Regional‐scale comparisons of vegetation and soil wetness with surface energy budget properties from satellite and in‐situ observations

Abstract

This study focuses on relationships between vegetation cover, soil moisture, precipitation and evapotranspiration for the Midwest and Great Plains region of the central United States in early June 1988, the year of the Midwest drought. The data base consists of: (1) microwave estimates of soil moisture made using data from the Special Sensor Microwave/Imager (SSM/I); (2) Normalized Difference Vegetation Index (NDVI) estimates made using data from the Advanced Very High Resolution Radiometer (AVHRR); (3) surface‐based Antecedent Precipitation Index (API) estimates made using precipitation records from surface reporting stations and; (4) estimates of sensible heating, evapotranspiration and an “effective” surface roughness made for the region using a combination of satellite‐measured surface “skin” temperatures from geostationary (GOES) satellites and radiosonde measurements of the time‐changes of the height of the planetary boundary layer (DH) from the synoptic network. The scatter diagram of the daytime range (time‐change) of surface skin temperature (DT_S) versus NDVI over the large region investigated shows a similar pattern to published results for absolute skin temperature measurements versus NDVI over smaller areas (generally the size of a single watershed). Over the entire region investigated, the correlation between DT_S and NDVI was poor. In the eastern section of the region, however, which was experiencing drought conditions by June 1988 (evidenced in very low API values), the correlation between DT_S and NDVI was much improved. In the western section of the investigation region, with generally higher values of API (suggesting higher soil moisture content), there was a good correlation of DT_S with API for areas with low NDVI. Values of a microwave‐based API derived from SSM/I data generally showed similar relationships to DT_S values as the surface‐based API, but with slightly higher correlation coefficients. Values of the surface Bowen ratio derived from DT_S and DH measurements were in qualitative agreement with the pattern of the developing drought conditions in the area, although quantitative comparisons with other the other components of the data base were less definitive. Values of an “effective” surface roughness derived from the same DT_S and DH measurements showed a good correlation to NDVI.