Data collection was conducted during the period of June 10 through June 18, 1992. The observations 
followed a period of very heavy rains over several weeks that ended on June 9. As a result, the initial 
conditions were typically saturated soils with standing water quite common. No rainfall occurred during the 
experimental period thus allowing the observation of drying conditions. 
4.2. Brightness Temperature Mapping 
ESTAR data were geolocated using ground control points identified from the aircraft video records. These 
tapes were reviewed using 1:24,000 scale maps and image data (SPOT and NS001) to determine time coverage 
of numerous points on each line. These points were then digitized (UTM coordinates) and recorded with the 
times. Due to the numerous road intersections in the region, the georeferencing accomplished using this method 
is considered to be quite accurate. Also, with the visual guidance the road network provided, the pilots were 
able to reproduce the same flightlines each day, thus yielding complete and contiguous coverage on each date. 
These data were then corrected for angular variations and resampled to a uniform grid. Figure 2 
includes gray scale brightness temperature images for each day. The pattern of variation exhibits a spatial 
structure that appears to be correlated to the soil texture (Jackson and Schiebe, 1993). The temporal change in 
brightness temperature between these dates retains the textural distinction. 
4.3. Large Scale Prediction of Soil Moisture 
As a first step in the Washita'92 ESTAR calibration and verification of the passive microwave - soil moisture 
algorithms, we examined the data on a watershed average basis. All of the soil moisture samples collected on 
a given day (nominally 350 points) were averaged for the study area. This same procedure was used for the 
200 m resolution brightness temperature data (nominally 20,000 cells) which were then converted to an 
emissivity estimate by normalizing with the averaged soil temperature data. This results in one pair of 
emissivity and soil moisture of the eight days. These values are plotted in Figure 3.for each of the eight days. 
A passive microwave simulation model (Jackson, 1993) was used with estimates of roughness and 
vegetation parameters to predict a soil moisture-emissivity relationship for the two dominant soil types in the 
watershed. The soil dielectric model used here is that described in Wang and Schmugge (1980). Comparing 
these relationships to the observations, we see a very close correspondence to the silt loam function. These 
preliminary results indicate that the data interpretation algorithms apply within this region. It is also interesting 
that even when the data is averaged over the entire area (740 km 2 ) the algorithms appear to apply. 
For comparison purposes, the data collected in the Arizona experiment have been included in Figure 
3. All of these data were averaged in the same manner as that described above. Both the 1990 PBMR and 
1991 ESTAR data are included. These results appear to follow a different functional relationship associated 
with a sandy soil and sparser vegetation cover, which were prevalent. 
5 - SUMMARY 
The ESTAR L band radiometer was evaluated for soil moisture mapping applications over the semiarid 
rangeland Walnut Gulch watershed located in southeastern Arizona and a larger watershed in a subhumid area 
of Oklahoma. Microwave brightness temperature data were used in conjunction with a microwave emission 
model to predict soil moisture and compared to ground observations of soil moisture. A second verification was 
conducted using an extensive data set collected with the PBMR radiometer. Both tests showed that the ESTAR 
is capable of providing soil moisture with the same level of accuracy as existing systems. Preliminary results 
of a large scale passive microwave study to map surface soil moisture were presented. Comparisons of model 
predictions and watershed averaged observations of the emissivity - soil moisture relationship showed close 
agreement indicating the reliability of the model in data interpretation and the validity of extending it to very 
large areas. These results show that the ESTAR is capable of providing the necessary data for soil moisture 
applications. 
6 - REFERENCES 
Allen, P. B. and Naney, J .W., 1991. Hydrology of the Little Washita River Watershed, Oklahoma: data and 
analyses. USDA, ARS-90, 74pp.