Military land managers are faced with the challenge of balancing threatened, endangered, and at-risk species conservation with testing and training activities. Compounding the challenge are the frequent shortages of data needed and the complexity of most quantitative approaches for in-depth effects analyses. Of necessity, then, these evaluations are typically qualitative and based largely on expert opinion. The manager’s job would be greatly facilitated by easy to use, scientifically-based tools to assess the effects of training and conservation programs on the viability of imperiled species.

The overall objective of this project was to develop a set of models and methods for prioritizing management of threatened, endangered, and at-risk species. Specifically, researchers sought to develop tools for evaluating effects of management on population viability using varying levels of data: (1) life history data when little is known about a species, (2) count data when only estimates of abundance through time are available, and (3) metapopulation models when there is a lot of demographic data about multiple populations.

A framework was constructed based on structured population models that can be used to evaluate possible effects of management actions on population growth rates for data-poor species. This approach was demonstrated with the arroyo toad. A time-series of count data was then used to fit stochastic models of population growth and assess population trend or viability. This approach was applied to evaluate extinction risk for the greater sage grouse at Yakima Training Center, Washington. For species with detailed information about vital rates, spatial structure, and metapopulation characteristics, methods were developed for conducting global sensitivity analyses of viability models for multiple populations. These methods were demonstrated using a pilot conservation incentive program for golden-cheeked warblers at Fort Hood, Texas. Finally, time-series abundance estimates for multiple populations were combined to evaluate the probability of persistence for Sonoran pronghorn at Barry M. Goldwater Range, Arizona.

This project demonstrated that arroyo toad population growth rates are highly dependent on the survival of toads during their terrestrial life stages. The strong effect of parameter uncertainty on population persistence estimates was also illustrated. A set of population growth models was applied to the greater sage grouse population at Yakima Training Center and showed that the population growth rate is negatively influenced by drought and increasing area of grassland habitat. Overall, the probability of this population dropping below an extinction threshold of 50 lekking males was high under the best growth models. In demonstrating the approach for applying sensitivity analyses to metapopulation models for golden-cheeked warblers, uncertainty in model structure and parameter estimates made it impossible to develop general guidelines for valuing habitat patches, but parameters that need to be targeted in future research were identified. Using time-series estimates of abundance for multiple populations, the probability of persistence for Sonoran pronghorn under three management scenarios was estimated and showed a clear and substantial benefit to supplementing the wild population with individuals from the captive population at Cabeza Prieta National Wildlife Refuge.

The tools developed by this project can be used to address management questions for many species of Department of Defense concern and to improve biological assessments, NEPA analyses, recovery planning, and endangered species management plans. Moreover, these tools can be used to provide direction as to how future research should proceed to focus on the factors that are most important to population viability. The demonstrations of these tools were designed to address real management issues and have clear and immediate use to the managers of focal installations.