3-D Simulations of the San Francisco Estuary with Subgrid Bathymetry to Explore Long-Term Trends in Salinity Distribution and Fish Abundance

The UnTRIM hydrodynamic model was applied to San Francisco Bay and the Sacramento–San Joaquin Delta (Delta) using a coarse-resolution model grid with bathymetry represented at a finer subgrid scale. We simulated a 35-year period, spanning from January 1, 1980 through December 31, 2014. This simulation was used to develop salinity distribution maps to facilitate visualization of fish distribution and abundance data. We compared predicted salinity from the coarse-grid UnTRIM Bay–Delta model to continuous salinity monitoring observations as well to the measured surface salinity from San Pablo Bay through the Delta at a total of 5,542 times and locations where surface salinity was observed as part of several long-term fish monitoring programs: the Fall Midwater Trawl, Summer Townet Survey, and San Francisco Bay Study. The coarse-grid UnTRIM Bay–Delta model was shown to accurately predict hydrodynamics and the spatial distribution of salinity over both a 3-year detailed validation period and over the full 35-year analysis period. The predicted salinity was used to calculate the daily position of X2 and the daily-averaged area of the Low Salinity Zone (LSZ) for each day during the 35-year simulation. Our analysis highlights the influence of multi-year climate patterns, shorter-duration weather patterns, and Delta outflow on salinity distribution. We used the predicted salinity to develop maps of salinity distribution over seven periods for six fish species, and combined the salinity maps with historic fish sampling data to allow for visualization of fish abundance and distribution for 33 years between 1980 and 2012. These maps can be used to explore how different species respond to annual differences in salinity distributions in the San Francisco Estuary, and to expand the understanding of the relationships among salinity and fish abundance, distribution, and population resiliency.