To assist in site identification for development of solar energy facilities in Australia, relevant spatial data were drawn from a range of sources. Along with infrastructure and development constraints, spatial solar irradiance estimates were a key requirement. Existing estimates, derived from satellite observations, were found to include significant bias when compared to ground station observations, likely due to the difference in measurement (spatial average for satellite and point location for ground station). Rather than revert to reprocessing the original data, post-processing corrections were developed to compensate for this bias. Investigations were also made into the influence of aerosols on the satellite estimates (aerosols were not included in the existing model), and whether any spatial bias existed in the satellite estimates. In both cases, no evidence was found to indicate that improvements in the data were possible by including these effects. (C) 2012 Elsevier Ltd. All rights reserved.