This 18-month field study addresses the seasonal and spatial patterns and processes controlling methylmercury (MeHg) production, bioaccumulation, and export from natural and agricultural wetlands of the Yolo Bypass Wildlife Area (YBWA). The data were collected in conjuntion with a Proposition 40 grant from the State Water Resources Control Board in support of the development of Best Management Practices (BMP's) for reducing MeHg loading from agricultural lands in the wetland-dominated Yolo Bypass to the Sacramento-San Joaquin River Delta. The four managemenr-based questions addressed in this study were:

Measurements of MeHg concentrations in sediment, water, and biota (plants, invertebrates, and fish) were made to assess management-level patterns in five wetland types, which included three type of shallowly-flooded agricultural wetlands (white rice, wild rice, and fallow) and two types of managed wetlands (permanently and seasonally flooded). To strengthen our understanding of the processes underlying the seasonal and spatial patterns of MeHg cycling, additional exploratory factors were measured including ancillary sediment and water quality parameters, stable isotope fractionation (oxygen, sulfur, carbon, and nitrogen), photodemethylation rates, and daily-integrated hydrologic budgets. Samples and field data were collected from May 2007 to July 2008, and nearly all sample analyses were completed by September 2008 as per the Quality Assurance Program Plan (QAPP) requirements.

Although wetland type was a major factor that drove the study design, within-field hydrology also proved to be an important factor controlling aqueous MeHg and total mercury (THg) concentrations and export. Overall, agricultural wetlands exhibited higher MeHg concentrations in overlying water, sediment, and biota than did managed seasonal and permanent wetlands. This appears to be partly due to higher rates of sediment in microbial production of MeHg on agricultural wetlands during the fall through spring period. Both sulfate- and iron-reducing bacteria have been implicated in the MeHg production process, and both were demonstrably active in all wetlands studied; however, sulfate-reducing bacteria were not stimulated by the addition of sulfate-based fertilizer to agricultural wetlands, suggesting that easily-degraded (labile) organic matter, rather than sulfate, was limiting their activity in these field types. The data suggest that agriculturally-managed soils promoted MeHg production through 1) enhanced microbial activity via higher temperatures and larger pools of labile carbon, and 2) enhanced pools of microbially available inorganic divalent mercury (Hg(II)) resulting from a decrease in reduced-sulfur, solid-phase minerals under oxic or only mildly reducing conditions.

MeHg mass balances were assessed by comparing filed-specific MeHg loads for inlets vs. outlet flows. The overall mass balance for MeHg in surface water during the summer irrigation period (June - September 2007) indicated little to no net MeHg export from the six agricultural wetlands taken as a whole. Of the six agricultural wetlands, there was net overall MeHg export from two fields (one fallow and one white rice) during August, and from four of the six fields (one fallow, one white rice, and two wild rice) during September) Over the entire summer irrigation period, two of the fields (one fallow and one wild rive) showed net MeHg export, and the other four fields showed wither net import or no significant change. Rates of measured photomethylation and exchange between sediment and water pools suggest that both processes may be responsible for the lack of MeHg export. Despite significant differences during winter months between fields in surface water concentrations of MeHg, MeHg loads were not calculated in mid-winter because flood waters had overtopped field boundaries and field fidelity could not be established.

During the summer 2007 irrigation season, surface water out-flows from agricultural wetlands were 9%-36% of inlet flows, and evaporation rates explained most of this water loss, with infiltration likely accounting for the remainder. Unfiltered aqueous MeHg concentrations increased from <1 ng L^-1 in source waters to up to 10 ng L^-1 in agricultural wetland drains during the summer irrigation period. Increases in solute concentration caused by evapoconcentration were estimated by determining concentration factors (outflow/inflow) for chloride (a conservative dissolved constituent) and by measuring oxygen isotope ratios (¹⁸O/¹⁶O, expressed as δ¹⁸O) in water. Increases in MeHg concentration from inflows-to-outflows exceeded those caused by evapoconcentration on several fields during the summer irrigation season. This was especially true when initial surface water MeHg concentrations were low, as seen in the southern block of fields receiving irrigation water directly from the Toe Drain. The northern block of fields received irrigation water from Greens Lake, which included Toe Drain water plus recirculated drain water from other agricultural fields within the Yolo Bypass and west of the Yolo Bypass; as such, the northern fields showed a smaller percentage increase in MeHg concentration because initial MeHg concentrations in surface water inflows were greater than in inputs to the southern fields.

Mercury concentrations in fish were greater in agricultural wetlands white rice and wild rice) than in the two permanently flooded wetlands. Additionally, Hg concentrations in biota showed a general increase from inlets to outlets withing agricultural wetlands, but not within permanent wetlands. This was particular evident in white rice fields where caged western mosquitofish at the outlets had Hg concentrations that were more than 4 times higher than in caged fish held at the inlets. Similar spatial patterns in Hg bioaccumulation in agricultural and permanent wetlands were seen for wild populations of western mosquitofish and Mississippi silversides. In contrast to fish, invertebrates, such as water-boatman (Corixidae) and back swimmers (Notonectidae), had greater Hg concentrations in permanent wetlands than in tempoarirly flooded agricultural wetlands, Fish THg concentrations were weakly correlated with water MeHg,a and not correlated with sediment MeHg. In contrast, invertebrate MeHg concentrations were more strongly correlated with sediment MeHg than with water MeHg concentrations. These results illustrate the complexity of MeHg bioaccumulation through food webs and indicate the importance of simultaneously using multiple biosentinels when monitoring MeHg production and bioaccumulation.

Despite high sediment production rates and water concentrations in agricultural wetlands, MeHg export was physically limited by hydrologic export for all wetlands studied. We suggest that load reduction is maximized by limiting water throughout, but that on-site biota exposure is maximized by this loner water residence time. While field-specific hydrologic loads could not be fully quantified during flood conditions in February 2008, we suggest that the primary period of MeHg export from Yolo Bypass Wildlife Area is during those winter flooding periods when overall microbial activity and MeHg production in agricultural soils is fueled by the decomposition of rice straw, and when hydrologic flowthrough is maximal.

Local stakeholders participated in two workshops related to this study, demonstrating an interest in understanding factors controlling MeHg production, export, and bioaccumulation. The results of this field study show that permanently flooded, naturally vegetated wetlands are unlikely to a large source of MeHg production within the YBWA, in contrast with agriculturally-managed wetlands. MeHg loading to Toe Drain waters of the Yolo Bypass may be reduced by lowering rated of hydrologic export from agricultural wetlands during the growing season and especially during rice harvest, However, under these water-holding conditions, biota living within agricultural wetlands may thus be exposed to higher MeHg concentrations in surface water, As observed in this study, rapid bioacculumaltion over a 2-month period led to MeHg concentrations in invertebrates and fish more than 6 and 11 times higher, respectively, than proposed TMDL target values to protect wildlife (0.03 ppm ww).

The results of this field study, together with the information from YBWA stakeholders, provide a more definitive understanding of how MeHg cycling and bioaccumulation respond to habitat differences and specific management practices. These results directly address 4 core components of CBDA's Mercury Strategy for the Bay-Delta Ecosystem (Wiener et al., 2003a):

a) Quantification and evaluation of THg and MeHg sources,

b) Quantification of effects of ecosystem restoration on MeHg exposure,

c) Assessment of ecological risk, and

d) Identification and testing of potential management approaches for reducing MeHg contamination.

In addition, the quantitative results reported here assess the effect of current land use practices in the Yolo Bypass MeHg production, bioaccumulation and export, and provide process-based advice towards achieving current goals of the RWQCB-CVR's Sacramento -- San Joaquin Delta Estuary TMDL for Methyl & Total Mercury (Wood et al., 2010b). Further work is necessary to evaluate biotic exposure in the Yolo Bypass Wildlife Area at higher trophic levels (e.g. birds), to quantify winter hydrologic flux of MeHg to the larger Delta ecosystem, and to evaluate rice straw management options to limit labile carbon supplies to surface sediment during winter months.

In summary, agricultural management of rice fields -- specifically the periodic flooding and production of easily degraded organic matter -- promotes the production of MeHg beyond rates seen in naturally vegetated wetlands, whether seasonally or permanently flooded., The exported load from MeHg from these agricultural wetlands may be controlled by limiting hydrologic export from fields to enhance on-site MeHg removal processes, but the tradeoff is that this impoundement increases Me Hg exposure to resident organisms.