of the art and comparison). In: Wanner H, Gyalistras D, Luterbacher J, Rickli R, Salvisberg E, Schmutz C (eds) Klimawandel im Schweizer Alpenraum. Hochschulverlag AG an der ETH, Zrich, p 197235. Gyalistras D., 2002. An uncertainty analysis of monthly temperature and precipitation scenarios for Switzerland. Internal report, Climatology and Meteorology Research Group, University of Bern. Gyalistras D., 2003. Development and validation of a highresolution monthly gridded temperature and precipitation data set for Switzerland (19512000). Clim Res 25(1):55 83. Gyalistras D., H. von Storch, A. Fischlin, M. Beniston, 1994. Linking GCM-simulated climatic changes to ecosystem models: case studies of statistical downscaling in the Alps. Clim Res 4:167189. Gyalistras D., C. Schr, H.C. Davies, H. Wanner, 1998. Future Alpine climate. In: Cebon P, Dahinden U, Davies HC, Imboden D, Jaeger JC (eds) Views from the Alps: regional perspectives on climate change. MIT Press, Boston, p 171223.
264
Halenka T., J. Kalvova, Z. Chladova, A. Demeterova, K. Zemankova, M. Belda, 2006. On the capability of RegCM to capture extremes in long term regional climate simulation comparison with the observations for Czech Republic, Theor. Appl. Climatol., 86, 2006, 125-145. Hanssen-Bauer I., E.J.Frland, J.E.Haugen, O.E.Tveito, 2003. Temperature and precipitation scenarios for Norway - Comparison of results from dynamical and empirical downscaling. Climate Research, 25, 15-27. Hanssen-Bauer I., C. Achberger, R.E. Benestad, D. Chen, E.J. Frland, 2005. Statistical downscaling of climate scenarios over Scandinavia, Climate Research, Vol. 29, 255-268 Haugen J.E. and Iversen T., 2007.Response in extremes of daily precipitation and wind from a downscaled multi-model ensemble of anthropogenic global climate change scenarios (Submitted, Tellus). Huszr T., J. Mika, D. Lczy, K. Molnr, . Kertsz, 1999. Climate change and soil moisture: a way of simulation. Physics and Chemistry of the Earth. (A) Vol. 24, 905-912. Huth R., 2005. Downscaling of humidity variables: A search for suitable predictors and predictands. Int. J. Climatol., 25, 243-250. Huth R., 2002. Statistical downscaling of daily temperature in central Europe. J. Climate, 15, 1731-1742 Huth R., R. Mldek, L. Metelka, P. Sedlk, Z. Huthov, S. Kliegrov, J. Kysel, L. Pokorn, M. Janouek, T. Halenka, 2003. On the integrability of limited-area numerical weather prediction model ALADIN over extended time periods. Studia geoph. geod., 47, 863-873 Huth R., J. Kysely, M. Dubrovsky, 2003. Simulation of Surface Air Temperature by GCMs, Statistical Downscaling and Weather Generator: Higher-Order Statistical Moments. Studia Geophysica et Geodaetica 47, 203-216 Jacob D. and Podzun R., 1997. Sensitivity Studies with the Regional Climate Model REMO, Meteorol. Atmos. Phys, 63, 119-129. Jacob D., K. Blow, M. Milliez 2003. Dynamische und statistische Erstellung von hochaufgelsten Klimaszenarien (1/6) als Basis fr wasserwirtschaftliche Handlungsempfehlungen im KLIWA-Projekt B 1.1.1 Klimaszenarien, Max-PlanckInstitut fr Meteorologie, Hamburg, im Auftrag des Arbeitskreis KLIWA KLIWA Final ReportJylh K., Fronzek S., Tuomenvirta H., Carter T.R. and Ruosteenoja K., 2007: Changes in frost and snow in Europe and Baltic Sea ice by the end of the 21st century. Climatic Change, (accepted). Jylh K., K. Ruosteenoja, P. Risnen, H. Tuomenvirta, 2006. Regional analysis of an extended set of climate model simulations for the Nordic CE project. European Conference on Impacts of Climate Change on Renewable Energy Sources, Reykjavik, Iceland, 5.-9.6.2006, Extended abstract, 21-24. Kotlarski S., U. Bhm, D. Jacob, K. Keuler, R. Knoche, A. Walter, 2005. Regional Climate Model Simulations as Input for Hydrological Applications: Evaluation of Uncertainties, Advances in Geophysics, 5:119125, 2005. Kovcs-Lng E., G. Krel-Dulay, M. Kertsz, G. Fekete, S. Bartha, J. Mika, I. Dobi-Wantuch, T. Rdei, K. Rajkai, I. Hahn, 2000. Changes in the composition of sand grasslands along a climatic gradient in Hungary and implications for climate change. Phytocoenologia 30, 385-407. Kysely J., R. Huth, M. Dubrovsky, 2001. Simulation of extreme temperature events using general circulation models, statistical downscaling and stochastic generator. [in Czech; Simulace extremnich teplotnich jevu globalnimi cirkulacnimi modely, statistickym downscalingem a stochastickym generatorem] Meteorological Bulletin 54, 73-82.
265
Mika J., J. Molnr, K.Tar, 2005. Effects of macro-circulation on local climatic conditions of plant development. Physics and Chemistry of the Earth vol. 30, 135-141243. Racsko P., L. Szeidl, Semenov M. 1991. A serial approach to local stochastic weather models. Ecological Modelling 57, 27-41. Sanchez E., C. Gallardo, M.A. Gaertner, A. Arribas, A. y Castro, 2004. Future climate extreme events in the Mediterranean simulated by a regional climate model: a first approach. Global and Planetary Change in press. Schmidli J., C. Schmutz, C. Frei, H. Wanner, C. Schr, 2002. Mesoscale precipitation variability in the region of the European Alps during the 20th century. Int. J. Climatol. 22: 10491074. Schmidli J., C. Frei, P.L.Vidale, 2006. Downscaling from GCM precipitation: a benchmark for dynamical and statistical downscaling methods. Int. J. Climatol. 26: 679689. Scire J.S., F.R. Robe, M.E. Fernau, R.J. Yamartino, 2000. A Users Guide for the CALMET Meteorological Model (Version 5), software manual, Earth Tech Inc., Concorde. Semenov M.A., R.J. Brooks, E.M. Barrow, C.W. Richardson, 1998. Comparison of the WGEN and LARS-WG stochastic weather generators in diverse climates. Climate Research 10, 95-107. Semenov M.A. and Brooks R.J., 1999. Spatial interpolation of the LARS-WG stochastic weather generator in Great Britain. Climate Research 11, 137-148. Skaugen T., M. Astrup, L.A. Roald, E.J. Frland, 2003. Scenarios of extreme daily precipitation for Norway under climate change. Nordic Hydrology, Vol.15, No.1, 1-13. Spekat A., W. Enke, F. Kreienkamp, 2006. Neuentwicklung von regional hoch aufgelsten Wetterlagen fr Deutschland und Bereitstellung regionaler Klimaszenarios auf der Basis von globalen Klimasimulationen mit dem Regionalisierungsmodell WETTREG auf der Basis von globalen Klimasimulationen mit ECHAM5/MPI-OM T63L31 2010 bis 2100 fr die SRES Szenarios B1, A1B und A2 Final Report Umweltbundesamt. Truhetz H., A. Gobiet, G. Kirchengast, 2007. Evaluation of a dynamic-diagnostic modeling approach to generate highly resolved wind fields in the Alpine region, Meteorol. Z., in press. van den Hurk B., M. Hirschi, G. Lenderink, E. van Meijgaard, A. van Ulden, B. Rockel, S. Hagemann, P. Graham, E. Kjellstrm, 2005. Soil control on runoff response to climate change in regional climate model simulations, Journal of Climate, Vol. 18, No. 17: 35363551. Von Storch H., 1995. Spatial Patterns: EOFs and CCA. In: Analysis of Climate Variability. Application of Statistical Techniques, H. von Storch and A. Navarra (eds.), Springer Verlag, 227-258. Wilks S.D., 1995. Statistical Methods in the Atmospheric Sciences, vol. 59, International Geophysics Series, Academic Press, 467pp. Wilby R.L and Wigley T.M.L., 1997. Downscaling general circulation model output: a review of methods and limitations. Prog. Phys. Geogr., 21, 530-548. Zaliwski A., Grski T., 1998. Numerical Maps of Temperature Sums in Poland. 2nd European Conference on Applied Climatology ECAC98, Vienna, Austria, 19-23 Oct. 1998. CD ROM, ZAMG. Zaliwski A., T. Grski, S. Lipski, R. Winiarski, E. Wrblewska, 1999. Numerical Maps of Profit Probability for Maize Production in Poland. EFITA/99 Conf. Bonn, vol. A, 217224.
266
7. Risk assessment and foreseen impacts on agriculture
7. RISK ASSESSMENT AND FORESEEN IMPACTS ON AGRICULTURE Jorgen Eivind Olesen, Miroslav Trnka, Kurt Christian Kersebaum, Pirjo Peltonen-Sainio, Arne Oddvar Skejvg, Federica Rossi, Jerzy Kozyra, Bernard Seguin, Fabio Micale Abstract The studies on anthropogenic climate change performed in the last decade over Europe indicate consistent increases in projected temperature and different patterns of precipitation with widespread increases in northern Europe and rather small decreases over southern Europe. The development in national grain yields for wheat in the period 1961 to 2006 for countries in Europe shows that yields in northern Europe are limited by cool temperatures, whereas yields in southern Europe are limited by high temperatures and low rainfall. Yields increased considerably during the period 1970 to 1990 in all countries due to improved technologies with the highest absolute increases in western and central Europe. The yield increases have levelled off considerably during the past 10-20 years. There is in recent years a tendency in many countries to lower yields and increased yield variability. A preliminary analysis shows that the yields in several European countries in recent years correlated well with the mean temperature during the main part of the growing season with observed yield reduction in warmer years. Grain yields in maize have been increasing over the period 1961-2006 in both central and southern Europe. The yields increases seem to be continuing in Belgium and Germany, even in recent years, where wheat yield increases have been levelling off. This has also resulted in a steadily increasing grain maize area in these countries. In order to gather information on perceived risks and foreseen impacts of climate change on agriculture in Europe we designed a set of qualitative and quantitative questionnaires that were distributed to leading experts in 26 countries. There were two types of questionnaires distributed to the COST 734 members and other experts: i) country based overview questionnaires and ii) climate region specific quantitative questionnaires. Europe was divided into 13 Environmental zones (EZ). In total, we had 16 complete national reports and 50 individual responses for specific EZ from 26 countries. The questionnaires provided both country and EZ specific information on the: 1) main vulnerabilities of crops and cropping systems under present climate; 2) estimates of climate change impacts on the production of nine selected crops; 3) possible adaptation options as well as 4) adaptation observed so far. In addition we also focused on the overall awareness and presence of warning and decision support systems.
267
The results show that farmers across Europe are currently adapting to climate change, in particular in terms of changing timing of cultivation and selecting other crop species and cultivars. The responses in the questionnaires show a surprisingly high proportion of negative expectations concerning the impacts of climate change on crops and crop production throughout Europe, even in the cool temperate north European countries.
7.1 Introduction Europe is one of the world's largest and most productive suppliers of food and fibre. In 2004 it accounted for 21% of global meat production and 20% of global cereal production. About 80% of this production occurred in the EU25 countries. The productivity of European agriculture is generally high, in particular in western Europe, and average cereal yields in the EU countries are more than 60% higher than the world average. The EU Common Agricultural Policy has during the last decade been reformed to reduce overproduction, reduce environmental impacts and improve rural development. This is not expected to greatly affect agricultural production in the short run (OECD, 2004). However, agricultural reforms are expected to enhance the current process of structural adjustment leading to larger and fewer farms (Marsh, 2005). The hydrological features in Europe are very diverse, and there is also a large diversity in water uses, pressures and management approaches. About 30% of abstracted fresh water in Europe is used for agricultural purposes, primarily irrigation (Flrke and Alcamo, 2005). The proportion of fresh water abstraction used for agricultural purposes is only 4% in northern EU, but as high as 44% in southern EU and projected to increase to 53% by 2030 under baseline conditions (Flrke and Alcamo, 2005). Intensive farming systems in western Europe generally have a low sensitivity to climate change, because a given change in temperature or rainfall have modest impact (Chloupek et al., 2004), and because the farmers have resources to adapt and compensate by changing management. These systems may therefore respond favourably to a modest climatic warming (Olesen and Bindi, 2002). On the other hand some of the low input farming systems currently located in marginal areas may be most severely affected by climate change (Reilly and Schimmelpfennig, 1999; Darwin and Kennedy, 2000). There is a large variation across the European continent in climatic conditions, soils, land use, infrastructure, political and economic conditions (Bouma et al., 1998). These differences are expected also to greatly influence the responsiveness to climatic change (Olesen and Bindi, 2002).
268
7.2 Observed and projected climate in Europe 7.2.1 Observed climate change in Europe Most of Europe has experienced increases in surface air temperature during 1901 to 2005, which amounts to 0.9 C in annual mean temperature over the entire continent (Kjellstrm, 2004; Alcamo et al., 2007). However, the recent period shows a trend considerably higher than the mean trend (+0.4C/decade for the period 1977-2001, Jones and Moberg, 2003). For the past 25 years, trends are higher in central, northeastern Europe and in mountainous regions, while the lowest temperature trends are found in the Mediterranean region (Klein Tank, 2004). Temperatures are increasing more in winter than summer (EEA, 2004; Jones and Moberg, 2003). An increase of temperature variability has been observed, primarily due to increase in warm extremes (Klein Tank and Knnen, 2003). There are indications of changes in the rainfall pattern as indicated by the frequency of drought events during spring and early summer. There has been an increase in frequency of droughts in large parts of western and eastern Europe, with particularly large increases in the Mediterranean region (Trenberth et al., 2007). Mean annual precipitation is increasing in most of Atlantic and northern Europe and decreasing along the Mediterranean (Klein Tank et al., 2002). An increase in mean precipitation per wet day has been observed in most parts of the continent, even in areas getting drier (Frich et al., 2002; Klein Tank et al., 2002). Severe flooding affected parts of Austria, the Czech Republic and Germany for three weeks during August 2002. Heavy rainfall from storms crossing central Europe during early August triggered sequential flood waves that moved down the Vltava, Labe and Elbe rivers in the Czech Republic and Germany, and down the Danube river in Austria, Slovakia, Hungary, Croatia, Serbia and Romania. Recent results using high-resolution regional climate models have shown that global warming may be linked with a shift towards heavier intensive summertime precipitation over large parts of Europe (Christensen and Christensen, 2003). The precipitation events over central Europe may therefore occur more frequently in the future (Pal et al., 2004). The severity of the floods was probably enhanced by human management of the river systems, e.g. diking and installation of reservoirs (Helms et al., 2002) and possibly by the agricultural land use in the river basins (van der Ploeg and Schweigert, 2001). Even though most model studies indicate a long-term tendency towards lower soil moisture, particularly at the end of the growing season (e.g. Seneviratne et al., 2006 or Beniston et al., 2007), the results of studies based on the data collected over the last two centuries are not conclusive for western and Central
269
Europe to be able to predict higher drought frequency or severity. However, several studies based on grided datasets (e.g. van der Schrier et al. 2006 or Dai et al., 1998 and 2004), that describe drought in terms of soil moisture anomaly, showed that in many regions of Europe did exhibit a severe decline (though in some cases not statistically significant) in the available soil water over the 20th century, both during summer and over the entire year. Other studies that were based on homogenized station series (e.g. Trnka et al., 2008 or Szinel et al., 1998) indicated that the number of stations with statistically significant trends towards drier conditions (in terms of available soil moisture) prevail in Central Europe over those where either no trend at all or a tendency toward wetter conditions was noted. These shifts in intensity and frequency of drought in the region were shown to be driven by changes in near surfaces temperatures rather than changes in precipitation (e.g. van der Schrier et al., 2007) and associated with changes in ciruclation patterns. A severe heat wave over large parts of Europe in 2003 extended from June to mid-August, raising summer temperatures by 3 to 5 C. The warm anomalies in June lasted throughout the entire month (increases in monthly mean up to 67C), but July was only slightly warmer than on average (1-3C), and the highest anomalies were reached between 1 and 13 August (+7 C) (Fink et al., 2004). Maximum temperatures of 35 to 40 C were repeatedly recorded in most southern and central European countries (Andr et al., 2004; Beniston and Diaz, 2004). This heat wave has been found to be extremely unlikely statistically under current climate (Schr and Jendritzky, 2004). However, it is consistent with a combined increase in mean temperature and temperature variability (Schr et al., 2004; Meehl and Tebaldi, 2004; Pal et al., 2004). As such the 2003 heat wave resembles simulations by regional climate models of summer temperatures in the latter part of the 21st century under the A2 scenario (Beniston, 2004; Beniston and Diaz, 2004). The heat wave was associated with annual precipitation deficits up to 300 mm, and this drought was a major contributor to the estimated reduction of 30% over Europe in gross primary production of terrestrial ecosystems (Ciais et al., 2005). 7.2.2 Projections of climate change in Europe Most of the recent global climate model (GCM) experiment results are based on coupled ocean-atmosphere models (AO-GCM). The main modelling uncertainties stem from the contrasting behaviour of different climate models in their simulation of global and regional climate change. These uncertainties are largely a function of the relatively coarse resolution of the models and the different schemes employed to represent important processes in the atmosphere, biosphere and ocean. There has recently been an increased effort
270
in downscaling the coarse GCM results using regional climate models with spatial resolutions of 50 km or less (Christensen and Christensen, 2007; Christensen et al. 2007). This has led to improved quality in projections of regional climate changes in Europe. The projections show marked seasonal and regional differences in the projected changes (Fig. 7.1). The warming is greatest over eastern Europe during winter and over western and southern Europe in June-July-August (Giorgi et al., 2004). A very large increase, in summer temperatures, is projected in the south-western parts of Europe (exceeds 6 C in parts of France and the Iberian Peninsula) by the end of the 21st century under the A2 scenario (Fig. 7.1). a
Figure 7.1: Simulated changes in mean air temperature (C) for the period 2071-2100 relative to 1961-1990 for the A2 scenario using the HIRHAM regional climate model for winter, DJF (a) and summer, JJA (b) (Christensen and Christensen, 2007)
271
Generally for all scenarios, the mean annual precipitation increases in northern Europe and decreases farther south (Fig. 7.2). But the change in precipitation varies substantially from season to season and across regions. There is a projected increase in winter precipitation in northern and central Europe, whereas there is a substantial decrease in summer precipitation in southern and central Europe, and to a lesser extent in northern Europe. a
Figure 7.2: Simulated changes in mean air precipitation (%) for the period 2071-2100 relative to 1961-1990 for the A2 scenario using the HIRHAM regional climate model for winter, DJF (a) and summer, JJA (b) (Christensen and Christensen, 2007)
272
It is very likely that the frequency of drought spells and their severity will increase at least in some regions of Europe (the south and centre in particular), and recent projections of climate change impacts support this hypothesis (e.g. Hayes et al. 2005; Calanca, 2007).Recent results also indicate that variability in temperature and rainfall may increase considerably over large parts of central Europe (Christensen and Christensen, 2003; Schr et al., 2004). Indeed heat waves and droughts similar to the 2003 situation may become the norm in central and southern Europe by the end of the 21st century (Beniston and Diaz, 2004). This heat wave led to substantial reductions in primary productivity of terrestrial ecosystems and large and widespread reductions in farm income (Fink et al., 2004; Ciais et al., 2005). 7.3 Current European cropping patterns The highest yields of both cereal and tuber crops are obtained in West Europe and the lowest yields in South and East Europe (Table 7.1 and 7.2). The highest variation in yields within the regions used in Table 1 was obtained in North Europe, where yields were considerably higher in UK, Ireland and Denmark than in the Baltic countries. By far the largest cropping areas are found in eastern Europe, in particular in the Russian Federation and Ukraine. The cropping areas of eastern Europe are larger than the total of all other regions for wheat, barley, potato and sugar beet. The development in national grain yields for wheat in the period 1961 to 2006 is shown in Fig. 7.3 for selected countries in northern, central and southern Europe. Yields in northern Europe are limited by cool temperatures, whereas yields in southern Europe are limited by high temperatures and low rainfall. Yields increased considerably during the period 1970 to 1990 due to improved technologies in all countries with the highest absolute increases in western and central Europe. The yield increases have levelled off considerably during the past 10-20 years. There seems still to be a small yield increase, if any, during the past 10-20 years in Finland, which is negligible compared to the available genetic gains in yield potential. Yields in Greece have been declining. Both effects may be climate related with increasing temperatures being beneficial in Finland, but negative in Greece. The wheat yields in Germany and Greece seem to indicate an increased yield variability, which mostly likely is also related to climate. There are also clear indications that increasing temperatures are causing grain yield reductions globally (Lobell and Field, 2007).
273
Table 7.1: Mean national yields (Mg ha-1) of selected cereal, tuber and root crops in Europe as average for the period 2001-2006 (FAOSTAT) Region North Europe Country Wheat Barley Rapeseed Maize Potato Sugarbeet Denmark 7.1 5.2 3.2 39 57 Estonia 2.4 2.1 1.6 14 Finland 3.5 3.3 1.3 24 36 Ireland 8.8 6.5 3.3 36 48 Latvia 3.0 2.0 1.7 13 37 Lithuania 3.4 2.5 1.6 12 37 Norway 4.5 3.6 1.7 26 Sweden 5.9 4.1 2.4 29 49 United Kingdom 7.8 5.7 3.1 42 55 Mean 5.2 3.9 2.2 26 46 Austria 5.1 4.6 2.7 9.3 31 64 Belgium 8.4 7.4 3.8 11.2 43 68 France 6.9 6.2 3.2 8.4 41 76 Germany 7.4 5.9 3.5 8.7 40 58 Luxembourg 6.1 5.3 3.5 8.0 31 Netherlands 8.4 6.0 3.8 11.6 43 61 Switzerland 5.7 6.1 3.2 8.5 37 72 Mean 6.8 5.9 3.4 9.4 38 66 Belarus 2.8 2.7 1.0 3.6 16 31 Bulgaria 3.1 2.9 1.5 4.1 15 21 Czech Republic 4.8 4.0 2.7 6.8 23 49 Hungary 4.0 3.4 2.1 6.1 24 48 Moldova 2.2 1.8 1.2 2.8 9 24 Poland 3.7 3.1 2.5 5.5 18 42 Romania 2.7 2.5 1.3 3.5 15 26 Russian Federation 1.9 1.9 1.1 3.2 12 26 Slovakia 4.0 3.5 2.1 5.5 15 45 Ukraine 2.6 2.3 1.1 3.7 12 22 Mean 3.2 2.8 1.7 4.5 16 33 Albania 3.0 2.8 0.0 4.4 16 80 Bosnia + Herzegovina 2.9 2.6 2.3 4.4 10 25 Croatia 4.0 3.2 2.0 5.7 11 42 Greece 2.3 2.4 2.0 10.0 21 58 Italy 3.3 3.6 1.5 9.0 24 57 Macedonia (FYROM) 2.8 2.5 2.1 4.2 14 32 Portugal 1.3 1.4 5.5 15 69 Serbia and Montenegro 3.4 2.8 1.8 4.6 11 41 Slovenia 4.4 3.6 2.3 6.9 21 44 Spain 2.7 2.5 1.6 9.7 27 68 Mean 3.1 2.9 1.6 6.4 17 52
West Europe
East Europe
South Europe
274
Table 7.2: Mean national area (1000 ha) of selected cereal, tuber and root crops in Europe as average for the period 2001-2006 (FAOSTAT) Region North Europe Country Wheat Barley Rapeseed Maize Potato Sugar-beet Denmark 651 727 105 39 50 Estonia 74 135 44 16 Finland 190 548 78 29 29 Ireland 95 176 3 13 31 Latvia 177 135 44 50 14 Lithuania 349 339 90 84 24 Norway 76 164 8 14 Sweden 379 378 71 31 50 United Kingdom 1860 1042 525 149 157 Sum 3851 3644 968 0 425 355 Austria 285 203 45 174 22 44 Belgium 200 44 6 51 64 90 France 5106 1672 1161 1736 159 402 Germany 3046 2018 1293 427 283 428 Luxembourg 12 10 4 0 1 Netherlands 134 54 1 23 161 99 Switzerland 90 39 16 21 13 18 Sum 8873 4040 2526 2432 703 1081 Belarus 364 662 102 20 519 77 Bulgaria 1113 288 12 351 35 1 Czech Republic 814 509 288 83 40 72 Hungary 1135 337 113 1204 30 58 Moldova 351 107 3 495 39 43 Poland 2342 1081 503 326 777 292 Romania 2131 434 64 2862 277 34 Russian Federation 22995 9384 223 731 3125 800 Slovakia 375 210 100 146 23 32 Ukraine 5765 4297 110 1643 1553 731 Sum 37385 17309 1518 7860 6418 2140 Albania 88 1 0 49 10 1 Bosnia + Herzegovina 85 22 1 197 43 0 Croatia 203 52 14 375 46 27 Greece 814 99 2 226 44 41 Italy 2229 324 9 1134 74 206 Macedonia (FYROM) 104 48 1 33 13 2 Portugal 168 22 131 57 7 Serbia and Montenegro 640 118 3 1210 98 57 Slovenia 34 14 1 44 7 5 Spain 2190 3138 7 451 102 102 Sum 6558 3837 38 3850 495 448
West Europe
East Europe
South Europe
275
8 Grain yield (Mg ha )

-1
Norway
Finland

-1
UK
Germany

-1
Spain
Greece
1960
1970
1980 Year
1990
2000
1960
1970
1980 Year
1990
2000
Figure 7.3: National grain yield of wheat in northern, central and southern European countries for the period 1961 to 2006 (FAOSTAT database)
276
Grain yields in maize have been increasing over the period 1961-2006 in both central and southern Europe (Fig. 7.4). The yield increases seem to be continuing in Belgium and Germany, even in recent years, where wheat yield increases have been levelling off. This has also resulted in a steadily increasing grain maize area in these countries. The yield of grain maize in France and Italy has not increased in recent years. This is most likely due to warmer climate and a higher frequency of droughts, which reduces the water available for irrigation, and since maize is predominantly an irrigated crop in these countries, this has impact on both maize yields and the area cropped with maize.
Belgium 30 Maize area (%) Germany Area Yield
12 10 8 6 Grain yield (Mg ha ) Grain yield (Mg ha )

-1 -1
20
10 4 2 12 10 8 6 10 4 2 1970 1980 1990 2000 1970 1980 1990 2000 Year Year
France 30 Maize area (%)
Italy
20
Figure 7.4: National grain yield of wheat in northern, central and southern European countries for the period 1961 to 2006 (FAOSTAT database)
277
7.4 Questionnaire A questionnaire on aspects of climate vulnerability, and impacts and adaptation of crops and cropping systems to climate change was submitted to members of the COST 734 action. The questionnaire asked for national based overviews of the current knowledge base within the following areas: 1. Main vulnerabilities of crops and cropping systems 2. Critical thresholds of climatic suitability and climate change 3. Assessments and studies of climate change impacts 4. Adaptation options 5. Observed adaptation 6. National impact assessments, adaptation strategies and awareness 7. Dissemination of information and recommendations 8. Warning systems These responses showed a wide range of communalities among responses from different countries. However, for some of the larger countries there were also considerations in the responses that impacts and adaptation would vary among climatic regions within countries. A more quantitative questionnaire was therefore developed and submitted to members of COST action 734 and a range of selected experts within agronomy and agrometeorology across Europe. These experts were asked to fill in this questionnaire for separate environmental zones within each country according to the zoning defined by Metzger et al. (2005) (Fig. 7.5).
Figure 7.5: Environmental zones in Europe (Metzger et al., 2005)
278
All of the questions involved some subjective assessments with a scoring that was based on the experts own evaluations (Tables 7.3, 7.4 and 7.5).
Table 7.3: Scale used for scoring present limitations of crops to climatic conditions Score NA NI 0 1 2 3 4 5 Explanation Not applicable (e.g. crop not grown) No information No problem Minor problem, occurs rarely, no detectable effects on regional production Small problem, occurs sometimes, small and rare effects on regional production Moderate problem, occurs occasionally, small effects on regional production Major problem, occurs frequently, moderate effects on regional production Large problem, occurs almost every year, major effects on regional production
Table 7.4: Scale used for scoring projected climate impacts by 2050 Score NA NI -2 -1 0 1 2 Explanation Not applicable (e.g. crop not grown) No information Large decrease Small decrease No change or no effect even if there is a change in the parameter Small increase Large increase
Table 7.5: Scale used for scoring how much different adaptation options would contribute to change in cropping systems Score NA NI 0 1 2 3 Explanation Not applicable (e.g. crop not grown) No information None Minor Moderate Large
For the descriptive questionnaires responses were received for 16 countries in Europe. More responses were received for the quantitative questionnaires, and this gave responses for all relevant European environmental zones, although the number of responses varied among zones (Table 7.6).
279
Table 7.6: Number of responses received within European environmental zones Zone ALN - Alpine North BOR Boreal NEM Nemoral ATN - Atlantic North ALS - Alpine South CON - Continental ATC - Atlantic Central PAN Pannonian LUS Lusitanian ANA Anatolian MDM - Mediterranean Mountains MDN - Mediterranenan North MDS - Mediterranean South Total Number of responses per zone 1 2 4 6 4 10 4 4 1 0 4 5 5 50 Number of countries/regions 1 2 4 5 4 9 4 4 1 0 4 4 4 46
Responses for the quantitative questionnaires regarded all relevant European environmental zones with the exception of the Anatolian zone (ANA) that can be found only in Turkey which was not included in the study. Although the number of responses varied among zones (Table 7.6) all of them were covered sufficiently and all zones with significant crop production being covered by at least 4 respondents. The total number of responses was 50 from 26 countries and in most cases a whole team of experts took part in the questionnaire answering. Where needed the respondents were contacted back to verify their responses (e.g. when any inconsistency occurred between responses within one environmental zone) and the overall results were consulted with the respondents as well. The list of all respondents and questionnaire co-authors is given in the Annex I.
7.5 Vulnerabilities and climate impacts on crop Biophysical processes of agroecosystems are strongly affected by environmental conditions. The projected increase in greenhouse gases will affect agroecosystems either directly (primarily by increasing photosynthesis at higher CO2 (Kimball et al., 2002) or indirectly via effects on climate (e.g. temperature and rainfall affecting several aspects of ecosystem functioning (Olesen and Bindi, 2002) (Table 7.7). The exact responses depend on the sensitivity of the particular ecosystem and on the relative changes in the controlling factors.
280
Table 7.7: Influence of CO2, temperature, rainfall and wind on various components of the agroecosystem Component Plants Animals Water Soil Pests/diseases Weeds Influence of factor CO2 Dry matter growth Water use Fodder yield Soil moisture SOM turnover Quality of host biomass Competition Temperature Growth duration Growth and reproduction Irrigation demand Salinization SOM turnover Nutrient supply Generation time Earliness of attack Herbicide efficacy Rain/wind Dry matter growth Health Groundwater Wind- and water erosion Disease transmission
Many studies have assessed effects of climate change on agricultural productivity in Europe (e.g. Harrison et al., 2000; Maracchi et al., 2005). However, relatively little work has been done to link these results across sectors to identify vulnerable regions and farming systems (Olesen and Bindi, 2002). Such assessments are needed to properly identify needs for change in agricultural policy caused by climate change. Increasing atmospheric CO2 concentration stimulates yield of C3 crops and to a lesser extent C4 crops (Fuhrer, 2003). However, recent estimates of the yield benefit from increasing CO2 are smaller than earlier ones (Ainsworth and Long, 2005), and the average annual increase of the next decades is marginal compared with what has been achieved through conventional crop management and breeding (Berntsen et al., 2006). Increasing temperature affects crops primarily via plant development. With warming, the start of active growth is advanced, plants develop faster, and the potential growing season is extended. This may have the greatest effect in cooler regions, and may be beneficial for perennial crops or crops, which do not reach their vegetative phase, e.g. sugar beets. However, increased temperature reduces crop duration for many annual crops. In wheat, an increase by 1 C during grain fill reduces the length of this phase by 5%, and yield declines by a similar amount (Olesen et al., 2000). Similar responses were found in Finland for spring barley. Compared to temperate crops, sensitivity to warming may be even greater in tropical crops, in particular when they are grown at the borders of their natural range.
281
7.5.1 Present climatic limitations and vulnerabilities The results of the questionnaire indicate that there are considerable differences between individual environmental zones as well as between crops that were examined in our study (Fig. 7.6). Some of the crops could not be evaluated in all zones as grain maize and grapevine are not grown in ALN and BOR and winter wheat is not present in ALN.
e
Figure 7.6: Present limitation of crop production by climate factors for 5 selected crops over the individual European Environmental Zones: a) winter wheat; b) spring barley; c) grain maize; d) grassland; e) grapevine
282
The length of growing season considerably limits growth in BOR and ALN zone and also in ATN zone in case of grapevine. The growing season duration is also a limiting factor for grassland and grapevine production in MDM zone. Grain maize and grapevine are not limited by this factor in LUS, MDN and MDS zones. Damage caused by late frost in the spring (or early frost during the fall) are seen as a limiting factor particularly in case of grain maize and grapevine in all cooler zones where these crops are grown (i.e. NEM, ATN and CON). In case of grapevine the late frosts seem to be a limiting factor in a number of seasons in the PAN environmental zone, and in case of ALN zone risk for spring barley production is reported for some seasons. High frequency of rainy conditions complicates sowing and harvest across most of the north-western zones with the highest impacts being reported for ALN and BOR zones. Also in BOR, ATN and ALS zones rain during key field operations seem to be important especially in case of harvesting cereals and hay. On the other hand precipitation is not perceived as a limiting factor in the harvest or sowing timing in MDN and MDS, where conditions are much drier. Flooding and water stagnation in the field is considered as a persistent problem in ALN, BOR, ATN and MDM zones, especially in case of grasslands, winter wheat and spring barley. Flooding and especially water stagnation at the field has been reported also from the PAN environmental zone for cereals. On the other hand respondents did not consider water stagnation or floods as limitations in ATC, LUS, MDN and MDS zones. Out of all crops grapevine seems to be least endangered by this phenomena due to terrain (frequently on slopes), soil conditions (often on light and permeable soils) and environmental zones (mostly warmer and drier) where it is grown. Overwintering and damage caused to the crops during winter is considered a major problem in ALN zone in case of grasslands and in BOR zone for grasslands and winter wheat production. Local to regional scale damage has been reported also from NEM, ATN, ALS, CON and PAN zones in case of winter wheat, whilst the effect on grassland is considered to be only marginal in these areas. The damage caused by snow cover and overall winter conditions (including severe frosts) are also threats to grapevine production in ATN zone and less in ALS, CON and PAN areas. The hail damage seems to be a minor problem that occurs rarely and has no detectable effect on regional production in case of wheat, barley and grassland production across most zones. However it is considered as a quite prominent risk even for all crops in PAN zone and particularly in Serbia. Lower risk
283
reported for small grain cereals and grassland is most likely given by their relatively high resistance to the damage compared to grain maize and especially grapevine. In case of these two crops the hail damage is seen as a moderate problem in case of ALS, CON, PAN, LUS and MDM zones. One of the most interesting findings in our survey has been the prominence of drought as a limiting factor. Whilst drought rarely scores as the severely limiting factor it seems to be of concern across the all zones and all crops with exception of grapevine. In case of grapevine the damage is considered to occur only rarely affecting production at the regional level only. However it is surprising that grassland or winter wheat production seems to be quite substantially limited not only within warm and dry zones (i.e. MDN or MDS) but also in the mountains (MDM and ALS) as well as in the cool and relatively wet zones in the north (ALN, BOR or NEM). In Finland drought interferes with grass crop establishment and also regrowth after the first cut in each season. For winter wheat, drought is particularly important during a critical period for yield determination (floret and thereby seed set), but winter types are far better able to escape from drought due to a deep root system compared to spring types that are extremely vulnerable. When the total ranking of the individual limiting factors is done across the zones and crops drought appears to be the single most significantly limiting factor. On the other hand the pattern of perceived limitations caused by heat stress is much more erratic with reported effects mainly in case of wheat, barley and maize in PAN and partly in the MDS zones. However, it should be noted that even under present climate conditions most of the respondents considered this factor as an important one even in zones that are not viewed as being threatened by this weather phenomena (e.g. BOR, NEM or ALS). 7.5.2 Climate change impacts Arable crops A climatic warming will expand the area of cereals cultivation (e.g. wheat and maize) northwards (Kenny et al., 1993; Carter et al., 1996). For wheat, a rise in temperatures will lead to a small yield reduction, which often will be more than counterbalanced by the effect of increased CO2 on crop photosynthesis. The combination of both effects will for a moderate climate change lead to moderate to large yield increases in comparison with yields simulated for the present situation (Ghaffari et al., 2002; van Ittersum et al., 2003). Drier conditions and increasing temperatures in the Mediterranean region and parts of eastern Europe may lead to lower yields there and the adoption of new varieties and cultivation methods. Such yield reductions have been estimated
284
for eastern Europe, and the yield variability may increase, especially in the steppe regions (Sirotenko et al., 1997). Potato, as well as other root and tuber crops, has shown a large response to rising atmospheric CO2 (Kimball et al., 2002). On the other hand warming may reduce the growing season in some species and increase water requirements with consequences for yield. Climate change scenario studies performed using crop models show no consistent change in mean potato yield (Wolf and van Oijen, 2003). For sugar beet yield the increasing occurrence of summer droughts may severely increase yield variability (Jones et al., 2003). However, increased duration of the growing season will increase the yield potential for this crop in northern Europe, where water supply is adequate. Climate-related increases in crop yields are only expected in northern Europe, while the largest reductions are expected around the Mediterranean and in the Southwest Balkans and in the South of European Russia (Olesen and Bindi, 2002; Maracchi et al., 2005; Alcamo et al., 2007). In southern Europe, particularly large decreases in yield are expected for spring-sown crops (e.g. maize, sunflower and soybeans) (Audsley et al., 2006). Whilst, on autumnsown crops (e.g. winter and spring wheat) the impact is more geographically variable, yield is expected to strongly decrease in the most southern areas and increase in the northern or cooler areas (e.g. northern parts of Portugal and Spain) (Olesen et al., 2007; Santos et al., 2002). However, these results vary between SRES scenarios and climate models (Olesen et al., 2007). Some crops that currently grow mostly in southern Europe (e.g. maize, sunflower and soybeans) will become more suitable further north or in higher altitude areas in the south (Audsley et al., 2006). The projections for a range of SRES scenarios show a 30 to 50% increase in suitable area for grain maize production in Europe by the end of the 21st century, including Ireland, Scotland, southern Sweden and Finland (Hildn et al., 2005; Olesen et al., 2007). Moreover, by 2050 energy crops show a northward expansion in potential cropping area, but a reduction in suitability in southern Europe (Schrter et al., 2005). Perennial crops Many fruit trees are susceptible to spring frosts during flowering. A climatic warming will advance both the date of the last spring frosts and the dates of flowering, and the risk of damage to flower buds caused by late frost are likely to remain largely unchanged (Rochette et al., 2004). Additionally the risk of damage to fruit trees caused by early autumn frosts is likely to decrease. However, there may very well be increased problems with pests and diseases (Salinari et al., 2006).
285
Grapevine is a woody perennial plant, which requires relatively high temperatures. A climatic warming will therefore expand the suitable areas northwards and eastwards (Jones et al., 2005). However, in the current production areas the yield variability (fruit production and quality) may be higher under global change than at present. Such an increase in yield variability would neither guarantee the quality of wine in good years nor meet the demand for wine in poor years, thus implying a higher economic risk for growers (Bindi et al., 1996). However, yields in grapevine may be strongly stimulated by increased CO2 concentration without causing negative repercussions on the quality of grapes and wine (Bindi et al., 2001). A climatic warming is also likely to lead to unsuitable conditions for currently economically important traditional varieties, at least at their current locations. Olive is a typical Mediterranean species that is particularly sensitive to low temperature and water shortage, thus both the northern and southern limits of cultivation are conditioned by the climate. The area suitable for olive production in the Mediterranean basin may increase with climate warming (Bindi et al., 1992), and several new fields are already established in some regions of Northern Italy, complementing traditional fruit tree assortment. Several perennial crops are candidates for bioenergy crops (Sims et al., 2006). This includes willow for coppice and reed canary grass and Miscanthus for solid biofuel crops to be used in providing biomass for fuel in combined heat and power plants (Clifton-Brown et al., 2004) or for use in second generation bioethanol production. The climatic suitability for many of these perennial bioenergy crops is projected to increase over most of Europe for the 21st century (Tuck et al., 2006). Grasslands Grasslands will differ in their response to climate change depending on their type (species, soil type, management). In general, intensively managed and nutrient-rich grasslands will respond positively to both the increase in CO2 concentration and to a temperature increase, given that water supply is sufficient (Thornley and Cannell, 1997). Nitrogen-poor and species-rich grasslands, which are often extensively managed, may respond differently to climate change and increase in CO2 concentration, and the short-term and long-term responses may be completely different (Cannel and Thornley, 1998). Climate change is likely to alter the community structure of grasslands (Buckland et al., 2001; Lscher et al., 2004), in ways specific to their location and type, and these changes will often depend on complex interactions between soils, plants and animals. Management and species-richness of grasslands may increase their resilience to change (Duckworth et al., 2000).
286
Fertile, early succession grasslands have been found to be more responsive to climate change than more mature and/or less fertile grasslands (Grime et al., 2000). In general, intensively managed and nutrient-rich grasslands will respond positively to both increased CO2 concentration and temperature, given that water and nutrient supply is sufficient (Lscher et al., 2004). As a general rule, productivity of European grassland is expected to increase, where water supply is sufficient (Byrne and Jones, 2002; Kammann et al., 2005). On the other hand an increased frequency of summer droughts will severely affect grassland production in the affected areas. Weeds, pests and diseases The majority of the pest and disease problems are closely linked with their host crops. This makes major changes in plant protection problems less likely (Coakley et al., 1999). Conditions are more favourable for the proliferation of insect pests in warmer climates, because many insects can then complete a greater number of reproductive cycles (Bale et al., 2002). Warmer winter temperatures may also allow pests to overwinter in areas where they are now limited by cold, thus causing greater and earlier infestation during the following crop season. Insect pests are also affected directly by the CO2 effect through the amount and quality of the host biomass (Cannon, 1998). Climate warming will lead to earlier insect spring activity and proliferation of some pest species (Cocu et al., 2005). A similar situation may be seen for plant diseases leading to an increased demand for pesticide control (Salinari et al., 2006). Unlike pests and diseases, weeds are also directly influenced by changes in atmospheric CO2 concentration. Higher CO2 concentration will stimulate growth and water use efficiency in both C3 and C4 species (Ziska and Bunce, 1997). Differential effects of CO2 and climate changes on crops and weeds will alter the weed-crop competitive interactions, sometimes for the benefit of the crop and sometimes for the weeds. However, interaction with other biotic factors may also influence weed seed survival and thus weed population development (Leishman et al., 2000). Changes in climatic suitability will lead to invasion of weed, pest and diseases adapted to warmer climatic conditions (Baker et al., 2000). The speed at which such invasive species will occur depends on the change of climatic change, the dispersal rate of the species and on measures taken to combat non-indigenous species (Anderson et al., 2004). The dispersal rate of pests and diseases are most often so high that their geographical extent is determined by the range of climatic suitability (Baker et al., 2000). The Colorado beetle, the European cornborer, the Mediterranean fruit fly and karnal bunt are examples of pests
287
and diseases, which are expected to have a considerable northward expansion in Europe under climatic warming. Environmental impacts Environmental impacts of agriculture under a changing climate are becoming more and more important. In particular, the role of nitrate leaching on the quality of aquifers, rivers and estuaries is globally recognized (Galloway, 2004). A warming is expected to increase soil organic matter turnover provided sufficient water is available, and experiment have shown that increases in net N mineralisation rates may be considerably higher than the increases in soil respiration (Rustad et al., 2001). Projections made at European level for winter wheat showed for the 20712100 time-slice that decreases in N-leaching predominate over large parts of eastern Europe and some smaller areas in Spain, whereas increases occur in the UK and in smaller regions over many other parts of Europe (Olesen et al., 2007). This in combination with longer growing seasons for the aquatic ecosystems would likely lead to higher risk of algal blooms and increased growth of toxic cyanobacteria in lakes (Moss et al., 2003; Andersen et al., 2006). The climate change scenarios could also lead to increases in GHG emissions from agriculture. Increasing temperatures will speed decomposition where soil moisture allows (Davidson and Janssens, 2006), so direct climate impacts on cropland and grassland soils will tend to decrease SOC stocks for Europe as a whole (Smith et al., 2006). This effect is greatly reduced by increasing C inputs to the soil because of enhanced NPP, resulting from a combination of climate change and increased atmospheric CO2 concentration. However, decomposition becomes faster in regions where temperature increases greatly and soil moisture remains high enough to allow decomposition (e.g. North and East Europe), but does not become faster, where the soil becomes too dry, despite higher temperatures (southern France, Spain, and Italy) (Smith et al., 2006). Climatic impacts on the key crops within individual environmental zones
Winter wheat
The impacts of climate change on winter wheat are thought to be negative across most of the zones (Fig. 7.7a). Whilst higher temperatures are expected to enable using late maturing cultivars in BOR, NEM and ATN zones it will also mean shortening of growth in remaining zones with the ATC, LUS, MDM and MDN being most affected. However, the damage during winter and risk of frost damage are expected to be lower in most of the zones with the
288
exception of ALS. Improved conditions for sowing and harvest are expected in NEM, ALS and LUS zones, whilst notable worsening of the conditions is expected only in MDM zone.
a)
b)
c)
d)
e)
Figure 7.7: Expected impacts of climate change on a range crop production limiting factors for 5 selected crops: a) winter wheat; b) spring barley; c) grain maize; d) grassland; e) grapevine. The scale used for scoring is presented in the Table 7.2 and colour-coding reflects positive effect (green) or negative effect (red). The gray colour represents area
289
The changes in the seasonal climate variability are considered to have neutral or negative effect on winter wheat production with the NEM and PAN regions being the most vulnerable. Not surprisingly risk of drought and heat stress are thought to increase over all zones except MDM. However, this increase is according to the used scale considered to be small with the exception of heat stress risk in PAN area where large increase is expected. The biggest threat for the northern and central European zones (BOR, NEM, ATN, ALS, CON and PAN) is thought to be higher risk from plant pathogens and pests, whilst in the southern zones this problem is considered to be only marginal. Higher intensity of weed occurrence or an introduction of new weed species has been picked up by most respondents with the exception of NEM and ALS zones. Risk from a soil erosion and nitrogen leaching is much higher in the zones where higher precipitation is expected (BOR, ATN or ALS) but also mentioned as a threat (however of lower importance) in the Mediterranean region (MDS and MDN).
Spring barley
As in the case of winter wheat the change in the evaluated parameters is expected to influence spring barley production mostly negatively (Fig. 7.7b). Increase of temperature is expected to prolong growth duration in the northern range of spring barley growing area (i.e. ALN, BOR and ATN) while a negative influence is expected especially in ALS, LUS, MDM and MDN. The frost risk is thought to decrease or remain the same in most of the zones. Improved conditions for harvest are expected in NEM, ALS and LUS and on the other hand in MDM and MDM zone a notable worsening of the harvest conditions is expected. The impact of changes in the seasonal variability is in general perceived as negative with the exception of Mediterranean area (MDM, MDN and MDS). The respondents expect the biggest changes for the PAN, NEM and ATN zones. Drought is perceived as a very prominent risk in most of the zones (except MDM) and spring barley production is thought to be more at risk compared to the winter wheat. Damage caused by the heat stress is also expected to rise in most of the zones, but with the cooler zones thought to be more vulnerable compared to the MDM, MDN or MDS. As it was in the case of wheat, changes in the hail risk do not show any significant pattern with the exception of pronounced risk increase in the PAN zone. On the other hand the expected damage caused by the phytopathogens and pests is significantly greater than in the case of winter wheat. Similar to this crop the negative impacts are thought to dominate in ALN, BOR, NEM, ATN, ALS, CON and ATC zones but in the remaining part of Europe no change or even improvement are expected. Higher occurrence of weeds or introduction of new weed species has been picked up by most respondents with the exception of
290
NEM zone. Similar to the wheat case the soil erosion and nitrogen leaching are expected to increase under the changed climate with the exception of MDS.
Grain maize
The estimates of impact of climate change on the grain maize production indicate that this only C4 crop species among the range of assessed crops is expected to perform much better than the previous two C3 cereal crops (Fig 7.7a-c). Increase of temperatures is thought to positively influence length of growing season in BOR and ATN zone whilst shortening of growth duration is expected in CON, LUS, MDM and MDS zones. On the other hand across all the zones the decrease of the late frost risk is expected as well as increase in the number of suitable days for harvest (although in many cases only small one). While the impact of drought and heat stress in the colder zones (e.g. NEM, ATN or CON) are thought to be smaller compared to wheat or barley, grain maize seem to be more vulnerable to drought and heat stress in PAN, LUS or MDS zones compared with the present conditions. Despite the fact that the impact of climate change is expected to lead to increase of hail damage, pest and disease risk, increase of weed pressure, higher soil erosion and nitrogen leaching, the intensity of these changes are in most zones smaller compared with small grain cereals.
Grassland
Out of our sample of five crops, grasslands seem to be least affected by the climate change. In all zones growth duration is expected to increase especially in ALN, BOR, NEM, ATN and ALS. In the same time damage during winter and those caused by frosts is expected to decrease and the number of days suitable for harvest is thought to increase (but mostly slightly) in all zones except ALN. Only marginal negative impact is expected from hail occurrence, heat stress, soil erosion and nitrogen leaching and weed occurrence with a notable exception of ALN zone where some of these parameter are changing to the worse (Fig. 7.7d). However drought and changes of seasonal climate variability is expected to cause negative impact across all zones with quite significant effect on ALN, BOR, NEM, ALS, CON, PAN and LUS zones. Interestingly the magnitude of changes (both positive and negative) is thought to be highest in the northernmost zones (ALN, BOR and NEM) with only subtle changes expected to grassland production in Mediterranean (LUS, MDM, MDN and MDS).
291
Grapevine
The changed climate conditions are expected to lead to decrease in winter and frost damage in the cooler ones of the wine-producing zones (ATN, CON or ATC) but increased frost risk is expected in the warmest areas (MDS and LUS). Number of days suitable for harvest also increases slightly in most of the zones with a decrease being reported only in MDM zone. The length of the growing season is expected to decrease in the LUS, MDM, MDN and slightly in the CON and MDS zones. On the other hand increased temperatures will lead to an increased period of growth in the ATN zone. Despite the fact that changes in all remaining parameters are expected to be negative across most of the zones, the changes are not expected to be large. Significant increase of drought and heat stress was estimated by our respondents only for the ALS zone and partly for MDN area, where irrigation is largely practiced for this crop. Grapevine production is thought to suffer from the increased hail damage risk across the zones and higher risk of diseases and pest occurrence is expected especially in the ATN and ALS zones. The soil erosion and nitrogen leaching is expected to rise, however not significantly probably with the exception of the CON zone where quite pronounced increase of soil erosion is expected.
7.6 Adaptation to climate variability and climate change To avoid or at least reduce negative effects and exploit possible positive effects, several agronomic adaptation strategies for agriculture have been suggested. Studies on the adaptation of farming systems to climate change need to consider all the agronomic decisions made at the farm level (Kaiser et al., 1993). The agronomic strategies available include both short-term adjustments and long-term adaptations (Olesen and Bindi, 2002). Autonomous adaptations The short-term adjustments include efforts to optimise production without major system changes. They are autonomous in the sense that no other sectors (e.g. policy, research, etc.) are needed in their development and implementation. Examples of short-term adjustments are changes in varieties, sowing dates and fertiliser and pesticide use (Ghaffari et al., 2002; Alexandrov et al., 2002; Tubiello et al., 2000; Chen and McCarl, 2001). In particular, in southern Europe short-term adaptations may include changes in crop species (e.g. replacing winter with spring wheat) (Minguez et al., 2007), changes in cultivars and sowing dates (e.g. for winter crops, sowing the same cultivar earlier, or choosing cultivars with longer crop cycle; for summer irrigated
292
crops, earlier sowing for preventing yield reductions or reducing water demand) (Olesen et al., 2007; Kaukoranta and Hakala, 2008). There are many plant traits that may be modified to better adapt varieties to increased temperature and reduced water supply (Sinclair and Muchow, 2001). The use of early ripening fruit tree species may reduce the water consumption, as proper management practices may be applied to orchards to improve adaptation (Rossi, 2006). However, the effectiveness of such traits depend on whether there is simultaneous change in climatic variability, and a combination of traits may be needed to stabilise yield in poor years, without sacrificing yield in good years (Porter et al., 1995; Sinclair and Muchow, 2001). In northern Europe new crops and varieties may be introduced only if improved varieties will be introduced to respond to specific characteristics of the growing seasons (e.g. length of the day) (Hilden et al., 2005; PeltonenSainio et al., 2008). Long-term adaptations The long-term adaptations refer to major structural changes to overcome adversity caused by climate change. This involves changes in land allocation and farming systems, breeding of crop varieties, new land management techniques, etc. This involves changes of land use that result from the farmer's response to the differential response of crops to climate change. The changes in land allocation may also be used to stabilise production. This means substitution of crops with high inter-annual yield variability (e.g. wheat or maize) by crops with lower productivity but more stable yields (e.g. pasture or sorghum). Crop substitution may be useful also for the conservation of soil moisture. Long lead times in crop substitution are present for the perennial crops (e.g. grapevine, olive and fruit trees), and here adequate and region specific information on climate change and suitable species and varieties are critical for efficient adaptation. Other examples of long-term adaptations include breeding of crop varieties, new land management techniques to conserve water or increase irrigation use efficiencies, and more drastic changes in farming systems (including land abandonment). Increasing the supply of water for irrigation may not be a viable option in much of southern Europe, since the projections show a considerable reduction in total runoff (Lehner et al., 2006). Changes in farming systems The farm is typically the entity at which adaptation to climate change and climatic variability must take place through introduction of new management methods and technologies. Because of the complexities of processes,
293
management and inter-relationships of land use within a farm, studies on farming systems require a holistic approach (Rivington et al., 2006). Climate change will not only affect crop yield, but total farm-level production through effects on altered carbon and nitrogen flows resulting from changed crop and residue quality, crop resource use, or mineralisation of soil organic matter (Dueri et al., 2006). Adaptation will have to deal with all of these issues, and the links with water availability may be among the most important ones, affecting the need for improving irrigation efficiencies (Tavakkoli and Oweis, 2004) or the need for terracing (Wadsworth and Swetham, 1988; Fuhrer et al., 2006). Recent studies at European level have demonstrated the need to include changes in climate and non-climate factors (technological, socio-economic, etc.) for assessing the changes in crop yield and suitability (Schrter et al., 2005). A different allocation of European agricultural land use seems to represent one of the major long-term adaptation strategies available. Rounsevell et al. (2005) estimate a decline of up to 50% in cropland and grassland areas under the A1FI and A2 scenarios. For the A1FI and A2 scenarios both the quantity and the spatial distribution of crops will change, whilst, for the B1 and B2 scenarios the pressures toward declining agricultural areas should be counterbalanced by policy mechanisms that seek to limit crop productivity. Changes in farming systems may also play a fundamental role in the adaptation of European agriculture to climate change. The interpretation of four IPCC-SRES scenarios suggests that different types of adaptation of farming systems (intensification, extensification and abandonment) may be appropriate for particular scenarios and areas (high latitude and altitude, marginal areas, etc.) (Berry et al., 2006). The sensitivity to climate change of farming systems may depend on the degree of diversification. However, based on data from a large number of operations in Canadian prairie agriculture, farms have recently become more specialized, and this trend is unlikely to change in the immediate future (Bradshaw et al., 2004). A similar situation is likely to take place in Europe, although the trend to organic farming in some areas and the urbanization of some rural areas may restrict this development. Finally, the substitution of food production by energy production through the widespread cultivation of bioenergy crops will affect land use (Tuck et al., 2006). Several temperate and Mediterranean crop species are suitable for various types of biofuels, including oilseed crops, starch crops, cereals and solid biofuel crops. All climate change scenarios show a northward expansion of these species with northern Europe becoming more favourable for most species. However, the choice of energy crops in southern Europe may be
294
severely reduced in future, both due to increased temperatures and reduced rainfall. Taking into account potential impacts and adaptive capacity, the vulnerability of agriculture based on Farmer livelihood (profit) have been analysed for EU15 (Metzger et al., 2006). The results show the agricultural sector in the Mediterranean region as vulnerable under most climate change scenarios starting at different time slices, depending on the scenario used. The A1FI and A2 scenarios anticipate greater vulnerability throughout; whilst the B2 scenario seems to be least harmful for farmers livelihood (Metzger et al., 2006). 7.6.1 Observed adaptation Observed adaptation based on quantitative questionnaires The quantitative questionnaire asked about ten specific adaptation responses and about five new crop species that were thought to be applicable through most of the environmental zones and would be picked up by respondents. As Fig. 7.8 shows, minor to moderate changes in the cultivation timing were observed in all environmental zones during past decade.
Figure 7.8: Observed adaptation responses by the farmers as reported by respondents in individual environmental zones
295
The most significant changes seem to be going on in the cooler zones (e.g. BOR, CON, ATC or PAN) compared to the Mediterranean (MDN or MDS). The change of cultivation includes not only changes in the tillage practices but mainly shifts in the sowing dates (e.g. tendency to earlier sowing of spring crops). The introduction of new crops to the crop rotation has been also reported in the study with increase in the area of silage and grain maize being the most notable changes. In case of silage and grain maize ATN and CON areas seem to show the largest change. The grain maize has also shown increase in the PAN zone. BOR, CON and PAN zones show increased interest in growing of sunflower as a response to changing climate conditions and this goes frequently in hand with introduction of soybeans to these zones. The respondents in NEM, ATN, CON, PAN, MDN and MDS noted tendency to more grapevine production. Whilst in the cooler zones the main driver seems to be more favourable climate conditions enabling introduction of the crop, in the southern zones the main reason seems to be higher tolerance of grapevine to drought compared with the field crops. Especially in the warmest zones (LUS, MDN and MDS) we can see a tendency of farmers to reduce crops that are unsuitable for the changing climate conditions. Despite the tendency to new cropping schemes farmers seem to be more interested in maintaining the present portfolio of crops that is documented by evident tendency to introduce new and more suitable cultivars of the presently grown crops across all zones. This is even more evident from a reported increase of interest in the cultivars that are able to cope better with drought and other weather extremes. As the drought has been spotted as one of the most pervasive crop growth limitations, there has been a wide spread effort to promote techniques that preserve soil water, especially in the most drought prone regions (PAN, MDN and MDS) but partly also in other zones. This response has been combined with the expansion of irrigated areas. Although the irrigation expansion seems to be an obvious response in very dry zones, where water resources are limited (LUS and MDS) we have seen quite marked drop in the area under irrigation in combination with change of crop structure. Almost all zones show quite pronounced effort in introduction of cultivation that reduces soil erosion that might be an indication of higher frequency of more intense precipitation (as a lead cause of water erosion) but also the result of more frequent droughts as a prerequisite of wind erosion over the area. The changes in rainfall patterns are most likely behind the reported improvements in the field drainage systems but this measure seems to be the least notable.
296
7.6.2 Future adaptation responses Cultivation timing Changes in the timing of cultivation (including sowing and harvest) are expected to have minor to moderate effect for all five model crops evaluated by the respondents (Fig. 7.9). Interestingly the largest changes are expected on the opposite sides of climate gradient i.e. ALN and BOR in case of wheat, barley and grassland production and MDM and MDN in case of barley, maize, grassland and grapevine.
e
Figure 7.9: Expected importance of adaptation measures under the expected climate conditions for individual crops: a) winter wheat; b) spring barley; c) grain maize; d) grassland; e) grapevine
297
In general, significant changes of the sowing dates (and consequently of other field operations) are expected in order to avoid dry periods during summer and use as much of winter precipitation as possible. The anticipation of large shifts in the cultivation timing in the northern zones is probably enhanced by the pronounced prolongation of growing season that will allow introduction of cultivars with longer growth period. Change in tillage practices Change of tillage practices is mostly focused on the soil water conservation and protecting the soil against soil erosion (both water and wind). Tillage practices focused on soil water conservation are considered by respondents as likely adaptation response by grapevine farmers in the MDM, MDS and MDS zone that is associated with expected increase of water stress. Introduction of water-conserving tillage practices are assumed to be an important adaptation measure especially in case of the wheat, barley and partly maize in the warm and dry zones (PAN, MDM, MDN and partly MDS and CON), the zones that are expected to experience increase of precipitation (e.g. BOR or ATN) put more stress on the erosion protection as soil water reserves are not expected to be replenished during winter months. In grassland case the most notable change will occur in ALN and BOR zones when the length of growing season will increase leading to increase of potential productivity that will facilitated changes in the tillage practices. Modification of the fertilization The expected shifts of fertilization patterns show an interesting north-south gradient with the northern most zones expecting moderate to major changes in the fertilization schemes both in field crops (wheat, barley or maize) as well as perennials (grassland and partly also grapevine). As the potential productivity of northern zones (ALN, BOR, NEM or ATN) is expected to be increased due to longer vegetation season it will also require adequate increase of available nutrients for the crops. However, the expected increase of precipitation (in combination with more strict EU environmental regulation) will lead to higher risk of nitrogen leaching resulting in necessary modifications of fertilization schemes used in these zones. The expectations for drier zones are much lower and only minor shifts are expected as far as fertilization is concerned in MDN, MDS, LUS, ATC and partly PAN in case of barley, maize and grasslands. In case of grapevine the picture is more complex with ATN, CON and MDM zones expecting larger changes than remaining zones.
298
Introduction of new cultivars The prospect of this adaptation response is obviously lowest in case of grassland (Fig 7.9d) whilst it is expected to be important in case of field crops and in some zones also for the grapevine. The new cultivars are expected to be more important in case of spring barley and grain maize. According to the respondents this measure is thought to be quite pronounced in case of NEM, AKS, CON, PAN, MDN and MDM zones. Modification of crop protection and pest and disease monitoring The expected change in the crop protection technologies is one of the most prominent adaptation measures especially in the case of wheat, barley, maize and partly grapevine. Obviously in case of grasslands the economic benefit of crop protection and monitoring is quite low (as are the risks) and thus the potential of these adaptation measure is considered to be small. However in case of wheat, barley and maize the changes in the crop protection schemes are expected to be major ones, especially in the BOR, NEM, ATN, ALS, CON and PAN zones. It is not surprising that within these zones also the introduction of pest and disease monitoring scheme has been emphasized as one of the key adaptation responses. Such a response could be explained when we realized that changed climate conditions will most likely cause spread of pests and diseases from warmer zones which threatens more zones in the north than those in the south. In addition, in BOR or ATN the expected higher precipitation might result in higher infestation pressure of some native diseases (e.g. Fusarium sp.). Overall the importance of some sort of monitoring system has been mentioned by respondents over all zones with the exception of LUS. Seasonal weather forecasting The changed climate conditions and according to some indications higher probability of unusual weather patterns lead most respondents to stress the role of seasonal forecasting as an adaptation tool. The seasonal forecasting is expected to be important in case of field crops and partly also grapevine, whilst in case of grasslands it is given minor priority probably for the same reasons as phytopathological monitoring. For wheat, barley and maize the biggest effect of seasonal forecasting is expected in zones where relatively large inter-seasonal differences are more likely, i.e. those with higher continentality (e.g. BOR, NEM or PAN) compared to ATN or ATC zones whilst in case of Mediterranean zones (MDM, MDN and MDS) the importance of seasonal forecast is stressed by the increasing uncertainty in the rainfall amounts.
299
Crop insurance As the crop insurance is seen as a quite effective tool mitigating the effect of unsuitable weather conditions during growing season (or occurrence of severe weather events) the highest importance has been reported for the zones where the climate impact are expected to be mostly negative. This explains why the respondents in the NEM, CON, ALS and especially PAN zones put large emphasis on this particular adaptation measure, whilst in ATC or LUS zone this option has not been seen as important. The expected benefit of the crop insurance is not seen as directly dependent on the value of the crop per unit of area as the expected importance of crop insurance for grassland is of the same magnitude as in the case of maize or grapevine with obvious regional differences.
7.7 Improving awareness and adaptation to climate change 7.7.1 National impact assessments, adaptation strategies and awareness Fig. 7.10 documents existence of top-town gradient in the collective knowledge about the imminent climate change impacts on agriculture with better level of understanding by the governmental offices and the lower one in case of extension service and individual farmers.
Figure 7.10: Reported level of climate change awareness among farmers, agriculture advisors and government officials in 26 countries and the status of agriculture adaptation strategy and education programs for farmers
300
Despite the upsurge of the information flow from the various sources the level of awareness seems to be relatively low compared to the level of observed adaptation reported in the survey (Fig. 7.8). Only in one-third of the countries the most affected group (i.e. farmers) are considered to have a good understanding to the consequences of climate change on their livelihoods. It also seems that the policy makers in majority of the governments are not sufficiently informed about the risks associated with the climate change impacts for agriculture. This could be a possible explanation of a large discrepancy between the claimed awareness among government officials as almost 2/3 of countries claim to have medium or good level of information about the possible impacts, but only 3 countries reported an existing agriculture adaptation strategy whilst the rest of them either does not have any or it is in the process of preparation. Although 75% of responding countries acknowledged activity aimed at increasing farmers awareness there seems to be quite long way to go before sufficient level of understanding is reached. 7.7.2 Dissemination and warning systems The level of information dissemination and existence of warning systems shows that a lot needs to be done in improving resilience of farming systems across Europe. The survey showed quite large differences between individual countries (and crops) but in general the use of decision support systems (DSS) is lower than expected (Fig. 7.11). It is even more surprising when we compared the use of DSS with the present climate limitation for crop production in individual zones. Whilst drought seems to be a pervasive problem across all of the zones and is expected to get worse under the changed climate only half of the countries have some sort of DSS system and only one fifth of countries have some sort of nation-wide drought monitoring scheme. Even worse is the situation in case of heat-stress and weed management. On the other hand, selection of suitable crops or cultivars, crop protection and fertilisation schemes seems to be quite well supported by existing DSS with more farm-based approach in case of fertilisation and regional approach in case of the crop protection. One of the possible explanations of the present state is that the listed factors play an important role in the economy of every farm and direct benefits of DSS is well understood by the farmer as it could be well measured in terms of cost savings. On the other hand the drought or heat stress DSS benefit is less straight forward and whilst it might provide essential information to the decision makers, farmers are less inclined to demand such systems. Surprisingly in case of the irrigation scheduling that has been always seen as one of the most efficient application of DSS, the results of the survey show rather mixed picture. Whilst in some countries quite sophisticated DSS are applied (mostly in drier zones), in many countries where irrigation is used
301
for wheat, maize, grasslands and grapevine production there is no DSS system in place. This is in particular troubling when the drought and irrigation trends are considered.
Figure 7.11: Present level of decision support system use for four key crops across 26 responding states: a) winter wheat; b) grain maize; c) grassland; d) grapevine
7.8 Implications and perspectives The projected changes and the perceptions of impacts and adaptation as seen from the questionnaires have some implications for agricultural and environmental policy, for research and for development of the agricultural sector. Policy will have to support the adaptation of European agriculture to climate change by encouraging the flexibility of land use, crop production, farming systems etc (Olesen et al., 2002). In doing so, it is necessary to consider the multifunctional role of agriculture, and to strike a variable balance between economic, environmental and economic functions in different European regions. Policy will also need to be concerned with agricultural strategies to mitigate climate change through a reduction in emissions of methane and nitrous oxide, an increase in carbon sequestration in agricultural soils and the
302
growing of energy crops to substitute fossil energy use. The policies to support adaptation and mitigation to climate change will need to be linked closely to the development of agri-environmental schemes in the EU Common Agricultural Policy. Research will have to deal with some unknown aspects that due to their complexity have not yet been studied in detail. These include the effect on secondary factors of agricultural production (e.g. soils, weeds, pests and diseases), the effect on the quality of crop and animal production, the effect of changes in frequency of isolated and extreme weather events on agricultural production, and the interaction with the surrounding natural ecosystems. It should also be noted that for obvious reasons most studies on climate change impacts have so far focused on crop production. However, some livestock production systems, especially those involving grazing systems or use of fresh fodder, may be severely affected by climate change, and more studies on these systems are warranted. There is a considerable need for an increased focus on regional studies of impacts and adaptation to climate change in agriculture, since effects and responses are likely to be regionally specific depending on interactions with soils, current climate and cropping systems (Olesen et al., 2006a). Such studies are now becoming more realistic with the arrival of detailed regional scale climate change scenarios (Christensen and Christensen, 2007). These studies should include assessments of the consequences on current efforts in agricultural policy for a sustainable agriculture that also preserves environmental and social values in the rural society. The research on adaptation in agriculture has not yet provided a generalised knowledge on the adaptive capacity of agricultural systems across a range of climate and socioeconomic futures. There is also a considerable need to better estimate the costs of various adaptation measures, and adaptation studies have to move from looking at potential adaptation to adoption, taking into account the complexity of farm-level decision-making, diversities at different scales and regions (including the entire food chain), and time-lags in responses and biophysical, economic, institutional and cultural barriers to change. The adaptation to climate change has in particular to be factored in as part of the ongoing technological development in agriculture, including plant breeding (also using molecular techniques), livestock feeding technologies, irrigation management, application of information and communication technology etc. This would be feasible utilising the main agricultural resources in agriculture (Olesen and Bindi, 2002). In some cases such adaptation measures would make sense without considering climate change, because they help to address current climate variability. In other cases, the measures must be implemented in anticipation of climate change, because they would be ineffective if implemented as a reaction to climate change (Smith and Lenhart, 1996).
303
7.9 Conclusions The results of the studies on anthropogenic climate change performed in the last decade over Europe indicate consistent increases in projected temperature and different patterns of precipitation with widespread increases in northern Europe and rather small decreases over southern Europe. These changes in climate patterns are expected to affect all the components of the European agricultural ecosystems (e.g. crop suitability, yield and production, etc.). Thus, adaptation strategies should be introduced to reduce negative effects and exploit possible positive effects of climate change. Both short-term adjustments (e.g. changes in crop species, cultivars and sowing dates) and long-term adaptations (e.g. water management, land allocation, farming systems and institutions) should be considered. However, the differences in climate exposure, sensitivity, and adaptive capacity will affect in a different way the agricultural eco-systems across Europe. In particular, agriculture in the Mediterranean region seems to be more vulnerable than in other European regions. This calls for a considerable effort in research and development to deal with the changes, both at the continental and regional levels.
7.10 Acknowledgements We gratefully acknowledge all the experts contributing to the questionnaires. M. Trnka acknowledges support of KONTAKT OC187 project for supporting COST 734 activities in the Czech Republic.
7.11 References
Ainsworth E.A. and Long S.P. 2005. What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytol. 165:351-372. Alcamo J., J.M. Moreno, B. Novky, M. Bindi, R. Corobov, R.J.N. Devoy, C. Giannakopoulos, E. Martin, J.E. Olesen, A. Shvidenko, 2007. Europe. In M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden, C.E. Hanson, (Eds), 2007. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK, p. 541-580. Alexandrov V., J. Eitzinger, V. Cajic, M. Oberforster, 2002. Potential impact of climate change on selected agricultural crops in north-eastern Austria. Global Change Biol. 8:372389. Andersen H.E., B. Kronvang, S.E. Larsen, C.C. Hoffmann, T.S. Jensen, E.K. Rasmussen, 2006. Climate-change impacts on hydrology and nutrients in a Danish lowland river basin. Sci. Tot. Environ. 365:223237.
304
Anderson P.K., A.A. Cunningham, N.G. Patel, F.J. Morales, P.R. Epstein, P. Daszak, 2004. Emerging infectious diseases of plants: pathogen pollution, climate change and agrotechnology drivers. Trends Ecol. Evol. 19:535-544. Andr J.C., M. Dqu, P. Rogel, S. Planton, 2004. La vague de chaleur de l't 2003 et sa prvision saisonnire. Geoscience 336:491-503. Audsley E., K.R. Pearn, C. Simota, G. Cojocaru, E. Koutsidou, M.D.A. Rounsevell, M. Trnka, V. Alexandrov, 2006. What can scenario modelling tell us about future European scale agricultural land use, and what not? Environ. Sci. Pol. 9:148-168. Baker R.H.A., C.E. Sansford, C.H. Jarvis, R.J.C. Cannon, A. MacLeod, K.F.A. Walters, 2000. The role of climatic mapping in predicting the potential distribution of non-indigenous pests under current and future climates. Agric. Ecosyst. Environ. 82:57-71. Bale J.S., G.J. Masters, I.D. Hodkinson, C. Awmack, T.M. Bezemer, V.K. Brown, J. Butterfield, A. Buse, J.C. Coulson, J. Farrar, J.E.G. Good, R. Harrington, S. Harley, T.H. Jones, R.L. Lindroth, M.C. Press, I. Symrnioudis, A.D. Watt, J.B. Whittaker, 2002. Herbivory in global climate change research: direct effects of rising temperature on insect herbivores. Global Change Biol. 8:1-16. Beniston M., 2004. The 2003 heat wave in Europe: A shape of things to come? An analysis based on Swiss climatological data and model simulations. Geophys. Res. Let. 31:L02202. Beniston M. and Diaz H.F., 2004. The 2003 heat wave as an example of summers in a greenhouse climate? Observations and climate model simulations for Basel, Switzerland. Global Planet. Change 44:73-81. Beniston M., D.B. Stephenson, O.B. Christensen, C.A.T. Ferro, Frei C, Goyette S, Halsnaes K, Holt T, Jylh K, Koffi B, Palutikof J, Schll R, Semmler T, Woth K. 2007. Future extreme events in European climate: an exploration of regional climate model projections. Climatic Change, 81: 71-95. DOI: 10.1007/s10584-006-9226-z Berntsen J., B.M. Petersen, J.E. Olesen, 2006. Simulating trends in crop yield and soil carbon in a long-term experiment - effects of rising CO2, N deposition and improved cultivation. Plant Soil 287:235-245. Berry P.M., M.D.A. Rounsevell, P.A. Harrison, E. Audsley, 2006. Assessing the vulnerability of agricultural land use and species to climate change and the role of policy in facilitating adaptation. Environ. Sci. Pol. 9:189-204. Bindi M., F. Ferrini, F. Miglietta, 1992. Climatic change and the shift in the cultivated area of olive trees. J. Agric. Mediter. 22:41-44. Bindi M., L. Fibbi, B. Gozzini, S. Orlandini, F. Miglietta, 1996. Modeling the impact of future climate scenarios on yield and yield variability of grapevine. Clim. Res. 7:213-224. Bindi M., L. Fibbi, F. Miglietta, 2001. Free air CO2 enrichment (FACE) of grapevine (Vitis vinifera L.): II. Growth and quality of grape and wine in reponse to elevated CO2 concentrations. Eur. J. Agron. 14:145-155. Bouma J., G. Varallyay, N.H. Batjes, 1998. Principal land use changes anticipated in Europe. Agric. Ecosyst. Environ. 67:103-119. Bradshaw B., D. Holly, B. Smit, 2004. Farm-level adaptation to climatic variability and change: crop diversification in the Canadian Prairies. Clim. Change 67:119-141. Buckland S.M., K. Thompson, J.G. Hodgson, J.P. Grime, 2001. Grassland invasions: effects of manipulations of climate and management. J. Appl. Ecol. 38:301-309. Byrne C. and Jones M.B., 2002. Effects of elevated CO2 and nitrogen fertilizer on biomass productivity, community structure and species diversity of a semi-natural grassland in Ireland. Proc. R. Irish Acad. 102B:141-150. Calanca P., 2007. Climate change and drought occurence in the Alpine region: Howe severe are becoming the extremes?. Global and planetary change. 57: 151-160. DOI:
305
10.1016/j.gloplacha.2006.11.001. Cannell M.G.R. and Thornley J.H.M., 1998. N-poor ecosystems may respond more to elevated [CO2] than N-rich ones in the long term. A model analysis of grassland. Global Change Biol. 4:431-442. Cannon R.J.C., 1998. The implications of predicted climate change for insect pests in the UK, with emphasis on non-indigenous species. Global Change Biol. 4:785-796. Carter T.R., R.A. Saarikko, K.J. Niemi, 1996. Assessing the risks and uncertainties of regional crop potential under a changing climate in Finland. Agric. Food Sci. Finland 3:329-349. Chen C.C. and McCarl B.A., 2001. An investigation of the relationship between pesticide usage and climate change. Clim. Change 50:475-487. Chloupek O., P. Hrstkova, P. Schweigert, 2004. Yield and its stability, crop diversity, adaptability and response to climate change, weather and fertilisation over 75 years in the Czech Republic in comparison to some European countries. Field Crops Res. 85:167-190. Christensen J.H. and Christensen O.B., 2003. Severe summertime flooding in Europe. Nature 421:805-806. Christensen J.H. and Christensen O.B., 2007. A summary of PRUDENCE model projections of changes in European climate by the end of this century. Clim. Change 81:7-30. Christensen J.H., B. Hewitson, A. Busuloc, A. Chen, X. Gao, I. Heid, R. Jones, R.K. Kolli, W.T. Kown, R. Laprise, V.M. Rueda, L. Mearns, C.G. Menndez, J. Risnen, A. Rinke, A. Sarr, P. Whetton, 2007. Regional climate projections. In: Climate change 2007: The physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tigor, H.L. Miller, (eds)). Cambridge University Press, Cambridge, UK and New York, NY, USA. p. 847-940. Ciais Ph., M. Reichstein, N. Viovy, A. Granier, J. Oge, V. Allard, M. Aubinet, N. Buchmann, C. Bernhofer, A. Carrara, F. Chevallier, N. De Noblet, A.D. Friend, P. Friedlingstein, T. Grnwald, B. Heinesch, P. Keronen, A. Knohl, G. Krinner, D. Loustau, G. Manca, G. Matteucci, F. Miglietta, J.M. Ourcival, D. Papale, K. Pilegaard, S. Rambal, G.Seufert, J.F. Soussana, M.J. Sanz, E.D. Schulze, T. Vesala, R. Valentini, 2005. Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature 437:529-533. Clifton-Brown J.C., P.F. Stampfl, M.B. Jones, 2004. Miscanthus biomass production for energy in Europe and its potential contribution to decreasing fossil fuel carbon emissions. Global Change Biol. 10:509-518. Coakley S.M., H. Scherm, S. Chakraborty, 1999. Climate change and plant disease management. Annu. Rev. Phytopathol. 37:399-426. Cocu N., et al., 2005. Geographical location, climate and land use influences on the phenology and numbers of the aphid, Myzus persicae, in Europe. J. Biogeography 32:615-632. Dai A., K.E. Trenberth, T.R. Karl, 1998. Global Variations in Droughts and Wet Spells: 19001995. Geophysical Research Letters 25 (17): 3367-3370. DOI: 10.1029/98GL52511. Dai A., K.E. Trenberth, Q. Taotao 2004. A Global Dataset of Palmer Drought Severity Index for 1870-2002: Relationship with Soil Moisture and Effects of Surface Warming. Journal of Hydrometeorology 5: 1117-1130. DOI: 10.1175/JHM-386.1. Darwin R. and Kennedy D., 2000. Economic effects of CO2 fertilization of crops: transforming changes in yield into changes in supply. Environ. Model. Assess. 5:157-168. Davidsson E.A. and Janssens I.A., 2006. Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature 440:165-173. Duckworth J.C., R.G.H. Bunce, A.J.C. Malloch, 2000. Modelling the potential effects of climate change on calcareous grasslands in Atlantic Europe. J. Biogeography 27:347-358.
306
Dueri S., P.L. Calanca, J. Fuhrer, 2006. Climate change affects farm nitrogen loss - A Swiss case study with a dynamic farm model. Agric. Syst. 93:191-214. EEA, 2004. Impacts of Europe's changing climate: An indicator-based assessment. EEA Report No 2/2004, European Environment Agency, Copenhagen, Denmark, 107 pp. Fink A.H., T. Brcher, A. Krger, G.C. Leckebusch, J.G. Pinto, U. Ulbrich, 2004. The 2003. European summer heat waves and drought - Synoptic diagnosis and impact. Weather 59:209-216. Frich P., L.V. Alexander, P. Della-Marta, B. Gleason, M. Haylock, A.M.G.K. Tank, T. Peterson, 2002. Observed coherent changes in climatic extremes during the second half of the twentieth century. Climate Res. 19:193-212. Fuhrer J., 2003. Agroecosystem responses to combinations of elevated CO2, ozone, and global climate change. Agriculture, Ecosyst. Environ. 97:1-20. Galloway J. N., F. J. Dentener, D.G. Capone, E.W. Boyer, R.W. Howarth, S.P. Seitzinger, G.P. Anser, C.C. Cleveland, P.A. Green, E.A. Holland, D.M. Karl, A.F. Michaels, J.H. Porter, A.R. Townsend, C.J. Vrsmarty, 2004. Nitrogen cycles: past, present, and future. Biogeochem. 70:153-226. Ghaffari A., H.F. Cook, H.C. Lee, 2002. Climate change and winter wheat management: A modelling scenario for South-Eastern England. Clim. Change 55:509-533. Giorgi F., X. Bi, J. Pal, 2004. Mean, interannual and trends in a regional climate change experiment over Europe. II: Climate Change scenarios (2071-2100). Clim. Dyn. 23:839858. Grime J.P., V.K. Brown, K. Thompson, G.J. Masters, S.H. Hillier, I.P. Clarke, A.P. Askew, D. Corker, J.P. Kielty, 2000. The response of two contrasting limestone grasslands to simulated climate change. Science 289:762-765. Hayes M.J., M. Dubrovsk, M. Trnka, M.D. Svoboda, D.A. Wilhite, Z. alud, D. Semerdov, 2005. Application of Drought Indices to Assess Drought Conditions in Changed Climate. In: AGU fall meeting, San Francisco, 5-9 December 2005, poster abstracts available: http://www.agu.org/meetings/ fm05/waisfm05.html. Helms M., B. Bchele, U. Merkel, J. Ihringer, 2002. Statistical analysis of the flood situation and assessment of the impact of diking measures along the Elbe (Labe) river. J. Hydrol. 267:94-114. Hilden M. and Lethtonen H., 2005. The practice and process of adaptation in Finnish agriculture. FINADAPT Working paper 5, Helsinki, Finnish Environment Institute Mimeographs 335. Jones, G.V., M.A. White, O.R. Cooper, K. Storchmann, 2005. Climate change and global wine quality. Clim. Change 73:319-343. Jones P.D. and Moberg A., 2003. Hemispheric and large scale surface air temperature variations: an extensive revision and an update to 2001. J. Clim. 16:206-223. Kaiser H.M., S.J. Riha, D.S. Wilks, D.G. Rossiter, R. Sampath, 1993. A farm-level analysis of economic and agronomic impacts of gradual climate warming. Am. J. Agric. Econ. 75:387-398. Kammann C., L. Grnhage, U. Grters, S. Janze, H.J. Jger, 2005. Response of aboveground grassland biomass to moderate long-term CO2 enrichment. Basic Appl. Ecol. 6:351-365. Kaukoranta T. and Hakala K., 2008. Impact of spring warming on sowing times of cereal, potato and sugar beet in Finland. Agric. Food Sci., in press. Kenny, G.J., P.A. Harrison, J.E. Olesen, M.L. Parry, 1993. The effects of climate change on land suitability of grain maize, winter wheat and cauliflower in Europe. Eur. J. Agron. 2:325-238. Kimball, B.A., K. Kobayahsi, M. Bindi, 2002. Responses of agricultural crops to free-air CO2
307
enrichment. Adv. Agron. 77:293-368. Kjellstrm E., 2004. Recent and future signatures of climate change in Europe. Ambio 33:193198. Klein Tank A.M.G. and Knnen G.P., 2003. Trends in indices of daily temperature and precipitation extremes in Europe. J. Clim. 16:3665-3680. Klein Tank A.M.G. et al. 2002. Daily dataset of 20th-century surface air temperature and precipitation series for the European Climate Assessment. Int. J. Clim. 22:1441-1453. Lehner B., Dll, P., Alcamo, J., Henrichs, H. and Kaspar, F. 2006, Estimating the impact of global change on flood and drought risks in Europe: a continental, integrated analysis, Clim. Change 75:273-299. Leishman M.R., G.J. Masters, I.P. Clarke, V.K. Brown, 2000. Seed bank dynamics: the role of fungal pathogens and climate change. Func. Ecol. 14:293-299. Lobell D.B. and Field C.B., 2007. Global scale climate-crop yield relationships and the impacts of recent warming. Env. Res. Lett. 2:014002. doi: 10.1088/1748-9326/2/1/014002. Lscher A., M. Daepp, H. Blum, U.E. Hartwig, J. Nsberger, 2004. Fertile temperate grassland under elevated atmospheric CO2-role of feed-back mechanisms and availability of growth resources. Eur. J. Agron. 21:379-398. Maracchi G., O. Sirotenko, M. Bindi, 2005. Impacts of present and future climate variability on agriculture and forestry in the temperate regions: Europe. Clim. Change 70:117-135. Marsh J., 2005. The implications of Common Agricultural Policy reform for farmers in Europe. Farm Pol. J. 2 (2):1-11. Meehl G.A. and Tebaldi C., 2004. More intense, more frequent, and longer lasting heat waves in the 21st Century. Science 305:994-997. Metzger M.J., M.D.A. Rounsevell, L. Acosta-Michlik, R. Leemans, D. Schrter, 2006. The vulnerability of ecosystem services to land use change. Agric. Ecosyst. Environ. 114:69 85. Metzger M.J., R.G.H. Bunce, R.H.G. Jongman, C.A. Mcher, J.W. Watkins, 2005. A climatic stratification of Europe. Global Ecol. Biogeogr. 14:549-563. Minguez M.I., M. Ruiz-Ramos, C.H. Daz-Ambrona, M. Quemada, F. Sau, 2007. First-order impacts on winter and summer crops assessed with various high-resolution climate models in the Iberian peninsula. Clim. Change 81 (suppl 1):343-355. Moss B., D. Mckee, D. Atkinson, S.E. Collings, J.W. Eaton, A.B. Gill, I. Harvey, K. Hatton, T. Heyes, D. Wilson, 2003. How important is climate? Effects of warming, nutrient addition and fish on phytoplankton in shallow lake microcosms. J. Appl. Ecol. 40:782792. OECD, 2004. Analysis of the 2003 CAP reform. OECD publications, Paris. Olesen J.E. and Bindi M., 2002. Consequences of climate change for European agricultural productivity, land use and policy. Eur. J. Agron. 16:239-262. Olesen J.E., T. Jensen, J. Petersen, 2000. Sensitivity of field-scale winter wheat production in Denmark to climate variability and climate change. Clim. Res. 15:221-238. Olesen J.E., T.R. Carter, C.H. Diaz-Ambrona, S. Fronzek, T. Heidmann, T. Hickler, T. Holt, M.I. Minguez, P. Morales, J. Palutikov, M. Quemada, M. Ruiz-Ramos, G. Rubk, F. Sau, B. Smith, M. Sykes, 2007. Uncertainties in projected impacts of climate change on European agriculture and ecosystems based on scenarios from regional climate models. Clim. Change 81 (suppl. 1):123-143. Pal J.S., F. Giorgi, X.Q. Bi, 2004. Consistency of recent European summer precipitation trends and extremes with future regional climate projections. Geophys. Res. Let. 31:L13202.
308
Peltonen-Sainio P., K. Hakala, L. Jauhiainen, K. Ruosteenoja, 2008. Comparing regional risks in producing turnip rape and oilseed rape - Impacts of climate change and breeding. Acta Agric. Scand., in press. Porter J.R., R.A. Leigh, M.A. Semenov, F. Miglietta, 1995. Modelling the effects of climatic change and genetic modification on nitrogen use by wheat. Eur. J. Agron. 4:419-429. Reilly J. and Schimmelpfennig D., 1999. Agricultural impact assessment, vulnerability, and the scope for adaptation. Clim. Change 43:745-788. Rivington M., K.B. Matthews, G. Bellocchi, K. Buchan, C.O. Stckle, M. Donatelli, 2006. An integrated modelling approach to conduct multi-factorial analyses on the impacts of climate change on whole-farm systems. Environ. Model. Software 22:202-210. Rochette P., G. Belanger, Y. Castonguay, A. Bootsma, D. Mongrain, 2004. Climate change and winter damage to fruit trees in eastern Canada. Can. J. Plant Sci. 84:1113-1125. Rossi F., 2006. Orchard-atmosphere exchange processes and sustainable management. In Crops: growing, quality and biotechnology. (R.Dris (ed)). WFL Publ. Finland. p. 25-62. Rounsevell M.D.A., F. Ewert, I. Reginster, R. Leemans, T.R. Carter, 2005. Future scenarios of European agricultural land use. II. Projecting changes in cropland and grassland. Agric. Ecosyst. Environ. 107:117-135. Rustad L.E., J.L Campbell, G.M. Marion, R.J. Borby, M.J. Mitchell, A.E. Hartley, J.H.C. Cornelissen, J. Gurevitch, 2001. A meta-analysis of the response of soil respiration, net nitrogen mineralization, and aboveground plant growth to experimental ecosystem warming. Oecologica 126:543-562. Salinari F., S. Giosue, F.N. Tubiello, A. Rettori, V. Rossi, F. Spanna, C. Rosenzweig, M.L. Gullino, 2006. Downy mildew (Plasmopara viticola) epidemics on grapevine under climate change. Global Change Biol. 12:1299-1307. Santos F.D., K. Forbes, R. Moita, (eds.) 2002. Climate change in Portugal: Scenarios, Impacts and Adaptation Measures. SIAM project report, Gradiva, Lisbon, Portugal, 456 pp. Schr C. and Jendritzky G., 2004. Climate change: Hot news from summer 2003. Nature 432:559-560. Schr C., P.L. Vidale, D. Lthi, C. Frei, C. Hberli, M.A. Liniger, C. Appenzeller, 2004. The role of increasing temperature variability in European summer heatwaves. Nature 427:332336. Schrter D., W. Cramer, R. Leemans, I.C. Prentice, M.B. Arajo, N.W. Arnell, A. Bondeau, H. Bugmann, T.R. Carter, C.A. Gracia, A.C. de la Vega-Leinert, M. Erhard, F. Ewert, M. Glendining, J.I. House, S. Kankaap, R.J.T. Klein, S. Lavorel, M. Lindner, M.J. Metzger, J. Meyer, T.D. Mitchell, I. Reginster, M. Rounsevell, S. Sabat, S. Sitch, B. Smith, J. Smith, P. Smith, M.T. Sykes, K. Thonicke, W. Thuiller, G. Tuck, S. Zaehle, B. Zierl, 2005. Ecosystem service supply and vulnerability to global change in Europe. Science 310:13331337. Seneviratne S.I., D. Lthi, M. Litschi, C. Schr, 2006. Land-atmosphere coupling and climate change in Europe. Nature, 443: 205-209. DOI: 10.1038/nature05095. Sims R.E.H., A. Hastings, B. Schlamadinger, G. Taylor, P. Smith, 2006. Energy crops: current status and future prospects. Global Change Biol. 12:2054-2076. Sinclair T.R. and Muchow R.C., 2001. Systems analysis of plant traits to increase grain yield on limited water supplies. Agron. J. 93:263-270. Sirotenko O.D., H.V. Abashina, V.N. Pavlova, 1997. Sensitivity of the Russian agriculture to changes in climate, CO2 and tropospheric ozone concentrations and soil fertility. Clim. Change 36:217-232. Smith J.B. and Lenhart S.S., 1996. Climate change adaptation policy options. Clim. Res. 6:193-201.
309
Smith J., P. Smith, M. Wattenbach, S. Zaehle, R. Hiederer, R.J.A. Jones, L. Montanarella, M.D.A. Rounsevell, I. Reginster, F. Ewert, 2006. Projected changes in mineral soil carbon of European croplands and grasslands, 1990-2080. Global Change Biol. 11:2141-2152. Szinell C.S., A. Bussay, T.Szentimrey, 1998. Drought Tendencies in Hungary. International Journal of Climatology 18: 1479-1491. DOI: 10.1002/(SICI)1097-0088(199804)18:5 Tavakkoli A.R. and Oweis T.Y., 2006. The role of supplemental irrigation and nitrogen in producing bread wheat in the highlands of Iran. Agric. Water Manage. 65:225-236. Thornley J.H.M. and Cannell M.G.R., 1997. Temperate grassland responses to climate change: an analysis using the Hurley pasture model. Ann. Bot. 80:205-221. Trenberth K.E., P.D. Jones, P. Ambenje, R. Bojariu, D. Easterling, A.K. Tank, D. Parker, F. Rahimzadeh, J.A. Renwick, M. Rusticucci, B. Soden, P. Zhai, 2007. Observations: Surface and atmospheric climate change. In: Climate change 2007: The physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tigor, H.L. Miller, (eds)). Cambridge University Press, Cambridge, UK and New York, NY, USA. p. 235-336. Trnka M., J. Kysel, M. Mon, M. Dubrovsk, 2008. Changes in the Central European Soil Moisture Availability and circulation patterns in 1881-2005, International Journal of Climatology, 28: (In print). Tubiello F.N., M. Donatelli, C. Rosenzweig, C.O. Stockle, 2000. Effects of climate change and elevated CO2 on cropping systems: model predictions at two Italian locations. Eur. J. Agron. 13:179-189. Tuck G., M.J. Glendining, P. Smith, J.I. House, M. Wattenbach, 2006. The potential distribution of bioenergy crops in Europe under present and future climate. Biomass Bioener. 30:83-197. van der Ploeg R.R. and Schweigert P., 2001. Elbe river flood peaks and postwar agricultural land use in East Germany. Naturwissenschaft 88:522-525. van Ittersum,M.K., S.M. Howden, S. Asseng, 2003. Sensitivity of productivity and deep drainage of wheat cropping systems in a Mediterranean environment to changes in CO2, temperature and precipitation. Agric. Ecosyst. Environ. 97:255-273. van der Schrier G., K.R. Briffa, P.D. Jones, T.J. Osborn, 2006. Summer Moisture Variability across Europe; Journal of Climate; 19: 2818-2834; DOI: 10.1175/JCLI3734.1 van der Schrier G., D. Efthymiadis, K.R. Briffa, P.D. Jones, 2007. European Alpine moisture variability 1800-2003; International Journal of Climatology 27: 415-427. DOI: 10.1002/joc.1411 Wadsworth R. and Swetnam R., 1988. Modelling the impact of climate warming at the landscape scale: will bench terraces become economically and ecologically viable structures under changed climates? Agric. Ecosyst. Environ. 68:27-39. Wolf J. and van Oijen M., 2003. Model simulation of effects of changes in climate and atmospheric CO2 and O3 on tuber yield potential of potato (cv. Bintje) in the European Union. Agric. Ecosyst. Environ. 94:141-57. Ziska L.H. and Bunce J.A., 1997. Influence of increasing carbon dioxide concentration on the photosynthetic and growth stimulation of selected C4 crops and weeds. Photosyn. Res. 54:199-208.
310
7.12 ANNEX I Overview of the respondents of the survey that were essential in compiling of the report.
Country Austria Zone(s) CON, ALS PAN CON, MDM, MDN CON Lead author J. Eitzinger
V. Kazandjiev
Contributors S. Thaler, H.Formayer, G. Kubu
Affiliation of the lead author University of Applied Life Sciences in Vienna, Meteorological and Hydrological Service
Contact josef.eitzinger@boku.a c.at vkazandjiev@abv.bg vucetic@cirus.dhz.hr
Bulgaria Croatia
V. Vueti M.Trnka L.Bartoov, M.Dubrovsk, P.Hlavinka, Z.alud
Czech Rep.
Mendel University of Agriculture and Forestry Aarhus University
mirek_trnka@yahoo.co m
Denmark Estonia Finland
ATN, CON NEM BOR
J.E. Olesen A. Ingver P. PeltonenSainio B. Seguin K. Hakala
JorgenE.Olesen@agrsc i.dk Anne.Ingver@jpbi.ee Pirjo.PeltonenSainio@mtt.fi bernard.seguin@avign on.inra.fr
MTT Finland INRA
Agrifood
France
Germany
Greece Hungary Ireland Italy
MDN, ALS, MDM, ATC, LUC ATN ATC CON MDS PAN ATC MDS, MDN
K.C. Kersebau m L. Toulios Z. Zoltan A. Donnelly F. Rossi
W. Mirschel, K.-O. Wenkel
Leibniz-Centre for Agricultural Landscape Research Nagref Hungarian Met. Service Trinity College CNR-IBIMET
ckersebaum@zalf.de
ltoulios@nagref.gr dunkel.z@met.hu donnelac@tcd.ie f.rossi@ibimet.cnr.it
M. Jones S. Orlandini, V. Di Stefano, L. Mariani, O. Failla, O. Facini, T. Maggiore
Latvia Lithuania Netherlands Netherlands
NEM NEM ATC ATC
A. Jermuss S. Lazauskas R. Rtter J. Verhagen
Zemkopibas Instituts Lithuanian Institute of Agriculture MTT Agrifood Finland WUR
aivaram@inbox.lv sigislaz@lzi.lt reimund.rotter@mtt.fi Jan.Verhagen@wur.nl
311
Norway
Poland Romania Scotland Serbia
ALN, ATN, CON, BOR CON CON, PAN ATN PAN,
A.O. Skjelvg
Norwegian University of Life Sciences State Research Institute, Pulawy
arne.skjelvag@umb.no
J. Kozyra E. Mateescu M. Rivington B. Lalic
kuzyr@iung.pulawy.pl mateescu.elena@gmail .com m.rivington@macaulay .ac.uk branka@polj.ns.ac.yu
Macaulay Institute D. Michajlovic, M. Malesevic Faculty of Agriculture University of Novi Sad Slovak Hydrometeorological Instittue, Bratislava ITACYL UPM SLU
Slovakia
CON
P. Nejedlk A. Utset Suastegui A. Iglesias H. Eckerstee n P. Calanca
J. Takac B. Siska
Pavol.Nejedlik@shmu. sk UtsSuaAn@itacyl.es ana.iglesias@upm.es Henrik.Eckersten@vpe .slu.se pierluigi.calanca@art.a dmin.ch
Spain Spain Sweden
MDN MDS NEM
Switzerland
CON ALS
Agroscope ReckenholzTnikon, Research Station ART, Zurich
312
Annex 1. List of contributors
ANNEX 1. LIST OF CONTRIBUTORS

Name Josef Eitzinger Heimo Truhetz Sabina Thaler Werner Schneider Wolfgang Wagner Valentin Kazandjiev Vesellin Alexandrov Darko Voncina Marko Vucetic Visnja Vucetic Cedo Brankovic Lenka Bartosova Miroslav Trnka Zdenek Zalud Petr Hlavinka Eva Kocmankova Martin Dubrovsky Tomas Halenka Jaroslava Kalvova Martin Mozny Petr Stepanek Jorgen Eivind Olesen Tim Carter Stefan Fronzek Affiliation Universitaet fuer Bodenkultur, Vienna Wegener Center, Univ. Graz. Universitaet fuer Bodenkultur, Vienna Universitaet fuer Bodenkultur, Vienna Technical University, Vienna National Institute of Meteorology and Hydrology National Institute of Meteorology and Hydrology University of Zagreb Meteorological and Hydrological Service of Croatia Meteorological and Hydrological Service of Croatia Meteorological and Hydrological Service of Croatia Mendel University in Brno Mendel University in Brno Mendel University in Brno Mendel University in Brno Mendel University in Brno Institute of Atmospheric Physics, Prague Charles University, Prague Charles University, Prague Czech Hydrometeorological Institute Czech Hydrometeorological Institute Danish Institute for Agroecology Science, Foulum Finnish Environment Institute SYKE, Helsinki Finnish Environment Institute SYKE, Helsinki Denmark Finland Czech Republic Croatia Bulgaria Country Austria
313
Pirjo Peltonen-Sainio Heikki Tuomenvirta Kari Tiilikkala Kaija Hakala Emmanuel Cloppet Bernard Seguin Kurt Christian Kersenbaum Wilfried Mirschel Karl-Otto Wenkel Martin Wegehenkel Hans Friesland Nicolas Dalezios Christos Domenikiotis Dimitrios Bampzelis Emmanouil Tsiros Efrossini Kanellou Leonidas Toulios Szabolcs Bella Zoltan Dunkel Janos Mika Maria Pustay Nicholas M. Holden Simone Orlandini Fabio Micale Federica Rossi Giampiero Maracchi Valentina di Stefano Osvaldo Facini
Agrifood Research, Jokioinen Finnish Meteorological Institute Agrifood Research, Jokioinen MTT Agrifood Research Jokioinen Meteo France, Toulouse INRA, Avignon Leibniz-Centre for Agricultural Landscape Research, Muencheberg Leibniz-Centre for Agricultural Landscape Research, Muencheberg Leibniz-Centre for Agricultural Landscape Research, Muencheberg Leibniz-Centre for Agricultural Landscape Research, Muencheberg German Weather Service University of Thessaly, Volos University of Thessaly, Volos University of Thessaly, Volos University of Thessaly, Volos University of Thessaly, Volos National Agricultural Research Foundation Larissa Hungarian Meteorological Service Hungarian Meteorological Service Hungarian Meteorological Service Hungarian Meteorological Service University College, Dublin University of Florence Joint Research Centre. Ispra CNR IBIMET, Bologna CNR IBIMET, Florence University of Florence IBIMET CNR, Bologna Ireland Italy Hungary Greece Germany France
314
Vincenzo Levizzani Domenico Vento Luigi Mariani Vittorio Marletto Fabio Maselli Alessandro Chiaudani Jan Verhagen
ISAC CNR, Bologna Unit di Ricerca per la Climatologia e la Meteorologia applicate all'Agricoltura-CRA University of Milan ARPA Emilia Romagna, Bologna CNR-IBIMET, Florence ARPA Veneto, Padova ASgrosystem Research Plant Research International, Wageningen Soil Science Centre, ALTERRA, Wageningen University of Norway Norvegian Crop Research Institute, Aas Institute of Meteorology and Water Management, Krakov Institute of Soil Science and Plant Cultivation, Pulawy Institute of Meteorology and Water Management, Krakov Oporto University Romanian Meteorological Service Romanian Meteorological Service Institute for field and vegetable crops, Novi Sad Hydrometeorological Service of Serbia University of Novi Sad University of Novi Sad Slovak Hydrometeorological Institute Soil Science and Conservation research Institute Slovak Agricultural University, Nitra Technical University of Zvolen Slovakia Serbia Portugal Romania Poland Norway Netherlands
Reimund Paul Rotter Arne Oddvar Skejvg Tor Haakon Sivertsen Malgorzata KepinskaKasprzak Jerzy Kozyra
Piotr Struzik
Ana Monteiro Gheorghe Stancalie Elena Mateescu Miroslav Malesevic Zoran Krajinovic Dragutin Mihailovic Branislava Lalic Pavol Nejedlik Jozef Takac Bernard Siska Jaroslav Skvarenina
315
Lucka Kaifez Bogataj Andreja Susnik Antonio Mestre Barcelo Federico Sau Margarita Ruiz-Ramos Roser Botey Jose Luis Garcia-Merayo Ernesto Rodrguez Camino Pierluigi Calanca Mark Danson Robin Matthews
University of Ljubljana Ministry of Environment and spatial Planning, Ljubljana Meteorological State Agency, Madrid Politechnic University of Madrid Politechnic University of Madrid Meteorological State Agency Meteorological State Agency Meteorological State Agency Environmental Protection and Climate Natural Sources, Zurich Telford University of Environmental Systems, Manchester Macaulay Institute
Slovenia
Spain
316
ANNEX 2. LIST OF QUESTIONNAIRES QUESTIONNAIRE 1 - Agroclimatic Indices and Models 1. Please name the agroclimatic indices (incl. statistical models) which are used in your country operationally for (please indicate if for forecasting (F), nowcasting(N) or pastcasting(P), which time step(T), spatial realization and resolution(S), for which aim they are used(A), main providers (PR) and users (US) : 2. Please name process-oriented models (crop models) which are used in your country operationally for assessment of any impacts of weather and climate variability (please indicate F, N or P, time step(T), spatial realization and resolution(S), for which aim they are used(A), (e.g. irrigation scheduling, yield estimate etc.), main providers (PR) and users(US) : 3. Please name the agroclimatic indices (incl. statistical models) which were used in your country as research tools in order to assess impacts of climate change and variability for (please indicate for which aim they were used (A) (e.g impacts on yield, water balance, yield quality, production technique, adaptation options, spatial resolution and realization etc..). 4. Please name process-oriented models (crop models) which were used in your country as research tools in order to assess impacts of climate change and variability for (please indicate for which aim they were used (A) (e.g impacts on yield, water balance, yield quality, production technique, adaptation options, spatial resolution and realization, etc..) : 5. Please indicate which model outputs in which time step of the named process oriented models are by your expertise most useful in order detect the impact on climate change and variability regarding : 6. Please indicate the main limitations in order to apply process-oriented models for operational use in your country. For input data, please name which inputs. 7. Please name if any model applications in your country (operational or research) were done with combination of GIS and/or Remote Sensing. 8. Please name if any model applications in your country (operational or research) were done beyond the national scale (e.g. bilateral, regional), and name involved countries (or regions). 9. Any other aspects to be reported (optional) QUESTIONNAIRE 2 - Trends in Agroclimatic Indices and Model Outputs
317
1. Please provide any information on past and current meteorological as well as remote sensing data applied in your country/region: 2. Please indicate which numerical weather and regional climate models or their related outputs are used in your country/region: 3. Please specify any weather generators applied in agrometeorological studies and/or services in your country/region: 4. Please name and shortly describe any homogenization tests/procedures applied to meteorological and agricultural related time series in your country/region: 5. Please provide any information on the statistical methods for analyses of meteorological and simulation model output related time series in order to evaluate mean and variability patterns as well as to determine respective trends in agroclimatic indices and simulation model outputs 6. Please list any possible (general/specific) constraints during the implementation of the WG2 tasks 7. Any additional information/problems/questions related to the COST 734 activities and especially the WG2 ones QUESTIONNAIRE 2.1 - Satellite Data Records Survey The national members of COST 734 are kindly asked to fill in the survey related to WG2 Remote Sensing Subgroup task as below: 1. climate/biophysical product 2. satellite/instrument 3. produced by 4. operational/experimental 5. since 6. responsible company/other 7. Country 8. Area covered QUESTIONNAIRE 3 - Climate change scenarios 1. Please name which types of climate change scenarios were used so far (in the last 10 years) in your country for research or operational purposes (obligatory: provide references) and give short overview regarding the purpose of investigation. Which strengths and weaknesse of scenarios were found in your country? Who was the provider of scenarios (models)? 2. Please name which downscaling techniques (spatial and temporal) were used in your country (again provide references) for assessment of
318
3.
4.
5.
6.
7.
possible impacts of climate variability (please indicate time step and spatial resolution). Please name which Regional climate models (if any) were used in your country (again provide references) for assessment of possible impacts of climate variability on agriculture? Please name which Sea level rise scenarios (if any) were used in your country (again provide references) for assessment of possible impacts on agriculture? Please give information regarding any attemps in your country to analyse the risk of weather and climate extremes in future regional forecast scenarios (again provide references) Was your country in any way involved in EU PRUDENCE, ENSEMBLES or similar projects? If so please give a description of your role and possible implementations (with references). Are you familiar with high-resolution PRUDENCE climate change scenarios for 2071-2100 for Europe? Were they already used in any studies of impacts? Please identify critical research needs that address development and use of climate change scenarios in your country. In your opinion how can WG3 of COST 734 help in meeting your needs?
QUESTIONNAIRE 4 Risk Assessment and Foreseen Impacts on Agriculture 1. Main vulnerabilities of crops and cropping systems. Please provide a description on the main vulnerabilities of crops and cropping systems in your country to current climatic variability as well as for projected climate change (provide references, if possible). Which crops and cropping systems are most vulnerable? 2. Critical thresholds of climatic suitability and climate change. Please provide information on critical thresholds of climatic suitability and climate change for crops and cropping systems in your country. The thresholds may be provided for both changes in climatic variables, agroclimatic indices and agroecosystem indicators. The rationale behind the thresholds should be briefly described (provide references, if possible). 3. Assessments and studies of climate change impacts. Please provide information on assessments of climate change impacts on crops and cropping systems in your country. This should include information on which types of scenario analyses were performed (sensitivity analyses, GCM or RCM based analyses), and whether both CO2 and climate changes were considered. Please provide information on the timescale
319
4.
5.
6.
7.
8.
9.
(projected year) of the scenarios applied in impact assessment. References on studies of climate change impacts on agricultural and horticultural crops in your country should be included (if reports are available on the web, please provide the web address (url)). Adaptation options. Which adaptation options within crops, cropping systems and crop man-agement would you consider the most appropriate for dealing with climate change in your country, both in the short and long term. Remember to consider also adaptation in related sectors (e.g. the water sector), which will also affect adaptation in agriculture. Please give references on studies of adaptation of crop and cropping system management to climate change in your country (if reports are available on the web, please provide the web address (url)). Observed adaptation. Are there already visible implemented adaptation measures related to climate change in your country (provide references, if possible)? National impact assessments, adaptation strategies and awareness. Please provide information on the status and plans for national impact assessments and adaptation strategies in your country. These assessments and studies often cover many sectors, and the answer should focus on the role of agriculture within this picture and identified relations with other sectors. What is the awareness and attitude to climate change among policy makers and stakeholders related to agriculture? Dissemination of information and recommendations. Please provide information on dissemi-nation of information and recommendations related to climate change impacts and adaptation in agriculture to farmers, extension services, plant breeders, policy makers and other stakeholders in your country. Warning systems. Please provide a description of any warning systems in your country that is designed to cope with climate/weather (interannual) variability or otherwise could be used for re-ducing vulnerability to climatic change. Contact persons for climate change impacts and adaptation. Please provide names and ad-dresses (including e-mail) of persons in your country, who are working on climate change impacts and adaptation for crops and cropping systems.
320
Finito di stampare presso la Copisteria Sangallo - Firenze (Italia) nel mese di Luglio 2008

4

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

4

Uploaded by

Copyright:

Available Formats

COST ACTION 734 IMPACT OF CLIMATE CHANGE AND VARIABILITY ON EUROPEAN AGRICULTURE

ESF provide the COST Office through an EC contract

COST is supported by the EU RTD Framework programme

ESSEM Earth System Science and Environmental Management

PREFACE Pavol Nejedlik and Simone Orlandini............................................................................1

4.6 4.7 4.8 4.9

ANNEX 1. LIST OF CONTRIBUTORS................................................................313 ANNEX 2. LIST OF QUESTIONNAIRES ............................................................317

PREFACE Pavol Nejedlik and Simone Orlandini

1. Climate change, impacts on agriculture, sustainability

Atlantic N** P** P** P** P** N* N* N* na P** N**

Central N*** N* P** P** N* N* N* N** P* N** P* P*to* N N** P* N** na

P** P** P*to* na N P*** P** P*to* N N* P** P*

1. Climate change, impacts on agriculture, sustainability

1. Climate change, impacts on agriculture, sustainability

1. Climate change, impacts on agriculture, sustainability

1. Climate change, impacts on agriculture, sustainability

2. Agroclimatic indices and simulation models

2. Agroclimatic indices and simulation models

2. Agroclimatic indices and simulation models

2. Agroclimatic indices and simulation models

Figure 2.1: Operational use of agrometeorological indices in European countries

Figure 2.2: Use of agrometeorological indices in research in European countries

2. Agroclimatic indices and simulation models

2. Agroclimatic indices and simulation models

2. Agroclimatic indices and simulation models

Soil water content for top 10 cm Index of hydrometeorological drought NPET

Water balance components

Soil moisture content VHI

daily, weekly monthly, yearly daily, weekly, monthly Monthly

from site specific to national 10x10 km to national National

CZ, F, I, NO, PL, SI, SK, SR DE , FI, IT, RO, SI, SR GR

on users request Monthly

Precipitation totals and anomalies Rainfall percentile

weekly, monthly, yearly monthly

Regional, National National

AT, I, PL, SI, SR HR

ES, I, SI, SR, GR NL, SK

2. Agroclimatic indices and simulation models

F P monthly, thirty, sixty and ninety days-daily calculation monthly decade

Regional Regional, national

Precipitation totals anomalies Palmer`s Z index

Regional national station network

estimating affected regions estimating affected regions

(F) forecasting, (N) nowcasting, (P) pastcasting

Daily monthly, growing period

temperature towards normal days with heat stress

national station network

Number of days with heat stress

AT, HR, SI, SR

daily, decade and monthly

national station network

Number of days with heat stress

estimation of regional frost risk frost occurrence

Daily Daily Daily

Snow cover National National 10x10 Regional

DE, SI HR, SI A, FI, SI, SR IT SK

(F) forecasting, (N) nowcasting, (P) pastcasting

2. Agroclimatic indices and simulation models

10x10 km 10x10 km National National National National national station network

FI, GR FI DE DE, SI, SK, HR DE DE, SK AT, HR, SR

(F) forecasting, (N) nowcasting, (P) pastcasting

Greeness index Phenomaps Crop yields Phenological stages

unregula r Seasonal ly Daily, unregula r Daily

AT I, SI All AT, CH, CZ, DE, HR, RO, SI, SK All

Atlantic N P P P P** N* N* N* na P N

Central N*** N* P P N* N* N* N** P* N** P* Pto N N** P* N** na

P P Pto na N P* P Pto N N* P** P*