Understanding CYGNSS wind performance against in-situ and satellite winds during low-to-moderate and cyclonic conditions

Abstract

An assessment of the performance of CYGNSS operational wind products is done against a multitude of reference datasets which include in-situ wind speed from Indian National Centre for Ocean Information Services (INCOIS) network of buoys (hereafter, INCOIS-buoys) in the Indian Ocean, SCATSAT-1 operational wind products and cyclone best track data from India Meteorological Department (IMD). The validation study is done for CYGNSS data version 2.1 for the period 17th March 2017 till 31st December 2017. The performance of the operational wind products is examined through a two-tier approach- first one covering low-to-moderate wind regimes (less than or equal to 15 m/s) where in-situ data from INCOIS-buoys and rain-free SCATSAT-1 products are used. While the former gives an idea of CYGNSS winds in the Indian Ocean basin, the latter can be treated as global reference dataset. In the second approach, high wind (greater than15 m/s) performance of CYGNSS winds are evaluated on a case study basis during some of the tropical cyclones (TC) of 2017. The first approach gives good accuracy of minimum variance estimator (MVE) wind speed products in terms of standard deviations of 1.85 m/s (INCOIS-buoys) and 2.07 m/s (SCATSAT-1). MVE is the best product amongst all the five different operational wind products viz. MVE, FDS-NBRCS, FDS-LES, YSLF-NBRCS, YSLF-LES of CYGNSS. Time series analyses of high wind performance during the TC- Mora and Ockhi show better sensitivity of young sea wind products especially YSLF-NBRCS compared to YSLF-LES. However, the fully developed wind product, FDS-NBRCS shows equally good sensitivity of wind speed.