Wildlife Habitat Modeling
Advanced approaches for estimating and conserving wildlife-habitat relationships
Our lab is collaborating with the Arizona Game and Fish Department, U.S. Forest Service, National Park Service, Bureau of Land Management, and other agencies and NGO’s in Arizona, Nevada, New Mexico, and California to develop spatially explicit predictive models of wildlife response to habitat and landscape change across extensive spatial scales. We use novel spatial modeling and statistical techniques to estimate occupancy, abundance, connectivity, and other key demographic parameters for multiple species of sensitive wildlife, including American pronghorn, Merriam’s Turkey, Northern Goshawk, songbirds, puma, black bears, desert bighorn sheep, desert tortoise, kit fox, small mammals, and numerous other listed or imperiled species.
Tools and data used to guide our work
Although the modification and loss of wildlife habitats is occurring globally at unprecedented rates and extents, the magnitude of these changes can be challenging to document or estimate. The persistence of most animal populations varies in part as a function of the amount and configuration of their habitats. Conservation efforts for these species can be guided effectively by accurate models and maps of the location, quantity, and quality of these habitats. In order to build the “best” wildlife-habitat relationships models we can, we integrate remote sensing applications, multi-date satellite imagery, and ground data to custom derive spatially explicit characterizations of habitat location and quality.
Indeed, animal movement and use of habitat is motivated by a complex suite of environmental cues and ecological processes, including the location of prey, conspecifics, and competitors in time and space. However, the movement patterns of species cannot be separated from the habitat and landscape features that connect individuals or populations. For multiple species on a heterogeneous landscape, connectedness depends on the organisms under investigation and how relevant habitat attributes are distributed. Due to differences in life and ecological histories (e.g., body size, mobility, migration rates), co-occurring species can simultaneously perceive a landscape as both connected and disconnected. Thus, the methods we use to quantify connectivity, such as circuit theory or Bayesian hierarchical models, involve careful consideration of how organisms deferentially interact with the landscape during movement events and how these events are influenced by management or land-use activities. With increasing levels of habitat loss and fragmentation, species are forced to navigate novel environments with features that may impede their distribution or persistence. In this context, novel and robust tools are required to measure and conserve connectivity processes for multiple species, and maintain or enhance biodiversity in threatened environments.
Coupled with data on animal demographic or movement data, robust statistical techniques and model selection approaches, these models of habitat also can be used to estimate or predict the response of species to forest treatment, fire, plant invasion, or alternative management scenarios (e.g., prior to their implementation).
Rigorously developed, empirically based models of wildlife-habitat relationships will better facilitate the provision and conservation of resources necessary to support populations of species that may be sensitive to environmental changes. In the face of global climate change and continued human population growth, for example, the tasks of identifying and assessing these relationships will only gain in importance.