use of AVHHR imagery for large-scale forest inventories

Abstract

Satellite-based remote-sensing observations provide a continuous source of data useful for evaluating forest vegetation over large areas. The Advanced Very High Resolution Radiometer (AVHRR) sensor aboard the National Oceanic and Atmospheric Administration's (NOAA's) series of polar-orbiting environmental satellites is well suited for large-scale studies, providing worldwide coverage daily in five spectral bands with a spatial resolution of 1.1 km. Research is being conducted by the Forest Inventory and Analysis unit of the Southern Forest Experiment Station, USDA Forest Service, on the use of AVHRR data for forest-inventory applications. Digital image data from the AVHRR sensor were used in an unsupervised clustering procedure to produce generalized land-cover classifications of three states in the southern United States. Statewide estimates of forest area were generated for Arkansas, Louisiana, and Mississippi and were compared with recent forest-survey estimates. All AVHRR-based estimates were within 5% of the ground-based forest-survey estimates, and the estimate for Louisiana was within 1%. A more detailed analysis of Louisiana revealed a very high correlation between AVHRR and forest-survey estimates at the parish (county) level. Though not appropriate for detailed forest cover-type mapping, the frequent repeat cycle, large area coverage, spectral characteristics, and relatively low cost of AVHRR data make it attractive as a potential component of a forest inventory-update model, especially when combined with higher-resolution imagery such as from Landsat. This type of data could also be useful for developing countries, which do not yet have a complete inventory of their forest resources, and for monitoring tropical deforestation. © 1990.