San Francisco Estuary and Watershed Science

Sacramento River winter-run Chinook Salmon (Oncorhynchus tshawytscha) are an endangered population that faces numerous challenges across its life cycle, including juvenile out-migration through the heavily anthropogenically modified Sacramento–San Joaquin Delta, or Delta. Water exports from pumping facilities in the Delta can alter local hydrology and influence movement of out-migrating juveniles, some of which are observed in or near pumping facilities. Monitoring and regulations, intended to protect out-migrating fish through restrictions on pumping, are predicated on assumed relationships among fish observations, water operations, and through-Delta migratory survival. In this study, we use a new conceptual model to review the current state of science for winter-run Chinook Salmon out-migration survival in the Delta, and use simulation modeling to address pertinent knowledge gaps. Results of this study highlight varying support for the influence of Sacramento River flow, temperature, and water exports on routing and survival in different regions of the Delta. The contributions of specific routing pathways to the interior Delta (e.g., through Threemile Slough) to survival, and the relationship between fish entrainment at pumping facilities and overall migratory survival, remain uncertain. Recommended future work includes continued fine-scale acoustic telemetry studies throughout the Delta, novel integrated modeling of monitoring data, and contextualizing the relevance of Delta-based survival to population viability by incorporating explicit uncertainties about survival into existing life-cycle models.

The restoration of native species-dominated ecosystems is critical for improving ecosystem health and meeting policy goals in the Sacramento–San Joaquin Delta and Suisun Marsh (upper San Francisco Estuary, collectively), one of the largest estuarine systems in North America. To accomplish large-scale restoration in this heavily altered system, a variety of projects, programs, and motivations inform restoration planning and implementation. Chapter 4 of the Delta Plan synthesizes restoration goals across these efforts to produce comprehensive ecosystem restoration targets of between 60,000 and 80,000 acres across seven ecosystem types by 2050, but a comprehensive review of restoration progress and planning to date is needed. To fill this gap, this paper analyzes the current state of ecosystem restoration in the upper San Francisco Estuary in the context of the Delta Plan targets. We review current scientific and management literature and implementation approaches, and synthesize acreage totals across completed, in-progress, and planned projects for four ecosystem types where substantial development of restoration in the system has occurred: tidal wetland, non-tidal wetland, riparian, and floodplain. We find that tidal wetland restoration has progressed more rapidly than other ecosystem types, motivated by mitigation requirements related to the federal Endangered Species Act. Across all ecosystem types, we identify both promising progress and clear needs for accelerated planning and implementation of restoration projects to meet Delta Plan 2050 targets, and discuss ongoing needs related to science, funding, and implementation.

Most managed wetlands in California are ephemeral and are purposefully flooded during the fall and winter for over-wintering waterfowl and are dry during the spring and summer waterfowl breeding season. Only semi-permanent and permanent wetlands remain flooded through the critical summer brood-rearing period for ducklings. We examined flooded wetland availability for breeding waterfowl in the brackish Suisun Marsh (California, USA) annually during the spring (April 27–May 17, during peak nesting) and summer (June 17–July 7, during peak duckling brood rearing), for a 38-year period using Landsat satellite imagery and spectral mixture analysis. Flooded wetland area increased 43% in spring and 48% in summer from 1984 to 2021 but varied among years (spring: 37.6–88.6 km²; summer: 17.7–57.5 km²). This increase in flooded wetland area over the past four decades was due to just a few sites, with only 24% (spring) and 15% (summer) of the 198 land-owner parcels increasing in flooded area. Flooded wetland area in the spring was unrelated to annual precipitation between October and April (range: 25–104 cm) or spring precipitation between January and April (range: 8-65 cm), whereas flooded wetland area in the summer was weakly correlated to both annual and spring precipitation. Flooded wetland area in spring and summer was also weakly correlated with the median daily outflow from the Sacramento-San Joaquin River Delta between March 15 and June 15, which corresponds to a critical period of wetland water management for breeding waterfowl. Our results indicate that spring and summer flooded wetland habitat for breeding waterfowl has slightly increased over the past four decades, varies annually, and is mostly dependent on local wetland management practices rather than on precipitation or Delta outflows. Managing habitats as semi-permanent wetlands would increase flooded wetland habitat in the spring and summer and provide habitat for nesting hens and ducklings.

Zooplankton are a key food source for juvenile fishes in estuaries worldwide, including California’s Sacramento–San Joaquin Delta (hereafter Delta); both zooplankton quality and quantity are critical to ecosystem health. Zooplankton may be affected by pesticides in water and the food web, and the Delta is known to contain complex pesticide mixtures. In this study, we evaluated pesticide concentrations in water and zooplankton in the northern Delta during (1) the summer–fall of 2017, 2018, and 2019, which included periods of augmented pulse flows from agriculture tailwater, and (2) across a full seasonal cycle from May 2019 to March 2020. We quantified changes in pesticide concentration in response to environmental factors. We found that zooplankton showed more frequent detections of hydrophobic pesticides compared to more frequent detections of hydrophilic compounds in water. Pesticide concentrations were influenced by flow, pesticide application, and season, but the effects of these environmental factors differed by habitat (Sacramento River or Yolo Bypass Toe Drain). Pesticides in water responded similarly to environmental factors in the Sacramento River and Yolo Bypass, whereas pesticides in zooplankton responded differently. In water, we found more detections and higher concentrations at higher flows in the Yolo Bypass and Sacramento River, but responses to pesticide application varied by habitat. Alternatively, pesticide concentrations in zooplankton increased in the Yolo Bypass with increasing flow (correlated with flow pulses) and changed seasonally; whereas, pesticide concentrations in zooplankton in the Sacramento River decreased at higher flows, and decreased with or did not respond to higher pesticide application in the watershed. Our study suggests that augmented flows—particularly those using agricultural tailwater—may have unintended negative ecological effects that could partially offset benefits to the food web and fishes in the northern Delta, underscoring the complex interplay among factors that drive increased pesticide exposure.

Strategies for endangered species conservation may have different outcomes depending on the habitat context in which they are implemented. Understanding these context-dependent effects can help optimize and target management efforts. In this analysis, we investigate how environmental and food-web conditions interactively affect condition and foraging of Delta Smelt (Hypomesus transpacificus), an endangered fish endemic to the San Francisco Estuary (the estuary). Food limitation, in terms of pelagic zooplankton availability, is considered a main factor that contributes to the decline in Delta Smelt abundance. Our overarching objective was to examine whether the effect of zooplankton on Delta Smelt depended on habitat context. Specifically, we hypothesized that zooplankton would less positively effect Delta Smelt condition—as measured by hepatosomatic index (HSI)—and foraging success in areas with nearby tidal wetlands, because these adjacent habitats may provide access to prey items from the epibenthos and fringing vegetation. In contrast, in regions with limited proximity to wetlands, we hypothesized that Delta Smelt would rely more on pelagic prey, which would manifest as a more positive effect of zooplankton on body condition and foraging success for Delta Smelt. Using models that accounted for habitat in multiple ways, we found little evidence that zooplankton and habitat interactively influenced Delta Smelt condition or gut fullness. Rather, the influence of zooplankton on HSI and gut fullness was generally positive across habitat contexts. Given the well-documented food limitation in the estuary, promoting the availability of pelagic zooplankton is a rational, albeit complex, management aim. Furthermore, our results suggest that efforts to increase zooplankton would broadly benefit Delta Smelt across a wide range of habitat contexts.

Sediment bulk density (ρ-dry) and particle size are two important parameters for predicting sediment bed erosion. ρ-dry, however, is difficult to measure accurately. The units of ρdry have not been consistently reported in the literature, leading to confusion, particularly in the calculation of sediment budgets that typically require integrating mass-based and volumetric components. Relationships between ρdry and sediment composition have been developed for multiple regions and differ between systems. Developing a system-specific predictive model for ρdry can help fill data gaps and improve sediment budgets, model accuracy, and estimates of quantities of sediment needed for restoration. In this study, we investigate whether ρdry in San Francisco Estuary can be predicted from organic carbon content or percent of fines, which are more easily or frequently measured than ρdry. We compiled sediment properties from samples collected over the past decade throughout the intertidal and subtidal regions of San Francisco Bay and the Sacramento–San Joaquin Delta to examine this relationship. Sample composition ranged from 2.18 to 99.97% fines (particles < 0.0625 mm), ρ-dry ranged from 0.22 to 1.60 g cm-3, and organic carbon ranged from 0.06 to 7.98%. Regression analysis indicates that the percent of fines explains 93% of the variation of ρ-dry (p-value < 0.05, N = 81). The coefficient of determination decreased by ~1% when organic carbon was incorporated in the regression analysis. Comparison of this predictive ρ-dry model to four published models based on samples from other regions supports previous findings that the relationship between ρdry and grain size may vary by system. We also examined additional factors that may affect sediment erodibility, such as hydrographic and oceanographic conditions. Classification of sample sites as intertidal vs. subtidal or wavy vs. non-wavy each significantly explained the residuals from the ρdry model, and both intertidal and wavy conditions were associated with higher ρ-dry values.