PS 60-23
Historical deforestation effects on stream fish assemblages in western Amazon

Friday, August 15, 2014
Exhibit Hall, Sacramento Convention Center
Gabriel L. Brejão, Department of Zoology and Botany, Sao Paulo State University, São José do Rio Preto, Brazil
Lilian Casatti, Department of Zoology and Botany, Sao Paulo State University, São José do Rio Preto, Brazil
María A. Pérez-Mayorga, Escuela de Ciencias Biológicas, Unidad de Ecología en Sistemas Acuáticos - UDESA, Universidad Pedagógica y Tecnológica de Colombia - UPTC, Tunja, Colombia
Carolina R. Bordignon, Sao Paulo State University, Sao Jose do Rio Preto, Brazil
Silvio F. B. Ferraz, Forestry Sciences, University of São Paulo, Piracicaba, Brazil
Background/Question/Methods

In tropical rainforest, historical deforestation pattern affects many ecological processes that occur in streams. There are some evidences about the importance of land use in the past as a determinant of streams species diversity observed in the present time. We sampled 75 stream reaches, selected due to their temporal deforestation patterns. Using land-use maps for 1984, 1999 and 2011, deforestation pattern was assessed by Forest Change Curvature Profile (FCCP), calculated by LUCAT (Land-Use Change Analysis Tools). We determined functional traits for each species (with abundance higher than five) according to habitat use (morphological traits). Functional traits were used to obtain the net relatedness index (NRI). If NRI > 0, assemblages are functionally clustered; if NRI < 0, assemblages are overdispersed. Multiple regression analyses were conducted to find the relation between forest dynamics and species richness (S) and NRI. 

Results/Conclusions

Regarding historical deforestation pattern effects, we found a significant relationship between FCCP and species richness (F = 6.199, t = -2.49, P < 0.0152), and between FCCP and NRI (F = 11.334, t = 3.37, P = 0.0013). When we separated fish species according to their habitat use, we found that generalist species tend to respond to long-term deforestation. Thus, nowadays, higher richness of generalists were registered in watersheds that were deforested around 1984 (F = 11.458, t = 7.308, P = 0.0012). Specialists, on the other hand, tend to be less rich in watersheds in which deforestation is recent (e.g. 1999 and 2011), occurring mainly at preserved areas (F = 6.837, t = 10.618, P = 0.0109 for 1999; and F = 12.600, t = 7.001, P = 0.0007 for 2011). Our findings indicate that impacts are more associated to functionally clustered and low species richness in the fish communities are. Hence, according to our results, the process of colonization and establishment of generalist species in degraded streams takes longer time than the loss of specialists, which occurs little after deforestation, so that fish assemblages respond to a temporal mediated habitat suitability-unsuitability gradient.