NOAA Integrated Cal/Val System for Radio Occultation Performance Monitoring and Data Quality Assurance

The quality of GNSS Radio Occultation (RO) is critical for weather forecast and climate change detection. RO data come from many different sources, including different countries, agencies, satellites and receivers, and different time periods. In addition, Commercial Weather Data Pilot (CWDP) and newly launched COSMIC II bring radio occultation technology into a new era. Therefore, there is a need to ensure that the RO data quality and consistency to meet user needs for their applications. To achieve this goal, NOAA has established an on-line integrated calibration/validation system (ICVS) for radio occultation to monitor RO instrument performance and validate various RO products against multiple references. Currently, RO data from 12 publicly available missions are included, from GPSMET collected in 1995 to CWDP data in 2019.
This paper will present the major components of the ICVS for RO which consists of three major sections, 1) RO data vs. model comparison, 2) inter-comparison between RO measurements, and 3) RO data comparison with radiosonde. In model comparison, GNSS-RO product are compared against a subset of weather model including GFS 6hour, GDAS, NCEP reanalysis R2, ECWMF reanalysis interim, MERRA2 and JRA-55. In RO inter-comparison, all RO data are compared against COSMIC/KOMPSAT5 processed by UCAR, which are considered the reference standard. In the radiosonde section, Vaisala RS80/RS92 are used as reference for comparison. We will demonstrate the functionality of the system using specific examples from KOMPSAT5, COSMIC, COSMIC2, and other RO missions to show analysis of global distribution, daily/monthly statistics, time series, anomalies, individual profiles, data quality in excess phase, bending angle/refractivity, and level 2 products. In preparation for COSMIC2, we are developing capabilities to use the ICVS for comparing COSMIC2 profiles with those from COSMIC in terms of global mean bias using both radiosonde and models. Monthly mean climatology will also be analyzed. Preliminary results and findings will be presented.