The Impacts of Assimilating Enhanced Atmospheric Motion Vectors (AMVs) on HWRF Analyses and Forecasts of Hurricanes

Atmospheric wind data is essential for hurricane prediction. The enhanced atmospheric motion vectors (AMVs), derived from multiple satellites, are produced by CIMSS, University of Wisconsin. In this study we evaluate the impact of these enhanced AMVs on numerical analyses and prediction of hurricanes with the NCEP operational hurricane weather research and forecasting (HWRF) model. A hybrid of 3-dimensional variational (3DVAR) and ensemble Kalman filter (EnKF) data assimilation system, known as the NCEP hybrid gridpoint statistical interpolation (GSI) system is used. Two types of AMV data products, with different quality control thresholds, have been assimilated into the HWRF model for Hurricane Edouard (2014) during its rapid intensification and mature stages. Various sensitivity studies have also been conducted. Early results show positive impacts with the use of AMVs on forecasts of Edouard. However, the degree of the impacts depends on data quality, horizontal and vertical coverage and the configurations of the data assimilation system. Specifically, the background error terms in the GSI hybrid system have an influence on the assimilation results.

Additional experiments are being performed for hurricane cases in recent seasons. The dropsonde observations from recent field programs (e.g., TCI, HS3) are used in forecast evaluation and are also being assimilated into the HWRF model for further comparison.