Annual sediment flux estimates in a tidal strait using surrogate measurements

doi:10.1016/j.ecss.2006.04.008

Estuarine, Coastal and Shelf Science

Volume 69, Issues 1–2, August 2006, Pages 165-178

https://doi.org/10.1016/j.ecss.2006.04.008 Get rights and content

Abstract

Annual suspended-sediment flux estimates through Carquinez Strait (the seaward boundary of Suisun Bay, California) are provided based on surrogate measurements for advective, dispersive, and Stokes drift flux. The surrogates are landward watershed discharge, suspended-sediment concentration at one location in the Strait, and the longitudinal salinity gradient. The first two surrogates substitute for tidally averaged discharge and velocity-weighted suspended-sediment concentration in the Strait, thereby providing advective flux estimates, while Stokes drift is estimated with suspended-sediment concentration alone. Dispersive flux is estimated using the product of longitudinal salinity gradient and the root-mean-square value of velocity-weighted suspended-sediment concentration as an added surrogate variable. Cross-sectional measurements validated the use of surrogates during the monitoring period. During high freshwater flow advective and dispersive flux were in the seaward direction, while landward dispersive flux dominated and advective flux approached zero during low freshwater flow. Stokes drift flux was consistently in the landward direction. Wetter than average years led to net export from Suisun Bay, while dry years led to net sediment import. Relatively low watershed sediment fluxes to Suisun Bay contribute to net export during the wet season, while gravitational circulation in Carquinez Strait and higher suspended-sediment concentrations in San Pablo Bay (seaward end of Carquinez Strait) are responsible for the net import of sediment during the dry season. Annual predictions of suspended-sediment fluxes, using these methods, will allow for a sediment budget for Suisun Bay, which has implications for marsh restoration and nutrient/contaminant transport. These methods also provide a general framework for estimating sediment fluxes in estuarine environments, where temporal and spatial variability of transport are large.

Introduction

Sediment supply to a subembayment of an estuary is determined by watershed sediment input and the sediment exchange with adjacent embayments. Sediment supply is a critical variable for investigations of habitat stability, restoration potential, and contaminant fate/transport. Suspended-sediment is needed to create and sustain valuable estuarine habitats such as tidal wetlands (Zedler and Callaway, 2001, Pont et al., 2002, Reed, 2002, Temmerman et al., 2003), though sediment-associated contaminants can also accumulate wherever sediment preferentially deposits (Hornberger et al., 1999, Arzayus et al., 2002, Taylor et al., 2004). In addition, nutrients and biota accumulate near estuarine turbidity maxima (ETM), where high suspended-sediment and contaminant concentrations are found (Peterson et al., 1975, Jassby and Powell, 1994). Enhanced biological activity in these areas may increase contaminant uptake by the food web (Kimmerer et al., 1998).

These issues converge in Suisun Bay, California (Fig. 1, Fig. 2). Over 90% of marsh area has been lost in San Francisco Bay since the 19th century, and current management goals in Suisun Bay include marsh restoration. Deposited sediment in Suisun Bay is thought to be high in mercury concentrations due to 19th century gold mining in the watershed (Hornberger et al., 1999), leading to concerns about resuspension of these deposits over long timescales (>10 yr). Cappiella et al. (1999) show net erosion in Suisun Bay since the first bathymetric surveys in the 19th century. The net sediment budget of Suisun Bay in the current era may shed light on the viability of habitat restoration as well as the magnitude of mercury introduction to the water column. In order to quantify this sediment budget, sediment flux at the landward and seaward boundaries of the subembayment must be determined.

Estimating sediment flux in estuaries can be complicated by the large range of tidal and riverine forcing. Spring and neap tides combined with periods of varying freshwater flow can introduce lateral and vertical variability of suspended-sediment concentrations (SSC) and water velocity. These variations can be induced by salinity gradients, bathymetric forcing, and secondary circulation (Ridd et al., 1998, Blanton et al., 2003). Suisun Bay, as the landward-most subembayment of San Francisco Bay, is subject to variable freshwater flow as well as semi-diurnal tides. The seaward boundary of Suisun Bay is Carquinez Strait, which is approximately 1 km wide and has a maximum depth of 40 m. Formation of an ETM has been noted in Carquinez Strait (due to a sill between Carquinez Strait and Suisun Bay; Jay and Musiak, 1994, Schoellhamer, 2001a), while density stratification has been documented during periods of high and low freshwater flow. The presence of a bend in Carquinez Strait may complicate hydrodynamic conditions.

Proper estimation of suspended-sediment flux through Carquinez Strait must account for the entire cross-section. While monitoring a large cross-section is both physically and financially impossible over the necessary time-frames, it may be possible to identify surrogate data that can be continuously monitored. The use of surrogates in estuaries has previously been explored; Uncles et al. (1998) suggested the use of salinity as a proxy for the location of the turbidity maximum in the Humber–Ouse Estuary, while Warner et al. (2005) used freshwater flow as a surrogate for the longitudinal salinity distribution in the tidal Hudson River. Allen (1990) highlighted the limitations of using marsh accretion rates as a surrogate for sea-level rise, as system characteristics may change over timescales longer than the detailed study periods. In this study, we occupied portions of the channel with autonomous equipment for 3 months, and calibrated those data to tidal-cycle cross-sectional measurements. The results from the 3-month deployment were extrapolated to quantify fluxes over water year 2004, and other years when surrogate data were available.

Section snippets

Site description

Suisun Bay spans from the Sacramento/San Joaquin River Delta (Delta) at the landward end to Carquinez Strait at the seaward end. Over one-half of the bay is shallower than 5 m at mean-lower-low-water (MLLW), though numerous deep-water (>10 m) channels run longitudinally through the bay. Two large shallow areas, Grizzly and Honker Bays, lie on the northern edge of the main channels. Maximum depths in Carquinez Strait approach 40 m. The landward end of Carquinez Strait immediately bifurcates at the

Spring monitoring period

Point SSC in the four quadrants was successfully measured for the duration of the spring monitoring period, though fouling and debris decreased the percentage of valid data (Fig. 3). At site NBen, SSC data yield was 79% and 94% for the upper and lower sensors, respectively. Data yield at site SBen was 64% and 91% for the upper and lower sensors, respectively. SSC was elevated at all sites near the beginning of the record, due to recent sediment supply from the Delta, overlapping with a spring

Accuracy of long-term prediction

The extrapolation of relations developed during the transition between wet and dry season introduces errors of an unknown magnitude, but this does not change the seasonal pattern of suspended-sediment transport. The most intensive measurement period was conducted during tidal-cycles in June and July 2004, when freshwater flow was minimized and salinity gradient was at a maximum. Those efforts result in relations for channel-average velocity, channel cross-sectional area, and velocity-weighted

Conclusions

Despite the highly variable nature of velocity, salinity, and SSC in estuarine cross-sections, reasonable measurements of suspended-sediment flux can be made over limited periods. Ideally, these measurements can be related to continuously monitored surrogates. Such a procedure was performed for Carquinez Strait, California, the seaward boundary of Suisun Bay. The calculated fluxes show a clear pattern of seasonality to the advective and dispersive fluxes; these patterns suggest that

Acknowledgments

This work was paid for in part with funds from California State Water Resources Control Board contract #01-281-150-2. Tidal-cycle measurements were performed with the assistance of Curt Battenfeld, Greg Brewster, Paul Buchanan, Jay Cuetara, Megan Lionberger, Heather Ramil, Greg Shellenbarger, and Brad Sullivan. ADCP data were analyzed by Randal Dinehart. Catherine Ruhl provided Delta flow data and advice. Jon Burau supplied invaluable input concerning the hydrodynamics of the area. Jan

References (38)

K.M. Arzayus et al.
Effects of physical mixing on the attenuation of polycyclic aromatic hydrocarbons in estuarine sediments
Organic Geochemistry
(2002)
J.O. Blanton et al.
Transport of salt and suspended sediments in a curving channel of a coastal plain estuary: Satilla River, GA
Estuarine, Coastal and Shelf Science
(2003)
K.R. Dyer
The salt balance in stratified estuaries
Estuarine and Coastal Marine Science
(1974)
M.I. Hornberger et al.
Historical trends of metals in the sediments of San Francisco Bay, California
Marine Chemistry
(1999)
A.D. Jassby et al.
Hydrodynamic influences on interannual chlorophyll variability in an estuary: upper San Francisco Bay–Delta (California, U.S.A.)
Estuarine, Coastal and Shelf Science
(1994)
R.G. Linville et al.
Increased selenium threat as a result of invasion of the exotic bivalve Potamacorbula amurensis into the San Francisco Bay–Delta
Aquatic Toxicology
(2002)
D.H. Peterson et al.
Location of the non-tidal current null zone in northern San Francisco Bay
Estuarine and Coastal Marine Science
(1975)
P.V. Ridd et al.
Density-driven secondary circulation in a tropical mangrove estuary
Estuarine, Coastal and Shelf Science
(1998)
A.F. Shchepetkin et al.
The Regional Ocean Modeling System (ROMS): a split-explicit, free-surface, topography-following coordinates ocean model
Ocean Modelling
(2005)
S. Temmerman et al.
Modelling long-term tidal marsh growth under changing tidal conditions and suspended sediment concentrations, Scheldt estuary, Belgium
Marine Geology
(2003)

J.C. Warner et al.

Floodtide pulses after low tides in shallow subembayments adjacent to deep channels

Estuarine, Coastal and Shelf Science

(2004)

J.R.L. Allen

Constraints on measurement of sea-level movements from salt-marsh accretion rates

Journal of the Geological Society

(1990)

P.A. Buchanan et al.

Summary of suspended-sediment concentration data in San Francisco Bay, California, water year 2003

U.S. Geological Survey Data Series

(2005)

Burau, J.R., Simpson, M.R., Cheng, R.T., 1993. Tidal and residual currents measured by an acoustic Doppler current...

California Department of Water Resources

DAYFLOW program documentation and DAYFLOW data summary user's guide

Cappiella, K., Malzone, C., Smith, R., Jaffe, B., 1999. Sedimentation and bathymetry changes in Suisun Bay, 1867–1990....

Ganju, N.K., Schoellhamer, D.H. Lateral displacement of the estuarine turbidity maximum in tidal strait. Proceedings of...

Jaffe, B.E., Smith, R.E., Torresan, L.Z., 1998. Sedimentation and bathymetric change in San Pablo Bay, 1856–1983. U.S....

D.A. Jay et al.

Particle trapping in estuarine turbidity maxima

Journal of Geophysical Research

(1994)

Cited by (40)

Uncertainties associated with in situ high-frequency long-term observations of suspended particulate matter concentration using optical and acoustic sensors
2019, Progress in Oceanography
Citation Excerpt :
Water clarity is an important parameter to understand marine ecosystems and is mainly controlled by suspended particulate matter concentration (SPMC). To detect variability in SPMC, networks of observational platforms have been installed worldwide using optical and acoustical sensors (e.g. Butman et al., 1979; Grabemann and Krause, 1989 Fettweis et al., 1998; Guézennec et al., 1999; Ganju and Schoellhamer, 2006; Badewien et al., 2009; Palinkas et al., 2010; Garel and Ferreira, 2011; Nauw et al., 2014; Anastasiou et al., 2015; Jalón-Rojas et al., 2015; Many et al., 2016; Druine et al., 2018) as well as sensors or samples that give information on the shape, size and composition of the SPM (e.g. Krivtsov et al., 2008; Cartwright et al., 2009; Gray and Gartner, 2009; Fettweis and Baeye, 2015; van der Hout et al., 2015; Chapalain et al., 2019). In addition to these sensors gravimetric measurements of filtered water samples are generally used as ground truth reference (e.g. Neukermans et al., 2012a).
Measurement of suspended particulate matter concentration (SPMC) spanning large time and geographical scales have become a matter of growing importance in recent decades. At many places worldwide, complex observation platforms have been installed to capture temporal and spatial variability over scales ranging from cm (turbulent regimes) to whole basins. Long-term in situ measurements of SPMC involve one or more optical and acoustical sensors and, as the ground truth reference, gravimetric measurements of filtered water samples. The estimation of SPMC from optical and acoustical proxies generally results from the combination of a number of independent calibration measurements, as well as regression or inverse models. Direct or indirect measurements of SPMC are inherently associated with a number of uncertainties along the whole operation chain, the autonomous field deployment, to the analyses necessary for converting the observed proxy values of optical and acoustical signals to SPMC. Controlling uncertainties will become an important issue when the observational input comprises systems of sensors spanning large spatial and temporal scales. This will be especially relevant for detecting trends in the data with unambiguous statistical significance, separating anthropogenic impact from natural variations, or evaluating numerical models over a broad ensemble of different conditions using validated field data.
The aim of the study is to present and discuss the benefits and limitations of using optical and acoustical backscatter sensors to acquire long-term observations of SPMC. Additionally, this study will formulate recommendations on how to best acquire quality-assured SPMC data sets, based on the challenges and uncertainties associated with those long-term observations. The main sources of error as well as the means to quantify and reduce the uncertainties associated with SPMC measurements are also illustrated.
Application of an unstructured 3D finite volume numerical model to flows and salinity dynamics in the San Francisco Bay-Delta
2017, Estuarine, Coastal and Shelf Science
Citation Excerpt :
Model estimates of discharge were compared to daily, tidally filtered cross-sectional discharge measurements. Modeled discharges were tidally filtered using a Butterworth filter with a cutoff frequency of 1/30 h, as performed in Ganju and Schoellhamer (2006) and Achete et al. (2015). Net Delta Outflow (NDO) was the flow metric used to assess model performance of seasonal fluvial outflow from the Delta into the Bay.
A linked modeling approach has been undertaken to understand the impacts of climate and infrastructure on aquatic ecology and water quality in the San Francisco Bay-Delta region. The Delft3D Flexible Mesh modeling suite is used in this effort for its 3D hydrodynamics, salinity, temperature and sediment dynamics, phytoplankton and water-quality coupling infrastructure, and linkage to a habitat suitability model. The hydrodynamic model component of the suite is D-Flow FM, a new 3D unstructured finite-volume model based on the Delft3D model. In this paper, D-Flow FM is applied to the San Francisco Bay-Delta to investigate tidal, seasonal and annual dynamics of water levels, river flows and salinity under historical environmental and infrastructural conditions. The model is driven by historical winds, tides, ocean salinity, and river flows, and includes federal, state, and local freshwater withdrawals, and regional gate and barrier operations. The model is calibrated over a 9-month period, and subsequently validated for water levels, flows, and 3D salinity dynamics over a 2 year period.
Model performance was quantified using several model assessment metrics and visualized through target diagrams. These metrics indicate that the model accurately estimated water levels, flows, and salinity over wide-ranging tidal and fluvial conditions, and the model can be used to investigate detailed circulation and salinity patterns throughout the Bay-Delta. The hydrodynamics produced through this effort will be used to drive affiliated sediment, phytoplankton, and contaminant hindcast efforts and habitat suitability assessments for fish and bivalves. The modeling framework applied here will serve as a baseline to ultimately shed light on potential ecosystem change over the current century.
Seasonal variations in suspended-sediment dynamics in the tidal reach of an estuarine tributary
2013, Marine Geology
Quantifying sediment supply from estuarine tributaries is an important component of developing a sediment budget, and common techniques for estimating supply are based on gages located above tidal influence. However, tidal interactions near tributary mouths can affect the magnitude and direction of sediment supply to the open waters of the estuary. We investigated suspended-sediment dynamics in the tidal reach of Corte Madera Creek, an estuarine tributary of San Francisco Bay, using moored acoustic and optical instruments. Flux of both water and suspended-sediment were calculated from observed water velocity and turbidity for two periods in each of wet and dry seasons during 2010. During wet periods, net suspended-sediment flux was seaward; tidally filtered flux was dominated by the advective component. In contrast, during dry periods, net flux was landward; tidally filtered flux was dominated by the dispersive component. The mechanisms generating this landward flux varied; during summer we attributed wind–wave resuspension in the estuary and subsequent transport on flood tides, whereas during autumn we attributed increased spring tide flood velocity magnitude leading to local resuspension. A quadrant analysis similar to that employed in turbulence studies was developed to summarize flux time series by quantifying the relative importance of sediment transport events. These events are categorized by the direction of velocity (flood vs. ebb) and the magnitude of concentration relative to tidally averaged conditions (relatively turbid vs. relatively clear). During wet periods, suspended-sediment flux was greatest in magnitude during relatively turbid ebbs, whereas during dry periods it was greatest in magnitude during relatively turbid floods. A conceptual model was developed to generalize seasonal differences in suspended-sediment dynamics; model application to this study demonstrated the importance of few, relatively large events on net suspended-sediment flux. These results suggest that other estuarine tributaries may alternate seasonally as sediment sinks or sources, leading to the conclusion that calculations of estuary sediment supply from local tributaries that do not account for tidal reaches may be overestimates.
The use of modeling and suspended sediment concentration measurements for quantifying net suspended sediment transport through a large tidally dominated inlet
2013, Marine Geology
Sediment exchange at large energetic inlets is often difficult to quantify due complex flows, massive amounts of water and sediment exchange, and environmental conditions limiting long-term data collection. In an effort to better quantify such exchange this study investigated the use of suspended sediment concentrations (SSC) measured at an offsite location as a surrogate for sediment exchange at the tidally dominated Golden Gate inlet in San Francisco, CA. A numerical model was calibrated and validated against water and suspended sediment flux measured during a spring–neap tide cycle across the Golden Gate. The model was then run for five months and net exchange was calculated on a tidal time-scale and compared to SSC measurements at the Alcatraz monitoring site located in Central San Francisco Bay ~ 5 km from the Golden Gate. Numerically modeled tide averaged flux across the Golden Gate compared well (r² = 0.86, p-value < 0.05) with 25 h low-pass filtered (tide averaged) SSCs measured at Alcatraz over the five month simulation period (January through April 2008). This formed a basis for the development of a simple equation relating the advective flux at Alcatraz with suspended sediment flux across the Golden Gate. Utilization of the equation with all available Alcatraz SSC data resulted in an average export rate of 1.2 Mt/yr during water years 2004 through 2010. While the rate is comparable to estimated suspended sediment inflow rates from sources within the Bay over the same time period (McKee et al., 2013-this issue), there was little variation from year to year. Exports were computed to be greatest during the wettest water year analyzed but only marginally so.
A step decrease in sediment concentration in a highly modified tidal river delta following the 1983 El Niño floods
2013, Marine Geology
Anthropogenic activities in watersheds can have profound effects on sediment transport through river systems to estuaries. Disturbance in a watershed combined with alterations to the hydro-climatologic regime may result in changes to the sediment flux, and exacerbate the impacts of extreme events (such as large-magnitude floods) on sediment transport. In the San Francisco Estuary, suspended sediment has been declining over the past 30 years as a result of declining sediment supply, contributing to dramatic changes in the ecology and geomorphology of the estuary. However, the decline has not been gradual. Recent observations of an abrupt decrease in suspended sediments in the San Francisco Bay have been explained by a model that suggests that the step change has occurred due to exceedance of a sediment regulation threshold that triggered the change from a sediment transport regime to a supply-limited system. We investigated structural changes in the historical record of total suspended solids (TSS) concentration measured in the upper estuary to verify the model predictions. TSS in the upper estuary exhibited an abrupt step decrease in 1983 corresponding to the record-high winter and summer flows from the 1982 to 1983 El Niño event. After this step change, TSS concentrations had a significant declining trend despite subsequent near-record high flows. The abrupt change in TSS followed by the declining trend provides evidence for the hypothesis of sediment supply limitation in the San Francisco Estuary.
Sediment transport in the San Francisco Bay Coastal System: An overview
2013, Marine Geology
The papers in this special issue feature state-of-the-art approaches to understanding the physical processes related to sediment transport and geomorphology of complex coastal–estuarine systems. Here we focus on the San Francisco Bay Coastal System, extending from the lower San Joaquin–Sacramento Delta, through the Bay, and along the adjacent outer Pacific Coast. San Francisco Bay is an urbanized estuary that is impacted by numerous anthropogenic activities common to many large estuaries, including a mining legacy, channel dredging, aggregate mining, reservoirs, freshwater diversion, watershed modifications, urban run-off, ship traffic, exotic species introductions, land reclamation, and wetland restoration. The Golden Gate strait is the sole inlet connecting the Bay to the Pacific Ocean, and serves as the conduit for a tidal flow of ~ 8 × 10⁹ m³/day, in addition to the transport of mud, sand, biogenic material, nutrients, and pollutants. Despite this physical, biological and chemical connection, resource management and prior research have often treated the Delta, Bay and adjacent ocean as separate entities, compartmentalized by artificial geographic or political boundaries. The body of work herein presents a comprehensive analysis of system-wide behavior, extending a rich heritage of sediment transport research that dates back to the groundbreaking hydraulic mining-impact research of G.K. Gilbert in the early 20th century.

View all citing articles on Scopus

View full text

Annual sediment flux estimates in a tidal strait using surrogate measurements

Abstract

Introduction

Section snippets

Site description

Spring monitoring period

Accuracy of long-term prediction

Conclusions

Acknowledgments

Organic Geochemistry

Estuarine, Coastal and Shelf Science

Estuarine and Coastal Marine Science

Marine Chemistry

Estuarine, Coastal and Shelf Science

Aquatic Toxicology

Estuarine and Coastal Marine Science

Estuarine, Coastal and Shelf Science

Ocean Modelling

Marine Geology

Estuarine, Coastal and Shelf Science

Constraints on measurement of sea-level movements from salt-marsh accretion rates

Journal of the Geological Society

Summary of suspended-sediment concentration data in San Francisco Bay, California, water year 2003

U.S. Geological Survey Data Series

DAYFLOW program documentation and DAYFLOW data summary user's guide

Particle trapping in estuarine turbidity maxima

Journal of Geophysical Research