Using METRIC to Estimate Surface Energy Fluxes over an Alfalfa Field in Eastern Colorado

The ability to estimate surface energy fluxes using remote sensing methods has enabled the determination of evapotranspiration over a large area with less cost. Several models that employ the energy balance method have been developed. The model discussed in this paper is the Mapping Evapotranspiration at high Resolution with Internalized Calibration (METRIC). It estimates net radiation (Rn), soil heat flux (G), sensible heat flux (H), and then determines latent heat flux (LE or ET) as a residual. Landsat 5 TM and Landsat 7 ETM images for the 2010 alfalfa growing season near Rocky Ford, CO, were processed and analyzed using Erdas Imagine 2010 Software. The accurate estimation of Rn and G is important as these two determine how much energy is available to be partitioned into H and LE. Hence, the focus on evaluating Rn, G and ET in this study. The results obtained from the processed images were compared with actual fluxes measured by instruments installed in the alfalfa field and performance indicators for each flux were determined. For Rn, the Mean Bias Error (MBE) was 17.8 W m (3.3%), and the Root Mean Square Error (RMSE) was 22.1 W m (4.1%), the Nash-Sutcliffe coefficient of Efficiency (NSE) value of 0.53 and R of 0.84. The G MBE was -3.0 W m (-5.8%), RMSE of 14.2 Wm (27.6%), NSE value of 0.79 and R of 0.96. Hourly ET resulted with an MBE of -0.08 mm h (-10.3%) and an RMSE of 0.14 mm h (17.6%), NSE of 0.71 and an R of 0.83. It was observed that the estimation of G had larger errors under smaller biomass surface conditions (e.g., when the alfalfa had just been harvested). However, overall the remote sensing model estimated well the heat fluxes and it seems to be suitable for applications at regional scales in hydrology studies.