STRATIFIED ESTIMATES OF FOREST AREA USING THE k-NEAREST NEIGHBORS TECHNIQUE AND SATELLITE IMAGERY

For two study areas in Minnesota, stratified estimation using Landsat Thematic Mapper satellite imagery as the basis for stratification was used to estimate forest area. Measurements of forest inventory plots obtained for a 12-month period in 1998 and 1999 were used as the source of data for within-strata estimates. These measurements further served as calibration data for a k-Nearest Neighbors technique that was used to predict forest land proportion for image pixels. The continuum of forest land proportion predictions were separated into strata to facilitate stratified estimation. The variances of the stratified forest area estimates were smaller than variances based on simple random estimates by factors as great as 5, and when including all plots over a 5-year plot measurement cycle, the forest area precision estimates may be expected to satisfy national standards.