Relationships between fishery catch rates and land cover along a longitudinal gradient in floodplains of the Amazon River

Abstract

Land cover is an important driver of patterns of fish diversity and assemblage structure in Amazonian floodplains. In the floodplain of the Amazon River, remarkable changes in the land cover composition across a longitudinal gradient, from the middle Amazon to the lower Amazon, likely affects fishery catch rates at local scales. Here, we investigated potential relationships between fishery catch rates and extent of land cover types in lakes of these two regions, the middle and lower floodplains of the Amazon River. Areas of flooded shrubs, herbaceous vegetation, and open water, within buffers of 1000 and 2000 m, were quantified around the areas of 23 lakes and modeled as predictors of fishing catch rate in the two subregions of the River i.e., the lower and middle Amazon, while these regions were also included as fixed effects of our models. Comparisons of the spatial extent of the land cover types between the two regions showed a predominance of flooded herbaceous vegetation in the downstream-lower portion, and a predominance of flooded shrubs and forests upstream-middle portion. In the models, the region was consistently the variable of greatest relative importance (Σwi > 0.9), and indicates higher fishing catch rate in the middle Amazon region when compared to the lower Amazon. Flooded shrub quantified on the smallest scale was the land cover type of strongest influence driving higher fishing catch rates, regardless of the region. In particular, our results support the idea that loss of flooded shrubs in adjacent areas of floodplain lakes can lead to reduced fishing catch rates at local and regional scales.