Hurricanes in the Regional-refined GFDL HiRAM: Inner-core Structure Simulation and Subseasonal Prediction

While significant progress over the past decades has been made in the representation of hurricane climatology and variabilities in global climate models, achieving realistic hurricane structure simulation in these models is still considered challenging. The recently developed two-way global-to-regional nesting approach in NOAA/GFDL’s HiRAM (High-Resolution Atmospheric Model) allows for increasing resolution over a limited area in the global model domain, and thus provides an efficient way to reach hurricane-resolving resolution over the region of interest. In this talk, we will first present HiRAM simulated hurricane structure from a suite of sub-seasonal (30-day duration) retrospective predictions. The results from the two-way-nested configuration (~8 km resolution) are evaluated against the global-uniform-resolution (~25 km resolution) configuration, as well as various observational datasets. We demonstrate that the two-way nesting method yields much improved hurricane inner-core size and structure simulation, which leads to improved representations of hurricane intensification rates and their lifetime maximum intensity. The prediction of the basinwide and regional hurricane activities on the subseasonal scale is then assessed. We find that both grid configurations show very promising performance in the hurricane genesis prediction. The two-way-nested configuration shows better performance in the prediction of major hurricane activities (including their formation and accumulated cyclone energy) because of the improved storm structure and intensity representation.