CALCULATIONS OF SURFACE RADIATION IN ARID REGIONS - A CASE-STUDY

The difficulties encountered in the derivation of surface radiation budget in arid regions are studied using the surface and satellite data measured during the preliminary field experiment for the Land-Atmosphere Interactions Experiment conducted at the Heihe River basin in western China. The surface radiation is derived by coupling theoretical radiative transfer calculations with satellite cloud retrievals. Comparisons with the surface measurements of solar and thermal IR fluxes show that a large error in the computed surface fluxes occurs in some cases. The error is attributable to the lack of aerosol data, the uncertainty in cloud retrievals, and the time difference between the surface and satellite measurements. For cloud-free cases, the modeled downward solar fluxes are systematically larger than the measured fluxes. The major cause of the error appears to be the failure to include aerosols in the calculations. The error is particularly large in the afternoon hours when the ground temperature is very high (> 50-degrees-C) and the atmosphere dust content is large due to an unstable boundary layer. We find that the error can be reduced and that a good agreement between the computed and measured surface solar fluxes can be obtained by using an aerosol single-scattering albedo of 0.5 and an optical thickness of almost-equal-to 0.2 in the afternoon hours. Nevertheless, the reason for the strong absorption of solar radiation in the atmosphere remains unclear. For all the cases studied when both surface and satellite data are available, the mean errors are 4.3 and -4.7 W M-2 for the net downward surface solar flux and the downward surface IR flux, respectively. The rms errors are 17.4 and 22.1 W m-2 for the respective surface fluxes. The relatively large errors found in the cases with small cloud amounts can be explained by the fact that aerosols are often misinterpreted as clouds in the cloud retrievals. Results of this study reemphasize the importance of aerosols in surface radiation calculations. Because the diurnal variation of ground temperature is very large in the arid regions, reliable calculations of surface IR radiation require high temporal resolution for temperature measurements. Aerosol and ground-temperature retrievals from satellite data should be the highest priority in the computations of surface radiation budget over arid regions.