Elevational diversity gradients of amphibians in Mexican mountain ranges: patterns, environmental factors, and spatial scale effects

Mountains are estimated to harbor the highest diversity in many taxonomic groups. However, patterns of elevational distributions and their underlying explanatory environmental variables have not been well studied in amphibians. We explore elevational patterns of species richness in amphibians and evaluate seven environmental models that may explain such patterns. We also test how the spatial scale of elevational bands affects the perceived patterns and the power of the explanatory variables. We utilize records from the most comprehensive amphibian database of Mexico. Gradients were selected across five Mexican mountain ranges, for both coastal and continental slopes, according to four criteria that allowed for standardization. Species richness-elevation relationships were classified into five patterns as described in the literature. Generalized linear models were applied to test six models, which include temperature, precipitation, water-energy dynamics, topographic heterogeneity (TH), their interaction and area. Species richness and environmental factors for each model and each gradient were calculated at four spatial scales of elevational bands (100, 200, 300, and 500 m). A total of 20 gradients were evaluated, for which all five patterns of species richness reported in literature were observed at least once. The pattern of mid-peak richness was the most commonly observed, followed by a decreasing pattern of richness with elevation. The water-energy dynamic was the best model at smaller spatial scales, while temperature, TH and area, became increasingly important in explaining richness patterns as the spatial scale increased. The elevational distribution of species richness in amphibians showed similar tendencies as in literature. This study demonstrated a larger variation in the power of explanatory variables between mountain ranges than between spatial scales. In general, the proportion of variance explained was high, showing that the chosen explanatory environmental variables are important in structuring the elevational patterns.