An Antenna-Coupled Simulation Tool to Explore the Optimal Configuration for the NOAA’s Future GNSS RO Architecture Missions

Global Navigation Satellite System (GNSS) Radio Occultation (RO) data is considered an essential long-term observable for weather and climate applications. NOAA incorporates about 12,000 occultations per day from COSMIC-2, commercial small satellites, and partner missions into its global weather forecast system. NOAA is also exploring optimal configurations for future RO missions to meet the needs of climate studies and weather forecasting. To fully exploit potential mission opportunities, examining the optimal spatial and temporal coverage of the backbone RO observations with the minimum RO constellation needed and minimum latency for weather forecasts is crucial. Due to varying receiver antenna designs, e.g., single patch versus phase array antennas, the quality of data collected from different RO missions can differ, especially in the lower troposphere. The signal-to-noise ratio (SNR) from various RO missions typically indicates the strength of RO signals penetrating the lower troposphere. Key factors affecting the SNR magnitude and its spatial distribution include the signal power of the GNSS emitter, the intermediate frequency bandwidth of the receiver, the design of the RO antenna, and the antenna gain pattern in related to the viewing geometry. Incorporating information on these factors into RO occurrence modeling is essential for determining the optimal configurations for future missions. In this study, we detailed the development of an antenna-coupled simulation tool to explore the optimal configuration for NOAA future GNSS RO architecture missions. We first developed an efficient algorithm for predicting RO limb-sounding events between GNSS transmitters and LEO RO receivers. The algorithm used the Simplified orbital Perturbation Model (SGP4) with Two-Line Element (TLE) inputs to predict the positions and velocities of LEO and GNSS satellites. We can predict RO limb-sounding occurrences for their time, location (longitude and latitude), viewing geometry, and occurrence of rising or setting. In addition, a given antenna design's gain pattern and frequency response are simulated with an antenna modeling tool. The modeled 3D antenna gain patterns for particular GNSS operating frequencies are transformed into RO versus GNSS viewing coordinates, allowing the spatial distribution of SNR for RO missions from a specific GNSS emitter to be modeled. We validated our prediction results by comparing the predicted COSMIC-2 RO events with actual COSMIC-2 RO observations, assessing the accuracy of the predicted ground tracks, RO event counts, and RO event distribution relative to antenna view angle. The simulation explains the latitudinal distribution of COSMIC-2 SNR and the distribution differences between COSMIC-2 SNRs from GPS and GLONASS. The difference between COSMIC-2 and Spire latitudinal SNR distributions is investigated. Furthermore, we employ the prediction tool to project temporal and spatial RO occurrences and SNR distributions for conceptual RO missions, accommodating different receiver antenna designs (single patch antenna vs. beam steering phased array antenna) and incorporating transmitters like GPS, GLONASS, Galileo, and BeiDou. We will first examine the RO occurrence coverage for existing mission orbits, including NOAA JPSS, COSMIC-2, EUMETSET METOP-SG, NASA CYGNSS and TROPICS constellations, and the international space station (ISS) with hypothetical GNSS receivers. We further simulate the hypothetical RO receiver constellations with different orbital inclination angles to explore the optimal spatial, temporal, and SNR coverage configurations. Our simulation tool is invaluable for optimizing RO coverage tailored to weather forecasting needs, considering both global and local time and SNR coverage. It can integrate planned NOAA missions with additional RO constellations for a comprehensive analysis.