ESTIMATING EVAPORATION COEFFICIENT DURING TWO-STAGE EVAPORATION FROM SOIL SURFACES.

Evaporation coefficient (KE) is defined as the ratio of actual evaporation over potential evaporation under the same atmospheric conditions. The use of KE provides an alternative means for evaluating soil moisture status and quantifying soil evaporation by reducing in situ measurements. However, little effort has been made to evaluate the estimation of KE over soil surfaces. In this study, a resistance-based method (KrE) and a temperature-based method (KtE) for estimating KE were compared with high-frequency measurements of two-stage soil evaporation. An alternative method (KaE) for estimating KE was presented in which soil surface resistance (rs) was computed with temperature differences between remotely measured surface and overlying air (Tr - Ta). Results indicated that more accurate estimates of KE were found in the first-stage evaporation from soil surfaces than in the second stage. Compared with the measurements, the KrE method produced the best estimates with a correlation coefficient (r2) of 0.88, a root-meansquare-difference (RMSD) of 0.04, and a mean-absolute-difference (MAD) of 0.03. The KtE method overestimated KE in both stages of evaporation with the lowest r2 (0.30) and largest RMSD (0.14) and MAD (0.13). The KaE method produced estimates of KE comparable to the KrE method, which was a significant improvement from the KtE method. The correlation coefficient between KaE and measurements was 0.64, and the RMSD and MAD were 0.05 and 0.04, respectively. By incorporating remotely sensed temperature data into physically based algorithms, the KaE method probably would be more practical than the KrE or KtE method.