Impact of Convection-Permitting Resolution and EnKF Data Assimilation on Atlantic Tropical Cyclone Forecasts

During the fall 2010 and 2011 hurricane seasons, the Center for Analysis and Prediction of Storms (CAPS) at the University of Oklahoma produced twice daily (00 and 12 UTC) 48-hour experimental hurricane forecasts in real time for the Atlantic Ocean basin on a single large 4-km resolution grid using the Advanced Research WRF (WRF-ARW). Two sets of forecasts were produced, initialized from the operational NCEP Global Forecast System (GFS) analyses based on the GSI (Gridpoint Statistical Interpolation 3DVAR system) and from the experimental global ensemble Kalman filter (EnKF) ensemble mean analyses produced by the NOAA Earth System Research Laboratory.

Verification is performed on these forecasts against best track data from the National Hurricane Center. The ARW forecasts are compared to corresponding deterministic forecasts from the operational GSI-initialized and experimental EnKF-initialized GFS. These forecasts allow us to examine the impact of the high cloud-resolving resolution and the EnKF initial conditions on hurricane track and intensity forecasts.

Statistically significant improvement is found with the high-resolution WRF forecasts compared with the GFS forecasts, in terms of 10 m wind speed and mean sea level pressure. In addition, track forecasts are also improved, especially for tropical cyclones below hurricane strength. Overall, the 4-km WRF with EnKF initial conditions produces the best forecasts in both intensity and track.