Comparison of Stocking Rates From Remote Sensing and Geospatial Data

Abstract

Remote sensing data from the Advanced Very High Resolution Radiometer (AVHRR) have coarse spatial resolution (1 km2 pixel size) and high temporal resolution, which can be used to estimate net primary production regionally. The normalized difference vegetation index (NDVI) is used to determine the fraction of absorbed photosynthetically active radiation, which is sensitive to differences in growth caused by a large year-to-year variation in precipitation. The 12-year average of net primary production was used to calculate stocking rates in animal-unit months per acre for the state of Wyoming. Stocking rates were also calculated for Wyoming from 1:500 000 scale soil and climate geospatial data layers based on stocking rates from the US Department of Agriculture Natural Resources Conservation Service Technician Guide to Range Sites and Range Condition. In a pixel-by-pixel comparison, there was a weak but significant correlation between the 2 methods based on the spatial distribution of precipitation. There were classes of vegetation type for which the AVHRR data predicted either much lower or much higher stocking rates. More work needs to be done to reduce geospatial data uncertainties for the determination of stocking rates from both NDVI and stocking rate tables. Remote sensing indicates the actual condition of vegetation, so this is an important step in the development of regional forecasting of range condition, trend, and projected stocking rates for decision support tools.