Xu Lu and Dr. Xuguang Wang- March 1- Convective Meteorology (Mesoscale Dynamics) Seminar

Convective Meteorology (Mesoscale Dynamics) Seminar

Impact of Vortex Relocation Strategies on Hurricane Inner-core Data Assimilation and Prediction in the HWRF EnVar DA System

Xu Lu and Dr. Xuguang Wang

Friday, March 1^st

3:00pm/NWC 5600

In a previous study, Chen and Snyder (2007) showed that a large location error in the background forecast can result in a poor performance of ensemble Kalman filter (EnKF) data assimilation (DA) due to the violation of the Gaussian assumptions.  One way to alleviate this issue is to apply vortex relocation (VR) before the ensemble-based DA.

During the real time forecast, TCVital is used to provide position guidance for hurricanes.  One strategy is to relocate vortices from all ensemble members to the TCVital location.  However, TCVital location can also have non-negligible errors.  Another strategy is to produce an ensemble of analyzed location by updating the first guess location through assimilating the TCVital location.   Additionally, the effect of relocation to reduce the violation of Gaussian assumption can be dependent on different DA methods such as 3DEnVar v.s. 4DEnVar.

Experiments are then designed and conducted in this study to investigate the impact of these different VR strategies on the inner-core DA during Hurricane Irma (2017).  Results suggested that the analyzed positions after assimilating TCVital location may still present large position errors, which, as a result of violating Gaussian assumption, produces anomalous 3DEnVar analysis increment.  The more accurate location derived from the flight-level wind observations (Willoughby and Chelmow, 1982) significantly improves the 3DEnVar analysis.  Such benefit manifests itself more in 3DEnVar than 4DEnVar.  Further diagnostics to understand how different relocation strategy benefits the analysis will be presented in the seminar