Does a lack of juveniles indicate a threat? Understanding body size distributions in a group of long-lived vertebrates

Turtles are declining globally, and absences of juveniles during surveys are often interpreted as evidence of threats to early life stages. In Australia, for example, it is widely argued that a low number of juveniles is likely due to nest predation by introduced red foxes (Vulpes vulpes). However, small sample sizes within populations, low detectability of juveniles and turtles' long lifespans often confound the conclusion that a paucity of juveniles indicates a declining population. Because turtles have long reproductive lifespans, we might intuitively expect most turtle populations to be heavily weighted towards large individuals, but a 'typical' or 'healthy' size distribution for turtle populations has not been well established. Therefore, we collated data on 41,021 freshwater turtles from 38 species and 428 populations located in parts of Australia both with and without introduced foxes, as well as populations in the United States of America, which naturally have raccoons (Procyon lotor), foxes and other nest predators. We examined population-level body size distributions to establish a baseline for 'typical' turtle populations and test whether populations that are exposed to introduced foxes have proportionately fewer juveniles compared to both AU populations that lack introduced foxes and USA populations that are naturally exposed to nest predators. We found that most turtle populations in AU and the United States were heavily skewed towards adults and had few juveniles, regardless of the presence of foxes or other nest predators. There were, however, clear differences among population survey methods: those that target shallow areas (e.g. crawfish traps) tended to capture proportionately more juveniles, and small sample sizes (∼<50) often produced inaccurate representations of size distributions. Additionally, we used a simulation to demonstrate that, given common turtle life history parameters, even stable populations should generally have low proportions of juveniles. Based on our results, we encourage caution when interpreting turtle size distributions. A small number of juveniles does not inherently suggest that a population is declining due to high egg and/or juvenile mortality, and researchers should pay careful attention to the biases in their methods and strive to capture a minimum of 50-100 turtles before drawing inferences.