GIS Analysis of Mountain Lion, Lynx, and Ocelot in North America.

Geography 565: Annotated Bibliography.

Steven W. Diehl, Jr.

Large Carnivores principally mountain lions are free roaming and once inhabited the entire North American Continent. Their numbers decreased steadily as deer populations shrunk, and they were hunted because they were seen as a pest or threat by the local human population. Now they inhabit a small section of southern Florida, and most land west of the Great Plains. In recent years, GIS has become available to analyze data about mountain lion movements and patterns, which is gathered by field studies, and can be used to draw interesting conclusions. That they may be coming back to areas they have not been for decades. Very few studies have been published about the mountain lion so data on other feline species, the lynx and the ocelot, has been included to show that more then one feline species has been affected by man in recent history and has been studied.

Articles.

Chetkiewicz, Cheryl-Lesley, Mark S. Boyce. 2002. Carnivores and Corridors in the Crownest Pass. Alberta Species at Risk Report No. 50. Alberta Sustainable Resource Development, Fish and Wildlife Division. December : 1-52.

It is a research paper whose main goal was to figure out “resource selection functions and movement models at different spatial and temporal scales” on mountain lions (cougars) and grizzly bears in Crowsnest Pass Alberta. The area focused upon was the Crowsnest River Valley in southwestern Alberta which includes a wide variety of habitat types and a major highway and railroad link. Data was gathered from individual cougars and grizzly bears that were captured and fitted with GPS collars. Their positions were taken every 4 hrs, and was downloaded by radio telemetry on a weekly basis in 2001 to 2002. The data was placed into a GIS system and the positions taken were plotted on remote sensored data images and on maps. The GIS then figured out the homerange sizes of the cougars and grizzly bears. The data from this study will be used to inform and support decisions by habitat management agencies and conservation organizations. Habitat and movement models of grizzly bears and cougars will identify and validate wildlife areas that could focus on habitat retention and protection efforts by the NCC, ACA and SRD. GIS models can be used for habitat enhancements efforts by trying different scenarios in the models and seeing what results they achieve to predict the consequences of different habitat management policies. The models can define wildlife habitat and movement corridors. and the models may assist managers develop migration options.

Clark, David W., Staffany C. White, Annalea K. Bowers, Leah D. Lucio, Gary A. Heidt. 2002. A Survey of Recent Accounts of the Moutain Lion (Puma Concolor) in Arkansas. Southeast Naturalist Volume 1, Issue 3 (September) : 269-278.

The article gives a brief history of mountain lion incidents within the state up to the present involving physical evidence. The study was to set up a clearinghouse of mountain lion information, to find out the relation between reports, location, and deer harvest, and examine geographic relationship between captive and wild mountain lion sightings. A series of questionnaires were mailed to hunting groups, the USDA, the Arkansas Fish and Game Commission (AFGC) records, and unsolicited reports from Arkansas citizens. They classified the data into 4 categories of importance, ranging from hard evidence to highly questionable reliability. The responses from the questionnaires were examined and their locations plotted in ARC INFO using maps with boundaries, roads, and lakes and rivers information. They included information from an earlier study for comparison. And sighting and sign location were calculated using Spatial Analysis extension for ARC/INFO. The GIS software was a small part of the study but it gave them useful data for their study. Most of the sightings were in the Ozark and Ouachita Mountains remote forested areas. And some were in the Gulf Coastal Plain. They concluded that there isn’t enough hard evidence available to say there is a permanent breeding population of wild mountain lions living in the state. They are most likely escaped captive mountain lions roaming free. The number of reports and physical evidence has been increasing in recent years though.

Cramer, Patricia C. Kenneth M. Portier. 2001. Modeling Florida panther movements in response to human attributes of the landscape and ecological settings. Ecological Modelling 140 : 51-80.

This was a study to help plan for the reintroduction of the Florida panther back into the state beyond the areas it now occupies. It is a highly endangered species and the state and local conservation agencies would like expand its current territory range and population in the state. They felt using GIS models would help them in the planning process. They created, PANTHER, “an individual-based spatially explicit module”. It would predict how Florida panthers would use the northern Florida area, to see the effects of human development and roadways on panthers, and to create conservation ideas for strategies to best preserve the landscapes native environment for panthers. They focus on the development of PANTHER and show the results of human influence upon panther movements. The study area included a 5 county rural area in northern Florida along the Florida-Georgia border encompassing Suwannee River Basin and the Osceola National Forest. Maps showing land cover, roads, deer densities, and human densities were loaded into the GIS simulation.

The program uses C++ object oriented computer language to simulate individual panther movements in the study model to theoretically follow the path of each panther released into the study area in the simulation. Maps showing land cover, roads, deer densities, and human densities were loaded into the GIS simulation as raster data. They are assigned different values ranging from most to least favorable to panthers. The panther then decides which terrain to walk into based on the values of the terrain. The panther “looks” at its environment , “chooses” where to move next, and then “moves” there. The panther interactions with one another is programmed into the model as well. Several scenarios were run, one without human presence, one as a baseline including all the information, and two future scenarios one for 2020 with as a worse case scenario with increased human presence, and one with the most conservation measures enacted in the year 2020.

The baseline scenario came out as expected. Scenario one with little human presence the cougars moved more freely and ranged wider. Cougars favored hardwood hammock land cover, followed by pine forests. When looking at connectivity issues, the cougars ranges were constricted significantly when roads and human presence were introduced acting as a barrier to their normal movement choices. The results show that as human population and settlement increase, panther population and habitat decrease. As humans change the environment it will affect the reintroduction of panthers to northern Florida. Territory that the panthers preferred were also preferred by humans and it’s the panthers who are forced out into less favorable habitats. It also showed the more frequently the panthers crossed roads the higher the mortality rate became for the panthers. It was concluded that PANTHER was successful in investigating conservation issues surrounding the Florida panther. And Florida state conservationists and planners can use this program to help form future conservation strategies.

PANTHER can be used for other species besides panthers by changing a few variables.

Harveson, Patricia M. , Michael E. Tewes, Gerald L. Anderson, and Linda L. Laack. 2004. Habitat use by Ocelots in south Texas: implications for restoration. Wildlife Society Bulletin, 32(3) : 948-954.

The study focused on examining the ocelot in the Lower Rio Grand River Valley (LRGCV). Its range has been vastly reduced by urbanization and land cultivation to the point it is greatly endangered. The study was exploring the relationship between canopy cover and soil composition to determine the best possible habitat for the ocelot. They used USDA-NRCS soil survey maps, and 8 yr old radiotelemetry location data collected from adult , subadult, and juvenile ocelots in the field in their analysis using a GIS. This study tried to match soil types with vegetation type to determine the best likely possible habitat sites for use by ocelots in the LRGV. Ocelots were found to prefer dense cover vegetation in their home ranges 95% of the time over open range plains. The type of cover they prefer covers less than 1 percent of the site study area. They concluded that they need to acquire land with certain soil types and reintroduce native plant species to recreate the dense canopy cover the ocelots prefer. The strategy needs to be part of an ocelot restoration plan in order to restore the ocelot to the LRGV.

Hunter, Richard D., Robert N. Fisher, and Kevin R. Crooks. 2003. Landscape-level Connectivity in Coastal Southern California, USA, as Assessed through Carnivore Habitat Suitability. Natural Areas Journal Volume 23 (4) : 302-312.

This initial “landscape level connectivity” study used a GIS system to develop data models focusing on two species, the mountain lion, and the bobcat in coastal southern California. It is felt mountain lions are a prime candidate for the evaluation of connectivity on large spatial scale. Bobcats are seen as ideal candidates for studies on a smaller spatial scale. They identified vegetation habitat for the mountain lion and bobcat. Then they designed GIS models to show the disturbance impacts of roads and development on their territory and finally combined the habitat relationship model with the disturbance impact model to create a “characterterization of habitat connectivity” for both bobcats and mountain lions. The models identified core mountain lion and bobcat territory, locate key connection areas between the core territories, and help show the level of protection needed by gap analysis.

The results of the models showed that the most suitable habitat for both species was in very mountainous terrain away from human settlement. Both species habitiat was fragmented into small patches by human settlement. They posed a significant barrier to mountain lion and bobcat movement. The results of these large scale analysis’s will be used to help inform coastal southern California planners and conservationists to develop management and conservation plans. They are preliminary studies to help focus future research efforts targeting connectivity studies.

Pike J. R., Shaw J. H., Lesle Jr. D. M., Shaw M.G. 1999. A Geographic Analysis of the Status of Mountain Lions in Oklahoma. Wildlife Society Bulletin 27, no. 1 : 4-11.

This study was a used to develop a statewide database of mountain lion sightings, determine which areas in Oklahoma had those sightings, compare the location and frequency of mountain lion sightings in different types of terrain. They mailed out survey questionnaires to natural resource professionals in the private, state, and federal sectors in Oklahoma in 1996. In each questionnaire was included a map asking the person to locate and date each sighting they observed. They obtained GIS base maps containing rivers, lakes, highways. Deer harvest units were digitized into the GIS and mapped for 1985-1995. General topography data was digitized into the GIS and divided into 3 elevation regions. Human population densities were assigned to each county. And the data from the surveys were plotted after they were received. They were all analyzed in the GIS utilizing regression analysis, descriptive statistics, and Statistical Analysis system and came out with some interesting results. As the number of deer increased so did the number of mountain lion sightings. Mountain lions preferred tall grassland with hilly terrain with steep slopes that they could use for cover. The Central Rolling Red Plains ecoregion provided the lions with such terrain. “Mountain lion sightings correlated negatively with human population density.” They felt the number of mountain lion sightings is dependant upon the size of the deer population and the number of people in the area. Areas with little human presence and large deer populations will see an increase of mountain lion populations. The study concludes that using questionnaires directed at natural professionals is a valid means to acquire data on a large scale.

Singleton, Peter H. , William L. Gaines, and John F. Lehmkuhl. Landscape Permeability for Large Carnivores in Washington: A Geographic Information System Weighted-Distance and Least-Cost Corridor Assessment. Research Paper. PNW-RP-549. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station : 1-89.

The study was done by the Washington State Department of Transportation and the USDA Forest Service, Pacific Northwest Research Station. They saw the states highway system as a barrier to large carnivore movement between forested areas in Washington State and souther British Columbia and wanted to study this effect using a GIS system analysis. This study “used GIS system weighted-distance and least-cost corridor analysis’s to evaluate landscape permeability” for the large carnivores: Grizzly Bear, Lynx, Gray Wolf, and the Wolverine. The results of the study found that the forests of north, central, and south cascades and coast range of Washington and the British Columbia Coast Range, the Selkirk-Columbia Mountain Ranges, and the Kettle Monashee Mountains of eastern Washington.were core areas for the focal species and that Snoqualmie Pass, Okanongan Valley, Fraser Coquihalla River valley, and the upper Columbia and Pend Oreille River valleys were significant barriers to these speicies movements between the core areas. Therefore these populations were isolated by one another because of human occupation and highways of the passes and valleys. The study is intended to provide information in the planning process for future conservation strategies, to help with future survey efforts and to help set management priorities.

Woolf, Alan, Clayton K. Nielsen, Theodore Weber, Tara J. Gibbs-Kieninger. 2002. Statewide modeling of bobcat, Lynx rufus, habitat in Illinois, USA. Biological Conservation 104 : 191-198.

This study was sponsored by the Illinois Department of Natural Resources (IDNR). They wanted information to review the bobcat as a threatened species, and develop a statewide management plan. They used “bobcat sighting data, habitat information from satellite imagery, multivariate statistics, and a GIS to model presence/absence, relative abundance, and habitat suitability” for bobcats. They used canonical discriminant function analysis to figure out the presence/absence and the “relative abundance” of bobcats in each county. The presence/absence model used 2 group canonical discrimination functional analysis. The relative abundance model used three group canonical discriminant function analysis. The results of the statistical analysis were compared to actual bobcat sighting locations for validation. They showed that the models were fairly accurate with their results in predicting the absence/presence of bobcats, and the amount of bobcats that were present in those counties that had them present. They used stepwise logistic regression (SLR) analysis to produce a suitability habitat module map for bobcats. The suitability habitat model showed that 29 % of the state had somewhat suitable habitat for bobcats with the most favorable in the southern most section of the state. The models were validated when sighting data was incorporated into the models and verified the accuracy of the results. The two class model was 75 % correct. The three class model was less accurate. The SLR suitability habitat model is likely to be more favored by managers because it provided a finer scale depiction of bobcat habitat relationships, and it “identified fewer significant variables” which made it easier for making conservation decisions. The models indicated that bobcats favored wooded habitat because they provide den sites and have a greater abundance of prey.

The studies results resulted in the bobcat being delisted as a threatened species in 1998 by the Endangered Species Protection Board. The bobcat was shown to be distributed widely enough across the state to support that decision in the models.

Online GIS Analysis:

Mangiamelihttp, John . Citizen-initiated Conservation Planning and GIS. Available from: http://gis.esri.com/library/userconf/proc99/proceed/papers/pap174/p174.htm.

The Sonoran Desert Protection Plan (SDPP) has decided to use GIS to look at possible locations of a focal species, Puma Concolor, that is endangered by suburban outgrowth and increased human encroachment upon its traditional home range. The study used base maps of landscapes, terrain types, government zoning, land ownership, parks, distribution of puma concolor habitat, to manufacture a habitat distribution, landscape characteristics beneficial and harmful to the focal species map, and combine the habitat distribution and beneficial and harmful landscape map into a final map to show the best possible locations that would be ideal for puma concolor in the state of Arizona. The findings state that Mountain Lion habitat is mainly within mountainous terrain, which lies within government lands, and that the species is fragmented and separated by human occupied lands, therefore connectivity issues need to be addressed for the benefit of the species. The final map will be used to as a tool to help form conservation plan suggestions to state and local agencies.

Informative online information:

Freeman, Debbie, Arizona Game and Fish Department. 2005. Mountain lion goes high tech. http://www.paysonroundup.com/section/localnews/story/20774

Grand Canyon National Park Foundation. 2005. Mountain Lion/Human Interactions. http://www.gcnpf.org/projects/mtnlions.html.

KGUN9. 2006. Researchers use satellite technology to track mountain lions. http://www.kgun9.com/story.php?id=1598.

Washington Department of Fish and Wildlife. 1999. The Gap Analysis Concept.

http://www.wdfw.wa.gov/wlm/gap/dataprod.htm

Contact me: dragon67892000@yahoo.com