Filling in Spatio-temporal Gaps in the GOES-R Land Surface Temperature Product

Abstract:

Land surface temperature (LST) is an important parameter used in a number of applications in hydrology, meteorology and climatology. Forecasters also use it to forecast the occurrence of fog and frost. LST is routinely retrieved from the GOES-R Advanced Baseline Imager (ABI) long wave spectral channels employing a split-window technique. Since the product is available only under clear sky conditions, large gaps can exist in the data stream which correspond to contamination by clouds. However, continuous estimates of LST data is vitally needed for several applications such as drought monitoring (i.e., Standardized Precipitation-Evapotranspiration Index or SPEI), vegetation growth and crop yield estimation to list a few. Studies have shown that LST tracks with corresponding changes in incident solar radiation or more specifically changes in surface absorbed solar radiation with good correlation irrespective of sky conditions (clear or cloudy). A recent validated study of near real time surface solar absorption parameter from GOES-R demonstrated the potential of filling in large LST data gaps. The performance of the scheme is validated against in situ observations over the NOAA surface radiation network (SURFRAD) and United States Climate Reference Network (USCRN) sites.