Abstract
Projected accelerations in the rate of sea-level rise (SLR) later this century have the potential to structurally reshape the coastal environment. We sought to assess the vulnerability of tidal marsh habitats to be lost or altered due to SLR across Don Edwards San Francisco Bay National Wildlife Refuge in south San Francisco Bay using extensive field datasets including baseline surveys of marsh surface elevation and vegetation as well as soil characteristics and accretion rates derived from soil cores. Across eight marsh parcels we collected three 1 m deep soil cores in low, mid, and high marsh zones, and processed them for bulk density (mean = 0.39 ± 0.05 g cm-3) and organic matter percent (mean = 12.7 ± 3.8%). A subset of cores was dated using gamma spectroscopy methods; four cores had usable data for the calculation of accretion rates (Mowry: 0.29 cm yr-1, Triangle: 0.38 cm yr-1, Guadalupe Slough: 0.91 cm yr-1, and Coyote Creek Lagoon: 0.67 cm yr-1). We had a low success rate using radioisotopes to date soil cores, likely due to extensive sediment resuspension that occurs across south San Francisco Bay. Data from these four sites were then used to calibrate WARMER3, a biogeomorphic model of wetland soils and projected future elevation, plant species composition, and carbon accumulation under four SLR scenarios and three sediment availability scenarios. The model showed that all marshes persisted under the low SLR scenario, which simulated the continuation of historic SLR without rapid acceleration. Under intermediate SLR (~+1 m by 2100), all four marshes persisted, however Mowry and Triangle were projected to convert from a pickleweed-dominant high marsh to Spartina-dominant low marsh. With intermediate high SLR scenario (~+1.5 m by 2100) Mowry and Triangle were projected to convert to an unvegetated state and Guadalupe Slough to Spartina-dominant, while with high (~+2 m by 2100) SLR, Mowry, Triangle, and Coyote Creek Lagoon were projected to become unvegetated by the end of the century; only Guadalupe Slough was projected to persist as a vegetated marsh under the highest SLR scenario. Changing sediment availability altered marsh vulnerability at Guadalupe Slough and Coyote Creek Lagoon but had minor effects at the other two sites. This study highlights marsh resilience under low or modest SLR projections, primarily due to high sediment availability and accretion rates, but vulnerability under higher rates of SLR. It also demonstrates the challenges in using radioisotopes to measure accretion rates in a part of the estuary that is subject to extensive sediment resuspension. Long-term contemporary monitoring of accretion in south San Francisco Bay is critical to understand SLR vulnerability as the estuary conditions continue to change.
