Mapping surface soil organic carbon for crop fields with remote sensing.

The soil organic carbon (C) concentration of surface soil can be used in agricultural fields to vary crop production inputs. Soil organic C is often highly spatially variable, so that maps of soil organic carbon can be used to vary crop production inputs using precision farming technology. The objective of this research was to demonstrate the feasibility of mapping soil organic carbon on three fields, using remotely sensed images of the fields with a bare surface. Soil samples covering the range in soil organic carbon were taken from each field to develop a satisfactory relationship between soil organic carbon content and image reflectance values. The regression equations differed between fields, but gave highly significant relationships with R² values of 0.93, 0.95, and 0.89 for the three fields. Two classification, arbitrarily classification and a minimum distance clustering, were applied in mapping soil organic carbon. A comparison of mapped soil organic carbon with an independent set of 21 soil samples taken on one of the fields gave highly satisfactory results for the two methods with r² = 0.81 and 0.87. [ABSTRACT FROM AUTHOR]