1. Executive Summary
2. Introduction
3. Results
4. Discussion
5. Conclusions
6. Methodology
7. Acknowledgements
8. Literature Cited
9. Printable Version

Methodology

To identify the extent, causes and distribution of mountain lion mortalities caused by humans we compiled mortality records and data on available mountain lion habitat in the 11 western states of Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington and Wyoming. Mortality records were obtained from state wildlife agencies and from state status reports published in the Proceedings of the Mountain Lion Workshops (Padley 1996, Becker et al. 2003, Harveson et al. 2003, Beausoleil et al. 2005). Mortality records were categorized by year [1], state, management area [2], reason (e.g., sport hunting, depredation, public safety, and unspecified other) and sex of animal.

The terminology used for management areas varies from state to state. Arizona reports mortality data by game management unit (GMU). Where GMUs had subunits we pooled data into the GMU to simplify analysis. California reports kill data by county. Colorado and Idaho report kill data by Data Analysis Unit (DAU) which are comprised of several game management units. Montana reports mortalities by GMU and Management Region; we used Regions. New Mexico reports data by Management Zone, which are comprised of several GMUs. Nevada, Utah and Wyoming report data by Hunt Areas. Oregon reports data by Management Zone, which are comprised of a number of GMUs. Washington reports kill data by Cougar Management Units, each of which contain several GMUs. We will simplify the terminology and reference areas as management units.

For states that reported mortalities by harvest year (which overlap calendar years) we report by the year in which the season began (e.g., 1998-1999 listed as 1998) to facilitate inter-state comparisons. States reporting lion mortalities by calendar year included Arizona, California, Idaho, Oregon and states reporting mortalities by harvest year included Colorado, Montana, Nevada, New Mexico, Utah, Wyoming, and Washington. This report focuses on human-caused mortalities from 1997 to 2004, the longest recent period for that we were able to acquire kill mortality reports records for all western states and management units.

Because states (and management units) differ with regard to the amount of suitable mountain lion habitat contained within their borders, comparing absolute number of mountain lions provides insight into the extent but not intensity of human-caused mountain lion mortality. Therefore, we standardized mortality by calculating an annual kill density (defined as number of mountain lions killed per 1000 square miles of suitable mountain lion habitat) for each state and management area . This approach provides a useful metric for evaluating the intensity of exploitation across the various western states and management areas. However, since methods of classifying and estimating the amount of suitable habitat vary among states it should be considered a more accurate measure of relative kill density when comparing management units within the states than across states.

To determine the extent of suitable habitat for mountain lions in each state we used a combination of two sources: 1) predictions and maps of suitable mountain lion habitat generated by the national gap analysis program (GAP; see www.gapanalysis.nbii.gov) when available; and 2) estimates derived by state wildlife agencies. Table 5 contains a summary of sources and methods used for deriving estimates of suitable habitat for mountain lions.

Table 5. Sources and methods for estimating amount suitable mountain lion habitat in states and management units in the American West

State	Sources	Method of Estimation
		Management Unit	State
Arizona	Arizona Game and Fish Department (unpublished data)	Estimates by AZGF regional managers	Sum of management unit estimates
California	GAP (Davis et al. 1998) and CDFG Wildlife Habitat Relationship Model (Torres and Lupo 2000)	Wildlife Habitat Relationships Model (CDFG) based on GAP	Wildlife Habitat Relationships Model (CDFG) based on GAP
Colorado	GAP (Schrupp et al. 2000) and Colorado Division of Wildlife	Based on CDOW regional manager estimates	Sum of management units
Idaho	GAP (Scott et al. 2002)	MLF estimate based on GAP	GAP
Montana	GAP (Redmond et al. 1998)	MLF estimate based on GAP	GAP
Nevada	Ashman et al. (1983) and Nevada Division of Wildlife	MLF estimate based on map by Ashman et al. (1983)	NDOW estimate and total of MLF estimate of management units
New Mexico	GAP (Thompson et al. 1996) and New Mexico Department of Game and Fish (unpublished data)	NDGF estimate based on GAP	NDGF estimate based on GAP
Oregon	GAP (Keister and Van Dyke 2002) and Atlas of Oregon Wildlife (O'Neil et al. 2001)	MLF estimate from Atlas of Oregon Wildlife	Atlas of Oregon Wildlife estimate
Utah	GAP (Edwards et al. 1995)	Utah Division of Wildlife data based on GAP estimate	GAP estimate
Washington	GAP (Cassidy 1997)	Washington Department of Fish and Wildlife data based on GAP estimate	GAP estimate
Wyoming	GAP (Merrill et al. 1996) and Wyoming Game and Fish Department (2006)	WGFD estimates when available, augmented with MLF estimate based on GAP	GAP estimate

At the time of our analysis, GAP reports that included numerical predictions of suitable mountain lion habitat were available for California (Davis et al. 1998), Colorado (Schrupp et al. 2000), Idaho (Scott et al. 2002), Montana (Redmond et al. 1998), New Mexico (Thompson et al. 1996), Utah (Edwards et al. 1995), Washington (Cassidy 1997), and Wyoming (Merrill et al. 1996). The GAP report for Oregon (Kagen et al. 1999) contained a map of predicted suitable habitat but no numerical estimates and GAP reports had not been completed for Arizona or Nevada. Consequently, alternative methods were used to estimate suitable habitat in these states as described below.

• For Arizona, we used estimates of suitable habitat provided by Arizona Game and Fish Department that were made by regional managers for game management areas under their jurisdiction. We summed these estimates to determine an estimate of all suitable mountain lion habitat in Arizona under the jurisdiction of Arizona Game and Fish Department. However, this Figure does not reflect the total amount of habitat in the state as it does not include lands under the jurisdiction of the National Park Service or tribal governments.
• For Nevada, we derived a an estimate of suitable habitat by overlaying a mountain lion habitat map created by Ashman et al. (1983) and a map of management areas provided by Nevada Division of Wildlife onto a 1-cm by 1-cm grid and estimating the percent of suitable habitat within each management area. We then summed these estimates to create a state wide estimate. Because our total estimate for the state was only slightly more than the 50 thousand square miles of suitable habitat estimated by Nevada Division of Wildlife (Woolstenhulme 2005) we use our estimate herein.
• For Oregon, we considered maps of mountain lion habitat produced by GAP (Kagan et al. 1999) and the Atlas of Oregon Wildlife (O'Neil et al. 2001) as well as unpublished data provided by the Oregon Department of Fish and Wildlife. The GAP and the Atlas of Oregon Wildlife maps estimated that roughly 17 and 50 percent of the state is suitable habitat, respectively. Alternately, ODFW estimates that 96 percent of the state contains suitable habitat. The reason for the large disparity between these estimates is unclear. To maintain relative consistency in methodology among states we chose to use the estimate based on the Atlas of Oregon Wildlife map, which was published subsequent to and based upon the Oregon GAP project.

Next, to estimate the amount of suitable habitat in individual mountain lion management units, we used the following procedures:

• For California, Colorado, New Mexico, Utah and Washington we used estimates provided by state agencies derived from GAP data (UT, WA) or were modified from GAP data (CA, CO, NM).
• For Idaho, Nevada, and Oregon, where state agencies were unable to provide suitable habitat estimates at the management unit scale, we developed an estimate by overlaying a statewide map of management units and a map of mountain lion habitat generated by GAP (Idaho) or other studies (Nevada, Oregon see above) onto a 1-cm by 1-cm grid and counted the number of cells in each management area. We then counted the number of cells that indicated suitable mountain lion habitat and determined the percentage of suitable mountain lion habitat within each management area. Finally, we multiplied the total area in square miles of each management area as provided by state agencies by the percentage of suitable habitat in that area to derive an estimate of suitable mountain lion habitat for each management area.
• For Wyoming, we used estimates of suitable habitat as derived by Wyoming Game and Fish (Wyoming Game and Fish Department 2006). For those management areas where this information was not available from WGDF, we used the same technique as described for Idaho, Nevada and Oregon above based on GAP.
• For Montana, where Montana Fish Wildlife and Parks was unable to provide either estimates of suitable mountain lion habitat or total area of management units, we overlaid a map of management regions and map of suitable habitat generated by GAP onto a 1-cm by 1-cm grid and counted the total number of cells in the state and in each region and the total number of cells in each region indicating suitable habitat. To estimate the total area of each region in square miles we divided the number of cells in each area by the number of cells statewide and then multiplied this number by the land area of the entire state according to the 2000 U.S. Census. Finally, to estimate the amount of suitable habitat in each area we multiplied the total estimated area of each area by the percentage of cells in the area indicating suitable mountain lion habitat and rounded the result to the nearest 1000 square miles.

 Next Section