Although natural populations have always experienced fragmentation to varying degrees, human activities in North America in recent centuries have greatly increased the extent of fragmentation for nearly all native species. Since the persistence of local populations is a function of population size, such fragmentation has significantly raised the probability of local—and therefore global—extinction for many species. The likelihood of species persistence is decreased directly through reductions in available resources such as food, breeding locations, and territorial home ranges; and indirectly through higher susceptibility of smaller populations to environmental variability such as drought years, storms, and unexpected low breeding events. Additionally, fragmentation can reduce the potential for recolonization of habitat patches where subpopulations have become extirpated. Fragmentation can also disrupt ecosystem processes on which species and communities depend. For example, when habitat is fragmented by urbanization or other changes in land use that disrupt fuel continuity, fragments can no longer burn with historic frequency or patterns.

Department of Defense (DoD) lands support more than 350 federally listed species—the highest density of threatened and endangered species of any federal landowner. Military use of the land can cause threats to listed species but does not, in general, result in the complete conversion of suitable habitat to unsuitable habitat, as has been the case on lands surrounding many DoD installations. Consequently, as habitat is lost and fragmented around installations, DoD carries an increasing responsibility for conservation of listed species. Some of the measures necessary to conserve listed species restrict military land uses, potentially compromising DoD’s mission. DoD needs to develop methods to identify the most ecologically important land parcels and conservation strategies on and near its installations to facilitate long-term species conservation benefits and avoid additional military training restrictions.

The objective of this project was to develop methods to identify optimal management for metapopulations of sensitive species occurring in highly fragmented landscapes where subpopulations may be isolated and not functioning as true metapopulations. This project used the best available data or data that could be assembled within a short period of time so that options were not foreclosed by land use conversion before decisions could be made. This project will help sustain military missions by linking habitat preservation and management on and off military installations so that species persistence can be maximized and the DoD’s proportion of the responsibility for management does not increase over time.

Technical Approach

This project developed a multi-species planning framework for fragmented landscapes using population viability analysis and habitat suitability modeling. Climate change was incorporated through habitat suitability projection. Quantitative conservation objective functions were used to select among management scenarios. Metapopulation models were used in a functional assessment approach to evaluate whether dispersal dynamics were likely to be operative in the landscape and thus inform management and reserve design. The test case involved four species of the maritime chaparral in coastal southern California with different population responses to one of the primary ecosystem drivers, wildland fire, and with different hypothetical responses to fragmentation. The species were Ceanothus verrucosus, a long-lived obligate seeding shrub; Chorizanthe orcuttiana, an annual plant; Neotoma macrotis, big-eared woodrat; and Quercus dumosa, a long-lived obligate resprouting shrub. Coastal southern California is a biodiversity hot spot with high levels of habitat loss and fragmentation.


A multi-species planning framework was developed entailing four phases: 1) identification of the species and landscape, 2) modeling, 3) selection of objective functions and ranking management scenarios, and 4) implementation in an adaptive management framework. Specific recommendations for the San Diego region included maintaining longer average fire intervals. The specific objective in terms of percent decline from initial population influenced which fire management scenarios ranked highest. Climate change was included through projected changes in species abundance and distribution in the landscape. For the most part, the patterns in management rankings established for the current climate were maintained under future climates. The species modeled, even those with greater dispersal ability, have few connections between populations; consequently, removal of dispersal from the models has little to no effect on model results. This result indicates that patch connectedness is less important than total habitat area in prioritizing habitat configurations for conservation in the highly fragmented landscape of coastal southern California. If dispersal is not effective, future management may require translocations to reestablish extirpated populations and maintain genetic diversity.


Specific conservation recommendations were presented for the target species throughout their geographic ranges. In addition, a multi-species conservation planning framework was presented that explicitly considers species tradeoffs in response to multiple threats and management. The resulting methods relied on a combination of single patch and metapopulation models, habitat suitability models, and quantitative conservation objective functions to explicitly address species tradeoffs under multiple threats from global climate change.