Impact of Kalpana-1 retrieved multispectral AMVs on Mahasen tropical cyclone forecast

Abstract

The atmospheric motion vectors (AMVs) retrieved from geostationary satellites are recognized as one of the important inputs for numerical weather prediction models to improve the tropical cyclone (TC) forecast. In this study, the weather research and forecasting (WRF) model, WRF three-dimensional variational (3D-Var) data assimilation system and WRF tangent linear and adjoint model are used to investigate the impact of multispectral Kalpana-1 AMVs on the simulation of Mahasen tropical cyclone (now known as cyclonic storm Viyaru) over the Indian Ocean. Three different sets of experiments are performed to evaluate the impact of Kalpana-1 AMVs. First, the impacts of Kalpana-1 AMVs are evaluated for different forecast lengths. The assimilation of Kalpana-1 AMVs improves the cyclone track prediction compared to control experiment. However, all the experiments are unable to capture the deep re-curvature of the TC. The next set of experiments is performed to evaluate the impact of Kalpana-1 AMVs derived from different multispectral channels (viz. visible, infrared and water vapor channels). More improvement is observed in TC track forecast when AMVs from water vapor channel are used for assimilation compared to infrared channel. Results also show degradation in short-range forecast when less-strict quality control is used for AMVs assimilation, but a considerable improvement is observed in long-range forecasts. Finally, the WRF tangent linear and adjoint model is used to compute the forecast sensitivity to Kalpana-1 AMVs observations. Upper- and lower-level circulation information provided by the Kalpana-1 AMVs influences the TC steering flow, and a positive impact on the track prediction is observed.