Efeito das alterações climáticas sobre interação peixe-floresta na Bacia Amazônica.

Abstract

Climate change impacts biological communities, altering their structure and dynamics. These effects ultimately reflect on ecological interactions, compromising the stability and resilience of ecosystems. To assess whether the fish-forest interaction may experience decoupling in future scenarios, we used species distribution models to obtain the present and future distributions of two Amazonian frugivorous fish species, Colossoma macropomum and Piaractus brachypomus, as well as 35 fruit-bearing tree species in the Amazon Basin. Models were projected for two carbon emission scenarios: a moderate scenario (SSP2-4.5) and a pessimistic scenario (SSP5-8.5), for the years 2050 and 2090. Decoupling was measured through connectance, calculated for each grid cell. In both evaluated scenarios, the two fish species experienced a significant contraction in their climatically suitable areas, concentrating in the central, central-northern, central-western, and central-southern portions of the Amazon Basin. The greatest reduction in tree species richness was observed under the pessimistic scenario (SSP5-8.5), with the suitable occurrence area for several tree species decreasing by up to 99%. Furthermore, our results indicate that by 2090, some of these tree species may completely lose their climatic suitability, while others may benefit from climate change. The connectance between species showed an alarming reduction throughout the projected scenarios. Under the moderate carbon emission scenario (SSP2-4.5), the average species connectivity dropped to 0.187 by 2050 and further declined to just 0.112 by 2090. In the pessimistic scenario (SSP5-8.5), the decline was even more pronounced, with values of 0.096 for 2050 and only 0.021 in 2090, representing a loss of up to 93.9% of connectance by 2090. Despite the overall decline in species occurrence probability and connectivity, high values of these metrics remain concentrated around the major rivers of the basin, such as the Amazon, Madeira, Negro, Tapajós, and Tefé Rivers. These predictions highlight the importance of large rivers as potential climate refugia, where ecological connectivity can be partially maintained, promoting the persistence of fish-forest interactions in the context of climate change. The modeling methodology, combined with ecological network analyses, played a crucial role in obtaining results and identifying interaction patterns between fish and plants observed in nature, such as the connectance index and potential decoupling in the biotic interaction process.