Anthropogenic threats increase vulnerability of Podocnemis unifilis nesting areas in the most protected Brazilian Amazon State

Abstract

Introduction: Spatial prioritization is essential for conservation strategies seeking to reverse population losses among Amazonian freshwater turtles, including the yellow-spotted river turtle (Podocnemis unifilis). Nesting areas represent critical habitats for populations of this species. Broad-scale assessments may, however, fail to detect anthropogenic threats to nesting areas. Therefore, acquiring local-scale data is crucial for accurately identifying these threats and subsequently informing effective, site-specific conservation and restoration measures. Methods: We assessed the vulnerability of P. unifilis nesting areas (n = 275) along 197 km of rivers in equatorial Brazilian Amazonia across a gradient of anthropogenic impacts. We adopted a multiscale approach that included variables associated with anthropogenic impacts and land cover change. We quantified vulnerability at three spatial scales (nesting area, 5 km river reaches, and ~50 km zones) by analyzing a combination of field and remote sensing data on nesting area size, land cover change, and anthropogenic threats. A multifactorial vulnerability index was calculated based on the variables analyzed. We then used this index to map nesting area vulnerability at different spatial scales. We then explored the main drivers associated with vulnerability through hierarchical-clustering-on-principal-components, which identified six clusters. Results: The most vulnerable nesting areas were in the downstream locations close to a town and hydroelectric dam. We found that 14 nesting areas (1.4% of the overall area) had very high vulnerability, while most nesting areas (206) had very low vulnerability. Discussion: We established conservation priorities based on a combination of nesting area size, location, and vulnerability. Our results can help inform conservation and restoration actions for P. unifilis. The approach could be useful to generate similar assessments across the species pan-Amazonian range.