福衛7與獵風者GNSS資料加值與應用-福衛7與獵風者GNSS資料加值與應用(1/3)

This project aims to optimize and improve the data quality, enhance and expandthe value and application of the data from the FormoSat 7 satellite constellationand the TRITON satellite in Taiwan. The core technology is radio occultationtechnology (GNSS-RO), and sea surface reflectometry (GNSS-R) for FormoSat 7and TRITON, respectively. The objectives of this project are (1) refinement andoptimization the quality of the Triton satellite remote sensing products, and (2)value-added applications of the Triton R and FormoSat-7 RO data.For the first objective, improvement or value-added strategies are proposedbased on the different stages of satellite data processing. To reduce theuncertainty of wind speed inversions and to improve the accuracy of high windspeed inversions, the strategies adopted are as follows: Level 0~1: Optimize thecorrection parameters of the existing DDMI; Level 1~2: Separate the original coremodules into two segments in series, centering on Mean Square Slope (MSS),establish the relationship between DDM and MSS through the new GMF at thefront end, and introduce the most-recent MSS vs U10 parameterization at theback end to reduce the inverse uncertainty by considering factors such asbreaking wave and associated whitecap coverage. To verify the reliability of thismethod, sea-truth MSS is essential as the observational evidence. This studyschedules to use the new MOST ocean research vessels to deploy clusters ofminiature data buoys around the warm pool of the Northwest Pacific Ocean in2021 summer. These miniature buoys would then form a spatial array for fourmonths of simultaneous measurements of the sea surface roughness duringtyphoons or during the high wind speed circulation around it. Moreover, we willalso use the GNSS-R software receiver stationed on the northeastern coast ofTaiwan or outlying islands to investigate the sea surface reflectance of GNSSsignals and sea surface roughness at microscopic scales, which will serve as thebasis for the corrections of the satellite borne GNSS-R data. We will also conductoffline verification and calibration based on inter-comparison with other satellite orweather observations, such as ASCAT and dropsonde.The second objective is carried out in two directions. Firstly, we will use the dataassimilation and the numerical weather prediction model to incorporate theGNSS-R retrieved surface wind data to improve the severe weather forecast. Wewill seek for different strategies to enhance the value of the GNSS R data bytransforming the wind speed data to wind vector or TC wind radii. In thisapplication, we will further integrate with the satellite observations of FormoSat7to investigate the importance of the air-sea moisture fluxes required for thedevelopment of tropical cyclone and rainfall systems. In addition, we will combinewith ground-based GNSS observations to develop a new generation of Formosat-7 data profile calculation system. Finally, we will combine deep learning withGNSS and other optical satellite data to estimate the soil moisture content andexpand the application of GNSS in the terrestrial hydrological cycle.