The impact of assimilating airborne Doppler radar radial velocity and dropsonde data on the analysis and prediction of Hurricane Earl (2010) using the ARPS EnKF system and WRF prediction model

Airborne radar radial velocity (Vr) data are assimilated for rapidly intensifying Hurricane Earl (2010) using the ensemble Kalman filter (EnKF) system originally developed for the Advanced Regional Prediction System (ARPS) at cloud-permitting 4 km grid spacing while WRF ARW model is used for the forecasting. The initial ensemble is created by adding perturbations from the global ensemble forecast system to the deterministic GFS analysis. Multiple microphysics (MP) and planetary boundary layer (PBL) parameterization schemes are employed in the ensemble to partly account for model uncertainty. The airborne Vr data are assimilated in a two-hour assimilation window before Earl intensifies rapidly. Dropsonde observations are also assimilated together with the airborne Vr. The impact of the data assimilation on Earl's track, structure and intensity is examined in terms of the analyses and 72 hour deterministic and ensemble forecasts.