Comparison of RUC condensate analyses and forecasts with satellite-derived cloud properties

The transportation industry, including aviation, has an important need for improved predictions of clouds, fog, ceiling/visibility and precipitation. Model-based predictions, which are a significant source for forecaster guidance, require accurate specification of cloud and hydrometeor fields in the model initial conditions. Recent improvements in satellite remote sensing techniques have led to the development of near real-time cloud products with sufficient accuracy to suggest the potential to improve NWP analyses and forecasts, if they can be adequately incorporated via data assimilation methods. Cloud parameters derived from geostationary satellite data could provide the greatest impact on high temporal resolution models because of their high temporal frequency (compared to polar orbiting satellites) and because they more accurately represent spatially continuous fields, versus the discrete point measurements obtained from surface observations. A pre-requisite step in designing an assimilation system for satellite cloud products is an evaluation of the model-based hydrometeor analyses and forecasts with respect to the coincident satellite–derived cloud products. The NOAA Rapid Update Cycle (RUC) cycles at full-resolution five microphysical species (cloud water, cloud ice, rain water, snow, and graupel) and has the capability for updating these fields from observations. Half-hourly 4-8 km resolution cloud properties are being derived from Geostationary Operational Environmental Satellite (GOES) radiances at NASA Langley Research Center (LARC) under the NASA Advanced Satellite Aviation-weather Products (ASAP) Initiative. The LARC cloud products include ice and liquid water path, cloud top height, cloud particle effective size, optical and geometric thickness among other parameters. In this paper, we present a statistical assessment of the RUC model output in terms of the analyzed and predicted condensate fields when compared to the satellite-derived cloud parameters. The comparisons are conducted for selected time periods in different seasons, and focus on the relative horizontal and vertical placements of the condensates, the magnitude of the differences, in addition to the frequency and location of total disagreement.