Abstract:
Map delineations of soil and vegetation for a 14,000
acre (5,800 hectare) site in the Oregon Coast Range were
compared. Research objectives were to ascertain the
types of information that could be extracted from delineation
comparisons and to develop a methodology suited
to this purpose. The latter objective was achieved in
a preliminary study involving a small number of soil-vegetation
complexes. Data on shape similarity and areal
correspondence were collected using a digital planimeter.
The methodology developed was then applied to three data
sets: all delineations of two soil mapping units which
differed only in dissection, and delineations from a
random sample of other mapping units. Nonparametric
statistical procedures were employed to analyze the
data in terms of soil mapping units and in terms of
physiographic position. Little association was found
between soil and vegetation delineations overall. However,
vegetation and soil delineation shapes, as measured
using an elongation ratio, appear to be more similar in
the uplands than in the lowlands. Greater upland contrasts
in factors significant to both soils and vegetation
may contribute to this trend. Areal correspondence was evaluated with the intersection/union ratio for vegetation
and soil polygons. Ranks of vegetation communities
according to their areal correspondence with the soil
body were generally inconsistent. The dissected soil
showed the greatest amount of consistency in areal correspondence
between upland and lowland complexes. This
raises the possibility that dissection may be a more important
determinant of community distributions than is
physiographic position. Finally, shape similarity and
areal correspondence trends were associated in the uplands
but not in the lowlands. The more consistent shape
comparisons in the uplands may contribute to this dichotomy
of results.
