During the second Atmospheric Radiation Measurement (ARM) Enhanced Shortwave Experiment (ARESE II), measurements were obtained during low-altitude passes of the Twin Otter aircraft directly over the central facility of the ARM Southern Great Plains (SGP) site. One set of measurements taken from this platform was the spectral irradiance pointing in the nadir and in the zenith with moderate spectral resolution from about 350 to 1670 nm using the NASA Ames solar spectral flux radiometer (SSFR). Routine measurements are made at the central facility (CF) of the ARM SGP site using two multifilter radiometer (MFR) heads with six narrowband filters covering portions of the spectrum between 415 and 940 nm. One measures upwelling spectral irradiance above an ungrazed pasture, and the other measures that above a wheat field. In addition, on one of the cloudy days during ARESE II (3 March 2000), measurements were made with a hand-held, commercial spectrometer above wheat, pasture, and dry corn stalks in and around the CF site between the wavelengths of 350 and 2340 nm. A needed input to radiation modeling at the top of the atmosphere and at the surface is the effective spectral surface albedo. In the calculation of downwelling radiation at the surface it is important to correctly specify surface albedo in overcast and clear conditions. This paper will examine the level of agreement among different spectral albedo measurements. The effect of the differences on calculated downwelling surface irradiance will be analyzed for thin and heavy overcast. Finally, the importance of spectral albedo versus a single-valued broadband albedo on modeled, clear-sky diffuse irradiance is demonstrated.