Toward Understanding the Contribution of Satellite-Derived Atmospheric Motion Vectors To Numerical Tropical Cyclone Analyses and Forecasts

Atmospheric Motion Vectors (AMVs) are derived using a sequence of three satellite images to track targets (including cirrus cloud edges, gradients in water vapor, and small cumulus clouds) every 30 minutes in operation, and these are processed hourly by CIMSS. The assimilation of AMVs has been shown to significantly improve numerical forecast skill. For example, improvements to tropical cyclone (TC) track forecasts have been made via the use of AMVs in better representing the large scale steering flow. When a TC is a threat to society, the Rapid-Scan mode is often activated to allow more vectors to be derived spatially and temporally. In this study, CIMSS hourly and Rapid-Scan AMVs are assimilated into the Weather Research and Forecasting (WRF) model using the Ensemble Kalman Filter (EnKF) with different strategies to investigate the contributions of AMVs to the track and intensity of TCs.

Several data-denial cycles are prepared, in which AMVs at specified height levels and cut-off distances from the TC center are withheld. Each cycle uses 84 ensemble members with a resolution of 27 km on the analysis grid (and 9 km in the forward forecast model) covering the lifetime of Typhoon Sinlaku (2008). To identify the relative contributions of the different layers of AMVs to the analyses and forecasts of the TC and its environment, three data-denial cycles are designed: eliminate AMVs between 150-350 mb; eliminate AMVs between 350-700 mb; and eliminate AMVs between 700-999 mb respectively. Two further data-denial cycles are prepared, with AMVs withheld within (outside) 10 degrees from the TC center, in order to investigate the contribution of AMVs on the TC structure. Initial results and insights from this series of data-denial studies will be presented.