Craighead Environmental Research Institute

Conservation Area Design Conservation Biology & Genetics Conservation GIS Conservation Education About Us Links Join Us The mission of the Institute is to increase humankind's understanding, appreciation, and protection of our natural environment; particularly wildlife populations and wild landscapes.  Our goal is to enable human beings to live in harmony with other species.

Coastal Forest and Mountains Ecoregional Plan (area shaded blue on map) CERI is involved with 5 interrelated Conservation Area Designs on the Pacific Coast of British Columbia and Southeast Alaska. These began as separate projects but have converged into 3 complementary CADs. Initially, Round River Conservation Studies developed a CAD for (1) the Central Coast of British Columbia: CERI was involved in review of the CAD and conducted site visits in conjunction with Round River and the Raincoast Conservation Society. Recently, Round River completed a (2) CAD for the North Coast of British Columbia along with a refinement of the Central Coast CAD. CERI has been involved in review, model inputs, and site visits. The methodology used for the Central and North Coasts will be applied, and refined, for the (3) Transboundary Watershed CAD which will complete a Conservation Area Design for the coast [areas outlined in red, magenta, and yellow] based upon grizzly bear habitat, salmon, and old growth forest as the primary data layers. At the same time, The Nature Conservancy and Nature Conservancy Canada are conducting (4) the Ecoregional Plan for the Coastal Forests and Mountains Ecoregion which is outlined in blue on the map. This Plan will incorporate three approaches: representational analysis, focal species analysis, and special elements analysis, and should be complementary to the Round River CAD. Comparison and contrast of the two approaches will teach us about the relative strengths and weaknesses of each approach, especially in relation to designing conservation areas on this vast scale with limited data resources. On a finer scale, Round River is completing a (5) Conservation Area Design for the Taku River drainage. This is a more detailed CAD incorporating high resolution wildlife habitat models, traditional ecological knowledge, and extensive field work. CERI is assisting with model development, field work, and CAD review.  The Nature Conservancy's Coastal Forests and Mountains Ecoregional Plan In the Coastal Temperate Rainforest regions conserving and maintaining habitat connections is critical for a high quality of life; for human communities, plant communities and animal communities, and implementing Conservation Area Designs in this region is the most effective way to accomplish this. The Craighead Environmental Research Institute is cooperating with, Round River Conservation Studies [RRCS], The Nature Conservancy, Nature Conservancy Canada, Ecotrust, the Yellowstone to Yukon Conservation Initiative, and the members of Transboundary Watershed Alliance. We also work with academic institutions (University of Northern British Columbia [UNBC] and Montana State University) and government agencies (US Forest Service, Alaska Department of Fish and Game, the BC Ministry of Sustainable Resource Management, and the Interagency Grizzly Bear Study Team [IGBST]). CERI's work is designed to dovetail with ongoing analyses that are being conducted with co-operators in these areas. Round River Conservation Studies is developing habitat suitability and least-cost-path analyses for the Transboundary Watershed region in collaboration with CERI, and the Transboundary Watershed Alliance. Lance Craighead is helping to direct GIS model development, and to conduct analyses to evaluate the correlation with known GPS locations of animals and the correlation with other models. CERI's work is designed to dovetail with ongoing analyses that are being conducted with co-operators in these areas. Nature Conservancy Canada and The Nature Conservancy (U.S.) are developing habitat suitability models for the Coastal Forests and Mountains Ecoregion in collaboration with Round River Conservation Studies, CERI, Ecotrust, and The U.S. Forest Service: Lance Craighead is directing the development of wide-ranging mammal habitat models for this project. CERI will help implement the results of regional-scale (1km2 grid) focal species models of core and movement habitat for a study area encompassing the Coastal Forests and Mountains Ecoregion from the north end of Vancouver Island to Glacier Bay (in cooperation with Round River Conservation Studies and Nature Conservancy Canada). The role of CERI biologists/conservationists includes: 1. Conducting site-specific ecological evaluations of the habitat and the movement barriers in areas that have been pinpointed by GIS-model results. 2. Conducting interviews with local biologists, sportsmen, outfitters, and landowners to gather additional information on animal movements and conservation possibilities. 3. Conducting workshops in local communities to disseminate the information, which is the best available scientific data, and to elicit additional comments and local participation and support. Providing data and other information into the planning processes of the U.S. Forest Service, Tongass National Forest; the Alaska Department of Fish and Game; the U.S. Fish and Wildlife Service, The BC Ministry of Sustainable Resource Management, and the B.C. Wildlife Branch, as well as local governments. Wide-ranging mammals strategy: Lance Craighead Justification Using a suite of wide-ranging mammals as focal species for ecoregional planning in the Coastal Forests and Mountains Ecoregion should serve at least three primary conservation goals. 1) it should ensure that populations of those species have sufficient intact protected habitat to persist for perpetuity, 2) as a suite of 'umbrella species' it should provide a measure of redundancy to coarse filter representation analysis for the protection of diverse arrays of plants and animals which are ecologically interrelated with the focal species, and 3) it should serve as an umbrella for ecological functions and processes. In regards to the latter goal we need to recognize that our understanding of ecosystems is incomplete; the use of coarse filter representation and optimization routines to determine portfolios do not include systems-type analyses of ecosystem processes and functions. Wide-ranging mammals, especially carnivores, can presumably be sustained only by healthy, functioning ecosystems, and so we use their populations as an index to ecosystem intactness, and we use their habitat requirements as a guide to protecting and maintaining biodiversity; especially those components of biodiversity which we cannot identify or measure separately. Overview Because of the diversity of the entire Coastal Forests and Mountains Ecoregion, it has been subdivided into 7 sub-regions based upon plant distribution and other biogeoclimatic differences. Five primary wide-ranging mammal focal species have been chosen: brown bear, black bear, black-tailed deer, gray wolf, mountain goat, and river otter. If time permits, other species may be modeled: wolverine, moose, etc. Habitat suitability (or capability) models will be used to delineate potential habitat restricted by known distribution (ie: Although there is wolf habitat on Admiralty Island there have been no wolves historically, so the wolf model will not be used there). Targets for focal species will be chosen based upon long-term viability of populations. Area requirements will be estimated for populations or metapopulations based on best available science for long-term genetic diversity and population persistence (ie: a genetic effective population size of about 500, or a census population of about 2000). Goals will be set for the desired number of populations in each subregion (to provide redundancy in case of local extinction). Currently, the most detailed and accurate habitat suitability models for this region have been developed for the Tongass Land Management Plan. Although these were developed from data for specific study areas within the Tongass, we feel that they will be applicable to the larger region. Data layers currently exist to run these models for the Alaska portion of the CFM ecoregion. Analagous data exist for the British Columbia portion of the ecoregion: differences in habitat classifications (forest structure and volume) will be crosswalked to provide a uniform classification layer across the ecoregion. Model parameters will be adjusted to recognize changes in input layers. The refined Tongass models will be used to create maps of potential habitat for each focal species across the ecoregion. These results will be compared with other current modeling approaches throughout the region (see below), and will be reviewed by regional biologists and modelers. If the refined Tongass models are deemed to be less accurate than other models during the review process, the better models will be used as the basis for the ecoregional plan. Brown/Grizzly bear models: · Tongass LMP model: based upon resource-selection-functions derived from radio-collared bear locations on Admiralty and Chicagof Islands. Habitat capability values were assigned to 20 habitats. This model uses habitat information available in the Tongass GIS. Very accurate for the ABC Island subregion. A human disturbance layer was incorporated into the model, and this will be updated to include recent human developments. Estuaries, avalanche chutes, salmon streams, and low-elevation old growth are important. · Round River Conservation Studies model: based upon stand structure, abundance of salmon, and human disturbance intensity. Currently completed for the mid- and north-coast of BC. Should be completed for the entire CFM ecoregion by June. This model should provide the best comparison for the CFM model. · McCrory-Paquet model: based upon relative habitat capability for grizzly and black bear of each BC biogeoclimatic subzone variant; values were determined by an expert panel for a study area on the mainland portion of the Spirit Bear preserve. Capability was based upon a plant food value layer overlaid layer of salmon values, estuaries, and old-growth winter den habitat. · BC Wildlife Branch grizzly bear habitat capability models: developed by Tony Hamilton for the Central Coast, and Kalum LRMPs. Models and data have been requested. Deer/Wolf models Because wolves are extreme habitat generalists and do not exhibit strong preference for any given habitat, wolf models are based upon the habitat needs of primary prey species. In Southeast Alaska the primary prey is white-tailed deer. Winter habitat, particularly snow depth, is the limiting factor for white-tailed deer populations, and thus wolf populations. · Tongass LMP model: based upon habitat coefficients derived from a number of habitat-use and food habits studies in Southeast Alaska plus slope, aspect, and elevation. A human disturbance layer was incorporated into the model, and this will be updated to include recent human developments. This model uses habitat information available in the Tongass GIS. · McCrory-Paquet model: based upon habitat used by deer in summer-fall and winter-spring seasons for the Spirit Bear Preserve area, modified by slope. Biogeoclimatic subzone variants were assigned values. Elevations above 500 m were omitted as well as slopes > 60 degrees. · Darimont-Paquet model: Chris Darimont, a graduate student of Paul Paquet is currently developing a deer-wolf model that incorporates more recent research findings; including the utilization of salmon by wolves along the coast, and the importance of estuaries for wolf den sites. Black Bear models · Tongass LMP model: based upon habitat coefficients derived from a number of habitat-use and food habits studies in Southeast Alaska plus slope, aspect, and elevation. A human disturbance layer was incorporated into the model, and this will be updated to include recent human developments. This model uses habitat information available in the Tongass GIS. · McCrory-Paquet model: based upon relative habitat capability for grizzly and black bear of each BC biogeoclimatic sub zone variant; values were determined by an expert panel for a study area on the mainland portion of the Spirit Bear preserve. Capability was based upon a plant food value layer overlaid layer of salmon values, estuaries, and old-growth winter den habitat. This is basically the same model used for grizzly bears on the mainland, but applied to the BC islands where black bears have historically excluded grizzly bears. Otter models · Tongass LMP model: based upon habitat coefficients derived from a number of habitat-use and food habits studies in Southeast Alaska. Food availability, although important, is difficult to define in a habitat database, so cover and proximity of habitat was the model focus. Old growth forest for cover and den sites, within 150 m of shorelines or in riparian habitat are important. A human disturbance layer was not incorporated into the model, and this may be included to reflect avoidance of heavily disturbed areas. This model uses habitat information available in the Tongass GIS. Goat models · Tongass LMP model: based upon habitat coefficients derived from a number of habitat-use and food habits studies in Southeast Alaska plus slope, aspect, and elevation. The primary factors are availability of food (plant community and aspect) and escape terrain. A human disturbance layer was incorporated into the model, and this will be updated to include recent human developments. This model uses habitat information available in the Tongass GIS. Other models The habitat capability model for gray wolves is tied directly to habitat capability models for Sitka black-tailed deer, moose, and mountain goat. The assumption is made in this model that gray wolves will first select large ungulates as prey and utilize beaver as maintenance prey when ungulates are not plentiful. There are currently no habitat capability models for moose or beaver in the region, and we assume that the deer model will be sufficient for identifying important wolf habitat throughout the region. A wolverine model may provide an index to additional aspects of biodiversity, but there is probably not enough data on habitat use, particularly denning habitat, in this ecoregion. It is likely that the bear models and the coarse filter approach will protect wolverine.