Nested regional multiscale 4dVar ocean state estimation for SWOT: From geostrophic balance to km-scale variability in the California Current System

Principle Investigator: Sarah Gille (Scripps Institution of Oceanography, UC San Diego)

Co-Investigator(s): Bruce Cornuelle, Matthew Mazloff, Ganesh Gopalakrishnan, Ariane Verdy

---

The Surface Water and Ocean Topography (SWOT) mission has launched a new era of ocean observations, allowing unprecedented access to physical processes at the 5-15-km scale. In this project we will take advantage of SWOT data to constrain regional models of the California Current System (CCS), using 4-dimensional Variational (4dVar) assimilation with the MITgcm. Traditional ocean assimilation focuses on motions at scales of about 50 to 100 km or larger, with time scales of days to weeks. At these scales motions are governed by the interactions between the Earth’s rotation and pressure gradients, referred to as geostrophic motions. In contrast, SWOT's spatial resolution captures variability at scales shorter than 50 km, with correspondingly faster temporal scales. The smallest scale motions resolved by SWOT are governed by nonlinear physical processes (e.g. baroclinic tides, internal waves, and surface gravity waves) that have traditionally been excluded from large-scale assimilation systems. These small-scale processes can excite instabilities that grow over time, leading to strong sensitivity to poorly observed small-scale features. Such instabilities potentially limit the duration of the assimilation window.

Our work will develop and deploy a multi-scale approach for the MITgcm, taking advantage of existing 4dVar capabilities to constrain large-scale geostrophic flows, and developing new 4dVar strategies that employ SWOT to constrain small-scale physics at times and locations where feasible. We employ a regional model configuration in order to achieve high-resolution assimilation, permitting us to carry out experiments that would incur prohibitive computational costs in a global domain. We specifically target the CCS because the region encompasses a SWOT one-day crossover and has a strong heritage of in situ observations, including mooring and glider data collected for SWOT cal/val as well sustained mooring, glider, and high-frequency radar measurements. The region is dynamically important in setting the characteristics of ocean conditions along the US West Coast.

Expected results of this work will address two important challenges. First, results of our multi-scale assimilation development effort will provide new insights into the predictability and controllability of small-scale motions, providing information that can guide global-scale assimilation efforts. Second, the model fields will allow us to evaluate the physical processes governing small-scale ocean dynamics, taking advantage of the fact that 4dVar assimilation produces dynamically consistent estimates of the ocean state. Model fields from our work will be freely available for the SWOT science team (and other investigators). While our primary research target is the CCS, we will use what we learn from this domain to inform similar regional assimilation efforts, including those in the Gulf of Mexico and the Gulf Stream, both of which also have extensive in situ observations and existing 4dVar MITgcm configurations. These locations are characterized by different dynamical regimes, offering the opportunity for further development and improved physical insight.