CURRENT RESEARCH

Biogeochemical cycling of selenium in the San Francisco Bay and Sacramento-San Joaquin River Delta

Funding Sources

National Science Foundation, Environmental Geochemistry and Biogeochemistry Program (http://www.nsf.gov/), Grant OCE 9707946

CALFED (California Bay-Delta Authority; http://calwater.ca.gov/), Grants 98-2015000-00096; ERP-01-C07

Background

San Francisco Bay is the second largest estuary on the western coast of North America where seawater enters through the Golden Gate and mixes in the "North Bay" with fresh water from the Sacramento and San Joaquin Rivers (these come together in an complex series of canals, islands, and marshes to the east of the SF Bay in a region called the "Delta") in the Central, San Pablo, and Suisun Bays. Seawater also moves into the tidally-dominated South San Francisco Bay.

San Francisco Bay is a habitat for many commercial and recreational fisheries, vital for international trade (major US port city), the site of many oil refineries that process crude from Alaska and California, and the home to more than seven million people.

Selenium is an essential, but also toxic, trace element whose chemical speciation (form) and concentration affect this dual role. Dissolved selenium exists as the oxyanions selenate (Se+6 as SeO42-) and selenite (Se+4 as SeO32-+HSeO3-), and as organic selenides (Se-2), primarily in the form of dissolved free amino acids and soluble peptides. In the particulate state, Se can be found as adsorbed selenate and selenite, particulate organic selenide, and as insoluble elemental selenium (Se0). The selenium cycle in the San Francisco Bay is depicted in the following diagram.

Dissolved selenium is introduced to the Bay via the Pacific Ocean, river inputs from the Sacramento/San Joaquin Delta, and from oil refinery effluents that enter the Bay in the region near the Carquinez Strait. This cycle includes the multistep regeneration of particulate organic selenide to dissolved selenate (via dissolved organic selenide and selenite) and the selective uptake of dissolved Se species (organic selenide = selenite >selenate) by autotrophs (phytoplankton) and perhaps heterotrophic bacteria. In the process of uptake, selenite and selenate are reduced to particulate organic selenides. This organic selenide can be recycled as above, transferred to grazers such as zooplankton or benthic invertebrates, or deposited in the underlying sediments. Dissolved Se can also exchange with the sediments via diffusion (depending on the concentration gradient), with reducing sediments producing elemental Se via the dissimilatory reduction of selenate and selenite. Of course all of this cycling is superimposed upon physical transport (Advective/Diffusive Transport) processes driven by river and tidal flows, and dispersion that are major features of the Bay.

<<Top

Major Research Topics and Questions

1. Until the late 1990's, oil refineries were a major input of dissolved selenium to the San Francisco Bay (Cutter, 1989; Cutter and San Diego-McGlone, 1990; Cutter and Cutter, 2004), but changes in their operations and effluent treatment changed the amount and chemical form of selenium entering the Bay. How has this affected the concentrations and speciation of dissolved selenium?
2. Similarly, selenium in suspended particles is the primary route of the element into higher organisms (see Se cycle diagram). Have the reductions in refinery Se inputs affected particulate Se in the Bay?
3. Selenium is also introduced to the Bay via the Delta, with Se concentrations in the San Joaquin River being at least 10 times higher than those in the Sacramento (Cutter and San Diego-McGlone, 1990; Cutter and Cutter, 2004). If state-mandated flows of the San Joaquin River into the Delta increase, how will this affect selenium concentrations in the Bay, and eventually in consumer organisms?
4. Do sediments in the Bay and Delta contain historical records of inputs and cycling of selenium (and other trace elements) in this ecosystem?
5. The tidal freshwaters of the Delta are an extremely dynamic ecosystem that are difficult to study because of their complexity (temporal and spatial variability). What are the processes affecting the transfer of selenium from the Delta into the Bay?

<<Top

Research Team

Old Dominion University

Gregory Cutter - Professor

Martina Doblin - Postdoctoral Researcher/Research Assistant Professor

Lynda Cutter - Lab manager/person who makes it all happen

Shannon Meseck - Ph.D. student

Susanna Mulikin - MUST undergraduate intern

United States Geological Survey, Menlo Park, California

Samuel Luoma - Senior Research Hydrologist (http://wwwrcamnl.wr.usgs.gov/tracel/people/sam.html)

Robin Stewart - Postdoctoral Researcher/Research Hydrologist (http://wwwrcamnl.wr.usgs.gov/tracel/people/robin_stewart.html)

And a host of others (see: http://wwwrcamnl.wr.usgs.gov/tracel/)

State University of New York, Stony Brook

Nicholas Fisher - Professor (http://www.msrc.sunysb.edu/people/fisher.htm)

Stephen Baines - Postdoctoral Researcher/Research Assistant Professor (http://www.msrc.sunysb.edu/people/baines.htm)

<<Top

Selected Results

Sampling the Bay and Delta:

Greg and Susanna obtaining a water sample with a 5L Go-Flo bottle.

Shannon filtering water from the Go-Flo bottles for dissolved and particulate selenium (on the right), and particulate carbon, nitrogen, sulfur, and chlorophyll (on the left).

The Delta sampling crew on board the RV David Johnson in 1999 (left to right): Martina Doblin, Greg Cutter, Robin Stewart, and Lynda Cutter.

Delta coring on the RV David Johnson in 2001; piston corer in use and box corer on the right.

Two meters of Delta history?

The aftermath of box coring; photographing a sub-core before processing (cutting into sections for storage and analyses). Porewaters are collected with another sub-core (see next picture).

Shannon uses a "whole core squeezer" (on the right) to extract sediment porewater samples for dissolved selenium analyses on board the RV David Johnson.

Dissolved selenium inputs from the Sacramento and San Joaquin Rivers :

Sixteen years of dissolved selenium data for the San Joaquin River (from Cutter and Cutter, 2004). These data show that selenate is the predominant form of dissolved selenium in the river and that the greatest temporal variability occurred in the 1980's.

Sixteen years of dissolved selenium data for the Sacramento River (from Cutter and Cutter, 2004). These data show that selenate and organic selenide are the predominant forms of dissolved selenium in the river and that there are no detectable changes in selenium behavior over this period.

Dissolved and Particulate Selenium in the Estuary :

These data (from Cutter and Cutter, 2004) show the dramatic decrease in dissolved selenium concentrations brought about by the 67% decrease in selenium discharge from oil refineries near the Carquinez Strait during the low river flow period typical of the fall (Sept-Nov). Most of the decrease is due to selenite (SeIV) which was the predominant form of dissolved selenium in refinery discharges until 1998 (Cutter, 1989; Cutter and San Diego-McGlone, 1990; Cutter and Cutter, 2004).

These data (from Doblin et al., submitted) for total particulate selenium inventory (per liter) and concentration (per gram) show that they remain relatively unchanged over the same period as that for dissolved selenium (preceding figure). In contrast, chlorophyll a concentrations dropped dramatically due to the introduction of the invasive clam Potamocorbula amurensis (refer to the USGS web site listed above).

For a detailed discussion of these and other data, you should read the Cutter and Cutter (2004) and Doblin et al. (submitted papers). In general, dissolved selenium has undergone a remarkable change in the estuary, largely due to the decreased selenium inputs from oil refineries. However, the concentration of selenium in particulate matter has not decreased equivalently, either due to the "buffering" of concentrations from sedimentary inputs or by the uptake of selenium in resident phytoplankton. The ramifications of these findings for the trophic transfer of selenium to higher organisms (e.g., from phytoplankton to clams to sturgeon) are currently being investigated. It is likely that changes in selenium inputs from the Delta could also affect the estuary and its biota, and this scenario was investigated with a simulation model of selenium in the Bay.

Modeling the Biogeochemistry of Selenium in the SF Bay:

Because of the complexity of the selenium cycle in the SF Bay Estuary (see diagram above), including multiple inputs, biotic cycling, transport processes and potential changes to these inputs (e.g., refineries; San Joaquin River), a simulation model for selenium in this ecosystem was developed by Shannon Meseck for her Ph.D. dissertation (2002). This model includes physical transport and mixing (driven by river flows and tides; the effects of wind are not simulated), suspended sediment inputs from river detritus and resuspension, primary production (light-limited), grazing (pelagic and benthic), and all of the Se processes depicted in the cycle diagram, plus refinery inputs. The model accurately reproduces all of the dissolved and particulate Se (concentration and speciation) behavior in the North Bay (Rio Vista to the Golden Gate) from 1986 to 1999 (Cutter, 1989; Cutter and Cutter, 2004) encompassing changes in river flows (e.g, El Nino years), refinery inputs, and benthic grazing with the introduction of Potamacorbula; the solid (red) lines in Figs. 1-3 are model-derived outputs for each sampling period. Because of its performance under differing climatic and biotic conditions, as well as variable point sources, the model was then used to predict changes in dissolved and particulate Se in the estuary for increased San Joaquin River flows. The scenarios included present day behavior, increasing the flow to levels measured at Vernalis (where Se monitoring occurs and before major withdrawals), but with the Delta removing a constant 60% of the dissolved Se (based on empirical data), and finally with the Vernalis flow, but turning off the Delta removal since the water residence time in the Delta will decrease dramatically. These simulations show that increasing the flow of SJ River water into the Bay increases dissolved and particulate Se for all extremes of river flow (Fig. 4 and 5). More significantly, in the fall when river flow is lowest, and freshwater residence time in the estuary is at a maximum, Se in the suspended particles themselves (per gram basis) rises above 1 µg g -1, well above typical concentrations in the Bay (Cutter, 1989; Doblin et al., submitted) and at levels where bio-accumulation studies have shown adverse effects to many consumer organisms such as the Sacramento splittail (Stewart et al., 2004).

Figure 1. North SF Bay dissolved selenium profiles, 23 Sept 1986 . Observed data and model simulations are shown.	Figure 2. North SF Bay dissolved selenium profiles, 5 Nov 1999 . Observed data and model simulations are shown. __
	Figure 3 . North SF Bay particulate selenium profiles, 12 Oct 1998 . Observed data and model simulations are shown.

	Figure 4. Simulations of dissolved and particulate Se in theNorth SF Bay in the spring and fall under 3 different SJ Riv flow and Delta removal conditions. Meseck (2002)

	Figure 5. Fall simulation of selenium concentrations in suspended particles of the North SF Bay under 3 different SJ Riv flow and Delta removal conditions. Meseck (2002)

Selenium Cycling in the Sacramento-San Joaquin River Delta:

The simulation model results demonstrate that the Delta is an important interface between selenium delivery from rivers and its behavior in the Estuary. We have been conducting studies in the Delta for the last 5 years and some of the results are described here. Examining the summary data in the table, only the concentration of selenium in the sediments is statistically different than those in the Bay (in this case Suisun Bay for comparison). However, dissolved selenium in the water column can be substantially higher than that in the Bay (see Table), likely due to inputs from the San Joaquin River (with its higher selenium concentrations; figure above). We also conducted an intensive study of selenium cycling in a flooded island habitat, Mildred Island in 2001 and these results are shown in the poster attached below.

Summary of Selenium Data for the Delta and Suisun Bay

Mildred Island Poster (PPT file) for the ASLO Meeting (MI Doblin ASLO Poster.ppt)

<<Top

Publications and Presentations

If available, links to the articles on the journal web sites are given below. If you cannot access them this way, then please email Dr. Cutter for a reprint file or hard copy.

Cutter, G.A. 1989. The estuarine behavior of selenium in San Francisco Bay. Estuarine Coastal Shelf Sci. 28: 13-34.

Cutter, G.A. and M.L.C. San Diego-McGlone. 1990. Temporal variability of selenium fluxes in the San Francisco Bay . Sci. Total Environ. 97: 235-250.

Baines, S.B., N.S. Fisher, M.A. Doblin, and G.A. Cutter. 2001. Uptake of dissolved organic selenides by marine phytoplankton. Limnol. Oceanogr., 46: 1936-1944.

Linville, R., S.N. Luoma, L.S. Cutter, and G.A. Cutter. 2002. Increased selenium threat as a result of invasion of the exotic bivalve Potamocorbula amurensis into the San Francisco Bay-Delta. Aquat. Toxicol., 57: 51-64.

Shannon L. Meseck. 2002. Ph.D. in Oceanography. Dissertation Title: Modeling the biogeochemical cycle of selenium in the San Francisco Bay.

Purkerson, D.G., M.A. Doblin, S.M. Bollens, S.N. Luoma, and G.A. Cutter. 2003. Selenium in San Francisco Bay zooplankton: possible effects of hydrodynamics and food web interactions. Estuaries, 26:956-969.

Baines, S. B., N.S. Fisher, M.A. Doblin, G.A. Cutter, L.S. Cutter, and B. Cole. 2004. Light dependence of selenium uptake by phytoplankton and implications for predicting selenium incorporation into food-webs. Limnol. Oceanogr., 49: 566-578.

Cutter, G.A. and Cutter, L.S. 2004. The biogeochemistry of selenium in the San Francisco Bay estuary: changes in water column behavior. Estuarine Coastal Shelf Sci. 61: 463-476.

Doblin, M.A., Baines, S., Cutter, L.S., and Cutter, G.A. 2005. The biogeochemistry of selenium in San Francisco Bay estuary: seston and phytoplankton. Submitted to Estuarine, Coastal and Shelf Science.

Meseck, S.C. and G.A. Cutter. Selenium in sediments of the San Francisco Bay. Submitted to Estuarine, Coastal and Shelf Science.

<<Top