Intensity forecasts for hurricane charley: the role of data assimilation

The intensification of hurricane Charley (2004) prior to landfall in southwest Florida was both rapid and unanticipated. While certainly not unprecedented, Charley serves to highlight the current difficulty in hurricane intensity forecasting. While steadily increasing computational resources permit higher-resolution models, the recently completed High-Resolution Hurricane test (HRH), a part of the Hurricane Intensity Improvement Project (HFIP), has demonstrated that higher model resolution doesn't necessarily lead to better intensity forecasts. Here, we investigate the potential of high-resolution data assimilation to address this problem. We employ an ensemble Kalman filter (EnKF) applied to the University of Wisconsin Nonhydrostatic Modeling System (UW-NMS) to assimilate both conventional (RAOB, METAR, etc.) and satellite (GOES, SSM/I, etc.) data to produce an ensemble of analyses of hurricane Charley and show that the improved representation of the hurricane structure resulting from satellite data assimilation leads to an increase in the predictability of intensity.