Abstract
This study reveals how geostationary satellite-derived minute-scale and mesoscale atmospheric motion vectors (MAMVs) offer a breakthrough in super typhoon forecasts. By leveraging high-frequency, high-resolution imagery, MAMVs provide unprecedented insights into wind patterns and atmospheric flows across multiple tropospheric layers, significantly enhancing the skill of numerical weather prediction models (NWP) in forecasting super typhoons. To evaluate the effectiveness of MAMVs in enhancing typhoon forecasts, this study assimilated MAMVs into NWP models and compared them with assimilating standard atmospheric motion vectors and control experiments without assimilation. Compared to assimilating standard AMVs, the assimilation of MAMVs significantly reduces typhoon track forecast errors by nearly 50% within 48 hours and improves forecasts up to 72 hours in advance. Case studies of recent super typhoons demonstrate the substantial enhancement in forecast accuracy, offering a promising approach to advance tropical cyclone prediction and enhance disaster preparedness, where improved forecasting can save lives and reduce damages.
 

Typhoon; Geostationary high-speed imager; Mesoscale atmospheric motion vector; Data assimilation.