Scaling patterns and drivers of species richness and turnover across the Afrotropics

[摘要] ENGLISH ABSTRACT: Understanding the broad scale patterns and environmental determinants of amphibian richness(α-diversity) and turnover (β-diversity) is becoming ever more crucial as communities,ecosystems and landscapes are increasingly threatened by anthropogenic processes and climatechange. Spatial scale, the grain at which sampling takes place or analysis occurs, confoundsunderstanding of such diversity patterns and the relationships these exhibit with environmentalprocesses. This is because various processes operate at different spatial scales resulting indifferent patterns emerging at different spatial scales. This is problematic, because patterns, andpattern-process' relationships which emerge at one scale may not hold at another scale therebyconfounding our understanding of how biotic patterns are generated and maintained, thus leadingto misguided conservation strategies and policies. The focus of the present study was thus toexamine the relationship between present patterns of anuran richness and turnover, therelationship of these with several contemporary environmental processes, and how these patternsand pattern-process relationships are influenced by spatial grain.Using IUCN range data for the Afrotropical region, amphibian richness and turnover patternswere generated using a Geographic Information System and quantified using the recentlyformulated zeta diversity partitioning method (ζ). These patterns were then related to severalcontemporary environmental variables/processes hypothesised to govern amphibian diversityacross a range of spatial scales using both the global Ordinary Least Squares (OLS) method andlocal spatial Geographically Weighted Regression (GWR) across three nested spatial grains.The results show that amphibian richness and turnover patterns (visual & empirical) and therelationships these exhibit with environmental conditions were sensitive to spatial grain.Visually, spatial patterns were more distinct at finer grains, but visibly smooth at the coarsergrains. The statistics describing these patterns suggest that the scaling behaviour follows anincreasing linear and nonlinear trend across spatial grain. Results from both the OLS and GWRmodels confirm that contemporary environmental conditions are significant determinants of bothanuran richness and turnover patterns across spatial grain. Precipitation was the strongestdeterminant of anuran richness while topographic complexity best explained turnover. Therelationship between both diversity components and environmental conditions, however, wasscale dependent with environmental conditions explaining a greater proportion of the variation inthese biotic patterns at coarser than finer grains. Studies conducted across multiple grains is thusrecommended to improve current understanding of biodiversity patterns and the relationship theyexhibit with environmental processes.

[发布机构] Stellenbosch University