Seasonal effects on the potential spatial distribution of Amazonian migratory catfishes

Abstract

The Amazon basin, spanning approximately 540,000 km², exhibits distinct fluviometric surfaces that differentiate between the dry and rainy seasons. This seasonality, along with hydrological connectivity and the creation of new habitats during the rainy season, significantly promotes the migration, reproduction, and feeding of potamodromous fishes. To estimate the realized niches of species, species distribution models (SDMs) employ the extrapolation of environmental predictors and species occurrence data. Our objective was to compare the spatial distribution of migratory fish species in the Amazon basin using SDMs based on variables characterizing the dry season, rainy season, and a combination of both. All evaluated treatments demonstrated high performance and exhibited different distribution ranges in the applied SDMs, particularly when combining environmental variables with occurrence data during the rainy season. These findings support the hypothesis that spatial distribution is influenced by seasonality. The increased fluviometric surface and enhanced connectivity of the rainy season favor both longitudinal and lateral migrations of Amazonian migratory catfishes. Moreover, the spatial distribution reveals four critical spatial overlap (CSO) regions with higher population densities regardless of the season. These CSOs primarily coincide with the Amazon alluvial plain, which exhibits the highest rates of endemism, species richness, and abundance of organisms. Considering the discontinuous and heterogeneous nature of fluviometry when performing niche modeling processes is pivotal, although SDMs applied in the Amazon generally ignore such regional seasonality.