How well do proxy species models inform conservation of surrogate species?

Context

Proxy species, which represent suites of organisms with similar habitat requirements, are common in conservation. Landscape Capability (LC) models aim to quantify the spatially-explicit capability of landscapes to support proxy species that represent suites of forest birds.

Objectives

We evaluated the North Atlantic Landscape Conservation Cooperative (NALCC) proxy models of LC and represented species framework across 13 states in the northeastern United States from Virginia to Maine. We validated a suite of questions related to co-occurrence of proxy and represented species with a compilation of independent datasets.

Methods

We tested proxy species LC models ability to explain represented species’ occurrences, including using multiple proxies together, and benchmarked against empirical data and land cover type classifications. We tested effect of several factors on predictive ability including relative range overlap and ecological and taxonomic dissimilarity between proxy and represented species.

Results

LC models performed variably, but represented species occurrences were rarely predicted as accurately as proxy species. Models improved predictions over macrohabitat classifications. Using multiple proxies together occasionally improved predictions of represented species. Considerable range overlap was needed for models to be predictive of represented species. Ecological and taxonomic similarity had no effect on predictive ability. LC models worked similarly to using empirical observations, suggesting shortcomings were because of imperfect surrogacy.

Conclusions

Conservation proxies as representatives of species groups that are associated with macrohabitats are useful, but empirical data are necessary to evaluate proxy species’ effectiveness. Habitat-based models can provide similar predictive ability as empirical observations of proxies and represent a useful tool in conservation planning.