Correlation of rangeland cover measures to satellite‐imagery‐derived vegetation indices

Abstract

Using field data from the U.S. Army's land inventory and monitoring program, a study was conducted to examine the utility in estimating the quantity of vegetation cover with satellite imagery across a large and complex rangeland. The study area used for this investigation was the U.S. Army's Yakima Training Center in central Washington. The Land Condition‐Trend Analysis (LCTA) program at Yakima Training Center has 202 permanent plots located in a randomly stratified manner across the installation. The principle measures taken along the transects were canopy cover and ground cover. Landsat Thematic Mapper imagery collected in May and August of 1992 was used in the analyses. The satellite data coincided with the beginning and end of the field data collection period and were used to derive various vegetation indices including Ratio, TVI, SAVI, and MSAVI. Analysis of correlation of rangeland cover measures and satellite imagery derived vegetation indices were performed. Correlation between the vegetation indices derived from the May image and the cover measures were found to be stronger than those between the August image and cover measures.