Forward Modeling of Bending Angles With a Two-Dimensional Operator for GNSS Airborne Radio Occultations in Atmospheric Rivers

April 16, 2025

A paper titled “Forward Modeling of Bending Angles With a Two-Dimensional Operator for GNSS Airborne Radio Occultations in Atmospheric Rivers” was recently published in the AGU’s Journal of Advances in Modeling Earth Systems. The study was led by Pawel Hordyniec (Institute of Geodesy and Geoinformatics, Wrocław University of Environmental and Life Sciences) and SIO Professor/CW3E research collaborator Jennifer Haase, with contributions from Michael Murphy (SIO, NASA Goddard, and UMBC), Bing Cao (SIO), Anna Wilson (CW3E), and Ivette Baños (NCAR).

Airborne Radio Occultation (ARO) has been a key component of Atmospheric River Reconnaissance (AR Recon) data collection since receivers were first added to the NOAA G-IV aircraft in 2018. ARO data complement the vertical profile data collected by the foundational AR Recon dropsonde data. ARO indirectly makes observations of the moisture and temperature in the atmosphere by measuring delays in GNSS (GPS) signal propagation. Because it is an indirect measurement of the atmospheric properties, it is necessary to have a means for simulating the observations in order to assimilate them in to a numerical weather prediction (NWP) model and improve forecasts.

This paper describes the observation operator that has been developed to assimilate the ARO observations into any weather model that has been interfaced to the new Joint Effort for Data assimilation Integration (JEDI) system. Because of the modular design of JEDI, the data can now be assimilated into the Model Prediction Across Scales developed by the National Center for Atmospheric Research and the Finite Volume 3 (FV3) model used at NASA and NCEP, among others. The observations measure the time of propagation of a near-horizontal ray path from a setting GNSS satellite to the aircraft and samples the larger region surrounding the flight path, and thus complements the reconnaissance dropsondes directly beneath the aircraft. The operator is developed to take into account the highly complicated two-dimensional structure of the AR traversed by the long horizontal ray path, so the data assimilation can reproduce high resolution features in combination with other datasets. An upcoming paper will describe the improved forecast of an AR in the pacific northwest using the operator with the MPAS model. The figure below illustrates the large errors that would be encountered at 5-6 km altitude in the core of the AR if the two-dimensional nature of the AR was not considered in simulating the observations.

Figure 1. (a) Each profile shows the difference between simulated 1D and 2D bending angles. When the atmosphere has large variations along the GPS signal raypath, the points in the profile will deviate greatly from a straight line, as they do specifically for the red points in the profile 025.21.52.R02 and 025.22.29.G19 that were measured along transect A2 near the core of the AR. Profiles are plotted sequentially along the horizontal axis where each grid box represents 10% bending angle difference. The labels A1, A2, A3, and A4 indicate in which part of the flight path the profiles are located in panel (c). Individual tangent points in the profile are color-coded by the integrated vapor transport beneath that point, and the size of each dot is scaled to corresponding IWV values as indicated in the legend. (b) Refractivity anomaly profiles (observation minus climatology) calculated from the dropsondes in transects A2 and A3. Dashed line indicates the location where the aircraft turned on the cold side of the AR core from transect A2 to A3. Blue shading indicates schematically the levels in the profiles with large dropsonde dew point depression (dry air), and red shading indicates levels with near saturation conditions in the dropsonde data. Profiles are offset by 10% for visibility. (c) Location of occultation profiles along transects A1 (outside the AR), A2 and A3 (inside the AR) and A4 (outside the AR). Figure 14 from Hordyniec et al. (2025).

Hordyniec, P., Haase, J. S., Murphy, M. J., Jr., Cao, B., Wilson, A. M., & Banos, I. H. (2025). Forward modeling of bending angles with a two-dimensional operator for GNSS airborne radio occultations in atmospheric rivers. Journal of Advances in Modeling Earth Systems, 17, e2024MS004324. https://doi.org/10.1029/2024MS004324