WikiHome

Wildlife Habitat Assessment

John C. Kilgo
Southern Research Station, USDA Forest Service

Introduction

Wildlife habitat assessment is the evaluation of the relative habitat conditions available to a focal group of wildlife. Assessments of wildlife habitat are predicated on the basic assumption that at some level wildlife is controlled by its habitat, because an organism’s ability to survive depends in large part on the resources (e.g., food, water, and cover) available to it. These resources are provided by the habitat in which organisms live. Thus, habitat conditions are often used as a surrogate to make inferences about the presence, abundance, fitness, or productivity of wildlife populations, species, or communities.

Each species is adapted to a fairly unique set of habitat conditions. For example, Swainson’s warblers (Limnothlypis swainsonii) most often are associated with moist lowland forests that have a dense shrub layer, whereas brown-headed nuthatches (Sitta pusilla) prefer mature upland pine forests. These associations between wildlife species and their preferred habitats are often referred to as wildlife habitat relationships. The range of habitat conditions that are acceptable varies considerably among species. Swainson’s warblers and brown-headed nuthatches tend to be restricted to the types described above, whereas many other bird species may be found virtually across the landscape. Species that occur only in a narrow range of conditions are habitat specialists, whereas those that tolerate a broader range are habitat generalists. It tends to be easier to assess or predict the habitat relations of specialists than of generalists.

The assumption that the presence or abundance of a species reflects the quality of a given location as habitat is not always valid. An important aspect of habitat quality is the contribution of the habitat to the reproductive fitness of the population, and hence its ability to persist through time. In this regard, habitat quality may not be very well correlated with density. For example, in birds, territoriality may limit the number of breeding individuals that can occur in a particular forest stand, even one of optimal quality, and relegate non-breeding individuals to areas of lesser quality. Density may actually be higher in the low quality stand where no reproduction occurs. Likewise, areas of potentially optimal quality may be completely unoccupied by species that are rare, simply because of a lack of individuals to fill available habitat. These considerations notwithstanding, as long as managers are alert to this possibility, the assessment of habitat conditions based on density still can be a useful tool in evaluating and understanding the impacts of habitat alteration.

Wildlife habitat assessment is conducted for a variety of purposes. A manager may simply want to assess the capability or likelihood that a given tract can support certain wildlife species. When declines in the abundance of a focal species are noted, wildlife managers may assess habitat conditions to determine if some degradation in quality may be responsible. Often, it is desirable to be able to predict in advance the impact of proposed land management activities, such as timber harvest or prescribed burning, on the quality of wildlife habitat. Following is an overview of some considerations and common approaches to wildlife habitat assessment.

Habitat Feature.--Before undertaking any assessment of habitat, it is essential to understand the habitat relationships and preferences of the species of interest. A review of the literature or consultation with experts familiar with the species often is sufficient to determine which habitat features are important to the focal species and when. Habitat characteristics can be considered as either macro or micro variables, roughly corresponding to the scale or perspective of measurement. Assessment of the quantity and quality of such variables, either formally (as with models, see below) or informally, can provide insights into the relative quality of habitat for wildlife.

Macro Features.--Characteristics of habitats that are described across relatively large areas or within a landscape context usually can (and sometimes must) be assessed remotely via the use of tools such as geographic information systems (GIS), satellite imagery, or aerial photography. Examples of such variables include stand or patch size, shape, and age, amount of edge, vegetation cover type, and distance to other important features such as water, roads, cliffs, caves, or neighboring nesting/roosting sites.

Micro Features.--Some habitat features that describe more site-specific conditions must be directly assessed on-site. Forest inventory databases are available in some situations (e.g., the U.S. Forest Inventory and Analysis database), but the data recorded in such inventories may be of limited use if collected for other purposes or if very specific features are needed. Micro habitat features most often include measures of vegetation structure and composition, though soil and topographic conditions are also critical to some species. In forested habitats, vegetative conditions are usually the most important habitat components for wildlife. Vegetation structure includes both vertical and horizontal aspects of density (e.g., stem density, foliage volume), and particular preferred plant species may need to be considered. Other specific features like snag size and density (for cavity nesting birds and small mammals), burrows (for small mammals, reptiles, amphibians, and invertebrates), or the availability and quality of preferred browse plants (for deer) may be measured.

A wide variety of methods have been developed for vegetation sampling, from plot- or transect-based designs to plotless methods. The sampling method must be carefully adapted to the objective of the study. See Higgins et al. (1994) for a review of vegetation sampling methods applicable to wildlife habitats.

Wildlife Habitat Models

A tremendous body of literature exists on wildlife habitat modeling. The purposes of such models range from enhancing our understanding of species’ ecology and the habitat factors that affect their distribution and abundance to predicting future distribution and abundance under different scenarios affecting the habitat. Many approaches to modeling and many types of models have been developed. General model types include (but are not limited to) statistical empirical models, habitat relationship models, simulation models, GIS-based models, and expert (or knowledge-based) models. Some of the more commonly used models are briefly described below.

Wildlife Habitat Relationship (WHR) Matrix Models.--WHR matrix models are simply tables that relate each of any number of wildlife species to habitat types. The quality of each habitat type for each species may be indicated simply as suitable or unsuitable, or it is often given as a qualitative categorical rank, such as marginal, suitable, or optimal. These ranks may be derived from field studies, professional judgment, or a combination of both. While WHR matrix models lack detail and tend to predict more species as present than actually are, they can be useful in quickly assessing potential habitat capabilities across many species.

Forest stand growth and yield models designed for silvicultural purposes and models of ecological succession are often used to assess wildlife habitat. Predictions of future vegetation conditions from such models can be cross-walked to the general habitat types of WHR matrix models to produce an assessment of future habitat potential.

Habitat Suitability Index (HSI) Models.--HSI models have been used by the U.S. Fish and Wildlife Service and other agencies to assess the quality of habitat for individual species of importance. The output is a suitability index that ranges from 0 to 1. This value represents the geometric mean of a set of environmental variables that are perceived to have the greatest impact on the species’ presence, abundance, or productivity. Thus, the basic structure of an HSI model is HSI = (V₁ · V₁ · … V_n)^1/n, where the V’s represent the environmental variables. Although HSI models have many shortcomings and have been heavily criticized, they constitute a repeatable assessment procedure and values can be compared among management alternatives.

HSI models frequently are used in habitat evaluation procedures (HEP), wherein an HSI score is determined for a habitat and multiplied by the area of the habitat to produce a “habitat unit.” Habitat units for the species can then be compared among locations or management alternatives.

Spatially Explicit Models.--GIS models that use geographically referenced information on habitat conditions in conjunction with models of species response can be used to simulate the impacts of habitat alteration, disturbance, and changes in landscape structure. Such models have gained popularity, as they can be combined in the GIS environment with different model types, both theoretical and quantitative, to produced much more specific output on species response than more general model types like WHR matrix or HSI models.

Model Accuracy and Applications

Perhaps the greatest shortcoming of most wildlife habitat models is the uncertainty of their accuracy. Validation of wildlife habitat models involves testing model predictions against actual field data. This process can be difficult or impossible when model structure and output is subjective or otherwise non-quantifiable in real-world terms such as density or survival or reproductive rates. However, there is growing recognition in the wildlife literature of the necessity of validating predictive models. Without an assessment of accuracy, model predictions remain only hypotheses to be tested. 
 

Nevertheless, assessments of wildlife habitat are crucial in order to understand wildlife and habitat relationships and to adjust our forest management practices to favor desirable species, deter undesirable species, and minimize adverse impacts of habitat modification on wildlife and biodiversity.  Accordingly, we will continue to use wildlife habitat assessments and models to assess our management, and continue to enhance the verification and improvement of those methods and models.

References

Anderson, S. H., and K. J. Gutzwiller. 1994. Habitat evaluation methods. Pages 592-622 in T. ABookout, ed. Research and management techniques for wildlife and habitats. The Wildlife Society, Bethesda, MD.

Higgins, K. F., J. L. Oldemeyer, K. J. Jenkins, G. K. Clambey, and R. F. Harlow. 1994. Vegetation sampling and measurement. Pages 567-591 in T. A. Bookout, ed. Research and management techniques for wildlife and habitats. The Wildlife Society, Bethesda, MD.

James, F. C., and H. H. Shugart. 1970. A quantitative method of habitat description. Audubon Field Notes 24:727-736.

Morrison, M. L., B. G. Marcot, and R. W. Mannan. 2006. Wildlife-habitat relationships: concepts and applications, 3^rd ed. Island Press, Washington, DC.

Nudds, T. D. 1977. Quantifying the vegetative structure of wildlife cover. Wildlife Society Bulletin 5:113-117.

Schamburger, M., A. H. Farmer, and J. W. Terrell. 1982. Habitat suitability index models: introduction. FWS/OBS-82/10. U.S. Fish and Wildlife Service, Washington, DC.

Scott, J. M., P. J. Heglund, M. L. Morrison, J. B. Haufler, M. G. Raphael, W. A. Wall, and F. B. Samson, eds. 2002. Predicting species occurrences: issues of scale and accuracy. Island Press, Washington, DC.

Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of Wildlife Management 47:893-901.

Verner, J., M. L. Morrison, and C. J. Ralph, eds. 1986. Wildlife 2000: modeling habitat relationships of terrestrial vertebrates. University of Wisconsin Press, Madison, WI.

Posted 30 September 2007