ESTIMATING SOIL TEMPERATURE REGIMES IN FOREST AND RANGELAND SOILS USING THERMAL INFRARED REMOTE SENSING (INFORMATION SYSTEMS, SOIL TAXONOMY, CALIFORNIA)

Thermal infrared remote sensing has been used to measure and monitor moisture conditions of vegetation and soils on croplands, but only rarely on forest and rangelands. Moreover, thermal infrared data have not been used much for estimating other soil properties or for supporting soil taxonomy research needs. Thus, my research objective was to determine the degree to which remotely sensed thermal infrared data can be used to spatially estimate forest and rangeland soil temperature regimes.;Thermal infrared spectral data were collected twice daily on two dates using the Thematic Mapper Simulator. Coincidentally acquired ground measurements of canopy and soil radiant temperatures and site variables were made to determine the relationship between remotely sensed thermal data and soil-vegetation properties. Diurnal temperature data provided improved methods for stratifying forested landscapes into mapping units having similar plant and soil temperatures. Of three field sampling methods used to measure radiant temperatures of several forest cover types, a circular plot method provided data having low variability and close correspondence to remotely acquired temperature data.;Two regression-based estimators were used to estimate mean annual soil temperature for several sites in the Plumas National Forest, California. The ability of a multiple regression model to predict the mean October soil temperature was limited due to the difficulty in estimating four of the eight variables over large remote areas.;A soil information system was developed to spatially estimate and map soil temperature regimes (STR). The overall percent correct of the STR map was 79.8 percent with commission errors of 78 percent and 5 percent for the frigid and mesic STR's, respectively.;Three studies were conducted in several forest cover types to: (1) evaluate the comparability of airborne- and ground-acquired thermal infrared data; (2) characterize soil temperature regimes using existing statistical models; and (3) map soil temperature regimes using a multi-thematic digital data base.;Pedological research should be directed toward the development of soil information systems for (1) soil survey using multivariate stratification of landscapes based on the soil forming factors; (2) soil survey interpretations, and (3) monitoring soil management practices in forest and rangeland environments.