Predicting suitable nesting sites for the Blackcaiman (Melanosuchus niger Spix 1825) in theCentral Amazon basin

Abstract

After many years of illegal hunting and commercialization, the populations of the Black caiman (Melanosuchus niger) have been recovering during the last four decades due to the enforcement of a legislation that inhibits their international commercialization. Protecting nesting sites, in which vulnerable life forms (as reproductive females, eggs, and neonates) spend considerable time, is one of the most appropriate conservation actions aimed at preserving caiman populations. Thus, identifying priority areas for this activity should be the primary concern of conservationists. As caiman nesting sites are often found across the areas with difficult access, collecting nest information requires extensive and costly fieldwork efforts. In this context, species distribution modeling can be a valuable tool for predicting the locations of caiman nests in the Amazon basin. In this work, the maximum entropy method (MaxEnt) was applied to model the M. niger nest occurrence in the Mamirauá Sustainable Development Reserve (MSDR) using remotely sensed data. By taking into account the M. niger nesting habitat, the following predictor variables were considered: conditional distance to open water, distance to bare soil, expanded contributing area from drainage, flood duration, and vegetation type. The threshold-independent prediction performance and binary prediction based on the threshold value of 0.9 were evaluated by the area under the curve (AUC) and performing a binomial test, respectively. The obtained results (AUC = 0.967 ± 0.006 and a highly significant binomial test P<0.01) indicated excellent performance of the proposed model in predicting the M. niger nesting occurrence in the MSDR. The variables related to hydrological regimes (conditional distance to open water, expanded contributing area from drainage, and flood duration) most strongly affected the model performance. MaxEnt can be used for developing community-based sustainable management programs to provide socio-economic benefits to local communities and promote species conservation in a much larger area within the Amazon basin.