AUTHOR=Mazzini Piero L. F. , Pianca Cassia , Pareja-Roman L. Fernando , Cole Kelly L. , Walter Ryan K. , Castelao Renato M. , Hunter Elias J. , Chant Robert J. 

TITLE=Spatio-temporal variability of San Francisco Bay Plume from space

JOURNAL=Frontiers in Marine Science

VOLUME=Volume 12 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2025.1588441

DOI=10.3389/fmars.2025.1588441

ISSN=2296-7745

ABSTRACT=As brackish turbid waters exit San Francisco Bay, one of the largest estuaries in the U.S. West Coast, they form the San Francisco Bay Plume (SFBP), which spreads offshore and influences the Gulf of the Farallones (GoF), an ecologically significant region in the California Current System that is also home to three National Marine Sanctuaries. This paper provides the first observationally based investigation of the spatio-temporal variability of the SFBP, using a plume tracking algorithm applied to more than two decades (2002-2023) of ocean color data from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor onboard satellites Aqua and Terra. The turbid SFBP spreads radially, extending 10-20 km offshore around 50% of the time, and during extreme discharge events (<1% of the time), the plume can reach nearly 60 km offshore to the shelf break. The greatest variability in frequency of plume occurrence was observed 10-20 km offshore and it was largely explained by the seasonal cycle (80% of total variance), linked primarily to seasonal changes in river discharge. Largest plume areas (determined by summing up all pixel areas weighted by their respective fraction of plume occurrence) were observed during winter and smallest during summer, occupying on average 24% and 1.5% of GoF area, respectively. Beyond 20-30 km offshore, variability in frequency of plume occurrence was dominated by the intraseasonal band (50-80% of total variance), attributed to plume response to synoptic wind-forcing and/or filaments and eddies, while the interannual band played a secondary role in the plume variability (<20% of total variance). Finally, a multivariable linear regression model of the turbid SFBP area was created to explore the potential predictability of the plume’s influence in the GoF. The model included the annual and semi-annual cycles and discharge anomalies (deseasoned and detrended), and despite its simplicity, it explained over 78% of total variance of the turbid SFBP area. Therefore, it could be a useful tool for scientists and stakeholders to better understand how management actions on freshwater supply can have consequences offshore beyond the Golden Gate and help guide future management decisions in this ecologically important region.