The life history of the Yellow-spotted Amazon River Turtle (Podocnemis unifilis) as told from the nests

Nesting site selection is of great ecological and evolutionary importance for turtles, because it is a key determinant of individual reproductive success. In this study we evaluated the effects of two types of nesting substrate (sand bank and clay banks) on the incubation success of a generalist nesting species, Podocnemis unifilis, in a floodplain area in the Brazilian Amazon. Nesting characteristics were recorded during the nesting seasons of 2012-2014, when 634 nests were monitored throughout the incubation period. Mean size of laying females did not vary between substrate types (sand = 24.71 cm; clay = 2 5.44 cm; P = 0.2054), nor was clutch size related to females' sizes. However, females that nested on sand bank exhibited a clearer pattern of nest positioning relative to distances to vegetation and water line than those laying eggs on clay banks. Mean egg size (40.24 +/- 2.68 mm) and weight (21.09 +/- 5.46 g) did not vary between types of nesting substrate, and there was no variation in clutch size across the reproductive seasons for either substrate. On the other hand, clutch sizes were higher in nests in clay banks (28 +/- 8) than those that were deposited in sand bank (22 +/- 7). Mean incubation time was shorter in sand bank (62 +/- 5.98) than in clay banks (77 +/- 4.33). Although eclosion success for sand and clay nests was high (84.82 and 7 2. 97 %, respectively), incubation success was relatively low in both substrate types (sand = 57.34%; clay = 25.80%). On sand bank, clutch losses were mainly due to human predation (16%), on clay banks, human predation also affected 16% of the nests, but floods compromised another 30%. Natural predation was mostly due to ovivory by the lizard Tupinambis teguixin, although fire ants, Solenopsis geminata, also contributed to egg predation on clay banks. Hatchlings incubated on sand bank were on average larger and heavier than those from clay banks nests. Our results suggest that differences found in the life histories of different populations of P. unifilis reflect flexible responses that are largely linked to the diversity of habitats this species utilizes for nesting. Furthermore, the presence of source-sink and habitat patches should be considered for short term population management, beyond the creation of networks of reserves that incorporate and protect a variety of nesting sites that these small local groups depend on for their survival.