Retrieving Hurricane Maximum Winds Using Simulated CYGNSS Power-Versus-Delay Waveforms

Abstract

A novel approach in retrieving hurricane maximum winds using simulated NASA Cyclone Navigation Satellite System (CYGNSS) data is presented. Five hundred fifty two hurricane wind fields, from the 2010–2011 Atlantic and Eastern pacific hurricane seasons, were used to test the algorithm. These wind fields have been obtained from the hurricane weather research and forecasting model (HWRF). Power-versus-delay waveforms associated with specular points located along CYGNSS tracks crossing these wind fields were simulated. These “storm” power-versus-delay waveforms were compared to “reference” power-versus-delay waveforms generated over a set of synthetic Willoughby storms with known maximum wind speeds. The retrieved maximum wind speeds are compared against the hurricane research division reanalysis data (Best Track) and HWRF. For Best Track maximum wind speeds less than 40 m/s and greater than 40 m/s, the overall bias against Best Track is 11.3 and 2.1 m/s, respectively. When comparing against HWRF maximum wind speeds less than 40 m/s and greater than 40 m/s, the overall bias is 11.5 and 3.0 m/s, respectively. These results are improved when translation effects were applied to these synthetic storms: compared against Best Track for maximum wind speeds less than 40 m/s and greater than 40 m/s, the biases are 9.0 and $-$ 1.13 m/s, respectively. When compared against HWRF, the biases are 8.6 and 0.4 m/s, respectively.