land degradation & development

Land Degrad. Develop. 21: 139144 (2010) Published online 16 April 2009 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/ldr.918

IS RAINFALL EROSIVITY INCREASING IN THE MEDITERRANEAN IBERIAN PENINSULA?

M. DE LUIS*, J. C. GONZALEZ-HIDALGO AND L. A. LONGARES
Department of Geography, University of Zaragoza, C/Pedro Cerbuna 12, 50009 Zaragoza, Spain Received 4 November 2008; Revised 19 January 2009; Accepted 30 January 2009

ABSTRACT The risk of erosion and desertication is one of the main environmental concerns in the Mediterranean Iberian Peninsula. Changes in precipitation are expected in Mediterranean areas because of climate change, but predictions are not certain. For this reason, dense precipitation databases are required to explore observed changes in the amount, concentration and variability of precipitation, to gain a clearer understanding of the dynamics involved in the main climatological agent of erosion. For this study, we took the recently developed MOPREDAMES dataset, which includes 1113 complete and homogeneous monthly rainfall series from the Mediterranean fringe of the Iberian Peninsula (IP) covering the period 19512000. These were used to calculate and analyse trends in Total Annual Precipitation (Pt), the Precipitation Concentration Index (PCI) and Modied Fourier Index (MFI). Our results show that, although there were decreases in annual rainfall, increases in the concentration of precipitation also predominated in the Mediterranean Iberian Peninsula during the period 1951 2000. However, spatial variability of these trends is high, and changes in rainfall erosivity exhibit a complex spatial pattern. Thus, decreases in rainfall erosivity are detected under semiarid conditions (Central Ebro basin and South East IP), while increases mainly occur in dry and subhumid areas. We present a detailed spatial description of the results and discuss their implication for the risk of erosion and desertication in different regions of the study area. Copyright # 2009 John Wiley & Sons, Ltd.
key words: rainfall erosivity; precipitation trends; Precipitation Concentration Index; Modied Fourier Index; Mediterranean Spain

INTRODUCTION Soil erosion and desertication are one of the main environmental problems in Mediterranean climate areas, and Mediterranean regions of Spain, in particular, have been described as the most threatened area in Europe (Vallejo et al., 2005). Most of the soil erosion in the Mediterranean areas is due to rainfall (splash and wash), but soil erosion by water is a complex phenomenon in which there is no exact relationship between soil erosion and the total amount of rainfall. Neither is there one with the intensity of rainfall and its distribution in time (Kirkby and Neale, 1987). Because of global warming, a higher risk of desertication has been predicted, due to increased aridity around the Mediterranean basin (Trenberth et al., 2007), but there is also great uncertainty about the future development of soil erosion by water, because the reliability of model outputs for precipitation are less accurate than those for temperature (Christensen et al., 2007).
* Correspondence to: M. De Luis, Department of Geography, University of Zaragoza, C/Pedro Cerbuna 12, 50009 Zaragoza, Spain. E-mail: mdla@unizar.es

As far as possible, regional analysis of potential erosion should take into account the variability of rainfall in space and time, and this should be achieved by using dense spatial information. At present, there are many datasets on a global or continental scale (Klein-Tank et al., 2002; Wijngaard et al., 2003), but these are not useful at sub-regional level because of the low density of observations. This situation is especially critical in areas with high variability in rainfall, such as those in the Mediterranean climate. Different indices of rainfall aggressivity (erosivity) have been proposed for analysing soil erosion, and the most appropriate ones seem to be those relating to soil erosion with kinetic energy of rainfall, such as the well-known EI30 (Wischmeier, 1959). However, there is scarce availability of high quality datasets on a regional scale, because EI30 requires rainfall data at intervals of one minute (Loureiro and Coutinho, 2001). To avoid this problem, other indices based on monthly data averages have been proposed, such as Fourier Index (FI) and its modication by Arnoldus (1980) (MFI). Agreement between FI or MFI with the USLE R factor (rainfall aggressivity factor) has been described on many occasions (Renard and Freimund, 1994; Gabriels, 2001; Loureiro and Coutinho, 2001; Diodato and Bellocchi,

Copyright # 2009 John Wiley & Sons, Ltd.

140

M. DE LUIS ET AL.

2007), and, as consequence, they are commonly used as the input aggressivity factor in the development of regional models (Gregori et al., 2006). Basically, the Modied Fournier Index (MFI) may be expressed as the product of Total Annual Precipitation (Pt) and monthly precipitation concentration (PCI) (MFI Pt PCI) (Apaydin et al., 2006). According to this relation, rainfall erosivity (MFI) is more intense where there are high values of precipitation concentration (high PCI) and Total Annual Precipitation (Pt). In this paper, we analyse the variability in space and time of the MFI, the Precipitation Concentration Index (PCI) and total precipitation (Pt) trends by using a dense precipitation database recently developed for the Mediterranean fringe of Iberian Peninsula (i.e., the runoff contribution area to the Mediterranean Sea). The objectives of this paper are to analyse the relationship between trends of Pt, MFI and PCI to describe the evolution of rainfall aggressivity during 19512000 in environments subject to a high erosion risk, and to look for spatial distribution patterns. MATERIAL AND METHODS We used the monthly precipitation database (MOPRE DAMES) recently presented by Gonzalez-Hidalgo et al. (2009) and De Luis et al. (2009). MOPREDAMES was produced after analysing the archives of National Meteorological Agency of Spain, where large amounts of monthly precipitation data are stored. After being exhaustively processed for quality control and homogenisation, 1113 monthly precipitation stations were reconstructed, covering the period 19512000, with an overall density of 1 observatory per 150200 km2 (Figure 1).

This database is, at present, probably the most dense and the highest quality monthly precipitation database available for the western Mediterranean area and comprises ve hydrological divisions which are drained into the Medi terranean Sea: Eastern Pyrenees, Ebro, Jucar, Segura and Eastern Andalusia. The Modied Precipitation Index (MPI) (Fournier, 1960; Arnoldus, 1980) and Precipitation Concentration Index (PCI) (Oliver, 1980) were calculated on an annual basis for each station, according to the following equations:
12 X p2 i i1 12 P

MFI

Pt

PCI 

i1 12 P i1

p2 i 2 pi

with pi being the monthly precipitation at month i, and Pt the annual precipitation. The annual time series of MFI, PCI and Pt were used to test for trends and to detect changes during the second half of the 20th century. The intensity of observed changes was estimated by using linear regression techniques (Suppiah and Hennessy, 1998; De Luis et al., 2000; Gonzalez-Hidalgo et al., 2001). The spatial distribution of results (rate of change measured as the slope of linear models) is shown by spatial interpolation (Inverse Distance Weighted Methods) of observed trends at each of the 1113 stations (Ninyerola et al., 2007). Finally, we studied surfaces affected by different trends by using raster maps and area statistics obtained for each variable, using ArcGis 911 software. RESULTS The mean annual precipitation for the whole study area is 562 mm; however, values higher than 2000 mm y1 (the Pyrenees and most SW mountain areas) and lower than 200 mm y1 (extreme South East) have been observed (Figure 2). The trend analysis of Total Annual Precipitation shows an overall decrease in annual precipitation (Figure 3). The exception is the extreme SE, where a very complex spatial pattern emerges, and includes patches of both positive and negative trends. Thus, on an annual scale, precipitation diminished over 901 per cent of the territory, with a decrease in the value of the global mean annual precipitation of 124 per cent during the period 19512000. Precipitation concentration (PCI) exhibits a different pattern from those observed for Total Annual Precipitation. Higher values are observed in the south and south-eastern part of the study area, while lower values are found in the Central Pyrenees (Figure 4). The trend analysis also shows a different pattern, and a general increase in precipitation concentration (PCI) is observed (Figure 5). Thus, precipiLAND DEGRADATION & DEVELOPMENT, 21: 139144 (2010)

Figure 1. Study area. Precipitation stations and hydrological divisions of Eastern Spain. EBR, Ebro; EPY, Eastern Pyrenees; JUC, Jucar, SEG, Segura and EAN, Eastern Andalusia. Copyright # 2009 John Wiley & Sons, Ltd.

RAINFALL EROSIVITY IN MEDITERRANEAN IP

141

Figure 2. Mean annual rainfall during the period 19512000 in the study area.

Figure 4. Mean value of the Precipitation Concentration Index during the period 19512000 in the study area.

tation concentration has increased over 805 per cent of the territory, with a mean increase of 88 per cent during the period 19512000. Rainfall aggressivity (MFI) is relatively high in most parts of study area, with higher values observed in the Pyrenees, inner coastland of the Valencia region and Southern Andalusia (Figure 6). However, because of opposite results

from annual precipitation and precipitation concentration, complex patterns are observed when analysing trends in rainfall aggressivity (MFI) (Figure 7). Thus, we found increases in rainfall aggressivity mainly in the Central Pyrenees and extensive areas of the regions of Valencia and Andalusia, while decreases are mainly located in the Central Ebro basin.

Figure 3. Trends in rainfall amounts. Percentage of change related to the mean value for 19512000. Copyright # 2009 John Wiley & Sons, Ltd.

Figure 5. Trends in the Precipitation Concentration Index (PCI). Percentage of change related to the mean value for 19512000. LAND DEGRADATION & DEVELOPMENT, 21: 139144 (2010)

142

M. DE LUIS ET AL.

Figure 6. Mean value of the Modied Fournier Index during the period 19512000 in the study area.

DISCUSSION Erosion and desertication are key elements to understanding the implications of climate change in Mediterranean climate environments. In this regard, not only changes in total precipitation, but also in its distribution and

Figure 7. Trends in the Modied Fournier Index (MFI). Percentage of change related to the mean value for 19512000. Copyright # 2009 John Wiley & Sons, Ltd.

concentration, are the key to understanding and quantifying these processes on a regional scale. The decrease in rainfall trends around the Mediterranean basin has been related to global warming and changes in pressure gradients between the islands of the Azores and Iceland associated with the North Atlantic Oscillation (Houghton et al., 2001). However, in the aforementioned study, the average spatial density of observations was low, and therefore, spatial variability and transient areas might not be correctly identied. Furthermore, the most recent global revision covering the Mediterranean basin highlighted large spatial variations and did not detect any clear spatial pattern of precipitation trends during the period 19512000 (Norrant and Douguedroit, 2006). Our results, based on a high-density data-set (1113 time series) and covering the ve hydrological divisions of the eastern IP, describe high variability in precipitation regimes and trends, and constitute a new tool for detailed spatial analysis. Furthermore, they allow us to identify some general patterns not previously described. Thus, annual trends are characterised globally by a general decrease in precipitation, but an increase in precipitation concentration during the period 19512000. Indices based on monthly data average such as the Fourier Index and its modication by Arnoldus (1980) (MFI) are often used to quantify the nature of rainfall variability and its effects on the distribution of soil erosion. As has been recently described by Apaydin et al. (2006), MFI can be expressed as the product of Total Annual Precipitation and the Precipitation Concentration Index described by Oliver (1980). Consequently, opposite trends observed in total rainfall and precipitation concentration draw a complex pattern for the trends of rainfall aggressivity in the Mediterranean IP. The importance of precipitation seasonality has been highlighted previously. Langbein and Schumm (1958) suggested that the available energy for erosion and transport increases positively with the amount of annual rainfall up to about 300 mm, at which perennial and annual vegetation cover increases surface protection and limits soil erosion. On the other hand, Kirkby and Neale (1987) proposed a seasonal erosion model in which peak maximum precipitation, plant cover and soil erosion are out of phase. Thus, the vegetation dynamic emerges as a key factor in quantifying and interpreting the risk of erosion and desertication. It is widely accepted that precipitation concentration and seasonality play a leading role in vegetation dynamics. Moreover, rainfall seasonality is directly related to disturbances like forest res, which have a critical effect on erosion. Thus, on one hand, the frequency and extent of forest res in Mediterranean areas are directly related to summer water stress (Moreno, 1998; Prosper-Laget et al., 1998; Rambal and Hoff, 1998). On the other hand, winter,
LAND DEGRADATION & DEVELOPMENT, 21: 139144 (2010)

RAINFALL EROSIVITY IN MEDITERRANEAN IP

143

and mainly spring, rainfall is linked to ne fuel growth, this fact being critical in determining the occurrence (Bessie and Johnson, 1995) and spread (Viegas and Viegas, 1994) of forest res. Last, but not least, the regeneration capacity of ecosystems and restoration activities after summer res are highly dependent on seasonal rainfall, particularly in autumn (Bautista et al., 1996; Vallejo, 1996; De Luis et al., 2001). In short, the observed decrease in total precipitation should not be interpreted as a decrease in the risk of erosion, because the process of soil erosion is closely linked to the vegetation dynamic and to the probability of disturbances, such as forest res, occurring, and these are mainly linked to rainfall seasonality and concentration.

CONCLUSIONS Trend analysis of a dense monthly precipitation database in the Mediterranean fringe of Spain (western Mediterranean basin) indicates opposite behaviour between annual precipitation (P) and seasonal precipitation concentration (PCI). Thus, during the period 19512000, decreases in annual rainfall, but increases in precipitation concentration, predominated in the Mediterranean Iberian Peninsula. As a consequence, and due to the opposing inuence of precipitation components, trends in rainfall aggressivity (MFI) exhibit a complex spatial pattern. Is spite of this general behaviour, there is high spatial variability among these trends, and we attach detailed spatial maps of these precipitation components. Our results may contribute to a better understanding of the local dynamic of the main climatological agents of erosion in Mediterranean areas, and may also be valid for identifying areas subject to different erosion and desertication risks in Mediterranean ecosystems in Spain when creating suitable mitigation strategies in the assessment of global warming.

ACKNOWLEDGEMENTS This study was supported by the Spanish Government, CGL2005-04270 and CGL2007-65315-CO3-01/CLI and by Gobierno Regional de Aragon, Grupo de Investigacion Consolidado Clima. Agua, Cambio Global y Sistemas Naturales (BOA 69, 11-06-2007). We thank National Meteorological Agency of Spain for providing the original data set for the study area. We also thank Cindy Verkijk for her work in improving the English of this manuscript. REFERENCES
Apaydin H, Erpul G, Bayramin I, Gabriels D. 2006. Evaluation of indices for characterizing the distribution and concentration of precipitation: A case for the region of Southeastern Anatolia Project, Turkey. Journal of Hydrology 328: 726732. Copyright # 2009 John Wiley & Sons, Ltd.

Arnoldus HM. 1980. An approximation of the rainfall factor in the Universal Soil Loss Equation. In Assessments of Erosion, de Boodts M, Gabriels D (eds). John Wiley and Sons Ltd, Chichester 127132. Bautista S, Bellot J, Vallejo VR. 1996. Mulching treatment for postre soil conservation in a semiarid ecosystem. Arid Soil Research and Rehabilitation 10: 235242. Bessie WC, Johnson EA. 1995. The relative importance of fuels and weather on re behavior in subalpine forests. Ecology 76: 747762. Christensen JH, Hewitson A, Busuioc A, Chen X, Gao I, Held R, Jones RK, Kolli W-T, Kwon R, Laprise V, Magana Rueda L, Mearns CG, Menendez J, Raisanen A, Rinke A, Sarr A, Whetton P. 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, Quin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL (eds). Cambridge University Press: Cambridge and New York, NY. 848 940. De Luis M, Garca-Cano MF, Cortina J, Raventos J, Gonzalez-Hidalgo JC, Sanchez JR. 2001. Climatic trends, disturbances and short-term vegetation dynamics in a Mediterranean shrubland. Forest Ecology and Management 147: 2537. De Luis M, Gonzalez-Hidalgo JC, Longares LA, Stepanek P. 2009. Seasonal precipitation trends in Mediterranean Iberian Peninsula in second half of XX century. International Journal of Climatology. DOI: 10.1002/ joc.1778. De Luis M, Raventos J, Gonzalez-Hidalgo JC, Sanchez JR, Cortina J. 2000. Spatial analysis of rainfall trends in the region of Valencia (East Spain). International Journal of Climatology 20: 14511469. Diodato N, Bellocchi G. 2007. Estimating monthly (R) USLE climate input in a Mediterranean region using limited data. Journal of Hydrology 345: 224236. Fournier F. 1960. Climat et Erosion. PUF: Pars. Gabriels D. 2001. Rain erosivity in Europe. In: Man and Soil in the Third Millenium, III International Congress of European Society for Soil Conservation, 3143. Gonzalez-Hidalgo JC, de Luis M, Raventos J, Sanchez JR. 2001. Spatial distribution of seasonal rainfall trends in a western Mediterranean area. International Journal of Climatology 21: 843860. Gonzalez-Hidalgo JC, Lopez-Bustins JA, Stepanek P, Martn-Vide J, de Luis M. 2009. Monthly precipitation trends on the Mediterranean facade of the Iberian Peninsula during the second half of the 20th century (19512000). International Journal of Climatology DOI:10.1002/ joc.1780. Gregori E, Costanza M, Zorn G. 2006. Assessment and classication of climatic aggressiveness with regard to slope instability phenomena connected to hydrological and morphological processes. Journal of Hydrology 329: 489499. Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Xiaosu D. 2001. Climate Change 2001. The Scientic Basis. University Press Cambridge, Cambridge. Kirkby MJ, Neale RH. 1987. A Soil Erosion Model Incorporating Seasonal Factors, In International geomorphology, Vol. II. Gardiner V (ed.). John Wiley and Sons Ltd, Chichester: 179210. Klein-Tank AMG, Wijngaard JB, Konnen GP, Bohm R, Demaree G, Gocheva A, Mileta M, Pashiardis S, Hejkrlik L, Kern-Hansen C, Heino R, Bessemoulin P, Muller-Westermeier G, Tzanakou M, Szalai S, Pals dotir T, Fitzgerald D, Rubin S, Capaldo M, Maugeri M, Leitass A, Bukantis A, Aberfeld R, Van Engelen AFV, Forland E, Mietus M, Coelho F, Mares C, Razuvaev V, Nieplova E, Cegnar T, Lopez JA, Dahlstrom B, Moberg A, Kirchhofer W, Ceylan A, Pachaliuk O, Alexander LV, Petrovic P. 2002. Daily dataset of 20th-century surface air temperature and precipitation series for the European Climate Assessment. International Journal of Climatology 22: 14411453. Langbein WB, Schumm SA. 1958. Yield of sediment in relation to mean annual precipitation. Transaction American Geophysical Union 39: 10761084. Loureiro ND, Coutinho MD. 2001. A new procedure to estimate the RUSLE EI30 index, based on monthly rainfall data and applied to the Algarve region, Portugal. Journal of Hydrology 250: 1218. Moreno JM. 1998. Large Forest Fires. Backbuys Publishers, Leiden. The Netherlands. LAND DEGRADATION & DEVELOPMENT, 21: 139144 (2010)

144

M. DE LUIS ET AL. Trenberth KE, Jones PD, Ambenje P, Bojariu R, Easterling D, Klein-Tank A, Parker D, Rahimzadeh F, Renwick JA, Rusticucci M, Soden B, Zhai P. 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, Solomon S, Quin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL (eds). Cambridge University Press: Cambridge, and New York, NY. Vallejo VR. 1996. La restauracion de la cubierta vegetal en la Comunidad Valenciana. CEAM-Generalitat Valenciana: Valencia, Spain. Vallejo VR, Daz-Fierros F, de la Rosa D. 2005. Impactos sobre los recursos edacos. In Evaluacion Preliminar General de Los Impactos en Espana por Efecto del Cambio Climatico, Moreno JM (ed.). Ministerio de Medio Ambiente:Madrid. Viegas DX, Viegas MT. 1994. A relationship between rainfall and burned area for Portugal. International Journal of Wildland Fire 4: 1116. Wijngaard JB, Klein-Tank AM, Konnen GP. 2003. Homogeneity of 20th century European Daily Temperature and Precipitation series. International Journal of Climatology 23: 679692. Wischmeier WH. 1959. A rainfall erosion index from Universal Soil Loss Equation. Soil Science Society of American Journal 24: 323326.

Ninyerola M, Pons X, Roure JM. 2007. Monthly precipitation mapping of the Iberian Peninsula using spatial interpolation tools implemented in a Geographic Information System. Theoretical and Applied Climatology 89: 195209. Norrant C, Douguedroit A. 2006. Monthly and daily precipitation trends in the Mediterranean (19502000). Theoretical and Applied Climatology 83: 89106. Oliver JE. 1980. Monthly precipitation distribution: A comparative index. Professional Geographer 32: 300309. Prosper-Laget V, Douguedroit A, Guinot JP. 1998. A satellite index of risk of forest re occurrence in summer in the Mediterranean area. International Journal of Wildland Fire 8: 173182. Rambal S, Hoff C. 1998. Mediterranean ecosystems and re: The threats of global change. In Large Forest Fires, Moreno JM (ed.). Backbuys Publishers, Leiden. The Netherlands: 187213. Renard KG, Freimund JR. 1994. Using monthly precipitation data to estimate the r-factor in the revised USLE. Journal of Hydrology 157: 287306. Suppiah R, Hennessy KJ. 1998. Trends in total rainfall, heavy rain events and number of dry days in Australia, 19101990. International Journal of Climatology 18: 11411164.

Copyright # 2009 John Wiley & Sons, Ltd.

LAND DEGRADATION & DEVELOPMENT, 21: 139144 (2010)