干旱区生态环境知识资源中心

Research in digital soil mapping has indicated that its methodologies could be successfully extended to a field setting where it could enhance the quality and scientific foundation of soil surveys, as well as save time and money. These assumptions are being put to the test in the Mojave Desert of California as part of a continuing effort within ongoing soil surveys. One of the questions being posed in this study is how comparable is a third-order soil map created by the conventional soil survey approach to one created via digital soil mapping. To compare the two mapping approaches the subgroup level of Soil Taxonomy was chosen as the response variable and the map unit as the unit of comparison. Within each map unit the proportion and number of subgroups predicted by each mapping approach was qualitatively compared. Within the intermontane basins the predictive model used for digital soil mapping, estimated a smaller proportion and number of soil subgroups associated with fan remnants. Whereas within the mountains the conventional approach predicted fewer soil subgroups, and a different composition of soil subgroups.