The role of physiological optima in shaping the geographic distribution of Spanish moss
ecological niche, ecophysiology
Aim Ecological niche modelling is being widely applied to help us understand the geographic distributions of species, despite challenges regarding the estimation of fundamental niches that limit model transferability over time and space. Mechanistic models are an alternative, but they can be difficult to implement owing to the detailed knowledge that they require about the organism for full parameterization. In this paper we explore the geographic projection of physiological measurements of optimal temperature, precipitation and relative humidity requirements, as measured under controlled conditions using a climate dataset with high temporal resolution, as a case study for Spanish moss (Tillandsia usneoides), and compare scaling effects with correlative niche models calibrated in Maxent. Location The Americas. Methods We used climate data with a high temporal resolution to understand how often and where populations of Spanish moss occur under optimal and suboptimal conditions with respect to different environmental variables across their geographic range. We used weather station data for the United States with higher spatial resolution to provide a finer-grained view. We also developed an ecological niche model to show how averaged climate data can present inaccurate views of physiological thresholds for the species. Results Few populations of Spanish moss are located at sites that present suboptimal conditions for more than two environmental parameters. The northern distributional limit of Spanish moss is set by minimum temperature requirements, whereas maximum temperatures are less limiting. However, when the same occurrences are analysed with respect to averaged climate data, 95% of populations appear to fall within the optimal physiological intervals. Main conclusions Our analyses revealed that most populations of Spanish moss do not experience optimal ecophysiological conditions for all environmental variables, even over long time-scales. Physiological data may be of limited utility in delimiting suitable areas for populations of species, but they offer unique perspectives on the causes of range limitation.