The Past as a Lens for Biodiversity Conservation on a Dynamically Changing Planet: The utility of body size as a functional trait to link the past and present in a diverse reptile clade

Understanding the relationships between functional traits and environment is increasingly important for assessing ecosystem health and forecasting biotic responses to future environmental change. Taxon-free analyses of functional traits (ecometrics) allow for testing the performance of such traits through time, utilizing both the fossil record and paleoenvironmental proxies. Here, we test the role of body size as a functional trait with respect to climate, using turtles as a model system. We examine the influence of mass-specific metabolic rate as a functional factor in the sorting of body size with environmental temperature and investigate the utility of community body size composition as an ecometric correlated to climate variables. We then apply our results to the fossil record of the Plio-Pleistocene Shungura Formation in Ethiopia. Results show that turtle body sizes scale with mass-specific metabolic rate for larger taxa, but not for the majority of species, indicating that metabolism is not a primary driver of size. Body size ecometrics have stronger predictive power at continental than at global scales, but without a single, dominant predictive functional relationship. Application of ecometrics to the Shungura fossil record suggests that turtle paleocommunity ecometrics coarsely track independent paleoclimate estimates at local scales. We hypothesize that both human disruption and biotic interactions limit the ecometric fit of size to climate in this clade. Nonetheless, examination of the consistency of trait–environment relationships through deep and shallow time provides a means for testing anthropogenic influences on ecosystems.

In healthy ecosystems, the distributions of functional traits should be sorted by environmental relationships (1–3) and are, therefore, important for conservation forecasting and response strategies (4–7). Trait–environmental relationships can be modeled across a range of communities and habitats, and the resultant models can be applied to future climate change projections to forecast expected trait distributions for communities adapted to particular future climatic conditions. Such distributions can then be used to identify which species are at risk for extirpation or extinction because their traits do not fall within the projected community distribution and, conversely, which species’ traits will be well suited for future climates.Key to utilizing traits is an understanding of the functional factors that govern performance. Hypotheses of trait function are determined through direct behavioral observation, experimental mechanistic approaches, or spatial correlation of trait distributions with environmental variables (8) with post hoc inference of functional relationships (9). The development of taxon-free analysis of functional traits (ecometrics) allows further tests of function across clades and through time (9–12). Application of modern ecometric models to trait distributions in fossil communities (hindcasting) allows us to evaluate histories of niche stability (13, 14). Comparing estimates from hindcasting to independent coeval paleoclimate proxies can provide data on trait resilience or shifts in functional factors over time, which should be established in order to use ecometric models to forecast community compositions in response to future climate change (11). Here we examine the utility of body size as an environmentally correlated functional trait in extant turtles, both in the context of metabolic scaling with temperature for species’ size maxima and as a community ecometric correlated to temperature and precipitation. We apply our ecometric models to the fossil record of turtles from the Plio-Pleistocene eastern African tropics to assess whether modern ecometric relationships persist through time, and are therefore potentially useful for conservation planning.Body size is an important functional trait for examining biotic responses to environmental change, with multiple functional factors sorting size, including metabolism, developmental rate, and changes in ecosystem structuring (15, 16). In poikilotherms, temperature-dependent metabolism has been proposed as a mechanism to explain either increases in body size with temperature for mass-specific metabolic rates or decreases in size based on increasing growth and metabolism (16, 17). Disentangling the relative roles of metabolism and life history processes in driving trait distribution is crucial for understanding potential responses to critical climate maxima and trophic responses to trait shifts (18, 19).Turtles are widely distributed and speciose (20) (Fig. 1A), and are a conservation concern, with over 50% of species classified as endangered due to habitat loss and human predation (21). Turtles are ecologically diverse, with habitats and habits ranging from fully terrestrial herbivores to obligately aquatic carnivores, and body sizes of extant species ranging from carapace lengths of 11 cm to 150 cm (Dataset S1). Previous analyses of body size distributions in extant taxa at a global scale have found positive correlations between size, latitude, geographic range, and temperature (22, 23), but have not found strong support for the functional factors underpinning these relationships. Analyses of within-lineage size trends have additionally found inverse relationships of body size to environmental temperature (24–26), further confounding functional inferences of the role of metabolism and the reliability of body size as a functional trait.Open in a separate windowFig. 1.(A) Geographic distributions of species richness (Top) and maximum body size, measured as carapace length (Bottom), for extant turtles. Rectangles denote ranges used for continental-scale ecometric analyses. (B) Turtle fossils from the Shungura Formation of Ethiopia (yellow star on Inset map). (Top) The aquatic trionychid turtle Trionyx cf. triunguis, carapace in dorsal view, rock hammer for scale, Member G. (Bottom) The terrestrial tortoise cf. Centrochelys sp., partial carapace, girdles, and appendicular skeleton in visceral view, with tape measure for scale, Member H.To determine whether turtle body size distributions are correlated to metabolic rate, we model expected maximum size for minimum mean annual temperature (MAT) within geographic ranges of aquatic and terrestrial species (SI Appendix) based on the size–MAT relationships for the largest living species, using a metabolic scaling model based on the Q₁₀ coefficient which describes oxygen consumption changes per temperature change by 10 °C (17). We compare model predictions with actual size–temperature relationships for living species. To model size as a community ecometric, we apply linear regression (LR) and maximum likelihood (ML) models to determine the predictive power of turtle body size distributions for estimating climate variables including MAT and mean annual precipitation (MAP). We use standard ecometric approaches and, additionally, train ecometric models using range predictions for turtle species derived from species distribution models (SDMs). SDMs model geographic ranges where environmental conditions are for each species, which minimizes the effects of anthropogenic extirpation and habitat loss on estimating trait–environmental relationships. Comparisons of these models reveal which input datasets and spatial scales capture the strongest ecometric relationships.We hindcasted modern ecometric and scaling relationships onto the turtle fossil record of the Plio-Pleistocene Shungura Formation of Ethiopia and compare results with other paleoclimate proxies to determine whether extant relationships are sufficient for predicting nonanalog climates of the past, and, by extension, the future. The Shungura Formation consists of fluviolacustrine sediments and intercalated volcanic tuffs cropping out west of the Omo River in southwestern Ethiopia (ref. 27 and references therein). It preserved a dense, well-sampled vertebrate fossil record representing both terrestrial and aquatic ecosystems, including a diversity of turtles (Fig. 1B), as well as paleoclimate estimates from paleobotanical, faunal, and geochemical proxy data (28–33). The Shungura Formation is precisely chronostratigraphically dated throughout from ∼3.6 Ma to 1.0 Ma, and thus records faunal responses of rich tropical vertebrate communities to climate parameters equivalent to projections of future anthropogenically mediated warming (34). As a result, it represents an excellent model system for testing functional traits beyond modern environments.