Mogollon Mtns strip.jpg

The Southwest

Arizona, New Mexico, Colorado & Utah

The technical papers and synoptic report featured on this page provide an overview of the history, current status, and future prospects of grizzly bears in Arizona, New Mexico, Colorado, and Utah. These states once had thriving grizzly bear populations, with individual bears surviving as last as the 1930s in Arizona and New Mexico. The last known grizzly bear in the Southwest was killed in 1979, in the San Juan Mountains of Colorado. Collectively, the analyses presented here describe ample opportunity for restoring grizzly bears to the Southwest, including robust populations in the Mogollon Highlands of Arizona and western New Mexico and the San Juan and Sangre de Cristo Mountains of Colorado and northern New Mexico.

You can download three technical papers and one report published by the Grizzly Bear Recovery Project on this page. The technical papers provide more detailed information on methods used to assess potential for recovery of grizzly bears in the Southwest: 1. a summary of contemporary grizzly bear densities and relations to habitat conditions; 2. a method for estimating potential densities as a function of habitat features; 3. a detailed geospatial analysis of recovery prospects. The report explores the deep history of grizzly bears and grizzly bear-human relations in the Southwest, a comprehensive appraisal of current potential for recovery, and an assessment of future prospects in the face of climate change and human population growth.

Sierra del Nido grizzly 1957.jpg

Modeling Restoration Areas for Grizzly Bears in the Southwest

Grizzly Bear Recovery Project Technical Report GBRP-TR-2021-3

by David Mattson & Troy Merrill

We appraised the suitability of Arizona, New Mexico, southern Utah, and southern Colorado for restoration of grizzly bears Ursus arctos horribilis by extending and integrating existing models of habitat capability and remoteness from humans, calibrated to historical grizzly bear locations in our Southwest study area. We applied previously published standards or new standards based on established concepts to identify areas productive enough and remote enough from humans to sustain grizzly bears locally, as well as habitat complexes that were capable of supporting robust grizzly bear populations because of large size and high quality. We identified three promising complexes of suitable habitat ranging in size from 8,000 to >15,000 km2 : the Mogollon, San Juan, and Sangre de Cristo Complexes, of which the last two potentially functioned as one conservation area. These complexes of suitable habitat were, in turn, the basis for delineating three candidate Recovery Areas with estimated average carrying capacities of 620, 425, and 281 grizzly bears, respectively. We also assessed these candidate Recovery Areas for overlap with areas offering additional protections (e.g., Wilderness Areas and National Parks) as well as additional prospects of human-grizzly bear conflict (e.g., private property and public land sheep grazing allotments).

Mattson Merrill_2021_Modeling restoration areas-1.jpg

Models of Grizzly Bear Density for Conservation Design in the North American Rocky Mountains

Grizzly Bear Recovery Project Technical Report GBRP-TR-2021-2

by David Mattson & Troy Merrill

We investigated relations between estimated grizzly bear Ursus arctos horribilis densities in 12 Rocky Mountain study areas and several potentially predictive or explanatory variables that included tassled cap transformed Thematic Mapper (TM) satellite imagery, the extent of whitebark pine Pinus albicaulis range, diet energy concentration, remoteness from humans, and study area size. Our objective was to develop models for predicting potential grizzly bear population sizes in areas currently unoccupied by bears, or supporting small and vulnerable populations. To test our models, we examined goodness of fit of predicted and observed densities in 5 additional study areas and determined whether predicted densities were spatially correlated with observations of grizzly bears in 2 regions. We also determined whether key predictive metrics correlated positively with a direct measure of habitat productivity in the Yellowstone region. Our best model included a single variable (Wetness from tasseled cap transformed TM imagery). This model optimized parsimony and fit and produced density predictions that correlated well with distributions of grizzly bear observations and fit observed densities in the 5 independent test areas. TM Wetness was also positively correlated with habitat productivity in the Yellowstone region. Based on this model, and without considering limiting human effects, we predicted that former 1850s range in Arizona and New Mexico could support 1905 bears (0–5059 prediction interval), that currently unoccupied but potential habitat in central Idaho could support 615 bears (443–757), and that habitat currently occupied by a very small (<40) and vulnerable population in northwestern Montana could support 362 bears (286–428).

Mattson Merrill_2021_Modeling GrizzlyBearDensities-1.jpg

Estimating Densities, Distributions, and Total Population Sizes of Extirpated Grizzly Bear Populations in the Contiguous United States

Grizzly Bear Recovery Project Technical Report GBRP-TR-2021-1

by David Mattson

In this paper I provide an estimate of pre-European grizzly bear (Ursus arctos) population sizes and distributions for the contiguous United States along with a rationale for this estimate based on relations between contemporary grizzly bear densities and coarse-grain environmental features. I also present a comprehensive database of density estimates made for grizzly bear populations throughout western North America that updates an earlier version presented by Mowat et al. (2013). Contemporary grizzly bear densities in North America are positively associated with average annual precipitation and temperature; evapotranspiration; terrain ruggedness; and areas dominated by herbs and shrubs. Densities are negatively associated with consumption of meat from terrestrial sources and densities of livestock. A weaker negative relation with human densities is also evident. At the time of first contact with Europeans, I estimate that there were roughly 47,300 (SD = 15,300) grizzlies in what was to become the contiguous United States, distributed over an area of approximately 2,864,000 km2 .

Mattson_2021_Estimating extirpated GB populations-1.jpg