Abstract's details
Toward Assimilating SWOT Altimetry Data into the Current and Next-Generation GMAO Subseasonal-to-Seasonal Forecast System
Event: 2025 SWOT Science Team Meeting
Session: Oceanography: Inversion/Assimilation
Presentation type: Poster
At NASA’s Global Modeling and Assimilation Office (GMAO), efforts are underway to assimilate wide-swath ocean altimetry data from the new Surface Water and Ocean Topography (SWOT) satellite’s Ka-band Radar Interferometer (KaRIn) instrument into our suite of coupled earth system models. To pave the way for such new observation types into coupled data assimilation (DA), the next version of the Goddard Earth Observing System Subseasonal-to-Seasonal (GEOS-S2S) prediction system will transition from legacy DA systems for the individual Earth system components to the new Joint Effort for Data assimilation Integration (JEDI) system, which provides a unified framework for multiple DA methods across various Earth system components.
As a first step, assimilation of conventional altimetry data from SWOT’s Jason-class nadir altimeter is being tested within both GEOS-S2S version 3 (S2S-3) and JEDI’s marine interface Sea-ice, Ocean and Coupled Analysis (SOCA). For ocean DA, S2S-3 uses a modified version of the local ensemble transform Kalman filter (LETKF), whereas SOCA testing at GMAO has thus far been limited to the 3DVar and 3D-FGAT methods. In this work, the impact of SWOT nadir data on analyses and forecasts in the two DA systems is explored using Observing System Evaluation (OSE) experiments. Each OSE is comprised of two component experiments: One in which the SWOT data are assimilated (i.e., the SWOT experiment), and one in which the SWOT data are withheld (i.e., the CONTROL). For both experiments, all other conventional data are assimilated (e.g., Argo, other nadir altimeters, satellite SSS, etc.), as would be the case in operational prediction scenarios. The impact of assimilating SWOT data is assessed by comparing the SWOT and CONTROL experiments for both the S2S-3 and SOCA DA systems, as well as long-range (e.g., ENSO) forecasts initialized from the two experiments.
Beyond SWOT nadir assimilation, for future work, we intend to show progress towards assimilation of wide-swath KaRIn data, which will likely require new methods which account for two-dimensional spatial correlations in the observation error covariance.
Back to the list of abstractAs a first step, assimilation of conventional altimetry data from SWOT’s Jason-class nadir altimeter is being tested within both GEOS-S2S version 3 (S2S-3) and JEDI’s marine interface Sea-ice, Ocean and Coupled Analysis (SOCA). For ocean DA, S2S-3 uses a modified version of the local ensemble transform Kalman filter (LETKF), whereas SOCA testing at GMAO has thus far been limited to the 3DVar and 3D-FGAT methods. In this work, the impact of SWOT nadir data on analyses and forecasts in the two DA systems is explored using Observing System Evaluation (OSE) experiments. Each OSE is comprised of two component experiments: One in which the SWOT data are assimilated (i.e., the SWOT experiment), and one in which the SWOT data are withheld (i.e., the CONTROL). For both experiments, all other conventional data are assimilated (e.g., Argo, other nadir altimeters, satellite SSS, etc.), as would be the case in operational prediction scenarios. The impact of assimilating SWOT data is assessed by comparing the SWOT and CONTROL experiments for both the S2S-3 and SOCA DA systems, as well as long-range (e.g., ENSO) forecasts initialized from the two experiments.
Beyond SWOT nadir assimilation, for future work, we intend to show progress towards assimilation of wide-swath KaRIn data, which will likely require new methods which account for two-dimensional spatial correlations in the observation error covariance.