Geographic Scope
Is there an answer to mapping old growth?
An examination of two projects conducted with remote sensing and GIS
Robert A. Norheim
As can be inferred from the "Geographic scope" sections in the previous Chapter for the two projects, Pacific Meridian Resources and The Wilderness Society did not map precisely the same areas.
The most obvious difference is that Morrison followed the ecological region and mapped three National Forests in California whereas Pacific Meridian followed the administrative/ political boundary of Region 6 and did not map the California forests. However, even within the nine Forests that both projects mapped, there are significant differences.
For example, four portions of the Mt. Baker-Snoqualmie National Forest are administered by other National Forests (see map 1). There is an area between the North Cascades National Park Complex and the Cascade crest that is administered by the Okanogan National Forest. The area was split off from the rest of the Mt. Baker-Snoqualmie National Forest when the park complex was designated. Since it was contiguous with the Okanogan National Forest and because much of the area is in the Pasayten Wilderness that also extends into the Okanogan National Forest, the area is administered by the Okanogan. This area was mapped by Pacific Meridian, but not by Morrison (perhaps because most of it is already protected in designated wilderness). Similarly, there are two areas of the Mt. Baker-Snoqualmie National Forest to the southwest of Mt. Rainier National Park that are administered by the Gifford Pinchot National Forest. One area is immediately adjacent to the park, the other is disjoint. Both of these areas were included in the Gifford Pinchot National Forest by Morrison. Pacific Meridian mapped these areas with the rest of the Mt. Baker-Snoqualmie National Forest. For the purposes of this thesis, the areas are included in the figures for the Gifford Pinchot National Forest only.
A fourth area of the Mt. Baker-Snoqualmie National Forest, east of Mt. Rainier, is east of the Cascade crest and is administered by the Wenatchee National Forest. Neither Pacific Meridian nor Morrison mapped the area. Like the Pasayten area, over half of this area is already protected by wilderness designation and so would be of less interest for the purposes of The Wilderness Society (you really should see map 1).
However, the two projects disagreed on including portions of the Gifford Pinchot and Mt. Hood National Forests area east of the Cascade Crest (see map 2). These were not mapped by Pacific Meridian, but were mapped by Morrison. An argument could be made that because most of that area drains directly into the Columbia, it might be considered arbitrary to define a "crest" in that area; however, Pacific Meridian used established county lines as a boundary.
The extreme southern portion of the Mt. Hood National Forest, inside the Mt. Jefferson Wilderness, was mapped by Morrison as part of Mt. Hood National Forest but by Pacific Meridian Resources as part of the Willamette National Forest (by whom it is administered).
The Siuslaw National Forest has three portions: a south unit, a north unit, and the Oregon Dunes National Recreation Area (NRA). These areas were handled differently by the two projects. The NRA was not mapped by Morrison, and the north unit was not mapped by Pacific Meridian Resources. Both mapped the south unit, which is the largest of the three.
The eastern border of the rest of the forests in Oregon are defined by the crest, so there are no discrepancies in the areas mapped by the two projects for those forests (note that parts of the Siskiyou and Rogue River National Forests are actually in California; both projects mapped the Forests in entirety).
There are also some ownership discrepancies. In November 1988, Olympic National Park and Olympic National Forest swapped some lands, moving the boundary from section lines to ridge tops in several different places. Morrison used the new boundaries for the Olympic National Forest, while Pacific Meridian used the old section line boundaries. However, Morrison apparently did not have current ownership information for inholdings on the Olympic and Gifford Pinchot National Forests, whereas Pacific Meridian did have such information. Thus, for those forests, Morrison mapped all lands within the proclamation boundary, even inholdings not actually owned by the Forest Service, whereas Pacific Meridian mapped only lands belonging to the Forest Service (other than the boundary discrepancy mentioned above). Morrison also had up-to-date ownership coverages for the other Forests, so the areas mapped there match up well, although there are still some small differences.
The scope of my analysis in this thesis will be on four Forests: the Olympic, Mt. Baker-Snoqualmie, Gifford Pinchot, and Mt. Hood National Forests; i.e., Morrison's "phase 3", photo-interpreted Forests, which he feels were done much better than the other eight Forests. I will compare old growth locations only in areas mapped by both projects. For example, the areas mapped by Morrison east of the Cascade crest in the Gifford Pinchot and Mt. Hood National Forests are excluded, as is the area including the Pasayten Wilderness area, which was mapped only by Pacific Meridian. Old growth mapped by Morrison inside the proclamation boundary but not in federal ownership (as defined by Pacific Meridian) is ignored. Once again, the areas southwest of Mt. Rainier that are officially in the Mt. Baker-Snoqualmie but are administered by the Gifford Pinchot are included in the figures for the Gifford Pinchot National Forest only.
Table 5-1 summarizes the differences between the areas mapped and gives corrected figures that are the acreage of old-growth on only the areas of these four Forests that both projects mapped. The second acreage column is the acreage of the original as supplied by the originator. The "conversion change" column indicates the small amount of acreage gained or lost when converting the two original coverages as supplied by the originators to coverages that could be directly compared (resampling the 57 sq. meter The Wilderness Society raster data to a 25 sq. meter grid, and converting the Pacific Meridian Resources data from polygons to a 25 sq. meter grid). The figures in the last column are a much better basis of comparison of the two projects than are the figures in Table 4-1.
Table 5-2 shows that the comparisons between the amounts of old growth found by each study get much worse than the originally published figures with the subtraction of the areas east of the crest on the Gifford Pinchot and Mt. Hood National Forests. While the original comparison for the Mt. Hood National Forest is fairly close, over 30% of the acreage that Morrison found is removed from the comparison. The comparison on the Gifford Pinchot National Forest was not very close to begin with, and gets much worse with the corrections. The change on the Olympic National Forest is negligible, and the overall comparison for the Mt. Baker-Snoqualmie National Forest actually improves with the corrections.