Traits in space: Integrating functional ecology into spatially explicit species distribution models

Despite the potential of functional traits for mediating ecological responses to environmental perturbations, they are rarely incorporated into Species Distribution Models (SDMs). Furthermore, complex ecological data often exhibit spatial autocorrelation, which, if ignored, can reduce model power and compromise robustness. Here, we present a hierarchical spatiotemporal Bayesian framework that incorporates trait-environmental relationships and explicitly accounts for spatial autocorrelation through the implementation of Integrated Nested Laplace Approximation (INLA) and Stochastic Partial Differential Equations (SPDE). We applied this framework to four case studies to examine how trait-based and, spatially explicit approaches affect single- and multi-species models. In both scenarios, we tested the hypothesis that incorporating functional traits (body size) and accounting for spatial autocorrelation improves model performance using simulations, and we reinforced these findings with empirical analyses of a single species, the European hake ( Merluccius merluccius ), and a multispecies approach using the Iberian demersal fish assemblages. In all cases, incorporating body size as a functional trait strengthened ecological interpretation of species responses. Moreover, explicit spatial models outperformed non-explicit spatial models, highlighting the importance of accounting for spatial autocorrelation. Results revealed a consistent relationship between body size and thermal sensitivity, with larger-bodied species showing stronger responses in their distributions to temperature gradients. The results underscore the importance of incorporating traits into spatial SDMs to account for both intraspecific and interspecific distribution variations, improve ecological understanding, and boost predictive accuracy. Our trait-based SDM framework provides a scalable, interpretable, and ecologically grounded tool to support population and biodiversity predictions and conservation planning under global change.