Quantifying wavelengths constrained by simulated SWOT observations in a submesoscale resolving ocean analysis/forecasting system. (March 2019)

• Record Type: Journal Article
• Authors: D'Addezio, Joseph M.; Smith, Scott; Jacobs, Gregg A.; Helber, Robert W.; Rowley, Clark; Souopgui, Innocent; Carrier, Matthew J.
• Is Part Of: Ocean modelling. Volume 135(2019)
• Page Start: 40
• Page End: 55
• Publication Date: 2019-03
• Subjects: OSSE -- SWOT -- Altimeter -- 3DVAR -- Mesoscale -- Submesoscale; Oceanography -- Periodicals; Océanographie -- Périodiques; Oceanography; Periodicals; 551.46
• DOI: 10.1016/j.ocemod.2019.02.001
• Languages: English
• ISSNs: 1463-5003

Abstract

Abstract: Using a suite of Observing System Simulation Experiments (OSSEs), the utility of simulated Surface Water Ocean Topography (SWOT) observations is estimated in a high-resolution (1 km) ocean analysis/forecasting system. Sampling a Nature Run provides observations for the OSSEs and the realism of the Nature Run is established by comparison to climatological data and an independent ocean analysis/forecast system. Each OSSE experiment assimilated different sets of simulated observations including traditional nadir altimeters, satellite sea surface temperature (SST), in situ profile data, and SWOT. OSSE evaluation metrics include area-averaged errors and wavenumber spectra with the latter providing much finer differentiation between experiments. 100 m temperature, sea surface height (SSH), and mixed layer depth (MLD) errors across the observed wavenumber spectra were reduced by up to 20% for OSSEs assimilating the simulated SWOT observations. The minimum constrained wavelength was found to be 130 km when both nadir altimetry and SWOT observations were used. The experiment using only nadir altimetry produced a value of 161 km. This 31 km gain in skill of predictable scales suggests that ocean forecasts can expect substantial gains in capability when utilizing the forthcoming SWOT data. Experimentation with the analysis decorrelation length scale suggests that emerging multi-scale assimilation methodologies will provide additional advancements in predictive skill. Highlights: An OSSE is constructed to quantify the utility of SWOT data for ocean prediction. Wavenumber spectral analyses show that SWOT decreases errors by up to 20%. SWOT, combined with nadir altimeters, constrains wavelengths of 130 km and greater. Experimentation shows the need for multi-scale assimilation to fully utilize SWOT.