Assimilating Global Synthetic Weather Radar in the Weather Research and Forecasting Model over Kwajalein Atoll
Report Number:
ARL-TR-9298
October 28, 2021
Approved for public release: distribution is unlimited.
Author(s):
Brian P Reen, Huaqing Cai, John W Raby, and James J Ramsey
Abstract:Global Synthetic Weather Radar (GSWR) data were assimilated into the Weather Research and Forecasting model to determine which assimilation methodology performs best and the difference in performance between a 1-km and a 3-km horizontal grid-spacing configuration. An evaluation of six GSWR assimilation methodologies on a 1-km domain over three case days was used to determine two methodologies that were applied to four additional case days; the two methodologies were also applied on a 3-km domain for all seven case days. The seven case days were used to choose a GSWR assimilation methodology for future use and to determine that for the verification metrics used, the 1-km and 3-km domains perform similarly. Based on this result and that the 1-km domain was approximately nine times slower to run, the 3-km domain will be used in future work to allow more frequently updated forecasts and quicker forecast availability. Use of GSWR is shown to improve the 0-h forecast column maximum reflectivity when convection is sufficiently large to be resolved by GSWR. Higher-resolution observation sources, such as satellite-derived data, should be pursued in the future to improve forecasts of small-scale convection.
