Local-scale validation of the Surface Observation Gridding System with in situ weather observations in a semi-arid environment

Abstract

Although the Surface Observation Gridding System (SOGS) provides spatially continuous models of meteorological conditions, little work has been done to validate SOGS data independently for site-specific research and, as a result, a single nearby weather station is commonly selected instead. This study sought to determine local-scale accuracy of SOGS data (1) by correlation with independent, in situ weather station measurements and (2) relative to a nearby weather station. Correlations between SOGS data and in situ weather observations and between in situ weather observations and a nearby weather station were examined in a semi-arid environment of southeastern Idaho over the 2006 growing season. The results indicate that both SOGS and nearby weather station data were significantly correlated with in situ weather station measurements. Although temperature correlations between in situ and the nearby weather station were slightly greater compared to SOGS, SOGS data were a better predictor of precipitation. This suggests that the use of a nearby weather station is appropriate for local temperature parameters but precipitation parameters are better estimated using SOGS data. Overall, the validation of the SOGS weather models agreed closely with independent, in situ weather measurements and, as a result, greater confidence can be placed in the accuracy of the productivity, biomass and global climate change models derived from these data.

Acknowledgements

This study was made possible by a grant from the NASA Goddard Space Flight Center. Idaho State University thanks the Idaho Delegation for their assistance in obtaining this grant. The authors gratefully acknowledge assistance with the SOGS datasets provided by Maosheng Zhao, The University of Montana.