Fj. Young et Rd. Hammer, Defining geographic soil bodies by landscape position, soil taxonomy, and cluster analysis, SOIL SCI SO, 64(3), 2000, pp. 989-998
The key Soil Taxonomy is based upon the idea that certain properties can be
used to define populations of soils from the soil continuum. The soil mapp
ing paradigm is that similar soil populations exist within landforms, High
taxonomic variability has been reported within numerous soil mapping units.
The hypotheses in this paper are that for large scale mapping (i) Soil Tax
onomy creates classes that are only partially related to landform and (ii)
more homogeneous soil classes exist if different defining constructs are us
ed. The objectives are to: (i) classify a sampling of soils within a 40-ha
upland pasture using both Soil Taxonomy and cluster analysis, (ii) identify
the distinctness and relationships of these samples to landforms, and (iii
) compare the geographic distributions of soil classes identified by Soil T
axonomy and cluster analyses. Ninety-four soil properties were measured fro
m 257 pedons along point transects. Cluster analysis identified three pedol
ogically and geographically distinct groups. A single cluster group was ide
ntified for soils within the "ridge" landform, whereas the "backslope" land
form was a mixture of all three groups, A significant relationship was foun
d for soil attributes and slope profile curvature within the backslope land
form, but predictive value was low. Soil Taxonomy produced 13 geographicall
y indistinct classes that were partially related to cluster groupings. Clus
ter analysis appears to be useful for revealing patterns of soil homogeneit
y and for identifying relationships among soil properties and landforms. Nu
merical analysis may be a helpful supplementary method for correlating soil
surveys with large soil databases, or for defining those soil attributes w
hich distinguish mappable bodies from the soil continuum.