Watershed Condition

Top conservation priorities for the coastal temperate rain forest bioregion include the protection and stewardship of large contiguous blocks of undeveloped land and water. Watersheds form a logical unit of analysis to identify such priority areas: the flow of water through a drainage basin integrates distribution of fish populations, flows of energy, and movement of materials. The analysis illustrated by Map 9 is based on a methodology developed by Keith Moore (1991) as part of an inventory of coastal watersheds in British Columbia.

The level of disturbance within the rain forest within primary coastal watersheds larger than 5,000 hectares is shown in relation to protected areas. Approximately 41 percent of the forested watersheds included in this analysis remain undeveloped. None of these largely pristine coastal temperate rain forest watersheds occurs outside of British Columbia and Alaska.

The analysis of watershed condition combined the human development layer with the boundaries of major watersheds within the region. Given the known area of coastal temperate rain forest in each watershed and the amount converted by human development within each, the portion affected by human activities was readily calculated as a percentage.

Highlights

Forty-one percent of the forested watersheds included in this analysis (343 of 827 watersheds) remain in a largely undeveloped condition, all of them in the more northern parts of the bioregion. These watersheds encompass roughly 3 million hectares of rain forest.
Only 11 of 46 watersheds larger than 100,000 hectares remain intact. Five are in Alaska: one on the Kenai Peninsula (the coastal temperate rain forest area is undeveloped, although the lower watershed has been settled), one in Prince William Sound, two in the Copper River Delta, and one on Icy Bay. These Alaska watersheds are somewhat anomalous, as much of the area is rock and ice, with a narrow strip of rain forest along the coast. Nonetheless, they are ecologically important as representative of this rare subpolar coastal temperate rain forest zone.
Six of the intact watersheds larger than 100,000 hectares are found in British Columbia: the Whiting, and the contiguous Unuk, Iskut, and Chickamin Leduc watersheds on the Alaska/British Columbia border, and the Cranberry and Kitlope watersheds on the north coast.
South of the Canadian border, only the Elwha River in Washington, Taylor Creek in Oregon, and the Big Sur River in California are less than 50 percent developed.
Of the 15 watersheds containing more than 100,000 hectares of rain forest, only the Wannock (8 percent developed) and the Nimpkish (46 percent developed) in British Columbia are less than 50 percent developed.
From Vancouver Island north, a substantial number of the coastal temperate rain forest watersheds are smaller than 5,000 hectares in size, and therefore not included in this analysis. They are nonetheless important, and any future analysis should include watersheds 1,000 hectares and larger.

Analyzing Watershed Condition

The analysis of watershed condition applied a definition of major coastal watersheds developed by Moore (1991) which includes all watersheds greater than 5,000 hectares having their terminus in salt water and encompassing the area from terminus to height of land. The lower limit of 5,000 hectares, though somewhat arbitrary, has been used as an appropriate minimum size for functional wilderness areas (Wilderness Advisory Committee 1986). Moore's definition was modified for those few major rivers extending inland through the coast ranges and flowing from headwaters in the interior valleys or the Cascades. Rather than delineating large watersheds which contain a small amount of coastal temperate rain forest relative to their full extent as single units, analysts delineated second-order streams (those flowing into the major river) for the portion of these watersheds within the coastal temperate rain forest zone.

The GIS was used to calculate the amount of coastal temperate rain forest in each watershed. The human development layer was then overlaid with the water-shed boundaries to calculate the amount of developed coastal temperate rain forest area within each watershed. This figure, divided by the total area of coastal temperate rain forest in each watershed, determined the percentage affected by human activity. Five development classes were distinguished according to the extent of human impact.

The Research Challange

Limits of the Present Analysis

While providing a reasonably accurate first approximation of the status of coastal temperate rain forest in the region, this study was limited by the type and availability of data. Two key elements of the analysis, the determination of original distribution of coastal temperate rain forests and the present extent of undeveloped forests, used data that provide indirect, rather than direct, measures of forest cover.

The distribution of forest ecosystems, results from the complex interactions of numerous factors. A combination of available moisture and mild temperatures is the primary determinant of the distribution of coastal temperate rain forests (Alaback 1991). An area receiving less rainfall but experiencing cooler year-round temperatures might provide growth conditions roughly comparable to another area receiving more rainfall but experiencing higher temperatures. Under ideal circumstances, the necessary preconditions would be observed directly, through physical measurements of soil moisture or evapotranspiration.

Unfortunately, such data are not collected on a regular basis. In the absence of direct physical measurements, available precipitation and temperature data sets, combined with auxiliary data sets including land cover and elevation, served as surrogate measures of environmental conditions. (British Columbia's data set -the coastal western hemlock zone — provides a better measure of the conditions of interest, since its classification scheme uses the distribution of vegetation, which integrates a host of climatic factors). Future studies can refine the distribution based on additional, more direct measures of environmental conditions.

One goal of this analysis was to determine the amount of remaining natural forest of any age (i.e. not just old growth) within the coastal temperate rain forest zone. It relied on existing data sets describing forest cover for various areas within the region. These data sets record timber age and size. crown closure, or specific indicators of old-growth habitat. They do not distinguish forest regrowth after logging from regrowth after natural disturbance. Studies of old-growth forests have also indicated the difficulty of distinguishing old growth from older second-growth forests, making the dis tinction between never-logged forests and very old regrowth on logged areas somewhat arbitrary. (This is not an issue in British Columb ia or Alaska, where logging activity is more recent.)

This analysis assumed that areas not specifically classified as mature or old-growth forest had been logged. While some of these areas may in fact be younger forests resulting from natural (windthrow or landslides) rather than industrial disturbance, this would not significantly affect the results of this study. Fire is common only on southern Vancouver Island and in the more southern portions of the North American coastal temperate rain forest. Conversely, some areas included as mature forest have likely been logged at some time, but may effectively be functioning as natural forests. Future studies that incorporate new data sets can provide a more accurate assessment of natural forests.

Two types of information, though beyond the scope of this analysis, are important from a conservation perspective. Distribution of key coastal species (salmon, the anadromous smelt known as eulachon, and marbled murrelets, among others) would indicate key watersheds for conservation and provide valuable information for management plans. Additional information on land ownership is indis pensable for development of an overall stewardship strategy for the bioregion.

Priorities for Future Research

This analysis, based on the best regional-scale data available, provides a valid region-wide picture of the status of coastal temperate rain forests. There is room to improve both data and methods, especially to support local-level analyses, as well as to update the rapidly changing status of critical areas within the bioregion. Specifically:

Consistent application of a comprehensive definition of this forest type would better delineate the coastal temperate rain forest bioregion.
Better data on evapotranspiration, soil moisture, or growing season degree-days would strengthen the analysis. Such data would support more sophisticated models of the original distribution of coastal temperate rain forests.
Stepping the elevation criteria for coastal temperate rain forest along a south-to-north gradient would provide a more accurate picture of forest area (This has already been done for the coastal western hemlock biogeoclimatic zone in British Columbia.)
More detailed and timely analysis of forest condition using images collected by satellites, particularly for areas outside national parks and national and provincial forests, is an urgent need. Improved means of distinguishing natural forests (at any age) from managed forests should be a goal of such analysis. (The Sierra Club of British Columbia is working on this type of analysis for the British Columbia coast.)
Coastal streams flowing from watersheds smaller than 5,000 hectares are especially common from the mid-coast of British Columbia north through the Gulf of Alaska. The exclusion of such areas from the analysis by no means represents a conclusion about their potential importance to regional conservation. Future analyses should delineate these watersheds.
The adequacy of protected areas within the region could be better assessed by indicating the levels of protection achieved by existing parks and other legally protected areas. An improved data set containing up-to-date information on new protected areas, revised boundaries, and altered legal status would improve the analysis.
Delineating key vegetation types within the coastal temperate rain forest zone (e.g. wetlands, peatlands, riparian, and subalpine types) and structural classifications would help to capture the interactions between forest and wildlife.