Legacy mercury contamination is problematic in California waterways due to historic gold and mercury mining. Even today, mercury from abandoned mines still leaches into the creeks and rivers in the Sierra and Coast Ranges, flowing downstream and into the Delta, contaminating the sediments.
With tens of thousands of acres of tidal wetlands planned for the Delta, the current assumption is that tidal wetlands are net exporters of methylmercury, which can result in expensive and time-consuming mitigation. However, studies undertaken by the Department of Water Resources have found that tidal wetlands are unlikely to be significant net exporters of methylmercury to receiving waters, although methylmercury concentrations of biota within wetlands still need to be considered.
At the 2021 Bay-Delta Science Conference, Petra Lee, a senior environmental scientist with the California Department of Water Resources, discussed the study results.
During the Gold Rush, mercury was mined in the Coast Range and brought to the gold mines in the Sierra and elsewhere in California where the mercury was mixed into the soil and the sediments to amalgamate the gold.
Today, California has about 47,000 abandoned gold and mercury mines; the red dots show the mercury mines; the gold mines are shown in gold. Many of these mines are still discharging mercury, which is considered a legacy contaminant. The mercury leaching from the mines is inorganic; however, bacteria present in the sediments of the Delta convert it to the more toxic form of methylmercury, which is capable of crossing the blood-brain barrier.
With tens of thousands of acres of restoration of tidal marshes and wetlands planned within the Delta, there is the concern of increased methylmercury production. The presence of inorganic mercury, along with plenty of organic carbon, and low oxygen in tidal marshes and wetlands provide nearly ideal conditions for making methylmercury. However, methylmercury is produced only when sediments dry out and then rewetted, and most tidal marshes don’t get very dry.
So are tidal wetlands sources of methylmercury? In other words, if we restore more tidal wetlands and marsh, will they be exporting methylmercury and increasing the toxicity?
The study had several hypotheses; the two that Ms. Lee will cover in this presentation:
Tidal wetlands are a net source of total methylmercury on an annual basis; and
Tidal wetlands are a net source of total mercury on an annual basis.
Four different wetlands were each studied for one year to test the hypothesis. The studies were done back to back, but it did take more than four years.
Ms. Lee explained why: “Finding wetlands was hard because we had to get a good accurate water flow measurement, which is really tricky,” she said. “We’re measuring the water entering and leaving each wetland. This meant that the wetland couldn’t be hydrodynamically leaky and could only have one or two breeches. So we put the equipment in those breeches so that we could measure the flow, take samples and readings, and capture the potential of exporting methyl mercury.”
There were other challenges as well. “Getting a good water balance in a tidal system is really tough; the water is sloshing back and forth, and we’re trying to see this tiny signal amongst that sloshing,” she continued. “We also hit all of the snags: equipment failure, weather patterns (do you remember that first part of 2017 with an epic storm and epic flooding?), and user error, but we did triumph in the end. We measured flow continuously as well as some basic water quality measurements like turbidity. Most importantly though, we took the mercury and methylmercury measurements about once a month.”
To determine whether the wetland was exporting methylmercury, flow and concentration were measured and used to calculate the mass or “load.” The flow is measured in cubic feet per second, a very large number, whereas the concentration is a very small number – nanograms per liter, equivalent to a full Lake Oroville with half a cup to a gallon of mercury. The load is measured in grams per tide.
The components measured:
Continuous temperature, dissolved oxygen, turbidity, specific conductance, total chlorophyll
Total mercury, methylmercury, organic carbon, total suspended solids monthly (approximately)
Ms. Lee pointed out that the sampling was very expensive and very difficult, which is why they sampled on a monthly basis.
“To reiterate, we’re measuring the amount from the tidal wetland and looking at whether the wetland was a net exporter and source of methylmercury or total mercury and whether the wetland was a net importer or a sink,” she said. “And to complicate it, the wetland is tidal, so we’re again measuring that water going back and forth and trying to see a net load of mercury in those large signals.”
Results and conclusions
Ms. Lee then gave the results and conclusions.
The first hypothesis was that tidal wetlands are a net source of total methylmercury on an annual basis.
“We looked at a few different lines of evidence,” she said. “First, I calculated that loads for each tidal cycle we sampled, and then I calculated monthly load estimates. Lastly, I looked at whether the concentration of mercury was significantly higher leaving the wetland than entering. Tricky considering the data is time series, but we figured it out.”
She then presented the results for methylmercury and mercury loads, noting that for this presentation, she won’t be talking about the concentrations, but those results did reflect the conclusions made from the load data.
The graph on the slide below shows the methylmercury load per 25-hour tidal cycle, one for each of the four wetlands that were studied. Above the x-axis, the wetland is a source and exporting methylmercury; below the x-axis, the wetland is a sink or importing methylmercury. The orange portion of the bar is the particulate fraction, and the blue is the filtered fraction.
“You can see that there isn’t a clear pattern,” said Ms. Lee. “There are spikes here and there; often those spikes coincided with when a couple of the wetlands were acting as floodplains rather than tidal wetlands.”
The graph below shows the estimated methylmercury load per month using the continuous flow data and approximately monthly mercury data. These graphs represent a monthly estimate rather than one tidal cycle shown on the slide above, although the tidal cycle data was used to estimate the monthly mercury data. Above the x-axis, the wetland is a source of methylmercury; below the x-axis, it’s a sink or importing. The orange portion of the bar is the particulate fraction, and the blue is the filtered fraction.
“As you can see, the results somewhat mixed,” said Ms. Lee. “There is a little bit more pattern, and sometimes the filtered fraction was a source or sink while the particulate fraction of the same month was the opposite.”
So what conclusion was the conclusion? “None of these four tidal wetlands nor either the tidal cycle loads nor the estimated monthly loads showed that any of these wetlands were a significant net annual source of methylmercury,” she said. “I say that very specifically because you can see some of the tidal events and some of the monthly estimates did show the tidal wetlands to be sources at times. But overall, for the net annual loads, none of them are sources exporting significant amounts of methylmercury. In fact, a couple of the wetlands were net sinks for methylmercury, which was kind of shocking.”
The second hypothesis is that tidal wetlands are a net source of total mercury on an annual basis.
The graphs below show the total mercury per event for the 25-hour tidal cycle; above the x-axis means the wetland is a source and exporting total mercury; below the x-axis, the wetland is importing mercury and is a sink. The orange portion of the bar is the particulate fraction, and the blue is the filtered fraction.
“The total mercury, which is mostly inorganic mercury, is predominantly in the particulate fraction,” said Ms. Lee. “That was much more the case with the total mercury than the methylmercury. And most of the mercury was found on the particles. And that’s a whole different talk, but it just seemed worth mentioning here.”
The graphs below show the estimated load of total mercury per month. Again, above the x-axis means the wetland is a source; below, it’s a sink. The orange is the particulate fraction; the blue is the filtered fraction.
“This data gave us a much more mixed picture about what total mercury imports and exports are,” said Ms. Lee. “Unlike the methyl mercury data, it didn’t show that the wetlands were all either that sources or sinks, and some of the wetlands actually were a source of total mercury. Two of those wetlands were a source of a fraction of total mercury, sometimes in the particulate form, sometimes in the filtered, so sometimes this hypothesis was true, and that mercury was mostly in the particulate fraction.”
“Obviously, there’s a lot of nuance and other data that I wasn’t able to cover, but the long and the short of it is that these four tidal wetlands are not net sources of methyl mercury,” she said. “Sometimes they were net sources of total mercury, and mostly in the particulate branch.”
STUDY: Mercury Imports and Exports of FourTidal Wetlands in the Sacramento-SanJoaquin Delta, Yolo Bypass, andSuisun Marsh for Delta MercuryControl Program Compliance
In 2010, the Central Valley Regional Water Quality Control Board (Regional Board) approved theSacramento-San Joaquin Delta Methylmercury (MeHg) Total Maximum Daily Load (TMDL) andBasin Plan Amendment (BPA) which established a Delta Mercury Control Program (DMCP) for the Sacramento-San Joaquin Delta (Delta) and Yolo Bypass. The California Department of Water Resources (DWR) was named as a discharger and wetlands were given a discharge allocation.
To calculate discharge allocations for wetlands, the TMDL/BPA relied on a small amount of existing MeHg wetland data, and dischargers such as DWR were called upon to develop management practices to decrease the amount of MeHg being discharged by tidal wetlands.
However, the data underlying the wetland allocations were not representative of tidalwetlands, partly because not enough characterization data existed; not enough tidal wetlandrestoration projects had been completed to characterize. Because of the lack ofcharacterization data, any management practices built upon the existing data were unlikely tobe effective. Consequently, DWR chose to characterize four tidal wetlands to determinewhether tidal wetlands were importing or exporting MeHg and by what mechanisms.
To characterize MeHg, DWR studied imports and exports of MeHg, total mercury (THg), andorganic carbon at four tidal wetlands: the Yolo Wildlife Area Tidal Wetland (Yolo) in the YoloBypass, Blacklock Tidal Wetland (Blacklock) in Suisun Marsh, North Lindsey Slough TidalWetland (North Lindsey) in the Cache Slough Complex, and the Westervelt Cosumnes RiverTidal Wetland (Westervelt) east of the confluence of the Cosumnes and Mokelumne Rivers.
DWR collected the concentration data from approximately monthly tidal events and combinedthat with flow data to calculate loads, which are masses of mercury; loads were calculated per25-hour tide as well as estimated per month. DWR also compared mercury ebb and floodconcentrations to determine if there was a significant difference.
Based on the collected data and analyses, none of the four wetlands appear to be a significantsource of MeHg to their adjacent waterbodies, nor are concentrations of MeHg significantlyhigher leaving the wetland than entering the wetland. Generally, the waters entering thewetlands are not meeting the Regional Board’s DMCP water quality criterion of 0.06ng/L, andthere does not seem to be a measurable annual increase in MeHg loads in receiving waters dueto the tidal wetlands.
While the four tidal wetlands do not appear to be a source of MeHg annually, Westervelt andNorth Lindsey appear to be a source of THg. Westervelt is a source of particulate THg, althoughbecause of how few times DWR was able to sample mercury, more data would beadvantageous to determine any patterns. Blacklock and Yolo appear to be sinks of THg,predominantly in the particulate form.
DWR did not see any strong mercury seasonal patterns, possibly because of the small data sets.Additionally, the majority of mercury was in the particulate fraction for most of the wetlands, except for North Lindsey which had a higher median percent of filtered MeHg than particulateMeHg.
Several future directions for studies present themselves: better resolution of the data toinclude more frequent sampling, a larger number of wetlands, and a longer period of timewould be incredibly beneficial to see if estimates are representative. This study did not look atthe toxicity of mercury in the biota, it only looked at water concentrations. Delving into thetoxicity and measuring concentrations of MeHg in native organisms is necessary to determineany direct effects of mercury on those organisms, and organisms that prey on them. Lastly, thisdataset provides a base for modeling tidal wetland mercury dynamics, which could help improve knowledge of mercury cycling.