Home - Table of Contents

Final Report Summary: Washington Gap Analysis Project

Christian E. Grue, Kelly M. Cassidy, and Karen M. Dvornich
Washington Cooperative Fish and Wildlife Research Unit,
University of Washington, Seattle

We conducted the Washington State Gap Analysis within the context of 31 vegetation zones: 9 steppe, 9 westside mesic-wet forest, 11 eastside dry-mesic forest, and 2 high-elevation zones. Data and results are reported in both hard copy and digital format. The hard-copy format is a five-volume report (in press). Volume 1 is a description of current land cover and its conservation status. Volumes 2, 3, and 4 are atlases for herpetofauna, mammals, and birds, respectively, and Volume 5 is the gap analysis. Digital data will be available through the Washington Department of Fish and Wildlife.

Land Cover

Actual land cover within each vegetation zone was mapped by on-screen digitization using spectrally clustered 1991 Landsat satellite Thematic Mapper (TM) imagery as a backdrop. The protection status of each zone was assessed using: 1) the percent of the zone in Conservation Status 1 and 2 lands, and 2) a Conservation Priority Index: CPI = ((100 - % protected)/(100 - % converted)) * log(total area in the zone) where "% converted" refers to the percentage of the zone converted to agriculture or development and "% protected" refers to the percentage of the zone in Status 1 and 2 lands.

Statewide, the percentage of lands in Conservation Status 1 and 2 is 12%, but protected lands are unevenly distributed among vegetation zones. The six steppe zones (all < 6%, four < 1%) and the four Puget-Willamette Trough zones (all < 3%) have the least Status 1 and 2 lands. The percentage of Status 1 and 2 lands in other zones generally increases with elevation, with the Permanent Ice/Snow zone having > 97% of its area on Status 1 and 2 lands. When vegetation zones are ranked by Conservation Priority Index (CPI), the four zones with highest priority based on low protection status, high conversion, and importance in terms of size, are three steppe zones (the Palouse, Big Sage/Fescue, and Wheatgrass/Fescue zones) and one westside zone (the Willamette Valley zone). Of the seven zones of moderately high CPI, four are steppe zones and three are the remaining Puget-Willamette Trough zones. Thus, seven of nine steppe zones and all four Puget-Willamette Trough zones have high or moderately high CPIs. Overall, 51% of the steppe zones has been converted to agriculture; 70-88% has been converted in the three steppe zones with the highest CPI. In the Puget-Willamette Trough zones (which encompass the major metropolitan areas of the state), 40-67% has been developed or converted to agriculture, and none of these zones have more than 15% of their area in conifer forest, the natural dominant cover.


Distributions of terrestrial vertebrate species were modeled by intersecting range limits with suitable habitats (Fig. 1). We assigned codes to indicate habitat quality for each species based on ecoregion, vegetation zone, and land cover within the zone. Vegetation zones within an ecoregion were designated as "core" or "peripheral"; core zones were those in which the species was most common and peripheral zones were those in which the species occurred, but was rare or the zone was believed to be a population sink. Land cover was designated as "good," "adequate," or "contingently suitable" (i.e., suitable, contingent upon the availability of habitats below our minimum 100-ha mapping unit).

We assessed the protection status of vertebrates by: 1) calculating each species’ total predicted distribution, the percentage of its distribution on Status 3 lands, and the percentage of its distribution of Status 1 or 2 lands; 2) mapping vertebrate species richness of various taxonomic groups and assemblages by overlaying predicted species’ distributions; and 3) mapping areas of high vertebrate richness according to Conservation Status (Fig. 2). The effects of basing vertebrate richness analyses on presence/absence versus the most suitable habitats for each species were also explored. We found that presence/absence-based maps obscured the relative importance of low-elevation zones and habitats unaltered by human activity. All subsequent vertebrate analyses were based on the most suitable habitats for each species.

Amphibians: The number of native amphibian species is highest in mid- to late-seral conifer forests in low- to mid-elevation westside forest zones. Mid- to late- seral conifer forests in the Western Hemlock zone on the southern Olympic Peninsula and the southwestern Cascades have particularly high amphibian richness.

Reptiles: Native reptile richness is highest in the steppe zones and low-elevation eastside forest zones in steppe habitats, open forests, and forest openings.

Mammals: Habitats with high numbers of mammal species are riparian areas and forests in the Western Hemlock and Olympic Douglas-fir zones of the westside, and the Interior Western Hemlock, Interior Redcedar, and Grand Fir zones of the eastside, but the patterns of species richness vary greatly among mammalian subgroups.

Birds: Native bird richness is generally highest in low-elevation forests of the eastside and low-elevation wetlands throughout the State; however, the patterns of species richness varies considerably among avian subgroups.

We chose 10% representation on Status 1 or 2 lands to compare the relative protection status of taxonomic groups of vertebrates. For each group, the number of native species with less than 10% of their predicted distribution on Status 1 or 2 lands was:

Amphibians 14 of 24 (58%)

Reptiles 18 or 21 (86%)

Mammals 45 of 102 (44%)

Birds 138 of 230 (60%)

Other groups of interest included low-disturbance associates, state and federally listed species, and Columbia Basin-dependents. For these groups, the percentage of species with less than 10% of their predicted distributions on Status 1 or 2 lands varied between 38 and 100%.

For each species, we also calculated its total modeled distribution in Washington and the percentage of the modeled distribution on Status 1 or 2 lands. Though some caution must be used in comparing modeled areas between species at different trophic levels and in habitats of greatly differing productivity, our data do allow us to determine which species have a combination of low protection status and limited distribution, a warning sign of potential risk of extirpation.

Highest Conservation Priorities

Steppe zones and Columbia Basin-dependents: The most glaring gap in protection of biodiversity in Washington is in the steppe zones. The vegetation zones with the highest Conservation Priority Index (CPI) are steppe zones. Vertebrate species that rely on steppe usually have a correspondingly low percentage of their distribution on areas managed primarily for biodiversity.

Puget-Willamette Trough zones: These zones include the Puget Sound Douglas-fir, Woodland/Prairie Mosaic, Willamette Valley, and Cowlitz River zones. All have been heavily converted to both agriculture and development. The remaining forests are now a patchwork of hardwood, mixed, and early-seral conifer forest. There are only a few small areas of high richness of low-disturbance associates, as most of these species have been extirpated from these zones.

Ponderosa Pine and Oak Zones: These lowest elevation eastside forest zones have moderately high CPIs with less than 4% of their areas in Status 1 and 2 lands. They are zones of high reptile and avian diversity. Reduction in natural disturbance via fire suppression is a significant conservation problem in these zones.

Sitka Spruce and Western Hemlock Zones: These wet to mesic, westside forest zones have relatively little of their areas in development or agriculture, but logging has been extensive. They are zones of high amphibian and mammal (especially bat) richness, and their remaining mid- to late-seral forests support large numbers of amphibian, mammal, and bird species that adapt poorly to anthropogenic disturbance. Our data indicate that less than 8% of the Sitka Spruce zone and less than 10% of the Western Hemlock zone remain in late-seral forest; an additional 14% of the Sitka Spruce zone and 20% of the Western Hemlock zone were estimated to be in mid-seral forest.

Home - Table of Contents