Assessing the Quality of GNSS Radio Occultation Observations in Tropical Cyclones and Implications for Future Observations

Abstract Tropical cyclone intensity forecasts remain a significant challenge in forecast models. In situ observations of tropical cyclone (TC) thermodynamics are generally limited to dropsondes and aircraft data. Vertical profiling of TC thermodynamics from passive microwave and infrared sensors has been limited due to coarse vertical resolution and signal degradation. High vertical resolution Global Navigation Satellite Systems (GNSS) radio occultation (RO) soundings are insensitive to clouds and precipitation and provide a unique opportunity to study TC thermodynamic vertical structure. However, the quality of RO observations in the TC environment has not been thoroughly evaluated. In this study, GNSS-RO profiles from COSMIC-1 (2006–19) and COSMIC-2 (2019–21) are analyzed in conjunction with collocated dropsondes in the TC environment. COSMIC-1 and COSMIC-2 demonstrate remarkable consistency in RO refractivity and bending angle for use across RO missions (less than ±0.25% differences in the troposphere). Overall median refractivity difference between GNSS-RO profiles and collocated dropsondes is within ±0.5% at altitudes of 8 km or higher, confirming the high quality of GNSS-RO measurements of the TC upper troposphere with minimal bias. Below 8 km, refractivity differences generally range between −3% and −1% depending on altitude as well as subset criteria and sampling characteristics. Refractivity bias is mostly unaffected by collocation criteria, in terms of distance in space and time between observations, but accounting for RO ray path orientation significantly reduces biases. Results from this study indicate that additional RO collocation criteria are required in regions, or around weather events, that have large gradients in temperature and moisture to accurately represent their effects in an analysis. Thorough understanding of GNSS-RO quality in the TC environment will allow for the use of GNSS-RO in TCs to fill in observational gaps to improve our understanding of TC thermodynamics and their evolution over time and provide more consistent observations for TC forecasting. Significance Statement Tropical cyclones (TCs) are frequent, destructive events that occur globally. Global Navigation Satellite Systems (GNSS) radio occultation (RO) observations provide high vertical resolution atmospheric observations that can be very useful for studying TCs, due to their global spatial sampling and disregard for the presence of clouds and precipitation. This research uses GNSS-RO to examine its potential use in filling in observational gaps in tropical cyclones and tests for biases by comparing RO profiles with dropsondes across various subsets. GNSS-RO biases can be reduced by ensuring that the ROs used have tangent points that pass through the TC. Other potential sources of bias in RO observations are also examined.