Improving the performance of indicator groups for the identification of important areas for species conservation.

Indicator groups may be important tools with which to guide the selection of networks of areas for conservation. Nevertheless, the literature provides little guidance as to what makes some groups of species more suitable than others to guide area selection. Using distributional data on all sub-Saharan birds and mammals, we assessed factors that influence the effectiveness of indicator groups. We assessed the influence of threatened, endemic, range-restricted, widespread, and large-bodied species by systematically varying their number in indicator groups. We also assessed the influence of taxonomic diversity by systematically varying the number of distinct genera and families within the indicator groups. We selected area networks based on the indicator groups and tested their ability to represent a set of species, which, in terms of species composition, is independent of the indicator group. Increasing the proportion of threatened, endemic, and range-restricted species in the indicator groups improved effectiveness of the selected area networks; in particular it improved the effectiveness in representing other threatened and range-restricted species. In contrast increasing the proportion of widespread and large-bodied species decreased effectiveness. Changes in the number of genera and families only marginally affected the performance of indicator groups. Our results reveal that a focus on species of special conservation concern, which are legitimate conservation targets in their own right, also improves the effectiveness of indicator groups, in particular in representing other species of conservation concern.