USE OF POPULATION VIABILITY ANALYSIS AND RESERVE SELECTION ALGORITHMS IN REGIONAL CONSERVATION PLANS

Abstract

Current reserve selection algorithms have difficulty evaluating connectivity and other factors necessary to conserve wide‐ranging species in developing landscapes. Conversely, population viability analyses may incorporate detailed demographic data, but often lack sufficient spatial detail or are limited to too few taxa to be relevant to regional conservation plans. We developed a regional conservation plan for mammalian carnivores in the Rocky Mountain region using both a reserve selection algorithm (SITES) and a spatially explicit population model (PATCH). The spatially explicit population model informed reserve selection and network design by producing data on the locations of population sources, the degree of threat to those areas from landscape change, the existence of thresholds to population viability as the size of the reserve network increased, and the effect of linkage areas on population persistence. A 15% regional decline in carrying capacity for large carnivores was predicted within 25 years if no addition to protected areas occurred. Increasing the percentage of the region in reserves from the current 17.2% to 36.4% would result in a 1–4% increase over current carrying capacity, despite the effects of landscape change. The population model identified linkage areas that were not chosen by the reserve selection algorithm, but whose protection strongly affected population viability. A reserve network based on carnivore conservation goals incidentally protected 76% of ecosystem types, but was poor at capturing localized rare species. Although it is unlikely that planning for focal species requirements alone will capture all facets of biodiversity, when used in combination with other planning foci, it may help to forestall the effects of loss of connectivity on a larger group of threatened species and ecosystems. A better integration of current reserve selection tools and spatial simulation models should produce reserve designs that are simultaneously biologically realistic and taxonomically inclusive.