A comparison of the effect of unit of analysis on the conservation status of terrestrial vertebrates in the western United States

Planning for conservation of animal and plant species requires knowledge of their distribution and conservation status. Assessing status within planning units that truncate species distributions provides highly variable results. Alternatively, rangewide distribution of a species provides the only accurate assessment of its status. I describe the “bio-assessment region” for a planning unit as the aggregation of rangewide distributions of all species of conservation interest which form the ecological context of the planning unit. I tested my premise using predicted habitat distribution maps for 59 terrestrial vertebrates from the Gap Analysis Program (GAP) in a study area bounded by the 11 western states. I first assessed the suitability of joining the individual state maps into regional distribution maps for conservation assessment. I found considerable variation in distribution extent, pattern, and grain when these state-based data sets are aggregated regionally, but their application to questions of approximate conservation status appears valid. Conservation assessments for species were conducted using the GAP method (gap analysis) for their mapped range within the study area and for each subdivision of that range in the form of counties, states, and ecoregions (Bailey's Sections). Counties approximated the rangewide conservation status of their collection of species only about 20% of the time, ecoregions 24%, and states 55%. A unit is likely to accurately assess its species' rangewide conservation status if it contains at least 90% of those species' total distribution, but not in every case. Characteristics of species and units of analysis are highly heterogeneous and largely unpredictable as I found in various correlation analyses. Case study analysis of 4 counties indicated that using analysis units other than the bio-assessment region may identify inappropriate species, amount of habitat, and locations for conservation. Therefore, my analyses indicate it is best to use the bio-assessment region to assess conservation need, whether for an individual species or for biodiversity in general. This does not preclude either analyses at the local (population) level or planning using other types of units. Instead such contextual information should be included in the mix of information used for planning and management.