Abstract
BACKGROUND: How species colonize new environments is still a fundamental question in ecology and evolution, assessable by evaluating range characteristics of invasive species. Here we propose a model approach to evaluate environmental conditions and species features to predict niche changes in non-equilibrium contexts. It incorporates potentially range-limiting processes (fundamental niche), hence allowing for better predictions of range shifts, differentiation of analog and non-analog conditions between the native and non-native (invaded) ranges, and identification of environmental conditions not currently available but likely in the future. We apply our approach with the worldwide invasive slider-turtle Trachemys scripta. METHODS: We estimated the native and non-native realized niches of T. scripta and built its fundamental niche based on key features of the turtle's temperature physiological tolerance limits and survival-associated factors. We next estimated response functions adjusted to the physiological predictor variables and estimated habitat suitability values, followed by a comprehensive set of analyses and simulations to compare the environmental conditions occupied by T. scripta (at its native and non-native ranges). RESULTS: Climatic space analysis showed that the T. scripta's non-native realized niche is 28.6% greater than the native one. Response curves showed that it does not use its entire range of temperature tolerances (density curves for native: 5.3-23.7 °C and non-native: 1.7-28.4 °C ranges). Whether considering the mean temperature of the warmest or the coldest quarter, it occupies a wider range of temperatures along its non-native distribution. Results of the response curves for worldwide (global) and across Mexico (regional) comparisons showed it occupies analog and non-analog conditions between its native and invaded ranges, exhibiting also unoccupied suitable climatic conditions. DISCUSSION: We demonstrate that T. scripta occupies a wider subset of its fundamental niche along its non-native range (within its physiological tolerances), revealing that the species observed niche shift corresponds to a different subset of its fundamental niche (niche unfilling). We also identified suitable environmental conditions, globally and regionally, where the slider turtle could potentially invade. Our approach allows to accurately predict niche changes in novel or non-equilibrium contexts, which can improve our understanding about ecological aspects and geographic range boundaries in current and potential invasions.