Advancing Microphysics Parameterizations In The Hurricane Weather Research and Forecasting (HWRF) System

Abstract

One of the main goals of the Developmental Testbed Center (DTC), a collaborative organization responsible for bridging the gap between the research and operational numerical weather prediction communities, is the advancement of modeling for hurricanes. A primary focus of the DTC is the Hurricane Weather Research and Forecasting (HWRF) model, which has several components for the regional analysis and prediction of tropical cyclones; the HWRF is a coupled atmosphere-ocean system with hybrid-ensemble-variational data assimilation, vortex initialization, and postprocessing components. It uses 27/9/3 km horizontal grid spacing, making it capable of taking advantage of advanced physics schemes to best represent the large-scale flow, as well as the inner core of hurricanes. The current operational configuration of the HWRF employs the Ferrier microphysics scheme, a single- moment parameterization, and the Geophysical Fluid Dynamics Laboratory (GFDL) radiation scheme as a part of its advanced physics suite. The Ferrier scheme predicts the mixing ratio of cloud, rain, and snow, and advects total condensate – the combination of all cloud and rain water, cloud ice, and precipitation ice (in the forms of snow, graupel, and sleet). With the latest HWRF release (v. 3.5a) came the capability to choose the Thompson et al. (2008) microphysics scheme, a partial double moment scheme for the prediction of both mixing ratio and number concentration of multiple water and ice species. The HWRF v. 3.5a release also contains a version of the Rapid Radiative Transfer Model (RRTMG) parameterization that takes advantage of the effective radius of cloud water, cloud ice, and snow calculated directly in the Thompson microphysics scheme to compute cloud optical depth. This is in contrast to the use of look-up tables for effective radius in non-coupled versions of the RRTMG. Case studies of Hurricanes Sandy and Tropical Storm Debby in the Atlantic using HWRF with the Thompson microphysics scheme indicated that a better representation of the physics at the HWRF grid scales led to improved track and intensity forecasts. The DTC and the Environmental Modeling Center (EMC) have conducted extensive HWRF tests with Thompson/RRTMG physics options. These tests indicate that the HWRF configured with the Thompson/RRTMG physics suite has reduced track and intensity error for the Atlantic storms, but surprisingly the results are not as promising in the Eastern North Pacific basin. A detailed analysis of the Thompson/RRTMG test and its effect on hurricane forecasts in each of the basins will be the focus of this presentation.