Search

After the surface water and ocean topography (SWOT) satellite launches in 2022, an in‐situ field campaign will be conducted to estimate the ocean state for calibration and validation (CalVal) purposes. It is demonstrably difficult to capture the sea surface height (SSH) features that are the focus of SWOT, with short time periods (<20 days) and fine‐scale spatial structures (15–150 km). Therefore, a critical component of the SWOT CalVal will be a data assimilation (DA) system coupled to a primitive equation numerical model that can estimate: (a) the 2D SSH over the SWOT swaths, and (b) the 3D dynamical (velocity) field. To explore the ability of DA to meet the challenges of SWOT fine‐scale observations, a multiscale DA system based on an extended 3D variational method has been developed. Here, we present a strategic evaluation of this DA system, with a focus to assimilate in‐situ data from an observing system to best represent the fine‐scale ocean variability at the CalVal site. The DA estimate is compared to independent observations taken during the 2019 pre‐launch field campaign 300‐km off Monterey Bay, California. The key result is this DA system can reconstruct the upper 500‐m steric height with O(1‐cm) error at hourly resolution, and subcentimeter error for periods longer than 2 days. Plain Language Summary: The surface water and ocean topography (SWOT) satellite mission is the first of its kind to launch. It will provide unprecedented measurements of the sea surface height in two dimensions (2‐D) and shorter spatial scales than previously possible. These measurements will allow oceanographers to study new scales of ocean variability and infer the ocean current velocity field. After SWOT launches, the accuracy of its measurements will need to be verified with observations made using standard instrumentation whose accuracy is well known. However, the cost to produce equivalent measurements for comparison with 2‐D high‐resolution SWOT observations is prohibitive, so the instruments to be deployed are only sufficient to evaluate a small portion of the satellite footprint. To better understand the satellite measurements, the plan is to combine the instrument measurements with a numerical ocean model that provides a four‐dimensional best‐estimate of the ocean state. This is called data assimilation (DA). In this paper, we evaluate a DA system developed to address these challenges, and show its successful performance using observations from a pre‐launch field campaign in 2019. Key Points: Data assimilation (DA) of routine observations of large/mesoscale circulation allows effective assimilation of mooring data at fine scalesMooring DA reconstructs 0‐ to 500‐m hourly steric height with O(1‐cm) error, and subcentimeter error for periods longer than 2‐dayDA performance relies on observing system design; capturing the 2D horizontal gradient field is important for SWOT evaluation [ABSTRACT FROM AUTHOR]