A STUDY OF RELATIONSHIPS BETWEEN SOIL SURVEY DATA AND AGRICULTURAL LAND USE USING INFORMATION THEORY

Abstract

The significant advances which have been made on inventories of agricultural land use and in updating soil survey information provide an opportunity to compare these new sources of information to identify which of the various land factors shown on soil survey maps correlate most closely with agricultural land use. Three study areas representing three levels of land use intensity were used. Statistics of information theory were evaluated for application in studies of this type. Results demonstrate that a statistically significant correlation (adjustment) between physical land conditions and land use has occurred in all three areas. However, the degree of correlation varied according to the land use and the types of land factors considered. Increased correlation was achieved when land uses were weighted by economic investment. The factor called "soil association" was identified as correlating best with land use, but factors such as surface texture, capability class, slope and "spatial interaction" were also significant. Key words: Land use, soil survey, statistics, information theory, capability, economic investment