A COMPARISON OF VARIOUS ESTIMATORS FOR UPDATING FOREST AREA COVERAGE USING AVHRR AND FOREST INVENTORY DATA

Various methods of adjusting low-cost and possibly biased estimates of percent forest coverage from AVHRR data with a subsample of higher-co st estimates from the USDA Forest Service's Forest Inventory and Analy sis plots were investigated. Two ratio and two regression estimators w ere evaluated. Previous work (Zhu and Teuber, 1991) finding that the e stimates from the two different data sources differed the most in high ly fragmented land-use areas led to an investigation into improving th e estimates through the use of independently derived estimates of popu lation density. It is concluded that reasonable updates of percent for est area could be obtained through both the ratio and regression estim ators and that use of population density as a surrogate for land-use f ragmentation could help improve the estimators, although more direct m ethods of measuring land-use fragmentation might provide further impro vement.