This study considered the distribution of buzzard ( Buteo buteo ) nest areas in upland regions of Argyll, Scotland. The primary goal of the study was to test the efficacy of GIS in generating a model to predict the location of buzzard nests. Vegetation cover data, derived from satellite imagery and digitized topographic data, was classified into a number of discrete categories, such as mixed woodland, pre-thicket forestry, and broad-leaved woodland, and incorporated into the GIS. The authors reported a close agreement between the predicted distribution of nests based upon the habitat criteria incorporated into their model and their actual distribution determined through field surveys. Because of this, they argued that their model was successful. Moreover, they concluded that, in general, employing GIS in this way has considerable potential in predicting how the distribution of species may alter following habitat changes.

 

Bourg, N. A., McShea, W. J. & Gill, D. E. 2005. Putting a cart before the search: Successful habitat prediction for a rare forest herb. Ecology. 86 : 2793-2804.

In this study, the authors employed classification tree analysis modeling in a GIS to predict suitable habitat for the rare understory herb turkeybeard (Xerophyllum asphodeloides) in northwestern Virginia . They included several digital data layers of environmental variables in the GIS, including elevation, slope, forest type, and fire frequency, and determined the actual distribution of the herb using previously sampled data and ground-truthing for the study. By comparing the known distribution data with the predicted distribution, they correctly classified 74% of the known presence areas and 90% of the known absence areas. Moreover, they successfully identified 8 new occupied habitat patches. As a result, they considered their model to be successful at both defining suitable habitat and discovering new populations of turkeybeard, and commented on the general efficacy of this approach in other systems.

 

Hatten, J. R., Averill-Murray, A. & van Pelt, W. E. 2005. A spatial model of potential jaguar habitat in Arizona . Journal of Wildlife Management. 69 : 1024-1033.

This study employed a GIS to identify potential habitat for the jaguar (Pantera onca) in the southwestern United States. Because of the rarity in which jaguars occur in the region, distribution data for the model came from historic jaguar sightings. These were overlaid on several habitat variables, including vegetation biomes, elevation, terrain ruggedness, human density and proximity to water sources, to characterize suitable habitat. Once characterized, the authors determined that between 21% and 30% of the region could be considered potential jaguar habitat - i.e., characterized as scrub grasslands, intermediate to extreme rugged terrain within 10km of a water source. Their results were subsequently used to identify suitable areas to focus future conservation efforts aimed at protecting this rare and elusive species.

 

Liu, J., Dunning, Jr., J. B. & Pulliam, H. R. 1995. Potential effects of a forested management plan on Bachman's sparrows (Aimophila aestivalis): Linking a spatially explicit model with GIS. Conservation Biology. 9 : 62-75.

In this study, the authors combined a spatially explicit, population simulation model with a GIS to examine the potential effects of a proposed forest management plan on the population dynamics of Bachman's sparrow ( Aimophila aestivalis ) in South Carolina. Using this combined approach, they simulated the effects of prescribed harvesting, burning and thinning, on the availability of suitable habitat for the species. The results from the study suggested that the major components of the management plan may be sufficient to allow the Bachman's sparrow to reach the management goal set for it, but only after an initial population decline and lag period, and with some potential for extinction (estimated to be 5% probability over 50 years). In light of this, the authors cautioned that management for one species - the management plan in this case was created to encourage the recovery of the endangered Red-cockaded woodpecker (Picoides borealis) - can potentially threaten other species of concern, such as the Bachman's sparrow.

 

Mace, R. D., Waller, J. S., Manley, T. L., Lyons, L. J. & Zuuring, H. 1996. Relationships among grizzly bears, roads and habitat in the Swan Mountains Montana. The Journal of Applied Ecology. 33 : 1395-1404.

In this study, a GIS was used to identify the relationship between grizzly bear distribution, habitat type, and the occurrence of roads in Montana . The movement and distribution of female bears was tracked using radio-collars and these data were entered into the GIS, along with digitized elevation data and satellite-derived vegetative cover type data. Road maps were constructed by digitizing all roads present at the beginning of the study period from orthophotographic quads, and two relevant measures were subsequently considered: total road density and traffic volume. Using this approach, the authors identified several important habitat characteristics, and concluded that the bears had a preference for low temperate and temperate elevation zones, areas with low road density and frequently associated with avalanche chutes. Unfortunately, they ultimately suggested that combined mortality from natural causes and human-induced mortality, which is directly influenced by road access, is to great to promote population growth in the region.

 

Musiega, D. E., Kazadi, S. & Fukuyama, K. 2006. A framework for predicting and visualizing the East African wildebeest migration-route patterns in variable climatic conditions using geographic information system and remote sensing. Ecological Research. 21 : 530-543.

This article addressed the potential impact of climate change on wildebeest (Connochaetes taurinus) migratory routes in the Serengeti-Mara ecosystem of East Africa. Because of the link between monthly rainfall patterns in the region, which are influenced by climate change, and vegetation profiles, which strongly affect the animals' choice of migratory path, the authors suggested that knowledge of seasonal weather patterns can inform managers about the specific routes the animals might take each season. To determine if this is so, they first used a GIS to characterize the relationship between rainfall patterns and vegetation cover, and then incorporated their knowledge of wildebeest habitat (vegetation) requirements into the model to predict the routes that the animals should take. Their predictions were then compared with migratory path data obtained previously from radio-collared individuals. The strong similarities between the predicted and actual routes revealed that paths could in fact be predicted with considerable confidence provided that the general weather patterns are known.

 

Norris, D. R., Theberge, M. T. & Theberge, J. B. 2002. Composition around wolf ( Canis lupus ) dens in eastern Algonquin Provincial Park, Ontario. Canadian Journal of Zoology. 80 : 866-872.

The aim of this study was to identify patterns in habitat use of forested ecosystems by wolves (Canis lupis) around dens in Algonquin Provincial Park, Ontario, Canada using GIS. The researchers radio-collared several individuals and observed their behavior around a total of sixteen dens. Using remote-sensing imagery, they identified eight habitat types, including pine, lowland conifer, wetlands and intolerant hardwoods, and incorporated each of these into their analysis. Their results indicated that wolves preferred to establish dens in areas dominated by pine forests and avoided those areas in or neighboring tolerant and intolerant hardwood stands. Although pine forests face considerable threat from logging within the park, the authors noted that available den sites were not limiting. They did caution, however, that there is still a need to protect den sites and their associated habitat across large tracts within the park.

 

Smith, A. P., Horning, N. & Moore, D. 1997. Regional biodiversity planning and lemur conservation with GIS in western Madagascar . Conservation Biology. 11 : 498-512.

In this study, lemurs were used as a test subject in Madagascar in an effort to develop a GIS-based approach towards rapid animal surveys and habitat modeling procedures designed to facilitate reserve selection. The distribution and abundance of lemurs was determined using stratified surveys throughout the study region in the rainforests of western Madagascar. Several environmental variables were incorporated into the GIS, including elevation, slope, and rainfall, in addition to a number of anthropogenic variables, such as village disturbance and domestic animal trails. From these data, the researchers observed that lemur abundance was highest in regions characterized by higher elevation and rainfall, and greater distances from villages and roads. Unfortunately, the areas that they identified as suitable lemur habitat did not correlate with the location of existing reserves. Their approach, however, revealed that stratified transect-based surveys in conjunction with GIS can be a useful and rapid approach to identifying potential habitat in areas under imminent threat from development.

 

Stoms, D. M., Davis, F. W., Cogan, C. B., Painho, M. O., Duncan, B. W., Scepan, J. & Scott, J. M. 1993. Geographic analysis of California condor sighting data. Conservation Biology. 7 : 148-159.

In this article, the authors used a GIS to examine the distribution of and habitat characteristics associated with the endangered California condor (Gymnogyps californianus). Their goals were to provide an inventory of condor habitats, to examine the relationship between condor activity patterns and specific habitat variables, and to characterize the spatial and temporal patterns in the distribution of wild birds. The primary habitat variable, land cover, was mapped over the entire historic range of the species by photointerpretation of available satellite imagery, and several land categories were defined, such as developed land, agricultural land and bare land. Condor point data were compiled from a number of sources, including field biologists, fire lookout personnel and ranchers, and entered into the GIS with their associated sighting info (e.g., date of sighting and bird behavior). From this, the researchers identified that the birds continued to use nearly 100% of their traditional range and that only five percent of this range is used for either urban or agricultural purposes. The recovery of the species, therefore, will depend upon the successful breeding and re-introduction programs already underway.

 

Yamada, K., Elith, J., McCarthy, M. & Zerger, A. 2003. Eliciting and integrating expert knowledge for wildlife habitat modelling. Ecological Modelling 165 : 251-264.

In this study, the authors were interested in comparing two methods for eliciting expert knowledge, and using these methods to model the distribution of sambar deer (Cervis unicolor) in a national park in Victoria, Australia. In the first method, they employed a quantitative GIS with a simplified graphical user interface that allowed knowledgeable individuals to directly enter their own deer sightings into the database by clicking on relevant maps. In the second method, the researchers asked the same individuals to comment on the habitat characteristics of areas where they most frequently encountered deer. Interestingly, the latter approach was deemed to be more useful in identifying potential habitat since near universal agreement was found in the description of habitat characteristics whereas individuals frequently disagreed on the actual location of sightings selected on the map in the former approach. Therefore, using the description of suitable habitat types provided by the expert individuals, the authors employed a GIS to identify areas in the park that could potentially support the animals. Their analysis indicated that the deer could potentially occur in all areas in the park.