Comparing environmental and biological surrogates for biodiversity at a local scale

A recent debate concerns the relative merit of the two major types of surrogates for biodiversity, biological Surrogates and environmental surrogates. Evidence, in the form of direct comparison of these two surrogate types, is scarce. We conducted a direct comparison of the performance of a series of biological and environmental surrogates, at a local scale (300 km(2)), which is often the relevant scale for land planning and management. Performance was referred to as the degree Of Surrogate congruence with a spatial pattern of diversity of woody species, of geophytes, and of land snails. "Environmental domains", surrogates based on numerical classification of environmental variables (topography, soil, and vegetation cover), outperformed other environmental surrogates (qualitatively delineated vegetation units and physiographic land types). The environmental domains surrogates were robust to subjective decisions on a number of classes and on input variables that drove the classification. The best biological surrogate was the woody species diversity pattern, with performance similar to that of the environmental domains. Our results support the notion that environmental domains may be reliable and cost-effective surrogates for biodiversity at small scales, particularly in data-poor regions.