Fish Communities and Fisheries

SYMPOSIUM PROCEEDINGS

Carlos Edwar de Carvalho Freitas Miguel Petrere Jr. Alexandre A. F. Rivas Don MacKinlay

International Congress on the Biology of Fish

Tropical Hotel Resort, Manaus Brazil, August 1-5, 2004

Copyright 2004 Physiology Section, American Fisheries Society All rights reserved
International Standard Book Number(ISBN) 1-894337-55-7

Notice This publication is made up of a combination of extended abstracts and full papers, submitted by the authors without peer review. The formatting has been edited but the content is the responsibility of the authors. The papers in this volume should not be cited as primary literature. The Physiology Section of the American Fisheries Society offers this compilation of papers in the interests of information exchange only, and makes no claim as to the validity of the conclusions or recommendations presented in the papers. For copies of these Symposium Proceedings, or the other 20 Proceedings in the Congress series, contact: Don MacKinlay, SEP DFO, 401 Burrard St Vancouver BC V6C 3S4 Canada Phone: 604-666-3520 Fax 604-666-0417 E-mail: mackinlayd@pac.dfo-mpo.gc.ca

Website: www.fishbiologycongress.org

ii

CONGRESS ACKNOWLEDGEMENTS
This volume is part of the Proceedings of the 6th International Congress on the Biology of Fish, held in Manaus, Brazil in August, 2004. Ten years have passed since the first meeting in this series was held in Vancouver, BC, Canada. Subsequent meetings were in San Francisco, California; Baltimore, Maryland; Aberdeen, Scotland; and again in Vancouver, Canada. From those meetings, colleagues from over 30 countries have contributed more than 2,500 papers to the Proceedings of over 80 Congress Symposia, all available for free viewing on the internet. We would like to extend our sincere thanks to the many people who helped us organize the facilities and program for this 6th Congress. The local arrangements team worked very hard to make this Congress a success. The leaders of those efforts were Vera Almeida Val, Adriana Chippari-Gomes, Nivia Pires Lopes and Maria de Nazare Paula Silva (Local Arrangements); Marcelo Perlingeiro (Executive Secretary) and Maria Angelica Laredo (Fund Raising). The enormous contribution of time and effort that was required has led to an unforgettable experience for the participants, thanks to the imagination, determination and dedication of this team. Many sponsors helped ensure the success of the meeting through both monetary and in-kind contributions, including: Fundao Djalma Batista, Honda, Merse, Cometais, Turkys Aquarium, Banco da Amaznia, Banco do Brasil, FUCAPI, SEBRAE/AM, IDAM/SEPROR, FAPEAM, SECT-AM, SUFRAMA, PETROBRS, CAPES, FINEP, CNPq, the Physiology Section of the American Fisheries Society, UFAM - Federal University of Amazonas, Fisheries and Oceans Canada and INPA - National Institute for Research in the Amazon. Travel arrangements were ably handled by Atlantic Corporate Travel (special thanks to Maria Espinosa) and Orcal Planet, and the venue for the meeting was the spectacular Tropical Hotel Conference Center in Manaus. The Student Travel Award Committee of the Physiology Section of the American Fisheries Society, led by Michael Redding, evaluated 65 applications from 15 countries and awarded 40 Travel Grants, after an ambitious and trying fund-raising effort. Special thanks must go the US Department of Agriculture, the US Geological Survey, US National Science Foundation and the World

iii

Fisheries Congress for providing funds. In addition, the American Fisheries Society contributed books to be used as prizes for the best student papers. The editorial team compiled the short abstracts into an abstract book and formatted and compiled the papers for the Symposium Proceedings. Thanks to Karin Howard, Christie MacKinlay, Anne Martin, Callan MacKinlay and Marcelo Perlingeiro. In particular, we would like to extend a sincere thank you to the organizers of the individual scientific Symposia and their many contributors who took the time to prepare a written submission for these proceedings. Their efforts are very much appreciated. We hope that their participation will result in new insights, new collaborations and new lines of research, leading to new papers to be presented at the 2006 Congress in St. John's, Newfoundland. Congress Chairs: Adalberto Luis Val National Institute for Research in the Amazon, INPA, Manaus, Brazil Don MacKinlay Fisheries & Oceans Canada Vancouver, Canada

iv

TABLE OF CONTENTS
Drought and trout sometimes less is more Jim Chadwick................................................................................................ 1 Common trends in fisheries time series in the Northwest Atlantic Jennifer A. Devine*, Richard Haedrich...................................................... 15 Population structure and historical demography of the red snapper (Lutjanus campechanus) fishery in the northern Gulf of Mexico J. R. Gold, E. Saillant, and C. R. Pruett..................................................... 23 Reproductive cycle and migration of blue shark (Prionace glauca) in South Atlantic Ocean. Legat, Jefferson........................................................................................... 25 Preliminary description of fishery in the Amazon estuary based on multivariate analysis. Andra Viana, Diogo Oliveira, Thierry Fredou, Flvia Lucena ................ 37 Bycatch Reduction and Associated Management of the Atlantic Shark Fishery in the United States Michael Clark ........................................................................................... 41 A tentative to disentangle the complexity of Amazon fisheries. Bernard de Mrona .................................................................................... 45 Changes in fishing landings in function hydrological cycle. Cardoso, R. S. & Freitas, C.E.C. ................................................................ 57 An experimental evaluation of altitudinal species zonation patterns in montane streams: do abiotic or biotic factors determine the distribution of native and nonnative trout in Utah, USA, rivers? Pete McHugh .............................................................................................. 63 The importance of connection channels between lakes and Amazonian rivers as side to migratory species. Raniere Garcez and Carlos E. C. Freitas ................................................... 69 A survey of the fish fauna along the floodplain of the Amazon River in Brazil Jansen Zuanon, Luiz Claro Jr., Fernando P. Mendona, Efrem J. G. Ferreira and Lcia H. Rapp Py-Daniel ................................................. 75 Composition and diversity of fish from a managed lake in central Amazon Yamamoto, K. C.; Soares, M. G. M.; Freitas, C. E. C.; Pereira, H. S ............................................................................................................. 79 Dam impacts on fish communities trophic structure (Araguari River, Upper Paran, Brazil) Volney Vono................................................................................................ 85

Inventory of fishes in Maraj archipelago, Western Amazonian (Par State, Brazil) Luciano Fogaa de Assis Montag and Adna Almeida de Albuquerque................................................................................................ 89 The ichthyofauna spatial distribution in the Fish Lagoon National Park (31 S), Brazil Daniel Loebmann & Joo Paes Vieira ..................................................... 101 The fishes of the forest streams of central Amazonia: a preliminary survey Fernando P. Mendona, Jansen Zuanon; Mizael S. Seixas & Andr V. Galuch .................................................................................................. 107 Ichthyofauna composition and diversity in aquatic macrophyte banks in a central Amazonian floodplain lake, AM, Brasil. Magalhes, E.R.S.; Soares, M. G. M.; Freitas, C. E. C.; Yamamoto, K.C. ........................................................................................ 111 Fish species commercialised in Manaus: a revision on the life cycle and biological parameters Batista, G.S.; Soares, M. G. M.; Freitas, C. E. C.; Yamamoto, K. C. ........................................................................................................... 115 Piramutabas (Brachyplatystoma vaillantii) age distribution along the Amazon River Lilianne Esther Mergulho Pirker............................................................ 119 Ecological feature of a Volga pike perch (Stizostedion volgense) population with reference to anthropogenic aspects in the upper Kuibyshev water reservoir of Russia Asiful Islam and V. A. Kuznetsov .............................................................. 125 The fish and fishermen from Encantada Lake (Ilhus, Bahia, Brazil). Gecely Rocha, Schiavetti, A. and Melo, V.G.V ......................................... 137 Environmental Factors Influencing the Distribution of Fish Groups in Headwater Streams, Ja National Park, AM B R Forsberg ............................................................................................ 145 The composition and structure of fish communities of floodplain lakes at lower stretch of Solimes river (Amazon Brazil). Flvia Kelly Siqueira de Souza................................................................. 155 Composition Of Fish Fauna In Salina Lagoon In The Estuary Of Rio Caet, Am, Brazil Ynglea G.F. Goch, Uli Saint-Paul, Jansen A.S. Zuanon; Uwe Krume ....................................................................................................... 167

vi

Fish assemblages in temporary ponds adjacent to "terra-firme" streams, central Amazon Victor Pazin .............................................................................................. 171 American intrusion: range extensions of some groundfish species due to recent climatic changes Alexei M. Orlov......................................................................................... 175 Quantitative changes of bottom trawl catch compositions in the Pacific waters off the Northern Kuril Islands and southeastern Kamchatka during past decades Alexei M. Orlov......................................................................................... 187 Taxonomic diversity and vertical distribution of ichthyoplankton off central America and Costa Rica Dome. Andrei V. Suntsov...................................................................................... 199 Diet and feeding strategy of fishes in a Ruppia maritima meadow, in the Patos lagoon estuary, Brazil. Marcelo B. Raseira ................................................................................... 205 Distribution and abundance of Nannostomus unifasciatus (Characiformes: Lebiasinidae) at Lago Aman, Amazonas, Brazil. Michel F. Catarino, Haroldo B. C. Rodrigues, Eduardo R. Paes & Jansen Zuanon .......................................................................................... 215 Feeding ecology of the Leaf fish Monocirrhus polyacanthus (Perciformes: Polycentridae) in the Aman Lake, Brazilian Amazon. Michel F. Catarino, Jansen Zuanon ......................................................... 219 Nyctemeral variations of the fish community in the seagrass beds of Guadeloupe Island Dorothe Kopp.......................................................................................... 223 Reproductive period and size of the first gonad maturation of the Hippocampus reidi sea-horse in the Brazilian northeast region Rosana Beatriz Silveira and N.F. Fontoura.............................................. 229 The use of mercury and stable isotopes to investigate food chain structure in aquatic ecosystems in the Amazon Mario J. F. 1Thom-Souza........................................................................ 235 Reproductive Biology Of The Roughneck Grunt, Pomadasys Corvinaeformis Anairam De Medeiros E Silva & Sathyabama Chellappa ........................ 241 Reproductive Behaviour And Biology Of Hybrid Red Tilapia Oreochromis niloticus X O. mossambicus A P. T. Medeiros & S. Chellappa ............................................................. 245

vii

Genetic Characterization of Peacock Bass (Cichla Monoculus) And Red-Breasted Pirana (Pygocentrus Nattereri) Populations Collected In Five Lakes In The Area Under The Influence Of Petroleum Transport Activity Between Tesol (Terminal Solimes) And The Reman (Refinery Of Manaus). Luciana E Castro, LK Cunha, M de N Paula-Silva, K LpezVsquez, V M F de Almeida-Val ............................................................... 249 Genetic variability studies of Piramutaba - Brachyplatistoma vaillantii and Dourada - B. Rousseauxii (Pimelodidae: Siluriformes) in the amazon: basis for management and conservation Batista, J. S.; Formiga-Aquino, K.; Farias, I.F. & Alves-Gomes, J.A ........................................................................................................... 253 Mitochondrial DNA population structure in whitemouth croaker (Micropogonias furnieri) along the Atlantic coast of Brazil Puchnick, Angela ...................................................................................... 259 Preliminary analysis of the molecular Phylogeny through mtDNA (12S) in captured species of Gymnotiformes in banks of grass in some points of the Central Amazon. Fernandes, Graciene do Socorro Taveira & Bentes-Sousa, Alexandra Regina .................................................................................... 271 Preliminary study of the Phylogeny through mtDNA (12S) of Siluriformes and Gymnotiformes. Fernandes, Graciene do Socorro Taveira; Bentes-Sousa, Alexandra Regina & Rapp Py-Daniel, Lcia Helena. ............................. 275 The metazoan fauna of Arapaima gigas collected from a floodplain area in the Upper Solimes River Ana Lcia Silva Gomes and Jos Celso de Oliveira Malta ...................... 279 Characterization and analysis of expressed sequence tags (est) from hypophisis and brain of Colossoma macropomum Alexandra Regina Bentes-Sousa et al. ...................................................... 283 Economic incentives, direct controls James Kahn and Zack Manis .................................................................... 285 The role of salinity on the fishes assemblages in the main channel of Paranagu Estuary South Brazil (Tropical/sub-tropical transition zone). Mario Barletta, Ulrich Saint-Paul ............................................................ 299

viii

DROUGHT AND TROUT - SOMETIMES LESS IS MORE J.W. Chadwick Chadwick Ecological Consultants, Inc. 5575 S. Sycamore Street Suite 101 Littleton, CO 80120 UNITED STATES Phone: (303) 794-5530 Fax: (303) 794-5041 Email: Chadeco@aol.com L.C. Bergstedt, D.J. Conklin, and S.P. Canton Chadwick Ecological Consultants, Inc. Abstract Drought has generally been associated with negative impacts to fish communities, as a result of cessation of flow, increased temperature, competition, and susceptibility to predation. In this study, we look at seven years of data on several high elevation trout streams in the southern Rocky Mountain ecoregion. The water year 2002 produced the lowest spring snowmelt runoff for the period of record in these systems (8% to 24% of average). Density and size structure of trout populations were compared to yearly peak runoff flow. Total trout density and density of trout 2+ were most often negatively correlated with spring snowmelt runoff in these systems. Density of YOY and age 1+ were not negatively correlated with runoff to the extent that was expected. Introduction The structuring of aquatic communities through abiotic (e.g., flow, temperature, oxygen) and biotic (e.g., predation, competition) processes and the relative importance of each process has been a long debated issue in aquatic ecology (Allan 1995). Much of the attention associated with abiotic factors structuring aquatic communities has centered on the flow regime, especially the frequency and predictability of high flow events (Poff and Ward 1989). Flooding has been shown to decrease densities of stream fish, especially YOY and age 1+ fish (Seegrist and Gard 1972, Anderson and Nehring 1985, Schlosser 1985, Nehring and Anderson 1993) The effects of drought on trout density have been less well studied. Drought is generally considered to have negative effects on trout densities due to cessation

of flow, increased temperature, competition, and increased susceptibility to predation. Extended drought was believed to be the cause for a significant decrease in brook trout abundance in a Wyoming stream (Binns 1994), the elimination of brook trout in the upper reaches of a California stream (Erman 1986), variation in YOY rainbow trout recruitment in an Appalachian stream (Freeman et al. 1988), and variation in brown trout recruitment in Pennsylvania streams (McFadden and Cooper 1962). However, evidence for the negative effects of drought in these studies was largely anecdotal. In 2002, severe drought conditions existed throughout much of the western United States and the snowmelt runoff period was characterized by extremely low flows. During routine biomonitoring conducted by the authors, densities of resident trout (excludes stocked trout) in several streams appeared to be higher than during previous years of higher runoff. It had been observed that years of higher than normal runoff appeared to result in lower densities of trout, especially younger age classes, as had been previously reported in the literature. We hypothesized that density of trout in Rocky Mountain streams would demonstrate a strong inverse relationship with peak spring runoff flow, with lowest densities being observed in years of high flow, as has been observed by others, and highest densities would be observed in years of low flow, which has not been widely documented. Study Area A total of eight sites were used in this analysis from two watersheds in the southern Rocky Mountain ecoregion (Omernik 1987). Four sites were in the Red River basin in north-central New Mexico. Two sites were on the mainstem of the Red River and two sites were on tributaries to the Red River (Columbine Creek and Middle Fork of the Red River). The most upstream site in the Red River study area is at an elevation of 2,706 m while the downstream site has an elevation of 2,155 m. The dominant resident trout species in the basin is Salmo trutta, with Oncorhynchus clarki, Salvelinus fontinalis, and O. mykiss x O. clarki hybrids also present. Four sites were located in the upper Arkansas River basin in central Colorado. The most downstream site was located on the mainstem of the Arkansas River at 2,963 m. Two sites were located on the East Fork, an upper tributary of the Arkansas River, at elevations of 3,018 m and 3, 030 m. The final site was located on Tennessee Creek, also an upper tributary, at an elevation of 2,999 m. The dominant resident trout species is S. trutta with S. fontinalis also present.

The eight sites were chosen because: 1) they had a relatively long-term, annual fisheries data base associated with them (five to seven years), 2) all sites are minimally impacted from anthropogenic stress and are used as reference sites in routine biological monitoring programs, and 3) all sites have natural flow regimes with no impoundments and minimal water withdrawals (i.e., little agricultural, industrial, or municipal water use) associated with them. Methods Fisheries data were collected using either bank or backpack electrofishing gear and a multiple pass removal method in late-summer or early fall. Population estimates were calculated using the maximum likelihood estimator in the Microfish program developed by the U.S. Forest Service (Van Deventer and Platts 1983, 1989). All fish sampled were identified, counted, measured for total length, weighed, and released. This sampling provided estimates of density and age-structure of the trout community that are used in this analysis. Flow statistics were based on USGS gaging station records from the Red River near Questa, NM (USGS gage 08265000) and from the Arkansas River near Leadville, CO (USGS gage 07081200). Relationships between trout density and peak runoff flows were analyzed by examining the relationships between YOY density, 1+ density, 2+ density, and total density with the highest average monthly flow (peak flow) during the runoff period, based on an average of each days flow. Peak flow generally occurs in May or June in these systems. Relationships between trout density and peak flow were analyzed using the nonparametric Spearman rank correlation (Zar 1999). The nonparametric correlation analysis was chosen due to both density and flow variables not being normally distributed for many of the sites. Correlations were considered significant at p < 0.05. In addition to the Spearman Rank correlation, each relationship was examined graphically with scatter plots in order to determine visually to what degree the density and flow parameters fit or did not fit our hypothesized relationship (Fig. 1). For example, a site with seven years of data would be expected to have the highest density (rank 7) associated with the lowest flow (rank 1) and follow a negative linear trend (density rank 6 with flow rank 2, etc.). To determine the overall fit of the data to our hypothesis, the number of points either on the line or within one point of the line was divided by the total number of data points to

determine the percentage of the data that fit with our hypothesis. We believe that a rank within one unit of the line still demonstrates a close fit and was considered in agreement with the hypothesis. We chose a 70% fit of the data to signify biological significance and be in agreement with the hypothesis. Figure 1. Hypothesized relationship between trout density and peak flow ranks.

Results Peak flow in the southern Rocky Mountains is highly variable. In 2002, the lowest measured flow on record was observed in both systems, with 78 years of data for the Red River and 35 years of data for the Arkansas River. Peak flow on the Red River was only 7.6% of the average peak flow in 2002, while peak flow in 1997 was 43.7% higher than average (Table 1). Peak flow on the Arkansas River varied from 24.1% of average in 2002 to over twice the average in 1997 (Table 1). Red River The highest total resident trout density observed occurred in 2002 at three of the four Red River basin sites (Table 2), corresponding to the year of the lowest peak runoff flows. At the most upstream mainstem Red River site, highest

densities were observed in 2000, the second lowest runoff year recorded during sampling. Three sites had the lowest total resident trout densities in 1999 while Columbine Creek had the lowest trout density in 1997. The highest runoff recorded was in 1997, but only two of the four sites were sampled in 1997. The upstream Red River site and the Middle Fork site showed significant negative correlations between YOY resident trout density and flow (rs = -0.89 and -0.90, p = 0.01 for each site, respectively). For density of age 1+ resident trout in the Red River basin, the downstream Red River site had a significant negative correlation (rs = -0.90, p = 0.04) between density and peak flow. Density of resident trout age 2+ or older showed a significant negative correlation at the upstream mainstem Red River site (rs = -0.89, p = 0.01). Table 1. Peak flow (cfs) for the Arkansas River, CO, and Red River, NM, for years when fisheries data have been collected, and long term-mean over the period of record. Year 1994 1996 1997 1998 1999 2000 2001 2002 2003 Long-term Mean Red River --227 104 186 38 152 12 81 158 Arkansas River 282 503 707 -392 -241 83 269 345

Table 2. Resident trout density (#/ha) from the Red River basin 1997 - 2003. Upper Red River 579 1586 436 3551 906 2726 1137 1560 Lower Red River --699 2240 1810 2471 2237 1891 Middle Fork --1165 4139 1647 9237 2125 3663 Columbine Creek 731 747 1170 3459 1178 6052 1310 2092

Year 1997 1998 1999 2000 2001 2002 2003 Mean

When total density of resident trout was analyzed, the upstream and downstream mainstem Red River sites showed significant negative correlations between total resident trout density and peak flow (rs = -0.99 and -1.00, p > 0.01 for both, respectively). Columbine Creek also had a significant correlation (rs = -0.78, p = 0.04). The Middle Fork site exhibited a perfect negative relationship (rs = 1.00, p < 0.01). When densities were averaged for all Red River basin sites, significant negative correlations were observed between peak flow and YOY density (rs = -0.89, p = 0.01), age 2+ and older density (rs = -0.86, p = 0.01), and total density (rs = 0.96, p < 0.01). A total of 11 of the 20 correlations between trout density parameters and peak flow had a significant negative correlation associated with them. When observing the scatter plots of the ranks of resident trout density and the ranks of peak flow (Table 3), 15 of the 20 comparisons had data which fit our hypothesis over 70% of the time (i.e., data points were on the predicted line or within one unit of the line), indicating better performance than the Spearman rank correlations. Mean density for all sites performed well for YOY, 2+, and total density (Fig. 2).

Table 3. Percentage of data points falling on or within one unit of the predicted density-peak flow line for the Red River basin data and the mean of all Red River basin sites for YOY, age 1+, age 2+ or older, and total density. Site Upper Red Lower Red Middle Fork Columbine Mean Density YOY 100% 20% 100% 57% 86% 1+ 43% 100% 80% 71% 43% 2+ 100% 100% 60% 71% 86% Total 100% 100% 100% 86% 100%

Arkansas River The highest resident trout density observed occurred in 2002 at all four of the Arkansas River sites (Table 4). This corresponds to the year of the lowest peak flow. The lowest resident trout densities observed were more variable. For the mainstem Arkansas River site, the lowest density was observed in 1996, the year with the second highest flow recorded in a year when this site was sampled. The most upstream site on the East Fork has the lowest resident trout density in 1994, a relatively low runoff year. The downstream East Fork site had its lowest density in 1999, a slightly above average runoff year. Tennessee Creek had the lowest density in 1997, the highest runoff year during the period sampled.

Figure 2. Scatter plots of mean YOY, 1+, 2+, and total trout density rank and peak flow rank for all Red River sites. : = data point does not fit hypothesis.

The downstream East Fork site showed a significant negative correlation between YOY density and flow (rs = -0.94, p < 0.01). For density of age 1+ brown trout in the Arkansas River basin, three of the four sites demonstrated significant negative relationships between density and peak flow. The downstream East Fork site and the Tennessee Creek site had identical correlations (rs = -0.83, p = 0.04). The mainstem Arkansas River was also significant (rs = -0.93, p < 0.01). Density of brown trout 2+ showed a significant negative correlation at the mainstem Arkansas River site (rs = -0.93, p < 0.01) and the upstream East Fork site (rs = -0.83, p = 0.04). When total density of resident trout was analyzed, the mainstem Arkansas River site and the downstream site on the East Fork showed significant negative correlations between total resident trout density and peak flow (rs = -0.79 and -0.89, p = 0.04 and 0.02, respectively). When densities were averaged for all Arkansas River basin sites significant negative correlations were observed between peak flow and 1+ density (rs = -0.86, p = 0.01), 2+ density (rs = -0.93, p < 0.01), and total density (rs = -0.93, p < 0.01).

Table 4. Brown trout density (#/ha) from the Arkansas River basin 1994 - 2003. Arkansas River 976 890 982 1239 1388 2993 1414 1412 Upper East Fork 1215 -1802 1422 2034 2492 1536 1750 Lower East Fork 1730 -1252 1238 1847 3084 1879 1838 Tennessee Creek 3057 -898 1648 2045 4019 1292 2160

Year 1994 1996 1997 1999 2001 2002 2003 Mean

A total of 11 of the 20 correlations between trout density parameters and peak flow had a significant negative correlation associated with them. When observing the scatter plots of the rank of brown trout density and the rank of peak flow (Table 5), 13 of the 20 comparisons had data which fit our hypothesis over 70% of the time, indicating slightly better performance than the Spearman rank correlations (Fig. 3). Discussion In general, the analysis of various trout density parameters and peak flow agreed with our hypothesis. However, some results were surprising. For the Arkansas River basin sites, YOY density was not affected to the extent that the older age classes and total density were. For the Red River basin sites, YOY, age 2+ and older, and total density were more likely to be affected than age 1+. The reasons for these differences are not clear. In order to verify that the observed relationships were not artifacts of the varying stream size (i.e., stream widths smaller in low runoff years, stream widths wider in high runoff years), Spearman correlation analysis was used to identify trends between stream width at time of sampling, area sampled, and peak flow. Of the 20 possible

correlations, peak flow and stream width were only significantly correlated at the Middle Fork and upper mainstem Red River site (rs = 0.81 and 0.69, p = 0.04 and 0.02). Sample area was not significantly correlated with peak flow at any site. All analyses were also conducted using fish density expressed as number of fish/km in order to remove any area effect. Results were nearly identical as density expressed as number of fish/ha for all analyses. Table 5. Percentage of data points falling on or within one unit of the hypothesized density-peak flow line for the Arkansas River basin data, and the mean of all Arkansas River basin sites for YOY, age 1+, age 2+ or older, and total density. Site Arkansas River Upper East Fork Lower East Fork Tennessee Creek Mean Density YOY 43% 50% 100% 33% 57% 1+ 100% 17% 83% 83% 86% 2+ 100% 83% 83% 83% 100% Total 71% 67% 100% 67% 100%

Figure 3. Scatter plots of mean YOY, 1+, 2+, and total trout density rank and peak flow rank for all Arkansas River sites. : = data point does not fit hypothesis.

10

Sites with greater habitat complexity may allow for greater resistance to flooding (Pearsons et al. 1992). The same habitat complexity may also alleviate the affects of drought (Binns and Remmick 1994) by providing more deep water habitats. Habitat data was not available for all years the sites were sampled. However, sites were chosen to be representative of average habitat conditions within the reach. This may partly explain the increases in density of older fish during drought years. While stream width was not significantly correlated with peak flow, the flow at time of sampling was significantly correlated with peak flow for both the Arkansas River and Red River sites (r = 0.81 and 0.91, p = 0.03 and < 0.01, respectively). In years of low flow, larger fish may move into areas of better habitat, resulting in decreases in density in areas of poor habitat, which were not sampled. Preliminary investigations have also demonstrated that trout density in streams with more regulated flows associated with water storage reservoirs or irrigation withdrawals do not demonstrate strong correlations with flow variables. This is most likely due to a dampening of the flow regime and more stable flow conditions (i.e., lower peak flow, higher low flow conditions). This analysis suggests that in extremely low runoff years, trout density responds in a positive direction. This may be coincidental carryover from previous strong age-classes or repositioning of adult fish from marginal habitats along with an actual gain in YOY trout. In 2002, the lowest flow on record, most of the sites had the highest YOY densities observed. When flows were well above average, YOY densities were generally among the lowest observed. The pattern generally did not fit as well during years of intermediate flows, suggesting density is controlled by other factors during intermediate peak flow years. Drought, like flooding, is a natural phenomena which has been part of the evolutionary history of fish species throughout the temperate regions of the world. Trout have developed both resistance and resilience mechanisms to overcome problems associated with extremes in flow.

11

References Allan, J.D. 1995. Stream Ecology: Structure and Function of Running Water. Chapman and Hall. New York, NY. Anderson, R.M., and R.B. Nehring. 1985. Impacts of stream discharge on trout rearing habitat and trout recruitment in the South Platte River, Colorado. Pages 59-64 in Olson, F.W., R.G. White, and R.H. Hamre, editors. Symposium on Small Hydropower and Fishereis. American Fisheries Society, Bethesda, MD. Binns, N.A. 1994. Long-term responses of trout and macrohabitats to habitat management in a Wyoming headwater stream. North American Journal of Fisheries Management 14: 87-98. Binns, N.A., and R. Remmick. 1994. Response of Bonneville cutthroat trout and their habitat to drainage-wide habitat management at Huff Creek, Wyoming. North American Journal of Fisheries Management 14: 669-680. Erman, D.C. 1986. Long-term structure of fish populations in Sagehen Creek, California. Transactions of the American Fisheries Society 115: 682-692. Freeman, M.C., M.K. Crawford, J.C. Barrett, D.E. Facey, M.G. Flood, J. Hill, D.J. Stouder, and G.D. Grossman. 1988. Fish assemblage stability in a southern Appalachian stream. Canadian Journal of Fisheries and Aquatic Sciences 45:1949-1958. McFadden, J.T., and E.L. Cooper. 1962. An ecological comparison of six populations of brown trout (Salmo trutta). Transaction of the American Fisheries Society 91:53-62. Nehring, R.B., and R.M. Anderson. 1993. Determination of population-limiting critical salmonid habitats in Colorado streams using the Physical Habitat Simulation system. Rivers 4: 1-19. Omernik, J.M. 1987. Ecoregions of the conterminous United States. Annals of the Association of American Geographers 77: 118-125.

12

Pearsons, T.N., H.W. Li, and G.A. Lamberti. 1992. Influence of habitat complexity on resistance to flooding and resilience of stream fish assemblages. Transactions of the American Fisheries Society 121: 427-436. Poff, N.L., and J.V. Ward. 1989. Implications of streamflow variability and predictability for lotic community structure: a regional analysis of streamflow patterns. Canadian Journal of Fisheries and Aquatic Sciences 46: 1805-1818. Schlosser, I.J. 1985. Flow regime, juvenile abundance, and the assemblage structure of stream fishes. Ecology 66:1484- 1490. Seegrist, D.W., and R. Gard. 1972. Effects of floods on trout in Sagehen Creek, California. Transactions of the American Fisheries Society 101:478482. Van Deventer, J.S., and W.S. Platts. 1983. Sampling and estimating fish populations from streams. Transactions of the North American Wildlife and Natural Resources Conference 48: 349-354. Van Deventer, J.S., and W.S. Platts. 1989. Microcomputer Software System for Generating Population Statistics from Electrofishing Data - User=s Guide for Microfish 3.0. General Technical Report INT-265/1989. U.S. Forest Service. Zar, J.H. 1999. Biostatistical Analysis Fourth Edition. Prentice Hall. Upper Saddle River, NJ. Acknowledgments We would like to extend our thanks to Molycorp, Inc. of Questa, NM, especially Anne Wagner, and Resurrection Mining Company of Denver, CO, for permission to present the data analyzed in this paper. We would also like to thank the Colorado Division of Wildlife, especially Greg Policky, for aiding in the collection of fisheries data on the Arkansas River. Finally, we thank Christopher Garrett and Rebekah Chadwick for aiding in the compilation of data and production of the graphics used in this paper.

13

14

TRENDS IN FISH POPULATIONS IN THE NORTHWEST ATLANTIC Jennifer A. Devine Memorial University, Oceans Science Center 4 Clark Place, St. Johns, NL A1C 5S7 Canada jdevine@mun.ca Richard L. Haedrich Memorial University, Oceans Science Center 4 Clark Place, St. Johns, NL A1C 5S7 Canada haedrich@mun.ca EXTENDED ABSTRACT ONLY DO NOT CITE Introduction Abundance, biomass and size of demersal fish species on the NewfoundlandLabrador Shelf has declined greatly over time. Annual landings of all groundfish species declined rapidly in 1978, stabilized in the 1980s, and declined sharply in the early 1990s (Sinclair and Murawski 1997). Most groundfish fisheries, including Atlantic cod Gadus morhua, were closed in 1992. Although fishing pressure has been reduced on most species, many species are still being captured as bycatch in other fisheries. Seal populations in Atlantic Canada have been increasing since the early 1970s; the population was estimated at approximately 6 million seals (four species) in 1996 (Hammill and Stenson 1997). There have been many arguments about what has caused the declines; the environment, overfishing, predation and changes in prey availability have all been pinpointed as the possible cause. The Newfoundland-Labrador Shelf system experienced severe temperature and salinity anomalies in the early 1970s, 1980s and 1990s. The mid-1980s to mid1990s saw temperature lows, with the early 1990s experiencing the lowest recorded temperature anomalies in sea surface waters (0-176 meters) since 1950. At the same time, the shelf system was experiencing a salinity anomaly that was similar in magnitude to the anomaly that occurred in the early 1970s (Belkin 2004). These salinity anomalies appear to occur on a decadal scale and may be linked to the North Atlantic Oscillation index (NAO), which is the pressure

15

difference between the Azores and Iceland. The NAO exerts a strong influence over the ocean and atmosphere of the North Atlantic Ocean. Our objective was to determine if the complex dynamics of the NewfoundlandLabrador Shelf system could be described using multivariate time series analysis. We used a newer type of time series analysis, designed for relatively short, nonstationary time series, to analyze trends in catch per unit effort of 13 fish species and one shark species and relate these trends to environmental variables, predator abundance and fishing effort. Previous studies have used a variety of methods to analyze possible causes of groundfish population and size structure declines with this same data; however, results have been inconsistent. Methods Study site The Newfoundland-Labrador Shelf is a unique ecosystem due to its topography and circulation patterns. The shelf is broad, ranging from 150-400 km wide, typically less than 200 meters deep and overlain by polar waters (Helbig et al. 1992). The Labrador Shelf topography is very complex; the shelf contains numerous shallow banks separated by deep saddles that provide channels from the deep-sea to the inner shelf and allow cross-shelf exchange (Drinkwater and Harding 2001). The northeast Newfoundland Shelf is broader and contains many deep bays inshore while offshore the shelf is separated into flat banks broken by a basin that deepens to 500 meters. To the south, the shelf forms the Grand Banks of Newfoundland, a relatively flat area with an average depth of 80 meters (Helbig et al. 1992). The Labrador Current forms two distinct branches over the Newfoundland-Labrador Shelf. The main branch flows offshore along the continental slope centered at the 500-meter isobath and the inshore branch flows along the inner half of the shelf at a much slower speed (Drinkwater and Harding 2001). The Labrador Current forms the cold intermediate layer, capped above and below by warmer waters, and has an effect on the distribution and migratory patterns of many fish species. Data The ECNASAP (East Coast North American Strategic Assessment Project) dataset was used as the source of records for the Newfoundland-Labrador Shelf (NAFO Divisions 2J3KL) for years 1978 through 1994. Prior to 1978, the survey was not based on a random stratified design and gear configuration was variable. In autumn 1995, DFO changed trawl gears and modified the survey

16

design to include deeper stations. Catches from the two gears are not comparable; therefore, the analysis does not include 1995 to the present. A mixture of important commercial, rare and non-commercial demersal fish species and one shark species were chosen for the analysis; Atlantic cod, American plaice Hippoglossoides platessoides, roughhead grenadier Macrourus berglax, rock grenadier Coryphaenoides rupestris, Greenland halibut Reinhardtius hippoglossoides, thorny skate Raja radiata, deepwater redfish Sebastes mentella, golden redfish Sebastes marinus, spinytail skate Bathyraja spinicauda, Atlantic wolffish Anarhichas lupus, northern wolffish Anarhichas denticulatus, spotted wolffish Anarhichas minor, blue hake Antimora rostrata, and black dogfish Centroscyllium fabricii. Catch per unit effort in weight was used as an index of size structure over time. Because trawl survey data often have a skewed distribution, data were log10(x+1) transformed. Environmental variables were sea surface temperature recorded to 100-meters depth, NAO annual index (www.cgd.ucar.edu/~jhurrell/nao.stat.ann.html), bottom temperature 250-1485 meters (the maximum depth of the survey), and salinity 0250 meters in NAFO Divisions 2J3KL, 1978-1994. Annual anomalies were estimated for all environmental data. Fishing effort data (number of days) in NAFO Divisions 2J3KL was obtained from the NAFO annual fisheries statistics database (www.nafo.ca) and harp seal abundance obtained from DFO, Newfoundland-Labrador region were also included. All data were standardized to assist with interpretation. Time series analysis Min/max autocorrelation factor analysis (MAFA) is a type of principal component analysis designed for time series to extract trends (www.brodgar.com). A trend is a long-term change in the mean level or a slow moving curve. One characteristic of a trend is high auto-correlation with time lag 1. Whereas the PCA algorithm will estimate axes (or components) that have a decreasing variance, the MAFA method estimates axes that have decreasing autocorrelation with time lag 1. Therefore, the first MAFA axis is the smoothest curve, or trend, underlying all the time series. Hence, the first axis represents the main trend in the time series and other axis represent less important trends. Loadings and canonical correlations, which are crosscorrelations between MAFA axes and CPUE time series, can be estimated to determine which species time series are related to a particular axis. MAFA analysis was implemented using the software package Brodgar 2.2.8 (www.brodgar.com). A randomization process was used to obtain p-values to determine how many axes to use.

17

Dynamic factor analysis (DFA) is a dimension reduction technique for multivariate time series analysis. DFA can be used to determine common patterns in time series, evaluate interactions between response variables and determine the effects of explanatory variables in short, nonstationary time series (Zuur et al. 2003, Zuur and Pierce 2004). We used DFA to model the time series as a function of a linear combination of common trends, a level parameter and noise. DFA has been applied to economic and psychology data and only recently has this technique been applied to fisheries data. DFA analysis was used by Zuur et al. (2003) to model Norwegian lobster Nephrops norvegicus from fishing grounds in northern European waters and Zuur and Pierce (2004) to model trends in Atlantic squid time series. DFA analysis was implemented using the software package Brodgar 2.2.8 (www.brodgar.com). Results and Conclusions Time series and correlations Cross-correlations between response variables and explanatory variables at lags up to 5 years, 10 years for fishing effort, showed some lagged explanatory variables had higher correlations than variables with no lags. CPUE of all species except Atlantic cod, golden redfish and thorny skate were significantly positively correlated with sea surface temperature lagged 3 years. CPUE of all species except Atlantic cod, American plaice, spotted wolffish, golden redfish and thorny skate were significantly negatively correlated with salinity data lagged 4 years. CPUE of all species except Atlantic cod was significantly positively related with fishing effort lagged 9 years. Bottom temperature, NAO annual index and harp seal abundance data with no lags had the highest proportion of significant correlations. All species except rock grenadier, blue hake and golden redfish were significantly positively correlated with bottom temperature, all species except Atlantic cod, Greenland halibut and blue hake were significantly negatively correlated with the NAO index, while all species except Atlantic cod were significantly negatively correlated with harp seal abundance. MAFA Analysis The MAFA analysis showed two main trends in the CPUE time series (Figure 1). The first MAFA axis, with an autocorrelation of 0.996 (p=0.050), represents a steady decline over time; this is the main pattern underlying the time series. The second MAFA axis (autocorrelation of 0.922, p=0.017) shows an increase from 1978 to 1987, followed by a decline. Canonical correlations between the

18

species and MAFA axes indicated that the first axis was important for all species except Atlantic cod, whereas the second axis was important for only American plaice, Atlantic cod and thorny skate. Because of collinearity between harp seal abundance and fishing effort, fishing effort was eliminated from the MAFA analysis. Explanatory variables used were salinity lagged 4 years, harp seal abundance (no lag), NAO annual index (no lag), bottom temperature (no lag) and sea surface temperature lagged 3 years. All explanatory variables except the NAO annual index were significantly correlated (p<0.05) with the first MAFA axis; harp seal abundance and salinity (lagged 4 years) were negatively correlated while bottom temperature and sea surface temperature (lagged 3 years) were positively correlated. The NAO annual index, while not significant at the p=0.05 level, was negatively correlated with the declining trend. Bottom temperature was the only explanatory variable significantly correlated with the second MAFA axis; bottom temperature was positively correlated with the increasing then decreasing trend.

0.6 0.4 0.2 Scores 0.0 -0.2 -0.4 -0.6 1978 1982 1986 Year
Figure 1. Two significant MAFA axes on CPUE time series of thirteen fish and one shark species from scientific random stratified surveys in NAFO Divisions 2J3KL, 1978-1994. Heavy line indicates main MAFA axis. DFA Analysis

1990

1994

19

DFA analysis with lagged explanatory variables gave a better fit than models with no lags. The Akaikes Information Criterion (AIC) was used initially to determine the goodness of fit and the number of parameters in the models; the model with the smallest AIC value was selected as being the most appropriate model. Additionally, fitted values and residuals were also used to determine goodness of fit of the model. Biological interpretation is also important for deciding which model is the best. Therefore, some variables with no lags, such as bottom temperature were selected as being the better model when running the analysis using two explanatory variables. The best models with only one explanatory variable were harp seal abundance lagged 1 year (1 common trend, AIC=275.0; 2 common trends, AIC=282.1) and fishing effort lagged 10 years (1 common trend, AIC=282.0; 2 common trends, AIC=291.6). Because of collinearity between harp seal abundance and fishing effort, these two explanatory variables could not be combined in the analysis. The best model with two explanatory variables, determined using the AIC and examining the plots of residuals, fits, common trends, factor loadings and canonical correlations, was harp seal abundance lagged 1 year and salinity lagged 2 years (1 common trend, AIC=172.7; 2 common trends, AIC=181.7; 3 common trends, AIC=196). The model with 3 common trends was chosen (Figure 2).
6 4 2 Trends 0 -2 -4 -6 1978 1980 1982 1984 1986 1988 1990 1992 1994 Year
Trend 1 Trend 2 Trend 3

20

Figure 2. The three main common trends of the CPUE time series for the DFA model with two explanatory variables, salinity lagged two years and harp seal abundance lagged one year in NAFO Divisions 2J3KL, 1978-1994. The first and main common trend is similar to the second MAFA axis; it showed an increase and then decrease over time. The second trend showed an oscillating pattern with a decrease until 1984 and then increased. The third trend declined until 1982, increased slightly until 1985, declined until 1989 and then remained fairly constant. Factor loadings and canonical correlations were used to determine which species were related to a particular trend. The first common trend was significantly, positively related to American plaice, Atlantic cod and thorny skate. Only American plaice was significantly and negatively related to the second trend; however, Atlantic cod, northern wolffish, spotted wolfish, Greenland halibut, black dogfish, deepwater redfish, golden redfish, spinytail skate and thorny skate were strongly and negatively related to the second common trend. All species except Atlantic cod were strongly and positively related to the third common trend; only Atlantic wolffish, northern wolfish, Greenland halibut, roughhead grenadier, rock grenadier, black dogfish, blue hake, deepwater redfish, and spinytail skate were significantly correlated to the trend. The second MAFA axis and first DFA trend are very similar and are strongly related to the same three species. This may be due to these species having similar generation times (8-11 years). Previous studies have also found these three species tend to be in the same assemblages and have similar bottom temperature preferences. The main MAFA axis and third DFA trend are also similar and are related to the same species; however, DFA analysis has shown that this trend oscillates over time. These techniques highlight the complexity of the Newfoundland-Labrador Shelf ecosystem and give insight into a rich system; many dynamics are occurring at the same time, often within the same species. Our analysis has illustrated a numeric technique to gain insight into the elements of a changing system. Time lags are important to consider when determining the effects of explanatory variables.

21

References Belkin, I.M. 2004. Propagation of the Great Salinity Anomaly of the 1990s around the northern North Atlantic. Geophys. Res. Lett. 31 (in press). Drinkwater, K.F. and G.C. Harding. 2001. Effects of the Hudson Strait outflow on the biology of the Labrador Shelf. Can. J. Fish. Aq. Sci. 58: 171-184. Hammill, M.O., and G.B. Stenson. 1997. Estimated prey consumption by harp seals (Phoca groenlandica), hooded seals (Cystophora cristata), grey seals (Halichoerus grypus) and harbour seals (Phoca vitulina) in Atlantic Canada. J. Northw. Atl. Fish. Sci. 26: 1-23. Helbig, J., G. Mertz, and P. Pepin. 1992. Environmental influences on the recruitment of Newfoundland/Labrador cod. Fish. Oceanogr. 1: 39-56 Sinclair, A.F., and S.A. Murawski. 1997. Why have groundfish stocks declined? In Boreman, J., B.S. Nakashima, J.A. Wilson, and R.L. Kendall. Northwest Atlantic groundfish: perspectives on a fishery collapse. American Fisheries Society, Bethesda MD. Zuur, A.F., and G.J. Pierce. 2004. Common trends in northeast Atlantic squid time series. J. Sea Res. (in press). Zuur, A.F., I.D. Tuck and N. Bailey. 2003. Dynamic factor analysis to estimate common trends in fisheries time series. Can. J. Fish. Aq. Sci. 60: 542-552. Acknowledgements This work was supported by Coasts Under Stress and NSERC. WE thank DFO, Newfoundland-Labrador region for supplying data and A. Zuur, Highland Statistics, for assistance with data analysis.

22

POPULATION STRUCTURE AND HISTORICAL DEMOGRAPHY OF THE RED SNAPPER (LUTJANUS CAMPECHANUS) FISHERY IN THE NORTHERN GULF OF MEXICO John R. Gold Center for Biosystematics and Biodiversity Texas A&M University College Station, Texas 77843-2258. USA Phone (979-847-8778 FAX (979) 845-4096 goldfish@tamu.edu Eric Saillant and Christin L. Pruett Center for Biosystematics and Biodiversity Texas A&M University College Station, Texas 77843-2258. USA EXTENDED ABSTRACT ONLY DO NOT CITE The Gulf red snapper, Lutjanus campechanus, is a highly exploited marine fish in the Gulf of Mexico (hereafter, Gulf). Current management of Gulf red snapper in U.S. waters is based on a single-stock hypothesis. Herein, results of a genetic survey of individuals from four cohorts (1995, 1997, 1999, and 2000) sampled offshore from three geographic localities in the northern Gulf are reported. The genetic survey included 19 nuclearencoded microsatellites and 590 base pairs of the mitochondrially (mtDNA) encoded ND-4 gene. Very low and generally non-significant levels of divergence among localities were found for both microsatellites and mtDNA. Maximum-likelihood estimates of current (variance) effective population size (NeV), based on the temporal method, ranged between 1,048 and 14,275 and were approximately three orders of magnitude lower than current estimates of census size (N). Such a ratio of effective to census size (NeV/N) is far below that expected in an ideal population. Potential causes for this low ratio are high variance in individual reproductive success and/or high temporal or spatial variance in productivity of reproductive subunits or demes. Estimates of NeV differed among localities; estimates of NeV from a locality in the northcentral Gulf were ~10 times higher than in localities to the east and west (Table 1).

23

Table 1. Estimates of variance effective size (NeV)* of red snapper sampled at three localities in the northern Gulf of Mexico. _______________________________________________________ ML NeV Texas Louisiana Alabama 95% low 1,048 14,275 1,273 95% high 622 1,998 803 2,576 >80,000 2,595 _______________________________________________________

_______________________________________________________ *Corrected for overlapping generations The differences in variance effective size are consistent with the hypothesis that different demographic stocks occur within the fishery in the northern Gulf. Nested-clade analysis of mtDNA haplotypes indicated a recurring history of range expansion due to short-distance dispersal and restricted gene flow due to isolation by distance or fragmentation. Results of nested-clade analysis, combined with differences in variance effective size and differences in ecology and hydrology across the northern Gulf, are compatible with the hypothesis that short-term, semi-isolated assemblages of red snapper exist in the northern Gulf. Over the long term these semi-isolated assemblages likely comprise a larger metapopulation tied together by sufficient, periodic gene flow to homogenize allele frequencies at selectively neutral genetic markers but not necessarily at genes responding to selective pressures and important to overall life history. Management of the red snapper fishery in the northern Gulf as a single stock in terms of resource assessment, allocation, and conservation may not be warranted. Acknowledgments Funding for the work was provided by the U.S. Department of Commerce (Marfin Program), the Gulf & South Atlantic Fisheries Foundation, and the Texas Agricultural Experiment Station. We thank the Cowan laboratory (University of South Alabama), the Wilson laboratory (Louisiana State University), and the Gold laboratory (Texas A&M University) for specimen procurement, and A. Blanchard, C. Bradfield, C. Burridge, and J. Patton for assistance in laboratory assays.

24

REPRODUCTIVE CYCLE AND MIGRATION OF THE BLUE SHARK (PRIONACE GLAUCA) IN SOUTH ATLANTIC OCEAN Jefferson F. A. Legat Empresa Brasileira de Pesquisa Agropecuria, Embrapa Meio-Norte. BR 343, km 35, Caixa Postal 341, Parnaba-PI. CEP 64200-970. Brazil. e-mail: legat@cpamn.embrapa.br Carolus M. Vooren Fundao Universidade Federal do Rio Grande, Laboratrio de Elasmobrnquios e Aves Marinhas, Caixa Postal 474, Rio Grande-RS. CEP 96201-900. Brazil. e-mail: doccmv@furg.br Abstract A study conducted in Southern Atlantic Ocean, between 27S and 35S, and researches developed by several authors propose the existence of an annual reproductive cycle for the blue shark, Prionace glauca, in the South Atlantic Ocean. Coupling takes place during spring and summer. Ovulation and egg fertilization occur between March and April, followed by parturition 9 or 12 months later. The existence of two distinct populations was proposed for South Atlantic: Population I, between 5 N and 7 S; and Population II between 20 and 35 S. Coupling, ovulation, fertilization and pregnancy of Population II occur near the southeastern and southern Brazil. Parturition and nursery areas may be situated respectively at the southeastern central Atlantic and southeastern Atlantic regions. Population I accomplishes coupling and ovulation in the northeastern Brazil, and pregnancy and parturition near Africa. Introduction The blue shark Prionace glauca is abundant worldwide in the epipelagic zone of tropical, subtropical and temperate seas. Throughout its area of distribution, the species has been caught in increasing numbers since the early 1960`s as bycatch or target species, in pelagic longline fisheries. In recent years the catch per unit effort has declined in several fishing areas. Due to its life characteristics the species is classified as especially vulnerable to overfishing (Castro et al., 1999). Thus, there is an urgent need to manage the fishery. One of the requirements for

25

such management is to define reproductive cycle, unit stocks and migratory pattern. P. glauca is viviparous, with a yolk-sac placenta. In the North Atlantic Ocean, reproductive cycle is annual. Coupling takes place during summer. Fertilization occurs in spring, followed by parturition 9 or 12 months later, young are born at a length of 35-50cm. After parturition, ovarian eggs are ready to another fertilization (Pratt, 1979). In the North Atlantic, individuals have shown a regular clockwise trans-Atlantic migration pattern with the current system of that hemisphere. P. glauca rides the Gulf Stream to Europe, takes various currents in the European and African coast and returns to North America in the Atlantic North Equatorial Current (Hazin, 1993; NOAA ,1998). In the southern Atlantic Ocean, Hazin et al. (2000) proposed an annual reproductive cycle. Coupling takes place between December and February. Ovulation and fertilization occur between February and April and parturition occurs next December and February. These authors proposed the existence of a single stock of P. glauca in the South Atlantic, with a clockwise migration as observed in North Atlantic. Females blue shark are supposed to copulate in Southern Brazilian waters during summer and then move to Northeastern Brazilian areas to ovulate and fall pregnant 4 months later. The pregnant females then move to African Coast and the parturition may occur at high latitudes. In this case, individuals dont follow the current system of South Hemisphere. Previous studies in southern and southeastern Brazilian coast ( lat. 20S and 33S and long. 39W and 50W) observed copulation, ovulation, fertilization, and pregnancy with embryos at different stages (Amorim, 1992; Guedes, 1999; Legat, 2001). These data indicated that blue sharks females of Southeast and Southern regions had the entire cycle except parturition at those areas and did not follow the theorized clockwise movement.

In this work we present a study of P. glauca reproduction in the Southern Atlantic Waters. According to these results and to researches developed by several authors, we propose the existence of an annual reproductive cycle and a new migration pattern for the blue shark in the South Atlantic Ocean.

26

Material and methods This study was conducted at the "Laboratrio de Elamobrnquios e Aves Marinhas da Universidade Federal do Rio Grande. The study area was the continental slope of southern Brazil, between lat. 27S and 34S, and long 46W and 51W (Figure 1). The samples were taken by R. V. Atlntico Sul, using tuna longline. Longline was baited with squid, set at 5 pm, and retried at 9 am of the next day. Samples were distributed randomly over depths of 200 to 1000 m, with few samples over greater depths: 7 sets in November and December 1996, 11 sets in July 1997, and 10 sets in March and April 1998.

Figure 1.Study area between lat. 27S and 34S and long. 46W and 51W. Spots represent the longline sets. A total of 227 blue sharks were caught, 60 females and 167 males. Data were collected immediately after capture. The total length (TL) was measured according to Compagno (1984). Weight of the body, liver and gonads were recorded. In males, clasper length was measured from the insertion to tip. For females, the following data were recorded: width of the nidamentary gland, width of the uterus, color and diameter of the largest ovarian follicle, presence of uterine eggs and embryos. In the 8 pregnant females caught, number of embryos, embryos TL and weight were recorded.

27

The presence of mating marks in female blue sharks is reported as copulation indicator (Stevens, 1974; Pratt, 1979; Hazin et al., 1994). Wounds can be semicircular impressions, tooth slaches and individual tooth cut. Fresh mating injuries indicates copulation period (Stevens, 1974). Statistical analyses were performed according to Campbell (1989) and to Sokal and Rohlf (1969). Results and Discussion In the southern Brazil, male blue shark reach maturity to start from 210 cm TL and all the individuals larger than 249 cm TL are adults (p< 0,000). The adult male has claspers larger than 15 cm, testis weightier than 200g and gonadosomatic indices higher than 0.4. The blue shark females reach maturity in this region to start from 195 cm TL. At 215 cm TL, 50% of population is mature and all females larger than 240 cm are adult (p< 0,000). Females were separated into the following sexual stages: Immature, (individuals smaller than 195 cm TL. Translucent uterine follicle, range from 0.4 to 0.7 cm diameter. Width of both nidamentary gland and uterus smaller than 2 cm); Subadult (females with TL between 197 and 240 cm; Nidamentary gland width bigger than 2.0 cm; Uterus width between 2.0 cm and 3.0 cm; Opaque or translucent ovarian follicles. These sharks are supposed to start the first reproductive cycle to enter the adult parcel of population); Adult (specimens with TL between 204 cm and 294 cm, nidamentary gland width larger than 2.0 cm, uterus width larger than 5.0 cm, presence of vitellogenic follicles larger than 1.0 cm diameter or presence of uterine eggs or embryos). From these data and previous work from Grotenbreg and Vooren (1999) works, male reach maturity between 4 and 7 years and female between 3.5 and 7 years. According to the results of this work and the studies from Amorim (1992), Hazin et al. (1994; 2000) and Guedes (1999), we propose the existence of an annual reproductive cycle for female blue shark in the South Atlantic Ocean (Figure 2) as descript bellow:

28

Mating occurs from November to March, mainly from December to February when females have vitellogenic follicles with 1.5 cm and 3.0 cm diameter. Ovulation and egg fertilization occur immediately after coupling in southern and southeast Brazilian waters. In northeastern Brazil, females ovulate and fall pregnant 3 months after mating. Thus, between the end of the summer and early fall, females blue shark have uterine eggs or embryos with TL ranging from 2.0 cm and 6.0 cm. From March to June, embryos grow to 10.0 and 25.0 cm TL. Embryos larger than 18.0 cm TL have placental connections with the uterine wall. In the winter, embryos reach 25.0 to 30.0 cm TL. Females begin to produce translucent ovarian follicles between 0.2 cm and 0.5 cm diameter. In the spring, embryos from 30.0 to 35.0 cm TL and translucent or opaque follicles between 0.2 and 0.8 cm are observed. In the late spring, vitellogenic follicles up to 1.0 cm diameter are present. In the following summer parturition occurs 9-12 months later. Embryos are full term with TL ranging from 35 to 50 cm. Females have vitellogenic follicles between 1.5 and 3.0 cm. After pupping, female reinitiate the copulationovulation-fertilization-gestation-parturition sequence.

29

Figure 2. Reproductive cycle of female Prionace glauca in South Atlantic Ocean According to this work and studies from Amorim (1992), Hazin et al. (1994; 2000), Castro and Mejuto (1995), Guedes (1999), and data from the Instituto Nacional de Pesca, Montevideo, Uruguay, we propose the existence of two stocks of P. glauca in the South Atlantic Ocean. Population I in Equatorial region between lat. 5N and 7S and Population II between lat. 20S and 40S (Figure 3). Beside the studies cited, the migration pattern of the two populations agrees with the geostrophic currents of South Atlantic Ocean determinate by Stramma and England (1999). The presence of mating, ovulation, fertilization and pregnant females with embryos at various stages of development at the southern and southeast Brazilian waters means that these females accomplish the entire reproductive cycle, except parturition, in the same region.

30

Figure 3. A: Proposed model of Prionace glauca migration in South Atlantic Ocean; Cp means copulation area, Ov ovulation area, Gt gestation area, Pt parturition area and Ns nursery area. Areas marked with an * are not confirmed. B: geostrophic currents of South Atlantic Ocean determinate by Stramma and England (1999).

31

In northeastern Brazil, lat. 2S and 7 S and long. 32W and 38 W, Hazin et al. (1994; 2000) observed ovulation, fertilization and gestation till embryos with 26.0 cm Fork length (FL). Those authors also observed mating scar at small scale in that region. Castro & Mejuto (1995) collected pregnant female with embryos from 3.0 cm to 35 cm FL, between lat. 5 N and 5 S and long. 5 E and 5 W. Hazin et al. (1994; 2000) suggest that female blue sharks copulate at southern Brazil and then move towards northeastern waters to ovulate about 3 months later. In this case, female must swim against the Brazil Current before start the ovulation and pregnancy and swim against the South Atlantic Current after parturition (Figure 3). P. glauca is a placental viviparous shark. Placental species establish a uteroplacental complex that supplies the embryo with nutrients and oxygen and removes wastes (Hamlett, 1997). In this way, movements with the stream become a form of maintaining energy for the gestation period. Furthermore, in their study about movements of P. glauca, Carey and Scharold (1990) affirm that none of the sharks with attached transmitters appeared to be swimming against a current. We proposed that in the population I, mating, ovulation, fertilization and initial stages of pregnancy occur at East region of South Atlantic Ocean, near northeastern Brazil. Females move towards the African waters while embryos developed. Parturition may occurs between lat. 5N and 5S and long. 5E and 10E, near the Angola Gyre (Figure 3). In the theorized population II, mating, ovulation, fertilization and pregnancy occur between lat 20 S and 40 S. Although it was been observed full term embryos in pregnant females, it was not possible to determinate a particular area where parturition occurs. We propose that pupping area is located in Uruguay and Argentina waters. Nursery area probably is located in African waters between lat. 30 S and 40 S (Figure 3). According to the migrate pattern proposed in this study, population I and II move following the geostrophic currents of South Atlantic Ocean. Population I moves with North Brazil Current, the South Equatorial Current, the Northern,

32

Equatorial, Central and Southern Branches and with the Equatorial Undercurrent. Population II moves with Brazil Current, Benguela Current and South Atlantic current. In this model, the species follow the current system to make a trans-Atlantic migration route as in North Atlantic Ocean. Further study is required to determinate the parturition and nursery areas of both population and to determinate the major mating area of population I. Therefore, data from another South Atlantic regions will permits verify the existence of unit stocks in South Atlantic. References Amorim, A.F. 1992. Estudo da biologia da pesca e reproduo do cao-azul, Prionace glauca L. 1758, capturado no sudeste e sul do Brasil. Instituto de Biocincias do Campus de Rio Claro, Universidade Estadual Paulista, Rio Claro, SP.205p. (Phd Thesis). Campbell, M. 1989. Statistics for biologists, 3nd edition. Cambridge University Press, NY. Carey, F.G. and Scharold, J.V. 1990. Movements of blue sharks (Prionace glauca) in depht and course. Mar. Biol., 106: 329-342 Castro, J.A. and Mejuto, J. 1995. Reproductive parameters of blue shark, and others sharks in the Gulf of guinea. Mar. Fresh. Res. 46: 967-973. Castro, J.L., Woodley,C. M. and Brudek,R.L. 1999. A preliminary evaluation of the status of shark species. FAO Fish. Tech.Paper, 380, 72p. Compagno, L.G..V. 1984. FAO species catalogue, volume 4. Sharks of the world. An annoted and illustrated catalogue of sharks species known to date. Part 2. Carcharhiniformes. FAO Fish. Synop., 125: 252-655. Grotenbreg, L. and Vooren, C.M. 1999. Projeto ARGO, Relatrio final, volume 2. Avaliao dos Recursos Pesqueiros dos Peixes Pelgicos de Grande Porte, parte 3. Age and growth of the blue shark, Prionace glauca, from southern Brazil. Fundao Universidade do Rio Grande, Rio Grande, RS.

33

Guedes, V. 1999. A pesca de espinhel de superfcie (longline) na regio sudeste-sul do Brasil. Programa Revizee / Score Sul, Avaliao das Capturas de Elasmobrnquios, CEPSUL/IBAMA, Itaja, SC. 183p. (Relatrio anual tcnico- cientfico). Hamlett, W.C. 1997. Reproductive models of elasmobranches in Shark News 9, June 1997. Nature Conservation Bureau Limited. Berkshire, UK. 16p Hazin, F.H.V. 1993. Fisheries-oceanographical study on tunas, billfishes and sharks in the southwestern Equatorial Atlantic ocean. Graduate School of Fisheries, Tokyo University of Fisheries, 286p. (PhD. Thesis). Hazin, F.H.V., Kihara, K., Otsuka, K., Boeckman, C.E. And Leal, E.C. 1994. Reproduction of the blue shark Prionace glauca in the southwestern Equatorial Atlantic ocean. Fish. Sci., 60 (5): 487491. Hazin F H.V., Pinheiro, P.B. and Broadhurst, M.K. 2000. Further notes on reproduction of the blue shark, Prionace glauca, and a postulated migratory pattern in the South Atlantic Ocean. Cincia e Cultura, 52 (2):114 a 120. Legat, J. F. A. 2001. Distribuio, abundncia, reproduo e morfometria de Prionace glauca no sul do Brasil. Fundao Universidade Federal do Rio Grande. Programa de Ps-graduao em Oceanografia Biolgica. Rio Grande 118p. (MSc. Thesis) NOAA / NMFS. 1998. Shark Tagger 1998 Summary. USA. Pratt, H.L.JR. 1979. Reproduction in the blue shark, Prionace glauca. Fish. Bull., 77 (2): 445-470. Sokal, R.R. And Rohlf, J.F. 1969. Biometry. W.H. Freeman and Company, USA. Stevens, J.D. 1974. The occrrence and significance of tooth cuts on the blue shark (Prionace glauca L.) from british waters. Journal of the Marine Biological Association of the United Kingdom, 54: 373-378.

34

Stramma, L. and England, M. 1999. On the masses and mean circulation of the South Atlantic Ocean. J. Geophys. Res., 104 (C9): 20 863-20 883. Acknowledgements The senior author carried out the present study as part of his MSc thesis in the Postgraduate Curse in Biological Oceanography of the University of Rio Grande, Brazil. The data were collected by the Project ARGO under grant n 62.0369/922 of the Conselho Nacional de Desenvolvimento Cientifico e Tecnolgico, CNPq .The authors thank the officers and crew of the R.V. Atlntico Sul for their cooperation during the longline cruises, the many students of Rio Grande University who assisted in the field work, and Prof. Dr. Tabajara Lucas de Almeida for his advice on statistical methods.

35

36

PRELIMINARY DESCRIPTION OF FISHERY IN THE AMAZON ESTUARY BASED ON MULTIVARIATE ANALYSIS Andra Viana, Diogo Oliveira, Thierry Frdou and Flvia Lucena GEMC- Grupo de Estudos Marinhos e Costeiros / Depto. de Oceanografia / UFPA. +55 (91) 211 1983 apv@ufpa.br The Amazon basin is the largest hydrographic basin of the world, with a size of 5.8 million of km2, being 3.9 million in Brazil. With 6.500 km of extension, the Amazon is responsible by the discharge of 20% of the total fresh water in the oceans. The Amazon estuary is constituted by the encounter of Amazon and Para rivers, which creates a huge zone of brackish water sizing 1200 nautical miles. This diversity is characterized by the strong seasonality of the region with changes in the salinity along the year. By the beginning of the year (rainy season), a great amount of fresh water flows through the estuary and fresh water species dominate. During the second semester (dry season) marine species dominate in the estuary. Fishery (and target species) is hence driven by this seasonality. Fishing activity used a great variety of gears. However, there is little information recorded concerning the activity of the types of gears in the region. This study aims to describe the fishing activity within the estuarine region of the state of Par using multivariate analysis aiming at describing an integrated picture of the fishing activity. Data was collected from a fishery company based in Belm-Par where landings in 2002 were reported for each trip. The CPUE (catch per unit of effort) were monthly obtained as kg per trip per gear (gill net, longline and trawlers). Statistical analysis was based on multivariate analysis in order to describe the typologies of the estuarine fishery along the coast of Par. The computational program PRIMER for Windows, version 5.2.4 was used. Cluster and MDS (Multidimensional Scaling) analyses were performed.

37

The landing of the estuarine fishery in the state of Par included 37 species of 16 different families and 5 orders. The order Perciforme dominates and the main fish families with the largest number of species reported were Pimelodidae, Scombridae, Scianidae and Ariidae. However, considering that the salinity and rain influence the fish community, the dominant species varied according to the semester of the year. The analysis grouped species considering the fishing gears. Considering the minimum of similarity of 40%, it was evident the formation of 5 groups within the estuarine fishery of the coast of Para (Figure 1). These groups were mainly characterized by their fishing gears. The group G1 encompassed the species captured by trawlers, with the fishing season lasting 3 months during the 2nd semester of the year. The gold (Brachyplatystoma flavicans) was the dominant species and represented by average abundance of 16770,52 of the group. In this group didnt observe significative values of by-catch. The group G2 was characterized by the specie Laulao catfish (Brachyplatystoma vaillantii) as target, with average abundance of 14911,68 of the group, and by B. flavicans as by-catch. These species were captured by trawler during the first semester of the year. Sharks (Carcharhiniformes) dominated the G3 group that was characterized by the longline as a gear mainly during the first semester and it had average abundance of 9294,39. In this group didnt observe significative values of bycatch. The G4 group, also characterized by longline, showed the Gillbacker sea catfish (Arius parkeri) as the dominant species. And the average abundance was of 3284,87. The main by-catch of this fishery was represented by the species Crucifix sea catfish (Arius proops) and Carcharhiniformes. Considering the gill net fishery, that is exclusive to the G5 group, Acoupa weakfish (Cynoscium acoupa) was the dominant specie showing average abundance of 3033,48. The A. parkeri, Amazon croaker (Pachyurus spp., Micropogonias furnieri), Fat snook (Centropomus sp.) and Catfish (Arius couma) were by-catch. The MDS analysis showed similar results than the obtained by the cluster analysis (figura 2). Samples tended to group species according to the gear and fishing season.

38

20

G1 Similarity
40

G2

G3

G4

G5

60

80

Samples Figure 1- Cluster analysis of the fishery in the Amazon estuary of the coast of Par. T- trawling; L- longline; G- gillnet.

Figure 2- MDS analysis of the fishery in the Amazon estuary of the coast of Par. L longline; G- Gill net; T-trawler.

39

The fisheries within both the estuarine and the marine environment at a artisanal and/or industrial level. The first modality (artisanal fisheries) is the most important due to its high representativity within the landings in the Amazon area and by a great variety of fishing gears and also several species are captured especially catfishes of the family species Pimelodidae, and Ariidae, and croakers of the Sciaenidae family. Industrial fishing is characterized by the catch using trawling mainly targeting B. vaillantii. References Barthem, R. B. 1985. Ocorrncia, Distribuio e Biologia dos Peixes da Baa do Maraj, Esturio Amaznico. Bol. Mus. Para. Emlio Goeldi. Serie Zoologia. Vol. 2 (1): 49-69. Clarke, K. R. & Warwick, R. M. 1994. Change in Marine Comunites: An Appoach to Estatistical Analysis and interpretation. Plymouth marine Laboratory. 144 p. Rodrigues, L. M. & Isaac, V. J. 2002. Relaes Morfomtricas de Peixes das famlias Ariidae e Engraulidade no Esturio do rio caet, Bragana, Par. Bol. Mus. Para. Emlio Goeldi. Serie Zoologia. Vol. 18 (1): 3-17. Torres, M. F. et al. 1996. O Gerenciamento de estoques Pesqueiros: O Caso da Piramutaba. Polticas Pesqueiras nos Pases Amaznicos: 279-359.

40

BYCATCH REDUCTION AND ASSOCIATED MANAGEMENT OF THE UNITED STATES ATLANTIC SHARK FISHERY Heather Stirratt, Karyl Brewster-Geisz, Chris Rilling, Greg Fairclough, and Michael Clark* Highly Migratory Species Management Division, F/SF1 National Marine Fisheries Service/NOAA/DOC 1315 East-West Highway Silver Spring, MD 20910 (301) 713-2347; FAX (301) 713-1917 Michael.Clark@noaa.gov * Presenting Author EXTENDED ABSTRACT ONLY DO NOT CITE Background The Highly Migratory Species (HMS) management division of the National Marine Fisheries Service (NOAA Fisheries) began managing Atlantic sharks in 1993. Currently, Atlantic sharks are grouped into three management complexes for commercial harvest: large coastal sharks (LCS), pelagic sharks (PS), and small coastal sharks (SCS). These complexes are assigned regional quotas (North Atlantic, South Atlantic, and Gulf of Mexico) on an annual basis consistent with the maximum sustained yield for each complex. Bycatch of non-targeted sharks within the directed shark fishery does occur in addition to sharks that are caught in other non-HMS fisheries such as the shrimp trawl and menhaden purse seine fisheries. Some of these species are marketable and retained, while others are discarded for economic or regulatory reasons, resulting in additional bycatch. Atlantic Shark Bycatch Shark bycatch occurs predominately in trawl, gillnet, and pelagic or bottom longline fisheries. Observer data in the Gulf of Mexico menhaden fishery indicates that many LCS have been caught incidentally, up to 7.5 percent of the total LCS harvest per year, and that 75 percent of these sharks die upon release

41

(Table 1) (DeSilva et al., 2001; Cortes, 2000). Annual estimates of SCS bycatch in the Gulf of Mexico shrimp fishery ranges from 570,165 kg dressed weight (dw) to 1,468,568 kg dw (Cortes, 2002). In 2001, bycatch of LCS in the pelagic longline and bottom longline fishery represented approximately 2.8 and 5.7 percent of total mortality, respectively (Cortes et al., 2002). Table 1. Estimates of Total Landings and Dead Discards for Large Coastal Sharks: 1991-2002 (Thousands of Fish). Source: Cortes et al.2002. Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Commercial Landings 200.2 215.2 169.4 228.0 222.4 160.6 130.6 174.9 111.5 111.2 95.7 118.0 Pelagic Longline Discards 7.5 20.9 7.3 8.8 5.2 5.7 5.6 4.3 9.0 9.4 5.6 2.4 Bottom Longline Discards 11.3 12.2 11.3 16.3 13.9 7.6 8.3 9.9 3.8 4.8 6.1 Menhaden Fishery bycatch 25.1 25.1 25.1 26.2 24.0 25.1 25.1 25.1 25.1 25.1 25.1 25.1 Total 280.5 273.4 213.1 279.3 265.5 199.0 169.6 214.2 149.4 150.5 132.5 150.2

4.7

42

Management Actions The Magnuson-Stevens Fishery Conservation and Management Act of 1996 defines bycatch, in part, as fish that are harvested in a fishery, but are not sold or kept for personal use, including economic and regulatory discards. This Act also mandates that conservation and management measures shall, to the extent practicable, (a) minimize bycatch and (b) to the extent bycatch cannot be avoided, minimize the mortality of such bycatch. Amendment 1 to the Fishery Management Plan for Atlantic Tunas, Swordfish, and Sharks (NOAA Fisheries, 2003) established a rebuilding plan for overfished LCS stocks and management measures to prevent overfishing of other shark stocks. Rebuilding LCS and preventing overfishing requires implementing measures to reduce bycatch and associated mortality of prohibited and overfished species of shark, including the neonate and juvenile lifestages. The HMS Management Division is in the process of developing regulations to improve and possibly expand current bycatch reduction efforts in HMS fisheries. Management actions that are currently in place include time and area closures, gear limitations, observer coverage on shark vessels, release equipment, and vessel monitoring system requirements. Challenges facing bycatch reduction in the domestic Atlantic shark fisheries include: (1) mixed shark fisheries (non-targeted and targeted shark species encountered on the same excursion), (2) lack of detailed life history data for some species, (3) life history characteristics of sharks (late sexual maturity and few young per brood), (4) inaccurate catch reporting data from both fish dealers and fishermen (e.g. species misidentification) (5) identification and protection of essential shark habitat, (6) the need to consider and coordinate state, federal, and international shark management efforts, and (7) outreach efforts to recreational and non-HMS fisheries that only catch sharks occasionally. Literature Cited Cortes, E. 2000. 2000 Shark Evaluation Report. NOAA, NOAA Fisheries, Southeast Fisheries Science Center, Panama City, FL. SFD-00/01-119. 23pp. Cortes, E. 2002. Stock Assessment of Small Coastal Sharks in the U.S. Atlantic and Gulf of Mexico. NOAA Fisheries, Southeast Fisheries Science Center. Panama City, FL. Sustainable Fisheries Division Contribution SFD-01/02152. 133pp.

43

Cortes, E., L. Brooks, and G. Scott. 2002. Stock Assessment of Large Coastal Sharks in the U.S. Atlantic and Gulf of Mexico: Final Meeting Report of the 2002 Shark Evaluation Workshop. National Marine Fisheries Service, Southeast Fisheries Science Center, Panama City, FL. 133pp. DeSilva, J. A, R. E. Condrey, and B. A. Thompson. 2001. Profile of shark Bycatch in the U.S. Gulf of Mexico menhadden fishery. North American Journal of Fisheries Management 21:111-124. NOAA Fisheries. 2003. Amendment 1 to the Fishery Management Plan for Atlantic Tunas, Swordfish, and Sharks. National Oceanographic and Atmospheric Administration, Highly Migratory Species Management Division, Silver Spring, MD.

44

AN ATTEMPT TO DISENTANGLE THE COMPLEXITY OF AMAZON FISHERIES. (THE MANAUS FISHERIES 1976-1988) Bernard de Mrona IRD-LEHF, Universit Cl. Bernard Lyon1 43 Bd. 11 novembre 1918, 69622 VILLEURBANNE, France phone. 33 (0)4 72 43 28 90. fax. 33(0)4 72 43 11 41 e-mail. Bernard.de-merona@univ-lyon1.fr Laure Bernardac-Gardel IRD, centre de Cayenne B.P. 165, 97323 CAYENNE Cedex e-mail. lgardel@popserver.cayenne.ird.fr Introduction The management of a fishery consists in adjusting the level of removal of an aquatic resource from the environment with its natural renewal. Even in the case of a single species exploited by a single fishing gear, this task is a difficult one. It implies a detailed knowledge of the biology of the exploited population and of the fishing gear efficiency including the natural variability and heterogeneityof these two components. In the case of artesanal tropical fisheries the complexity is at its maximum. Most of these fisheries target a number of species, involve many different actors using a variety of fishing gears and making use of many fishing strategies. Whatever the method used to analyze such fisheries (global or analytical models), the identification of relatively homogeneous subsets is unavoidable. This approach of decomposing the fishery was developed by Ralston and Polovina (1982) to analyze the long-line fishery in Hawaii, and was explored in many recent works on tropical fisheries (Laloe et al, 1998; Pech et al. 2001). Amazon fisheries are probably the most complex in the world (Meschkat, 1961; Petrere, 1998b; Mrona, 1990; Mrona and Bittencourt, 1991, 1993). They operate on a very large scale, in a heterogeneous and variable environment; the fish fauna is composed of more than 2000 species, most of them potentially exploitable; fishing practices are very diverse, with boats of capacity from 3 to almost 100 tons and utilizing a number of fishing gears. Analyses of some aspects of this fishery have already used subsets of data based on empirical observations (Petrere, 1986a, b; Mrona and Bittencourt, 1988).Indeed, the grouping of some landings exclusively composed of a species captured by a unique fishing gear is obvious. It is notably the case of the

45

tambaqui captured by gillnets. However, most landings are multispecificmultigear and the management of the fishery must take into account the interactions between the different fishing strategies developed in the fishery. In that way the purpose of this work is to explore the possibility to identify homogeneous subsets of the Manaus fishery in an objective way. We use the notion of metier or fishing strategies, as defined as the association of a species or group of species, a fishing gear or an association of gears, a fishing area and a fishing strategy. To proceed with this attemp we consider a set of continuous landing data in Manaus harbor from 1976 to 1988. Methods Landing data collection was established by Petrere in 1976 and described in detail (Petrere, 1978b). Data were collected every day from interviews with fishermen. At that time, the fish was landed at 11:00 p.m. after bargaining between the fishermen and the middlemen, leaving time for the collector to visit every boat and check the information given. Data included the date of landing, the number of days traveling, the number of days fishing, the fishing ground and its distance to Manaus, the number of fishermen participating, the number of canoes used, characteristics of the fishing gears and the list of species with their respective weight or number. For the fish commercialized by count, a transformation in weight was achieved based on measures of the mean weight of the fish in the market. In order to conduct the analysis some transformations were performed on this original file. We introduced 3 new variables: 1) the code of the main species, the one with the greatest quantity captured, 2) the percentage of the main species capture in relation to total capture and 3) the gears association, a concatenation of the codes when different gears were used in the fishing trip. We calculated the fishing effort following Petrere (1978a), multiplying the number of days of fishing by the number of fishermen participating. In order to introduce quantitative (capture, distance to harbor, effort) as well as qualitative (coded species and gears) variables in the analysis, we transformed all variables in dummy variables. That is each variable is represented by n modalities with values 0 or 1. From a careful examination of the distribution of data we coded: - the distances from Manaus harbor in 4 classes : 0-99 km ; 100-267 km ; 268-468 km ; 469-2000 km - the effort in 4 classes : 0-18 ; 19-34 ; 35-57 ; 58-900 fishermen.days

46

the capture in 4 classes : 0-1487 ; 1488-3468 ; 3469-7082 ; 7083892079 kg

We used only the 5 more abundant species which are: jaraqui (1), tambaqui (2), curimata (3), pacu (4) and tucunar (5) and gave the code 6 for other species. The 6 main fishing gears were coded 1: open water seine, 2: gillnet, 3: beach seine, 4: line, 5: long line and 6: traditional gears including trident, harpoon, bow and arrow, etc. The data were treated by a Correspondence analysis using SAS software. A hierarchical clustering using Euclidean distance and average link was then applied on the coordinates on the first factors of the analysis in order to identify individual subsets. Results The three first factors of the correspondence analysis extracted 45.86% of the total inertia of the data set. Each of the following factors extracted less than 10%. The first factor separated the species tucunar associated with lines and traditional gears (Fig. 1).

Figure 1. Projection of variables on the two first axis of the correspondence analysis.

47

Factor 2 separated the species tambaqui associated with gillnets and large efforts and distances from harbor. The third factor (12.9% of the total inertia) contrasted the seines and the gillnets, but did not clearly isolate a group of variables. The hierarchical classification of the variables shows a clear identification of three main groups (Fig.2). A group associated the tucunar with lines and traditional gears, another group the tambaqui with gillnets and large efforts and distances and a last large group was composed of all other species, gears and fishing trips characteristics.

Figure 2. Dendrogramme of the hierarchical clustering of variables. This preliminary analysis was then able to identify 3 homogeneous fisheries. One targeted the tucunar which was captured with trident and hand lines. Another focused on the tambaqui captured by gillnets and worked on fishing grounds localized far from the harbor. Finally a more complex fishery targeted some migratory species as the jaraquis, the curimata and the pacus and used two types of seines. Altogether these 3 identified fisheries accounted for 81% of the

48

landings and 79% of the quantities landed (Fig. 3). Indeed, correspondence analysis as well as hierarchical clustering focused on the main associations masking the weaker ones. An analysis of the remaining landings showed that they are mainly composed of tambaqui which represented more than 70% in weight and that 63% of these landings were using open-water seines. This observation allows the identification of a fourth fishery associating tambaqui with open-water seines.

Figure 3. Relative importance of the different fisheries. The tucunar fishery The tucunar fishery represented 10.3% of the fishing trips over the period of study but it produced only 3.3% of the quantities landed. The tucunar made up 78.7% of the capture. Other species included the aruan (10%), acaras (7.6%), and pescadas (1.5%) all species known to be sedentary. The most used fishing gear was the trident. Used alone it captured more than 82% of the total quantity and in association with other gears, 13%. Effort was relatively low, generally lower than 100 fishermen.days with teams from 4 to 11 fishermen and campaigns of 10 days or less (Fig.4). This fishery operated on fishing grounds near Manaus in varzea lakes. The most frequented site is the lago do Rei situated at a distance of 85 km from Manaus at the confluence of the Solimes with the Negro. On average, during the observation period, the capture did not show a strong seasonality, but the highest captures occurred in the low-water season. During the flood, the fish is more dispersed and more difficult to capture.

49

Figure 4. Parameters of the tucunar fishery. a: repartition of fishing trip length; b: repartition of effort; c: interanual variations of capture and effort (Two months of data are missing in 1979); d: seasonality of capture. The production followed a decreasing tendency except in 1979 when the production was multiplied by three. This phenomenon is even more accentuated than it appears if we consider the two months of data missing for that year. Total yearly effort followed the same decreasing tendency. The tambaqui-gillnet fishery This fishery was concerned with more than 7000 landings, representing 13% of the fishing trips and 13% of the quantities landed over the period. It was the activity which developed the greatest effort with long trips to go to fishing grounds situated at more than 300 km from Manaus (Fig 5). Some boats made the trip as far as the Peruvian border at about 2000 km. The average effort was 70 fishermen.days but it went until 400 fishermen.days. The species landed together with the tambaqui represented less than 5% of the quantities. These secondary species were of two types. On one hand there were sedentary species caught by active gears like trident, harpoon or lines that the fishermen used during the time left between the gillnets visits. These species were mainly the pirarucu (1.65%), the tucunar (1.1%), and the aruan (1.0%). On another hand there was a species closely related to the tambaqui: the pirapitinga which made 0.7% of the total capture.

50

Figure 5. Parameters of the tambaqui-gillnet fishery. a: repartition of fishing trip length; b: repartition of effort; c: interanual variations of capture and effort. (Two months of data are missing in 1979); d: seasonality of capture Gillnets alone were associated with 84.5% of the landings. In the remaining landings gillnets were associated with active gears like the trident (10.5% of the cases), the harpoon (1.9%), hand-lines (1.4%) or the open-water seine (0.9%). Capture was minimum during the low-water season in October and November and maximum during the intermediate periods of rising and falling water levels. During the time of high water when the floodplain forests are inundated, the tambaqui enters in the forests to eat the fruits and seeds falling from the trees (Goulding, 1980). The between year variability of the flooding intensity, and especially the appearance of El Nio episodes, led to large variations in monthly captures. The production of this fishery was maximal the two first years of observation but stabilized afterwards around 1500 tons.year-1. The total effort dedicated followed the capture tendencies except in 1985 where large effort did not produce any increase in the production. It seems then that the fishermen adjust their effort to the production potential of the floodplain. The migratory fish fishery The migratory fish fishery was the most important fishery in term of number of trips as well as in quantities landed. It represented 56% of the trips and 65% of the landings.This strategy was very active in the vicinity of Manaus inside a

51

radius of about 200km but is regularly carried out up to 500km from the harbor (Fig. 6).

Figure 6. Parameters of the migratory fish fishery. a: repartition of fishing trip length; b: repartition of effort; c: interanual variations of capture and effort (Two months of data are missing in 1979); d: seasonality of capture. In general the trips were short and the number of fishermen participating less than 10. The species were almost exclusively migratory. The jaraqui represented 39.7% of the captures, the curimata 14.4%, and the pacu 12.1%. Beside these 3 main species which were regularly brought to the market, many other migratory species were captured occasionally. These were the piramutaba (7.9%), the sardinhas (7.0%), the matrinch (6.7%), the pirapitinga (5.9%), the aracu (2.4%) and the branquinhas (1.4%). To catch such species fishermen used the seines, either in open water, or along the beaches. 83.6% of the landings used open water seines and 12.45% beach seines. The majority of the remaining landings were performed with long-lines.The fishery was seasonal. There was a large capture of jaraqui en May-June at the peak of the floods and an increase of the production at the end of the falling of the water when many fishes leave the floodplain. The annual production of the migratory fish fishery appeared to be stable over the observation period, although the two last years showed a decrease in the capture. The total capture followed roughly the capture of the three main species. The main exceptions were in 1976 and 1979 when large captures of sardinhas and piramutaba respectively were observed. The total effort was apparently correlated with the capture.

52

The tambaqui-seine fishery In terms of number of trips as well as quantities landed, the tambaqui-seine fishery was more important than the tambaqui-gillnet fishery. It represented 21% of the total number of trips and 19% of the capture. The tambaqui made 72.1% of the capture, a figure which was related to the use of open-water seine in 63.3% of the trips. Other species were the aruan (8.3%), the pirarucu (4.9%), the pescada (3%) and 9 other of diminutive importance. These were captured mainly with gillnets (14% of the trips) and trident (11%). To catch the tambaqui with seine, fishermen did not go very far from Manaus. The fishing grounds visited were localised at a distance lower than 500 km from the harbor with the majority of the trips included in a radius of 200 km (Fig. 7). The fishing effort developed in this fishery was relatively weak. The number of fishermen was most often 10 or lower and the number of days fishing was generally lower than 15. The quantities captured in the high water season were 4 times less than those in the low water season. This figure appears to be opposed to that observed for the tambaqui-gillnet fishery. Indeed, unlike the gillnets, seines cannot operate in the flooded areas because of the presence of trees. Capture was very variable from one year to another, but no tendency could be detected over the period. After a peak in 1979 when the capture was over 7000 t, the landings decreased and stayed low during the period 1981 to 1986. In 1987 a new peak was observed. The annual effort dedicated to this fishery is extremely high. Even in the period of low capture the effort was increasing. This is probably due to the high price of the tambaqui on the market.

53

Figure 7. Parameters of the tambaqui-seine fishery. a: repartition of fishing trip length; b: repartition of effort; c: interanual variations of capture and effort (Two months of data are missing in 1979); d: seasonality of capture. Conclusions and Discussion At first sight, the Manaus fishery appeared to be extremely complex with many different target species and fishing strategies and this complexity makes any tentative of management difficult. However, some strategies were relatively easy to identify and, in the past, allowed some interesting analyses. Data on the tambaqui- gillnet fishery as well as the tucunar-trident fishery were used to analyze variations in the relative abundance of these two species between 1976 and 1978 (Petrere, 1986a & b). Mrona & Bittencourt (1988) conducted a preliminary analysis on the tambaqui and the jaraqui based on the global model of production by effort. Nevertheless, identification of subsets in the fishery was then based on a subjective appreciation of the characteristics of landings. The approach proposed here, based upon multivariate analyses techniques, allowed a more objective identification of the landings repartition. It revealed the existence of 4 types of fishing strategies. Three of them targeted a single species or, at least, in the case of tucunar, two closely related species. The other strategy was more complex. It targeted a group of species, with homogeneous size, and which share the particularity to carry out migrations. More than 15 years passed from the end of the period studied in this work and the present. It is doubtless that many changes occurred in the fishing activity in the Amazon region. However, it is likely that the main fishing strategies continue to be operational even if

54

adaptations (tactics) have taken place to cope with changes in abundance and repartition of the stocks. In that way, this contribution aims to stimulate the studies of Amazon fisheries. Of course, identification of individual fisheries is only the first preliminary step in the process of managing a fishery. Application of models will require much more detailed studies, particularly on the ecology of the species, the interactions between different fisheries and the effects of environmental parameters (see Mrona & Gascuel, 1993). Acknowledgements The authors are grateful to ORSTOM (now IRD) and INPA for funding and supporting this research. This work could not be done without the dedication of the collectors on the market. Many thanks to Raimundo Freitas de S, Francisco Fonseca da Silva, Raimundo Soteiro da Silva and Arlindo Batista do Nascimento. References Goulding, M. 1980. The Fishes and the Forest. University of California Press, Berkeley, Los Angeles, London: 280 p. Lalo, F., Pech, N.,Sabatier, R. and A. Samba. 1998. Model identification for flexible multifleet-multispecies fisheries: a simulation study. Fish. Res. 37 : 193-202 Mrona, B. de. 1990. Amazon fisheries : general characteristics based on two case-studies. Interciencia 15(6) : 461-468 Mrona B. de and D. Gascuel. 1993. The effects of flood regime and fishing effort on the overall abundance of an exploited fish community in the Amazon floodplain. Aquat. Living Resour. 6 : 97-108 Mrona B. de and M. M. Bittencourt. 1988. A pesca na Amazonia atraves dos desembarques no mercado de Manaus : resultados preliminares. Mem. Soc. de Cienc. Nat. La Salle XLVIII (Sup.) : 433-453 Mrona, B. de and M.M. Bittencourt. 1991. La pche artisanale en Amazonie centrale : approches et difficults. In : La Recherche face la pche artisanale Symposium International, Montpellier France 3-7 juillet 1989.

55

J. R. Durand, J. Lemoalle and J. Weber (eds.). Paris, ORSTOM, t. 1 : 433441 Mrona, B. de and M.M. Bittencourt. 1993. Facteurs et contraintes de la pche de march en Amazonie centrale : le cas dun lac de plaine inonde (le lago do Rei , Amazonas, Brsil). Amazoniana XII(3-4) : 443-465 Meschkat, A. 1961. Fisheries of the Amazon region. Report to the Government of Brazil. FAO, 77p. Pech, N., Samba, A., Drapeau, L., Sabatier, R. and F. Lalo. 2001. Fitting a model of flexible multifleet-multispecies fisheries to Senegalese artisanal fishery data. Aquat. Liv. Res. 14 : 81-98 Petrere, M. Jr. 1978a. Pesca e esforo de pesca no Estado do Amazonas. I Esforo e captura por unidade de esforo. Acta Amazonica 8(3) : 439-454 Petrere M. Jr. 1978b. Pesca e esforo de pesca no Estado do Amazonas. II Locais, aparelhos de captura e estatisticas de desembarque. Acta Amazonica, 8(3) (sup. 2) : 54 p. Petrere M. Jr. 1986a. Amazon fisheries. I Variations in the relative abundance of tambaqui (Colossoma macropomum CUVIER, 1918) based on catch and effort data of the gill-net fisheries. Amazoniana IX(4) : 527-547 Petrere M. Jr. 1986b. Amazon fisheries. II Variations in the relative abundance of tucunar (Cichla ocellaris, C. temensis) based on catch and effort data of the trident fisheries. Amazoniana X(1) : 1-13 Ralston, S. and J.J. Palovina. 1982. A multispecies analysis of the commercial deep-sea handline fishery in Hawai. Fish. Bull. 80 : 435-448

56

CHANGES IN FISHING LANDINGS AS A FUNCTION OF THE HYDROLOGICAL CYCLE. Renato Soares Cardoso Master Student, Freshwater Biology and Inland Fishery, Amazon Institute National Research. 2936 Andr Arajo Avenue, Aleixo, CEP 69023-000. Manaus, Amazonas, Brazil. Phone: (92) 644-7787 E-mail: renato4@inpa.gov.br Carlos Edwar de Carvalho Freitas Fisheries Science Department, Amazon Federal University. E-mail: cefreitas@ufam.edu.br EXTENDED ABSTRACT ONLY - DO NOT CITE Introduction The Madeira River is the first tributary of right margin of the Amazon River in water flow and second in watershed basin. Its fishery resource has been exploited continuously for the fishery fleet of both towns Manaus and Porto Velho (Goulding, 1979), sustain too the small-scale fishery destined for supplying of municipal district and riverines communities in your basin. In 1990s the fishery of Madeira River basin was responsible for approximately 14% of all fishing landing in Manaus town (Batista & Petrere Jr., 2003), nevertheless this region is unprovided on information about fishery, and in this sense the present study aimed to assess the changes occurred in the fishing landing in function of hydrological cycle in the medium Madeira River. Methods Were sampled the fishing landings between January 2002 until December 2002 in Manicor town, located in the southeastern portion of the Amazonas State, seeking to evidence patterns of exploration of fishes species and fishing grounds

57

for fishers in accord with hydrological cycle of Madeira River, using for this the Correspondence Analysis (McGarigal, 2000). Data of fishing landing had been acquired daily using structuralized questionnaires. The information was about caught species and fishing grounds. The Madeira River level data was acquired from Brazilian Water Agency. Results The Manicor fishing fleet it is composed of boats and canoes with outboard machines. It operates in the municipal geographic limits, in main channel of Madeira River and tributaries including streams and lakes. The agents responsible for fishing landing are fishers of boats, motorized canoes and fish purchasers. The municipal fleet landed 31 species or groups of species, with special importance for jaraqui (Semaprochilodus spp), pacu (Mylossoma spp) curimat (Prochilodus nigricans), sardinha (Triphorteus spp) and jatuarana (Brycon cephalus) responsible for 75% of fish harvest. This five species are catch mainly in the drought season, however jatuarana it is present in the harvest of flood season. The twelve principal species in term of biomass are presented in Table 1. Table 1. Principal species landed in Manicor city (weight in Kg). # 01 02 03 04 05 06 07 08 09 10 11 12 13 Common name Jaraqui Pacu Curimat Sardinha Jatuarana Branquinha Dourada Caparari Aracu Surubim Pescada Tamoat Other species Specie Semaprochilodus spp Mylossoma spp; Metynis spp Prochilodus nigricans Triportheus spp Brycon cephalus Potamorhina spp Brachyplatystoma flavicans Pseudoplatystoma tigrinum Schizodon fasciatum; Anostomus spp Pseudoplatystoma fasciatum Plagioscion spp Hoplosternum litoralle Weight 54,534 42,111 26,235 24,099 18,436 16,258 6,774 4,395 4,057 3,640 2,482 2,443 14,286

58

The fishers exploit 31 fishing grounds, the most important in frequency terms are Madeira and Manicor rivers, Matupiri Stream and Mirautuba Lake. The Madeira River it is exploited for fishing throughout year, mainly in drought season. The Manicor River is the most exploited in flood season. The lakes usually are utilized in flood season for motorized canoes. In the flood season, boats use mainly rivers and motorized canoes use lakes as fishing grounds. Stream not presented standard of use for fishing agents (Figure 1).
0,16 0,14 Dimension 2; Eigenvalue: 0,00164 (3,050% of Inertia) 0,12 0,10 0,08 0,06 0,04 0,02 CANOE 0,00 -0,02 BOAT -0,04 -0,06 -0,4 river lake stream

PURCHASER

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

Dimension 1; Eigenvalue: 0,05203 (96,95% of Inertia)

Figure 1. Graph of two first dimension of Correspondence Analysis with data of seasonality (flood season). For drought, this standard is altered (Figure 2) with canoes fishing mainly in rivers. The purchaser do not present standard for both flood and drought season.

59

0,10

Dimension 2; Eigenvalue: 0,00117 (2,478% of Inertia)

0,08

PURCHASER

stream

0,06

0,04

0,02 river 0,00

CANOE

-0,02

-0,04

BOAT

lake

-0,06 -0,8

-0,6

-0,4

-0,2

0,0

0,2

0,4

0,6

Dimension 1; Eigenvalue: 0,04620 (97,52% of Inertia)

Figure 2. Graph of two first dimension of Correspondence Analysis with data of seasonality (drought season). Discussion The use of correspondence analysis in the fishery research is recent in Amazonian region (Freitas et al., 2002, Yamamoto, 2004). This approach was useful for to analyze the changes occurred in fishing landing in this region of Madeira River. The absence of the use of lake by boats of municipal fleet fishing is a characteristic of the region, this confers a great advantage for motorized canoes because exploit this fishing grounds. The pattern shown by fishers in the use of fishing grounds is explained by facility in displacement, mainly in rivers, because of nonexistence of prohibition for fishing by riverine communities as occur in lakes.

60

Acknowledgements The authors would like to thank CNPq, for financial support and all fishers of Manicor town for information that possibility the conclusion of study. References Batista, V.S. and Petrere Jr., M. 2003. Characterization of the commercial fish production landed at Manaus, Amazonas State, Brazil. Acta Amazonica, 33(1): 53-65 Freitas, C.E.C.;Batista, V.S. and Inhamuns, A.J. 2002. Strategies of the smallscale fisheries on the Central Amazon floodplain. Acta Amazonica, 32(1): 101-108. Goulding, M. 1979. A ecologia da pesca no rio Madeira. CNPq/INPA. Manaus. 172p. McGarigal, K.; Cushman, S. and Stafford, S. 2000. Multivariate statistics for wildlife and ecology research. Library of Congress, 283p. Yamamoto, K.C. 2004. A estrutura de comunidades de peixes em lagos manejados da Amaznia Central. Master Thesis. Instituto Nacional de Pesquisas da Amaznia/Universidade Federal do Amazonas. 71p.

61

62

AN EXPERIMENTAL EVALUATION OF ALTITUDINAL SPECIES ZONATION PATTERNS IN MONTANE STREAMS: DO ABIOTIC OR BIOTIC FACTORS DETERMINE THE DISTRIBUTION OF NATIVE AND NONNATIVE TROUT IN UTAH, USA, RIVERS? Peter McHugh* Utah Cooperative Fish and Wildlife Research Unit Department of Aquatic, Watershed, and Earth Resources Utah State University Logan, UT 84322-5210 PH: 435-757-9401 FAX: 435-797-4025 Email: petemchugh@cc.usu.edu Phaedra Budy Utah Cooperative Fish and Wildlife Research Unit Department of Aquatic, Watershed, and Earth Resources Utah State University Logan, UT 84322-5210 PH: 435-797-7564 FAX: 435-797-4025 Email: phaedra.budy@.usu.edu EXTENDED ABSTRACT ONLY - DO NOT CITE Introduction Patterns of species replacement commonly occur along altitudinal gradients in montane streams. In Utah, introduced brown trout (Salmo trutta) replace threatened cutthroat trout (Oncorhynchus clarki utah) progressively from headwater to downstream reaches. Three hypotheses regarding causes of species-zonation patterns are applicable to montane streams, where thermal gradients parallel elevation. First, fish only establish where their physiological tolerance permits (Moyle and Light, 1996). Thus, zonation patterns reflect the response of species to the arrangement of suitable temperatures along the gradient. Second, condition-specific competition (Dunson and Travis, 1991)

63

between upstream and downstream species confers invasion resistance to either one at temperatures near their optima. Therefore, one species dominates under warm- while the other does under cold-temperature conditions (DeStaso and Rahel, 1994; Taniguchi and Nakano, 2000); zonation is due to interactions mediated by physiology. Third, the distribution of one species is limited by temperature and the other by competition with the former. Though experimental data addressing these hypotheses are generally lacking, distinguishing between them has implications for native fish conservation. Therefore, our goal was to assess the influence of biotic and/or abiotic factors on the species-zonation pattern observed in the Logan River, Utah. Methods To address our goal, we conducted an experiment during the summer of 2003. We replicated three treatments (1. allopatric brown trout; 2. allopatric cutthroat trout; 3. both species in sympatry) each at six sites arrayed along the altitudinal gradient present in the Logan River. We assigned all treatments to enclosures at each site. All fish were stocked at ambient density using wild age-1 trout. Before the experiment, fish were held in a common environment for 8-d, at which time we weighed, measured, and tagged them. We introduced fish in early July and subsequently cleaned enclosures during a 42-d trial. After this, we removed, weighed, and measured all fish. We quantified performance of both species using standard growth and condition metrics. Ending condition was assessed on an individual basis using relative weight (Wr). We evaluated relative growth based on cage-averaged weights; we could not evaluate it individually due to tag loss. We assessed whether temperature controls distribution by evaluating allopatric fish performance as a function of elevation. To assess the strength of interspecific competition, we compared growth and maximum relative weight between allopatric and sympatric treatment groups for each species separately using ANOVA or ANCOVA. Results Patterns of allopatric condition indicate that summer conditions do not preclude brown and cutthroat trout from high- and low-elevation sites, respectively (Figure 1). When reared alone, both brown and cutthroat trout grew similarly across all elevations, a trend discordant with their distributional pattern.

64

However, cutthroat trout suffered an apparent cage effect, with condition values generally less than 100%. Relative growth averaged 37% and 48% for allopatric brown and cutthroat trout, respectively, indicating positive growth in enclosures. In sum, considering performance as a function of elevation suggests that summer temperatures do not limit the distribution of either species (at age-1) or cause species zonation. Competition may explain the replacement of cutthroat by brown trout in the downstream direction. The presence of brown trout caused suppressed cutthroat performance (maximum Wr, ANOVA, F1,9 = 9.58, P = 0.013, Figure 2; relative growth, ANOVA, F1,9 = 2.85, P = 0.126); cutthroat trout condition was 5-15% and relative growth ~25% higher in the absence of brown trout. Conversely, the presence of cutthroat trout had no effect on relative growth (ANCOVA, F1,8 = 1.89, P = 0.206) or condition (ANOVA, F1,9 = 1.57, P = 0.242, Figure 2) of brown trout. These results suggest that brown trout can displace cutthroat trout when they co-occur and are unaffected by their presence. Conclusion We conclude that the brown-cutthroat trout zonation pattern is the result of biotic interactions and physiological limitations. We demonstrated that cutthroat trout grew similarly at all elevations in allopatry - a pattern discordant with their absence from lower elevations. Further, we show that brown trout can have a negative effect on their performance. Thus, the deficiency of cutthroat trout in low-elevation reaches is likely due to competition with brown trout, demonstrating a biotic control on cutthroat distribution. Conversely, the existence of a high-elevation limit for brown trout is likely due to abiotic controls affecting recruitment during a lifestage not considered here. Supporting observations are: 1) brown trout performed well at all sites, suggesting neither summer temperatures nor competition with cutthroat precludes them from high altitudes; 2) an inspection of brown trout gonads suggests that they can grow well enough to mature/spawn at high-elevation sites; 3) there are no barriers to brown trout dispersal into upper sites. Thus, brown trout can access/spawn at high-elevation sites, but do not attain high abundance there because of recruitment failure occurring between egg deposition and the age-1 lifestage. Given the concordance of thermal changes and species replacement, temperature is a likely candidate factor.

65

References De Staso, J. III, and F.J. Rahel. 1994. Influence of water temperature on interactions between juvenile Colorado River cutthroat trout and brook trout in a laboratory stream. Transactions of the American Fisheries Society 123:289-297. Dunson, W.A., and J. Travis. 1991. The role of abiotic factors in community organization. The American Naturalist 138:1067-1091. Moyle, P.B., and T. Light. 1996. Fish invasions in California: do abiotic factors determine success. Ecology 77:1666-1670. Taniguchi, Y, and S. Nakano. 2000. Condition-specific competition: implications for the altitudinal distribution of stream fishes. Ecology 81:2027-2039.

66

120 110 100 90 80

brown trout

Percent relative weight

70 60 50 110 100 90 80 70 60 50 1400 1500 1600 1700 1800 1900 2000 2100 2200

cutthroat trout

Elevation (m )
Figure 1. Median (symbols) and range (upper and lower whiskers) of brown trout (upper panel) and cutthroat trout (lower panel) relative weight values plotted as a function of elevation in the Logan River. Circles correspond to enclosures where as species was reared alone (allopatry), while squares represent those where both species were together (sympatry). For reference, the horizontal line at Wr = 100% indicates that enclosure fish performed as well as wild, river fish.

67

110 105 100 95 90

brown trout

Maximum percent Wr (+ SE )

85 80 75 70 110 105 100 95 90 85 80 75 70 Allopatry Sympatry

cutthroat trout

Figure 2. Maximum relative weight (mean + 1SE) for brown trout (upper panel) and cutthroat trout (lower panel) reared in the presence (sympatry) or absence (allopatry) of the other species. An asterisk denotes statistical significance at = 0.10. See text for statistical test details.

68

THE IMPORTANCE OF CONNECTION CHANNELS BETWEEN LAKES AND AMAZONIAN RIVERS AS SIDEWAY TO MIGRATORY SPECIES. Raniere Garcez C. S.1 Federal University Amazon Av. Gen. Rodrigo Otvio, 3000 Manaus Amazon Brazil (092) 647-4067 raniere.garcez@bol.com.br Carlos E. C. Freitas. 1 Federal University Amazon (092) 647-4067 cefreitas@horizon.com.br Abstract The ecology of large rivers systems with adjacent flooplains are markedly influenced by floodpulse. In the Amazonian rivers, many fish species, mainly Characiformes, developed migratory strategies to explore successfully different environments. Some of these species are the most relevant to the regional fishing landings. We sampled the connection channel between two lakes and the Solimes River aiming to identify the use of this transient environment as sideway to migratory route of commercially important fish species. Introduction The floodplains, including its lakes and igaraps, have fundamental importance to life cycle of many fish species (Cox-Fernandes & Petry, 1991). This environment is used by several fish species to reproduction, nursery, feeding and refuge (Lowe-McConnell, 1987). The aim of this study is to identify the migratory and non-migratory inhabitant fish species of the connection channels during two season of hydrological cycle and identify the main patterns of these communities.

69

Materials and Methods This study was accomplished between to the Solimes-Amazonas river to the channel of Cururu and channel of Jacar lake.Experimental fisheries were performed two times at two season of hydrological cycle: To minimize the effects of selectivity, gillnets with different mesh size were used. Overall, the gillnets staying 18 hours in the water. In general, fishes were identified, measured and weighed soon after the catch. After formalin fixation, some fishes were kept in plastic bags and carried out to identification by experts. A principal component analysis (PCA) was accomplished to identify driving factors of patterns of community structure (Manly, 1994). The subjects are amostral units composed by connection channel (C = channel of Cururu lake and J = channel of Jacar lake) and season of experimental fishery (1, 2 = rising and 3, 4 = receding). Results A total of 1107 specimens representing 6 orders, 20 families, 56 genera and 78 species were collected. Characiformes was the most abundant order with 744 specimens or 67,27%, follow by Siluriformes with 28,12%, Osteoglossiformes with 1,99%, Perciformes with 1,45%, Clupeiformes with 1,08% and Gymnotiformes with 0,09%. Also Characiformes was the most diverse group with 44 species or 56,41% of total. The greater diversity was observed at channel of Cururu lake during dry season and the smaller at the same channel during rising. The greater dominance was noticed at channel of Cururu lake during rising and the smaller at channel of Jacar lake during receding. Evenness was smaller at channel of Cururu lake during receding and greater at Jacar lake during flooding season. The principal component analysis (Figure 1) allowed the identification of three groups of composed units. The first group is composed by samplings developed at channel of Cururu lake during high water period (C1 and C2 units). The second group is composed by samplings accomplished at channel of Cururu Lake during dry season (C3 and C4). The last group is composed by samplings accomplished at channel of Jacar lake in both seasons.

70

Assessing the factor scores in the PCA, the species driving groups formation are: Psectrogaster rutiloides, Potamorhina latior and Semaprochilodus insignis that were abundant at channel of Cururu lake during the high water period. At the same place, Parauchenipterus galeatus, Mylossoma duriventre, Schizodum fasciatum and Rhaphiodon vulpinus are determinants to group composed by samples accomplished during drying season. The samples accomplished at channel of Jacar lake were jointed mainly by Colossoma macropomum, Hoplosternum litoralle, Prochilodus nigricans, Potamorhina pristigaster and Triportheus flavus.
1,0

J4 0,5 J2 Factor 2 : 21,44% J1

J3 C3 C4

0,0

-0,5

C2 C1

-1,0 -1,0 -0,5 0,0 Factor 1 : 23,74% 0,5 1,0

Figure 1. Ordination of composed units: C = channel of Cururu lake; J = channel of Jacar lake;1 and 2 = rising; 3 and 4 = receding.

Discussion Possibly, the stochastic pattern was a result of environment transience. In the Amazonian basin, the channels connecting floodplain lakes and rivers quite dry during the low water phase. Beside, it constitutes a sideways to migratory fish

71

species (Cox-Fernandes & Petry, 1991). Thus, these assemblages are composed of (i) little resident species (e.g. Chaetobranchopsis orbicularis, Cichla monoculus and Plagiosciom squamossissimus) which exhibit seasonal changes of abundance; (ii) opportunist species that quickly colonize these environments after dry periods (e.g. Pygocentrus natterreri, Serrasalmus calmoni and S. eigenmanni); and migratory species which use it as a sideway and are commercially important to regional fish landings (e.g. Prochilodus nigricans, Semaprochilodus insignis and S. taenirus). Once channel of Cururu lake exhibit difference of species composition between flooding and drying seasons, while channel of Jacar lake showed similar composition in both seasons, the discrepancies between lakes channel are due, probably, differences in landscape characteristics of these channels that determine changes in colonization timing. References Cox-Fernandes, C.; Petry, P. 1991. A importncia da vrzea o ciclo de vida dos peixes migradores na Amaznia Central. In: Val, A. L.; Fligliuolo,R.; Feldberg, E. (eds.) Bases Cientficas para Estratgias de Preservao e Desenvolvimento da Amaznia: Fatos e Perspectivas. V. 1, Parte IV Animais da Amaznia. Capitulo 12 - Recursos Pesqueiros. Manaus. p. 315320. Lowe-McConnell, R. H. 1987. Ecological studies in tropical fish communities. Cambridge University Press, Cambridge (UK). 382p.

72

73

74

A SURVEY OF THE FISH FAUNA ALONG THE FLOODPLAIN OF THE AMAZON RIVER IN BRAZIL Jansen Zuanon, Instituto Nacional de Pesquisas da Amaznia INPA, Coordenao de Pesquisas em Biologia Aqutica - CPBA, CP 478, Avenida Andr Arajo, 2936, Petrpolis, 69083-970, Manaus, AM. zuanon@inpa.gov.br Luiz H. Claro-Jr., INPA/CPBA, clarojr@inpa.gov.br Fernando P. Mendona, INPA/CPBA, mendonca@inpa.gov.br Efrem J. G. Ferreira, INPA/CPBA, efrem@inpa.gov.br Lcia H. Rapp Py-Daniel, INPA/CPBA, rapp@inpa.gov.br EXTENDED ABSTRACT ONLY: DO NOT CITE The Amazon basin is composed by countless rivers, which form the largest hydrographic basin of the world with ca. 7 x 106 km2. These rivers differ in the origin and morphology of its courses, as well as in the physico-chemical properties of its waters (Goulding et al., 2003). An annual flood pulse that results in high biological productivity in the large and turbid rivers such as the Amazon, Madeira and Purus, marks the dynamics of the aquatic environment. During the flood season these rivers overflow and invade enormous adjacent areas, locally known as vrzeas. These areas in the Brazilian section of the Amazon River account for approximately 92,390 km2 (Sippel et al., 1992). The human populations have used the vrzea in a predatory way, notably in the last few decades. These activities may jeopardize the vrzea environmental

75

integrity and lead to an impoverishment of the regional diversity, or even decimate the populations of certain economically valuable species. This is especially true for some fisheries resources, which have been exploited carelessly for centuries in the region. Nevertheless, despite its regional importance for the commercial and subsistence fisheries, very few is known about the fish species richness and distribution along the vrzea floodplains. This lack of information can impair the efforts to preserve its fish diversity, since no comparative data is available about the patterns of species occurrence along the Amazon River floodplain. Forsberg (2000) proposed a division of the Amazon River floodplain in five zones, according to geomorphologic and landscape characteristics: 1o) an estuarine zone, from Santana (Amap State) to Almeirim (Par); 2o) a low stretch from Almeirim to Barreirinha (Amazonas); 3o) a central portion, from Barreirinha to Manaus (Amazonas); 4o) a mid section from Manaus to Tef (Amazonas); and 5o) a upper stretch from Tef to Tabatinga (Amazonas), close to the Colombian and Peruvian borders. The aim of this study was to make a survey of the fish diversity along the Amazon River floodplain, investigating the existence of different fish assemblages associated to that zones as a tool to help determine priority areas for conservation. The samples were obtained in the dry season (September and October) of 2003, along approximately 3,000km of the Brazilian Amazon River floodplain. The fish were collected using seine nets (12x3m, 5mm mesh size), in two habitats (beaches and floating meadows) with a standardized collecting effort. The fishes collected were preserved in 10% formalin and later identified to the most precise level and counted. Fish diversity was measured using the Shannon-Wiener and Simpson indexes. The similarity among samples was checked by means of the Morisitas Index (Krebs, 1999). A two-way ANOVA was performed to test for differences in the fish diversity, species richness and abundance among beach and floating meadow samples and the proposed zones of the vrzea floodplains. Thirty-one sand beaches and 26 floating meadows were sampled. Beach samples harbored 118 species and 5,107 specimens, and in the floating meadows 184 species and 13,367 specimens were collected, totaling 280 fish species. The ANOVA didnt disclose significant differences in fish diversity among the five proposed zones (p=0.30), however the fish fauna associated to floating meadows were significantly different from that of the beaches (p=0.01). This difference was detected in fish abundance (p=0.03; F=4.68), species richness (p <0.01;

76

F=13.99) and diversity as measured by the Shannon-Wiener Index (p=0.04; F=4.05), but not for Simpson's (p=0.18; F=1.88). Beaches and floating meadows exhibited different species assemblages. The nine dominant species in beaches composed 82% of the fish fauna: the characins Aphyocharax alburnus, Moenkhausia lepidura, Roeboides affinis, Prionobrama filigera, Thoracocharax stellatus and an unidentified species; the catfishes Pimelodus blochii and Trachydoras steindachneri; and the puffer Colomesus asellus. The 10 most abundant species in floating meadows (61% of the fishes) were characins (Ctenobrycon hauxwellianus, Moenkhausia sp. 6, Cyphocharax spiluropsis and Hoplias malabaricus; the cichlids Apistogramma aff. eunotus, Mesonauta sp., Laetacara curviceps and Cichlasoma amazonarum; and the knifefishes Eigenmannia aff. virescens and Brachyhypopomus pinnicaudatus. Sand beaches are considered poorly structured habitats with few shelter areas and limited food resources. In contrast, the floating meadows are known as nursery grounds for young fishes that use the submerged roots as refuge from predation and foraging substrate, characteristics that probably were responsible for the higher fish abundance and species richness found in that habitat. The apparent absence of differences among the vrzea zones in our results may be an artifact of the indexes utilized, which do not compare directly the species composition among areas. Initial analyzes have shown that there is a highly variable degree of similarity in species composition among the samples taken at each zone, with the midsection (Manaus Tef, zone 4) apparently being the more homogeneous. Further analyzes will be done to test for differences in species composition among the proposed vrzea zones. Acknowledgments The authors wish to thanks to IBAMA-ProVrzea Project and INPA for the financial and logistical support; to A. K. M. Albernaz and L. Costa for his enormous efforts to make possible this study. References Forsberg, B. 2000. Matriz para o Projeto Manejo Sustentvel dos Recursos Naturais da Vrzea. Relatrio Final. IBAMA. 165p.

77

Goulding, M.; R. B. Barthem and E. Ferreira. 2003. The Smithsonian Atlas of the Amazon. Smithsonian Books. Washington. 253 p. Krebs, C.J. 1999. Ecological Methodoly. 2nd ed. Benjamin/Cummings. Menlo Park, CA, USA. 620p. Sippel, S.J.; S. K. Hamilton and J. M. Melack.1992. Inundation area and morphometry of lakes on the Amazon River floodplain, Brazil. Arch. Hydrobiol, 123(4): 385-400.

78

COMPOSITION AND DIVERSITY OF FISH FROM A MANAGED LAKE IN CENTRAL AMAZON Kedma Cristine Yamamoto Instituto Nacional de Pesquisas da Amaznia INPA Avenida Andr Arajo, 02936, cep 69068-011, Manaus-AM Tel+33 (0) 643-3269/ Fax + 33 (0) 236-7817 yamamoto@inpa.gov.br Maria Gercilia Mota Soares Carlos Edwar de Carvalho Freitas Henrique dos Santos Pereira EXTENDED ABSTRACT ONLY DO NOT CITE Introduction The fish fauna abundance and composition are related to the water level in the Amazonian floodplain lake. Fish are the main animal protein source of Amazonian riverine populations. However, with the technological innovations of fishing facilities and the rising demand for fishes made the conflicts between riverine and commercial fishermen tend to proliferate. The lake management through the fishing agreements became an alternative to minimize or avoid fishing pressure under local fish stocks. This study evaluated the structure of fish communities in lake managed by Amazonian riverine populations. Material and Methods This study were performed in Comand Lake (0310'84"S e 5815'02" W) loctaed in the Island do Risco, municipiality of Itacoatiara, Amazonas. This lake is managed by riverine communities for subsistance fishery mainly. Fishes were sampled in 2002, during high and low water period using one group of gill net of 20-90 mm mesh size. Total weight (g) and Standard length (cm) were taken for each fish collected. Fish diversity were determined by Shannon - Wienner index and by reverse Berger-Parker dominance index (Berger & Parker, 1970).

79

Results In Comand Lake, richness, diversity and evenness varied between periods. During high water level recorded richness was 28 species increasing to 43 in low water level period. Shannon-Wienner index did not vary from high to low water level period (3.8 and 3.5 respectivelly), however estimatives of BergerParker index showed a decrease from 3.94 (high water period) to 2.98 (low water period). This pattern was observed in evenness also, which has shown 0.79 in high water period and 0.66 during low water period (Table 1). Table 1. Adjusted data of diversity indexes of fishes collected in Comand Lake. Comand Lake High water Low water 28 43 122 901 16885 52475 3.81 3.58 3.94 2.98 0.79 0.66

Period Number of species (S) Number of individuals (N) Biomass (g) Shannon-Wienner index (H) Berguer-Parker index (1/D) Evenness (E)

Despite the differences in species richness between periods, the indexes have relatively colsed values, which can be related to dominance by Hemiodus rabo vermelho, Potamorhina altamazonica, Triportheus elongatus, Hemiodopsis sp. The indexes decreased during low water period due to high number of Triportheus elongatus specimens, and high difference in species number. Ambos os ndices reduziram na seca por causa do alto nmero de exemplares de e devido a elevada diferena no nmero de espcies.

80

Relative f requency (% )

0 H.rabo vermelho P. altamazonica T. albus P.flavipinnis M. aureum P. nattereri H.immaculatus S. fasciatus R.microlepis A. ucaylalensis A. ocellatus P. nigricans C. macropomum M. lepidura C. macropterus T. elongatus Hemiodopsis sp P nattereri S. elongatus P. altamazonica T. angulatus A. falcirostris Rineloricaria sp C. vittata S. rhombeus A. falcatus P. rutiloides Anodus H. edentatus P. pristigaster C. meyeri Hemiodus sp. S. rhombeus P. Chalceus P. amazonica P. galeatus R. microlepis P.blochii P. latior

20

40

60

80 High Low

100

Figura 1. Species relative abundance during the periods of high and low water periods in Comand Lake. Species are organized in increasuing order of abundance.

81

Total abundance in number of individuals (N) and weight (g) during high water peiod were three times greater than the values recorder for low water period. During high water period 122 specimens were collected accounting for a biomass of 16885g while in the low warter period 901 individuals represented 52475g. Characiformes were the most abundant group in number of individuals and in biomass. They accounted for more than 50% of biomass captured in both periods. During high water period Characiformes were followed by Siluriformes and Clupeiformes orders. In the low water period Osteoglossiformes has shown the highest biomass, followed by Siluriformes and Perciformes. The species with highest biomass during the high water period was Potamorhina altamazonica, Pellona flavipinnis, Hemiodus rabo vermelho , Prochilodus nigricans and Colossoma macropomum; and during the low water period: Pygocentrus nattereri, Psectrogaster pristigaster, Hemiodopsis sp., Arapaima gigas and Triportheus elongatus. Discussion Specific composition and biomass of the ichthyofauna in Comand Lake varied seasonally between high and low water periods. The Characidae, Hemiodidae and Curimatidae families were dominant along hydrological cicle. Seasonal changes in fish composition were also observed in Incio (Saint-Paul et al., 2000) Rei (Merona & Bittencourt, 1993) lakes floodplains. Comparing richness and diversity indexes available in others studies on Central Amazonian floodplains, lakes allow us to state that Comand Lake still shows good environmental conditions for fish communities, although this lake was subject of intense commercial fisheries. Besides, this can indicate that fisheries do not affect irreversibly fish communities since habitat conservation is implemented allowing the fish populations to recolonize the lakes. In these context community fishery agreements are of key importance when the aim of management is to maintain fish community structure. Acknowledgements The riverine communities of Risco Island for their collaboration during study, and also to BAMCOPE/PNOPG project.

82

References Berger, W.H.; Parker, F. L. 1970. Diversity of planktonic foraminifera in deep sea sediments. Science, n. 168: 1345-1347. Mrona, B.; Bittencourt, M.M. 1993. Ls peuplements de poissons du lago do Rei, um lac dinondation dAmazonie Central: description gnrale. Amazoniana, 12: 415-441. Saint-Paul, U.; Zuanon, J.; Villacorta-Correa, M.A .; Garcia, M. Fabr, N.N.; Berger, U. & Junk, W. J. 2000. Fish communities in Central Amaznia white- and the backwaters floodplains. Environmental Biology of Fishes 57:235-250.

83

84

DAM IMPACTS ON TROPHIC STRUCTURE OF FISH COMMUNITIES (ARAGUARI RIVER, UPPER PARAN, BRAZIL) Volney Vono Fish Passage Center. Federal University of Minas Gerais. Av. Antnio Carlos, 6627. 31270-901 Belo Horizonte, MG. Brazil. Phone: +55 31 3499-2831 / 31 9633-2186 e-mail: lagovono@dedalus.lcc.ufmg.br EXTENDED ABSTRACT ONLY DO NOT CITE Introduction Nova Ponte (19o10S; 47o30W) and Miranda (19o2 S; 55o30W) dams were built in 1993 and 1997, respectively, in Araguari River, Upper Paran Basin, Brazil. This basin is responsible for providing 70% of Brazil energy demand. Both dams were constructed primarily for the generation of hydropower and produced two contiguous reservoirs. The total inundated area of these two reservoirs is 496 km2. This study evaluated the effects of these dams on the trophic structure of the fish communities. Data were collected before and after dam construction: from 1986 to 2001. My primary hypotheses were environmental perturbations caused by the dams construction shifts fish abundance and the feeding habits of the fishes and influencing the trophic guilds. Methodology From 1986 to 1988, seventeen samples were collected at four Araguari River sites before construction of the dams. From 1993 to 2001, forty-five sampling trips were collected after dam construction was completed. On each sampling trip: three sample sites were visited in each reservoir (2 reservoirs X 3 sites = 6 sites in reservoirs), one site immediately below the most downstream dam (Tailrace Site) and two sites further downstream in the remaining free-flowing segment of the river. Stomach contents of 2,570 individuals representing 34 species were analyzed. For each food item type, I calculated the occurrence frequency and its relative weight. Frequency and relative weight were combined into alimentary indices (IAi) after Kawakami & Vazzoler (1980). I used the IAi values to determine the Euclidian distance for similarity coefficients. To distinguish the trophic guilds for each studied period and site, I applied Cluster

85

Analysis with the complete linkage method. The guilds abundances were estimated using the capture per unit effort (CPUE) frequency data. Results and Discussion Considering both study periods, most species consumed a great variety of alimentary items. This is a common pattern in tropical regions where there is a high diversity of food for freshwater fish (Hahn et al., 1997). The mandiamarelo Pimelodus maculatus consumed the greatest diversity of alimentary items, including those of autochthonous and allochthonous origin (Gerking, 1994). Fruits were the most consumed item in the free flowing segments of the Araguari River before and after damming. But in both reservoirs, fruit consumption was remarkably reduced. In the reservoirs and at the Tailrace site, fishes were the most consumed item. Vegetation, detritus, and aquatic insect larvae were consumed in both reservoir, but mainly at Nova Ponte reservoir. Terrestrial insects were the most consumed in the Araguari river before damming. Zooplanctonic organisms and filamentous algae were consumed very little in all periods and at all sample sites. Eight trophic guilds were recognized among the fish communities. Twenty-two species, representing 65% of the analyzed species, opportunistically shifted their feeding habits due to the environmental changes that occurred after damming. Some of these shifts were considerable. Some examples, the Characiform Astyanax fasciatus (terrestrial insects in the free-flowing segments of the river before and after damming; but omnivorous in the reservoirs and zooplanctivorous in the Tailrace area) and Leporinus friderici (frugivorous in the river before and after damming; piscivorous in Nova Ponte Reservoir and in the Tailrace Area and herbivorous in Miranda Reservoir) and the Siluriformes Pimelodus maculatus (omnivorous in the river before damming, herbivorous in Nova Ponte Reservoir, benthivorous in Miranda Reservoir, piscivorous in the Tailrace Area and, after damming, frugivorous/herbivorous in the free-flowing segment river stretch). Before dam construction some species that were piscivorous and detritivorous didnt shift their alimentary habits after damming. Detritivorous species dominated, in number (57%) and biomass (54%), the fish community of the river before damming, while herbivores and piscivorous species became the most abundant in the reservoirs after dam construction (Figure 1).

86

% NUMBER

% BIOMASS

86/88 93/94 95/96 96/97 1998 99/00 97/98 99/00 2001 97/98 99/00 2001 97/01 Before dam. Nova Ponte Reservoir Miranda Reservoir Insectivorous Benthivorous Zooplanctivororous Omnivorous Tailrace site River after

Piscivorous Herbivorous Frugivorous Detritivorous

Figure 1. Proportion of the trophic guilds abundance at different period and sample sites for the fishes collected in Araguari River Basin, before and after damming, from 1986 to 2001. Conclusion Due to the environmental changes promoted by the Araguari river damming, most of the species presented great capacity of shifting their alimentary attributes. Even so, only a small portion of the fish community remained successfully in the new environments (Vono, in press.), indicating flaws in the recruitment, possibly due to suppression of important habitats for feeding and spawning and refuge for the juveniles as presented before damming. Thus, in this study, damming had a remarkably influence in the trophic structure and maintenance of the entire fish community of Araguari River.

87

References Gerking, S.D. 1994. Feeding ecology of fish. Academic Press. London. 416 p. Hahn, N.S., R. Fugi, V.L.L. Almeida, M.R. Russo & V.E. Loureiro. 1997. Dieta e atividade alimentar de peixes do reservatrio de Segredo. In: Agostinho, A.A. & Gomes, L.C. (eds.). Reservatrio de Segredo. Bases ecolgicas para o manejo. EDUEM. Maring. p. 141-162. Kawakami, E. & G. Vazzoler. 1980. Mtodo grfico e estimativa de ndice alimentar aplicado no estudo de alimentao de peixes. Bolm. Inst. Oceanogr., S. Paulo, 29 (2): 205-207. Acknowledgements Financial support provided by CEMIG.

88

INVENTORY OF FISHES IN MARAJ ARCHIPELAGO, WESTERN AMAZONIAN (PAR STATE, BRAZIL) Luciano Fogaa de Assis Montag1 , Adna Almeida de Albuquerque1, Valdilene do Socorro de Egito Sena1, Fabio Ribeiro Silva2, Ronaldo Borges Barthem1 Division of Ichthyology, Depto. of Zoology, Museu Paraense Emlio Goeldi (MPEG) 2 Instituto Nacional de Pesquisa na Amaznia (INPA) Introduction The Marajoara Archipelago is a fluvio-marine complex composed for sets of ten of islands in the estuary of the River Amazon (Par - Brazil), where the numerous islands and canals form the known region as Breves punctures of and the bay of the Maraj. Such archipelago had its savannah areas indicated as high priority in the Workshop on the Biomas Cerrado and Pantanal. This is an area disjoin of bioma Closed in the Amaznia. In the present work, it was aimed at to congregate scientific knowledge on ictiofauna of the ecorregio of the Maraj, in order to serve of base for the establishment of strategies of territorial management, ecological and partnerambient zoning, handling of resources and conservation of the nature, being such objectives base for the project "ecological Evaluation and election of with priority areas to the Amazonian savannah conservation, archipelago of the Maraj, Belm of Par". Objective The general objectives of the proposal of "Ecological evaluation and election of with priority areas to the Amazonian savannah conservation, archipelago of the Maraj, Belm of Par" are: (a) Ecological and biogeogrfica characterization of the savannah spots of the Archipelago of the Maraj; (b) Election and indication of with priority areas to the conservation of the biodiversity of savannahs of the Archipelago of the Maraj; e (c) Mapping, in extended scale (1:25.000), of the priority areas selected, with legends for importance degree, based in integrated analyses. The specific objectives will be concentrated in: to inventory the biological groups inside of the tipologias and the units of selected landscapes; to select, to mapear and to indicate the areas of high value for conservation, based
1

89

on the integrated analyses multi-thematic of the diverse biological groups and the degrees of threat. Materials and Methods First the present work congregated given of national ictiolgicas collections as well as international (MPEG, FURG, UF, ZMH, CU, NMR, NMRJ, FMNH, and MCZ), verifying the localities of collection and the validity of the specific epithets. The attainment of these data was mainly effected through the vestibule fishbase (www.fishbase.org). The collections will concentrate in the cities of Chaves, Salvaterra, Ponta de Pedras, Muan, Cachoeira do Arari and will be carried through fish experimental using mainly of nets of wait and nets of hand. The captured specimens will be fixed, conserved and deposited in the ictiolgica collection of the Paraense Museum Emilio Goeldi (Par). Results Until the present moment, the survey between the collections entered 1240 tumbles with 188 species of fish, where 175 species are exclusive registers of the ictiolgica collection of Museu Paraense Emlio Goeldi (MPEG). These species are contained in 18 orders and 59 famlias.all the units had been collected in the cities of Vigia, Cachoeira do Arari, Santa Cruz doArar, Porto Salvo, Salvaterra, Monsars, Jubim, Joanes, Condeixa, Soure, Pnta de Pedras, Santa Brbara do Par and Colares. Being that all these cities are situated in the northeast region littoral of the island, next to the city to Belm (Figure 1).

90

F Figure 1: Archipelago of the Maraj, indicating the localities of collections of ictiofauna of data deposited in cientfcias collections. As complement of the gotten data, had been carried through collections recently in ten points in December of 2003 in the city of Ponta de Pedras (01; 08"S) and (48 58' 20, 8"O) collected in the total 992 units with 62 species pertaining to the 25 families, being that of these, 6 families and 45 species are new registers to the region of the Maraj, what represents an increment of 24% of species (Table1), Of these collected a total of 17 species new already is registered in some of the collections raised. Barthem, R. B. 2001. Componente biota aqutica. Pginas 60-78 in: J.P.R. Capobianco et al.(org.). Biodiversidade na Amaznia brasileira. So Paulo. Estao Liberdade/Instituto Socioambiental. 540pp. Brasileira.. So Paulo. Estao Liberdade/Instituto Socioambiental. 540pp. Baensch, H.A. & Fischer. G. H. 1998. Aquarien Atlas Photo Index. Mergus Verlag GmbH. 1211p.

91

Bucher, E. H., Bonetto, A., Boyle, T., Canevari, P., Castro, G., Huszar, P. & Stone, T. Hidrovia: An initial environmental examination of the Paraguay Parana waterway. Wetlands for the Americas, Manomet, Mass., EUA e Buenos Aires, Argentina. 72 pp. Burgess, W. E. An atlas of freshwater and marine catfishes. A preliminary survey of the Siluriformes. TFH Publications. Neptune City. 1989. 784 p.

Table 1- List taxonomic of species inventoried for the bay of the Maraj Carcharhiniformes Carcharhinidae Carcharhinus limbatus (Mller & Henle, 1839) Carcharhinus porosus (Ranzani, 1840) Carcharhinus leucas (Mller & Henle, 1839) Isogonphodon oxyrhynchus (Mller & Henle, 1839) Sphyrnidae Sphyrna tudes (Valenciennes, 1822) PRISTIFORMES Pristidae Pristis perotteti Mller & Henle, 1841 RAJIFORMES Dasyatidae Dasyatis guttata (Bloch & Schneider, 1801) Potamotrygonidae Paratrygon aiereba (Mller & Henle, 1841) Potamotrygon histrix (Mller & Henle, 1834) Potamotrygon motoro (Mller & Henle, 1841) Potamotrygon orbignyi (Castelnau, 1855) Potamotrygon scobina Garman, 1913 Clupeiformes Clupeidae Odontognathus mucronatus Lacepde, 1800 Sardinella janeiro (Eigenmann, 1894)

92

Engraulididae Anchoa spinifer (Valenciennes, 1848) Anchovia clupeoides (Swainson, 1839) Anchovia surinamensis (Bleeker, 1866) Anchoviella brevirostris (Gnther, 1868) Anchoviella cayennensis (Puyo, 1946) Anchoviella elongata (Meek & Hildebrand, 1923) Anchoviella guianensis (Eigenmann, 1912) Anchoviella jamesi (Jordan & Seale, 1926) Cetengraulis edentulus (Cuvier, 1829) Lycengraulis batesii (Gnther, 1868) Lycengraulis grossidens (Agassiz, 1829) Pterengraulis atherinoides (Linnaeus, 1766) Pristigasteridae Pellona flavipinnis (Valenciennes, 1836) Pellona sp. LEPIDOSIRENIFORMES Lepidosirenidae Lepidosiren paradoxa Fitzinger, 1837 LOPHIIFORMES Ogcocephalidae Ogcocephalus radiatus (Mitchill, 1818) OSTEOGLOSSIFORMES Osteoglossidae Arapaima gigas (Schinz, 1822) Osteoglossum bicirrhosum (Cuvier, 1829) CHARACIFORMES Acestrorhynchidae Acestrorhynchus altus Menezes, 1969 Acestrorhynchus falcatus (Bloch, 1794) *Acestrorhynchus falcirostris (Cuvier, 1819) Acestrorhynchus isalineae Menezes & Gry, 1983 Anostomidae Leporinus friderici (Bloch, 1794) *Leporinus affins Gnther, 1864 Leporinus fasciatus (Bloch, 1794) Leporinus melanostictus Norman, 1926 Schizodon fasciatus Spix & Agassiz, 1829 Schizodon vittatus (Valenciennes, 1850)

93

Characidae Astyanax bimaculatus (Linnaeus, 1758) Astyanax saltor Travassos, 1960 *Bryconops alburnoides Kner, 1858 *Bryconops caudomaculatus (Gnther, 1864) Bryconops sp. Charax gibbosus (Linnaeus, 1758) Hemigrammus guyanensis Gry, 1959 Hemigrammus levis Durbin, 1908 Hemigrammus ocellifer (Steindachner, 1882) Hemigrammus rhodostomus Ahl, 1924 Hemigrammus unilineatus (Gill, 1858) *Hyphessobrycon copelandi Durbin, 1908 *Hyphessobrycon herbertaxelrodi Gry, 1961 *Hyphessobrycon sp. Hyphessobrycon stegemani Gry, 1961 *Metynnis mola Eigenmann & Kennedy, 1903 *Moenkhausia oligolepis (Gnther, 1864) Pristobrycon aureus (Spix & Agassiz, 1829) Pristobrycon calmoni (Steindachner, 1908) Pygocentrus nattereri Kner, 1858 Pygocentrus scapularis Roeboides dayi (Steindachner, 1878) Roeboides myersii Gill, 1870 Roeboides sp. Serrasalmus elongatus Kner, 1858 Serrasalmus marginatus Valenciennes, 1836 Serrasalmus rhombeus (Linnaeus, 1766) Serrasalmus spilopleura Kner, 1858 Triportheus albus Cope, 1872 Triportheus angulatus (Spix & Agassiz, 1829) Triportheus paranensis (Gnther, 1874) *Crenuchus spirulus Gnther, 1863 Crenuchidae *Microcharacidium sp. Ctenoluciidae *Boulengerella lucius (Cuvier, 1816) Curimatidae *Curimata inornata Vari, 1989 Curimata cyprinoides (Linnaeus, 1766)

94

*Curimatopsis myersi Vari, 1982 Psectrogaster rutiloides (Kner, 1858) Erythrinidae Erythrinus erythrinus (Bloch & Schneider, 1801 Hoplerythrinus unitaeniatus (Agassiz, 1829) Hoplias malabaricus (Bloch, 1794) Gasteroplecidae Gasteropelecus sternicla (Linnaeus, 1758) Hemiodontidae *Hemiodus unimaculatus (Bloch, 1794) Lebiasinidae Copella arnoldi (Regan, 1912) Copella nattereri (Steindachner, 1876) *Nannostomus eques Steindachner, 1876 *Pyrrhulina filamentosa Valenciennes, 1846 Pristigasteridae *Pellona sp. CYPRINODONTIFORMES Anablepidae Anableps anableps (Linnaeus, 1758) Anableps microlepis Mller & Troschel, 1844 Poeciliidae *Pamphorichthys sp. Poecilia parae Eigenmann, 1894 Rivulidae *Rivulus punctatus Boulenger, 1895 ELOPIFORMES Elopidae Elops saurus Linnaeus, 1766 Megalopidae Megalops atlanticus Valenciennes, 1847 GYMNOTIFORMES Apteronotidae Apteronotus albifrons (Linnaeus, 1766) Apteronotus apurensis Fernndez-Ypez, 1968 Apteronotus sp Sternarchella terminalis (Eigenmann & Allen, 1942) Sternarchogiton nattereri (Steindachner, 1868) Sternarchogiton porcinum Eigenmann & Allen, 1942 Sternarchogiton nattereri (Steindachner, 1868)

95

Sternarchorhamphus muelleri (Steindachner, 1881) Gymnotidae Electrophorus electricus (Linnaeus, 1766) Gymnotus carapo Linnaeus, 1758 Gymnotus cataniapo Mago-Leccia, 1994 Hypopomidae Brachyhpopomus brevirositris (Steindachner, 1868) *Brachyhpopomus occidentalis (Regan, 1914) Brachyhypopomus pinnicaudatus (Hopkins, 1991) Hypopygus lepturus Hoedeman, 1962 *Hypopygus sp. Steatogenys elegans (Steindachner, 1880) Rhamphichthyidae Gymnorhamphichthys sp. Rhamphichthys apurensis (Fernndez-Ypez, 1968) *Rhamphichthys marmoratus Castelnau, 1855 Rhamphichthys rostratus (Linnaeus, 1766) Sternopygidae Eigenmannia humboldtii (Steindachner, 1878) Eigenmannia virescens (Valenciennes, 1842) Rhabdolichops troscheli (Kaup, 1856) Sternopygus macrurus (Bloch & Schneider, 1801) Perciformes Carangidae Caranx hippos (Linnaeus, 1766) Oligoplites palometa (Cuvier, 1832) Oligoplites saurus (Bloch & Schneider, 1801) Centropomidae Centropomus parallelus Poey, 1860 Cichlidae *Acaronia nassa (Heckel, 1840) *Aequidens tetramerus (Heckel, 1840) *Aequidens pallidus (Heckel, 1840) *Apistogramma commbrae (Regan, 1906) *Apistogramma luelingi Kullander, 1976 Apistogramma sp. Astronotus ocellatus (Agassiz, 1831) Chaetobranchopsis orbicularis (Steindachner, 1875) Cichla ocellaris Bloch & Schneider, 1801 Cichla temensis Humboldt, 1821

96

*Crenicichla cincta Regan, 1905 *Crenicichla notophthalmus Regan, 1913 *Crenicichla reticulata (Heckel, 1840) Crenicichla saxatilis (Linnaeus, 1758) *Crenicichla sp. *Crenicichla strigata Gnther, 1862 Geophagus camopiensis Pellegrin, 1903 *Geophagus surinamensis (Bloch, 1791) *Satanoperca jurupari (Heckel, 1840) Eleotridae Guavina guavina (Valenciennes, 1837) Ephippidae Chaetodipterus faber (Broussonet, 1782) Gobiidae Awaous flavus (Valenciennes, 1837) Bathygobius soporator (Valenciennes, 1837) Boleophthalmus sp. Eleotris pisonis (Gmelin, 1789) *Eleotris sp. Evorthodus lyricus (Girard, 1858) Gobioides broussounetti Lacepde, 1800 Gobionellus oceanicus (Pallas, 1770) *Gobionellus sp. Haemulidae Diagramma sp. Genyatremus luteus (Bloch, 1790) Lobotidae Lobotes surinamensis (Bloch, 1790 Mugilidae Mugil curema Valenciennes, 1836 Mugil sp. Polycentridae *Polycentrus schomburgkii Mller & Troschel, 1848 Sciaenidae Cynoscion acoupa (Lacepde, 1801) Cynoscion microlepidotus (Cuvier, 1830) Macrodon ancylodon (Bloch & Schneider, 1801) Micropogonias furnieri (Desmarest, 1823) *Pachypops furcraeus Plagioscion auratus (Castelnau, 1855) Plagioscion squamosissimus (Heckel, 1840)

97

Plagioscion surinamensis (Bleeker, 1873) Plagioscion sp. Sciaena bathytatos Chao & Miller, 1975 Stellifer brasiliensis (Schultz, 1945) Stellifer microps (Steindachner, 1864) Stellifer rastrifer (Jordan, 1889) Scombridae Scomberomorus brasiliensis Collette, Russo & ZavalaCamin, 1978 Stromateidae Peprilus paru (Linnaeus, 1758) Trichiuridae Trichiurus lepturus Linnaeus, 1758 PLEURONECTIFORMES Achiridae Achirus achirus (Linnaeus, 1758) Achirus lineatus (Linnaeus, 1758) Apionichthys dumerili Kaup, 1858 Cynoglossidae Symphurus plagusia (Bloch & Schneider, 1801) Paralichthyidae Citharichthys macrops Dresel, 1885 Citharichthys spilopterus Gnther, 1862 *Paralichthys brasiliensis (Ranzani, 1842) Pristiformes Pristidae Pristis perotteti Mller & Henle, 1841 Dasyatidae Dasyatis guttata (Bloch & Schneider, 1801) Potamotrygonidae Paratrygon aiereba (Mller & Henle, 1841) Potamotrygon histrix (Mller & Henle, 1834) Potamotrygon motoro (Mller & Henle, 1841) Potamotrygon orbignyi (Castelnau, 1855) *Potamotrygon scobina Garman, 1913 SILURIFORMES Ariidae Arius rugispinis Valenciennes, 1840 Arius phrygiatus Valenciennes, 1840 Aspistor quadriscutis Valenciennes, 1840

98

Bagre bagre (Linnaeus, 1766) Cathorops spixii (Spix & Agassiz, 1829) Hexanematichthys couma (Valenciennes, 1840) Hexanematichthys parkeri (Traill, 1832) Hexanematichthys proops (Valenciennes, 1840) Notarius grandicassis (Valenciennes, 1840) Aspredinidae Aspredinichthys filamentosus (Valenciennes, 1840) Aspredo aspredo (Linnaeus, 1758) Platystacus cotylephorus Bloch, 1794 Auchenipteridae Ageneiosus inermis (Linnaeus, 1766) Ageneiosus ucayalensis Castelnau, 1855 Auchenipterus nuchalis (Spix & Agassiz, 1829) Pseudauchenipterus nodosus (Bloch, 1794) Tatia intermedia (Steindachner, 1877) Trachelyopterus galeatus (Linnaeus, 1766) Trachelyopterus galeatus (Linnaeus, 1766) Trachelyopterus striatulus (Steindachner, 1877) Trachelyopterus striatulus (Steindachner, 1877) Callichthyidae Hoplosternum littorale (Hancock, 1828) Megalechis thoracata (Valenciennes, 1840) Doradidae Anadoras weddellii (Castelnau, 1855) Doras eigenmani (Boulenger, 1895) Hassar sp. Lithodoras dorsalis (Valenciennes, 1840) Hepapteridae Rhamdia quelen (Quoy & Gaimard, 1824) Pimelodella cristata (Mller & Troschel, 1848) Loricariidae *Ancistrus sp. Hemiodontichthys acipenserinus (Kner, 1853) Hypostomus plecostomus (Linnaeus, 1758) Hypostomus watwata Hancock, 1828 Liposarcus pardalis (Castelnau, 1855) Loricarichtys sp. *Chaetostoma sp.

99

Pimelodidae Brachyplatystoma vaillantii (Valenciennes, 1840) Hypophthalmus marginatus Valenciennes, 1840 Leiarius marmoratus (Gill, 1870) Pimelodus altipinnis Steindachner, 1864 Pimelodus ornatus Kner, 1858 Zungaro zungaro (Humboldt, 1821) Pimelodus blochii Valenciennes, 1840 Pseudopimelodidae Batrachoglanis raninus (Valenciennes, 1840) SYNBRANCHIFORMES Synbranchidae Synbranchus marmoratus Bloch, 1795 TETRAODONTIFORMES Tetraodontidae *Colomesus asellus (Mller & Troschel, 1848) Colomesus psittacus (Bloch & Schneider, 1801) Anguilliformes Ophichthidae Myrophis punctatus Ltken, 1852 Batrachoidiformes Batrachoididae Batrachoides surinamensis (Bloch & Schneider, 1801) * New species

100

THE ICHTHYOFAUNA SPATIAL DISTRIBUTION IN THE FISH LAGOON NATIONAL PARK (31 S), BRAZIL Daniel Loebmann Bolsista DTI Empresa Brasileira de Pesquisa Agropecuria - Embrapa Meio Norte. BR-343, Km 35, Parnaiba, PI, Brazil. Cx. Postal 341, CEP 64200-970 e-mail: pinguimfiel@yahoo.com.br Joo P. Vieira Fundao Universidade Federal do Rio Grande, Departamento de Oceanografia, Av. Itlia Km 8, Rio Grande, RS, Brazil. Cx. Postal 474, CEP 96201-900 e-mail: vieira@mikrus.com.br EXTENDED ABSTRACT ONLY DO NOT CITE Introduction The Lagoa do Peixe National Park (LPNP) is located at Rio Grande do Sul coastal plain, between Patos Lagoon and Atlantic Ocean. The Fish-Lagoon (3121'26'' S; 05102'27''W) is a intermittent lagoon 31 km long and no more than 6 km across (35 km2) with a mean deep of about 0.3 m. The objective of this paper is to describe the spatial distribution of Fish-Lagoon. Material and methods Ten sites were sampled seasonally between 2001 and 2002. Four of those sites could be classified as shallow areas (Barra, Manduca, Guaritas e Talha Mar) and six were classified as deep zones (Chica, Capito - Rosa, Paiva, Lagamarzinho, Costa e Via Terra). Two sampling gear were used. 129 samples were taken using a experimental Beach Seine Net (1,2m high; 9 m long, 5 mm mesh at center (3 m) and 12 mm at extremities) and other 48 samples were collected using fishermans fyke-net used to catch shrimps. It is a passive net used at night with and attractive light.

101

These nets get working to about 12 hours. The mesh at the funnel is 12 mm. In this study we survey 1632 nets. We used Index of Relative Importance (IRI= FO% x (PN% + PW%), with takes into account the frequency of occurrence (FO%), the relative numerical abundance (NP%) and the relative weight (PW%). Based on IRI% values a cluster analysis was rum in order to group sites with similar fish structure. Results The use of the fishing gear into ten sampling sites resulted in the capture of 47,805 organisms, distributed in 56 species. At the shallow zone, Mugil platanus was the most abundant species (76.4%), followed by Jenynsia multidentata (9.3%) and Astyanax eingenmaniorum (4.4%). At the deep zone, Brevoortia pectinata was the most abundant species (42.5%), followed by Micropogonias furnieri (16.7%) and J. multidentata (14.6%). Regarding biomass, M. platanus represented 38.3%, A. eingenmaniorum 23.9% and J. multidentata 14.2% from total at the shallow zone. At the deep zone, B. pectinata represented 42.8%, Geophagus brasiliensis 13%, Hoplias malabaricus 11.8% and M. furnieri 11.5% from total. The cluster analysis on the RII showed that the fish assemblage can be divided in four sectors: Prelimnic Shallow Zone and Prelimnic Deep Zone, represented by the extremes North and South of FLagoon, and by the Estuarine Shallow Zone and Estuarine Deep Zone, represented by central sites of FLagoon, next to mouth.

102

Complete Linkage City-block (Manhattan) distances

Barra Guaritas Costa Lagamarzinho Paiva Manduca Talha mar Capito Rosa Chica Veia Terra
0 10 20 30

Estuarine Shallow Zone

Estuarine Deep Zone

Prelimnic Shallow Zone

Prelimnic Deep Zone

40

50

60

70

80

90

100

(Dlink/Dmax)*100

Figure 1 - Cluster based on Index of Relative Importance from the sampled sites of FLagoon. Discussion The FLagoon is composed by limnic, estuarine and marine species, since a saline gradient (salinities raging from 0 to 34) exists in that rea. Due to this stratification, a diferenciation on ichthyofauna composition can be found, revealing a tendency of the limnic species concentrate at the extremes and estuarine and marine species occupy the central area. Furthermore, the ichthyofauna captured in the extremities is mainly represented by adult organisms, whilst the central zone is mainly inhabited by juveniles. The estuarine shallow zones shelter many representants of the typically estuarine ichthyofauna. These areas are inhabited by small fishes (adults generally smaller than 10 cm) or marine species juveniles, constituted mainly by families Atherinidae, Mugilidae, Gerreidae, Cyprinodontidae and Anablepsidae (VIEIRA & MUSICK 1993). Such faunistic composition was confirmed at the shallow estuarine zone of FLagoon, since the RII% values showed that the most

103

important species for this assemblage were Atherinella brasiliensis and Odontesthes argentinensis, both Atherinidae, M. platanus (Mugilidae) and J. multidentata (Anablepsidae). Otherwise, the species found at FLagoon deep zone, represented mainly by B. pectinata, J. multidentata, M. furnieri and O. argentinensis, differ from patterns found by other estuaries, where J. multidentata and B. pectinata are not part of that assemblage, which is represented mainly by Scianidea and Ariidea (VIEIRA et al. 1998). It can be explained by the fact that, in the FLagoon, the deep zone is not so representative as in Patoos Lagoon, where the channel area are deeper. Thus, at the deep zone in FLagoon, there is a mixture of species that compose the assemblages of mid water (B. pectinata), shallow zone (J. multidentata) and deep zone (M. furnieri). Thus, the ichthyofauna composition along the sampled sites from estuarine to preliminc zones presented the following patterns: 1) a gradual change on dominance of the main species could be found; 2) there was a gradative replacement from marine to freshwater species; and 3) at the shallow zone, there was a drastic reduction of abundances of the species. This way, the results found in this study are in accordance with patterns found to other estuaries worldwide (PETERSON & ROSS 1991; WINEMILLER & LESLIE 1992; WAGNER & AUSTIN 1999). References Peterson, M.S. & S.T. Ross. 1991. Dynamics of littoral fishes and decapods along a coastal river-estuarine gradient. Estuarine, Coastal and Shelf Science, 33: 467-483. Vieira, J.P., J.P. Castello & L.E. Pereira. 1998. Ictiofauna. pp. 56-68. In: Seeliger U, Odebrecht, C. & Castello, J.P. (eds.) Os ecossistemas costeiro e marinho do extremo sul do Brasil, Editora Ecoscientia, Rio Grande, RS. 326p. Vieira, J.P. & J.A. Musick. 1993. Latitudinal patterns in diversity of fishes in warm-temperate and tropical estuarine waters of the western atlantic. Rev. Atlantica, Rio Grande. 115-133

104

Wagner, C.M. & H.M. Austin. 1999. Correspondence between environmental gradients and summer littoral fish assemblages in low salinity reaches of the Chesapeake Bay, USA. Marine Ecology Progress Series, 177: 197-212. Winemiller, K.O. & M.A. Leslie. 1992. Fish assemblages across a complex, tropical freshwater/marine ecotone. Environmental Biology of fishes, 34: 29-50.

105

106

THE FISHES OF THE FOREST STREAMS OF CENTRAL AMAZONIA: A PRELIMINARY SURVEY Fernando Pereira de Mendona, Instituto Nacional de Pesquisas da Amaznia (INPA/CPBA), CP 478, Av. Andr Arajo,2936, Petrpolis Manaus, AM, Brazil, CEP: 69083-970. Phone: 55 92 643-3253. mendonca@inpa.gov.br Jansen Zuanon, Instituto Nacional de Pesquisas da Amaznia (INPA/CPBA), zuanon@inpa.gov.br Mizael dos Santos Seixas, Universidade Federal do Amazonas (UFAM), bactres@yahoo.com.br Andr Vieira Galuch, Instituto Nacional de Pesquisas da Amaznia (INPA/CPBA), avgaluch@hotmail.com EXTENDED ABSTRACT ONLY - DO NOT CITE Amazonian streams hold a rich and diverse fish fauna that depends on the organic material as its main energy source. Despite some studies dealing with aspects of natural history of very few species or with ecological communities, there are no comprehensive results about the characteristics of the stream fish fauna as a whole. We made a survey of published papers, unpublished theses, dissertations and databanks in order to gather information about occurrence, species richness and general characteristics of central Amazonian stream fishes. Approximately 120 first- to fourth order streams were included, from data gathered from 11 sources and totaling 1352 registers of species occurrence. A total of 237 species were registered in the streams, belonging to seven Orders and 36 Families. Ostariophysan fishes composed almost 80% of the ichthyofauna and the most species-rich order was Characiformes (111 spp; 47%), Siluriformes (52 spp.; 22%) and Perciformes (42 spp.; 17%). The richest

107

family was Characidae (61 spp.) followed by Cichlidae (39 spp.). Some families (e.g., Ctenoluciidae, Serrasalmidae, Hemiodontidae) were registered only at larger order streams while others (e.g., Lebiasinidae, Characidae) predominated at first- and second order streamlets. The mean species richness by stream was 12 7.4 sd, with a modal number of 9 species (minimum = 5, maximum = 44) (Figure 1).

16 14 12 10 8 6 4 2 0 1 5 9 13 17 21 25 29 33 37 Number of species 41

Figure 1. Frequency of species richness registered in 120 Central Amazonia forest streams. The cichlid Aequidens pallidus was the species with higher number of occurrences (65), followed by the erythrinid Erythrinus erythrinus (59), and the lebiasinid Pyrrhulina brevis (59). The study also evidenced some unexpected omissions in the surveyed species lists, notably of some small species living among the submersed leaf litter banks. This denotes a biased underestimate of some components of the stream ichthyofauna, notably of the nocturnal, secretive gymnotiform species. Although far from complete, the survey revealed a dominance of small characid species on the ichthyofauna of these forest streams. This contrast with published results for Atlantic Forest streams in Southeastern Brazil, which faunal composition seems to be dominated by Siluriform species. Although biogeographic patterns may be responsible for this apparent discrepancy, it is most probably related to differences in the substrate composition. The

Frequency (N)

108

predominantly rocky bottom of Atlantic forest streams possibly provides a more complex substrate that is utilized as shelter by catfishes, which could explain the differences in faunal composition. A common problem to comparative studies of species richness in streams and rivers is the methodological differences found on most of published accounts. This impairs direct comparison among the datasets and can lead to erroneous interpretation of diversity patterns. The common underestimate of the fish species richness in most of studies (due mainly to inadequate sampling) causes an overestimate of Beta diversity, with obvious negative implications for the definition of conservation strategies involving these aquatic systems. Nevertheless, the strong modal distribution of species number observed in the analyzed studies indicates that these data are reliable and represents adequately the Alfa diversity in central Amazonian forest streams. The overall high species richness and the variable degree of similarity among river basins (although not analyzed in the present study) points to the necessity of large areas to guarantee the effectiveness of biological conservation of the regional stream fish fauna in Central Amazonia.

109

110

ICHTHYOFAUNA COMPOSITION AND DIVERSITY IN AQUATIC MACROPHYTE BANKS IN A CENTRAL AMAZONIAN FLOODPLAIN LAKE, AM, BRAZIL Esner Robert Santos Magalhes Instituto Nacional de Pesquisas da Amaznia - INPA, Departamento de Biologia Aqutica e Limnologia , Diviso de Biologia e Ecologia de Peixes - INPA Av: Andre Arajo 2936, Petrpolis Manaus, 69083000, AM Brasil Tel+33 (0) 643 3269 / Fax+33 (0) 236 7817 esnersantos@bol.com.br Maria G. M. Soares Carlos E. C. Freitas Kedma C. Yamamoto Introduction The Amazonian floodplain lakes are colonized by many species of aquatic plants that occupy extensive open areas in the rising water and high water level periods. The associations of these plants constitute the banks of aquatic macrophytes, inhabited by many species of fish, being considered important nurseries in floodplain lakes for migratory species (sold in the local markets) and for the non-migratory ones of economical value (Bayley, 1983; Goulding, 1980). Considering the scarcity of information about the ichthyofauna from the macrophytes banks in regions with high antropic impact on the environment, this study aimed to determine the structure of the fish community in floodplain lakes used for subsistence fishing. Material and methods The fishes were collected in the banks of macrophytes of the Comand lake in June (high water period) of 2002. The Comand is located in the Risco Island (031084 S) in Itacoatiara Amazonas, Brasil. This lake is managed by the riverine communities from the Island, where it is only subsistence fisheries is allowed. In this study a 25 x 5 m gill net (mesh: 10mm) was used. The fishes had been fixed in a 10% formalin solution and carried to the laboratory at INPA. Later, they were transferred to 70% alcohol for identification. The diversity of

111

ictiofauna in the aquatic macrophytes banks of Comand Lake was determined through the Index of Shannon-Wienner (Krebs, 1989) and inverse Berger-Parker dominance index (Berger & Parker, 1970). Results In this study 63 fish were captured, distributed in 4 orders, 8 families, 11 genus and 14 species. Characiformes is the predominant order with 63,49% of specimens, followed by Perciformes with 22.22%, Gymnotiformes with 11.11% and Clupeiformes with 3,17%. These orders contributed with the high biomass of the ichthyofauna from this lake. The highest amount of individuals collected were Characidae (39,76%) (Characiformes), Cichlidae (22,22%) (Perciformes) and Hipopomidae (11,11%) (Gynmotiformes). Therefore these are the most usual families of aquatic macrophytes banks. Shoal of fish had added more than half captured fishes in the Comand Lake. The Shannon-Wienner diversity index and Berger & Parker diversity index are similar (Table 1). The uniform number of individual among the species increases the evenness value to 0,62. Table 1.Parameters of the structure of the communities of fish in banks of aquatic macrophytes of Comand lake. N=63. Indexes Dominance d Berger-Parker 1/d Shannon-Wienner (individuals) Evenness (E) Biomass (g) Richmess (S) Number of individuals (N)

0.38 2.63 2.98 0.62 287.5 14 63

112

Table 2. Frequency of occurrence (F.O.%) orderly decreasing of the species captured in the Comand lake. N=63.
Common name Sardinha-comum Acar Sarap Trara Branquinha Pacu Apap Branquinha Branquinha Acar Piranha-preta Sardinha-papuda Acar-boar Acar Scientific name
T. albus C. amazonarum Brachyhypopomus sp. H. malabaricus P. latio M. duriventre P. castelnaeana Curimata vitata Curimata sp. Cichlasoma sp. S. rhombeus T. angulatus M. festivum S. jurupari

F. O. (%)
38,10 15,87 11,11 6,35 4,76 4,76 3,17 3,17 3,17 3,17 1,59 1,59 1,59 1,59

The Triportheus albus, Brachyhypopomus sp., Hoplias malabaricus, Mylossoma duriventre and Mesonauta festivum species (Table 2) usually found Comand lake machrophyte banks, were also identified in the lake Mamirau (Hendersosn, 1995), Camaleo (Souza-Pereira et al., 2000) and Janauac, of Rei and Camaleo (Sanchez-Botero, 2000). Discussion The dominant orders in the aquatic macrophytes banks of Comand Lake (richness, number of individuals and biomass) were Characiformes, Perciformes and Gymnotiformes according to what has already been reported in other studies. This study presents lower richness of species and lower number of individuals when compared with studies by Souza-Pereira et al. (2000), with 42 spp, and Sanchez (2000) with 91 spp. However, it is important to point out that in these works the fishing effort was greater than here indicating that Comand Lake is an exclusive subsistence fishery lake, having distinct characteristics in both richness and number of individuals as any other floodplain lake. The differences in the specific composition can be related not only to the management, but also to its localization in the island. In the rising water and high water level period there is a direct connection with the Amazon River.

113

This can be one of the factors that can be contributing for the high abundance of sardine (Triportheus albus), a species of high economical value which migrates in large shoals of fish. Acknowledgements This study was supported by the Instituto Nacional de Pesquisas da Amaznia (INPA) and BAMCOPE (CNPQ/PNOPG, n 550562701-0). References Bayley, P.B. 1983. Central Amazon fish populations: biomass, production and some dynamic characteristcs. - Phd Thesis, Dalhousie University. Halifax, Nova Scotia, Canada, 330 pp. Goulding, M. 1980. The fishes and The Forest: Explorations In Amazonian Natural History. University of California Press; 280 p. Henderson, P. A. & Hamilton , H. F. 1995. Standing crop and distribution of fish in drifting and attached floating meadow within and Upper Amazonian varzea lake. Journal Fish Biology. 47: 266-276. Snchez-Botero J. I. 2000. Distribuio espacial da ictiofauna associadas s razes de macrfitas aquticas em relao ao oxignio dissolvido, temperatura e tipo de planta na Amaznia Central. Dissertao de Mestrado, INPA-FUA, Manaus, 46 p. Souza-Pereira, A . C; Darwich, A., Soares, M.G.M. 2000 . Anlise nictemeral de variveis limnolgicas e a composio da ictiofauna em bancos de macrfitas aquticas. Anais da IXI Jornada de Iniciao Cientfica do PIBIC/INPA: 5-8.

114

FISH SPECIES COMMERCIALISED IN MANAUS: A REVISION ON THE LIFE CYCLE AND BIOLOGICAL PARAMETERS Gelson da Silva Batista Instituto Nacional de Pesquisas da Amaznia - INPA, Departamento de Biologia Aqutica e Limnologia , Diviso de Biologia e Ecologia de Peixes - INPA Av: Andre Arajo 2936, Petrpolis Manaus, 69083000, AM Brasil Tel+33 (0) 643-3269/ Fax + 33 (0) 236-7817 Maria Gerclia Mota Soares, Carlos Edwar de Carvalho Freitas, Kedma Cristine Yamamoto

EXTENDED ABSTRACT ONLY DO NOT CITE Introduction With a fishery potential estimated in about 420,000 t/year in 1991 (Bayley, 1998), fishing is the main protein source for the population. Fishing performed by professional and riverine dwelling fishermen supplies the local urban centers, and markets in other states and/or countries (Bayley & Petrere, 1989; Soares & Junk, 2000). The fishing activity exploits about 200 species (Barthem & Petrere, 1995), however, it is concentrated on only 11 groups, which are responsible for over 95% of all fish landed in Manaus. Given that fish is a limited resource and its market demand is ever increasing, the present paper proposes to provide a general view regarding the scientific findings on biology (feeding reproduction) and the population dynamics of five fish species marketed in Manaus-AM, based on literature data. These data are fundamental for preparing proposals and management policies concerning these species. Material and methods Bibliography survey and systematization (monographies, dissertations, theses, papers, (published in congresses and or specialized joirnals, reports, and books) available on Semaprochilodus insignis (jaraqui escama grossa), Prochilodus

115

nigricans (curimat), Cichla monoculus (tucunar), Schizodon fasciatum (aracu) and Triportheus elongatus (sardinha comprida) were conducted in the libraries at the Federal University of Amazonas (UFAM), National Researce Institute of Amazonas (INPA) and search in the Internet. Issues related with the life cycle and biological parameters were read and analyzed in order to seek data referent to the work theme. Results Jaraqui and curimat are iliofagous fishes (detritivorous), feeding on debris (amorphous material from organic material or litter) and micro-flora associated with the sediment. Tucunar is a piscivorous fish when adult, yet it eventually feeds on shrimp, when a minnow in plantofagous and insectivorous. Aracu is an herbivorous fish, feeding on roots, leaf-stem, fruits-seeds, algae and plant debris. Sardinha is na omnivorous fish, feeding on fruits-seeds, invertebrates from the flooded forest. Jaraqui, curimat, aracu belong to the fish group with seasonal reproductive strategy, they are fish that perform migration, with spawning time in the early rising water, external fecundation, with no care for the offspring and total spawning. Length of first sexual maturation (L50% ) in centimeters (cm) for jaraqui is between 24-26; for curimat between 26.2-35 and for aracu between 19-30. Tucunar belongs to the fish group with reproductive strategy in equilibrium, spawns various months of the year, mainly in the high-water season, in lentic environments. Tucunar L50% is between 19-28.5. Sardine reproduction time is from September to January, and the L50% is 12.2 cm. The better fitting growth model for tucunar is the Gullsnd-Holt one, because it provides higher L estimates, relative to Ford-Walford and Von Bertalanfy methods. Discussion Based on revised papers, there are data on feeding habit, diet and spawning time of the five species. There is little information regarding the L50% and the distinct locations in the Amazon region, and in addition to that there is no standard length measure for the studied species. Currently there is a government policy establishing the capture minimum size for jaraqui, curimat and tucunar. Given this policy and comparing it with the L50% of each species, it is not coherent for jaraqui, for the length of 20 cm (policy) is below the length of 24-26 cm

116

described in the papers revised for L50%. For curimat and tucunar the length established by the policy is coherent with that described in the papers. For sardinha e aracu no data was found on policy on account of the paucity of studies regarding population dynamics. One may conclude that knowledge on life cycles is still too incipient to serve as scientific basis for the establishment of legal mechanisms for preserving fish resources. Acknowledgements This investigation was funded by the Instituto Nacional de Pesquisas da Amaznia (INPA) and The project Influence of the Flood Pulse on the Ecological Dynamics in Floodplains granted by the Directed Research projects of the Pilot Program for protection of the Brazilian Rain Forest (PPG7). References Barthem, R. & Petrere, M. Jr. (1995): Fisheries and population dynamics of Brachyplatystoma vallantii (Pimelodidae) in the Amazon Estuary.- In: Armantrout, N.B & Wolotira R.J. Jr. (eds.): Condition of the Worlds Aquatic Habitats.- Orford & IBH Publishing, Bombay, pp. 329-240. Bayley, P.B. & Petrere, M. Jr. (1989): Amazon fisheries: Assessment methods, current status, and management options.- In: Dodge, D.P. (ed.): Proceedings of the International Large River Symposium.- Canadian Special Publication of Fisheries and Aquatic Sciences 106: 385-398. Bayley, P.B. (1998) Aquatic Biodiversity and Fisheries Management in the Amazon. Oregon State University. Soares, M.G.M.; Junk, W.J. (2000). Commercial fishery and fish culture of the state of Amazonas: Status and perspectives. In: The Central Amazon Floodplain: Actual Use and Options for a Sustainable Management. Backhuys Publishers, Leiden, The Netherlands, 433-461.

117

118

PIRAMUTABAS (BRACHYPLATYSTOMA VAILLANTII) AGE DISTRIBUTION ALONG THE AMAZON RIVER Pirker, Lilianne Esther Mergulho Museu Paraense Emlio Goeldi MPEG Trav. Mariz e Barros, 1384-Pedreira. CEP: 66.080-660 Belm-PA-Brasil Phone (91) 2265818 e-mail: pirker@museu-goeldi.br pirker@bol.com.br EXTENDED ABSTRACT ONLY DO NOT CITE Introduction Piramutaba (Brachyplatystoma vaillantii) is a freshwater catfish from the Pimelodidae family that occurs in the Amazon estuary and in the channels of Amazon river and its tributaries of white water and is exploited along the Amazon system. (Barthem & Goulding, 1997). The piramutaba is one of the Amazon species more exported. In the Par state, its exportation had already yielded US$ 13 millions for the state in 1980, which had occupied the ninth in the list of exportations products of the state (Banco do Brasil-CACEX, 1980). The piramutabas fishery is realized for two types of fisheries: traditional (along the Amazon river) and industrial fishery (restrict to Amazon estuary) (Barthem, 1990). Material and Methods The samples were obtained from the disembark of experimental fisheries in the Amazon river in the periods of dry and rainy seasons of 2002. The study area was divided into five regions: Belm, Santarm, Manaus, Tef and alto Solimes (Figure 1).

119

 Tef alto Solimes Manaus

Belm Santarm

Figure 1. Study area divided into five regions. From each region at least 250 furcal length of piramutaba were obtained. The total sample was 6.686 specimens of piramutaba that were measured. In this total sample was take a way a sub-sample with 963 specimens of piramutaba which the furcal length, total or eviscerate weight, sex maturity and vertebrae were collected. The piramutabas age was estimated through two different methods called direct and indirect: ring counting on vertebrae and length frequency analysis (LFA), respectively. In laboratory, the vertebrae were cleaned, sectioned, processed with a battery of chemistry processes and analyzed in a biological microscope. The vertebrae were observed through reflected light and black dish. Each ring observed was measured through a milimetric ruler with a unit of 0,1 mm.

120

The furcal lengths were aggregated through two centimeters of classes intervals. The modal decomposition was made by the Bertalanffys model by the FISATs program (FAO/ICLARM Stock Assessment Tools) (Gayanilo & Pauly. 1997). Results Along the Amazon river were observed seven age classes (Table 1). More than eight percent of the samples analyzed were three and four years old. Although, 57.58 percent of piramutabas were three years old. Too little that ten percent were captured with one, two, five, six and seven years old. The age class more frequent observed in all regions was the third. Piramutabas with seven years old were observed only in Belm, Santarm and alto Solimes. Belm was the only region that presented samples with all the age classes. And also the only region that presented piramutabas with one year old. This region presented the greatest frequency (62.94%) of three years old. Santarm presented the greatest percentage of piramutabas with six and seven years old and Tef with four and five years old. Belm can be representative of the smallest individuals and consequently the smallest ages. Table 1. Age frequency of piramutaba (Brachyplatystoma vaillantii) in Amazon system. Age (years) 1 2 3 4 5 6 7 Regions Belm 0.10 13.82 62.94 10.00 6.57 6.18 0.39 Santarm 0.00 4.27 51.36 20.18 12.16 9.96 2.07 Manaus 0.00 13.71 59.94 22.76 2.82 0.76 0.00 Tef 0.00 13.14 37.47 34.31 13.14 1.95 0.00 Alto Sol. 0.00 3.56 58.88 32.60 3.92 0.93 0.11

121

Discussion The most frequent age class disembark of piramutaba from the traditional fishery was the third, so this type of fishery was acting with more frequency in piramutabas with three year old before its age maturity (Barthem & Goulding, 1997). Belm is the region that needs to be managed separately of all the rest, because in this region was observed piramutaba with only one year old captured by the traditional fishery that is one type of selective fishery. However, the industrial fishery acts and exploits piramutaba with drags nets in Belm region. Consequently, more juveniles piramutabas are captured by this type of fishery without selectivity and any study about this type of fishery. Conclusions Along the Amazon river the traditional fishery exploits piramutabas with more frequency of three years old. In Belm the traditional fishery is exploiting piramutabas of the seven age classes, although the greatest frequencies are the specimens with three and four years old. Belm is a particular region of the Amazon region. The piramutabas fishery in this region needs to be managed differently of all the others regions in the Amazon system. References Banco do Brasil. Cacex. 1980. Principais produtos exportados pelo estado do Par. GEP Secretria de Estado da Fazenda. Coordenadoria de Informaes Econmico Fiscais. Barthem, R. B. 1990. Descrio da Pesca da Piramutaba (Brachyplatystoma vaillantii, Pimelodidae) no Esturio e na Calha do rio Amazonas. Bol. Mus. Par. Emlio Goeldi 6 (1): 119-131 p. Barthem, R. B. & Goulding, M. 1997. The catfish connection. Columbia Press. New York. 144 p.

122

Gayanilo Jr., & Pauly, D. 1994. The FAO ICLARM Stock Assessment Tools (FISAT) Users Guide. FAO Computerized Information Series, N 6 (Fisheries). 186 p. Acknowledgements Thank to Projeto Manejo dos Recursos Naturais da Vrzea - Pro-Vrzea for the opportunity to work in this research and the Museu Paraense Emlio Goeldi MPEG for the support.

123

124

ECOLOGICAL FEATURE OF A VOLGA PIKE PERCH (STIZOSTEDION VOLGENSE) POPULATION WITH REFERENCE TO ANTHROPOGENIC ASPECTS IN THE UPPER KUIBYSHEV WATER RESERVOIR OF RUSSIA Asiful Islam Department of Vertebrate Zoology, Kazan State University, Kazan, Russia 420008; Kazan Institute of Biochemistry and Biophysics, Russian Academy of Science, Kazan, Russia 420111 Phone: +7-8432-315258; Fax: +7-8432-387418; Email: asifuli@yahoo.com V. A. Kuznetsov Department of Vertebrate Zoology, Kazan State University, Kazan, Russia 420008 Phone: +7-8432-315146; Fax: +7-8432-387418; Email: Vjatscheslav.Kuznetsov@ksu.ru Abstract Volga pike perch introduced as a commercial fish in Kuibyshev water reservoir from 1973. This species play an important role in the ecological balance of this water reservoir. In 1980's decade, the production of S. volgense was 435.1 tonnes per annum (8.8% of the total production of fishes) but in 2001, the production was declined to 34 tonnes per year (1.2 %). Spawning efficiency of S. volgense is not influenced by the fluctuation of water level. Correlation coefficient was high in between the number of fry and fingerlings and the biomass of zooplankton in spring. Decreasing tendency of length and weight was observed in comparison to 1980's decade. Young fishes (2-4 years) were often found in our control fishing. In the last 5 years, the production and the growth of adult fishes were reduced abruptly in this water reservoir than the other water reservoirs of Europe. Commercial production of this species influence by different factors e.g. over exploitation, competition of food in the feeding ground, environmental factors, pollution and imbalance in the ecosystem etc.

125

Key words: Ecology, Volga pike perch, Stizostedion volgense, Kuibyshev water reservoir. Introduction Volga and zander pike perches (Stizostedion volgense Gmelin, 1788 and S. lucioperca Linnaeus, 1758) are the important commercial predatory fish species (Family: Percidae) of Kuibyshev water reservoir in Russia. Zander pike perch is usually found in most of the rivers and waters of Eastern and Western European countries and also in Russia (Wheeler, 1978). On the other hand, S. volgense is usually seen in delta Volga and its nearer tributaries e.g. Kuibyshev water reservoir, Kama, Ural, Don, Dnepr rivers and the deltas of Caspian sea (Berg, 1949; Dmitrieva, 1973; Tikhomirova, 1973) and also in Hungary and some other Eastern European rivers (Pinter, 1996). Though, Volga pike perch is adapted to cold environment, they prefer warm water for breeding. They perform two proportional spawning periods with a high rate of individual absolute fecundity. In Kuibyshev water reservoir, they start their natural spawning (7-30% of hatching) in May at a minimum temperature of 12-14 0 C and afterwards in June (Kuznetsov, 2001). In comparison to S. lucioperca, the growth rate of Volga pike perch is rather slower. Though they habituate with S. lucioperca at the same geographical and ecological condition of the water reservoir, have a diversified feeding habit. They are predator fish but, in the early stage (first year of their life cycle), they usually take phyto and zooplankton as their food. Volga pike perch was not treated as a commercial species from the beginning of the formation of Kuibyshev water reservoir (established in 1954) (Smidtov, 1956). But, its biological importance in this water reservoir is gradually increasing (Braslavskia, 1972). From 1973, the commercial production of this species began in this reservoir. Although many papers have been written about Volga pike perch (Berg, 1949; Lukin, 1952; Nikitina and Moskul, 1999), the ecological studies of this species is else insufficient. Several well-known pertaining works on morphology, feeding and fecundity of S. volgense of Kuibyshev water reservoir (Iashanin, 1978, 1982, 1986), biology of this species in Cibiaga bay (Smirnov, 1984, 1986), production and general morphology of S. volgense in Tcimiliansk, Vesalovsk and Proletarian water reservoirs of Russia (Tuniakov, 1971, 1974, 1976) and reproduction of Volga pike perch in Zaporozhsk (Danilenko, 1991) and ecological feature of them in Denaprovsk (Noviskii, 1999) water reservoirs in Ukraine.

126

The purpose of this study was to describe the ecological feature and the anthropogenic effects on S. volgense in Kuibyshev water reservoir. Moreover, there have been giving a concentration on the study of population biology of this species e.g., their commercial production, length-weight and age relationship, spawning efficiency and growth rate. Materials and Methods Materials of this study were collected from the lower Cibiaga bay of Kuibyshev water reservoir (Kazan, Russia- 55.8 N 49.2 E) during 1998-2002. The materials were also collected during 1981-1991 were used for the comparative experimental study. Fry and fingerlings were collected during 1964-2001. Adult fishes were caught by gill nets (diameter 24-65 mm). Larvae and fry were collected from the open waters of the reservoir by special dragnets (length 80 cm: Model 'Gas 15' and length 12 m: diameter 2.5 mm, Russia). Age of fish was determined by the transverse section of the abdominal fins and the observation of scale rings (Chugunova, 1959; Pravadin, 1966). Growth of fish was detected by direct proportional dependency method with the measurement of the radius of scale rings in relation to body length (Pravadin, 1966). Developmental stages of fish were studied by following the works of Basnesov (1953). Statistical methods were conducted by the method of Lakin (1990) and by using Excel 7.0 computer program. Results and Discussion Commercial production Volga pike perch was not treated as a commercial fish species in Kuibyshev water reservoir up to 1973. Though, the reservoir was created (for the establishment of hydroelectric station) in 1954, the production of S. volgense was not increased up to the commercial level for about 2 decades. In 1954, the total production of the commercial fishes was 66.8-74.6% among them, Volga pike perch was only 0.1% (Smidtov, 1956). In 1973, the total production of Volga pike perch was 135 tonnes (3.4% of the total production of fishes) and after five years in 1978, the production was 435.1 tonnes (8.8 %) i.e., the production of this species was increased 3 times by 5 years. On the other hand, in 1990s decade, the production of this species decreased to 2.1-5.1% from the total production of fishes. During 1999-2001, the production was sharply declined to 25.3-34.0 tonnes per year (0.9-1.2 %) (Fig. 1). Therefore, after 30

127

years, the commercial value of Volga pike perch is further reduced because of its rapid production declination.

500 400 300 200 100 0 1973

Production, /yr.

1978

1983

1988

1993

1998 Years

Figure 1. Commercial production of Volga pike perches during 1973-2002 in Kuibyshev water reservoir. Spawning efficiency of S. volgense was not influenced by the fluctuation of water levels in spring. The correlation between the number of fry and water temperature in May had a linear relationship with degrees of freedom at 0.05, which had a positive trend (r = 0.11; y = 0.22x0.96) with the abotic factors. A high correlation coefficient (0.48 0.20, P=0.0003) was revealed in between the number of fry and fingerlings and the biomass of zooplankton in May (Fig. 2).
No. of fingerlings

10 8 6 4 2 0 0 2

y = 0.6777x + 0.5184 R2 = 1

4 6 Zooplankton, g/m3

Figure 2. Relationship between the biomass of zooplankton and the number Volga pike perch fry and fingerlings in Kuibyshev water reservoir. Abiotic factors (e.g., temperature) had no influence at the efficiency of reproduction of this species. However, the breeding and number of fry and fingerlings of other fish species are closely related to the fluctuation of water level and temperature in Kuibyshev water reservoir (Kuznetsov, 1980). Length-weight and age relationship

128

In 1981-1991, the average length of S. volgense was 29 cm. In 1993, it was decreased to 27.79 + 0.60 cm. In 1998-2001, the average length was 25 cm and it reduced abruptly than the last 2 decades (Table. 1). Decreasing tendency of the average length of Volga pike perch was started in 1990s decade and eventually, it effected on the overall production of this species in the reservoir. Similarly, the average body weight of male and female fishes was 454.0 32.8 g in 1981 but, in 2002 it was found 166.3 9.8 g (Tab. 2). The correlation coefficient of length and weight of this species in between the 2 decades (1980s and 1990s) revealed a high variation with degrees of freedom, 12.3. During 1991 and 1993, 5-6 years old fishes were 12.8 % and 75.9% but, during 1998-2000, 4-5 years old Volga pike perch was 67.9-91.8%. In 2001 and 2002, 3 and 4 years old fish was 84.8% and 73.1%. In our controlled fishing, 3-5 years old S. volgense was captured - 68.8% during 1991-2002 (Fig. 3). Decreasing tendency of the production of adult fishes was revealed during the last ten years. Young fishes were captured more in this time and they were dominated in the reservoir with relatively lower length and weight of the body. Growth of Volga pike perch In 1991, the growth rate of S. volgense was high. During 1993-2002, the growth rate of this species was abruptly declined in relation to their body length (Fig. 4). We have investigated different growth rates of S. volgense in several water reservoirs of Europe (Tab. 3). Growth rates of S. volgense in different water reservoirs were different. In north white lake, S. Volgense had low growth rate. But in southern lake, it was found relatively high growth rate. Basically, the growth rate depends on biotic and abiotic factors of the water bodies. In Kuibyshev water reservoir, the growth rate of this species was comparatively so low. Growth rate of Volga pike perch of this reservoir was influenced by different factors e.g. competition of food in the feeding ground, environmental conditions, water quality, ecosystem etc.

129

Table 1. Average length of S. volgense in the upper part of Kuibyshev water reservoir during 1981-2002. Year Sex Min. length Max. length (cm) 1981 1991 1993 1998 1999 2000 2001 2002 Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average 20-40 20-34 20-40 22-30 14-32 14-32 16-34 16-32 16-34 14-28 14-28 14-28 20-34 20-32 20-34 20-31 20-29 20-31 20-34 16-29 16-34 20-29 20-27 20-29 30.53 + 1.88 29.15 + 0.83 29.83 + 1.35 25.80 + 0.60 25.88 + 0.63 25.86 + 0.62 28.91 +0.53 26.55 + 0.68 27.79 + 0.60 24.10 + 0.40 23.10 + 0.40 23.50 + 0.45 26.10 + 0.50 24.90 + 0.40 25.50 + 0.45 25.68 + 0.39 24.77 + 0.30 25.25 + 0.26 25.73 + 0.37 24.58 + 0.23 25.01 + 0.25 25.50 + 0.60 25.00 + 1.00 25.30 + 0.60 22.2 10.3 23.9 6.1 12.2 13.8 9.9 13.1 16.0 5.2 6.7 11.1 15.2 13.3 20.3 8.6 6.9 8.0 10.9 6.9 10.6 10.0 10.6 12.3 14 14 28 8 25 33 29 26 55 11 16 27 63 69 132 32 29 61 57 55 112 19 8 27 M+m CV, % n

Here, M + m Average arithmetic length and its error; CV, % - coefficient variant and n No. of fishes.

130

Table 2. Average weight of Volga pike perch of Kuibyshev water reservoir during 1981-2002. Year 1981 1991 1993 1998 1999 2000 2001 2002 Sex Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average Female Male Average Min. weight Max. weight (g) 200-1100 200-700 200-1100 120-360 20-360 20-360 160-400 150-390 160-400 70-220 50-300 50-300 100-340 130-340 100-340 100-400 100-300 100-400 100-650 50-300 50-650 100-320 85-190 85-320 M+m 500.0 + 39.2 421.4 + 18.8 454.4 + 32.8 214.4 + 17.9 232.2 + 15.1 230.9 + 15.8 290.2 + 14.0 221.3 + 15.4 257.0 + 11.0 184.5 + 4.1 169.4 + 4.4 176.9 + 4.1 230.9 + 8.6 205.8 +5.8 217.8 + 7.1 237.5 + 9.5 200.2 + 7.3 219.3 + 6.5 237.3 + 12.0 211.4 + 4.0 214.8 + 7.1 173.3 + 8.6 137.2 + 13.4 161.3 + 98 CV, % 28.2 16.1 3..3 2.1 3.5 3.,3 2.5 3.5 31.5 7.7 10.4 12.0 29.6 23.4 40.6 22.6 19.6 23.1 38.2 14.0 35.0 21.1 25.8 31.6 n 14 14 28 8 25 33 28 26 33 11 16 27 63 69 132 32 29 61 57 55 112 19 8 27

Here, M + m Average arithmetic weight and its error; CV, % - coefficient variant and n No. of fishes.

131

Table 3. Growth rate of Volga pike perches in different reservoirs (in relation to length) of Russia and former USSR. Reservoirs White lake Vachilevsk water reservoir Chimlianska ya water reservoir Diniprovskay a water reservoir Kuibyshev water reservoir Conclusion In the last decade of the 20th century, the destabilization of the ecosystem in Kuibyshev water reservoir influenced the production of all commercial fishes. Commercial production of Volga pike perch and other predatory fishes like: pikes, zander pike perches and bream were decreased sharply in this reservoir. It is an essential task to uplift the production of this species by analyzing biotic and abiotic factors of the waters. Fingerlings prefer to take zooplankton in their early stages. For successful spawning needs adult and mature fishes. These limiting factors influence the production of this newly introduced commercial species in this reservoir. Moreover, the declination of the production indicates the negative trends of S. volgense population in Kuibyshev water reservoir. As it is treated as a biological monitor like other predator fishes of this reservoir, so the ecological balance and proper exploitation of adult fishes could revive this fish population in this reservoir. 1 6.6 10. 4 16. 2 2 10.1 19.6 22.0 20.1 3 14. 9 23. 8 26. 2 23. 0 20. 1 Age, year 4 5 20.0 24.8 27.0 29.0 28.6 28.4 31.2 31.1 Authors 6 27. 7 29. 0 33. 7 32. 8 7 31. 3 29. 5 35. 4 32. 5 Tikhomirov a, 1973 Tuniakov, 1974 Tuniakov, 1971 Noviskii, 1999 Our results, 2000-2002

4.6

11.8

24.0

132

Acknowledgements We would like to thank the expedition group of Zoological station of Kazan State University for their technical support during the study period.

% 100 0

1991 yr. n=33 2 3 4 5 6 7

50 0 2
80 40 0 2

1993 yr. n=29 3 4 5 6 7

1999 yr. n= 3 4 5 6 7

50 2001 yr. n= 0 2
100 0

2000 yr. n=61 2 3 4 5 6 7

Figure 3. Ages of S. volgense of Kuibyshev water reservoir in spring during 1991-2002.

133

Length, cm 30

20 10 0 0 1 2 3 4 Age, years 5

1 2 3 4 5 6

Figure 4. Growth of Volga pike perch (in relation to body size) of Kuibyshev water reservoir. (1-1991 yr.; 2-1993 yr.; 3-1998 yr.; 4- 1999 yr.; 5- 2000 yr.; 6- 2001 yr.) References Braslavskia, L.M. 1972. Volga pike perch. Territory Tatarstan. Dept. of Fisheries. Pub. 12: 164-169. Basnesov, B.B. 1953. Stages of development of bony fishes. Essays on general questions of ichthyology. Pub. ASUSSR. 207-208. Berg, L.C. 1949. Fresh water fishes of USSR and its neighboring countries. Moscow, Leningrad. ASUSSSR. 2: 458, 3: 456. Chugunova, N.E. 1959. Manual for studying age and growth of fishes. Moscow. Pub. ASUSSR. 164. Danilenko, T.P. 1991. Sexual cycle of Volga pike perch, Stizostedion volgensis (Gmelin) of Zaporozhskii water reservoir. Hydrobiol. 27 (2): 33-40. Dimitrieva, E.N. 1973. Spawning ground of zander and Volga pike perches of Ural river. Ichthyology. 13 (5).: 934-937. Iashanin, E.E. 1978. Feeding of Volga pike perch of Central trenches of Kuibyshev water reservoir in spring. Fish Ecology of Kuibyshev water reservoir. Ulianovsk. 43-49.

134

Iashanin, E.E. 1982. Reproduction of Volga pike perch (Lucioperca volgensis Gmelin) of Central trench of Kuibyshev water reservoir. Changing of fish biology in condition to regulation of Volga river flow. Ulianovsk. 64-71. Iashanin, E.E. 1986. Sexual dimorphism of Volga pike perch Lucioperca volgensis (Gm.) of Kuibyshev water reservoir. Ecology and fish physiology of Kuibyshev water reservoir. Ulianovsk. 114-120. Kyznetsov, V.. 1980. Fluctuation of number of commercial fishes in conditions with regulation of river flow (e.g., Kuibyshev water reservoir). Ichthyology questions. Moscow. 20(5) : 805-811. Kyznetsov, V.. 2001. Fishes of Volga-Kama rivers. KSU Pub. Kazan. 5: 6990. Lakin, G.F. 1990. Biometry. .: High. Sch. 350. Lukin, A.V. 1952. Fisheries of Tatarstan and it's developmental prospectives. Kazan. : Tatarstan Govt. Press. 1952. 107. Nikitina, N.K. and N.G. Moskul. 1999. Number and stock of Volga pike perch in Krasnaderck and Shapcug water reservoir. 2nd Int. Sym. "Aquacultural resources technology" Mat. abst.156-157. Noviskii, R.A. 1999. Ecological feature of Volga pike perch, Stizostedion volgensis (Pisces, Percidae) of Dineprovskii water reservoir. Fam. Zoology. 33 (6): 63-72.

Pinter K. 1996. Hungary-present state and problems of recreational fishery. European inland fisheries advisory commission. Report of the Workshop on Recreational Fishery Planning and Management Strategies in Central and Eastern Europe. ilina, Slovakia, 2225 August 1995. EIFAC/OP32. Rome, FAO. 92. Pravadin, I.F. 1966. Manual-studying fishes. Food Indust. oscow. 373. SMIDTOV, .E. 1956. FISH POPULATION AND THEIR NUMBER IN THE REGION OF KUIBYSHEV WATER RESERVOIR. STD. RES. KSU. 116

135

(1): 221-226. Smirnov, G.M. 1984. Growth, feeding and economic value of Volga pike perch of Kuibyshev water reservoir. Changing of water based animal ecology in conditions of the water reservoir. Pub. KSU. Kazan: 94-103. Smirnov, G.M. 1986. Volga pike perch - Ecology of particular fish and fishing ground of Kuibyshev water reservoir. Kazan: Pub. KSU. 111-114. Tikhomirova, L.P. 1973. Volga pike perch, Lucioperca volgensis (Gmelin) of White lake. Ichthyology questions. Moscow. 13 (5): 932-934. Tuniakov, V.M. 1971. On the condition of spares and seasonal capturing of Volga pike perch in Tcimilianckii water reservoir. err. Volgagrad. Dept of fish. 5: 136-139. Tuniakov, V.M. 1974. Biology and production of Volga pike perch in Veselovskii water reservoir. err. Volgagrad. Dept of fish. 8: 153-161. Tuniakov, V.M. 1976. Volga pike perch [Lucioperca volgensis (Gm.)] of Proletarian water reservoir. err. Volgagrad. Dept of fish. 10 (2): 141-145. Wheeler A. 1978. Key to the fishes of Northern Europe. London: Frederick Warne, 613.

136

THE FISH AND FISHERMEN FROM LAKE ENCANTADA (ILHUS, BAHIA, BRAZIL) Rocha, G.R.A. Universidade Estadual de Santa Cruz, DCB. Rodovia Ilhus Itabuna, km 16. Ilhus BA, 45650-000, Brazil gecely@uesc.br Schiavetti, A. Universidade Estadual de Santa Cruz, DCAA. aleschi@hotmail.com Melo, V.G.V Universidade Estadual de Santa Cruz, DCAA vicentedovalle@zipmail.com.br

EXTENDED ABSTRACT ONLY DO NOT CITE Introduction The loss of biodiversity is one of the problems that development might cause. Therefore, it is necessary to create conditions for a sustainable use of the natural resources leading to the maintenance of species richness in present and future. The knowledge of species composition and abundance allows a most direct assessment of fundamental management goals such as maintenance of viable populations and native diversity. The objectives of our study were to identify fish species in the Lake Encantada and to determine the traditional knowledge about the local ichthyofauna and its use by the fishermen. The Basin of Almada River has a drainage area of 1.545 km2 and 252 km of perimeter and it is located in the Bahia Sate. The main water course is the Almada river with an extension of 138 km. Lake Encantada is formed by a depression in the basin of Almada river. An area of 11.800 ha around the lake was considered as an Environmental Protection Area (APA) in 1993.

137

The region has an average annual temperature of 24.5C and a highly seasonal pattern of rainfall, with an annual average of 1935 mm occurring mainly from November to March (Encarnao et al., 2000). Material and Methods Monthly samples were obtained with gillnets from 10 p.m. to 6 a.m. between May 2002 and February 2004. Gillnets with a stretched mesh size of 75 to 90 mm and a length of 34 to 61 m were used. Interviews and observations on food habits and traditional knowledge were made at Areia fishermen community. In the laboratory specimens were sorted, identified, counted, weighed (nearest 0.1 g) and measured (total length). The state of maturity was determined when possible. Species diversity measures were based on the proportional abundance of species. The Shannon information function (H), the richness (d), and the evenness (J) indices were calculated according to Magurran (1988) and the species dominance index according to Berger and Parker (1970). Results A total of 09 families comprising 11 fish species was collected during the sampling period (Table 1). Traditional fishermen mainly catch carpa, tilapia and a few other species. Table 1. Fish species from Lake Encantada, Ilhus. Scientific name Anchoviella sp Centropomus paralellus Centropomus undecimalis Clarias gariepinus Diapterus rhombeus Eugerres brasilianus Genidens genidens Myleus micans Prochilodus costatus Rhamdia quelen Oreochromis niloticus Local name Piaba bocuda Robalo Cambriau Ja, bagre africano Carapeba Carapeba Bagre Peixe-galo Carpa, curimba Jundi Tilapia Family Engraulidae Centropomidae Centropomidae Clariidae Gerreidae Gerreidae Ariidae Characidae Prochilodontidae Heptapteridae Cichlidae

138

Total abundance in number of individuals and weight are in Fig. 1 and 2.

Prochilodus costatus

Genidens genidens

Oreochromis niloticus

Eugerres brasilianus

Clarias gariepinus

Centropomus paralellus

Anchoviella sp.

Figure 1. Number of individuals by species. Prochilodus costatus was the most abundant species in number (46 %) and weight (44 %), followed by Genidens genidens in number (31 %) and O. niloticus in weight (15 %). Prochilodus costatus is caught at sizes between 15 to 36 cm with an average of 27.5 cm. Genidens genidens is caught at lengths varying from 16 to 32 cm, with an average of 22.6 cm. The African catfish Clarias gariepinus, an introduced species, was the largest species caught in the lake, with sizes between 37 to 88 cm and an average of 51.1 cm.

139

Genidens genidens Centropomus paralellus Clarias gariepinus Anchoviella sp Eugerres brasilianus Oreochromis niloticus Prochilodus costatus Centropomus undecimalis
Figure 2. Total weight by species Species diversity by sample, measured by Shannon-Wiener index, ranged from 0.578 to 1.606 natural bels per individual. No clear pattern appeared, however higher values were observed during the rainy season. The richness index fluctuated between 0.31 and 1.85. Evenness index reached peaks at different months, and varied from 0.46 to 0.99. Dominance index ranged from 0.53 to 1.00, with an average value of 0.77. The highest values of dominance were observed during the dry season. These low diversity values were related to low richness and high dominance observed as a result of a great abundance of species such as Prochilodus costatus, and Genidens genidens. A seasonal diversity pattern was only observed when cumulative samples were considered, with H equal to 1.368 and 1.620 natural bels in dry and rainy season, respectively. Both evenness and richness had a similar pattern. Dominance index had an average of 0.81 in dry season and of 0.74 in rainy season. The index of species diversity known by fishermen was equal to 2.84, revealing a high knowledge about the local fish assemblage. Discussion Freshwaters have their own faunas of primary freshwater fishes, being those which have evolved in freshwater and are unable to tolerate brackish water conditions. Of the world total of c. 8500 freshwater species about 93 % are ostariophysian fishes. These fishes include the carps (Cyprinoidei), characins (Characoidei) and catfishes (Siluroidei) of numerous families. Secondary

140

freshwater fishes, species evolved in freshwater but of marine groups and able to withstand brackish waters, include the cichlids so important in the lake faunas of Africa and in Central and South America rivers and lakes (Lowe-McConnell, 1977). Given the greater concern and awareness of conservation of aquatic biodiversity, which generally conflict with maximizing the production of a system, knowledge on the effects of an expanding fish with highly selective gears on the fish community structure and diversity is essential both from an ecological and fisheries management point of view. The fishery of Lake Encantada is a small-scale fishery. Gillnets are made by the fishermen mostly from monofilament lines obtained from nylon rope. One to two persons work on a canoe. Hook and line and tarrafa are used occasionally. The major reason for this has been the low level of technology employed by the traditional fishermen and a lack of storage, distribution, and marketing facilities. The characteristics of the canoes imply that it is mostly a subsistence fishery, whose product is partially sold in Ilhus. Margalef (1974) argues that a comparison between the cumulative sample diversity with the average diversity of individual samples can indicates the heterogeneity of the samples. In Lake Encantada, the overall diversity index (1.45 natural bels/individual) was higher than the average diversity (1.18), indicating a high sample heterogeneity. Introduction of exotic species can have negative impacts on local ecosystems. These include the introduction of parasites and pathogens, the introduction of hitchhiker species, and biological competition with native species, which results ultimately in a loss of biodiversity. Conversely, biological invasions by exotic species can alter the composition and community structure of invaded areas. Biological invaders change ecosystems by differing from native species in resource acquisition and/or resource use efficiency, by altering the trophic structure of the area invaded, or by altering disturbance frequency and/or intensity. At least 20 fish species have been introduced in Brazil (Agostinho and Julio Jr., 1996). Our study shows the presence of two introduced species, Oreochromis and Clarias. The introduction of different species popularly known as tilapia have long been reported in Brazil. The first of them might be that of the genus

141

Tilapia in 1953 (Nomura et al., 1972). Many of these species occur now in natural and artificial reservoirs of different river basins in many Brazilian States. The official occurrence of the African catfish Clarias gariepinus was reported in the River Paraopeba (So Francisco river basin), the River Grande (Paran river basin), and the River Doce (Doce river basin), three important Brazilian river basins (Alves et al., 1999). More recently the range of its occurrence reached the estuary of Patos Lagoon, in the south of Brazil (Braun et al., 2003). This is the first record in Almada river basin, in the State of Bahia. The advisability or otherwise of introducing exotic fish species for food raises problems that call for much careful research, for once introduced, exotic fishes are generally impossible to eradicate, and in many cases such introductions have led to extinction of indigenous species. However, the Brazilian fish fauna is still poorly know and only a preliminary list of endangered fish species is available (Rosa & Menezes, 1996). No case of native fish species extinction caused by introduction of species is reported in Brazil. As a result of the introduction of these species, and of the price that fish has on the local tourist activity, a cultural change on the food habits of the fishermen from Lake Encantada have been observed, who are consuming proteins coming from other sources but fishing. The food diet of the fishermen families shows an ingestion of fish varying between 11 and 55%, being the tilapia the most consumed fish. In most of the examples of intentional or accidental introductions we have no clear evidence of the effects of the introduction on other species, for the most part because we have no adequate baselines of pre-introduction condition for evaluating the impact of introductions. The biology of the species in the Lake Encantada is being studied to determine the effects on its local ichthyofauna. References Agostinho, A.A. and H.F. Julio Jr. 1996. Ameaa ecolgica: peixes de outras guas. Cincia Hoje, 21(124): 36-44. Alves, C.B.M.; V. Vono and F. Vieira. 1999. Presence of the walking catfish Clarias gariepinus (Burchell) (Siluriformes, Clariidae) in Minas Gerais state hydrographic basins, Brazil. Revta bras. Zool., 16(1): 259-263.

142

Braun, A.S.; P.C.C. Milani and N.F. Fontoura. 2003. Registro da introduo de Clarias gariepinus (Siluriformes, Clariidae) na Laguna dos Patos, Rio Grande do Sul, Brasil. Biocincias, Porto Alegre, 11(1): 101-102. Berger, W.H. and F.L. Parker 1970. Diversity of planktonic Foraminifera in deep sea sediments. Science, 168:1345-1347. Encarnao, A.M.V.; R.A.C. Miranda and W.F. Ferreira. 2000. Perfil pluviomtrico nos municpios da Bacia do Rio Cachoeira no sul da Bahia (srie 1984-1998). Reunio Brasileira de Manejo e Conservao do Solo e da gua, XIII. Ilhus, 2000. Resumos. p 94-95. Lowe-McConnell, R.H. 1977. Ecology of fishes in tropical waters. Edward Arnould, London. 64 p. Magurran, A.E. 1988. Ecological diversity and its measurement. London, Croom Helm. 179 p. Nomura, H.; A.R. Alves; A.M. Bonetti and D.E. Iost. 1972. Identificao especfica da Tilapia Smith, 1840 (Pisces, Cichlidae), introduzida no Brasil em 1953. Revta bras. Biol., 32(2): 157-168. Rosa, R.S. and N.A. Menezes. 1996. Relao preliminar das espcies de peixes (Pisces, Elasmobranchii, Actinopterygii) ameaadas no Brasil. Revta bras. Zool., 13(3): 647-667. Acknowledgements Financial support from CNPq (APQ 478874/2001-4) and UESC is gratefully acknowledged.

143

144

GONAD MATURATION SCALES FOR FISHES FROM THE UPPER TOCANTINS RIVER (GOIS, BRAZIL). Maria Fernanda Nince Ferreira Universidade de Braslia/ICB/GEM, Campus Asa Norte, Braslia, DF, Brasil, CEP: 70910-900 phone: 61 3072169, fax: 61 2734942 email: mfnf@unb.br Maria de Ftima Valentim and rica Pellegrini-Caramaschi Universidade Federal do Rio de Janeiro/ICB/Laboratrio de Ecologia de Peixes, Ilha do Fundo, RJ, Brasil, CEP: 21941-590. Introduction The settlement of maturation stages is one of the first steps in fish reproduction studies, providing information to detect the spawn season or the proximity of it. Frequently were used microscopic maturation scales but the importance of histological analysis in recognizing and/or confirmation of those stages has been emphasized (Vazzoler, 1996). On Tocantins River very few works have presented or discussed gonad maturation scales for the local species (Braga, 1990;Valentim, 1998; Anto, 2000; Lima, 2000). Furthermore, a very few works were produced based on data obtained from male individuals and presented histological descriptions about the reproductive system (Andrade and Godinho, 1983; Narahara, 1983; Agostinho et al., 1987a; Benedito-Ceclio and Agostinho, 1991a; Romagosa et al., 1993; Menezes and Pellegrini-Caramaschi, 1994). This work approached reproductive biology aspects of representative species of upper Tocantins River. The gonad maturation scales are assessed and discussed and suggested the spawning season of the analysed species.

Methods Ageneiosus brevifilis, Hypostomus emarginatus, Leporinus friderici, Pimelodus blochii, Plagioscion squamosissimus, Prochilodus nigricans, Rhaphiodum vulpinius and Serrasalmus rhombeus (which its importance in the region was assessed in values higher than 1% of the ponderal index, that associates number

145

of individuals to biomass) were analysed.. The capturess took place each two months, between December 1995 and February 2000, with nets in different sizes and types, witch were checked every 8 hours during 24 hours. Each individual was dissected and by visual exam of the gonads settled the gender and maturation stage. Based on the macroscopic features it was defined a maturation scale. The classification was made in account of the following set of macroscopic characteristics: gonad position in coelomic cavity; percentage occupation in coelomic cavity; forms; irrigation; color or colorless (testicle); turgidity; visibility, size and color of oocytes (ovary) and spontaneous elimination of sperm, under light or heavy pressure (testicle). For the histological analysis using light microscopy, the gonads were fixed in phormol at 10%, buffered , processed according to the routine for light microscopy and stained with Hematoxilin and Eosin (HE). The morphology and proportion of the germinative epitellium cells were observed in accordance with Wallace and Selmans nomenclature (1981) for females and Griers et al. (1978) and Grier and Taylors (1998) for males. Results Females Maturation Scale The stages considered microscopically consistent were: Immature, Initial Maturation, Advanced Maturation, Ripe, Partially Spawned/Spawned, Recovering and Recovered, described here under: Immature: macroscopically ovaries occupy < 1% of coelomic cavity, are transparent, color from white to rosy and without oocytes visible at naked eyes. Microscopically, only OI and OII can be observed, and the OI present themselves in nests, most of the time. Initial Maturation: the ovaries become slightly wide and appear translucent and yellow (P.squamosissimus e L. friderici ), cream color (A. brevifilis, P. blochii e H .emarginatus) or orange (S. rhombeus e R. vulpinius). The irrigation is discreet in all of the species, exception made to S. rhombeus, in which its more conspicuous. Oocytes are visible at naked eye as whitish granules. This stage was identified in the field as maturation 1 and/or 2. In histological sections, OII and some OIII are observed. Advanced Maturation: they occupy larger volume, except in P. squamosissimus, which appears proportionally inferior to the other species. The color is yellow, except in S. rhombeus (orange color). In H. emarginatus, P. blochii, A. brevifilis and S. rhombeus the oocytes are heterogeneous in regard to the opaque or

146

translucent aspect. The irrigation is evident. On the other side, in L. friderici, R. vulpinius and P. nigricans, just homogeneous oocytes (same size) with similar aspect can be observed. The irrigation is discreet. In P. squamosissimus the oocytes are small and the size heterogeneity is not visible at naked eye. The irrigation is evident. S. rhombeus, P. blochii, H. emarginatus, A. brevifilis and P.squamosissimus present some OII, OIII, OIV and some OV. But in L. friderici, R. vulpinius and P. nigricans OIV predominate; there are rare OV. Ripe: it reaches maximum volume and maintains the same macroscopic features as in previous phase, except for the greater proportion in translucent oocytes visible at naked eye. Nevertheless, microscopically, the species A. brevifilis, P. blochii, S. rhombeus, P. squamosissimus and H. emarginatus presented, concomitantly, OII, OIII, OIV and OV, with predominance of the last one. The species L. friderici, R. vulpinius and P. nigricans presented only OII and OV, again with predominance of the last one. Semi-spawned: the ovaries present characteristics referring to shape, color and visibility of oocytes similar to the ones presented in the previous two phases, but they are less voluminous, present hemorrhagic areas, evident irrigation and moderate flabbiness. We noticed OII, OIII, OIV, OV and follicle post ovulatory (FPO) . The stage was clearly recognized for A. brevifilis, P. blochii, S. rhombeus, P. squamosissimus and H. emarginatus. Spawned : the ovaries are flabby and hemorrhagic, with some translucent oocytes and some others opaque and white. Such characteristics were observed in L. friderici, R. vulpinius and P. nigricans. Its noticed tissue disorganization and predominance of OII, a number of OII in absorption process, post spawning follicles (FPO) and atresic follicles (FA). Recovering: they appear hemorrhagic and with scattered white oocytes. Predominate OII, the vitelogenic oocytes in absorption, follicle scars and further, tissue disorganization Recovered: the ovaries are similar to Immature stage, however they appear larger and longer in all of the species. Microscopically, they are identified by the predominance of OII, distended ovuligerous lamellas and by the presence, sometimes, of follicle scars. This stage was verified in all of the studied species, exception made to H. emarginatus, in which it could not be identified in histological exam. For this species, in the ovaries still in the Recovering stage,

147

observations could be made regarding tissue disorganization, predominant presence of OII (with increased volume) and some OIII. Macroscopically, the stages Recovering and Recovered get confused concerning the species P.nigricans, R. vulpinius, P. blochii and A. brevifilis. Recovered can be confused with the Initial Maturation stage in the species with very thick ovarian capsule (P. blochii and A. brevifilis). In A. brevifilis it was noted, in histological sections, the presence of spermatozoids among the oocytes in all of the stages, except for the Immature stage. Males Maturation Scale A scale with 07 (seven) macroscopic maturation stages was considered microscopically consistent and described hereunder: Immature : dorsal position in H. emarginatus, A. brevifilis, and P. blochii and lateral or dorso-lateral for the other species. Occupation of less than 1% of the celomatic cavity. Finger-like format in P. blochii and A. brevifilis. Discreet irrigation. The gonad is transparent and without coloration. Absence of sperm. Histologically the testicles are constituted by spermatogonias , wrapped up by abundant conjunctive tissue, dispersed or forming a very few seminiferous tubules. Maturation: more and more lateral e with greater volume in all of the species. Regarding H. emarginatus, it always maintains a dorsal position. Occupation of the coelomic cavity from 5% to 30% . Tubular format, due to the increase of the volume and foliaceus as regards S. rhombeus. The irrigation is discreet; the color is white; opaque. The sperm (evidenced macroscopically by white coloration), may be eliminated under some pressure. The seminiferous tubules are already defined and may be observed germinative cells during the different spematogenesis phases. At first, there is an increase of the secondary spermatogonia cysts. Progressively, spermatocyte cysts 1 and 2 and spermatids are formed. It occurs the production of spermatozoids which fulfill the tubules lumen and drainage ducts. Ripe: increase of the volume with elimination of sperm under light pressure. The tubules have large lumens and are fulfilled by spermatozoids. Many of them are ramified and anastomosed. Few cysts are detected. The conjunctive tissue that covers the tubules is scarce.

148

Partially spent : with the partial liberation of sperm, the testicles turn back to the lateral , latero-dorsal or dorsal position, depending on the species. The occupation is between 20% and 50%. The irrigation is hemorrhagic. The coloration varies between white and rosy, due to hemorrhage. The sperm is eliminated under pressure. Histologically, the main characteristic is the presence of tubules emptying. In R. vulpinius, P. nigricans and S. rhombeus the release of sperm does not happen completely, leaving the gonads with heterogeneous aspect. Spent: it occurs a remarkable decrease of the occupation volume (10% up to 20%). The hemorrhagic irrigation provides a rosy coloration. The sperm is little visible, without elimination. Typical flabby appearance. Tubules smaller and empty. The germinative epithelium is scarce, the cysts are absent and the spermatozoids are rare. Recovering: small occupation ( 2% up to 5 %). The irrigation is a little hemorrhagic, transparent and colorless. Macroscopically the sperm is not visible. The tubules present both regular aspect , small size and spermatogonias coating as emptied, without coating and rare spermatozoids. Recovered : similar to Immature with 1% to 2% of occupation. Tubules well defined, with spermatogonias coating and a few cysts of secondary spermatogonias. Macroscopical versus microscopical Regarding the males, 87,5% of the species presented accordance between macro and microscopic diagnoses over 50%. This percentage is lower for the females (37,5%) (Table 1). Table 1. Accordance percentage (%) between macro and microscopic diagnoses of testicles and ovaries of the studied species (N= number of cases; F = female; M= male). Species N Concordncia % M F M F A. brevifilis 14 29 100 51.7 H emarginatus 20 74 50 39.2 L. friderici 13 79 692 49.4 P. blochii 12 42 333 50 P. squamosissimus 58 204 97.6 37.2

149

P. nigricans R. vulpinius S. rhombeus

14 10 20

97 118 156

100 80 80

47.4 28.8 66

In Table 2 the accordance percentage is presented by stage for the total of the studied species. Table 2. Accordance percentage of macro and microscopic diagnoses of testicles and ovaries during the considered Maturation stages (N = number of cases; F = female; M = male; * only males). Stages N Concordncia % M F M F Immature 16 151 94.1 59.6 Initial Maturation 105 12 Advanced Maturation 40 57.5 Maturation* 44 93.2 Ripe 20 30 90 63.3 Partially Spawned 42 33 Partially Spent* 16 87.5 Spawned 38 13.2 Spent* 13 92.3 Recovering 24 161 91.6 46 Recovered 16 267 12.5 56.9 The classification error referring to gender was of 8%, concentrating on Immature and Recovered stages. This percentage reaches 29% for P. blochii e 20% for S. rhombeus and H. emarginatus. Reproductive period From the incidence of Mature stage the spawning season is indicated (Table 3).

150

Table 3 . Incidence of individuals, male and female, in the ripe stage. () % of ripes referring to the total of adult individuals during the period; * < 1%
Species Feb Sex M A. brevifilis H. emarginatus L. friderici P. blochii P. squamosissimus P. nigricans R. vulpinus S. rhombeus Apr Jun Aug Oct Dec

0 0 2 (6) 1 (1) 11 (6) 0 1* 0 0 1 (20) 0 0 18 (14) 21 (13) 30 (21) 6 (9) 0 0 9 (6) 1 (1) 0 6 (2) 8 (2) 6 (2) Feb Apr Jun Sex F A. brevifilis 0 1 (1) 2 (5) H. emarginatus 39 (45) 22 (31) 12 (15) L. friderici 3 (2) 1* 0 P. blochii 23 (19) 1 (1) 0 P. squamosissimus 4 (4) 6 (6) 14 (17) P. nigricans 2 (3) 1 (1) 2 (5) R. vulpinus 17 (12) 5 (2) 0 S. rhombeus 3 (1) 18 (5) 6 (2)

10 (29) 21 (32) 0 6 (2) 5 (1) 8 (5) 0 0 1* 0 0 1 (2) 37 (24) 43 (19) 36 (15) 0 4 (8) 15 (18) 0 4 (3) 28 (11) 4 (2) 6 (2) 15 (4) Aug Oct Dec 3 (9) 3 (20) 1 (6) 120(52) 74 (46) 28 (58) 0 4 (3) 10 (4) 1 (2) 1 (2) 26 (23) 14 (12) 10 (5) 8 (4) 0 0 6 (9) 0 21 (11) 39 (20) 16 (8) 10 (4) 14 (4)

Discussion We concluded that the proposed scales were efficient and can be used in the field. The characteristics used for the description of gonad proved adequate once they comprehend the morphological variety of the analysed species. (belonging to taxonomically distinct groups) and the variations originating from the reproductive process. For the females, it was registered the natural occurrence of two groups with distinct ovarian development, pointing to different reproduction strategies. The occurrence of the Partially Spawned stage is related to the species which present

151

asynchronic development in more than two groups and spawn in more than one lot, along the reproductive period. They are: A. brevifilis, H. emarginatus, P. blochii, P. squamosissimus and S. rhombeus. The scale with Spawned stage characterizes the species which present synchronic development in two groups (Vazzoler, 1996) which spawn just in one lot at each reproductive season. To this category belong the following species: L. friderici, P.nigricans and R. vulpinius. The macroscopic identification can be done through means of observation , in the stages Advanced Maturation and Ripe, of the ovocytes size. The histological analyses will confirm the assessment. The absence of the Recovered Stage in the ovaries of H. emarginatus suggests continuous cell maturation process (Pellegrini-Caramaschi et al., 1982). It is advisable the histological confirmation for the gender identification of individuals from P. squamosissimus and S. rhombeus with standard length inferior to 15,0 cm (Valentim, 1998) and 12,0 cm (Anto, 2000), respectively, and for ovaries in the Advanced Maturation, Partially Spawned and Spawned stages. With the males few variations occurred as regards to constitution and development of the germinative epithelium, representing a common pattern for the species, coherent with the literature (Grier et al., 1978; Menezes and Pellegrini-Caramaschi, 1994; Grier and Taylor, 1998). The macroscopic characteristics utilized proved being very effective in the stages identification since it can be noticed that, except for Pimelodus blochii, the remaining species presented an accordance percentage over 50%. The lowest value observed in Recovered stage is due to the small size of the testicles in this phase, inducing to error in the distinction between Immature and Recovered stages and in the gender identification as well. It is advisable a complementary histological analyses with S. rhombeus and H. emarginatus. One can notice that the mature females are available during a greater number of months comparison made with the males. The period ranging from December until February proved being the one with the greatest number of species with ripe individuals occurrence. Acknowledgements

152

We thank to the partnership with FURNAS CENTRAIS ELTRICAS S/A; SERRA DA MESA ENERGIA S/A and FUNDAO BIORIO/UFRJ which sponsered the project Basic Studies on Ictiofauna of Aproveitamento Hidreltrico Serra da Mesa, GO of which the present study makes part. References Agostinho, A. A.; Barbieri , M. C.; Agostinho, C. .S.; Barbieri, G. Biologia reprodutiva de Rhinelepis aspera (Agassiz, 1829) (Teleostei, Loricariidae) no rio Paranapanema. I. Estrutura dos testculos e scale de maturidade. Rev. Brasil. Biol. , v. 47, n. 3, p. 309-317, 1987a. Andrade, R. A.; Godinho, H. Annual male reproductive cycle of the brazilian teleost fish Leporinus silvestrii (Boulenger, 1902). Arch. Biol., Bruxelles, v. 94, p. 1-14, 1983. Anto, H. B. Estrutura populacional e biologia reprodutiva das females de Serrasalmus rhombeus (Linnaeus, 1766) (Teleostei: Characiformes) nas fases anterior e posterior ao represamento do rio Tocantins, pela UHE Serra da Mesa, GO. 2000. Dissertao (Mestrado) - PPGCB, Universidade Federal de Juiz de Fora. 2000. Benedito-Cecilio, E.; Agostinho. A. A. Biologia reprodutiva de Hypophthalmus edentatus (Spix 1829) (Osteichthyes, Siluriformes) na reserva de Itaip-PR. I. Estrutura dos testculos e scale de maturidade. Rev. UNIMAR, v. 13, n. 2, p. 196-209, 1991a. Braga, F. M. de S. Aspectos da reproduo e alimentao de peixes comuns em um trecho do rio Tocantins entre Imperatriz e Estreito, Estados do Maranho e Tocantins, Brasil. Rev. Brasil. Biol., v. 50, n. 3, p. 547-558, 1990. Grier, H. J.; Fitzsimons, J. M. ; Linton, J. R. Structure and ultra structure of the testis and sperm formation in goodeid teleosts. J. Morphol., v. 156, p. 419438, 1978. Grier, H. J.; Taylor, R. G. Testicular maturation and regression in the common snook. J. Fish Biol., v. 53, p. 521-542, 1998.

153

Lima, D. Influncia da formao do reservatrio do AHE Serra da Mesa, alto rio Tocantins, GO, na abundncia, estrutura da populao e reproduo de duas espcies de Leporinus (Teleostei, Anostomidae). 2000. Dissertao (Mestrado) - PPGE, Universidade Federal do Rio de Janeiro, Rio de Janeiro, 2000 Menezes, M. S.; Caramaschi, E. P. Caractersticas reprodutivas de Hypostomus grupo H. punctatus no rio Ubatiba, Maric, RJ (Osteichthyes, Siluriformes). Rev. Brasil. Biol., v. 54, n. 3, p. 503-513, 1994. Narahara, M. Y. Estrutura da populao e reproduo de Rhamdia hilarii (Val. 1840) (Osteichthyes, Siluriformes, Pimelodidae). 1983. Tese (Doutorado em Cincias) Universidade de So Paulo, So Paulo, 1983. Pellegrini-Caramaschi, E., Godinho, H. M. and Foresti, F. Reproduo de Hoplias malabaricus (Bloch, 1794) (Teleostei, Erythrinidae) na represa do rio do Pardo (Botucatu, SP) I. Histologia e scale de maturation do ovrio. Rev. Brasil. Biol., v. 42, n. 3, p. 635-640, 1982. Romagosa, E.; Narahara, M. Y.; Talmelli, E. F. A.; Braga, F. M. S. Mudanas morfolgicas dos testculos de pacu, Piaractus mesopotamicus (Holmberg, 1887), em condies de confinamento. Rev. UNIMAR, v. 15, n. esp, p. 1-17, 1993. Valentim, M. F. M.. Biologia reprodutiva das females da pescada de gua doce P. squamosissimus (Heckel, 1840) (Teleostei: Sciaenidae), antes e durante a formao do reservatrio da UHE Serra da Mesa, no alto rio Tocantins, GO. 1998. Dissertao (Mestrado) - PPGBA, Universidade Federal Rural do Rio de Janeiro, Itagua, 1998. Vazzoler, A. E. A. de M. 1996. Biologia da reproduo de peixes telesteos: teoria e prtica. Maring: Universidade Estadual de Maring, 1996. Wallace, R. A.; Selman, K. Cellular and dynamic aspects of oocyte growth in teleosts. Amer. Zool., v. 21, p. 325-343, 1981.

154

THE COMPOSITION AND STRUCTURE OF FISH COMMUNITIES OF FLOODPLAIN LAKES AT LOWER STRETCH OF SOLIMES RIVER (AMAZON BRAZIL). Flvia Kelly Siqueira - Souza Federal University of the Amazonas Av. General Rodrigo Otvio, 3000 Coroado (092) 647 4064 fla_kel@yahoo.com.br Carlos Edwar de Carvalho Freitas Department of Fisheries Sciences - Federal University of the Amazonas Av. General Rodrigo Otvio, 3000 Coroado (092) 647 4064 cefreitas@horizon.com.br EXTENDED ABSTRACT ONLY DO NOT CITE Introduction There are approximately 8,500 freshwater fish species Lowe-McConnel (1999), and most of it occurs in rivers and connected alluvial floodplains. These communities show a dynamic structure that reflects the characteristics and alterations that interact with biotic processes, specially predation and competition (Perrson, 1997; Jackson et al., 2001). The richness in species in the Amazon Basin is well known, Bolke et al., (1978) it esteems the total number of present species in the basin in about 2000, being 30% of these still ignored. Methodology From June 2001 to April 2003, we accomplished bimonthly experimental fisheries in nine floodplain lakes (Lake Sacambu, Sumauma, Preto, Iauara, Samauma, Campina, Marac, Poraque and Lake Arua) at lower stretch of Solimes river that crosses Amazonas state from West to East for 463 km since Coari city until its confluence with the Negro river. The sampling gear were

155

gillnets of standardized dimensions (20 x 2m) and several mesh sizes. The fishes captured were identified with help of ictiological keys of reference or by specialists of the area and were measured the length standard (cm) and weight (g). The used methodology was the capture for unit of effort (CPUE) and the constancy as Bodenheimer (1955) to identify the constant, accessory and accidental species. Results and Discussion We registered 118 species belonging to 21 families in 5 orders. In the season of the hight water was observed the smaller value of richness (15 species). The Campina lake presented the biggest richness in the receding with 40 species and in the rising with 41 species. The low water season was better represented by the Samama lake with 32 species. The constancy (persistence) of the presence of the 21 families of fish in the showed lakes, demonstrated that 15 families were constant (occurring in at least half of the occasions of collections) and the remaining families divided in 03 accessory and 03 accidental ones (less of of the collection places). The data below (Figs. 01 and 02) are presented in the form of percentile values of CPUEn. The graphics are commanded in decreasing way in relation to the constancy and increasing in relation to the variation coefficient. In relation to the hydrological pulse, figure 01 presents the result of the constancy of the families of fish in floodplain lakes, as Clupeidae and Hypophthalmidae example was the families which occurred in a broad view more homogeneously way manner along the years. You can observe that from the families witch occurred in the collection occasions, 5 them had shown more regular. The five families are: Clupeidae (cv=42,36%), Hypophthalmidae (cv=46,26%), Callichthyidae (cv=52,17%), Pimelodidae (cv=58,24%) and Doradidae (cv=58,63%), being that the last four ones belong to the order of the Siluriformes. The families that showed less regular were Characidae (cv=104,55%), Sciaenidae (cv=95,84%), Curimatidae (cv=95,11%), Serrasalmidae (cv=89,54%) and Osteoglossidae (cv=85,29%), of these the four first ones

156

concentrated in the period of the receding. Some families not presented capture in at least one station, such as Arapaimidae, Cynodontidae, Erythriniidae, Gastropelieidae, Hemiodontidae, Ageneiosidae and Auchenipeteridae, therefore it wasnt possible to evaluate the variation coefficients, more it is possible to notice that the majority of these were concentrated in the receding and rising season. In general, the families had shown higher values in the periods of receding and rising, with predominance of the first one. This probably occurs according the largest area available. The Characidae and Sciaenidae run away from this standard, with low values in the flood. In figure 02, it is perceived distribution for family, of the values of CPUE in the nine lakes floodplain showed. Of the families, 15 had been constant, appearing in all the showed lakes. We used as criteria to evaluate the variation coefficient of the families which had appeared in at least 7 lakes. The Ageneiosidae family presented better distribution between all the lakes, while Loricariidae got greater concentration in the Iauara lake. In short, the Campina lake presented the biggest values of CPUE of all the lakes (56%). From the families which appeared in at least 78% of the most regular lakes in one hand Ageneiosidae (cv=67,70%), Anostomidae (cv=67,96%), Pimelodidae (cv=71,74%), Characidae (cv=75,41%) and Doradidae (cv=82,89%) were the more representative. While in the other hand, in the same condition of constancy Loricariidae (cv=215,36%) Serrasalmidae (cv=134,02%) Cichlidae (cv=133,33%), Clupeidae (cv=121,41%) and Prochilodontidae (cv=112,84%) were less regular. Acknowledgments The authors would like to thank UFAM, CNPq and FINEP (PIATAM Project) for financial support. References Bodenheimer, F.S. 1955. Prcis dcologie animale. In: Silveira Netto, S.; Nakano; D. Barbin; N.A. Villa Nova. 1976. Manual de ecologia dos insetos. Piracicaba, Ed. Agronmica Ceres. 419p.

157

Bohlke, J.E.; Weitzman, S.H. & Menezes, N.A. 1978. Estado atual da sistemtica dos peixes de gua doce da Amrica do Sul. Acta Amazonica, 8 (4) 657-677. Jackson, D.A.; Peres-Neto, P.R. & Olden, J.D. 2001. What controls who is where in freshwater fish communities the roles of biotic, abiotic, and spatial factors. Canadian Journal of Fisheries and Aquatic Science, 58:157170. Lowe-McConnell, R. 1999. Estudos ecolgicos em comunidades de peixes tropicais. EDUSP, So Paulo SP, 524p. Persson, L. 1997. Competition, predation and environmental factors and environmental factors structuring forces in freshwater fish communities: Sumari (1971) revisited. Canadian Journal of Fisheries and Aquatic Science, 54:85-88.

158

Fig. 1 Varincia da dominncia relativa (CPUE) das famlias de peixes capturados nos lagos amostrados no perodo de 2001-2003. C = cheia, V = vazante, S = seca e E = enchente.
clupeidae 100

arapaimidae
osteoglossidae

100

100

50

50

50

0 C V S E

0 C V S E

0 C V S E

anostomidae 100

curimatidae
characidae 100

100

50

50

50

0 C V S E

0 C V S E

0 C V S E

cynodontidae 100

gastropelieidae
eryt hrinidae

100
100

50

50

50

0 C V S E

0 C V S E

0 C V S E

hemiodontidae 100
100

serrasalmidae
prochilodontidae

100

50

50

50

0 C V S E

0 C V S E

0 C V S E

ageneiosidae
auchenipteriidae

callichthyidae 100

100

100

50

50

50

0 C V S E

0 C V S E

0 C V S E

159

doradidae 100

locirariidae
hypopht halmidae 100

100

50
50
50

0 C V S E

0
0 C V S E

pimelodidae 100

cichlidae
sciaenidae

100
100

50

50

50

0 C V S E

0 C V S E

0 C V S E

Fig. 1 Varincia da dominncia relativa (CPUE) das famlias de peixes capturados nos lagos amostrados no perodo de 2001-2003. C = cheia, V = vazante, S = seca e E = enchente.

160

clupeidae 100
10 0

arapaimidae
osteoglossidae 100

50

50

50

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

anostomidae 100

characidae 10 0
100 curimatidae

50

50
50

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

cynodont idae
erythrinidae
gast ropelieidae 100

100

100

50

50

50

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

hemiodont idae 100

100 prochilodontidae
100 serrasalmidae

50
50
50

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

ageneiosidae 100

auchenipteriidae 100
100 callichthyidae

50
50
50

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

doradidae 100
100 hypophthalmidae
100 locirariidae

50

50

50

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

161

pimelodidae 100
100

cichlidae
100

sciaenidae

50

50

50

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

0 I II III IV V VI VII VIII IX

ig. 2 Varincia da dominncia relativa (CPUE) das famlias de peixes capturados nos nove lagos amostrados. (I = Lago Sacamb, II = L. Sumama, III = L. Preto, IV = L. Iauara, V = L.Samama, VI = L.Campina, VII = L. Marac, VIII = L. Poraqu e IX = L. Aru.)

162

COMPOSITION OF FISH FAUNA IN SALINA LAGOON IN THE ESTUARY OF RIO CAET, AM, BRAZIL Ynglea Georgina de Freitas Goch Universidade Federal do Par - UFPA Ncleo de Estudos Costeiros - NEC Alameda Leandro Ribeiro, s/n, Bragana, Par, Brasil. Cep. 68600000 Phone (55) 093 425-1209 Fax (55) 93 425-1745 e-mail: ynglea@yahoo.com.br ynglea@inpa.gov.br Uli Saint-Paul, Uwe Krume, Jansen Zuanon EXTENDED ABSTRACT ONLY DO NOT CITE Introduction The knowledge of a communitys of fish composition represents a good indicator of the variations of the environmental factors. Studies accomplished in estuary in the Brazilian southeast show that less than 50% of the species of fish studied spawn in the growth of mangroves and they complete his life cycle there; and that the reproductive strategies used by the fish in the estuary can vary in agreement with geographical characteristics and existent abiotics in that ecosystem (Chaves & Bouchereau, 2000). Traditionally, investigations about the composition of the species, dominance structure and spatial and temporal distribution of the ictiofauna have been made being used methods of experimental fishing. The results already obtained inside of the project MADAM (Berger et al., 1999; Barletta, 1999; Barletta-Bergan, 1999), they show that the swamp plays an important part in the food supply and protection against predators of the ictioplancton, guaranteeing like this low mortality rates and high growth rates. The present research aimed to determine the composition of fish fauna the Salina lagoon in the estuary of Rio Caet, on the northern coast of Brazil. The seasonal differences in the distribution of fish species and the influence of physical and chemical factors on the structure of the fish community were observed and analyzed in relation to the probable spawning period of the most abundant fish species.

167

Material and Methods In the coastal plain Bragantina, located in the northeast of the state of Par, is the Saline lagoon, which is placed beside the highway that ties BraganaAjuruteua (PA-253), between the meridians of 460 50'W and 460 30'W and the parallel of 00 45'S and 10 07'S, with an area of 0,19 Km2 and circumference of 2,3 Km. For this work, samples were collected during three months of the rainy season (September, October and November of the year 2000) and three months of the dry season (February, March and April of the year 2001). The samples of fish were collected with the nets (different sizes among us opposite) aid, which they were observed in intervals of three hours, with a total period of 24 hours, in addition measured of salinity, temperature and dissolved oxygen were also made in the same intervals of time. The Capture For Unit of Effort-CPUE was calculated in biomass (grams) and exemplares/m2/24 number hours, through the division of the total area of the nets for the weight and number of the exemplary captured in each collection. Results and Discussion Nineteen different species were captured in the Salina lagoon. The total CPUE biomass was of 4705,17 g/m2/24 hours and 26,3 m2/24 hours CPUE in number of individuals. There were no differences in CPUE values occured between the rainy and dry seasons (CPUE Biomass-F1,8 = 1,779; p = 0,219; CPUE n of individuals-F1,8 = 2,507; p = 0,152); however a difference in capturation period (between day and night) was detected (CPUE Biomass- F1,8 = 8,869; p = 0,018 / Tukey p = 0,017; CPUE n of individuals-(F1,8 = 15,668; p = 0,004/Tukey p = 0,004) (Fig. 1). Rooker & Dennis (1991), they observed that a variety of species migrates of the habitats of the day for other habitats during the night, in search of food, and this migration facilitates the capture. The absence of significant modifications temporary/seasonal in the occurrence of the species of fish of the lagoon, it can be related with the dominance in all of the months of collection of the most abundant species and with characteristics eurihalinas (C. pectinatus, C. undecimalis, and M. curema) and to the isolation of the lagoon in most of the year.

168

4,0

900
A

3,5

800

CPUE Biom. (g/m /24h)

CPUE Indiv. (n/m /24h)

3,0 2,5 2,0 1,5 1,0 0,5

Day

700 600 500 400 300 200 100

Day

Dry

Rainy

Night

Dry

Rainy

Night

Figure 1. Values of CPUE for station and collection period in n of individuals (A) and biomass (B). Observe that the values of CPUE are higher at night. The average values of diversity and dominance in the Salina lagoon was that of 1.62 and 0.27, respectively. Among the abiotic factors studied (dissolved oxygen, temperature and salinity), salinity showed the highest seasonal difference, varying between 0.7 and 25 ppm during the six months collection period (Fig. 2).

28

40 35

22

30

Salinity (ppm)

Salinity (ppm)
sep oct nov dec jan feb mar apr

16

25 20 15 10 5

10

-2

may jun jul aug sep oct nov dec jan feb mar apr may

Figure 2. Monthly average of salinity of the Saline lagoon for the six months of collection (A) and for a cycle of one year (B) (may/2000 the may/2001). (Source: U. Krumme, MADAM).

169

The four most abundant species identified were: Achirus Achirus, Centropomus pectinatus, Centropomus undecimalis and Mugil curema with A. achirus and M. curema showing peaks of IGS in april and september, respectively. The structure and composition of the fish fauna of the Salina lagoon are highly dependent on the gradients of salinity and the rainfall, but significant temporal and seasonal modifications in the occurrence of fish species of the lagoon were not found. Acknowledgments The present work was supported by MADAM (Mangrove Dynamics and Management) project, CNPq-for the bag concession and UFPA Campus de Bragana. References Barletta, M. 1999. Seasonal changes of density, biomass and species composition of fishes in different habitats of the Caet estuary (North Brazil Coast East Amazon). PhD thesis, Univ. Bremen. 206 p. Barletta-Bergan, A. 1999. Structure and seasonal dynamics of larval and juvenile fish in the mangrove-fringed estuary of the Rio Caet in North Brazil. PhD thesis, Univ. Bremen. 220 p. Berger, U.; Glaser, M.; Koch, B.; Krause, G.; Lara, R. J., Saint-Paul, U.; Schories, D. & Wolff, M. 1999. An integrated approach to mangrove dynamics and management. J. of Coastal Conser. 5: 125-134. Chaves, P. T. & Bouchereau, J. 2000. Use of mangrove habitat for reproductive activity by the fish assemblage in the Guaratuba Bay, Brazil. Oceanol. Acta 23 (3): 273-280. Rooker, J. R. & Dennis, G. D. 1991. Diel, lunar and seasonal changes in a mangrove fish assemblage of southwestern Puerto Rico. Bull. Mar. Sci. 49 (3): 684-698.

170

FISH ASSEMBLAGES IN TEMPORARY PONDS ADJACENT TO TERRA-FIRME STREAMS IN CENTRAL AMAZONIA Victor Fernando Volpato Pazin Coordenao de Pesquisas em Ecologia (CPEC), INPA Av. Andr Arajo, 2936, Cep: 69011-970, Manaus AM, Brazil. victor@inpa.gov.br William Ernest Magnusson Coordenao de Pesquisas em Ecologia (CPEC), INPA bill@inpa.gov.br Jansen Zuanon Coordenao de Pesquisas em Biologia Aqutica (CPBA), INPA zuanon@inpa.gov.br Introduction: Some species of fish occur mainly or exclusively in temporary ponds, but most temporary-pond fish species are recruited from permanent water during heavy rainfall or when streams and rivers overflow their banks. Despite the fact that small streams are intimatelly connected physically, chemically and biologically to their riparian zones (Murphy and Meehan 1991) and that streams can serve as sources of colonization for adjacent ponds, few studies of aquatic communities have explored both streams and temporary ponds. The colonization and aspects of physical environment, such as habitat diversity and physical-chemical gradients, are important in structuring fish communities, and are the main factors that influence the species distribution at a local scale (Tonn and Magnuson 1982, Capone and Kushlan 1991). In this study, we investigated abiotic factors affecting fish assemblages composition in temporary ponds around small forest streams in a central Amazonian rainforest reserve.

171

Material and Methods: The study was carried in a terra-firme humid tropical forest. The Reserva Florestal Adolph Ducke (0255 N, 5959 W) covers 10.000 hectares. A central plateau divides the stream systems into eastern and western drainage basins. Western streams drain to the Negro River and eastern streams drain to the Amazon River. Information on the fish faunas in the streams, collected by Mendona (2002), is available in the project data bank. Pools were sampled beside the 28 stretches of streams, which were distributed throughout the reserve Plots included a 50 meters stretch of length along the stream canal. Plot width corresponded to the widht of the floodplain, and generally did not exceed 40 meters. Each pool was sampled 5 times at intervals of 2 months. The fishes were collected with scoop nets and fish traps in each pond. The effects of environmental factors on the fish assemblage structure in temporary ponds were investigated at three different spatial scales: micro-scale (between ponds), meso-scale (between streams [plots]); macro-scale (between drainage basins). The fish assemblage structure was represented by Principal Coordinates Analysis (PcoA) axes. The relations between independent variables and the species composition and richness, were investigated by multivaried multiple regressions. Nested subsets analysis were carried out in theNested Temperature Calculator program (Atmar & Patterson 1995). Results and discussion: Were found 18 fish species in the ponds, distributed in six orders and nine families. Of these, 14 had been recorded from stream in Reserva Ducke. Characiformes was the group with highest species richness. Cypridodontiformes, represented by two species of Rivulus had greater abundance, with 51% of 1486 collected individuals. The five more abundant species (Rivulus sp, 42,7%; Pyrrhulina brevis, 13,1%; Copella nigrofasciata, 12,2%; Rivulus compressus, 8,3%; Hyphessobrycon melazonatus, 6%) represented 82.4% of all specimens. Although the Reserva Ducke streams serve as dispersal routes for the species found in adjacent temporary ponds, there was no relationship between the abundance of species in streams and ponds. The assemblages had a hierarchical structure (nested subsets) with smaller ponds containing subgroups of the species found in larger ponds. The results suggest

172

that the repeated fish colonization in the ponds is a strong mechanism in the generation and maintenance of the hierarchic structure, due the temporary connections between the ponds and streams during strong rains. Most ponds are acessible to most species during flooding, and three species, Rivulus sp, Rivulus compressus and Pyrrhulina brevis can colonize areas beyond the direct reach of floods by jumping across humid ground. In micro-scale analysis, the species composition, based on abundance data, was related with the pond area. For presence-absence data, the PCoA axis were related with some structural habitat variables, such as canopy cover average depth of the water and pond permanence. There was no significant relationship between physical-chemical characteristics of the water in the ponds and the fish composition. The number of species per pond was significantly related to pond area, canopy cover and hydroperiod. In meso-scale analysis, for quantitative data, there was no significant relationship between species composition and abiotic variables. For presenceabsence data, fish composition tended to vary with pond area. Species richness was significantly related to pond area and water depth. In a macro-scale, the composition and species richness were similar between drainage basins. In this study, the pond fish assemblages was related to the structural physical gradients, generally associated with water availability, such as area, depth, hydroperiod, and canopy cover, but had little relationship with chemical characteristics of the water. Acknowledgements: We are grateful to Coordenao de Pesquisas em Ecologia (CPEC) of Instituto Nacional de Pesquisas da Amaznia (INPA). This study is part a masters thesis by the first author and was supported by Conselho Nacional de Desenvolvimento Cientfico e Tecnolgico (CNPq). Thanks to Sr. Domingos and Sr. Z, the fieldwork assistants. References: Atmar, W., Patterson, B. D. 1995. The nestedness temperature calculator: a visual basic program, including 294 presence absence matrices. AICS Research, Inc. University Park, NM and the Field Museum, Chicago.

173

Capone, T. A., Kushlan, J. A. 1991. Fish community structure in dry-season stream pools. Ecology, 72(3): 983-992. Mendona, F. P., 2002. Ictiofauna de igaraps de terra-firme: estrutura de comunidades de duas bacias hidrogrficas, Reserva Florestal Adolpho Ducke, Amaznia Central. Dissertao de Mestrado. Instituto Nacional de Pesquisas da Amaznia/ Universidade do Amazonas. 43 p. Murphy, M. L., Meehan, W. R. 1991. Streams Ecosystems. In: Influences of Forest and Rangeland Management on SalmonidFishes and Their Habitat. W.R. Meehan, ed. Special Publications 19. American Fisheries Society. Bethesda, Maryland. Tonn, W. M., Magnuson, J. J. 1982. Patterns in the species composition and richness of fish assemblages in Northern Wisconsin lakes. Ecology, 63: 1149-1166.

174

AMERICAN INTRUSION: RANGE EXTENSIONS OF SOME GROUNDFISH SPECIES DUE TO RECENT CLIMATIC CHANGES Alexei M. Orlov Russian Federal Research Institute of Fisheries & Oceanography 17, V. Krasnoselskaya, Moscow, 107140, Russia (095) 264-91-43 / (095) 264-91-87 orlov@vniro.ru Abstract Until present only continental slope of the Bering Sea was considered to be the way, along which some typical representatives of American ichthyofauna are able to migrate to or its pelagic eggs/larvae may be transported to Asian coasts (Pacific halibut Hipposlossus stenolepis, arrowtooth flounder Atheresthes stomias, rex sole Glyptocephalus zachirus, sablefish Anoplopoma fimbria). Recent studies demonstrated that exchange between Asian and American ichthyofauna takes place along Kuril and Aleutian Islands. Some species (northern rockfish Sebastes polyspinis, dusky rockfish Sebastes ciliatus, arrowtooth flounder, and rex sole) extended their ranges from the Aleutians to southwestern Kamchatka and Kuril Islands due to recent climatic changes that may be considered as result of climatic changes, which affect reorganization of fish communities and therefore may be used for fisheries management. Introduction It is a number of evidences of climatic regime changes, which impact composition of fish communities and leads to their reorganization that can be used for forecast of abundance trends of commercially important species and therefore may be useful in management of exploitable fish populations. One of evidence of regime shift is considerable range extension of groundfish species. In the past, common understanding was that ichthyofauna exchange between Asia and America took place in both directions along Bering Sea continental slope and Aleutian-Kuril arch (Kodolov et al., 1991). At present, only continental slope of the Bering Sea is considered to be the main way of intrusion of typical American fish species (Pacific halibut Hippoglossus stenolepis,

175

arrowtooth flounder Atheresthes evermanni, sablefish Anoplopoma fimbria, rex sole Glyptocephalus zachirus) into the Asian waters (Novikov, 1961; Orlov, 2001) (Figure 1). In addition, Wilimovsky (1964) hypothesized possibility of westward range extension of some Aleutian fishes although there was no suggestion how far this extension may be. Furthermore, it has been recently suggested (Dudnik et al., 1998) that juvenile sablefish is able to migrate actively from the Aleutians into the Kuril Island and Kamchatka waters. Studies conducted during 1992-2002 in the western Bering Sea and the Pacific waters off the northern Kuril Islands and southeastern Kamchatka have provided new data on intrusion of some representatives of American ichthyofauna from Aleutians to Asian waters. Material and Methods Catch data were obtained from 21 bottom trawl surveys (total over 1,500 bottom trawl stations made during 1992-2002) and numerous bottom trawl hauls conducted during commercial operations in the Pacific Ocean off the Southeast Kamchatka and North Kuril Islands. The total investigated area was within 4730 N to 52 N and 15420 E to 15850 E. Samples were collected from three chartered commercial Japanese trawlers (Tomi-Maru 53, Tomi-Maru 82, and Tora-Maru 58). Catch data were also sampled aboard Japanese trawler Kayo-Maru 28 in the western Bering Sea (168 E 178 W) during bottom trawl survey and commercial fishing operations in summer 1997. Bottom trawl surveys were conducted during the daytime, commercial fishing operations were conducted around the clock at depths 76-833 m using bottom trawl net constructed from 100-120 mm (stretched mesh) polyethylene net with horizontal and vertical trawl mouth openings of 25-30 m by 5-7 m, respectively. Capture data on species were also taken from published sources.

176

65 60 55

SIBERIA

BERING SEA
KA M CH AT KA

ALASKA

SEA OF 50 OKHOTSK

Aleutian Islands

PACIFIC

OCEAN

45

Ku

ri l

s nd sla

150

160

170

180

7 190

200 16

Figure 1. Scheme of distribution of sablefish, Pacific halibut, arrowtooth flounder, and rex sole from eastern to western Bering Sea via active migrations and range extension of adults or via transfer of pelagic fry by currents.

Result and Discussion A possibility of distribution of American fishes into the Asian waters by means of range extensions was identified during second half of 1990s. It was noted that abundance of arrowtooth flounder in the Pacific waters off the northern Kurils and southeastern Kamchatka had sharply increased since 1997 (Figure 2). In 1994-1995, only incidental captures of large fish were observed in this area. However, in 1997-1998, catches of arrowtooth flounder in the study area dramatically increased. During this time mostly juveniles represented fish in catches. In the southeastern Sea of Okhotsk, arrowtooth flounder was found only once in 1996, its maximum abundance was recorded in 1997 and subsequently its catches substantially decreased (Chetvergov, 2001). The possible reason of arrowtooth flounder penetrating into the Kuril Islands and Kamchatka waters may be associated with range extension of the species from the Bering Sea along the continental slope. Studies conducted in the western Bering Sea during 1990s (Orlov, 2000) indicated that main arrowtooth flounder aggregations within the Russian EEZ were located to the east of 180. Fish caught in this area

177

was represented exclusively by large fish with length of 43 to 72 cm (mean length 54.8 cm). It means that expansion of fish from the Bering Sea cannot be taken as a cause of observed increase in the arrowtooth flounder abundance off Kurils and Kamchatka. Another cause of increase in the abundance of arrowtooth flounder in the study area may be associated with transfer by currents of pelagic flounder yearlings from adjacent regions. Dolganov (2000) hypothesized that currents may transport eggs of this species very far from reproductive areas. This hypothesis looks doubtful because arrowtooth flounder juveniles switch to the bottom way of life at length of 40-43 mm (Bouwens et al., 1999a). Catches from the Pacific waters off the Kurils and Kamchatka did not contain such individuals. Minimal length of arrowtooth flounder was 28 cm that correspond to age 3+ (Bouwens et al., 1999b). When compared by number of caudal vertebrae and gill rakers, arrowtooth flounder caught off Kuril Islands and Kamchatka is very similar to fish from the northeastern Pacific Ocean (Mukhametov and Orlov, 2002). Comparison of fish size composition from Kurils and Aleutians waters showed its similarity in both areas (Orlov, 2000). Taking in account all the facts mentioned above, the reason of observed significant increase in arrowtooth flounder abundance off Kamchatka and Kuril Islands is related to westward range extension of the species range from the Aleutians.

178

60

55

SEA OF OKHOTSK

KAMCHATKA

50
ri Ku sl lI

PACIFIC OCEAN
ds an

45 150 155 160 165 Figure 2. Map of capture localities of arrowtooth flounder (asterisks) off the Kurils and Kamchatka.

In the second half of 1990s considerable number of northern rockfish Sebastes polyspinis were observed in catches (Figure 3). This species was not previously recorded in the study area. In 1990s, northern rockfish was not found in the western Bering Sea as well. Size composition of this species in the Pacific waters off Kamchatka and Kurils, same as in case of arrowtooth flounder, was very similar to size composition of Aleutian fish (Orlov, 2000).

179

52

SEA OF OKHOTSK

KAMCHATKA

51

50

PACIFIC OCEAN

49

48 154 155 156 157 158

Figure 3. Map of capture localities of northern rockfish (asterisks) off the Kurils and Kamchatka. Since 1993 dusky rockfish Sebastes ciliatus started to appear occasionally in catches as a by-catch from Kamchatka and Kuril Islands waters (Figure 4). In 1993 and 1996 it was recorded in catches taken off Commander Islands; in 1997 it was found in catches taken off the southern tip of Kamchatka (Sheiko and Tranbenkova, 1998), and later it continued to be recorded within the entire study area from 4816 N to 5125 N (Biryukov, Nemchinov, Tokranov, Zolotov, unpubl. data).

180

65

SIBERIA
60
AT K A

BERING SEA

SEA 55 OF OKHOTSK
50

KA M

CH

Aleu

tian I

slan ds

45 150

PACIFIC
160

OCEAN
170

Figure 4. Map of capture localities of dusky rockfish (asterisks) in the northwestern Pacific. Recently, catches of typically eastern Pacific flatfish, rex sole Glyptocephalus zachirus, in the Pacific waters off the northern Kuril Islands and south-east Kamchatka have became more frequent (Tokranov and Vinnikov, 2000; Orlov et al., 2002) (Figure 5). There are several possible hypotheses, which may explain recent records of this species in study area. It appears that rex sole has a continuous distribution from

181

65

SIBERIA

60
SEA OF OKHOTSK
ds an sl lI

KA

55

CH

AT K

50
ri Ku

PACIFIC OCEAN

45 150

155

160

165

170

Figure 5. Map of capture localities of rex sole (asterisks) off the Kurils and Kamchatka.

California to the north of Kuril Islands. The absence of species records before 1980s and in 1990s may be due to the fact that the inshore fish fauna of the region has been poorly studied. Because rex sole has pelagic larvae and juveniles (Matarese et al., 1989), they may be carried from the Aleutian Islands or Bering Sea to the Kuril Islands by currents. Bottom settlement of rex sole may occur at length 49-72 mm (Alstrom et al., 1984). Rex sole caught in study area were 20-40 cm long, which corresponds to an age of 3-12 years (Novikov, 1974). Therefore, it may be possible to suggest that after bottom settlement the fish remain in the area for several years, although no individuals less than 20 cm long had ever been caught. For this reason, we have doubts that specimens reported from this area originated from pelagic young fish carried out by currents into this area. We think that rex sole found in the Pacific waters off the northern Kuril Islands and southeastern Kamchatka may have originated from the Aleutian Islands or Bering Sea due to considerable westward range extension.

182

An illustration of range extensions for a number of Aleutian fish species westward to the Kuril Islands is presented in Figure 6. We suggest that the main reason of range extensions considered are significant warming of the northwestern Pacific Ocean in late 1990s and considerable weakening of the East Kamchatka current (Khen, 1997; Hare and Mantua, 2000).
N

65 60 55 50 45
E
SEA OF OKHOTSK
r il Ku d an Isl s

SIBERIA

M CH

AT K

BERING SEA
Aleutian Islands

KA

PACIFIC

OCEAN

150

160

170

180

7 190

Figure 6. Hypothetic scheme of range extension of dusky and northern rockfishes, rex sole, and arrowtooth flounder from the Aleutians to Kurils.

Acknowledgements I would like to thank my friends and colleagues O.A. Nemchinov, I.A. Biryukov, A.M. Tokranov, O.G. Zolotov for capture data provided. I greatly appreciate the assistance of Dr. E.N. Sabourenkov (Commission for the Conservation of Antarctic Marine Living Resources, Hobart, Tasmania, Australia) who significantly improved manuscript.

183

References Ahlstrom, E.H., Amaoka, K., Hensley, H.G. et al. (1984). Pleuronectiformes: development. P. 640-669 in: Onthogeny and systematics of fishes. Amer. Soc. Ichthyol. Herpetol. Spec. Publ. Allen Press, Lawrence. Bouwens, K.A., Paul, A.J., and Smith R.L. (1999a). Growth of juvenile arrowtooth flounders from Kachemak Bay, Alaska. Alaska Fish. Res. Bull., 6(1): 35-40. Bouwens, K.A., Smith, R.L., Paul, A.J., and Rugen, W. (1999b). Length at and timing of hatching and settlement for arrowtooth flounders in the Gulf of Alaska. Alaska Fish. Res. Bull., 5(1): 41-48. Chetvergov, A.V. (2001). On occurrence of the arrowtooth flounder Atheresthes stomias (Jordan & Gilbert) in the eastern Okhotsk Sea. P. 106-108 in: Conservation of biodiversity of Kamchtaka and coastal waters. Mat. II Sci. Conf. Petropavlovsk-Kamchatsky, April 9-10, 2001. PetropavlovskKamchatsky, Kamshat (In Russian). Dolganov, V.N. (2000). Spawning of arrowtooth flounder Atheresthes stomias in the northwestern Bering Sea. Vopr. Ikhthyologii, 40(3): 411-412 (In Russian). Dudnik, Yu.I., Kodolov, L.S., and Polutov V.I. 1998. On distribution and reproduction of sablefish Anoplopoma fimbria off the Kuril Islands and Kamchatka. Vopr. Ikhtiologii, 38(1): 16-21 (In Russian). Hare, S.R. and Mantua, N.J. (2000). Empirical evidence for North Pacific regime shifts in 1977 and 1989. Progr. Oceanogr., 47(2-4): 103- 145. Khen, G.V. (1997). Main regularities of multi-year changes in ice cover of the Bering Sea. P. 64-67 in: Complex studies of ecosystem of the Sea of Okhotsk. VNIRO Publ., Moscow (In Russian). Kodolov, L.S., Kulikov, M.Yu., and Syusina, T.I. (1991). Distributional patterns of continental slope and seamount fishes of the North Pacific. P. 21-38 in: Biology of fishes and invertebrates of the North Pacific Ocean. Far East State University Publishing, Vladivostok (In Russian).

184

Matarese, A.C., Kendall, A.W., Jr., Blood, D.M. and Vinter B.M. (1989). guide to early life history stages of northeast Pacific fishes. U.S. Department of Commerce, NOAA Technical Report NMFS, 80: 1-652. Mukhametov, I.N. and Orlov A.M. (2002). The morphology of arrow- toothed flounders of the genus Atheresthes of Pacific waters of the northern Kuril Islands and southeastern Kamchatka. Biologiya Morya, 28 (3): 196-202 (In Russian). Novikov, N.P. (1961). New data on the distribution of halibuts and some other commercial fishes in the Bering Sea. Zool. Zhurnal, 40 (10): 15101515 (In Russian). Novikov, N.P. (1974). Commercial fishes of the northern Pacific Ocean continental slope. Pishchevaya Promyshlennost, Moscow. 308 pp. (In Russian). Orlov, A.M. (2000). Representatives of Oregonian ichthyofauna off the Asian coasts. P. 187-214 in: Commercial and biological studies of fishes in the Pacific waters of the Kuril Islands and adjacent areas of the Sea of Okhotsk and Bering Seas. VNIRO Publ., Moscow (In Russian). Orlov, A.M. (2001). Features of spatial and vertical distribution of representatives of the oregonian ichthyofauna off the Asian coasts. Bull. Mosk. Ob-va Ispyt. Prirody. Otd. Biol., 106(4): 23-37 (In Russian). Orlov, A.M., Tokranov, A.M. and Biryukov I.A. (2002). New records of rex sole Glyptocephalus zachirus Lockington, 1879 (Teleostei: Pleuronectidae) from the north-western Pacific. Aqua, J. Ichthyol. Aquat. Biol., 5(3): 89-98. Sheiko, B.A. and Tranbenkova A.G. (1998). New for Russian fauna and rare marine fishes from Kamchatka, Kuril and Commander Islands. P. 62-63 in: Actual Problems of Fish Taxonomy. Abs. Int. Conf., St. Petersburg, November 17-19, 1998. Zool. Inst., St. Petersburg (In Russian). Tokranov, A.M. and Vinnikov A.V. (2000). Capture of rex sole Glyptocephalus zachirus Lockington (Pleuronectidae) in waters of the southeastern Kamchatka. Vopr. Ikhtiologii, 40(3): 397-398 (In Russian).

185

Wilimovsky, N.J. (1964). Inshore fish fauna of the Aleutian Archipelago. Proc. 14th Alaskan Sci. Conf., 14: 172-190.

186

QUANTITATIVE CHANGES OF BOTTOM TRAWL CATCH COMPOSITIONS IN THE PACIFIC WATERS OFF THE NORTHERN KURIL ISLANDS AND SOUTHEASTERN KAMCHATKA DURING PAST DECADES1 Alexei M. Orlov Russian Federal Research Institute of Fisheries & Oceanography 17, V. Krasnoselskaya, Moscow, 107140, Russia (095) 264-91-43 / (095) 264-91-87 orlov@vniro.ru Abstract Quantitative analysis of multi-annual data on bottom trawl catch compositions may be useful for understanding of present state and tendencies of changes occurred in groundfish communities as a result of fishery and environmental impact. Based on bottom trawl surveys conducted in 1993-98s and similar published data of 1960-80s changes of rates and compositions of bottom trawl catches in the Pacific waters off the northern Kuril Islands and southeastern Kamchatka are analyzed. In the last decade groundfish catch rate off the southeastern Kamchatka was quite similar with that of 1970s but species composition principally differed. Catch rates of Pacific cod Gadus macrocephalus, Kamchatka flounder Atheresthes evermanni, Pacific halibut Hippoglossus stenolepis, flathead sole 1 The paper was presented at the International Symposium Quantitative Ecosystem Indicators for Fisheries Management, 31 March 3 April 2004, Paris, France 187

Hippoglossoides elassodon, northern rock sole Lepidopsetta polyxystra, and skates (Rajidae) were comparable with that of 1970s. In the same time, during the whole period considered catches of Greenland halibut Reinhardtius hippoglossoides matsuurae dramatically decreased. Opposite, catch rates and proportions of eelpouts (Zoarcidae) and snailfishes (Liparidae) considerably increased. Maximum rockfish catches (mostly Pacific ocean perch Sebastes alutus) off Eastern Kamchatka were observed in the late 1960s early 1970s with further considerable decreasing due to intensive unregulated fishery. Present total catch rate of groundfishes off the northern Kuril Islands is close to that in 1960s but its composition during both periods considerably differed. During time frame analyzed catches of Greenland halibut became ten times less. Present catch rate of northern rock sole was the least during whole investigated period. During recent years Pacific cod catches significantly decreased, which rate was quite similar with that of 1960s. Rate of catches of Atka mackerel Pleurogrammus monopterygius was almost the same as in 1975-79s but considerably less then in early 1970s, when its maximum abundance occurred. Permanently low abundance during long period is characteristic of flathead sole, which proportion in bottom trawl catches in the second half of 1960s was one fifth. Present catches of Kamchatka flounder became maximal during the whole period of study. In comparison with 1970s current rockfish catches essentially increased though its rate is very far (almost ten times lesser) from that of 1960s. Catches of snailfishes and skates significantly increased (two-three times) and its rates became maximal during whole period considered. Introduction Rational use of fish resources and correct estimation of fish productivity of exploitable ecosystems are impossible without authentic view on composition and structure of communities. Besides, such data may serve as indicators of the state of exploitable ecosystems, which qualitative and quantitative composition fluctuates caused by various reasons including anthropogenic impact (Fadeev, 1971; Blagoderov et al., 1982; Batytskaya, 1984; Blagoderov and Kolesova, 1985). Multi-annual changes of rates and compositions of bottom trawl catches on the continental slope of the Russian Far East seas are poorly studied. There are some publications available (Ilinsky, 1990a,b; 1991) described such changes within the upper bathyal (200 400-500 m) based on bottom trawl surveys conducted in 1960-80s. Comparison of current composition of groundfish catches within the upper bathyal in the northwestern Pacific with that

188

of past decades and analysis of its multi-annual dynamics were not performed till present. The main purpose of the paper is comparative analysis of bottom catch rates and composition within the upper bathyal (200-500 m) of the Pacific waters off the northern Kuril Islands and southeastern Kamchatka based on results of bottom trawl surveys conducted in 1960- 90s in relation to management considerations. Material and Methods Results of bottom trawl surveys on the upper part of continental slope conducted aboard chartered Japanese trawlers during summer-autumn 1993-1998 were used. The survey area was the Pacific waters off the northern Kuril Islands and southeastern Kamchatka between 4750 and 5210N within depth range of 200-500 m. Total number of hauls was 734. Horizontal and vertical openings of trawl were 25-30 and 5-7 m respectively, mesh size was 60-100 mm. Bottom trawl surveys were conducted mostly in summer (occasionally in late spring) and autumn, where the majority of fishes do not perform considerable migrations within study area and is distributed rather uniformly. Bottom trawl stations were allocated within surveyed area evenly. To prevent influence of vertical diurnal migrations on catch rate and composition, all hauls were conducted during light time, when the majority of groundfishes is on the ground. Hauling speed was about 3 knots, duration of hauls varied 20 minutes to one hour. Subsequently all catches were recalculated to standard fishing effort, i.e. catch per hour trawling. All catches were sorted by species and all individuals were counted and weighed. Data on rates and compositions of groundfish catches of winter-spring bottom trawl surveys in 1960-80s are taken from the paper of Ilinsky (1991). Our data are obtained mostly in summer-autumn period that makes comparison performed not quite correct due to different pattern of seasonal distribution of some species. Nevertheless, due to lack of other data such comparison may be helpful to understand present state and trends of changes occurred in communities considered and also may serve as the basis to develop recommendations on its rational exploitation.

189

Results and Discussion During past decade groundfish catch rate off southeastern Kamchatka was similar to that of 1970s but proportion of species or species groups was principally different (Figure 1). Catch rates of Pacific cod, Kamchatka flounder, Pacific halibut, northern rock sole and skates were comparable with that of 1970s (Figure 3). In the same time, during the whole study period catches of Greenland halibut, sablefish and greenlings decreased that is probably associated with natural fluctuations of abundance of these species. Catch rates and proportions of eelpouts and snailfishes significantly increased that also is possibly related to natural causes given similar changes of catch compositions in other areas of the North Pacific (Hoff, 1998). Considerable increasing of catches of grenadiers may be partly associated with different seasons of survey conducted in 1960-80s and 1990s because shift of main aggregations of grenadiers from greater depths to shallower waters in summer period occur (Tuponogov, 1997). Moreover, it is possible to suggest changes of distributional patterns of schoolings of grenadiers under influence of recent climate regime shift similar event is occurred recently in the North Atlantic (Gerber et al., 2004). Catches of rockfishes (mostly of Pacific Ocean perch Sebastes alutus) off the southeastern Kamchatka were maximum in the second half of 1960s first half of 1970s with subsequent considerable decline due to intensive unregulated fishery (Snytko, 1986). In 1990s rockfish catches slightly increased in comparison with previous period probably due to long lack of targeted fishery. However they are still three times less than in the period of maximum catches.

190

100% 80% Proportion, % 60% 40% 20% 0% 1965-69 Pacific cod Pacific halibut Other flatfishes Eelpouts Atka mackerel 1970-74 1975-79 Years Greenland halibut Flathead sole Snailfishes Grenadiers Greenlings 1985-87 1993-98

Arrowtooth flounders Northern rock sole Skates Sablefish Rockfishes

Figure 1. Multi-annual changes of compositions of bottom trawl catches within the upper bathyal off the southeastern Kamchatka. In the upper bathyal of the northern Kuril Islands total catch rate was similar to that of the second half of 1960s but catch compositions of both periods compared differed considerably (Figure 2). During period analyzed catch rates of Greenland halibut reduced almost ten times (Figure 3). Analogous trend of catch decline of this species off the southeastern Kamchtaka may indicate the same cause of event observed. Recent catch rate of northern rock sole is the least during the whole study period. The reason of that is probably natural abundance decline because the fishery of above species was not intensive. Catches of Pacific cod were recently reduced significantly reaching the level of 1960s. Catch rates of Atka mackerel were almost the same with that of 1975-79s

191

100% 80% Proportion, % 60% 40% 20% 0% 1960-64 Pacific cod Pacific halibut Other flatfishes Grenadiers 1965-69 1970-74 1975-79 1980-84 1993-98 Years Greenland halibut Arrowtooth flounders Flathead sole Northern rock sole Snailfishes Skates Greenlings Rockfishes

Figure 2. Multi-annual changes of compositions of bottom trawl catches within the upper bathyal off the northern Kuril Islands. but considerably less than in the early 1970s, when its maximal abundance occurred (Zolotov, 1984). Abundance increasing of this species in 1990s in many areas of its range (Yang, 1999) will probably result in continuous rise of its catches in the main fishery grounds including the Pacific waters off Kamchatka and Kuril Islands. Increasing of catches of grenadiers is associated with reasons similar to that of southeastern Kamchatka coast. Catches of flathead sole remain low during the long period. This species comprised one fifth of the total groundfish catch in the second half of 1960s. Increasing number of its juveniles during some recent years allows suggesting rise of abundance of this sole in the nearest future. Kamchatka flounder catches became highest during entire study period. It is possible to hypothesize that abundance of Greenland halibut and Kamchatka flounder fluctuate in antiphase. Rockfish catches considerably increased in comparison with 1970s but their rate is far (almost ten times less) from catches of 1960s. Their reduction was caused by intensive unregulated fishery (Snytko, 1986). Some increasing of abundance of rockfishes in the study area was probably associated with the lack of specialized fishery during long period. Though their low reproduction rates do not allow rapid recovery of stocks. Catches of snailfishes and skates increased significantly (2-3 times). Their rates were highest during the whole study period.

192

Conclusion Despite rather high catch rates of groundfishes current commercial importance of the upper bathyal is fairly limited because commercially important species (cods, rockfishes, greenlings, and flounders) comprise only about quarter of total catch. Abundance decline of the number of most commercially important species in 1990s and respective increasing of some presently unexploited fish stocks resulted in significant decreasing of importance of study area for fishery in comparison with that of 1960s, when their commercial exploitation was begun. Similar species composition of catches will probably remain in the nearest future that requires involvement in fishery of prospective species (grenadiers, skates, eelpouts, large sculpins) and development of their processing technologies.

193

Eelpouts Zoarcidae gen. sp., SE Kamchatka

12 3

Catch
10

Proportion
2,5 2

Catch, kg per h

1,5

4 2

1 0,5

0 1965-69 1970-74 1975-79 1985-87 1993-98

Years

fimbria , Sablefish Anoplopoma fimbria, SE Kamchatka

25 4 3,5 20 3 2,5 2 10 1,5 1 0,5 0 1965-69 0

Catch, kg per h

15

1970-74

1975-79

1985-87

1993-98

Years

monopterygius, Atka mackerel Pleurogrammus monopterygius , SE Kamchatka

12 10 3 2,5 2

Catch, kg per h

6 4 2 0 1965-69 1970-74 1975-79

1985-87 1993-98

1,5 1 0,5 0

Years

194

Proportion, %

Proportion, %

Proportion, %

P a c ific c o d G a d u s m a c ro c e p h a lu s , S E K a m c h a tk a
600 600 500 500 80 80 70 70 60 60 50 50 40 40 30 30 20 20 100 100 0 0 1 9 6 5 -6 9 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 7 5 -7 9 1 9 8 5 -8 7 1 9 8 5 -8 7 1 9 9 3 -9 8 1 9 9 3 -9 8 10 10 0 0
16 16 14 14

P a c ific h a lib u t H ip p o g lo s s u s s te n o le p is S E K a m c h a tk a P a c ific h a lib u t H ip p o g lo s s u s s te n o le p is ,, S E K a m c h a tk a

4 4 4 3 ,5 3 ,5 3 ,5 3 3 3 2 ,5 2 ,5 2 ,5 2 2 2 1 ,5 1 ,5 1 ,5 1 1 1 0 ,5 0 ,5 0 ,5 0 0 0

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

10 10 8 8 6 6 4 4 2 2 0 0

300 300 200 200

Years

1 9 6 5 -6 9 1 9 6 5 -6 9

1 9 7 0 -7 4 1 9 7 0 -7 4

1 9 7 5 -7 9 1 9 7 5 -7 9

1 9 8 5 -8 7 1 9 8 5 -8 7 1 9 8 5 -8 7

1 9 9 3 -9 8 1 9 9 3 -9 8 1 9 9 3 -9 8

Y ears Y ears

P a c ific c o d G a d u s m a c ro c e p h a lu s , N K u ril Is l.
450 450 400 400 350 350 70 70 60 60
18 18 16 16 14 14

P a c ific h a lib u t H ip p o g lo s s u s s te n o le p is N K u ril Is l. P a c ific h a lib u t H ip p o g lo s s u s s te n o le p is ,, N K u ril Is l.

1 ,6 1 ,6 1 ,6 1 ,4 1 ,4 1 ,4

Catch, kg per h Catch, kg per h

Proportion, % Proportion, %

12 12 10 10 8 8 6 6 4 4 2 2 0 0 1 9 6 0 -6 4 1 9 6 0 -6 4 1 9 6 5 -6 9 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 7 5 -7 9 1 9 8 0 -8 4 1 9 8 0 -8 4 1 9 8 0 -8 4 1 9 9 3 -9 8 1 9 9 3 -9 8 1 9 9 3 -9 8

250 250 200 200 150 150 100 100 50 50 0 0 1 9 6 0 -6 4 1 9 6 0 -6 4 1 9 6 5 -6 9 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 7 5 -7 9 1 9 8 0 -8 4 1 9 8 0 -8 4 1 9 9 3 -9 8 1 9 9 3 -9 8

40 40 30 30 20 20 10 10 0 0

1 1 1 0 ,8 0 ,8 0 ,8 0 ,6 0 ,6 0 ,6 0 ,4 0 ,4 0 ,4 0 ,2 0 ,2 0 ,2 0 0 0

Years

Y ears Y ears

Green lan d h alib u t Reinh ard tiu s hip p og losso ides , SE Kam ch atka
30 30 25 25 7 7 6 6
50 50 50 45 45 45 40 40 40

Fla thea d s ole Hippoglos soides ela ssodon ,, SE Kam chatka ela ssodon SE Kam chatka
8 8 8 7 7 7 6 6 6 5 5 5 4 4 4 3 3 3 2 2 2 1 1 1 0 0 0 1965-69 1965-69 1965-69 1970-74 1970-74 1970-74 1975-79 1975-79 1975-79 1985-87 1985-87 1985-87 1993-98 1993-98 1993-98

Catch, kg per h Catch, kg per h

4 4 15 15 3 3 10 10 5 5
0 1965-69 1970-74 1975-79 1985-87 1993-98

30 30 30 25 25 25 20 20 20 15 15 15 10 10 10 5 5 5 0 0 0

2 2
1 0

Yea Y e a rs

Ye ars Ye ars

Green lan d h alib ut Reinh ard tiu s hip po glo ssoid es , N Kuril Isl.
45
40 35

Flathe ad sole Hippogloss oide s elass odon ,, N Kuril Is l. s elass odon N Kuril Is l.
7
6
100 100 120 120

10 10 10 9 9 9 8 8 8

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

25
20 15 10 5 0 1960-64 1965-69 1970-74 1975-79 1980-84 1993-98

4 3 2 1 0

80 80 60 60

6 6 6 5 5 5 4 4 4 3 3 3 2 2 2 1 1 1 0 0 0
1960-64 1960-64 1965-69 1965-69 1970-74 1970-74

40 40
20 20 0 0

Yea Y e a rs

1975-79 1975-79 1975-79

1980-84 1980-84 1980-84

1993-98 1993-98 1993-98

Arro Ath A rro wto o th flo u n d ers , A th e resth es sp p ., S E K am ch atka

8 7 6 1,8 1,6 1,4
300 300 250 250

Northern rock sole Lepidopsetta polyxystra ,, SE Kam chatka Lepidopsetta polyxystra SE Kam chatka
60 60 60 50 50 50 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 1965-69 1965-69 1970-74 1970-74 1975-79 1975-79 1975-79 1985-87 1985-87 1985-87 1993-98 1993-98 1993-98

Y e a rs Y e a rs

Catch, kg per h Catch, kg per h

Proportion, % Proportion, %

5 4 3 2 1 0

200 200 150 150 100 100 50 50 0 0

1 0,8 0,6 0,4

0,2
0

1 96 5 -69

1 97 0-7 4

1 97 5-7 9

1 98 5-8 7

1 99 3-9 8

Y e ar s ars

Y ears Y ears

Arro Ku A rro wto o th flo u n d ers A th ere sth e s s p p ., N K u ril Is l.

10 9 8 2 2 ,5
350 350 300 300

Northern rock sole Lepidopsetta polyxystra ,, N Kuril Isl. Lepidopsetta polyxystra N Kuril Isl.
35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0
1960-64 1960-64 1965-69 1965-69 1970-74 1970-74

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

6 5

1 ,5

200 200 150 150 100 100 50 50 0 0

4
3 2 1 0 1 96 0-64 1 96 5-6 9 19 70 -7 4 19 75 -79 1 98 0-84 1 99 3-9 8

0 ,5

1975-79 1975-79

1980-84 1980-84

1993-98 1993-98

rs Y ea r s

Years Years

195

Proportion, % Proportion, %

250 250

Proportion, % Proportion, %

Proportion, % Proportion, %

1,2

Catch, kg per h Catch, kg per h

Proportion, % Proportion, %

30

Proportion, % Proportion, %

7 7 7

Proportion, % Proportion, %

20 20

Catch, kg per h Catch, kg per h

Proportion, % Proportion, %

5 5

35 35 35

Proportion, % Proportion, %

300 300

Catch, kg per h Catch, kg per h

50 50

1 ,2 1 ,2 1 ,2

Proportion, % Proportion, %

400 400

Proportion, % Proportion, %

12 12

O th e r fla tfis h e s , P le u r o n e c tid a e g e n . s p ., S E K a m c h a tk a O t h e r f la t f is h e s , P le u r o n e c t id a e g e n . s p ., S E K a m c h a t k a

30 30 25 25 6 6 5 5
25 25

G re nad ie rs M ac ro urid ae g e n. s p ., S E K am c hatka G re nad ie rs M ac ro urid ae g e n. s p ., S E K am c hatka

4 ,5 4 ,5 4 4 20 20 3 ,5 3 ,5

Catch, kg per h Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

Proportion, % Proportion, % Proportion, %

20 20 15 15 10 10 5 5 0 0 1 9 6 5 -6 9 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 7 5 -7 9 1 9 8 5 -8 7 1 9 8 5 -8 7 1 9 9 3 -9 8 1 9 9 3 -9 8

4 4 3 3 2 2 1 1 0 0

15 15

2 ,5 2 ,5 2 2 1 ,5 1 ,5 1 1 0 ,5 0 ,5

10 10

5 5

0 0 1 9 6 5 -6 9 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 7 5 -7 9 1 9 8 5 -8 7 1 9 8 5 -8 7 1 9 9 3 -9 8 1 9 9 3 -9 8

0 0

Y ears Y ears

Y e ars Y e ars
G r e n a d ie r s M a c r o u r id a e g e n . s p . G r e n a d ie r s M a c r o u r id a e g e n . s p .
0 ,9 0 ,9 0 ,8 0 ,8 0 ,7 0 ,7
16 16 14 14 12 12 3 ,5 3 ,5 3 3 2 ,5 2 ,5 2 2 1 ,5 1 ,5 1 1 0 ,5 0 ,5 0 0 1 9 6 0 -6 4 1 9 6 0 -6 4 1 9 6 5 -6 9 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 7 5 -7 9 1 9 8 0 -8 4 1 9 8 0 -8 4 1 9 9 3 -9 8 1 9 9 3 -9 8

O th e r fla tfis h e s , P le u r o n e c tid a e g e n . s p ., N K u r il Is l. th f la t f is t id

3 ,5 3 ,5 3 3 2 ,5 2 ,5 2 2 1 ,5 1 ,5 1 1 0 ,5 0 ,5 0 0 1 9 6 0 -6 4 1 9 6 0 -6 4 1 9 6 5 -6 9 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 7 5 -7 9 1 9 8 0 -8 4 1 9 8 0 -8 4 1 9 9 3 -9 8 1 9 9 3 -9 8

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

0 ,5 0 ,5 0 ,4 0 ,4 0 ,3 0 ,3 0 ,2 0 ,2 0 ,1 0 ,1 0 0

Proportion, % Proportion, %

0 ,6 0 ,6

10 10 8 8 6 6 4 4 2 2 0 0

Y ears

Y e ars Y e ars

S n a ilfis h e s L ip a rid a e g e n . s p ., S E K a m c h a tk a
18 0 18 0 16 0 16 0 14 0 14 0 40 40 35 35
18 18 16 16 14 14

Greenlings Hexagram m os spp., SE Kamchatka Greenlings Hexagram m os spp., SE Kamchatka

3 3 2,5 2,5

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

25 25 20 20 15 15 10 10 5 5 0 1 96 5-6 9 1 97 0-7 4 1 97 5-7 9 19 85 -8 7 19 9 3-9 8

Proportion, % Proportion, %

12 0 12 0 10 0 10 0 80 80 60 60 40 40 20 0

10 10 8 8 6 6 4 4 2 2 0 0 1965-69 1965-69 1970-74 1970-74 1975-79 1975-79 1985-87 1985-87 1993-98 1993-98

1,5 1,5 1 1 0,5 0,5 0 0

Years

Years Years

S n a ilfis h e s L ip a rid a e g e n . s p ., N K u ril Is l.

300 300

Greenlings Hexagrammidae gen. sp., N Kuril Isl.

35 35 30 30 25 25 20 20 150 150 15 15 100 100 50 50 0 0 1960-64 1960-64 1965-69 1965-69 1970-74 1970-74 1975-79 1975-79 1980-84 1980-84 1993-98 1993-98 10 10 5 5 0 0

70 60

16 14
250 250

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

40 30 20 10 0 1 96 0-6 4 19 6 5-6 9 19 70 -74 1 97 5 -79 1 98 0-8 4 19 93 -9 8

10 8 6 4 2 0

Proportion, % Proportion, %

50

Years

Years

S k a te s R a jid a e g e n . s p ., S E K a m c h a tk a
120 100 25

Rockfishes Sebastidae gen. sp., SE Kam chatka

30 30 25 25
20

8 8 7 7 6 6 5 5 4 4 3 3 2 2

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

15 60 10 40 20 0 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 5 -7 9 1 9 8 5 -8 7 1 9 9 3 -9 8 5

Proportion, % Proportion, %

80

15 15 10 10 5 0 1965-69 1970-74 1975-79 1985-87 1993-98

1 0

Y ears

Years

S k a te s R a jid a e g e n . s p ., N K u ril Is l.
140 120 100 80 15 60 40 20 0 1 9 6 0 -6 4 1 9 6 5 -6 9 1 9 7 0 -7 4 1 9 7 5 -7 9 10 5 0 30 25

Rockfishes Sebastidae gen. sp., N Kuril Isl.

800 700 600 90 80 70

Catch, kg per h Catch, kg per h

Catch, kg per h Catch, kg per h

Proportion, % Proportion, %

20

500 400 300 200 100 0 1960-64 1965-69 1970-74 1975-79 1980-84 1993-98

50 40 30 20 10 0

1 9 8 0 -8 4

1 9 9 3 -9 8

Y ears

Years

Figure 3. Multi-annual changes of catch rates and compositions of groundfishes within upper bathyal of study areas.

196

Proportion, % Proportion, %

60

Proportion, % Proportion, %

20 20

Proportion, % Proportion, %

12

200 200

Proportion, % Proportion, %

30 30

12 12

2 2

Proportion, % Proportion, %

Proportion, % Proportion, %

3 3

References Batytskaya, L.V. 1984. Changes of composition of bottom fish communities on the eastern Sakhalin shelf under conditions of intensive fishery // Biol. Morya 2: 45-53 (In Russian). Blagoderov, A.I., Kolesova, N.G. 1985. Qualitative and quantitative changes of bottom fish composition on the shelf of western coast of Kamchatka // Vopr. Ikhtiologii 25(4): 590-596 (In Russian). Blagoderov, A.I., Zadorina, L.G., Kolesova, N.G. 1982. Effect of fishery on the structure of bottom fish communities of the western Kamchatka shelf // Rybn. Khoz. 4: 45-47 (In Russian). Fadeev N.S. 1971. Changes of composition of bottom ichthyofauna during trawl fishery // Zool. Jurn. 50(4): 532-536 (In Russian). Gerber, Ye.M., Biryukin, S.N., Zimin, A.B. et al. 2004. Russian fishery researches in the Mid-Atlantic ridge area in 2003 // Working document presented to the ICES Working Group on the Biology and Assessment of Deep-Sea Fisheries Resources. Copenhagen, Denmark, 18-24 February 2004. 17 p. Hoff, G.R. 1998. Life history aspects of noncommercial species in the eastern Bering Sea // Alaska Fish. Sci. Center Quart. Rep. July-August-September 1998. P. 19-21. Ilinsky, E.N. 1990a. Long-term changes of bottom fish catch composition on the continental slope of the Sea of Okhotsk and Sea of Japan // Biol. Morya 6: 12-18 (In Russian). Ilinsky, E.N. 1990b. Long-term changes of catch composition of dominant bottom fish on the continental slope of the Far East seas // Izv. TINRO 111: 67-78 (In Russian). Ilinsky, E.N. 1991. Long-term changes of bottom fish catch composition on the continental slope of the western Bering Sea, Pacific coast of Kamchatka and Kuril Islands // Vopr. Ikhtiologii 31(1): 73-81 (In Russian).

197

Snytko, V.A. 1986. Rockfishes // Biological resources of the Pacific Ocean. Moscow: Nauka, P. 281-310 (In Russian). Tuponogov, V.N. 1997. Seasonal migrations of giant grenadier Coryphaenoides pectoralis in the Sea of Okhotsk and adjacent waters // Biol. Morya 23(6): 362-369 (In Russian). Yang, M.-S. 1999. The trophic role of Atka mackerel, Pleurogrammus monopterygius, in the Aleutian Islands area // US Fish. Bull. 97(4): 10471057. Zolotov, O.G. 1984. Biology of Atka mackerel Pleurogrammus monopterygius (Pallas) in the waters off Kamchatka and Kuril Islands. Ph.D. Thesis. Moscow: VNIRO, 22 p. (In Russian).

198

TAXONOMIC DIVERSITY AND VERTICAL DISTRIBUTION OF ICHTHYOPLANKTON OFF CENTRAL AMERICA AND COSTA RICA DOME Andrei V. Suntsov Institute of Aquatic Resources of the Arctic, Russian Academy of Sciences Alexander Nevskii Pr. 50 Karelia, Petrozavodsk, 185067 Russia, asuntsov@mail.ru EXTENDED ABSTRACT ONLY DO NOT CITE In 1987 Russian R/V Ak. Mstislav Keldysh carried out multidisciplinary investigations in the Pacific ocean off central America. This contribution is based on the analysis of 91 plankton samples from six stations along transect running parallel to the coast (Fig. 1). Two stations were located near the southwestern periphery of the Costa Rica Dome (3498,3499), two in the BIOSTAT area (3500, 3501), and two in the area of well developed oxygen minimum layer (3503, 3504). Samples from two types of plankton nets covered discrete depth levels as well as upper 200 m in toto. In general, the available ichthyoplankon collection is represented by rather common species, due to relatively small water volume filtered with standard vertical tows. These, however, provide a good estimate for the most abundant and common species in the region sampled. In total, I identified larvae of 38-40 fish species and 8-10 types of fish eggs. The list below provides a glimpse into overall ichthyoplankton diversity (arranged taxonomically) encountered at this not so commonly sampled region.

199

Table 1. List of Ichthyoplankton Species Ophichthidae: Ophichthus zophochir; Congridae: Heteroconger canabus; gen. sp.; Nettastomatidae: Hoplunnis sicarius; Bathylagidae: B. nigrigenys; Gonostomatidae: Diplophos proximus; Phosichthyidae: Vinciguerria lucetia; Idiacanthidae: Idiacanthus antrostoma; Scopelarchidae: Scopelarchoides nicholsi; Paralepididae: gen. sp.; Myctophidae: Bolinichthys longipes; Diaphus pacificus, Diogenichthys laternatus, Gonichthys tenuiculus, Lampanyctus parvicauda, Myctophum aurolineatum Melanocetidae: Melanocetus sp.; Bregmacerotidae: Bregmaceros sp.; Hemiramphidae: Oxyporhamphus micropterus; Exocoetidae: Exocoetus sp.; Melamphaidae: Melamphaes lugubris, Scopelogadus mizolepis; Fistulariidae: Fistularia corneta; Scorpaenidae: Pontinus sp.; gen. sp.; Eleotridae: Dormitator latifrons; gen. sp.; Scombridae: Katsuvonus pelamis; Nomeidae: Cubiceps pauciradiatus, Psenes sio, Psenes sp. Bothidae: Bothus leopardinus, Bothus sp.; Paralichthyidae: Syacium ovale, Citharichthys platophrys, Citarichthys sp.; Cynoglossidae: Symphurus elongatus, Symphurus sp., gen. sp. The most abundant were larvae of Vinciuerria lucetia and Diogenichthys laternatus (50 % of total collection), followed by Diaphus pacificus, Oxyporhamphus micropterus, Cubiceps pauciradiatus, Lampanyctus parvicauda, Bathylagus nigrigenys, Citharichthys platophrys, Pontinus sp. The remaining 28 species were represented by 1-2 individuals. The most abundant were eggs of V. lucetia and O. micropterus. The highest taxonomic diversity, as well as relative abundance of fish larvae was noted for Costa Rica Dome (based on all samples collected in this area). The distribution by ecological preferences of the adults was (no. species/no. family) as follows: epipelagic 5/4; mesopelagic 14/9; shallow water 17/9;

200

bathypelagic (Melanocetus sp.) 1/1. Due to relative proximity to the coast, the collection included larvae, not so commonly found as part of oceanic ichthyoplankton assemblages e.g. leptocephalii of Heteroconger canabus, Hoplunnis sicarius, Ophichthys zophochir. The majority of all fish larvae were recorded above or close to upper thermocline limit. Fish eggs exhibited broader depth distribution but were, nevertheless, principally limited by the lower thermocline limit (Fig. 2). There was an abrupt restriction of ichthyoplankton distribution by the strong oxygen minimum layer (with upper limit at 125-100 m) at two northernmost stations 3503 and 3504.

201

202

203

204

DIET AND FEEDING STRATEGY OF FISHES IN A Ruppia maritima MEADOW, IN THE PATOS LAGOON ESTUARY, BRAZIL Marcelo Bassols Raseira Departamento de Oceanografia, Laboratrio de Ictiologia Fundao Universidade do Rio Grande (FURG) CP 474, Rio Grande RS, 96201-900, Brasil. Phone (92) 613 6246, FAX: (92) 237 5616, e-mail: mbraseira@yahoo.com.br. Alexandre Miranda Garcia Departamento de Oceanografia, Laboratrio de Ictiologia Fundao Universidade do Rio Grande (FURG) CP 474, Rio Grande RS, 96201-900, Brasil. Phone: (53) 233 6539, E-mail: amgarcia@mikrus.com.br. Joo Paes Vieira Departamento de Oceanografia, Laboratrio de Ictiologia Fundao Universidade do Rio Grande (FURG) CP 474, Rio Grande RS, 96201-900, Brasil. Phone: (53) 233 6539 Email: vieira@mikrus.com.br. Abstract We analyzed the diet composition of the ichthyofauna, and the feeding strategy of the white croaker, in a Ruppia maritima meadow in the Patos Lagoon estuary, southern Brazil. The studied species seem to have generalist-opportunist feeding habits, showing different strategies and a wide spectrum of prey. Introduction Seagrass beds are considered important nursery grounds, providing food and shelter for juvenile fish and invertebrates. Studies involving feeding habits of fishes that utilize seagrass meadows help to illustrate the role of such places in the ecology of these species and to comprehend trophic relationships between organisms in a community (Adams, 1976).

205

In the Patos Lagoon estuary, seagrass beds are dominated by Ruppia maritima, representing common habitats in the shallow waters. They have perennial or annual growth cycles with peak of abundance during summer months, and with biomass showing strong local variation between-years (Seeliger et al., 1998). The importance of R. maritima beds for fishes inhabiting the Patos Lagoon estuary seems to vary according to the specie. For instance, several species have a pattern of occurrence and abundance associated with the vegetated area (Jenynsia multidentata, Mugil platanus and Geophagus brasiliensis), particularly the pipefish Syngnathus folletti that seems to occur throughout its entire life in the seagrass beds. In contrast, other species are more abundant outside the seagrass bed (Lycengraulis grossidens and Netuma barba). Finally, some fishes apparently occur indistinctly on both habitats, such as Micropogonias furnieri, Gobionellus schufeldti, Atherinella brasiliensis and Platanichthys platana (Garcia & Vieira, 1997). Garcia & Vieira (1997) suggested that feeding habits of the fish species occurring in the seagrass meadows could explain the structure of the fish assemblage inhabiting the Patos Lagoons seagrass meadows. Although R. maritima beds seems to play a key role as nursery grounds for fishes and decapod crustaceans (Garcia et al. 1996; Garcia & Vieira, 1997), there is a lack of information on the role that these environments play as feeding area for juvenile fishes in the Patos Lagoon estuary. In such context, the present study intends to contribute to the understanding of the diet composition and feeding strategies of fishes inhabiting these vegetated habitats. Materials and Methods The studied area was located in a protected shallow water of Patos Lagoon estuary (Figure 1), where usually occurs dense R. maritima beds in summer and fall (Seeliger et al., 1998). A total of 144 samples were taken between Dec.94 and Mar.95. Two different habitats, inside R. maritima meadow (VG) and a nearby non-vegetated area (NV), were sampled every 15 days in two periods (day: 2 - 6 PM and night: 8 - 11 PM). Samples were taken using a beam-trawl (mouth 1 m wide and 0.5 m high; body mesh 13 mm and cod-end mesh 3 mm).

206

Figure 1 Patos Lagoon estuary study area showing the sampling sites (IN = vegetated; OUT = non-vegetated). Based on their frequency of occurrence, 10 (out of 22) fishes were considered dominant: Micropogonias furnieri, Syngnathus folletti, Jenynsia multidentata, Gobionellus schufeldti, Geophagus brasiliensis, Netuma barba, Platanichthys platana, Mugil platanus, Lycengraulis grossidens and Atherinella brasiliensis. The gut content of eight different species was analyzed: M. furnieri (N=205 individuals), S. folletti (N=124), J. multidentata (N=20), G. schufeldti (N=20), G. brasiliensis (N=20), N. barba (N=20), P. platana (N=20) and A. brasiliensis (N=17). Each item found in the stomach was identified to the closest taxonomic category and classified in 10 different food categories (infauna, epifauna,

207

zooplankton, phytoplankton, nekton, inorganic material, aloctone vegetable, aloctone animal, algae and non identified organic material). Total area (mm2) and total length (mm) were taken for each food item. The following calculations were made for each feeding category: a) Frequency of Occurrence (FO%i): contribution in percentage of the number of stomachs in which a certain feeding category was found (Ni), divided by the total number of analyzed stomachs (NE) and multiplied by 100; b) Percentage of prey-specific abundance (PAEi% sensu Amundsen et al. 1996): percentage a prey taxon comprises (Si) of all prey items in only those stomachs in which the actual prey occurs (S); c) Percentage of the occupied area (PAi%): total area of a certain feeding category found in all the analyzed stomachs (Ai), divided by the total area of all the items found in the stomach (AT) and multiplied by 100. The description of M. furnieris feeding strategy was based on modifications of the Amundsen et al. (1996) and Costello (1990) methods, where values of FOi% (x axis) were plotted against PEAi% or PAi% (y axis).

Results Three fish groups were assembled together according to their diet similarity: GROUP I due to its preference on epifauna, GROUP II because of the dominance of zooplankton and GROUP III for large quantity of infauna (Figure 2). GROUP I - The pipefish S. folletti had a diet composed basically of the epifaunal isopod Munna peterseni and zooplanktonic copepods, while J. multidentata fed mainly on the amphipod Mellita mangrovi, the polichaete Laeonereis acuta and algae.

208

100 80 60 40 20 0 100 80 60
Mean area (%)

INFAUNA Laeonereis acuta Nephtys fluviatilis Kalliapseudes schubartii Erodona mactroides Others

40 20 0 100 80 60 40 20 0 100 80 60 40 20

EPIFAUNA Tanais stanfordi Melita mangrovi Munna peterseni Others

ZOOPLANKTON Copepoda Cirripedia larvae Others

OTHERS Algae Animal aloctone Inorganic matter Others

A. brasiliensis

J. multidentata

G. brasiliensis

G. schufeldti

M. furnieri

Figure 2 Percentage of mean area occupied by each item in the analyzed species separated in terms of food categories. GROUP II - The sardine P. platana consumed mainly zooplankton, with copepods comprising over 60% of the stomach contents. The silverside A. brasiliensis had a diet mainly composed by cirripedia larvae, the polichaete

209

P. platana

N. barba

S. folletti

Nephtys fluviatilis and algae. It was the only specie feeding on aloctone animal, which were composed basically of insects. GROUP III - The infaunal tanaidacean Kalliapseudes schbartii was the main item consumed (50%) by the croaker M. furnieri, followed by the less dominant Laeonereis acuta, Erodona mactroides (especially siphons) and few individuals of N. fluviatilis. The marine catfish N. barba fed mostly of K. schbartii, followed by L. acuta. The cichlidae G. brasiliensis consumed predominantly L. acuta, followed by E. mactroides, K. schbartii and N. fluviatilis. The goby G. schufeldti fed mainly on K. schbartii, N. fluviatilis, and Mellita mangrovi. This fish had the largest amount of sand in its stomach content (PA%=14%). The croaker fed in both vegetated and non-vegetated habitats on similar preys. Kalliapseudes schbartii, was abundant and frequent in the stomach content of the croaker captured inside and outside the vegetation, the isopod Munna peterseni was consumed frequently outside the seagrass, while copepods were consumed mainly in the interior of the vegetation. Sand occurred in the stomachs in similar amounts in both vegetated and non-vegetated habitats. An analysis of the feeding strategy indicates that the croaker had generalist feeding strategy inside and outside the seagrass bed. However, the population outside the bed showed a tendency to be specialist preying upon the infauna, especially K. schbartii, whereas in the interior of the vegetation there was opportunism towards nektonic preys. An attempt to compare the diet of M. furnieri between habitats (VG vs. NV) and period (day vs. night) was made. In the non-vegetated area, during the night, K. schbartii was the only prey being both abundant and frequent, whereas M. peterseni was frequent but not abundant. During the day, in contrast, Tanais stanfordi become important and copepods and M. peterseni become frequent (Figure 3).

210

Figure 3 - Feeding strategy and diet of the croaker M. furnieri between habitats (Outside and Inside) and period (Day and Night). The croakers feeding strategy outside the meadow was of a generalist in daytime and at night, showing a tendency to be specialist in consuming K. schbartii at night, and an opportunistic behavior in capturing fish during the day. Otherwise, inside the meadow, K. schbartii and copepods were frequent and abundant during day and night. During the night, M. peterseni was more frequent and M. mangrovi was important in terms of abundance, although not in

211

frequency. Its strategy was of a generalist at night, consuming specially copepods and less frequently Kalliapseudes. During the day it continued to behave as a generalist preying on K. schbartii and copepods, with an opportunistic feeding on fish and M. mangrovi. Discussion A hypothesis concerning the association of fish with seagrass beds suggests that fish assemblage structure would be primarily determined by the availability, composition and pattern of larvae settlement, being only secondarily influenced by predation (Bell & Westoby, 1986). Garcia & Vieira (1997) speculates that, in the shallow waters of Patos Lagoon estuary, fishes select the habitat (inside or outside R. maritima beds) because of the food availability, and not as a function of predation pressure (Garcia & Vieira 1997). In the Patos Lagoon estuary, the macro-invertebrate benthos shows a low number of species (30-40), and most of them have high seasonal and/or between-year fluctuations in their abundance. Among macro-invertebrate of low mobility there is no specie occurring exclusively in the seagrass meadow (Geraldi, 1997). The preferable food items consumed by the eight fish species analyzed in the present work could be explained by their higher abundance and greater availability. This hypothesis seems to be corroborated by previous findings showing the predominance of carnivore among several fishes that inhabit the Patos Lagoon estuary (Vieira et al, 1998). Higher predation over infauna is probably because the infauna abundance is three folds higher than epifauna (Geraldi, 1997). It could also be because fishes could move easier near to the bottom than in the plant canopy and its associated algae, which could influence the preys capture. The larger capture of copepods inside the meadow might be due to its greater abundance when compared to non-vegetated areas. The meadows reduce the water speed redirecting the flux around and over the bed (Fonseca et al., 1982), consequently working as a trap concentrating planktonic organisms of low mobility, therefore facilitating its capture by predators like fishes. Several authors suggest that seagrass beds are habitats where the benthonic invertebrates, particularly the infauna, are protected from the macro-predators (Adams, 1976). However, this hypothesis does not seem to be totally true for R. maritima meadow in the Patos Lagoon estuary, where fishes feed on similar items in both vegetated and non-vegetated habitats. For example, the infaunal K.

212

schbartii, theoretically more protected between the complex of leaves, roots and stems, which limit predator access (Asmus, 1984), was an abundant and frequent item on the croaker's diet in both studied habitats. The croaker M. furnieri amplifies its feeding spectrum during the day, both inside and outside the meadow, showing a generalist strategy, and some opportunism specific preys. At night, the croaker continues to have a generalist strategy, although with a tendency to being a specialist in capturing K. schbartii, outside, and copepods inside the seagrass bed. The studied species showed different strategies in obtaining food and most of them had a wide feeding spectrum. Even though they prey upon the same feeding categories, they consume different items within these categories, or differ on the feeding strategy used to obtain these items. Also, the high abundance of the available preys, seem to suggest that there is no niche overlap, or competition, between the species of juvenile fish.

Literature Cited Adams,S.M. Feeding ecology of eelgrass fish communities. Trans. Am. Fish Soc. 4: 514 519, 1976. Amundsen, P. A.; Gabler, H.M., Staldvik, F.J. A new approach to graphical analysis of feeding strategy from stomach contents data modification of Costello (1990) method. Journal of fish biology, 48:607 614, 1996. Asmus, M.L. Estrutura da comunidade associada Ruppia maritima no esturio da Lagoa dos Patos, Rio Grande do Sul, Brasil. Tese apresentada Universidade do Rio Grande, para obteno do ttulo de Mestre em Cincias, Oceanografia Biolgica, 1984, 154p. Bell, J. B. & Westoby, M. Abundance of the macrofauna in dense seagrass is due to habitat preference, not predation. Oecologia, Springer Verlag, Berlim, p.205 209, 1986. Costello, M.J. Brief Communications. Predator Feeding Strategy and Prey Importance: a New Graphical Analysis. J. Fish Biol., 36:261-263, 1990.

213

Fonseca, M.S., Fisher,J. S., Zieman, J.C. & Thayer, G.W. Influence of seagrass, Zostera maritima L., on current flow. Estuarine, Coastal and Shelf Science , 15: 351 364, 1982. Garcia, A. M., Vieira, J.P., Bemvenuti, C. E., & Geraldi, R.M. Abundncia e diversidade da assemblia de crustceos decpodos dentro e fora de uma pradaria de Ruppia maritima L., no esturio da Lagoa dos Patos (RSBrasil). Nauplius, Rio Grande 4: 113- 128, 1996. Garcia, A. M. & Vieira, J. P. Abundncia e diversidade da assemblia de peixes dentro e fora de uma pradaria de Ruppia martima L., no esturio da Lagoa dos Patos (RS- Brasil). Atlntica, Rio Grande 19 : 161 181, 1997. Geraldi, R. M. Estrutura da assemblia de macroinvertebrados bentnicos em fundos com e sem vegetao macrfita na regio estuarial da Lagoa dos Patos, Rio Grande , RS- Brasil. Tese de Mestrado. Fundao Universidade de Rio Grande, 1997, 208p. Seeliger, U., Odebrecht C. & Castello,J. P. Os ecossistemas costeiro e marinho do extremo sul do Brasil, Rio Grande, RS, Editora Ecoscientia, 341p, 1998. Vieira, J. P., Castello, J. P., & Pereira, L.E. Ictiofauna (O ambiente e a biota do esturio da Lagoa dos Patos), In Seeliger, U., Odebrecht, C. & Castello, J. P. Os ecossistemas costeiro e marinho do extremo sul do Brasil. Rio Grande, RS, 60 -67p, 1998.

214

DISTRIBUTION AND ABUNDANCE OF NANNOSTOMUS UNIFASCIATUS (CHARACIFORMES: LEBIASINIDAE) AT LAGO AMAN, AMAZONAS, BRAZIL Michel F. Catarino, Instituto de Desenvolvimento Sustentvel Mamirau, Avenida Braslia, 197. Bairro Juru. Tef Amazonas CEP: 69470-000, (92) 647-4233. michel@mamiraua.org.br Haroldo B. C. Rodrigues, INPA/CPBA, Manaus, Amazonas, Brazil. benjamin_haroldo@hotmail.com Eduardo R. Paes, INPA/CPBA, Manaus, Amazonas, Brazil. eduslipknot@bol.com.br Jansen Zuanon, INPA/CPBA, Manaus, Amazonas, Brazil. zuanon@inpa.gov.br EXTENDED ABSTRACT ONLY DO NOT CITE The pencilfish Nannostomus unifasciatus is widely distributed in the Amazon and frequently found in the international ornamental fish trade. Nevertheless, despite its economic importance, very few informations are available about its biology and ecology under natural conditions. During a survey of the fish fauna of the Aman Sustainable Development Reserve - RDSA (130 - 305S, 6250 6500W) conducted in 2002 and 2003, 344 specimens of N. unifasciatus were collected. Sixteen sampling stations were included, eight in black water (from forest streams not submitted to the annual flood pulse), and another eight in mixed water (receiving water from forest streams but seasonally flooded by the white waters of the Japur River). The sampling covered the four main periods of the hydrological cycle: (receding, low, flooding and high water periods), with two samples per habitat and hydrological phase. The fishes were collected with hand nets among the flooded marginal vegetation. The collecting effort

215

employed at each sampling station and phase was of 120 hand net throws along 100 meters of shore line, totalling 1.920 throws and 1.600 meters of river margins. The results are expressed in terms of Catch Per Unit of Effort (CPUE) as the mean number of specimens collected in 120 hand net throws. Nannostomus unifasciatus was found throughout the area of Aman Reserve and its abundance in the catches varied along the hydrological cycle. In black water environments the mean abundance of N. unifasciatus was higher in the receding and low water periods, although in mixed water environments expressively high abundance was just observed during the low water period (Table 1). Table1. Mean values of abundance of N. unifasciatus in black- and mixed water environments in the RDS Aman, Brazilian Amazon. Environment/water Black water Period of the flood cycle Flooding High Receding Low Flooding High Receding Low Mean abundance 22,5 5,0 89,0 26,0 1,5 1,5 1,0 24,5

Mixed water

Our results indicate that the relative abundance of N. unifasciatus in the flooded marginal vegetation at RDSA is influenced by the variation of the water level along the year, as predicted for fish species that inhabit the floodplain areas of the Amazon (Junk et al., 1989). The flooding season also improves the availability of shelter and food resources for the fish fauna and the large area for dispersion results in lower abundance of fish in the samples, whereas the reverse is true for the receding and low water periods. N. unifasciatus seems to be more common in small to medium sized black water tributaries when compared to white- or mixed water environments (JZ, pers. obs.). Although the mean abundance of the species in flooded marginal vegetation in black waters was higher, another 460 specimens of N. unifasciatus were collected in mixed waters by means of other fishing methods applied to floating meadows, mainly during the receding and low water phases. This

216

suggests that floating meadows represent an important habitat for the species during the periods were the predation pressure is higher, both in function of the restricted habitat available for shelter and the high biomass of predatory fishes present in the turbid waters of the Japur River. The pencilfish utilises the complex structure of stems, branches, twigs and roots of the flooded forest as shelter, were its body shape, swimming posture and contrasting longitudinal black stripe probably renders it cryptic. In this sense, the different strategies of habitat occupation of N. unifasciatus in black and mixed waters reflect not only the seasonal variation of the water level but also the availability of specific habitats and shelter grounds along the margins of the water bodies of the RDA Aman. Acknowledgements The authors wish to thanks to Instituto de Desenvolvimento Sustentvel Mamirau and Instituto Nacional de Pesquisas da Amaznia for the financial and logistical support. References Junk, W. J.; P. B. Bayley & R. E. Sparks. 1989. The flood pulse concept in river-floodplain systems. In D. P. Dodge (ed.) Proceedings of the International Large River Symposium. Can. Spec. Publ. Fish. Aquat. Sci., p. 110-127.

217

218

FEEDING ECOLOGY OF THE LEAF FISH MONOCIRRHUS POLYACANTHUS (PERCIFORMES: POLYCENTRIDAE) IN THE AMAN LAKE, BRAZILIAN AMAZON. Michel F. Catarino, Instituto de Desenvolvimento Sustentvel Mamirau, Avenida Braslia, 197. Bairro Juru. Tef Amazonas CEP: 69470-000, (92) 647-4233. michel@mamiraua.org.br Jansen Zuanon, Instituto Nacional de Pesquisas da Amaznia (INPA CPBA), CP 478, Av. Andr Arajo,2936, Petrpolis Manaus, AM, Brazil, CEP: 69083-970. Phone: 55 92 643-3253. zuanon@inpa.gov.br EXTENDED ABSTRACT ONLY DO NOT CITE Monocirrhus polyacanthus is a leaf-mimicking fish that occurs along most of the Amazon basin. Although widespread, Leaf fishes are uncommon in museum collections and apparently not abundant in natural environments. As a consequence of this apparent rarity, few published information is available about its natural history. During a fish faunal survey conducted at Aman Lake (localized in the Aman Sustainable Development Reserve, 130 - 305S, 6250 6500W) between 2002 and 2003 we collected 52 individuals of M. polyacanthus of 31.0-82.0 mm SL, and analyzed the stomach contents of 35 specimens. Males and females were of similar sizes (males: mean = 51.1 mm SL, range 32,3-73,1; females: mean = 53.3 mm SL, range 31,0-82,0;n = 22). The smallest adult male measured 39.2 mm SL, whereas the smallest ripe female attained 50.0 mm SL. Leaf fish diet was composed by fishes and invertebrates (Table 1).

219

Table 1. List of preys consumed by Leaf fish specimens in Aman Lake, Brazilian Amazon, during 2003-2004. Total of preys (n) =35. Total of stomachs with food= 18. Prey type Taxon Hemigrammus bellottii Hemigrammus sp. Nannostomus eques Fishes Nannostomus unifasciatus Nannostomus trifasciatus Unidentified Characiformes Unidentified Cichlidae Unidentified fishes Palaemonetes sp. (shrimp) Euryrhynchus sp. (shrimp) unidentified shrimp Invertebrates Conchostraca (micro crustacean) Coleoptera (adult beetle) Hymenoptera (ant) Odonata (dragonfly nymph) Ephemeroptera (mayfly nymph) n 6 1 3 2 1 2 2 4 1 1 1 3 1 1 2 4 % 17,1 2,9 8,6 5,7 2,9 5,7 5,7 11,4 2,9 2,9 2,9 8,6 2,9 2,9 5,7 11,4

Up to five prey items were recorded per stomach. Of the 18 specimens with identifiable prey in the stomach, 10 contained a single prey item; five contained two preys and three other specimens had four, five and six preys each. Eleven Leaf fish specimens contained only fish in the stomach; four specimens had just invertebrates and three other presented both fish and invertebrates. Fish constituted the main prey of the M. polyacanthus (60%, n = 35), of which 88.2 % were Characiformes and 11.8 % were Perciformes (Cichlidae). Among the identified characiform preys (n = 15) pencil fishes (Nannostomus spp., Lebiasinidae) constituted 40.0 % of the prey specimens whereas small tetras (Hemigrammus spp., Characidae) represented other 46.7 %. These prey fishes

220

measured 11.0 - 33.0 mm SL. Invertebrates (shrimps, mayfly and dragonfly nymphs) were preyed mainly by smaller M. polyacanthus individuals (mean = 45.2 mm SL; range: 28.0 54.2; n = 10), whereas fish predominated in the stomach contents of larger specimens (mean: 57.8 mm SL; range: 28.5 82.0; n = 19). There was a significant relation between Leaf fish size and prey size (R2 = 0,438, F = 13.239, p = 0.002, n = 19; Figure 1).

40

30
Prey size (mm)

20

10

0 20

30

40 50 60 70 Leaf fish size (mm)

80

90

Figure 1. Relation between predator size and prey size for Leaf fish specimens collected at Aman Lake, Brazilian Amazon, during 2003-2004. N = 18. Also, the mean proportional size of the prey ingested by the Leaf fish was large and corresponded to 35.2 % of predators length (range: 7.0 63.6 %; n = 19). The large and extremely protractile jaws of the Leaf fish allow the capture of proportionally large preys, which are swallowed whole. Large preys were occasionally found occupying up to the lower portion of the esophagus of the Leaf fish, especially when there were several preys in the stomach. Also, in these cases the prey were found bent in U or even curled and tightly packed in the stomach. Bent prey fishes were also observed in the stomach contents of

221

other piscivores such as catfishes (e. g., Phractocephalus hemioliopterus, Pimelodidae and Hemicetopsis macilentus, Cetopsidae; pers. obs.) and of a marine sardine (Chirocentrodon bleekerianus, Pristigasteridae; Sazima et al., 2004). Our results also suggest that Leaf fish exploits the dominant prey fish species occurring in its habitat; nevertheless, prey availability was not analyzed in the present study. Field observations indicate that Leaf fish inhabits the shallow marginal areas of the streams, close to submersed leaf litter banks, where its dark brown coloration and mottled pattern probably renders it cryptic among the drifting soaked dead leaves. The general appearance and slow movements of the Leaf fish may have evolved as adaptations that allows it to mimic drifting dead leaves and so permitting a dissimulated approach to potential preys. Acknowledgements. The authors wish to thanks to Instituto de Desenvolvimento Sustentvel Mamirau and Instituto Nacional de Pesquisas da Amaznia for financial and logistical support. References Sazima, C., R. L. Moura and I. Sazima. 2004. Chirocentrodon bleekerianus (Teleostei: Clupeiformes: Pristigasteridae), a small predaceous herring with folded and distinctively oriented prey in stomach. Brazilian Journal of Biology, 64 (1): 165 - 168.

222

DIEL VARIATIONS OF THE FISH COMMUNITY IN THE SEAGRASS BEDS OF GUADELOUPE ISLAND Dorothe Kopp Universit des Antilles et de la Guyane Laboratoire de biologie marine, BP 592 97159 Pointe--Pitre France phone: (590)489217 fax: (590)489211 email: dorothee.kopp@univ-ag.fr Yolande Bouchon-Navaro Universit des Antilles et de la Guyane email: yolande.bouchon@univ-ag.fr Claude Bouchon Universit des Antilles et de la Guyane email: claude.bouchon@univ-ag.fr Max Louis Universit des Antilles et de la Guyane email: max.louis@univ-ag.fr

EXTENDED ABSTRACT ONLY - DO NOT CITE Introduction The bay of the Grand Cul-de-Sac Marin, closed by a barrier reef 30 km long, is located in the north of Guadeloupe Island (Lesser Antilles). From the shore, bordered by mangroves, to the barrier reef, the bottoms of the lagoon are covered by seagrass beds composed mainly of Thalassia testudinum and less frequently of Syringodium filiforme. The ichtyofauna of these seagrass beds is already known (Aliaume et al., 1990; Bouchon-Navaro et al., 2004). However, the nyctemeral variations of the ichtyofauna have not yet been studied although some data are available for other Caribbean islands (Robblee & Zieman, 1984; Nagelkerken et al., 2000).

223

The purpose of the work was to study the day-night variations of the fish community at two stations: Caret Islet where seagrass beds are close to the barrier reef and Baie-Mahault where they are located near mangroves and to investigate fish exchanges between seagrass beds and neighbouring ecosystems like mangroves and coral reefs. Material and methods A fish trap constituted by a fence net terminated by three hoopnets at each of its extremities was settled in the seagrass beds. The hoopnets were visited every day at dawn and dusk. As this device is mostly efficient during the night, day samples were completed by the use of a seine net (32 trap samples / 96 seine samples). Presence-absence data, as well as number of individuals and biomass per species were treated with cluster and factorial analyses. Results Out of a total of 98 species belonging to 36 families collected in the two studied seagrass beds, 71 species were observed in the seagrass beds close to the coral reef and near mangroves, the fish community was composed of 50 species. Results of factorial and cluster analyses are represented in Figure 1. The first factorial plan gathers 36% of the information, with 21% of the variance explained by axis 1 and 15% by axis 2. Axis 1 clearly separates two types of stations: on the one hand, all the samplings made close to the reefs and on the other hand, those from coastal mangroves (Fig.1). The hierarchical classification confirms these results since it also separates the stations into two distinct groups on the basis of their location in the lagoon (Fig.1). Axis 2 opposes the samples collected by day to those collected by night. This day-night opposition is much more marked for the seagrass fish community close to the barrier reef than for the one located near mangroves (Fig.1). The hierarchical classification confirms these results by separating, in each station, the diurnal samplings from the nocturnal ones.

224

Figure 1: Results of the cluster and factorial analyses on the biomass data. (BM: Baie-Mahault; CA: Caret islet; J: Day; N: Night).

225

This phenomenon suggests that the exchanges between seagrass beds and reefs are more significant than those occurring between seagrass beds and mangroves. Examination of the species composition in the day and the night collections revealed that in Caret islet, the fish community was composed of 34 diurnal species, 29 nocturnal species and 8 indifferent species. Near the coastal mangroves, the community included 24 diurnal species, 10 nocturnal species and 16 indifferent species. Figure 2 illustrates the separation between day and night assemblages and between the fish community of the two stations.

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Caret Islet Day

Caret Islet Night

Baie-Mahault Day

Baie-Mahault Night

Miscellaneous Sparidae Lutjanidae Clupeidae

Diodontidae Haemulidae Holocentridae Muraenidae

Sphyraenidae Gerreidae Engraulidae Dasyatidae

Figure 2: Relative abundance in biomass of the main families in the two stations, day and night. During the day, Lutjanidae, Gerreidae and Clupeidae constitute the main biomass at Caret Islet. At night, Diodontidae, Holocentridae, Clupeidae and Haemulidae coming from the reef become dominant. At Baie-Mahault, the fish community is dominated both day and night by Sparidae and Gerreidae. Lutjanidae forms an important biomass by day whereas Muraenidae and

226

Haemulidae are preponderant at night. Discussion and conclusion The present study has revealed a clear day-night difference in the fish community living in the seagrass beds, both in the station close to the reef than in that located near mangroves. Differences are more important for the communities living near the reef areas. The appearance of new species at night in the Thalassia seagrass beds is responsible for this phenomenon. Trophic migrations of the reef ichtyofauna take place at night : fishes leave the reef at dawn in order to forage in the seagrass beds and come back at sunrise (Weinstein & Heck, 1979; Robblee & Zieman, 1984). That was the case in the present study for Haemulidae, Holocentridae and Apogonidae. The diurnal fish community was dominated by Scaridae, mainly Sparisoma radians, and by Labridae as also observed by Robblee & Zieman (1984). In the seagrass beds located near mangroves, some species typical of that biotope (Bairdiella ronchus, Bairdiella sanctaluciae, Mugil curema, Diapterus rhombeus, Chloroscombrus chrysurus), were found at night in the seagrass meadows. Apart from the latters, the majority of the species was found by day as well as by night in the seagrass beds. More investigations on the possible migrations of the ichtyofauna from the seagrass beds to the mangroves should be made. References Aliaume, C., G. Lasserre, and M. Louis. 1990. Organisation spatiale des peuplements ichtyologiques des herbiers Thalassia de Grand Cul-de-Sac Marin en Guadeloupe. Revue Hydrobiologie tropicale, 23 (3): 231-250. Bouchon-Navaro, Y., C. Bouchon and M. Louis. 2004. Lichtyofaune des herbiers de Phanrogames marines des Antilles franaises : intrt de leur protection. Revue Ecologie (Terre Vie), 59: 253-272.

227

Nagelkerken, I., M. Dorenbosch, W.C.E.P. Verberk, E. Cocheret de la Morinire and G. van der Velde. 2000. Day-night shifts of fishes between shallowwater biotopes of a Caribbean bay, with emphasis on the nocturnal feeding of Haemulidae and Lutjanidae. Marine Ecology Progress Series, 194: 5564. Robblee, M.B. and J.C. Zieman. 1984. Diel variation in the fish fauna of a tropical seagrass feeding ground. Bulletin of Marine Science, 34 (3): 335345. Weinstein, M.P. and K.L. Heck. 1979. Ichtyofauna of seagrass meadows along the caribbean coast of Panama and in the Gulf of Mexico: composition, structure and community ecology. Marine Biology, 50: 97-107.

228

REPRODUCTIVE PERIOD AND SIZE OF THE FIRST GONAD MATURATION OF THE HIPPOCAMPUS REIDI SEA-HORSE IN THE BRAZILIAN NORTHEAST REGION Rosana Beatriz Silveira Laboratrio de Aqicultura Marinha-LABAQUAC/Projeto Hippocampus Rua Saber s/n, Praa Sete, Porto de Galinhas, Ipojuca, PE, Brasil 55590-000 Phone: 558135522990 e-mail: labaquac@yahoo.com Nelson Ferreira Fontoura Faculdade de Biocincias PUCRS Caixa Postal 1429, CEP 90.6190-900, Porto Alegre, RS, Brasil e-mail: nfontoura@pucrs.br EXTENDED ABSTRACT ONLY DO NOT CITE

Introduction
Although they are listed as vulnerable by the IUCN Red List, the Hippocampus reidi sea-horses need studies about their basic biology. Some aspects of population dynamics, such as fertility and growth, have been developed in laboratory (Vincent, 1990; Silveira, 2000), however, population dynamic data in natural environments, such as the reproductive period, age of the first gonad maturation, were unknown until that moment. Registered only for the Americas (Lourie et al., 1999), H. reidi inhabits the coastal and estuary waters on Brazilian coast including the south of the country (Chao et al., 1982), existing from tropical to temperate zones. This study examined, for the first time, the reproductive period and the size of first maturation of the H. reidi in a natural environment in tropical Brazil. Methods The field studies were developed in the estuary of Maracape River, 83214.9S and 350017.8W, in the City of Ipojuca, State of Pernambuco, in

229

the northwest region of Brazil. The estuary of this river presents a vegetal community composed mostly of Rhizophora mangle and it communicates with the sea through a short channel forming the Pontal de Maracape, whose beach length is 570 m. Data were collected through dives and observations performed weekly, from June 2001 to July 2003. The animals, when viewed, were collected manually, conditioned in plastic recipients with local water and taken to the laboratory for height measurement (top of the head to the edge of the stretched tail) and weight measurement in a semi-analytical scale (0.001g). The pregnant males (n=96) were kept in the laboratory until the offspring were born. The reproductive period and peak were determined through the relative frequency of the pregnant males in relation to the total number of sampled males along the months that the collection was done. The age of the first gonad maturation was calculated through the absolute and relative frequencies of pregnant males per interval of height class. Results The reproductive period of the H. reidi in the Maracape estuarine area is extended along all the year, with the presence of pregnant males in several height classes. Although the reproductive period is extensive, there was a variation in the reproductive intensity, whose peak happened between June and October, comprehending the winter and part of the spring. It shall also be noted that, during the months of May and November, 50% of the sampled male population were pregnant. In the remaining months, H. reidi showed a moderate reproductive activity, reaching the lowest frequencies in the February and March, with no pregnant male detected in the month of April (Figure 1). The size of the first gonad maturation (L50) in the H. reidi is approximately 13.5 cm high. The population showing a higher size than this range starts to increase gradually its reproductive potential, reaching, over 17.0 cm, 100% of reproductive activity. The smallest pregnant male presented height of 12.0 cm, while the biggest one presented 17.7 cm (Figure 2).

230

1,00
Pregnant male

0,80 0,60 0,40 0,20 0,00

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Sampling months
Jan Fev Mar Abr Mai Jun Jul Ago Set Out Nov Dez

Figure 1. Relative frequency of pregnant male Hippocampus reidi sea-horse in the Maracape River estuary, in Pernambuco, Brazil, between June/2001 and July/2003.

1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00
1 10, 1 1 ,1 11 2 -1 1 12, 13 ,1 13 14 1 14, 1 5 1 15, 16 ,1 16 17 1 17, 1 8

Pregnant male

Height Class (cm)

Figure 2. Relative frequency of pregnant male Hippocampus reidi sea-horse, per interval of height class, in the Maracape River estuary, in Pernambuco, Brazil, between June/2001 and July/2003.

231

Discussion and Conclusion

According to Hatcher et al. (1989), the reproductive season for fishes of temperate zones is short, and determined by external factors, such as the light and temperature, while for the tropical species, it is usually more extensive and frequently governed by two seasons of monsoons, based on the winter, rainy seasons and standards of currents. Vincent (1990) observed that H. reidi in the laboratory kept reproductive for more than eight months, and that the reproductive period, in this case, was not influenced by monsoons or temperature. According to Lourie et al. (1999), the Hippocampus shows a variation in its reproductive period, from two to nine months, depending on the species and geographical distribution. In this study, H. reidi keeps the reproductive activity during all the year, although its reproductive peak is between June and October, including, for the Brazilian northeast region, the winter (rainy season) and part of the spring. This reproductive intensity related to the rainy season may be associated with a higher alimentary potential presented by the young ones, with an intensification of the secondary production. The size of the first gonad maturation calculated for the first time in the H. reidi of 13.5 cm allows a suggestion of a minimum size for commercial fishing, fulfilling the species preservation needs, and it allows the human populations that live on such fishing to use this resource in a sustainable way.
Acknowledgements

To the Research and Development National Council (CNPq - Conselho Nacional de Pesquisa e Desenvolvimento), for the doctor scholarship of the first author, and the City Administration of Ipojuca-PE, for the partnership with the Project Hippocampus, that allowed this study to be done.
References

Chao, L.N., Pereira, L.E., Vieira, J.P., Bemvenuti, M.A., Cunha, L.P.R., 1982. Relao preliminar dos peixes estuarinos e marinhos da Lagoa dos Patos e regio costeira adjacente, Rio Grande do Sul, Brasil. Atlntida 5:67-75.

232

Lourie, S.A., Vincent, A.C.J. and Hall, H.J. 1999. Seahorses: an identification guide to the words species and their conservation. Project Seahorse, London, UK. 224 p. Silveira, R.B.2000. Comportamento reprodutivo e desenvolvimento inicial de Hippocampus reidi Ginsburg, 1933 em laboratrio. Biocincias 8 (1): 115-122. Vincent, A.C.J, 1990. Reproductive Ecology of Seahorses. PhD Thesis, University of Cambridge. 101p.

233

234

THE USE OF MERCURY AND STABLE ISOTOPES TO INVESTIGATE FOOD CHAIN STRUCTURE IN AQUATIC ECOSYSTEMS IN THE AMAZONIA

Mario J. F. Thom-Souza Instituto Nacional de Pesquisas da Amaznia Alameda Cosme Ferreira, 1756, Manaus, AM thome@inpa.gov.br Bruce Rider Forsberg Instituto Nacional de Pesquisas da Amaznia forsberg@horizon.com.br Richard Carl Vogt Instituto Nacional de Pesquisas da Amaznia Maria das Graas Pires Sabbayrolles Universidade Federal do Par; Campus-Santarm. Jos Reinaldo Pacheco Peleja Instituto Nacional de Pesquisas da Amaznia
Introduction

15N has been used as an indicator of trophic level in many food chain studies because of the large isotopic fractionation which occurs as nitrogen passes between trophic levels (Leite, et al., 2002; Hamilton et al., 1992). The natural variation in stable carbon isotope ratios, 13 C, has also been used as a tracer of autotrophic carbon sources in these studies (Araujo-Lima et al., 1986; Forsberg et al., 1993). 15N has been shown to be a poor indicator of trophic level in Amazonian ecosystems due to the large isotopic variation encountered between plants at the base of the aquatic food webs (Victoria et al., 1992). Total mercury concentrations also accumulate consistently through the aquatic food chain and have been shown to vary little among aquatic plants in the Amazon. Here we demonstrate the potential of using mercury as a trophic level indicator and the value of using mercury together with 13C to investigate the structure of aquatic food chains in the Amazon River basin.

235

Methodology

To test this methodology, we collected aquatic plants and animals along the main channel of the Rio Negro and analyzed them for 13 C, 15N and total mercury concentration. In total, 180 samples were collected, including 33 flooded forest tree leaf, 33 periphyton, 68 benthic fish, 25 large catfish and 21 turtle samples. Most of these samples were collected during the high water periods of 1997, 1999 and 2000. Periphytic algae were collected on the margins of rivers and streams where they grew attached to submerged tree branches. Benthic fish and large catfish were collected with a deep trawl deployed at depths varying from 5 - 15 m. Twenty eight species of benthic fish were collected, predominantly Gymnotiformes and Siluriformes, but with numerically significant contributions from the Clupeiformes and Perciformes as well. Four species of large catfish were collected, Pseudoplatystoma trigrinum, Pseudoplatystoma fasciatum, Brachyplatystoma filamentosum and Phractocephalus hemiliopterus. Two species of turtles were collected, Podocmenis erythrocephala and Peltocephalus dumeriliana. Fish and turtle flesh, tree leaves, and algae were dried at 60C and then ground with mortar and pestle to a fine powder. Subsamples were analyzed for total Hg, following acid digestion, by cold vapor atomic fluorescence Spectroscopy (CVAFS) at the Universidade Federal do Para in Santarm, Para. Additional sub samples were analyzed for stable isotopes of carbon and nitrogen by mass spectrometry at the Centro de Energia Nuclear na Agricultura in Piracicaba, So Paulo, following methodology described by Forsberg et al., (1993). Results for different groups of plants and animals were compared using the SNK test (P<0.05).
Results

Mercury concentrations in the tissues of the organisms demonstrated bioaccumulation along the food chain with average values (+ standard error) of 0.030.01, 0.040.01, 0.110.01, 0.160.02 and 0.530.05 ppm for forest leaves, periphytic algae, turtles, benthic-fish and big-catfish, respectively. The SNK test indicated that the mercury levels of periphyton, tree leaves and turtles were not significantly different but were all significantly lower than those of large catfish. The mercury concentrations of benthic fish were significantly higher than those of both plant groups, significantly lower than those of the large predatory catfish and similar to those of turtles. A rough positive correlation was found between the levels of mercury and 15N for all samples (Figure 1).

236

1.6
Hg (ppm) weigth wet
C - big-catfish B - benthic-fish T - turtles p - periphyton F - flooded fo rest

1.2

0.8
C CC C BB C CC BCC C C C CC C B CBC BB BC BB B B B C B B BCB BT C B T T T BBB B BB B T BB B B B p T TT BB BB BB B Tp T TB B BBB B B B pT T p F FF pF p FTT T BBB BBB B B B F FFFFpFppFpF p p T T p p BB F pp F F p Fp F F T p B B F F F

0.4

0.0 -3

15

15N
Figure 1. Relationships between total mercury (Hg) and 15N levels for plants and consumers of the Rio Negro.

The small variation in mercury relative to 15N values for plants together with the consistent increase in mercury levels with known trophic level apparent in this figure indicates the superiority of mercury as an indicator of trophic level in this system. The two turtle species included in the study were predominantly herbivorous. Assuming that the average increase in mercury in these species relative to aquatic plants (0.075 ppm) represents a constant bioaccumulation rate per trophic level in the food chain, we classified turtles, benthic fish, and large catfish as herbivores, omnivore/predator 1 and predator 5, respectively, based on the average mercury level of each group (Figure 2). The average 13C values (+ SE) determined for leaves, periphyton, benthic-fish, big-catfish and turtles were -30.240.25, -38.400.7, -33.650.38, -32.260.37 and -31.360.39,

237

respectively. The 13C values of all consumers all between the average values for plants indicating that their carbon was derived from a mixture of these two autotrophic sources. Turtles and large catfish derived their carbon predominantly from igapo forest trees while benthic fish showed a much larger contribution from periphytic algae. When the results for both 13C and mercury were plotted on the same graph (Figure 2) the trophic level and autotrophic carbon sources of aquatic consumers were clearly evident.

0,5 0,4 0,3 Benthic-fish Turtles Igap forest 0,1 0 0,2

Perifton

-40

-38

-36

-34

-32

-30

13

Figure 2. Average levels of total Hg and 13C (+ Standard Error) for aquatic plants and consumers of the Rio Negro food web, indicating the trophic level and autotrophic carbon sources of consumers.
Conclusion

These preliminary results demonstrated the superiority of total mercury as an indicator of trophic level in aquatic food webs in the Amazon when compared to the traditional 15N index. When total mercury and 13C were used together, they provided a clear picture of both the trophic level and autotrophic carbon sources of aquatic consumers. Results from the rio Negro indicated that turtles were herbivores and derived their carbon predominantly from flooded forest trees. The majority of the fish analyzed were predators and derived their carbon either predominantly from flooded forest trees (large catfish) or from a mixture of this source and periphytic algae.

238

Hg (ppm) weigth wet

Hg (ppm) - Trophic level 0.03 Plants 0.11 Herbivore or Detritivore 0.19 Predator 1 0.27 Predator 2 0.35 Predator 3 0.43 Predator 4 0.51 Predator 5

Big-catfish

0,6

References

Araujo-Lima, C.A.R.M.; Forsberg, B.R.; Victoria, R.L. and Martinelli, L. 1986. Energy sources for detritivores fishes in the Amazon. Science 234: 12561258. Forsberg, B.R.; Araujo-Lima, C.A.R.M.; Martinelli, L.M.; Victoria, R.L and Bonassi, J.A. 1993. Autotrophic carbon sources for fish of the Central Amazon. Ecology 74: 643 652. Hamilton, S.K., Lewis, W.M., Jr. and Sippel, S.J. 1992. Energy sources for aquatic animals in the Orinoco River floodplain: evidence from stable isotopes. Oecologia 89: 324-330. Leite, R.G., Arajo-Lima, C.A.R.M., Victoria, R.L., and Martinelli, L.A. 2002. Stable isotope analysis of energy souces for larvae of eight fish species from Amazon floodplain. Ecology of Freshwater Fish. 11: 56-63. Victoria, R.L., Martinelli, L.A., Matsui, E., Forsberg, R.B., Richey, J. E. and Devol, A.H. 1992. The use of stable isotopes in studies of nutrient cycling: Carbon isotope composition of Amazon varzea sediments. Biotropica 24(2b): 240-249.
Acknowledgements

We thank Drs. Marcelo Moreira, Luiz Martinelli and Reynaldo Victoria for support and orientation in the analysis of stable isotopes at CENA and CNPq/Pronex N 466098/2001-4 and Dr. Ning Labbish Chao for financial and logistic support.

239

240

REPRODUTIVE BEHAVIOUR AND BIOLOGY OF HYBRID RED TILAPIA, OREOCHROMIS NILOTICUS X O. MOSSAMBICUS

Ana Patrcia Targino de Medeiros and Maria Emlia Yamamoto Post-Graduate Programme in Psychobiology, Department of Physiology/CB/Universidade Federal do Rio Grande do Norte, Lagoa Nova, CEP 59078-970, Natal, RN, Brazil. E-mail anapatriciatargino@yahoo.com.br Sathyabama Chellappa Post-Graduate Programme in Aquatic Bioecology, Department of Oceanography and Limnology, Universidade Federal do Rio Grande do Norte, Praia de Me Luiza, s/n. Via Costeira. CEP 59014-100, Natal, RN, Brazil. email:bama@dol.ufrn.br

Introduction

Red strains of two-species hybrid tilapia (Oreochromis niloticus x O. mossambicus), commonly known as red tilapia, are being used for semiintensive fish culture purposes in north-eastern Brazil (Sena et al., 1993; Chellappa et al., 1996). However, limited information in available regarding the reproductive biology of these cichlid fishes. These hybrids are aggressive towards conspecifics and male reproductive aggression is related to territorial defence, especially during the reproductive phase. This study was conducted to investigate reproductive aggression, territorial defence and gonadal development of adult males and females.
Materials and Methods

Adult male hybrid red tilapia were acquired from the Experimental Fish culture Station, Macaba, RN of the Federal University of Rio Grande do Norte, Brazil. Twenty fights were staged between adult males of red tilapia of different body sizes, and each test was conducted in triplicate. As the male fish settled down in the individual aquaria, they started to establish and defend territories. Fish defending territories and those without territories were used in the tests. Encounters were programmed between five resident adult males with established territories (resident status), each maintained in separate glass

241

aquarium, with five adult non-resident males (non-resident status). Resident fishes were those that established and defended territories. Fights were also staged in neutral aquaria, between adult males, which held territories, with those, which never established territories. In this test category, the two males were introduced simultaneously in a neutral aquarium. At the beginning of the tests, the measurements and weighs of the fishes were registered. The observations were conducted using the continuous focal method. Forty males and females of the hybrid red tilapia were used to determine the reproductive biology. The fishes were measured, weighed and dissected. The gonads were macroscopically classified. Histological examinations were carried out on sections of ovaries and testes stained with haematoxylin and eosin. Fecundity, gonadosomatic index and relative proportion of oocytes were estimated.
Results and Discussion

In all fights, one fish emerged as a loser as the agonistic interaction proceeded. The bigger males were more aggressive in agonistic encounters and acquired good spawning territories and were preferred by the females. In fights with a resident and an intruder, aggression manifested by the resident male was significantly higher than that of the intruder male. However, in fights on neutral territories, the larger fish won all fights. The number of attacks given by the winner was lowest in fights on neutral ground and highest in fights with a large resident. Resident males of similar size were more aggressive during the encounters (Figure 1), similar to the observations of Turner and Huntingford (1986). In territorial species, prior residence was the major determinant of the outcome of territorial encounters between males of hybrid red tilapia. Possession of a territory and relative body size of the males were determinant factors of territorial encounters.

242

Mean frequency of aggression

100 80 60 40 20 0 resident nonresident neutral territory resident territory

resident non-resident neutral territory resident territory

Figure 1- Mean frequency of aggression presented by males of red hibrid tilapia, Oreochromis niloticus x O. mossambicus The macroscopic observations of the gonads of hybrid red tilapia showed four stages of development: immature, maturing, mature and partially spent. The mature females showed simultaneous occurrence of oocytes of different sizes, indicating partial spawning (that each individual spawns more than once within a breeding cycle). (Cmara and Chellappa, 2000). The microscopic analyses, it was possible to identify the immature stage, phase I and II corresponding to the maturation stage, the mature stage and the partially spent stage with some empty follicles, mature oocytes with a large quantity of immature and maturing oocytes. The mean fecundity was 1213 mature oocytes. The microscopic analyses of testes showed the primary and secondary spermatogonia; primary and secondary spermatocytes; spermatids and spermatozoa. The mean gonadosomatic index of males was 0,73 and that of females was 2,75. It is a iteroparous species with a low fecundity due to its partial spawning habit coupled with the high degree of parental care.

243

References

Cmara, M. R. and Chellappa, S. 2000. Reproduo nas fmeas do hbrido vermelho de tilpia Oreochromis niloticus x Oreochromis mossambicus (Osteichthyes: Cichlidae). Revta Ecol. Aqua tropi., 10, 77-86. Chellappa, S. Chellappa, N.T. Silva, E. A. Huntingford, F.A. and Beveridge, M.C. M. 1996. The diet of hybrid red tilapia, Oreochromis niloticus (L) x Oreochromis mossambicus (Peters) reared in the freshwater ponds of northeastern Brazil. Aquaculture Research. 27, 945-952. Sena, R. F., Fernandes, M. O. and Volpato, G. L. 1993. Heterogeneous growth in the Nile tilapia: social stress and carbohydrate metabolism. Physiology, Behavior. 54: 319-323 Turner, G. F. and Huntingford, F. A. 1986. A problem for game theory analyses: assessment and intention in male mouth brooder contests. Animal Behavior. 34: 961-970.

244

REPRODUCTIVE BIOLOGY OF THE ROUGHNECK GRUNT, POMADASYS CORVINAEFORMIS (OSTEICHTHYES: HAEMULIDAE)

Anairam de Medeiros e Silva and Sathyabama Chellappa Post-Graduate Programme in Aquatic Bioecology, Department of Oceanografhy and Limnology, Universidade Federal do Rio Grande do Norte, Praia de Me Luiza. s/n. Via Costeira. CEP 59014-100, Natal/RN, Brazil. E-mail: bama@dol.ufrn.br
Introduction

Brazil has approximately eight thousand kilometers of spectacular coastline, with mangrove-swamps and coral-reefs. A larger part of this coastline is in the tropical and sub-tropical regions, which favours an exuberant diversity of marine fish species. Coastal fishing is practised with local crafts and 75% of the coastal fish production is from the north-eastern Brazil (Ogawa and Koike, 1987; Paiva, 1997). The roughneck grunts are small sized fishes, abundant in the coastal waters of northeast and contribute to the fishery production. Gonadal development of marine fish species is an important information to maintain the fishery stocks (Fonteles-Filho, 1989). The present study was carried out with an objective to determine the reproductive aspects of the roughneck grunt, Pomadasys corvinaeformis Steindachner, 1868 (Osteichthyes: Haemulidae) from the coastal waters of Ponta Negra, Rio Grande do Norte.

Materials and Methods

Fish samples were collected on a monthly basis during August, 2002 to July, 2003 .from the coastal waters of Rio Grande do Norte, Brazil. The fishes were measured, weighed, dissected and the gonads were removed, weighed, examined macroscopically to separate the males from the females and to determine the stage of maturation. Histological studies were carried out using hematoxilineosin stains for microscopic characterization of the gonads (Michallany, 1990). The gonadosomatic index (GSI), fecundity and type of spawning were determined besides sex ratio and the size at first gonadal maturation of P. corvinaeformis (Vazzoler, 1996).

245

Results and Discussion

The results show that the study species have a sex-ratio of 1M:2,2F with a predominance of females (69%). The males attained gonadal maturity at 103 mm and females at 104 mm of total length. The macroscopic characteristics of the ovaries and testicles revealed four stages of gonadal development, such as, immature (I), maturing (II), mature (III) and spent (IV). (Figure 1). The mean fecundity was 44716 mature oocytes, varying from 15056 to 83316. P. corvinaeformis is a total spawner and the gonadosomatic index ranged from 0.01 to 5.49 for females and from 0.01 to 0.85 for males. The probable spawning season occurred in the months of December, 2002 to June, 2003, coinciding with the raining season of the region. The microscopic analyses revealed seven phases of gonadal maturation for the females (immature, initial maturing phase, final maturing phase,, mature (initial, intermediary and final) and spent. Four phases for the males were identified similar to the macroscopic stages.
References

Fonteles Filho, A. A. Recursos pesqueiros: biologia e dinmica populacional. Fortaleza: Imprensa Oficial do Cear, 1989. 296p. Michalany, J. Tcnica histolgica em anatomia patolgica com instrues para o cirurgio, enfermeiro e citotcnico. Ed. So Paulo, SP, 1990, 247p. Ogawa, M. and Koike, J. (Eds). Manual de Pesca. Associao dos engenheiros de pesca do Estado do Cear. Fortaleza: Grfica Batista, 1987. 799 p Paiva, M. P. Recursos pesqueiros estuarinos e marinhos do Brasil. Fortaleza: Editora da Universidade Federal do Cear, 1997. 278 p. Vazzoler, A. E. A. M. Biologia da reproduo de peixes telesteos: Teoria e Prtica. Maring: Eduem, 1996, 169p.

246

Figure 1 Gonads of the males and females of P. corvinaeformis showing the four stages of maturation: I immature, II maturing, III mature and IV - spent.

247

248

GENETIC CHARACTERIZATION OF PEACOCK BASS (CICHLA MONOCULUS) AND RED-BREASTED PIRANA (PYGOCENTRUS NATTERERI) POPULATIONS COLLECTED IN FIVE LAKES IN THE AREA UNDER THE INFLUENCE OF PETROLEUM TRANSPORT ACTIVITY BETWEEN TESOL (TERMINAL SOLIMES) AND THE REMAN (REFINERY OF MANAUS).

Castro, L.E.A.; Cunha, L.K.H.; Silva, M.N.P.; Lpez-Vsquez, K.; Santana, A.D. and Almeida-Val, V.M.F. National Institute for Research in the Amazon Av. Andr Arajo, 2936, Aleixo, 69060-001, Manaus/Am, Brazil Phone/fax 55 92 643-3187. E-mail: luca_evelyn@hotmail.com

EXTENDED ABSTRACT ONLY - DO NOT CITE Introduction

Species living at the Amazon region have their biological cycles synchronized around the flood pulse, dryness and inundation, but it is still unknown how natural populations of continental water bodies face environmental perturbations like those caused by accidental petroleum spill, or similar man interferences. This work aims to evaluate the degree of disturbance of the genetic composition of the populations affected by man activities through genotypic frequency determination of 14 loci and 18 enzymatic alleles in populations of the species Cichla monoculus and Pygocentrus nattereri, which occur along Solimes River, in the region of petroleum transport in the Amazonas state (between Coari and Manaus).

249

Material and Methods

Animals were collected with a waiting net in five points at the work area of PIATAM II: TESOL, ESPERANA II, MATRINCH, BOM JESUS and LAGO PRETO. Fish tissues (muscle, heart, liver, brain and retina) were removed at the collect site, frozen in liquid nitrogen and, afterwards, stocked in a -80C freezer. Samples were homogenized and electrophoresis carried through according to the procedure described in Val et al. (1981). Histochemical staining was made according to well established protocols for fishes. Loci and alleles nomenclature followed respectively Shaklee et al. (1989) and Allendorf & Utter (1978). Gels were photographed and polymorphic loci count and frequency were submitted to appropriate statistical treatment. From electrophoretic studies and isozymic systems it is possible to differentiate genotypes and estimate parameters as genotypic frequencies and allele frequencies, which allowed us to determine gene diversity and heterozigosity coefficients.
Results

We analyzed the distribution of 14 gene loci pertaining to 8 enzymes of C. monoculus and P. nattereri. Polymorphic loci were found in PGI gene family from C. monoculus: PGI-A* - two alleles (A100 and A88) and PGI-B* - two alleles (B100 e B152). While PGI-B* alleles are distributed according to HardyWeinberg equilibrium (2(=0,05; gl=1)=3,36), PGI-A* alleles are not (2(=0,05; gl=1)=7,56). The species P. nattereri presented two polymorphic enzyme systems - PGI and IDH: PGI-B* - alleles B100 e B30 distributed according to the HardyWeinberg equilibrium (2(=0,05; gl=1)=1,55) and the PGI-A* - alleles A100 e A111 not in equilibrium at the analysed populations (2(=0,05; gl=1)=4,16). Locus IDHA* presented three phenotypes - A100, A134 e A100/A134. 2 calculation showed this locus is at genetic equilibrium (2(=0,05; gl=1)=0,71) while locus IDH-B* with the alleles B100 e B121 shows the population is not at Hardy-Weinberg equilibrium (2(=0,05; gl=1)=74,90) presenting significant differences from the expected by genetic equilibrium.

250

Table 1. Calculations of 2 to the locus PGI-A* of C. monoculus

Phenotypes A100/A100 A100/A88 A88/A88 N

O 25 24 24 73

E 18,76 36,49 17,75 73

D 25,74 11,99 5,75

D2/E 1,76 3,94 1,86 7,56

Table 2. Calculations of 2 to the locus IDH-B* of P. nattereri

Phenotypes B100/B100 B100/B121 B121/B121 N

O 104 2 7 113

E 97,59 14,85 0,56 113

D 25,74 11,99 5,75

D2/E 2,29 10,27 62,35 74,90

Conclusion

We may suggest the species studied present a high level of structural gene conservation, maintaining an extremely low level of mean heterozigosity when compared to other fish species. Besides, the allele distribution of locus PGI-B* in the species C. monoculus, IDH-B* and PGI-B* in the species P. nattereri suggest a break at the gene flow between the sampled lakes, which may be related to a bigger fishing effort of these species in some lakes. The low heterozigosity and the lack of distribution according with Hardy-Weinberg suggest a high inbreeding rate in the sampled populations, which reflect a decreasing of the population size in these lakes.

251

Acknowledgements

This work was supported by National Institute for Research in the Amazon (INPA), Federal University of Amazonas, FINEP/PETROBRAS (Project PIATAM) and The National Research Council of Brazil (CNPq), LEAC, LKHC; KLV and ASD are the recipients of fellowships from CNPq/Brasil. VMFAV is a senior research fellow from CNPq/Brasil.
References

Allendorf & Utter (1978) Population genetics. In: W.S.Hoar and D.J.Randall, eds. Fish Physiology, vol. 8: 407-454. Academic Press, New York. Shaklee, J.B.; Allendorf, F.W.; Morizot, D.C. & Whitt, G.S. (1989) Genetic Nomenclature for Protein-Coding Loci in Fish: Proposed Guidelines. Trans. Amer. Fish. Society, 118: 218-227. Val, A.L.; Schwantes, A.R.; Schwantes, M.L.B. & De Luca, P.H. 1981. Amido hidrolisado de milho como suporte eletrofortico. Ciencia e Cultura, 33:737-74.

252

GENETIC VARIABILITY STUDIES OF PIRAMUTABA -BRACHYPLATISTOMA VAILLANTII AND DOURADA - B. ROUSSEAUXII (PIMELODIDAE: SILURIFORMES) IN THE AMAZON: BASIS FOR MANAGEMENT AND CONSERVATION

Jacqueline da Silva Batista Instituto Nacional de Pesquisas da Amaznia INPA Coordenao de Pesquisas em Biologia Aqutica - CPBA Laboratrio Temtico de Biologia Molecular - LTBM Avenida Andre Arajo, n 2936. 69.060-001, Manaus, AM, Brasil Phone: + 55 92 6433382/ Fax: + 55 92 6433347/ e-mail: jac@inpa.gov.br Kyara Formiga-Aquino (COPE/LTBM/INPA) formiga@inpa.gov.br Izeni Pires Farias (ICB/UFAM) izeni@argo.com.br Jos Antnio Alves-Gomes (CPBA/INPA) puraque@inpa.gov.br

EXTENDED ABSTRACT ONLY DO NOT CITE

Piramutaba - Brachyplatystoma vaillantii - and Dourada - B. rousseauxii are the two most important catfish species exploited by commercial and artisanal fishing fleets in the Amazon. These two pimelodid catfish have been intensively captured for decades at fishing grounds distributed over four countries and along the entire length of the Solimes-Amazonas River system, which spans more than 4000 km. These species have a complex and still not completely understood life cycle, including distinct feeding/growing and reproductive areas, separated by several thousand kilometers (Barthem & Goulding, 1997). Together, these two species

253

represent about 70% (>20,000 tons/year) of all catfish commercially caught in the region, forming a major protein source for thousands of families in several Amazonian countries. The comparisons of capture data for different years in Belem indicate a reduction of the yield (JICA/MPEG/IBAMA 1998 Technical Report), but no study has been conducted to evaluate this decrement. In addition to recording any declines in annual harvests, it is also important to determine if the fishing effort is concentrated on a single stock or if several fish stocks are exploited. This study used the complete mitochondrial DNA control region with the main objective to genetically characterize the inter- and intra-specific genetic variability of both species along the main Solimes/Amazonas channel, and to verify the potential existence of one or more stocks for both species. It is not known if over-harvesting has had any impact on the genetic diversity of these two economically important catfishes. This information is necessary to augment conservation and management policies. A muscle tissue from 15 individuals of each species was collected at five localities (Belm, Santarm, Manaus, Tef, and Tabatinga) along the east-towest axis of the Amazon River, covering a linear distance of about 3000 Km. A total of 150 individuals were collected. The control regions of the two species had different sizes; dourada has 911 pb. Piramutaba has 942 pb. Matrices and pair-wise distances were generated in PAUP 4.0 (Swoford, 2001). Analysis of molecular variance (AMOVA) and genetics parameters such as, FST, nucleotide diversity (Pi), haplotype frequency (H), total number of mutations (ETA), polymorphic sites (S), number of singletons (NS) were estimated using Arlequin 2.01 (Schneider et al 2000) and DNAsp 4.0 (Rozas & Rozas 1999) programs. We also calculated a genetic variability index, GVI (number of singletons divided by the total number of individuals sampled). The neutrality tests of Tajima and Fu were applied to examine whether populations are at genetic equilibrium with respect to mtDNA. We estimated the females effective population size (Nef) and the age of the most recent common ancestor of the current population, both based on coalescence theory using Wattersons estimate of the population genetic parameter (S) and assuming a mutation rate of 5 x 10-8 mutations/site/generation. The combined interpretation of the results should give us a comprehensive overview of the distribution of genetic variability of these two species along the Solimes-Amazonas axis.

254

The results are summarized as follows: (1) The genetic variability level for both species was moderate to high. However, the genetic indexes showed higher genetic values in piramutaba compared to dourada in all localities sampled except in Tefe. Figure 1 and Table 1 show comparisons of main genetic patter between the two species.
Esturio
Santarm

Manaus
60 50 40 30 20 10 0

60 50 40 30 20 10 0

60 50 40 30 20 10 0

H
60 50 40 30 20 10 0

NS

S
60 50 40 30 20 10 0

NS

NS

Tef

Tabatinga

Dourada Piramutaba
H NS S

NS

Figure 1. Comparison of genetic parameters between B. vaillantii and B. rousseauxii collected at five localities along the Solimes/Amazonas axis. H = haplotype frequency; NS = number of singletons and S = number of polymorphic sites.

(2) There is no genetic structure in dourada and piramutaba as shown by AMOVA analyzes (p>0.05). Piramutaba and Dourada seems to comprise a single genetic stock along the Amazonas/Solimoes axis; (3) variance effective population size estimated from the population genetic parameter and identity-by-descent F suggests approximately ~139,500 female individuals for Dourada and ~162,300 female individuals for Piramutaba; Table 1. Summary of genetic parameters estimated from the control region of the two catfish species. An asterisk (*) indicates significant differences between the two species. K is the average number of nucleotide differences, Pi is

255

nucleotide diversity, HD is the haplotypes diversity and GVI is the genetic variability index. For each parameter, 95% CI is calculated. Species Piramutaba Belm Dourada Piramutaba Santarm Dourada Piramutaba Manaus Dourada Piramutaba Tef Dourada Piramutaba Tabatinga Dourada Piramutaba All Genetic parameters K Pi (95% CI) (95% CI) 11.009 0.012 (8.326(0.00913.694) 0.015) 9.067 0.010 (6.828(0.00711.305) 0.013) 14.133* 0.015 (10.733(0.01117.534) 0.019) 7.514* 0.008 (5.633(0.0069.396) 0.011) 14.133* 0.015* (10.733(0.01117.533) 0.019) 6.114* 0.007* (4.554(0.0057.674) 0.009) 12.629 0.013 (9.573(0.00915.684) 0.017) 9.943 0.011 (7.504(0.00812.382) 0.014) 14.933* 0.016* (11.349(0.01218.517) 0.019) 6.495* 0.007* (4.848(0.0058.143) 0.009) 13.580* 0.014* (12.184(0.01314.976) 0.016) HD (95% CI) 0.990 (0.9761.004) 0.981 (0.9650.997) 0.981 (0.9650.997) 0.981 (0.9650.997) 1.000* (0.9881.012) 0.914* (0.8860.942) 0.971 (0.9510.991) 1.000 (0.9881.012) 0.971 (0.9510.991) 0.933 (0.9100.956) 0.991* (0.9900.992) GVI (95% CI) 0.840* (0.7620.918)

Localities

256

Dourada

7.770* (6.9428.598)

0.009* (0.0070.009)

0.964* (0.9610.967)

0.520* (0.3580.682)

(4) A coalescent time analysis resulted in a relative age estimate of approximately 242,600 and 335,200 years in Dourada and Piramutaba, respectively. (5) Significant disequilibrium was observed in Fus Fs test but not in Tajimas D test suggesting that both species have been impacted by fishing, however, not severely. (6) The parallel trend suggests that both species are not differentiated in terms of exploitation pattern. Considering the greater age of Piramutaba and thus greater genetic diversity, both species have similar effective population size and similar exploitation patterns. The differences in the genetic variability indexes may reflect more differences in their life histories than any potential differences in anthropogenic patterns of exploitation. While both species still have relatively high level of genetic variability and possibly also relatively large census sizes, both species are beginning to show signs of genetic disequilibrium. This may be the result of over fishing. However, this hypothesis must to be verified by field studies before any firm conclusions may be drawn.
Acknowledgements

The present work was supported by International Foundation for Science-IFS, INPA (grant # 1-3550), Provarzea Project (IBAMA) and CNPq. We thank undergraduate Pirada group on helping with laboratory work, the Pyr group for collecting samples, and Tomas Hrbek for data interpretation.
References

Barthem, R. B.; Goulding, M. 1997. Os bagres balizadores: Ecologia, Migrao e Conservao de peixes amaznicos. Braslia: Sociedade Civil Mamirau, CNPq. JICA/MPEG/IBAMA. 1998. Relatrio Tcnico: The fishery resources study of the Amazon and Tocantins river mouth areas in the Federative Republic of Brazil. Final Report.

257

Swofford, D. L. 2001. PAUP: Phylogenetic Analysis using parsimony, version 4.0. Ilinois Natural History Survey, Champaign. Schneider, S.; Roessli, D.; Excoffier, L. (2000). Arlequin Version 2.000: A software for population genetic data analysis. Laboratrio de gentica e biometria. Universidade de Geneva, Suia. Adquirido de: http:// anthropologie. Unige.ch/arlequin. Rozas, J.; Rozas, R. DnaSP version 4: an integrated program for molecular population genetics and molecular evolution analysis. Bioinformatics. 15, p.174-175, 1999.

258

MITOCHONDRIAL DNA POPULATION STRUCTURE IN WHITE MOUTH CROAKER (MICROPOGONIAS FURNIERI) ALONG THE ATLANTIC COAST OF BRAZIL

A. L. Puchnick1 1Empresa Brasileira de Pesquisa Agropecuria, Embrapa Meio-Norte, UEP de Parnaba, Lab. Biotecnologia Aqutica. Caixa Postal 341, Parnaba, PI 64200-970, Brazil. Phone/Fax 55-86- 3151200; e-mail: angela@cpamn.embrapa.br. J.A. Levy2 2Fundao Universidade Federal do Rio Grande, Departamento de Qumica, Lab. Bioqumica Marinha. CP 474, Rio Grande, RS 96201-900, Brazil.
Abstract

Variation in mitochondrial DNA (mtDNA) was examined among 149 white mouth croakers (Micropogonias furnieri) from six geographic locations in the Atlantic coast of Brazil (1 S 34 S). Restriction fragment length polymorphism (RFLP) analysis of D-loop region detected a total of five composite haplotypes. Heterogeneity tests revealed no evidence of geographic differentiation in mtDNA haplotype frequencies within the region between 23 S and 34 S (P =0.263). However, significant heterogeneity occurred between northern and southern region of 23 S (P <0.003). Average sequence divergence revealed genetic differences in white croaker of the northern versus southern region (= 0.593%, p <0.05). Analysis of molecular variance (AMOVA) indicated relatively high gene flow (Nem = 3-44) among south-central localities, but restricted gene flow (Nem= 1-2) between northern and southern regions. Collectively, these data are consistent with 1) a single stock of M. furnieri within the southern Brazilian coast and 2) separate, weakly differentiated stocks in the northern and the southern Brazil.

259

Introduction

The white mouth croaker (Micropogonias furnieri) is a euryhaline sublittoral sciaenid fish that uses estuarine and coastal waters as nursery and feeding grounds for larvae and juveniles (Isaac, 1988). M. furnieri supports an important demersal fishery along the Southwestern Atlantic. The species is distributed from the Yucatan Peninsula, Gulf of Mexico, at 20 N, to the Gulf of San Matias, Argentina, at 39 S (Chao, 1978), but it is particularly abundant on the southeastern shelf of Brazil (south of 23 S) and the shelf of Uruguay (Figure 1a). Recently, concern has been expressed over the apparent decline in the white mouth croaker fishery along the coast of Brazil, Uruguay and Argentina (Haimovici, 1998). A thorough understanding of white mouth croaker population structure is essential for effective management of the fishery along the South Atlantic Ocean. The status of fish populations north of 23 S has not yet been investigated and the available information on the population structure south of 23 S appears contradictory. Variation in life history patterns and population dynamics suggest the existence of a population in the region between 23 S and 29S (population I) and another between 29 S and 33S (population II) (Vazzoler, 1991). Similar studies suggest that fishes inhabiting Rio de la Plata region probably form a third population (population III) on the coastal waters of Uruguay and that those further south form another (population IV) on the coast of Argentina (for review see Isaac, 1988). Populations II and III off southwestern Atlantic migrate southward (35 S) during summer and northward (27-28 S) during winter according to the seasonal displacements of the subtropical convergence system (Figure 1b). Previous genetic studies of M. furnieri using starch-gel electrophoresis of allozymes do not support the distinctstock hypothesis. A high degree of genetic homogeneity in allele frequencies and also high levels of gene flow were found among white mouth croaker sampled from regions between 23 S and 40 S, and confirmed the hypothesis of a single population unit along the southwestern Atlantic coast (Maggioni et al. 1994; Levy et al., 1998). Recent studies have shown that restriction fragment length polymorphism (RFLP) analysis of mitochondrial DNA (mtDNA) is useful in differentiating populations within several economically important fish species that exhibit little protein variation (Avise, 1994). The purposes of this paper were (1) to document the magnitude and spatial distribution of mtDNA variation in M. furnieri along the Brazilian coast, (2) to test further the hypothesis that white croaker

260

populations are subdivided, and (3) to estimate levels of genetic distance and gene flow among geographic localities.

Figure 1. Distribution of M. furnieri on the Atlantic Ocean. (a) Distribution of assumed white croaker populations on the southwestern Atlantic. Spots indicate collection sites in the coast of Brazil. (b) Distribution of water currents on the South Atlantic Ocean (modified from Stramma and England, 1999). Materials and Methods
Micropogonias furnieri of Brazilian coastal areas were obtained from commercial fishery during late 1999 and early 2000. Freshly caught tissue samples were removed and preserved at 95% ethanol. Dates and locations of capture and the size composition of the collections are presented in Table 1. Sample localities represent the northern (N), southeastern (SE) and southern (S) portions of the species Brazilian coast range, and also northern and southern regions of the hypothesized population boundary. Representative collections of

261

M. furnieri along the Northeastern (NE) coast were scarce.

Table 1. White mouth croaker Micropogonias furnieri collection data. Sample CH RG SC SPR RJ PA Location 33 41 S 53 27 W 32 00 S 52 20 W 29 20 S 49 43 W 25 31 S 48 30 W 22 54 S 43 13 W 0103 S 46 46 W Date 3/2000 12/1999 12/1999 1/2000 2/2000 4/2000 n 16 32 25 37 28 11 Standard length Mean, range (mm) 150, 130200 320, 285330 325, 295360 315, 290340 315, 285325 310, 285320

Genetic population structure in M. furnieri was examined by using polymerase chain reaction (PCR) and restriction fragment length polymorphism analysis (RFLP) of mitochondrial DNA (mtDNA). The studies were conducted at the Laboratrio de Bioqumica Marinha (LBM) Fundao Universidade Federal do Rio Grande (FURG) in Rio Grande, RS. Mitochondrial DNA was obtained by the isolation procedure of Chow and Inowe (1993). Primers L-THR and 12 SAR-H were used in the polymerase chain reaction to amplify a 1450 bp product containing the entire D-loop region, tRNA-Pro and tRNA-Phe genes, and portions of the 12S rRNA and tRNA-Thr genes (Lankford Jr. et al., 1999). Amplifications were performed in 50 l reaction volume containing 1X PCR Buffer, 0.2 mM of dNTP, 2.0 mM of MgCl2, 30 pmol of each primer, 1.5 U of Taq polymerase, and 20 ng of DNA template. PCR reactions were programmed for 35 cycles at 94 C for 1 min, 65 C for 1 min and 72 C for 2 min, including a final extension at 72 C for 8 min in the last cycle. PCR products were digested with the following nine restriction endonucleases: BglI, EcoRI, EcoRV, HaeIII, Hinf I, HpaII, PstI, RsaI, and SacII. Variant RFLP patterns were separated by gel electrophoresis on 1.5 % agarose gels and visualized after ethidium bromide with UV light illumination. Fragment sizes were estimated by using 1 Kb DNA ladder and 174 Hae III molecular weight markers, and UVIdoc software program, ver. 98.01. Distinctive restriction fragment patterns were identified by

262

letter codes and combined to produce composite mtDNA haplotypes for each individual fish. Statistical analyses were performed with the Arlequin Package, ver. 2000 (Schneider et al., 2000). The gene (nucleon) diversity and nucleotide diversity were calculated for each sample and for the pooled samples, according to Nei (1987). Percent mean nucleotide sequence divergences within and among white croaker samples were estimated by the average number of pairwise differences within and between populations (Nei and Li, 1979; Nei, 1987). Phenetic analysis of sample nucleotide sequence divergence matrix was carried out using UPGMA clustering (Sneath and Sokal, 1973). The distribution of haplotype frequencies was evaluated for homogeneity between samples using P-exact test of population differentiation (Raymond and Rousset, 1995) and a total of 10,000 steps in Markov chain. P-values < 0.003 were considered as significantly different, after Bonferroni sequential procedure. Population structure in M. furnieri was also calculated by using a hierarquical analysis of molecular variance (AMOVA, Excoffier et al., 1992). Samples were stratified by locality (PA, RJ, SPR, RSC, RG, and CH) and nested within region (N, SE, and S). Total genetic variation was partitioned into within geographic localities, among geographic localities, and between regions. Significance of these components was tested using 1,000 random permutations. An Euclidian distance matrix between pairs of haplotypes was used for the calculation of FST values as an approximation of F-statistic (Weir and Cockerham, 1984). Gene flow (Nem) among localities was estimated as FST = 2Nem + 1. A matrix correlation (Mantel test) was carried out between the sample genetic distance matrix and a matrix of geographic distances (in miles) between all pairs of samples. Results and Discussion RFLP analysis of the D-loop regions of M. furnieri collected in the Brazilian coast revealed a total of five composite haplotypes (Table 2). Only two of the nine enzymes employed (HinfI and RsaI) produced variant patterns. The digestion profiles of variants were consistent with the hypothesis of single gains or losses of restriction sites.

263

Table 2. Distribution of Micropogonias furnieri mtDNA haplotypes based on restriction endonucleases among different collections. The order of restriction enzyme morphs, represented from left to right, is HinfI and RsaI. Sample Haplotype h1 h2 h3 h4 h5 (AA) (AB) (AC) (BA) (BB) CH 12 0 4 0 0 16 RG 27 1 2 2 0 32 RSC 17 3 1 3 1 25 SPR 33 0 3 1 0 37 RJ 22 0 1 5 0 28 PA 5 0 0 6 0 11 Total 116 4 11 17 1 149

Total

Of the 149 individuals surveyed, 116 (78%) shared the same composite mtDNA haplotype. The common haplotype (h1) was numerically dominant (> 0.70% frequency) at all localities except northern sample (PA, 0.45% frequency). The haplotype 4 occurred at 0.55% frequency in PA, at low (< 0.18%) frequencies in the other samples, and it was absent in CH sample. The haplotype 3 was present at 25% frequency in CH, at very low (< 0.08%) frequencies in the other samples, and it was absent in PA. The haplotype 2 occurred at low (< 0.12%) frequencies in RG and RSC samples, and the haplotype 5 at 0,04% frequency only in RSC sample. Nucleon diversity averaged 0.388 0.097 (mean SE) for the pooled sample, and ranged from 0.203 0.084 in SPR to 0.546 0.072 in PA. Nucleotide diversity also varied geographically, ranging from = 0.086 0.071 in SPR to = 0.216 0.141 in SC. Estimates of mtDNA nucleon and nucleotide sequence diversities indicated that genetic variation in M. furnieri is amongst the lowest than those reported in other marine fish species (Gold and Richardson, 1991, Graves et al., 1992, Lankford Jr. et al. 1999). Percent nucleotide sequence divergences within and among samples are shown in Table 3. Average nucleotide divergence within (1.362%) and among (0.248%) pooled samples indicated that most of the observed mtDNA variation occurred within geographic localities. AMOVA also revealed that the majority (87.24%) of mtDNA variation in M. furnieri occurred within sample localities (p<0.002), but a significant portion (3.72%) was attributable to differences among localities (p=0.047). Variation between regions (9.04%) was unstructured (p=0.112).

264

Table 3. Average number of pairwise differences within population (iX., diagonal elements) and between populations (iXY, above diagonal); corrected average pairwise difference (below diagonal) (Nei and Li, 1979). Bolded elements were estimated as significantly different (p < 0.05). CH RG RSC SPR RJ PA CH 1.600 0.066 0.107 0.056 0.208 1.018 RG 1.375 1.018 0.014 -0.021 0.012 0.636 RSC 1.880 1.496 1.947 0.061 0.020 0.441 SPR 1.243 0.876 1.422 0.775 0.045 0.748 RJ 1.607 1.121 1.593 1.032 1.198 0.313 PA 2.636 1.963 2.233 1.953 1.731 1.636

The distribution and relative mtDNA haplotype frequencies support the general conclusions that white croaker population is weakly subdivided, with semiisolated populations occurring in the north and south region of 23 S. Heterogeneity tests revealed significant differences in haplotype frequencies between PA and SPR samples, and between PA and CH samples (P < 0.003), which represent both limits of the total geographic species range in this study (1 S 34 S). Genetic differentiation, as estimated by average nucleotide divergence, was detected between pooled northern and southern localities (= 0.593%, p < 0.05). Cluster-analysis of mtDNA genetic distances indicated that white croakers from these regions were distinguishable from one another (Figure 2). AMOVA indicated high level of population subdivision (ST = 0.322, p< 0.003) and restricted gene flow (Nm = 1-2 effective female migrants per generation) between pooled northern and southern region. The matrix correlation analysis (Mantel test) revealed a geographic component to the distribution of the white croaker mtDNA haplotypes (correlation value = 0.883, p < 0.003).

265

PA

RSC

RJ CH

0.1

SPR

RG

Figure 2. Phenetic tree showing the genetic distance (Nei and Li, 1979) among white mouth croaker sample localities. The observed genetic heterogeneity between northern and southern Brazilian coast suggests that the number of migrants per generation is not sufficient to preclude genetic divergence of populations by random drift (Slatkin, 1987). The tropical and the subtropical circulation of different water masses within the South Atlantic Ocean may affect white croaker dispersal capability and restrict gene flow between regions. Stramma and England (1999) demonstrated that subtropical South Atlantic is governed by the subtropical gyre with a southward shift of the northern part of the gyre, while the tropical circulation shows several depth-dependent zonal current bands (Figure 1b). In the northern Brazil, the South Equatorial Current (SEC) in the near-surface layer reaches the shelf of Brazil (near 16 S) and separates into the southward flow of Brazil Current (BC) and at subsurface depth into the northward of the North Brazil Undercurrent (NBUC). Near the equator, the SEC overrides the subsurface of the NBUC and forms the surface intensified North Brazil Current (NBC). The southern Brazil is

266

influenced by the Subtropical Convergence system, made up of the confluence zone of southward flow of BC with the Falkland Current (FC). There was, however, no genetic evidence of white croaker population subdivision within the southern Brazilian coast. Frequency- and distance- based analyses both suggested a single, panmictic population of white mouth croaker. The exact test of population differentiation revealed no heterogeneity in mtDNA haplotype frequencies among sample localities (P= 0.263). Low levels of mtDNA divergence among localities (= 0.011%), although not statistically significant, were more consistent with a pattern of semi-isolation by distance rather than marked subdivision by the oceanographic patterns of the subtropical convergence system. Low FST value (ST = 0.008) and high gene flow (Nm = 3-44 effective female migrants per generation) indicated a lack of geographic structure in the southeastern (23 S 29S) and southern (29 S 34 S) regions. These results are consistent with allozyme data reported by Levy et al. (1998), suggesting that the number of migrants per generation among populations is high enough to maintain the homogeneity in gene frequencies and therefore, avoid differentiation by genetic drift. The lack of genetic heterogeneity found within the white croaker in both mtDNA and nuclear gene frequencies is supported by many other studies, despite the geographical variation of morphological and life history characters observed by Vazzoler (1991). Lima et al. (1996) showed that the general form of the circulation in the southern Brazilian shelf can be characterized by a combination of different processes that can restrict population subdivision and present high levels of gene flow in this area. There are, in fact, several aspects of white croaker biology and life history, which should facilitate dispersal and minimize geographic subdivision within south-central area of the Brazilian coast. White croakers are relatively strong swimmers, and adults form large schools offshore and are capable of extensive migration (Vazzoler, 1991). M. furnieri are long-lived pelagic spawners, meaning that individuals could spawn at multiple localities throughout their lifetimes (Isaac, 1988). Combined with these observations, the present mtDNA analysis supports that species dispersal capability results from interaction between physical environment conditions and species ecological requirements, and that life history variation when present, has an ecophenotypic basis. At the level of genetic resolution employed in our study, we cannot disprove the null hypothesis that M. furnieri share a common gene pool. RFLP technique proved sufficiently powerful to detect white croaker population structure in the northern versus southern region of 23 S, but did not reveal structured genetic stocks of white croaker within the southern-central area of the Brazilian coast. The integrity of the hypothesized genetic stock boundary could be tested further by using fine-scale markers such as microsatellites, or direct sequencing of the

267

mitochondrial D-loop region, to provide greater resolution of geographic structure than RFLP analysis. Acknowledgements We are grateful to Coordenao de Aperfeioamento de Pessoal de Nvel Superior CAPES for a Masters degree to A.L.P., and also to Fundao Universidade Federal do Rio Grande FURG, for supporting this research. References Avise, J.C. 1994. Molecular Markers, Natural History and Evolution. Chapman & Hall, New York, 511p. Chao, L.N. 1978. In: W. Fisher (ed.), FAO species identification sheets for fisheries purposes. Western Central Atlantic (fishing area 31), vol. 6. Chow, S., and Inoue, S. 1993. Intra- and interspecific restriction fragment polymorphism in mitochondrial genes of Thunnus tuna species. Bull. Natl. Res. Inst. Far Seas Fish., 30: 207-225. Excoffier, L., Smouse, P.E., and Quattro, J.M. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics, 131: 479-491. Gold, J.R., and Richardson, L.R. 1991. Genetic studies in marine fishes. IV. An analysis of population structure in the red drum (Sciaenops ocellatus) using mitochondrial DNA. Fish. Res., 12: 213-241. Graves, J.E., Mcdowell, J.R., and Jones, M.L. 1992. A genetic analysis of weakfish, Cynoscion regalis stock structure along mid- Atlantic coast. Fish. Bull., 90: 469-475. Haimovici, M. 1998. Present state and perspectives for the southern Brazil shelf demersal fisheries. Fish. Manage. Ecol., 5: 277-289. Isaac, V. J. 1988. Synopsis of biological data on the whitemouth croaker Micropogonias furnieri (Desmarest, 1823). FAO Fisheries Synopsis, 150: 35p.

268

Lankford Jr., T.E., Targett, T.E., and Gaffney, P.M. 1999. Mitochondrial DNA analysis of population structure in the Atlantic croaker, Micropogonias undulatus (Perciformes: Sciaenidae). Fish. Bull., 97: 884-890. Levy, J.A., Maggioni, R., and Conceio, M. B. 1998. Close genetic similarity among populations of the white croaker (Micropogonias furnieri) in the south and south- eastern Brazillian coast. I. Allozyme studies. Fish. Res., 39: 87-94. Lima, I.D., Garcia, A.E., and Mller Jr., O.O. 1996. Ocean surface processes on the southern Brazilian shelf: characterization and seasonal variability. Cont. Shelf Res., 16: 1307-1317. Maggioni, R., Pereira, A.N., Jerez, B., Marins, L.F., Conceio, M.B., and Levy, J.A. 1994. Estudio preliminar de la estructura gentica de la corvina Micropogonias furnieri entre Rio Grande (Brasil) y El Rincn (Argentina). Frente Maritimo, 15 (Sec.A): 127-131. Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press, New York, 512p. Nei, M., and Li, W. 1979. Mathematical model for studying variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. USA, 76 (10): 5269-5273. Raymond, M., and Rousset, F. 1995. An exact test for population differentiation. Evolution, 49: 1280-1283. Schneider, S., Roessli, D., and Excoffier, L. 2000. Arlequin ver. 2000: A software for population genetics data analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland. Slatkin, M. 1987. Gene flow and the geographic structure of natural populations. Science, 236:787-792. Sneath, P.H.A., and Sokal, R.R. 1973. Numerical Taxonomy. W.H. Freeman, San Francisco, CA. Stramma, and England, 1999. On the water masses and mean circulation of the South Atlantic Ocean. J. Geophys. Res., 104(9): 20,863-20,883.

269

Vazzoler, A.E.A. de M. 1991. Sntese de conhecimentos sobre a biologia da corvina, Micropogonias furnieri (Desmarest, 1823), da costa do Brasil. Atlntica, 13 (1): 55-74. Weir, B.S., and Cockerham, C.C. 1984. Estimating F- statistics for the analysis of population structure. Evolution, 38: 1358-1370.

270

PRELIMINARY ANALYSIS OF THE MOLECULAR PHYLOGENY THROUGH MTDNA (12S) IN CAPTURED SPECIES OF GYMNOTIFORMES IN BANKS OF GRASS IN SOME POINTS OF THE CENTRAL AMAZON

Fernandes, Graciene do Socorro Taveira Instituto Nacional de Pesquisas da Amaznia COPE - Laboratrio Temtico de Biologia Molecular Av. Andr Arajo, 2936 - Petrpolis Fone: (**92) 643-3347 E-mail: taveira@inpa.gov.br Bentes-Sousa, Alexandra Regina Instituto Nacional de Pesquisas da Amaznia COPE - Laboratrio Temtico de Biologia Molecular Av. Andr Arajo, 2936 - Petrpolis Fone: (**92) 643-3347 E-mail: alexa@inpa.gov.br The Order Gymnotiformes is a monophyletic group of eletrogenic freshwater fish (Fernandes-Matioli, 2000). This Order comprises a relatively small group of fish endemic to the Neotropical region, whith 28 genera and more than 100 species currently described (Lundeberg et al., 1996; Mago Leccia, 1994). Morphologically, gymnotiforms are quite distinct from other neotropical fish and are easily identifiable. Their representatives possess an elongeted and laterally compressed body shape, dorsal and caudal fins are absent, apteronotids have a reduced caudal fin as an autapomorphic condition. But this group is its ability to generate and detect eletric potentials in water by means of an eletric organ and several kinds of specialized sensory cells (eletroreceptors). In recent years, however, gymnotiform systematics has major adveances, as several authors have sorted out the phylogenetic relationships among and between species using morphological, molecular, neuroanatomical, physiological, and fossil data (Albert & Campos-da-Paz, 2000; Alves-Gomes et al., 1995; MagoLeccia, 1994). Phylogenetic preliminary was based upon the information contained one gene fragments of the mitochondrial genome, the 12S rRNA. Total DNA (genomic and mitochondrial) was extracted by standard

271

phenol:cloroform protocol. To obtain sufficient template for DNA sequencing, the two gene fragments were amplified through the polymerase chain reaction (PCR) method. PCR products 12S rRNA fragments were obtained in reactions with a final volume of 25l [16,15 l Mili-Q water , 2,5 l T10XAB, 2,5 l dNTP at 0.1mM, 0,5 units of AmpliTaq DNA polymerase, 0,8 l each primer at 5mM. PCR amplifications were carriede out in 30 cycles, with desnaturation fro 30s at 93C, annealing for 30s at 54C and extension for 30s at 72C. The quality of the PCR product was tested by running 3 l of the amplified reaction in 0,8% agarose gel.DNA sequences were obtained with a 356 autometed sequencer (MegaBACE 1000) at Laboratrio Temtico de Biologia Aqutica, INPA. For phylogenetic searche of 12S gene were condensed into unique data matrix. Searche basead upon maximum parcimony methods were conducted using test version 4.0d61a of PAUP*. The generated cladogramas (distance and parsimony) demonstrate the monophyletic of the Apteronotidae family (Gymnotiformes). Although the evidences take the monophyletic the levels of Bootstrap show low values between the sorts Apteronotus and Sternachogiton. Maximum parsimony sample that the Apternotus species bonapatti and Sternachogiton schotii, in different are clados confirmed for the low values of Bootstrap. Ahead of the molecular analysis we can infer that these had not corroborated the morphologic ones. The mitochondrils genes can decide phylogeny problems, however suggest studies add for these species, perhaps with a bigger amostral number.
References

Albert, J. S. & R. Campos-da-Paz. 2000. Phylogeny and Classification Fishes. Part 4 Gymnotiformes Alves-Gomes, J. A., G. Orti, M. Haygood. W. Heilinghberg & A. Meyer. 1995. Phylogenetics analysis of South American Eletric Fishes (order Gymnotiformes) and the evolution of their electrogenic system: a synthesis based on morphology, eletrophysiology, and mitochondrial sequence data. Mol. Biol. Evol., 12 (2): 298-318. Fernandes-Matioli, F. M. C.; Matioli, S. R. & Almeida-Toledo, L. F. 2000. Species diversity and geographic distribution of Gymnotus (Pisces: Gymnotiformes) by nuclear (GGAC)n microssatellite analysis. Genetics and Molecular Biology, 23, 4 803-807.

272

Lundberg, J. G., C. C. Fernandes, J. S. Albert & M. Garcia. 1996. Magosternarchus, a new genus with two new species of eletric fishes (Gymnotiformes: Apteronotidae) from the Amazon River Basin, South American. Copeia, 1996 (3): 657-670. Mago-Leccia, F. 1994. Eletric fishes of continental waters of Amrica. Biblioteca de la Academia de Ciencia Fisicas, M. E. N. Clemente editores: Caracas.

273

274

PRELIMINARY STUDY OF THE PHYLOGENY THROUGH mtDNA (12S) OF SILURIFORMES AND GYMNOTIFORMES.

Fernandes, Graciene do Socorro Taveira Instituto Nacional de Pesquisas da Amaznia COPE - Laboratrio Temtico de Biologia Molecular Av.: Andr Arajo, 2936 - Petrpolis Fone: (**92) 643-3347 E-mail: taveira@inpa.gov.br Rapp Py-Daniel, Lcia Helena Instituto Nacional de Pesquisas da Amaznia CPBA - Laboratrio de Ecologia de Peixes Av.: Andr Arajo, 2936 - Petrpolis Fone: (**92) 643-3226 E-mail: rapp@inpa.gov.br Bentes-Sousa, Alexandra Regina Instituto Nacional de Pesquisas da Amaznia COPE - Laboratrio Temtico de Biologia Molecular Av.: Andr Arajo, 2936 - Petrpolis Fone: (**92) 643-3347 E-mail: alexa@inpa.gov.br Of the evolutionary point of view the orders Siluriforms and Gymnotiforms form a monophyletic group (Alves-Gomes, 1995; Fink & Fink, 1981, 1996). One high diversity of species of Siluriforms exists, which present great variety of adaptations and evolutionary strategies, guaranteeing the success of this group in the Amazonia (Alves-Gomes, pess.). As example of this diversity we can cite the variety of sizes, existing species with less than 3 cm of length until those with 2 meters (de Pinna, 1993). In this group it has a great variety of alimentary habits, since generalistas until those highly specialized, as the eaters of scales or teething rings of blood of other species of fish. Some 4,000 more than are migratory species and travel km to multiply (Barthem & Goulding, 1998). Many of these species possess high value commercial, either for the feeding or as ornamental. Especimens of these groups (Gymnotiformes and Siluriformes) can be found in the most varied habitats, and presents specific

275

adaptations for most different environments. In recent years, however, gymnotiform and siluriform systematics has major adveances, as several authors have sorted out the phylogenetic relationships among and between species using morphological, molecular, neuroanatomical, physiological, and fossil data (Albert & Campos-da-Paz, 2000; Alves-Gomes et al., 1995; Mago-Leccia, 1994). Phylogenetic preliminary was based upon the information contained one gene fragments of the mitochondrial genome, the 12S rRNA. Total DNA (genomic and mitochondrial) was extracted by standard phenol:cloroform protocol. To obtain sufficient template for DNA sequencing, the two gene fragments were amplified through the polymerase chain reaction (PCR) method. PCR products 12S rRNA fragments were obtained in reactions with a final volume of 25l [16,15 l Mili-Q water , 2,5 l T10XAB, 2,5 l dNTP at 0.1mM, 0,5 units of AmpliTaq DNA polymerase, 0,8 l each primer at 5mM]. PCR amplifications were carriede out in 30 cycles, with desnaturation fro 30s at 93C, annealing for 30s at 54C and extension for 30s at 72C. The quality of the PCR product was tested by running 3 l of the amplified reaction in 0,8% agarose gel. DNA sequences were obtained with a 356 autometed sequencer (MegaBACE 1000) at Laboratrio Temtico de Biologia Aqutica, INPA. For phylogenetic searche of 12S gene were condensed into unique data matrix. All methods of phylogenetic inference resulted in largely congruent topologies, differing only in minor rearrangements. The best obtained by maximum parsimony. It was obtained using Paup program. The phylogenetic cladograms generated by program PAUP, confirm the morphologic identification and differed the species of Gymnotiforms clearly. Being that the species Petalodoras punctatus the results shows a separation it enters two distinct groups molecularly. While for Ageneiosus brevis clade had been grouped in different, probably an error of identification of the species of this sort occurred, a time that one of the individuals of this species was in clade with others of the same genu. Considering the established difficulties of identification only in morphology the molecular tool, in special the gene 12S, revealed sufficiently efficient in the taxonomics resolution of question.
References

Albert, J. S. & R. Campos-da-Paz. 2000. Phylogeny and Classification Fishes. Part 4 Gymnotiformes Alves-Gomes, J. A., G. Orti, M. Haygood. W. Heilinghberg & A. Meyer. 1995. Phylogenetics analysis of South American Eletric Fishes (order Gymnotiformes) and the evolution of their electrogenic system: a synthesis

276

based on morphology, eletrophysiology, and mitochondrial sequence data. Mol. Biol. Evol., 12 (2): 298-318. de Pinna, M. C. C. de, 1993. Higher-level phylogeny of Siluriformes, with a new classification of the order (Teleostei, Ostariophysi). Ph.D, Disssertation, City University of New york, 482p. Fernandes-Matioli, F. M. C.; Matioli, S. R. & Almeida-Toledo, L. F. 2000. Species diversity and geographic distribution of Gymnotus (Pisces: Gymnotiformes) by nuclear (GGAC)n microssatellite analysis. Genetics and Molecular Biology, 23, 4 803-807. Fink, S. V., and W. L. Fink. 1996. Interrelationships of Ostariophysan Fishes (Teleostei). In M. L. J. Stiassny, L. R. Parenti and G. D. Johnson (eds.), Interrelationships of Fishes. Academic Press, San Diego. Lundberg, J. G., C. C. Fernandes, J. S. Albert & M. Garcia. 1996. Magosternarchus, a new genus with two new species of eletric fishes (Gymnotiformes: Apteronotidae) from the Amazon River Basin, South American. Copeia, 1996 (3): 657-670. Mago-Leccia, F. 1994. Eletric fishes of continental waters of Amrica. Biblioteca de la Academia de Ciencia Fisicas, M. E. N. Clemente editores: Caracas.

277

278

THE METAZOAN FAUNA OF ARAPAIMA GIGAS COLLECTED FROM A FLOODPLAIN AREA IN THE UPPER SOLIMES RIVER

Ana Lcia Silva Gomes Uninilton Lins. Av. Prof. Nilton Lins, 3000. Manaus.AM. Phone. + 55 (92) 643-2103. 643-2107. 248-8396 E-mail. argulus@bol.com.br Jos Celso de Oliveira Malta Instituto Nacional de Pesquisas da Amaznia INPA Laboratrio de Patologia e Parasitologia de Peixes E-mail. jcmalta@inpa.gov.br
EXTENDED ABSTRACT ONLY DO NOT CITE Introduction

Arapaimidae gigas (Cuvier, 1829), known in Bazil as pirarucu, is the only species in its genus and is endemic to the Amazon Basin. Its considered to be a primitive animal, which is very important for the study of the teleostei evolution assessment. Fish parasites, constitute an excellent model for studies on community ecology. The possibility of acquiring numerous parasites and quantifying the totality of these organisms that are distributed in several habitats (infection sites), facilitates the detection of population dynamics and specific relationship patterns. The biology and feeding habits of pirarucu reveal its great potentiality for participating as a definite host in trophically transmitted parasite systems. The present study presents an analysis of the parasite communities of pirarucu, for the purpose of evaluating the swim-bladder and body-cavity parasite

279

infrapopulation dynamics outlining aspect such as parasite indexes and community status.
Material and Methods

Three hundred twenty-two (322) A. gigas specimens from the Mamirau Sustainable Development Reserve in the Amazonas State were necropsied from June 2000 to September 2001. From these the free parasites in the body-cavity and the those present in the swim-bladder.
Results and Discussion

The examined fish measured from 52 to 207 cm in total length and weighed from 1.5 to 90 kg. A total of 1,371 parasite specimens belonging to three different species: Nesolecithus janicki Poche, (1922) and Schizochoerus liguloides Diesing, 1850 (CESTODA) and representatives not yet identified of the Nematoda phylo. Parasite indexes were made according to Margolis et al., (1982) as revised by Bush et al., (1997). From the analysed fishes 61% were infected by a metazoan species at least. The two most prevailing species, N. janicki 33% and S. liguloides 22% were found free in the body-cavity of their host and the nematoides were involved in the swim-bladder. Of all examined fish 34% of them did not present any kind of parasite (Tabela 1). An ecological approach of the parasite communities found in the body-cavity and swim-bladder was made from the components of their infracommunities. Those components were classified according to Bush & Holmes (1996), into central species (present in over two-thirds of the hosts), secondary species (present in one to two thirds of the hosts) and satellite species (present in one third of the hosts at least).

280

Table I . Prevalence (P%), intensity range (IR), mean intensity of infection (MII), mean abundance (MA) and community status (CS) of the parasite metazoans present in the body-cavity and swim-bladder of the pirarucu. Parasite P (%) IR MII MA CS*
Cestoda Nesolecithus janicki 33 0-15 3,1 0,4 S Schizochoerus 22 0-18 8,3 1,2 S liguloides 9 0-111 17,9 2,6 As Nematoda * (C) central species, (S) secundary species, (Sa) satellite species.

According to their prevalence, the Cestoda species were considered to be secondary species and the Nematoda Phylo representatives were considered to be satellite species (Figure 1). Figure 1. Pirarucu swim-bladder and body-cavity parasite metazoans prevalence and abundance indexes.

Prevalence
N. janicki 51,6%

Abundance
3 2,5 2 1,5 1 0,5 0

Nematod S. liguloides14,1% 34 4%

N. janicki S. liguloidesNematoda

Baylis, (1927); Kritsky et al (1985) e Thatcher (1991) studied the A. gigas parasite fauna, being that they all only emphasise the taxonomic aspect. On a

281

qualitative viewpoint Thatcher (1991) found the same parasite taxa found in the present paper, however, quantitatively this is the first study that shows the parasite indexes for the parasite communities that were found. The findings obtained in the present study, point out Nematoda as the main parasite community component in pirarucu (Figure 2).
Acknowledgements

Im indebted to Dr Marle Anglica for making it possible to collect the material in the field and to Mr. Lito Braga Rabelo for his help on the material analyses.
References

Baylis, H. A.. 1927. Some Parasitic Worms From Arapaima Gigas (Teleostean Fish) With A Description Of Philometra Senticosa N. Sp. (Filarioidea). Parasitology, V.19 P. 35-47. Bush, A.O.; Lafferty, K.D.; Lotz, J.M.; Shostak, A.W. 1997. Parasitology Meets Ecology On Its Own Terms: Margolis Et Al. Revisited. Journal Of Parasitology, Santa Barbara , V. 83, N. 4, P. 575-583. Bush, A.O & Holmes, J.C 1986. Intestinal Helminthsof Lesser Scaup Ducks: An Interactive Community. Canadian Journal Zoology. 64: 142-152. Kritsky, D.C; Boeger, A. & .Thatcher, V.E. 1985. Neotropical Monogenea. 7. Parasites Of The Pirarucu, Arapaima Gigas (Cuvier), With Descriptions Of Two New Species And Redescription Of Dawestrema Cycloancistrium Price And Nowlin, 1967 (Dactylogyridae: Ancyrocephalinae). Proc. Biol. Soc. Wash.. V.98, N.2, P.321-331. Thatcher, V.E. 1991. Amazon Fish Parasites. Amazoniana,v.11, p.263-572 .

282

CHARACTERIZATION AND ANALYSIS OF EXPRESSED SEQUENCE TAGS (EST) FROM HYPOPHISIS AND BRAIN OF COLOSSOMA MACROPOMUM (CHARACIFORMES: SERRASALMINAE).

Alexandra Regina Bentes-Sousa INPA-COPE, Molecular Biology Lab, Manaus-AM, Brazil. jirporto@inpa.gov.br Enedina N. Assuno UFAM-CAM, Manaus-AM, Brazil Ldia M.P. Moraes UNB, Braslia-DF, Brazil Spartaco Astolfi-Filho UFAM-CAM, Manaus-AM, Brazil Jorge Ivan Rebelo Porto INPA-CPBA, Fish Genetics Lab, Manaus-AM, Brazil. jirporto@inpa.gov.br
EXTENDED ABSTRACT ONLY DO NOT CITE

Tambaqui (Colossoma macropomum), one of the largest fish species from the Amazon rivers, is economically very important either as a natural or hatcheryreared fish. Domestication of tambaqui started in the 50s and since then its culture has grown faster, including in the Amazon. Despite the fifty years or so of aquaculture of tambaqui its genetic improvement programs are totally neglected. Only recently a basic question related to the estimation of its genetic diversity along the Amazon basin was achieved, in this case using both nuclear and mitochondrial DNA. Taking into account our participation in the Brazilian National Genome Research Consortium we developed competence in the field of Genomics and now we intent to apply them in several scientific questions.

283

Thus, we aimed to construct a cDNA library of hypophysis and brain of an Amazonian fish species to analyse their expressed sequence tags (ESTs) and to contribute to the advancement of molecular genetics in the aquaculture mainly the genetic improvement of Amazonian fish stocks. mRNA was isolated using two steps. First, total RNA was isolated from tissues or cells using the TRIzol Reagent and the pellet was dissolved in DEPCtreated and autoclaved milli-Q water then stored at 70. Second, Poly (A)+ RNA was isolated from total RNA using the Fast Track 2.0 Kit (Invitrogen). The library was constructed by using SuperScript Plasmid System with Gateway Technology (Invitrogen). Complementary DNA was constructed from mRNA using a primer consisting of a poly (dT) sequence with a Not I restriction site. Sal I adapters were ligated to the blunt-ended cDNA fragments followed by Not I digestion. The cDNA fragments were fractionated and cloned into the Not I-Sal I restriction site of the plasmid vector pCMV SPORT6 (Invitrogen). The ligated cDNA fragments were transformed into E. coli bacteria and the number of primary recombinants was determined. Pure plasmid DNA was isolated from bacterial colonies by the lysis miniprep method, and the cDNA inserts sequenced in automated DNA on the MegaBACE (Amersham Biosciences) sequencer using DYEnamic ET dye terminator Kit (Amersham Biosciencies) by using M13 sequencing universal primers. Partial cDNA libraries generated an average read lenghth of 200-600 pb, sequences of each clone was compared with nucleotide and protein sequence databases of GenBank (NBCI- National Centre for Biotechnology Information) using BLAST algorithm. Overall, 200 ESTs were generated from ~500 clones. This preliminary survey generated putative sequences (ESTs) assigned to several organisms including teleosts, zebrafish Danio rerio and catfish Ictalurus punctatus, as well as plants and humans. Since our research goal was to provide the first step to gather scientific and technological information for improving the tambaqui broodstocks, this powerful strategy and the EST data obtained thus far provides the first data for a Characiform fish species and many implications related to the development of aquaculture genetics, improvement of a regional fish genomics, and genetic engineering using beneficial genes is expected for the near future.
Acknowledgements

We thank San Diegos fish hatchery for providing fish specimens.

284

ECONOMIC INCENTIVES, DIRECT CONTROLS AND ECOSYSTEM MANAGEMENT OF FISHERIES: OUT WITH THE OLD AND IN WITH THE NEW

James R. Kahn, Ph.D Director, Environmental Studies Program John F. Hendon Professor of Economics Washington and Lee University and Professor Colaborador Centro do Cincias do Ambiente Universidade Federal do Amazonas kahnj@wlu.edu
Abstract:

ITQs, non-transferable quotas and other regulatory mechanisms have failed dismally in managing and protecting fishery resources. The vast majority of oceanic fisheries are in a state of collapse, as species-by-species regulation of catch had not been effective. This paper looks at a new set of economic incentives that are capable of dealing with catch issues, but going beyond to consider species inter-relationships, incidental catch, habitat destruction and other fishing externalities. Policies that are evaluated include ecosystem reserves (marine or freshwater), fishery recovery or protection bonds, concession agreements, and performance bonds, among other policies.
Introduction

In a world of environmental catastrophes, the collapse of oceanic ecosystems stands out as a monumental failure of policy. Despite decades of regulatory oversight, the productivity of oceanic ecosystems continues to decline as the population of species after species collapses. Similar problems exist in estuarine and freshwater fisheries.

285

A convincing case can be made that the failure of fisheries management is due to quotas that are too high to be sustainable, illegal and unreported catch, and the simply unwillingness of fishing nations to enforce politically unpopular restrictions that limit catch. However, the lack of limitation of catch is only one factor in the collapse of fisheries. This paper will argue that both the underlying goals of fishery management and the methods of fishery management are fundamentally flawed because they do not recognize that fish are part of ecosystems. Impacts on a target species have impacts on the whole ecosystem, as do additional types of damages caused by fishing such as by-catch and destruction of bottom habitat. Fishery management has focused on addressing what Tietenberg refers to as the contemporaneous externality and the inter-temporal externality, where both are forms of open-access externalities. The contemporaneous externality refers to the problem of an inefficiently high level of effort devoted to catching a given level of fish, and the intertemporal externality refers to catching an inefficiently high level of fish relative to the availability of fish in the future. A series of policies have been implemented over time, including both direct controls and economic incentives. Direct controls focus mostly on the intertemporal externality and include restrictions on how many fish can be caught, when they can be caught, where they can be caught and how they can be caught. Economic incentives such as individual transferable quotas limit either the number of fish that can be caught or the number of boats in the fishery, and also address the contemporaneous externality by limiting entry into the fishery. Despite the implementation of complex restrictions on fishing activity, fishery management has failed. There are two primary sets of reasons for this failure, the first having to do with the targets of regulation and the second having to do with the actual implementation of the regulation. The targets of regulatory efforts have been less than ambitious. Political considerations and lobbying efforts have lead to the setting of quotas which are much higher than the population can realistically support. Fishermen and fishing nations are constantly arguing that fish stocks are actually much greater than scientists estimate, so quotas can be correspondingly larger. Even though there is often no scientific evidence to support their claims, quotas are increased unless the scientists can prove that the populations are smaller than the fishing proponents claim. The onus is on scientists to prove that larger quotas are

286

destructive, rather than the burden of proof being placed on the fishing industry to show that larger quotas are benign.
Fishery management and the target level of catch

Even if we have good estimates of fish stocks, our setting of quotas is process that is conceptually flawed. We have relied on the goal of maximum sustainable yield as a target of fishery management, but if catch is only a small increment above maximum sustainable yield, it will begin the population on a downward spiral. Moreover, even if catch is equal to maximum sustainable yield for a typical season, natural shocks could cause the growth function to shift downward, making what was previously a maximum sustainable yield to become a collapse-inducing level of catch. In fact, although all the population-catch combinations on the equilibrium catch function in Figure 1 are equilibrium catches, only those combinations with population levels greater than P1 are stable equilibria. If population levels are to the left of P1, a shock that temporarily reduces population combined with an unchanged level of catch will cause the population to collapse towards zero. In other words, if the equilibrium consists of (P2, C2) and a shock reduces population to P3 and catch remains at C2, then the growth associated P3 is only G3 (which is less than C2), causing population to further decline. If catch continues to exceed the equilibrium level for the lower population, then population will continue to move towards zero. If the equilibrium catch function is not logistic in nature and has an inflection point to the left of P1 (as in Figure 2) this movement towards collapse can be rapid. If the function has a minimum viable population level (as in Figure 3) the collapse may be irreversible.

287

Figure 1: Equilibrium Catch Function catch

C1

P1

Fish population

The critical point here is that maximum sustainable yield should not be a target of management, because its pursuit can lead to collapse. In the fisheries economics literature, it is also emphasized that maximum sustainable yield should not be the target of management, but for another reason. It is argued that an equilibrium catch should be chosen to maximize the present value of the stream of income. However, it is likely that this catch-population combination could be a point on the equilibrium catch function to

288

Figure 2: Depensated Equilibrium Catch Function catch

C1

P1

Fish population

289

Figure 3: Critically Depensated Equilibrium Catch Function catch

C1

Minimum viable population

P1

Fish population

the left of P1. This makes the fishery even more likely to enter a regime characterized by population collapse. Therefore, a guiding principle for choosing a target level of catch must be the precautionary principle. In this case, the precautionary principle means that the population of the fish stock must be greater than the population level associated with maximum sustainable yield and the catch level must be less than maximum sustainable yield. This proposition will be examined in greater detail in the Section 2. Additionally, an important shortcoming of fisheries policy is that is has focused only on regulating the catch of fish on a species-by-species basis. This assumes that the growth function of one species is exogenously determined, and the only choices are where to move along the function. In fact, the growth function is endogenous to fishing activity for two reasons. First, different fish populations are related through a complex set of ecological relationships. Although some conceptual models of fishery economics explicitly recognize predator-prey relationships, fishing activity has impacts on the ecosystem independent of the removal of a particular species or set of related species.

290

Ecosystem Impacts

In addition to the contemporaneous and intertemporal externalities, there are a host of environmental externalities associated with fishing effort that must be addressed by policy. These include: Ecological ties across species By-catch Destruction of habitat through contact with fishing gear Since ecological relationships can be quite complex and far-reaching, it is difficult to predict the total ecosystem impacts associated with removing a large proportion of the biomass of a particular species, particularly when the species may be an apex predator (e.g. tuna, sailfish, swordfish or shark) or an important forage species (e.g. menhaden, anchovies, shrimp, squid or sardines). These ecosystem effects may be slight for relatively low levels of catch, but nonlinearities in ecological relationships and the existence of thresholds or bifurcations may lead to collapse for small additional increments in catch. This is illustrated in Figure 4, where the health of the ecosystem is measured along the vertical axis, and the population of the target species is represented along the horizontal axis. Since ecosystem collapse occurs in the vicinity of Pc , the precautionary principle would indicate that the target level of catch should be at a level to keep population above Pc , with an adequate margin of safety. Of course, if Pc is less than the maximum sustainable yield level of population, the goal of fishery management should be to keep population greater than this level, rather than Pc.

291

Figure 4. Ecosystem health and population levels

Ecosystem Health

Pc

Population level of a given species

By-catch is also known as incidental catch, but its consequences are hardly incidental. By-catch is associated with many different fishing technologies, including purse seines, gill nets, long-lines and trawling. An example of the significance of the impact of by-catch can be found by looking at trawling for shrimp. In addition to the well-known impacts on marine turtles, the shrimping activity kills an incredibly large number of fish, both adult and juvenile. For the purpose of analysis, we will focus on the impacts on juvenile species. First, even if the fish that is being killed is not a commercially important species, it could have impacts on the ecosystem, of the type diagrammed in Figure 4. Additional policy implications occur if the juvenile fish that is being killed is also commercially harvested as an adult. Although there may be some compensating reductions in natural mortality associated with the by-catch kill, the by-catch will have an impact on the population dynamics of the species. This is illustrated in Figure 5, where the removal of a large proportion of the juvenile population implies that there will be less growth associated with a given adult population, shifting the equilibrium

292

catch function downward. The more severe the depletion of the juvenile fish, the more severe the downward reduction in the equilibrium catch equation. The interaction of the downward shift in the equilibrium catch function with policy is quite significant. If policy is made without taking into account the impact of the depletion of juvenile fish through by-catch, then policy will be made with the idea that the equilibrium catch function more closely approximates E1 than E2. Of course, if this is the case, then what is thought to be an equilibrium catch will in actuality be an excess catch and the population will crash towards zero. Environmental impacts from fishing gear would have the same type of impact as Figure 5: The impact of by-catch on population dynamics

Equilibrium Catch

E2

E1 Population

illustrated in Figure 5. Disruption of bottom structure, benthic communities, submerged aquatic vegetation and coral reefs through contact with fishing gear would make shift the equilibrium catch equation as in Figure 5, with a diminution of the carrying capacity and a reduction in the amount of growth (and thereby the equilibrium catch) associated with a given population level.

293

Policy Recommendations

Based on the analysis presented above, fishery regulation and fishery policy must deal with the following set of issues, as whole: Establishment of an appropriate target level of harvest, taking into account the precautionary principle. Elimination of illegal and unreported catch and other cheating and enforcement issues Ecosystem impacts through overly aggressive harvesting policies, bycatch and disruption of ecosystem habitat. Although individual transferable quotas are often touted as a way of limiting entry and protecting the fish stock, they simple cannot deal with all these issues. The two major problems associated with ITQs are that they are not successful in restricting catch (because of high-grading, off-loading and other forms of cheating) and they simply do not address the broader ecosystem impacts of fishing effort. Enforcement is virtually non-existent, as there is no political will to do so in developed countries, and in developing countries there is the same problem coupled with a lack of enforcement resources. However, a wide range of economic incentives and command and control techniques are available to deal with both catch restriction and the more far-reaching consequences of fishing effort. Economic incentives include: Fishery Recovery Bonds Performance Bonds Leasing of fishery areas to sole operators Important command and control techniques include: Prohibition of fishing techniques that generate by-catch Establishment of marine reserves (and similar areas in freshwater) Increased enforcement and penalties for illegal fishing, illegal offloading of catch, high-grading and unreported catch An immediate question comes to mind and that is why these techniques have the potential to be successful, whereas traditional command and control techniques or ITQs are not successful. The answer is that the techniques suggested in the above two sections of bullets can treat the fishery as an integrated ecological community and do not just attempt to regulate the catch of a particular fish species.

294

For example, fishery recovery bonds have been suggested by Steve Sloan as a means of restoring a degraded fishery. The basic concept underlying fishery recovery bonds is to pay the fishermen not to fish in order to allow a depleted and degraded fishery to recover. The first step is the creation of the fund, which could be established by an international agency (e.g. FAO, UNEP, World Bank or regional development bank), a national government (if the fishery is freshwater or contained entirely in the Exclusionary Economic Zone) or a subnational government. An additional possibility is that a private corporation or group of corporations establish the fund in exchange for either future payments (from higher fishery income after the fishery recovers) or for exclusive rights to sustainably operate the fishery in the future. Of course, it could be established as a public investment opportunity sold on the stock market where shareholders would benefit from future fishery income. The fishery recovery bond would need to generate income to provide a sufficient level of income for the fishers who will have to cease fishing until the fishery recovers. As Sloan stresses, an important determinant of the sufficiency of income is that it be high enough to maintain the payments on the fisher's boat. This is important because the fisher's house is often collateral for the boat. If he or she cannot make the payments on the boat, the house is lost as well. In terms of the diagrams presented earlier in this paper, a fishery recovery bond would move the equilibrium catch/population relationship rightward along the equilibrium catch function. In addition, since other types of ecosystem damages could have time to heal, it could shift the equilibrium catch function up and to the right. Contracts to lease exclusive rights to exploit a resource in a sustainable fashion have been discussed at length in the tropical forestry literature. Kahn has shown how a combination of performance bonds and contractual leasing of harvesting rights can be designed with the potential to generate sustainable forestry and protection of forest ecosystems. The same type of system can be used to generate sustainable fisheries. The basic premise behind such a system is that firms pay for the right to sustainably harvest, but must follow restrictions to protect the ecosystem. Such restrictions take the form of limitations on the volume of the stock that is harvested in each time period. In addition, restrictions on how the stock is harvested serve to prevent collateral ecosystem damages. For example,

295

restrictions could be placed on gear to limit by-catch and to limit impacts on bottom habitat. Fishery territory could be worked on a rotational basis, including fallow periods where no harvesting activity takes place. Economic incentives are structured to make sure that both types of restrictions are followed. The first type of economic incentives would be performance bonds. A sum of money is placed in an escrow account and subject to forfeiture. If firms are in compliance with the environmental restrictions or harvest restrictions in the current period, the bond is not forfeited and they are allowed to move into the next period of the contractual lease. If firms are not in compliance, they forfeit the performance bond and lose the right to continue their leased harvesting rights. Direct controls are likely to be less effective, as has been demonstrated in the pass. The exception to this is the creation of marine reserves. Marine reserves operate in a similar vein to fishery recover bonds, in that they eliminate both harvest and disturbance in a given area. More importantly, they allow the existence of an area where fish stocks and the marine environment can remain undisturbed and serve as a site where biomass can be grow and disperse to nonreserve areas. Marine reserves should be used in conjunction with other policies. In particular, marine reserves could be used in combination with a leasing/performance bonding system as diagrammed in Figure 6. The marine reserve serves as the core of the fishery management area, and absolutely no disturbances are allowed in this area. The marine reserve could be surrounded by a catch and release sport fishing area. In addition to the catch and release restrictions, there should be other restrictions such as absolutely no bottom contact (no anchoring). The catch and release area could then be surrounded by a leasing area for sustainable commercial fishing, where different parts of the sustainable fishing area are left fallow for periods of time.

296

Sustainable Fishery area

Catch and release reserve

Marine reserve

297

Conclusions

It is clear that current fishery policy is ineffectively constructed and ineffectively implemented. A fundamental part of this problem is that it focuses on regulating catch on a species-by-species basis, and ignores important ecosystem-wide effects. Additionally, current regulations are inadequately enforced, with cheating occurring on a massive scale. In areas where they have been applied, individual transferable quotas have not led to fishery recovery. The development of innovative policies can aid in the process of fishery recovery. Fishery bonds, performance bond and leasing systems and marine reserves can be critical parts of a reconceived approach to fisheries. There are many issues associated with the design and implementation of these innovative policies, which need the attention of fishery and policy analysts.
References

Anderson, Lee 1986. The Economics of Fisheries Management Baltimore: Johns Hopkins University Press, Batabyal, Amit, James Kahn and Bob ONeill, 2003. On the scarcity value ecosystem services Journal of Environmental Economics and Management 46:334-352, Kahn, James 2002. The Development of Markets and Market Incentives for Sustainable Forestry: Application to the Brazilian Amazon, Organization for Economic Cooperation and Development, Environment Directorate, ENV/EPOC/GSP/BIO/(2001/6)/FINAL, 46 pages, 28 March Kahn, James, 2005. The Economic Approach to Environmental and Natural Resources, 3rd edition, New York: Southwestern Publishing, Sloan, Stephen 2003. Ocean Bankruptcy, Tietenberg, Thomas 2002. Environmental and Natural Resource Economics, New York: Addison-Wesley,

298

THE ROLE OF SALINITY IN THE DISTRIBUTION OF FISH ASSEMBLAGES IN THE MAIN CHANNEL OF PARANAGU ESTUARY SOUTH BRAZIL (TROPICAL/SUB-TROPICAL TRANSITION ZONE).

Mrio Barletta Laboratory of Ecology and Management of Estuarine and Coastal Ecosystems, Departamento de Oceanografia (DOCEAN)- UFPE, Cidade Universitria 50740-550, Recife, Pernambuco, Brazil. E-mail: barletta@ufpe.br Ulrich Saint-Paul Zentrun fr Marine Tropenkologie ZMT. Fahrenheidstr. 06, D-28359 Bremen, Germany Camila S. Amaral M. Phil. Student of Oceanography Post-Graduation Program DOCEAN/UFPE Adriana C.R. Couto M. Phil. Student of Oceanography Post-Graduation Program DOCEAN/UFPE
EXTENDED ABSTRACT ONLY DO NOT CITE Abstract

This study investigates the seasonal changes of the fish species composition in relation to biomass, density and number of species in three areas of the main channel of the Paranagu Estuary. Between June/2000 and July/2001 237 samples were taken in the main channel of Paranagu Estuary. During this time 81 species and 65,000 fish weighing 2 tons were captured. The total mean density and biomass were estimated for this habitat (1513 ind. ha-1; 34 kg ha-1). Analysis of the catch data showed that in the Paranagu Estuary the number of species and total mean density differed significantly amongst areas and seasons. However, the total mean biomass differed significantly only amongst areas.

299

Most of this differences were detected during the late rainy season. During this time, the fishes species concentrated in the middle of estuary. In the Paranagu Estuary the salinity is stable in the middle- and lower estuary even during the rainy season. For that reason independently from season the estuarine resident fish species remain in the estuary.
Introduction

Several authors have emphasised the importance of estuaries for marine fisheries by demonstrating that a large part of the landings around the world is made up of species that spend part of their lives in estuarine waters. Estuarine areas and shallow waters in the tropics-, sub-tropics- and temperate regions considered as nurseries for invertebrate and fish species (Blaber 2000). Studies in estuarine habitats in south Florida, Mexico, Solomon Islands and Australia have described fish assemblages distribution and structure in relation to seasonal variations in species number, biomass and density and discussed their importance as nursery areas. Along the Brazilian coast, studies of fish assemblages composition and seasonal variation in estuarine ecosystems and adjoining marine waters were undertaken in tropical North Brazilian in the Amazon River estuary, Caet Estuary and in the warm temperate and sub-tropical South and Southeast. However, studies about fish assemblage composition in relation to seasonal variation of species numbers, density and biomass as well as fish migration in the Paranagu Estuary (South Brazil) are still poorly understood. The main question addressed in the present study was: How does the fish assemblage (number of species, density and biomass) vary in relation to seasonal fluctuations of salinity in the main channel of Paranagu Estuary?
Materials and Methods

Paranagu Estuary is the largest estuary on the south Brazilian coast (2515 25 35 S and 48 20 - 48 45 W). The main channel of the east/west main axe of the estuary was divided into three areas (upper, middle and lower estuary) according to the salinity gradient.

300

Water temperature and salinity were recorded before samples were taken. In the three areas of the estuary, samples were taken with an otter trawl net (Barletta et al., 2004). Six monthly replicate samples were collected during first quarter moon between June/2000 and July/2001 in the upper, middle and lower estuary. All caches are expressed in density (ind. ha 1) and biomass (g ha 1). Two-way ANOVA was used to test differences in fish community parameters (density, biomass, and number of species) among areas (upper-, middle- and lower- estuary) and seasons (early dry-, late dry-, early rainy- and late rainyseason). Cochrans test was used to check the homogeneity of variances. Tukeys HSD test (p<0,05) was used whenever significant differences were detected.
Results

Salinities showed a seasonal trend, principally in the upper estuary. At the beginning of the rainy season (January/2001) salinity values decreased in the upper estuary. The lowest values (0 12) were observed between January and March 2001. After this time, rainfall decreased and salinities rose again. Independently from season, the upper estuary showed always the lowest salinity (0 16) and the lower estuary the highest values (> 25). Water temperatures showed the same seasonal trends in all estuary. Two hundred and thirty four samples were collected representing a total sampled area of 541,375 m2. Approximately 60,000 fish weighing 2 tons and consisting of 81 species of 29 families were collected from the axis east-west of Paranagu Estuary. The absolute mean density and biomass estimated from all samples was 1,513 ind. ha-1 and 34 kg ha-1 respectively. The upper estuary showed the highest mean values of density (2,147 ind ha-1), and the middle estuary showed the highest mean values of biomass (49.5 kg ha-1) (Table 1). The estuarine species Cathorops spixii, Stellifer rastrifer, and Anchoa parva ,comprised 70% of the total density and 65% of the total biomass. Table 1. Density and biomass for the most important species of Paranagu Estuary (upper, middle and lower estuary). The results are expressed in individuals and grams per hectare (ha).

301

Species Cathorops spixii Stellifer rastrifer Anchoa parva Total no. of species sampled area (m2)

Density -1 ind. ha 686,39 288,73 85,81 1512,79 83 541375

% 45,37 19,09 5,67

Biomass -1 g ha 20145,58 1659,73 55,01 33836,67

% 59,54 4,91 0,16

Upper 29,74 32,35 7,30

Density (%) Middle Lower 72,62 8,51 0,59 1,03 2,27 20,05

Upper 61,28 9,81 0,31

Biomass (%) Middle Lower 82,89 3,03 0,05 0,60 1,86 0,29

2146,60 1981,44 450,72 63 68 62 179614 174214 187547

32774,42 49503,55 14007,00

Two-way ANOVA showed that the mean number of species and total density differed significantly amongst seasons and areas (Table 2). The variable biomass has shown significant differences only for the factor area. Table 2. Summary of the ANOVA test results for number of species and total density and biomass. Analysis performed on log (x+1)-transformed data. Differences among areas and season were determined by Tukey`s HSD test post hoc comparisons. Where *, p < 0.05; **p < 0.01; NS, not significant; EDS, early dry season; LDS, late dry season; ERS, early rainy season; LRS, late rainy season; UE, upper estuary; ME, middle estuary; LE, lower estuary.
Parameters Number of species Source of Season (1) Variance Area (2) Interactions

** EDs LDs LRs ** LDs Ers LRs NS

** ERs (UE ~ ME) > LE

NS

Density (ind. m-2)

** EDs (ME ~ UE)>LE

NS

Biomass (g m-2)

** (ME ~ UE) >LE

NS

302

Discussion

In spite of the high number of species, the fish assemblage in this estuary is composed of a few species only. In addition, among the upper, middle- and lower estuary significant differences were observed for the number of fish species (season and area), density (season and area) and biomass (area), resulting in a rejection of the null hypothesis. Rain Paranagu Estuary falls mainly from January to May and this is reflected as higher river discharge into the estuary. During the peak of the rainy season (March/April) the Paranagu Estuary principally the upper estuary was flushed . However, the middle- and lower estuary keep the salinity values stable. For that reason, most of the estuarine fish species (adult and young-of-the-year) concentrated in the middle of estuary. It explain the significant differences among season and areas for variables number of species and density. According to Barletta et al. ( 2003 and 2004) and Barletta-Bergan et al. (2002) the Caet Estuary was a totally flushed system at the end of the early rainy season and beginning of the late rainy season and at this time the fish assemblage moved downstream to inshore areas. The authors, based in their and other studies realized in other estuaries in the equatorial region, suggested that this behaviour may be generalized for estuaries in northern South America. Blaber (2000) suggest that, this phenomenon occurs on a different scale in tropic and sub-tropic estuaries and it is triggered by seasonal variations of the salinity gradient. In the Paranagu Estuary the salinity is stable in the middle- and lower estuary even during the rainy season. For that reason independently from season the estuarine resident fish species remain in the estuary. Barletta et al. (2004) and this study agree with this theory, and add that the seasonal changes of estuarine fish assemblages may be determined by a combination of temporal fluctuations of the fish species abundance induced by rainfall, reproduction and recruitment of estuarine species and the recruitment of marine and freshwater species.
Acknowledgements

303

This work resulted from the cooperation between the Centre for Tropical Marine Ecology (ZMT) Bremen, Germany and the Centro de Estudos do Mar (CEMUFPR), Pontal do Paran - PR, Brazil under the Governmental Agreement on Cooperation in the Field of Scientific Research and Technological Development between Germany and Brazil financed by the German Ministry for Education, Science, Research and Technology (BMBf) [Project number: 03F0154A, Mangrove Management and Dynamics - MADAM] and the Conselho Nacional de Desenvolvimento Cientfico e Tecnolgico (CNPq - Brazil)". The first author was supported by CNPq grant (Nr. 30041900-NV)
References

Barletta, M., Barletta-Bergan, A., Saint-Paul, U. & Hubold G (2003) Seasonal changes in density, biomass, and diversity of estuarine fishes in tidal mangrove creeks of the lower Caet Estuary (northern Brazilian coast, east Amazon). Marine Ecology and Progress Series, 256:217-228. available online at www. Int-res.com Barletta, M., Barletta-Bergan, A., Saint-Paul, U. & Hubold G (2004) The role of salinity in structuring the fish assemblages in a tropical estuary (Caet River East Amazon Brazil). Journal of Fish Biology, in press. Barletta-Bergan. A., Barletta, M. & Saint-Paul, U. (2002) Structure and seasonal dynamics of larval and juvenile fish in the mangrove-fringed estuary of the Rio Caet in North Brazil. Estuarine, Coastal and Shelf Science, 56, doi: 10.1006/ecss.2001.0842, available online at http://www.idealibrary.com Blaber, S.J.M. (2000) Tropical estuarine fishes: ecology, exploitation & conservation. Blackwell, Oxford

304