An Assessment of CYGNSS Normalized Bistatic Radar Cross Section Calibration

A cyclone global navigation satellite system (CYGNSS) \sigma _o calibration analysis is presented using version 2.0 of the Level 1 dataset available on PO.DAAC. Three separate analyzes are conducted, namely, an examination of the specular bin location (in delay) and \sigma _o relationship, an investigation of the impact of recently improved characterizations of the GPS effective isotropically radiated power on CYGNSS \sigma _o, and an intersatellite \sigma _o calibration analysis. We first noted a correlation between the specular delay bin location and \sigma _o, where an increase in the specular delay bin resulted in an increase in \sigma _o regardless of the wind speed level; a specular delay bin location ranging from 4.75 to 5.00 and from 7.00 to 7.25 resulted in a 14.74 and 17.72 dB median \sigma _o, respectively. Noticeable improvements in the median \sigma _o were present in the version 2.0 dataset, when separating the data by GPS block type: for blocks IIR, IIF, and IIR-M, median \sigma _o were 15.39, 15.42, and 15.10 dB, respectively (compared to 19.38, 20.53, and 21.38 dB in version 1.1). Finally, an unexpected correlation between the instrument noise floor and \sigma _o was observed for all eight observatories while conducting the intersatellite \sigma _o calibration analysis. Approximately 0-1 dB absolute \sigma _o difference biases (with up to \sim1 dB standard deviation) between spacecrafts were observed. A report of this analysis was presented to CYGNSS scientists and engineers, who eventually found an issue with the D/A DDM scaling algorithm. We expect better statistical performances in future releases of the data.