A data collection experiment was held on 10 May 2006 in which various types of ground truth were collected. During analysis of the data, it became apparent that a good deal of temporal and spatial variability existed both between the two collection sites and between each of the flight lines. The variability was primarily manifested in the total sky radiance spectra collected during each flight line. These effects were believed to be due to a rapidly changing sky environment populated by sparse clouds that grew thicker as the collection continued. After discussing the theoretical considerations of remotely sensing in the presence of clouds and other objects, the results of our analysis are presented. It was observed that temporal variability resulted in approximately 2-15% error in sky radiance across the visible wavelengths, and that spatial variability in sky radiance could be shown to be responsible for up to 5% error in retrieved reflectance. Additional theoretical work in modeling the expected radiance contributions of a cloudy sky and the impacts of cloud-induced variability is on-going and will be supported by these results.