Select a country... Brazil Canada Central Africa Chile Indonesia Russia United States Venezuela < Home

English Français Español Bahasa

United States: Initiatives · Overview · Forests · Publications & Maps · News

United States' Forests

Excerpts from Low-Access Forests and Their Level of Protection in North America (PDF)

This analysis presents the results of a map-based analysis of the location and status of North American forests (excluding Mexico) that remain mostly undivided by roads and other access routes, so called low-access forests.  It provides a regional look at where large tracts (larger than 200 square kilometers [km²]) of low-access forest are located, as well as an assessment of the degree to which these tracts are currently protected.  Because of data limitations, these results create only a coarse picture at a continental scale, of the location and status of large, low-access forest tracts.  The results are useful for identifying forests that, due to their limited development, offer opportunities for expanding protected area networks and/or for restoration, as well as priority areas for future mapping to characterize intact forests at finer scales.

Why Identify Low-Access Forests?

Fragmentation and Degradation of Natural Forests

Approximately 20% of North American forests have been permanently cleared for agriculture and other uses, primarily within the last two centuries (Bryant, et al. 1997).  Currently, forest cover is stable (Matthews, et al. 2000); however, in most of the lower 48 states and southern Canada, remaining forests have experienced significant human disturbance and do not possess the same degree of ecological integrity as the original forest.  As human populations grow, forest fragmentation and degradation continues.  One result has been the loss of extensive areas of old-growth forest. According to one estimate, stands of century-old forest now account for only 7% of forest cover in the United States (USDA-FS 2000).  Consequently, remaining large tracts of relatively undisturbed natural forest are increasingly important, for several reasons:

1. Conservation value.  Large patches of natural forest provide sufficient area for natural ecological processes which shaping the forest ecosystem that cannot be sustained across smaller areas.  For instance, large forest areas are able to preserve some habitat intact in the face of periodic natural disturbances, such as fires.  Large tracts of forest also provide habitat for far-ranging species, such as wolves, wood buffalo, and elk.  As such, these areas serve as a reservoir for the successful colonization of smaller patches of habitat, especially those too small to maintain themselves over the long term.
2. Ecosystem goods and services value.  Forests provide a range of products and life support services essential to humans and other species.  Among the ecosystem services provided by forests are the maintenance of water quality, storage of carbon (which might otherwise contribute to global climate change), and regulation of local climatic processes (e.g., rainfall patterns).  Even intensely modified forests, such as plantations, supply such benefits, to some degree.  However, because of their condition and extent, large tracts of relatively undisturbed forest possess the greatest potential value in terms of ecosystem services for a given forest type.  In addition, these areas often include our greatest reservoirs of mature and old-growth timber stands and other commercially significant natural resources.  In many countries, including Canada, the timber industry relies heavily on harvest of old-growth and primary forest stands, which are a dwindling resource.  In order to balance timber needs with the need for non-market ecosystem services, many of which are associated with relatively undisturbed forests, it is important to know how much natural forest remains and where these areas are located.
3. Recreational, aesthetic, and heritage values.  As populations grow and natural forest is converted to other uses, the remaining large tracts of relatively undisturbed forest are increasingly valued for their natural heritage and for the opportunities they afford in terms of recreation and the experience of wilderness.

As forests in North America continue to be fragmented and degraded, there is growing public debate concerning the management of large tracts of relatively undisturbed natural forest, particularly on public lands. Some argue that remaining areas should be closed to further development in order to maintain their biological diversity and for their recreational and wilderness values; others urge development and use of the natural resources these forests contain. This debate is exemplified by the current controversy over the Clinton Roadless Rule in the United States , as well as by the lobbying of environmental groups in the United States and Canada against continued logging in what various call primary, old-growth, or intact forests.

Currently, continental-scale forest monitoring efforts track only changes in forest cover. Little integrated information exists about forest condition, especially the location and status of large tracts of relatively undisturbed forest. In the absence of such information, two indicators provide a useful proxy for data on forest condition:

• Access.  The presence of roads and other access routes is an excellent indicator of human disturbance of the forest. Roads, deforestation, and forest fragmentation are intimately related. Through physical, chemical, and biological mechanisms, roads affect the terrestrial and aquatic environment of the forest in many significant ways. Road building in natural forest areas is accompanied by increased erosion, air and water pollution, spread of invasive exotic species, increased animal and plant mortality, and habitat fragmentation (Trombulak and Frissell 2000). Even more important than the direct damage to natural ecosystems, the access to forest areas provided by roads leads to subsequent human disturbances from activities such as logging, mining, grazing, agriculture, and urban development. These disturbances result in substantial declines in native species and an overall degradation of ecosystem integrity.

• Size of forest blocks.  As noted above, large blocks of forest are more likely to contain and support a full complement of native species, including wide-ranging mammals. Large tracts also permit natural disturbance regimes, such as fire, to shape these ecosystems (Bryant et al. 1997, Crowley 1978).

Approach

The objective of the analysis reported on here is to provide a coarse-scale picture of the location of North American forests (excluding Mexico) that have been only minimally disturbed by recent human activity, such as logging, other commercial-scale activities, or development-induced fragmentation. To this end, we first identified all forests at excluding logging roads, and designated these areas as low-access forest. Evidence indicates that, in boreal and deciduous forests, fragmentation and conversion often occur at 0.3-0.9 km from human infrastructure, including roads (UNEP 2001).

Therefore, for the purposes of this assessment, we assumed that 1 km is the critical distance beyond which human pressures are less prevalent (Canadian Council of Forest Ministers 1997).  Next, we identified forested areas larger than 200 km² and classified these as large tracts of low-access forest. The rationale for selecting a threshold of 200 km² is as follows.

1. Because of the coarse scale of the analysis, small roads, logging roads, and other access routes are not included in the regional transport data. Thus, the actual extent of truly low-access forests is overestimated here. The use of relatively large block sizes in the analysis will tend to counterbalance this effect.

2. To maintain populations of far-ranging species and/or preserve some habitat intact in the face of periodic natural disturbance (e.g., fires in the boreal forests), forest areas must occupy a least 500 m from a road or other access route, minimum patch size, which varies considerably according to ecosystem type and species native to the area. For example, World Wildlife Fund-Canada (1999) suggested that a single protected area should exceed a minimum threshold of 500 km² in order to accommodate landscape-scale ecological processes in the boreal forest. Similarly, Greenpeace Russia and Global Forest Watch used a threshold of 500 km² for mapping intact forest landscapes in northern Europe, a region characterized by boreal forests (Yaroshenko, et al. 2001). The Nature Conservancy (2001) applied a block size of at least 60 km² for “ecological land units” in its Central Appalachians ecoregional planning process.

For this continental-scale analysis, a conservative threshold of 200 km² was selected. This midrange value was chosen in part to account for the wide variety of ecosystem types and, as indicated above, to offset the impact of incomplete road-access data. Table 1 and Figure 1 present estimates and proportions of how results would change if a higher (500 km²) or lower (less than 100 km²) threshold for tract size were used.

It is important to note that this approach provides only one measure of forest condition. For other useful indicators—such as, stand age, tree species types, patch size, and patch shape—continental scale data are difficult to acquire. Several of these indicators have been applied in other regional assessments.

Methods

Using Geographic Information Systems (GIS), we analyzed at coarse level forest blocks, their relative intactness, and their protected status. The approach entailed:

• Developing a 1- km² transportation grid, effectively creating a 500 m buffer around access routes. Transportation layers were at a scale of 1:100,000 and included primary and secondary roads, pipelines, and utility lines, where possible;

• Overlaying the access-routes grid on a spatial land-cover dataset (1992-3) at a resolution of 1 km². The land cover data were modified to depict two forest types: forests and woodlands (canopy cover of 10-60%) and dense forests (canopy cover greater than 60%);

• Identifying as “low-access forests” all forests outside of the access-routes grid;

• Delineating “large tracts” (larger than 200 km²) of low-access forest;

• Identifying the degree to which these large blocks of low-access forest are protected. For this purpose, we used existing datasets for Canada and the United States, classified by degree of protection according to internationally recognized World Conservation Union (IUCN) standards (see Box 2 ). IUCN Categories I-II are strictly protected, while Categories III-V are moderately protected. This standard was chosen so that these results can be compared with those from other countries in which similar Global Forest Watch mapping is underway.

For details and caveats regarding the analysis of this assessment, please refer to the Technical Notes in the report.

Results

Almost half of today’s forests and woodlands in North America (excluding Mexico) still qualify as large tracts of low-access forest; however, all but a fraction (5%) of this area is located in the northern most regions of the continent, namely boreal Canada and Alaska.  Most forests of southern Canada and the lower 48 states have been extensively disturbed by human activity.  In the lower 48, only 6% of forest cover remains in large, low-access tracts, mostly located in the Rocky Mountains and the Pacific Northwest , in portions of the Cascade Range and in the Klamath-Siskiyou of southwest Oregon and northern California.  Relatively undisturbed large tracts of dense forest, generally the most productive, account for15% of all remaining forests in the United States and Canada.

Major Findings for the United States

• About three-quarters of low-access woodlands and forests in the lower 48 are in tracts smaller than 100 km².

• Two-thirds of all large tracts (larger than 200 km²) of low-access forests in the United States are found in Alaska. About one-quarter (26%) has protected status, either moderate or strict. Close to 11%—including wilderness areas— are found within the Tongass and Chugach national forests (74% of the Pacific Coast forests in Alaska is managed by the US Forest Service).

• Just six states account for more than 60% of large, low-access forest tracts in the lower 48. Ranked in order of relevant land area, they are Idaho, Montana, Washington, California, Wyoming, and Minnesota.

• In the Pacific Northwest, most low-access forest blocks are located in portions of the Cascade Range as well as in the Klamath-Siskiyou in southwest Oregon and northern California. East of the Rocky Mountains, low-access forest blocks are located primarily in northern Minnesota, Maine, and the Adirondack region of New York.

• Approximately 30% of large, low-access forest blocks in the lower 48 states are found within national forests in the states of Idaho, Montana, Wyoming (northern Rocky Mountains), and Washington. Idaho, Montana and Wyoming house 43% of all Inventoried Roadless Areas (IRAs) in the lower 48 states. For some of the large-low access forests blocks, their protected status is under review, under the roadless-area logging ban (By definition, many of the large, low-access forest tracts identified by this analysis are included within the US Forest Service’s IRAs). Approximately 40% is strictly or moderately protected in parks and reserves.

Major Findings for Canada

• Two-thirds of Canada’s forests remain in large, low-access tracts.¹ Much of this area is in the central and northern portions of the country, characterized by slow-growing forests and open woodland. Dense forests account for just under a third of this area.

• About half the country’s low-access forests are located in Quebec, the Northwest Territories, and Manitoba.  Alberta, British Columbia, and Ontario house over half of Canada’s large, low-access tracts of dense forests, i.e., those with greatest potential timber value.

• Only 4% of large, low-access forest tracts are strictly or moderately protected.

For details by region, country, and state or province, please refer to Tables 1 and 2.

Protected Status at the Regional Level

Most (91%) large, low-access forest tracts in North America are located outside strictly or moderately protected areas, i.e., parks and reserves classified in IUCN categories I-V. About 36% of large tracts dominated by dense forests are strictly or moderately protected. However, as indicated above, the degree to which these tracts are protected varies significantly between the United States and Canada. Although vast areas of these forests remain in the far northern regions, most of these tracts are threatened not only by logging but also by mining and oil and gas development. In the United States as a whole, about 34% of large tracts of low-access forests are strictly or moderately protected.

One recent analysis concluded that the current system of protected areas in the United States fails to preserve a representative sample of the country’s biological richness. Most of the protected areas are located at higher elevations and on low-productivity soils (Scott, et al. 2001). That study indicated that most types of forest communities in the United States are underrepresented in the protected area system. Not surprisingly, this analysis shows that large blocks of protected, low-access forest are located mostly in mountainous areas or at high latitudes.

Another study determined that, if granted permanent protected status, IRAs could play a significant role in creating a representative network of conservation reserves in the United States (DeVelice and Martin 2001). As mentioned above, by definition, many of the large, low-access forest tracts are included within IRAs. National forests contain almost a third of all remaining large, low-access forest tracts in the lower 48. Although some such tracts are already strictly or moderately protected, granting permanent status to IRAs would significantly extend protection of large, low-access forests in the United States .

In the Kalamath-Siskiyou in southwest Oregon and northern California, roadless areas provide a wide range of important ecological attributes that maintain ecosystem integrity. For example, they contain significant numbers of threatened and rare species, special habitats— such as old-growth forests and serpentine geology. They also sustain key watersheds for aquatic biodiversity, and provide connectivity between existing wilderness areas (Strittholt and DellaSalla 1999).

Because of their size and distance from transport routes and associated human pressures, large tracts of low-access forest often offer important conservation opportunities. For example, significant areas of low-access forest—in the Cascades and the Rocky Mountains of Washington as well as in parts of northern Maine and northern Minnesota — remain outside of the protected areas system. Here, wilderness, recreational, and biodiversity values might be further enhanced through establishment of new parks and reserves or through management regimes that emphasize conservation values as well as extractive activities (e.g., certified logging operations).

Next Steps

This analysis provides only a coarse picture, using proxy measures, of the location and status of large, relatively undisturbed forests in North America. Careful management and stewardship of those forest areas in North America and elsewhere that are not yet significantly degraded or fragmented requires data of greater accuracy at a much finer resolution.

To address this information gap, Global Forest Watch is in the process of mapping forest conditions in several global regions. The first step in a two-phased approach entails identifying large, low-access forest blocks (as was done here), using existing datasets on land cover and roads and other transport routes. Similar analyses have been completed for forests in Central Africa and Indonesia. These coarse-scale assessments provide a rough picture, comparable across major regions, of the location of remaining large tracts of relatively undisturbed forest. This assessment for North America builds on earlier work by Global Forest Watch-Canada (Smith and Lee 2000) and extends it to provide a region-wide picture, including neighboring forests of the United States.

Coarse-scale assessments provide a starting point for the second phase of GFW mapping efforts, which features finer-scale analysis. This second-phase work usually incorporates high resolution satellite imagery, which is used to identify logging and other transport routes not shown in existing roads datasets and then further eliminate forest tracts accessed by these logging roads, including those that have undergone recent logging and other extractive activities (e.g., oil and gas development). Final maps depict at detailed scales what we refer to as “intact natural forests”—i.e., forests with few or no signs of recent, commercial-scale human activities and of sufficient size to maintain viable populations of resident species in the face of periodic natural disturbance. Such mapping has been completed for Chile and Russia and is now underway for Alaska and Canada. Similar work is planned for portions of the lower 48 states of the United States (Note that, in this phase of work, threshold tract sizes for “large” blocks of intact forests are defined according to forest type).

Caveats

As a coarse-scale analysis, this study attempts to provide only a rough picture of forest areas that are potentially intact and to assess their protection status. The accuracy of our results is affected by the limitations of those input datasets available for use in a continental-scale assessment. For example:

• Our analysis fails to pick up disturbance from nonlinear features, such as clearcuts and mines. There are no existing comprehensive datasets, at a national or regional level, that depict the location of forests that have been impacted by such disturbances.

• Because of the input datasets and scale, our analysis does not evaluate disturbances from small roads, logging roads, and seismic lines, which significantly degrade and fragment forests. This is particularly true for forests in rural areas of the United States , where existing transportation data, in many cases, underestimate the presence of roads (Karl, et al. 2001). Preliminary Global Forest Watch analysis for the forests of New England provides an indication of the degree to which coarse-scale mapping underestimates the extent of forest access. Here, we found a 40% increase in access routes when switching from medium-scale (1:100,000) to a finer-scale (1:24,000) mapping, due in large part to logging roads not appearing in medium-resolution coverages.²

• Because the land cover used for this assessment depicts percentage of woody vegetation per grid cell, pixels do not depict patterns at a finer scale (resolution below 1 km). Consequently, when examined at a finer scale, some areas showing a high percentage of  tree cover might prove to be highly fragmented; to the extent that this is true, the area of low-access forests is overestimated.

• Moreover, since the dataset was created (1992-3), the distribution of forest cover has changed, due to recent agricultural, urban, and other development.

• Maps are views of the surface of the earth, and representations on paper of a three dimensional object will always be somewhat distorted in shape, size, or direction, depending on the map projection used.

Sources

Bryant, D., D. Nielsen, and L. Tangley. 1997. “The Last Frontier Forests: Ecosystems and Economies on the Edge.” 42. Washington, DC : World Resources Institute. (Sections of the report available online at http://www.igc.org/wri/ffi/lff-eng/lff-toc.htm). (3/15/02 ).

Canadian Council of Forest Ministers. 1997. “Criteria and Indicators of Sustainable Forest Management in Canada.” 137.  Ottawa, ON: Canadian Council of Forest Ministers. (Available online at: http://www.nrcan.gc.ca/cfs/proj/ppiab/ci/pdf/ci_e.pdf). (3/15/02 ).

DeFries, R.S., J.R.G. Hansen, A.C. Janetos, and T.R. Loveland. 2000. “A New Global 1-km Data Set of Percentage Tree Cover Derived from Remote Sensing.” Global Change Biology 6. 247-54.

DellaSalla, D.A., N.L. Staus, J.R. Strittholt, A. Hackman, and A. Iacobelli. 2001. “An Updated Protected Areas Database for The United States and Canada.” Natural Areas Journal 21. 124-35.

Dent, B.D. “Cartography Thematic Map Design, Fifth Edition”. 1999. WCB/McGraw-Hill.

Karl, J., P. Morrison, L. Swope, and K. Ackley. 2001.“Wildlands of the United States.” Winthrop, WA: Pacific Biodiversity Institute. (Available online at:http://www.pacificbio.org/pubs/wildlands_of_the_united_states.htm).

Matthews, E., R. Payne, M. Rohweder, and S. Murray . 2000. “Pilot Analysis of Global Ecosystems, Forest Ecosystems.” 90. Washington, DC : World Resources Institute. (Available online at: http://www.wri.org/wri/wr2000/forests_page.html). (3/18/02).

McGhie, R.G., J. Scepan, and J.E. Estes. 1996. “A Comprehensive Managed Areas Spatial Database for the Coterminous United States.” Photogrammetric Engineering and Remote Sensing 62. 1303-06.

Scott, M.J., F.W. Davis, B. Csuti, R. Noss, B. Butterfield, C. Groves, H. Anderson, S. Caicco, F. D’Erchia, T.C.J. Edwards, J. Ulliman, and R.G. Wright. 1993. “GAP Analysis: A Geographic Approach to Protection of Biological Diversity.” Wildlife Monographs No. 123: The Wildlife Society.

 back to top