Mercury and rice in the Delta: Lessons linking wetlands to water and wildlife

Dr. Lisamarie Windham-Myers is an ecosystem ecologist specializing in wetlands.  As an ecosystems ecologist, she works across a broad spectrum of wetland types on issues related to carbon, nutrient, and mercury cycling.  While her focus on mercury began in New Jersey meadowlands, Lisa is an active researcher in the biogeochemistry on all U.S. coasts, and for the past 12 years, she has played a leading role in providing science advice to agencies seeking to reduce methylmercury exposure in the San Francisco Bay.  In this presentation, she discusses mercury and rice in the California Delta.

I am going to tell you a story about the interaction of these concepts –mercury and rice, starting with the more global context for mercury and global context for rice, bring it down to a very specific study we did in the Yolo bypass, which involved a whole lot of interacting processes,” began Dr. Lisamarie Windham-Myers.  “Then take it to what that means for the California Delta as a whole.”


When we talk about mercury, most people think of it as a local San Francisco Bay story because we have a lot of contamination, but it’s really a global pollutant,” she said.  In 2013, 128 signatories signed the Minamata Convention to reduce their emissions of mercury.  While mercury has never been shown to have a biological purpose, it’s used in all sorts of things, such as light bulbs, switches, batteries, and thermometers.

Dr. Windham-Myers said that because mercury has a really low boiling point, it can be in a vapor form in the environment; that is why emissions are a concern.  Emissions turn mercury into a nonpoint source.

When mercury is emitted, it can take one of three forms:

Hg-zero, which is elemental mercury; this form is fairly inert.  It goes up into the atmosphere, stays there for a while, and makes the way up to the poles – that is how it is getting into the polar bears.  This form is capable of long-distances and being transported globally.

Mercury 2-plus, or divalent (sometimes called reactive mercury) is basically a sticky gas.   When mercury gets oxidized, it can glom onto raindrops, particles, or aerosols and it can deposit out from the atmosphere, so it has a less long-distance transport.

Particulate mercury is released in smoke, such as from a coal-burning power plant.  It has a short transport; it’s basically gravity.  It’s going to move until it falls out of the sky.

Currently, coal is the second most important source of mercury emissions, and that is mostly coming from China.  But the biggest source of mercury emissions is actually artisanal gold mining.  “We discovered this basically through remote sensing,” Ms. Windham-Myers said.  “Whenever the price of gold goes up, people go out into the jungles of the Amazon, into Africa, or wherever, and start just trying to pick up flecks of gold.  They bring mercury with them.  So we can see that basically whenever the price of gold grows, we see more and more artisanal mining.  And the current rates of emissions are about 700 metric tons per year through that process.”

Dr. Windham-Myers noted that Peru has declared a state of emergency in 11 jungles for mercury contamination to people there.  “So it’s a definite health hazard as well as just entering the global system.”

But before we wag our finger at those artisanal gold miners, that was us 150 years ago, she pointed out.  “We did the artisanal mining up in the Sierras.  And the reason we were so successful in the Gold Rush is we had really quick, cheap proximity to mercury, because the mercury was here in the Coast Range.  Right under our feet are mercury ores that can be used to get mercury out of and then carry that mercury up into the gold-bearing sediments of the Sierra.”

The mercury ore in the Coast Range is mostly cinnabar, or mercuric sulfide.  It is really stable.  In order to get the mercury out of the ore, you have to roast it a high temperature to volatize the mercury out of the ore.  Once it has been volatized, it can be condensed into quicksilver or elemental mercury, bottled and taken wherever it needs to go.

The amount of mercury that came out of New Almaden mine alone was huge.  “From this one place alone, 7,000 metric tons were released in a decade at the height of the mining period,” Dr. Windham-Myers said.  “That’s metric tons – it’s the current emissions of the world right now globally.  By around 1920 or so, not so much.  Most of it had been used and then already brought up to the mountains by that point.  That said, 99% of what was brought up there is still up there.  It hasn’t all run out. It is still up there.”

Even today, if you go out with a pan or dig around in some rocks or sand, you can still find blobs of elemental mercury out there; she said it’s not that hard to find.  “The concern is that what happened in the Sierra is not staying in the Sierra.  It is coming out in a pulsed fashion, but eventually it’s all making its way down into San Francisco Bay.  We have the gold mining-based mercury coming out from the Sierra, and then from the Coast Range, we have this roasting leftover mercury coming out through, for example, Guadalupe River there.  And so the whole bay is basically pickled in mercury.  It’s just out there in the sediments.  There’s just nothing we can really do about it.”

We actually use a term, ‘bay clean,’” she said.  “It basically means sediments that are sort of the base level – anything less than about 400 nanograms per gram.  That’s really high. But that’s our base level.  And it’s just the way it is.”

It’s going to take a long time – far longer than 100 years – for it all to get flushed out to the ocean, and it would be very disruptive to try to dig it out, so it’s basically there, she added.

Mercury levels in fish are also high; the regional monitoring programs measure mercury levels in striped bass, about a foot long.  “When you look at the concentrations, they tend to be high in many places, but particularly sort of in the edgy type environments of the Bay Delta.  The value of 0.43 is actually really close to a can of white albacore tuna.  The FDA recommends that for pregnant women and children, just one can a week.  We don’t even have numbers for the monster fish, which is really what everybody’s out there to catch.  And the bigger the fish, the more biomagnifications; the more concentration, the older the fish, of course.  And there just is more mercury in it.  So we do have an issue with mercury in our fish.”

The point here is that the Bay Delta is higher than everybody else.  “If you go to other estuaries in the country, they tend to be lower,” said Dr. Windham-Myers.  “Any place that’s been monitoring striped bass will basically have lower average mercury concentrations than we do.  There are populations of folks in the Delta that basically have concerning concentrations in their hair.  So we know there’s a local population concern, and that’s why you see these signs everywhere in the Bay-Delta telling you what fish you might want to eat and how many times a week and maybe what kind of fish you might want to avoid.  But this puts a real wet blanket on the fishing and tourism industry for the delta. It’s a $2 billion industry, and it’d be great if we didn’t have this problem.”

The other big industry in the Delta is farming.  At the same time the Gold Rush was putting all the mercury out into the environment, they were also reclaiming the wetlands of the Delta for farming by building levees. “It used to be one huge freshwater wetland; now the majority of the Delta lands is actually farmed now.  And we have very channelized rivers that – or channels, I should say, in the Delta that are great for boating.”

It is great farmland, and the farming industry is huge – it’s about $27 billion for California, she said.  “A hundred years ago, all of that red would have been wetland.  Just for clarity, it’s a really large amount of wetlands that we have lost. So these delta levees have just changed the whole function and appearance of the Delta.”

The Delta is a lucrative part of the state; there are signature crops in the Delta like asparagus and pears.  Increasingly, rice is starting to be grown in the Delta, she said.  “People like to put rice out there because it helps prevent peat oxidation,” she said.  “If you put rice out there, you can keep the water levels higher, and then you don’t have as much of the loss of those peaty soils.  They’re not blowing off.  So there’s an interest in putting rice out there.”


Rice is an important crop in California; the state is the fifth-largest producer in the world, and more than half of that is exported.  It is high-quality rice and draws top dollar.  A lot of it is shipped to Japan.  “So the big barges come in with Sony TVs into Sacramento, dump their load off, and go back with rice,” she said.  “It’s pretty common.”

Most of the rice is grown in the Sacramento Valley, with some coming into the Delta.  “The most important feature of rice for me is its role as a surrogate wetland,” said Dr. Windham-Myers.  “Since we lost so many wetlands – today, for every one acre of natural wetland, we have two acres of rice.  So rice is by far the most common wetland type in California.”

During droughts when natural wetlands are getting wet, water is still being put on the rice fields, and so that’s where the wildlife goes.  “For the last two years, there’s been some really good documentation by the Nature Conservancy about where the birds are going, and many of them are going to these rice fields on the Sacramento River because that is where the water is, that is where the wetland is.  So they play this really important role in California as a wetland surrogate, and they are effectively an agricultural wetland.

From a global perspective, rice fields are the most abundant type of wetland in temperate and tropical latitudes.  All rice fields have Oryza sativa; for the last 4,000 years, they have bred Oryza sativa to grow white rice, black rice, red rice, long-grain, aromatic, short-grain, medium, or otherwise.

If you think about the importance of that, the world is dominated by this one plant; these wetlands are dominated by this one plant,” she said.  “It’s a really productive plant.  It’s incredible breeding that’s gone on to make this plant. Half of the biomass in these plants is edible. It’s the grain; the grain is very weighty.  So it actually makes sense to do things like these crazy terraces on these steep slopes because you get so much food out of a very small area.”

In California, rice is harvested in October; the Mississippi Valley harvests around September.  In the U.S., there is just one crop a year.  But in tropical wetlands, they can have up to three crops per year; it only takes about 30 to 60 days to grow a crop; then they pull it out and then put the next one in.  So a lot of rice is grown around the world, and it’s a very important food source.

The benefit of rice and wetlands is of course the food supply, she said.  “We have a growing population.  This is a really excellent plant for making lots of food fast.  However, the negative side are public health issues, such as mosquitoes.  Also rice is currently responsible for 11% of the global methane emissions – a huge source.  And thirdly is methylmercury; people are concerned about it, because when you look for it, you tend to find it.”

Wetlands make methylmercury; this has been demonstrated over and over, Dr. Windham-Myers said.  “At the micro scale, the bacteria that actually do the methylation of mercury tend to be in wetland soils – anyplace that has lots of organic matter and oxic conditions.  At the macro scale, the best predictor of methylmercury concentrations in a watershed’s surface water is the percent of wetlands that is in that watershed.  Dissolved organic carbon is a good predictor of how much methylmercury is in the water, but that’s also a wetland product.  So wetlands play an important role.”

She presented a diagram of how methylmercury is produced in a wetland to emphasize why, of all anoxic soils, wetlands are so good.  “It’s because they have this sort of sub-oxic zone; it’s not too oxygenated, it’s not too anoxic or anaerobic – it’s just sort of right in the middle,” she said.  “You have the right conditions for active bacteria, but the mercury itself is not glommed onto a sulfide molecule.  If you think about mercuric ore, that’s a strong a bond – mercury and sulfide.  The more sulfide you have produced, the less mercury is available.  So this sort of sub-oxic zone where sulfide is low but you still have the active bacteria, that’s the good thing.”

We’ve learned that roots of wetland plants are really good at kind of building up that sub-oxic zone.  They’re leaking out oxygen. They’re also leaking out carbon.  So this is a really good hotspot for methylmercury to be formed.”

For methylmercury to be a problem, it’s not just the production of it, but it leaves the soil, enters the water column, and ultimately the food chain.  Methylmercury is bioaccumulative pollutant which concentrates as it moves up the food chain, from algae to zooplankton to prey fish and the predators that eat them, such as trout and bass.  “So the mercury just keeps getting more and more locked into tissues as it moves up the food chain, and it’s hard to get rid of once you have it in your system,” she said.

What’s interesting about rice is that it doesn’t need the transport to get to bioaccumulation; the transport just happens in the plant itself,” said Dr. Windham-Myers.  “It’s really good at moving methylmercury up its stems and just right into the rice grain itself.  People have been concerned about arsenic for a while in rice, and that tends to be in the brown part – the rice bran – the part that goes around the rice – what you call brown rice.  You can actually polish that away, and then it’s free of arsenic or, you know, somewhat cleaned. If you polish the grains to make white rice, you don’t get rid of the methylmercury, though, because it’s in that endosperm.”

This is becoming in issue, particularly in China, because they grow a lot of rice as well as do a lot of coal combustion and other mining practices, and rice is a main staple of the country’s diet.  The dominant way that some populations in China are picking up methylmercury is through their rice consumption, not their fish consumption, she said.

Most of the research is being done in China as they are trying to come up with a solution.  One potential solution is genetic varietals that don’t move the methylmercury so quickly; another is the idea of aerobic rice cropping or basically keeping the soil not so wet so not to stimulate methylmercury production and that it doesn’t move up into the plant grain itself.

The rice fields in California have been studied for awhile.  Dr. Windham-Myers presented a chart showing methylmercury production produced with data from Mark Marvin-DiPasquale.  “There’s over a thousand data points in here just summarized into a very, very simple diagram which has basically methylmercury production on the bottom and percent methyl in the soil on the Y axis,” she said, noting that the graph is using logarithmic scales.  “The key factor here is that wetlands, shown in the green, are really the sites that are making methylmercury; they are where the high production is happening.  It’s not necessarily happening in the bay sediments or sloughs or lotic – meaning flowing – streams.  It’s really happening in these wetland soils.”

Looking across the systems, the big driver tends to be the abundance of organic matter and extreme wetting and drying.   “It’s not a tidal system where it’s a little wet, a little dry, a little wet, but it doesn’t really ever really dry out.  It’s the sites that really dry out – those are the ones that seem to be making the most methylmercury.  It might be related to resetting that sulfate iron clock – getting those to the right form to support the sulfate and iron-reducing bacteria.”

These wetlands are mostly around the edges of the Delta like the Yolo Bypass, Cosumnes River, and around the southern Delta – but not in the Central Delta.  “There’s a pattern that’s just been very persistent through time in fish,” she said.  “It’s called the hole in the doughnut.  Not doughnut hole – that’s the part that comes out – but the hole in the doughnut.  And if you were going to go fishing, this is where you want to catch your fish because the fish tend to be cleaner right in that central zone.


Ms. Windham-Myers presented data for silversides, explaining that silversides are often used as a sentinel species because they don’t range very far; they have high site fidelity.  “The fish concentrations tend to be low in the Central Delta for this sentinel species and then higher in these outer areas.”

One might think the wetlands are responsible because that’s where the methylmercury is being made, but that is only one of four possible explanations.  The second possibility is simply proximity to mining, and more mercury from the tributaries coming in from the Sierra and the Coast Range then the San Joaquin.  The third possibility is that the water comes in through these tributaries already has methylmercury and by the time it gets into the Delta and the more open-water type environments, the particles settle out and the mercury drops to the bottom, or it gets degraded by exposure to sunlight.  The fourth possible explanation is just pure hydrodynamics; that the Central Delta is a really well-flushed system.  The water comes in, and it gets cleaned out through the tidal system.

Dr. Windham-Myers then focused on methylmercury production issue and whether that’s a possible explanation for these high concentrations.  “We’ve been specifically working with the State Water Resources Control Board to get a handle on if there are any management practices that could be done to reduce the methylmercury concentration in the waters coming off this land.”

The State Water Board is trying to implement the first-ever total maximum daily load (TMDL), or regulation for export for methylmercury. “Nobody’s ever tried this before,” she said.  “They’ve done it for total mercury, but this is a species of mercury.  It’s really hard to measure, it’s expensive, and it’s always changing so it’s a very tough thing to pull off.  But their argument is there’s so much mercury out there, we really want to figure the methylmercury load because that’s the part we’re concerned about.”

The TMDL is trying to get from an average of 0.31 nanograms per liter in the Yolo Bypass down to 0.06, an 80% reduction.  “I’m not sure if we’re going to get there, but that’s the goal,” she said.

The Yolo Bypass is a stunning place along Highway 80 with a lot of migratory waterfowl and a wide variety of wetland types: seasonal and permanent wetlands, as well as agricultural wetlands, including rice, wild rice, and fallowed land.   “We looked at the wetting during the summer and then the dry-down and then the wetting during the winter, and then again the dry-down; we looked at basically all these processes at those different time periods over a whole year,” Dr. Windham-Myers said.  “We also looked at a permanent wetland. It was flooded all year.  And we then looked at a seasonal wetland that was only flooded half a year.”

Several agencies and organizations have been participating in the studies, which looked at many different components that can roughly be categorized into sediment, water column, plants, and biota.  “We were looking at processes,” she said.  “We wanted to understand how the mercury was moving through these different pools and how that changed over the course of the year.”

It was a million-dollar project with 12 PIs with the goal to link the processes in a box model and figure how these systems work.  They measured the methylmercury coming into the system and the methylmercury leaving the system; the difference between those would be net export.  “We measured that all the time because we just wanted to get a sense of how things were moving.  But we also measured all of these internal processes of creation of methylmercury, degradation of methylmercury, moving into the soil, moving into the water, moving into the plants, moving into the fish – the whole – the whole Full Monty here.”

Dr. Windham-Myers then summarized the results of the study.  “The probably most important thing was timing,” she said.  “We found that all of those processes were related if you step back and looked at how they were, but it was all lagged. It didn’t always happen at the same time.  Methylmercury was produced all year long; it didn’t really matter if it was the summer or the winter.  It was just being produced whenever there was water on the land.  However, export was really low in the summer.  It just was not leaving the system.”

Then lastly, biotic uptake was high in the summer. That’s when we saw the fish really picking up lots of methylmercury on the fields. That study was done with caged fish: we locked them in place and saw how much methylmercury they picked up.  It was a completely opposite pattern in winter where export was high, and then the fish uptake was low.  This is the type of patterns that people don’t think about.  They’re important to choosing when you’re going to monitor a site to understand actually how the system is working.”

They also learned that hydrology was really important.  “The timing in flooding led to hot moments we would get sometimes.  You put the water on; you get a big spike of methylmercury coming out of the soil.  If you know when it’s going on and off, you can know when you should be monitoring it to understand what the net export is going to be.”

Also important was the source water.  “If there was methylmercury in the water already going onto the rice fields, in some cases, it actually cleaned up the water; they provided some temporary sinks for that methylmercury, so it went out a little bit cleaner than it came in,” she said.  “When the water came in on the cleaner side, it could leave dirtier because it had more to pick up along the way. So the type of water that went in sort of drove what was going on.”

Finally, it really stood out that the plants were not just passive players in these fields.  “They are really dense; they are a huge physical presence out there,” she said.  “They were controlling the hydrology – flow paths, movement through the plants through transpiration.  They were affecting carbon availability and therefore the production of methylmercury.  And then they were also shading the landscape.  We had thought in summer we’d see a lot of photodegradation – a lot of sunlight hitting the water and then it would break up the methylmercury – it should be great.  But the plants basically shaded it by about 90%, so there was just very little sunlight actually reaching the water.”

Dr. Windham-Myers presented a graph showing a year of measurements of the level of unfiltered methylmercury in the water in terms of nanograms per liter.  “Every sample – it did not matter if it was the inflowing sample or the outflowing sample or a sample from the middle of the field.  They were all high. So everything was above the target goal of 0.06 nanograms per liter.  This is a really tough goal to try to get down to 0.06 when your incoming water isn’t even there.”

She explained that triangles are the inlets, circles are the outflows; she will just be focusing on the outflow data because it’s the most important point to make; the blue line is the hydroperoid: wetted, dried, wetted, dried.  “When you look at the patterns that we saw, the first one that probably leaps out at you are these big screamers here for the wild rice,” she said.  “There were high concentrations in the output water of the wild rice fields.  Those were the highest I think anybody had measured in the Delta, period.  It seems to be related to the fact that they wet harvest.  They beat on the wild rice, and then it floats, kind of like cranberries.  Then they collect all the rice.  It’s different than white rice, which you dry down and then harvest.”

Another thing that stood out was that with white rice, the big spikes tended to happen right on flood-up.  “So as soon as you put water – it didn’t matter if it was the summer flood-up or the winter flood-up – that’s when this big pulse of methylmercury would come out of the system.  It looks like this is basically legacy mercury.  It was methylmercury that was made in the previous season, and it’s just sort of sitting there in the dry sediment. When you put the water on, the flush comes out.  So it’s not so much an active production at that time, but more of a physical release.

Another important thing which is starting to become a general rule is that the permanent wetlands were always low at the outlet.  “All these processes of production and degradation are always happening in all of these wetlands, but degradation seems to really be a bigger player in permanent wetlands.  We have less methylmercury coming out from the permanently flooded wetlands than the seasonally flooded wetlands.”

In terms of bioaccumulation of mercury in fish, they found that even after 30 days of exposure, they were picking up a lot of methylmercury from the water column.  “These are really high rates of bioaccumulation,” said Dr. Windham-Myers.  “These are little mosquito fish; they’re really, really tolerant little buggers. … Basically, by day 60 at the end of the experiment they were 12 times higher than initial concentrations in the white rice fields, but only about 3 times higher in the permanent wetlands.  So there was accumulation in permanent wetlands, but it wasn’t that high, and it was almost in keeping with what you would expect.”

She noted that this level of accumulation in fish is well above the FDA guidelines of a fish that you should eat.  “These are little tiny mosquito fish, so they are at the base of the food web.  Anything eating them is just going to get more and more mercury in them,” she said.

One possibility of why methylmercury was so high in the white rice fields was that the fields were making so much methylmercury.  They expected to see high methylmercury production and they did.  But they wanted to figure out why.  “So we looked at the two main factors that drive methylmercury production: one is the availability of reactive mercury or that bioavailable pool that can get methylated and then the other is the bacterial activity; we looked at these two in concert to see which one of those were driving the story.”

It’s often called a ‘Goldilocks’ model – it’s the idea that there’s this just-right condition; that when it is in this zone, there is high methylmercury production.  When there is lower rates of sulfate reduction or more sulfide in the soil, there is less methylmercury production because the teeter-totter is pushing on one direction or another, she said.

Dr. Windham-Myers did not go into great detail on the data, but she noted that they did not see a difference between the different wetlands; they all effectively made methylmercury at the same rate.  “It wasn’t the methylmercury production rate that was driving the differences between the fields and the wetlands in general.  They were similar but for totally different reasons.  There was a lot of reactive mercury available in the rice soils, but they didn’t have as high bacterial activity, actually. Whereas, the managed wetland had very low mercury available in the soils and then much higher bacterial activity.  So for different reasons, they were both sort of in the moderate range.”

She clarified that the numbers were not low compared to other places in the country, but for the Delta, they were lower than they expected.  What explains these differences that they saw through time?

We basically started looking at physical processes.  We thought about what is happening and how is this stuff moving around?  It turned out to be a really interesting story with the plants.  Transpiration turned out to be 90% of how water was moving through these fields in the summer.  It’s shown in the black.  When people think about ETevapotranspiration – they just think of it as one lump term; that this is how water leaves the system and moves into the atmosphere.  But they’re very different processes. Evaporation leads to concentration of constituents in the water; whereas transpiration is pulling that water down into the soil, and then water is leaving through the plants, and so you’re getting a concentration of any constituent in the soil.  So where the mercury is is completely different depending on which one of those is the driver.”

She pointed out the difference in the seasons.  “In the summer, transpiration is a huge driver in these fields.  It looks like a lawn; it’s just completely dense – 2,000 grams of biomass per meter squared.  It’s a huge amount of material here.  In the winter, very little biomass on the fields, so therefore not a lot of transpiration.  So what they really looks like is that just from a pure physical process, something as inert as chloride – we use that often as a conservative tracer – would be pulled down into the soil during the  summer, and then released in the winter.   And that’s exactly what we saw with methylmercury.  It’s being pulled down in the summer and then released in the winter.”

This leads to a limitation on methylmercury export in rice fields for two reasons.  “One, it’s physically pulling water out of the surface into the soil, so some of that’s bringing some methylmercury with it. And then the second is that advective pull-down prevents the diffusion up of a higher concentration that’s in the soil.  The diffusion can’t compete with the strong advective pressure.  So there’s a very strong physical story there.”

Shading is also important, Dr. Windham-Myers said, comparing the two pictures of the rice field in May and again in August.  “By August – this is white rice – you can have this hugely dense canopy of shading. And it basically made the amount of sunlight hitting the water be equal between the winter and the summer.  The summer – even though there was, on average, on a daily flux, it was 50 mols per meter squared per day; by the time you factored in the plant absorption of that light, only 5 mols were hitting the water, which is about what you find in the wintertime.”

They also did ‘de-vegetation’ experiments where they didn’t let rice grow and then studied what happened.  “In those de-vegetation experiments, we measured tons of things in the soil.  And what we found is that basically, yes, if you did not have plants there, you could reduce methylmercury production by about 60%.  So you want to shut down methylmercury production?  Get rid of your plants.”

They determined that it was mostly being driven by a reduction in the bacterial activity in these sites.  In particular, they observed things like sulfite reduction was also depressed.  “What they found was that it was porewater acetate as a carbon source that was really what was driven down.  So the plants were bringing carbon to the system.  When you took out the plants, they had less carbon.  That looks like why the bacteria were less active.  We definitely saw less sediment methylmercury in those sites.  The proof is in the pudding.  There was less methylmercury in the soil.

Then they measured other things, such as as porewater chloride as a conservative tracer and iron speciation.  “What we found is some evidence, again, of this rhizo concentration.  So, again, concentrating in the soils.  Then also rhizo oxidation basically keeping what was sometimes called the ferrous wheel going, where the iron is basically moving from ferric to ferrous, from three to two, and we get more moving back into three – the more oxidized form.  We get that happening in the vegetated sites.”

She then presented a graph of the data for porewater acetate, plotted with a log scale on the x-axis, and the bacterial production rate on the Y axis.  “Basically the more carbon in the soil, the more bacterial activity – that’s definitely true for the summer.  And then interestingly, it’s true for the winter as well.  The highest rates of methylation were actually in the winter.  So this decaying rice straw, we believe, is part of the carbon supply in the winter that’s making these fields such good producers of methylmercury in the winter as well as the summer.”

They also looked at the actual concentrations in the seeds to see what it was from a wildlife perspective; are they picking up a lot of mercury by eating these rice and wild rice grains?

The answer was yes.  There was more in the grains of rice and wild rice than in the cattails and tule seeds, which are also basic food web in some of these systems.  So these values, they’re higher than the native seeds.  But just to put this in perspective, they’re way lower than the median values coming out of China of this district that’s getting a lot of attention and making concerns about human consumption.  It’s basically more than 100 times less than you would find, for example, in a striped bass.  So they’re fairly low concentrations, but they still were higher than what you would find in native population.”


They have done some studies in the Cosumnes River Preserve and on Twitchell Island, both places where they’re growing rice.

In the Cosumnes River Preserve, they’re growing it in organic fashion,” Dr. Windham-Myers said.  “They’re adding less sulfate, for example. Sulfite is a really common carrier anion for ammonium sulfate, zinc sulfate, copper sulfate – whatever it is you want to add to your crop often tends to have sulfate attached.  Not so much in the Cosumnes River Preserve.  And then Twitchell Island, they’re also growing rice out there.  But this is a tough place to grow rice because it’s actually pretty cold in the Delta compared to, for example, the Sacramento Valley.”

They have been studying how much is coming in and how much is going out from the rice fields there.  “What we’re finding is the same basic pattern,” she said.  “All of those rice fields are holding their mercury in the summer.  They’re producing it, but they’re not letting it out.  And then in the winter, you get a flush out.  And so they’re basically sources in the winter but not in the summer.

They have been partnering with UC Davis on research in the Sacramento Valley because a large amount of acreage is growing rice up there.  “They have much less mercury in the soil,” she said.  “So it’s really a question of how important are they for export down the tributary – Sacramento River to the Delta.  There’s a lot of the land, but there’s maybe not that much mercury.”

Christy Tanner, a graduate student, has finished studies looking at the role of agricultural drainages on the concentrations of mercury in the Sacramento River at the point of the drainages as well as downstream.  “She’s saying basically the same thing.  The concentrations in the system aren’t as high as what we saw in the Yolo Bypass, but it’s basically a story that the winter is important, and the summer is not so much.”

It’s great that they know the general patterns, and understand when export is coming and not coming, but it doesn’t really get us to the point of fixing it, Dr. Windham-Myers said.

So she’s seeing a significant effect of the agricultural drains in the winter, but not in the summer. So this is all fine and dandy, and we can say that we have these general patterns, we understand when the export is coming, when it’s not coming. But it doesn’t really get us to the point of fixing it.  “So the real question that we’ve been interested in is, can we do anything about it? Now that we know all these little knobs that might be able to turn, could we actually reduce the amount of methylmercury being produced and then how much is coming out of the system?

At the Cosumnes River Preserve, they have a partner in the Bureau of Land Management who allows them to work with the farmers and do work on some of their seasonal wetlands.  “We basically paid the farmers two years ago or three years ago to treat their rice straw differently.  We said, either leave it on the surface like you usually do and flood it, till it into the soil, or bail it up; cut it up and get it off the land, just to see if we could make a difference there.  We were really hoping that we would see a big effect, but we didn’t.  It was not a slam dunk.  There was a lot of background variability between these fields,  so we didn’t get a really strong signal of that carbon removal.  And so there’s promising aspects to it, but it’s – the winter removal was not the real slam dunk we were hoping for.”

They are also trying to see what happens by building a settling pond at the end of the seasonal wetlands; this could be done in a rice field as well.  “The idea is that the water coming off these shallow seasonal wetlands is loaded with methylmercury,” she said.  “If you give it a big deep area to go through, the methylmercury can settle out as particles, if it gloms onto particles. Or it can get photodemethylated in that open water column and then come out cleaner.  And this is actually looking to be somewhat promising.”

She noted that there seeing the same type of results in a Yolo Bypass study being conducted by Moss Landing Marine Labs.  “I would stay tuned.  I think it’s going to pan out to be a valuable approach.”


Dr. Windham-Myers said we’ve learned a lot about how seasonal wetlands and rice operate.  “Rice is a really good surrogate for seasonal wetland,” she said.  “In a seasonal wetland, you never know when the water is going to come or how much is going to come.  And with rice, you know the day, you know exactly  what’s going to happen, so you can study it very, very well.  And so we feel like what we’ve learned in rice actually applies to a lot of seasonal wetlands.”

She said you need to study the whole system; if you treat it like a black box – import and export, you’ll get a story, but you won’t really know why. “You won’t know why a site, for example, had production of methylmercury or a loss term.  And that’s what we’re trying to tweak. That’s what we’re trying to understand.”

The box models that we use today tend to be very big-scale, annual averages.  In these pulsed systems where you have water coming on at different times, the Delta and the wetlands in particular, you need timing. You need to know when it’s coming and what you’re measuring at what point in time.  The reason being, a lot of these processes are concentration-dependent, and those concentrations are changing through time as well as the drivers of those processes.”

We found out that rice-dominated wetlands make methylmercury all year.  They transport it mostly in the winter.  The real exposure to wildlife is in the summer, and that’s not good for the wildlife because that’s their sensitive breeding time. So this is not very good for wetland wildlife to have so much methylmercury on the landscape.”

We saw that, as you increase the amount of mercury in the system through pollution, you did see more methylmercury in the grains.  The plants were critical players in the system: physically – in terms of controlling transport and shading; geochemically – the carbon exudation, the oxygen exudation; and then biologically – some of them accumulated mercury and methylmercury at different rates.”

But there are reasons to be optimistic, she noted.  “We do think actually mitigation is possible.  We’re seeing that there are some knobs that you can turn, and there are some wetlands that are better than others, and we think we know why.  And that’s leading us to think that we can do some wetland management to reduce methylmercury export.”

That said, will this wetland management actually help reduce delta mercury fish levels? And I have to conclude that the solution is not in this piece of chalk.  The solution is not in this talk. We don’t know.  We know that that’s one possible reason.  And what we’re trying to do now is use our better modeling and monitoring tools to pull these pieces together.”


In terms of advancements, the bottom right corner of the slide is a map of Suisun Bay, and is the first map of methylmercury in surface water that has been made using a new airborne instrument out of NASA’s Jet Propulsion Laboratory.  “The bottom line is that methylmercury tends to track dissolved organic carbon, which is basically the brown-colored stuff in water.  So when you see tea-colored water, that’s DOC, or dissolved organic carbon.  Because you can see that from space, or you can see that from a airplane, we found that there’s a strong relationship, and you could actually map it.  And this might be a monitoring tool that we can use to figure out where it’s going at what points in time, where the concentrations are high.”

She then showed a visualization from the CASCADE model, which stands for Computational Assessments of Scenarios of Change for the Delta Ecosystem.  She said it’s a great model that lets you link a lot of processes together.

The visualization was of the physical transport model, to show how methylmercury might move around in Franks Tract, a Delta island that flooded about 30 years ago; it was never reclaimed so it’s now a big lake in the Central Delta.  The model shows where the methylmercury might move if it was in there and wasn’t going through any processes.

You can see the tide’s coming in and out and in and out.  Twice a day, we get a tide.  In this case, it is summer and we’ve got the pumps on.  What you can see is that, even with very low tidal flushing but with these pumps on, Franks Tract gets cleaned out in less than five days.  So anything produced in that site, it’s out of there very quickly.  So that’s the kind of thing that we need to do to figure out if a management process is going to have an effect at this large Delta scale.  And this is the type of stuff we’re trying to push forward on.”


Just in case you’re having rice for dinner tonight, I just want to point out that these are the kind of data we have,” Dr. Windham-Myers said.  “If you look across the world, we are the only people doing anything ecosystem-level in rice fields.  The kind of data that’s out there is pretty much only grain data  – what’s the methylmercury or mercury concentration in a grain of rice.   I just added these new points have come in since we published this paper.  The Yolo Bypass stands out, for example, for total mercury in rice grain.  But it’s the methylmercury that you really should be concerned about, and that’s actually pretty low.  So we’re not having lots of high concentrations here, but in China is really where these hotspots of methylmercury are showing up in the rice grain.”

Rice is expanding across the globe right now, and there are global models people are trying to put together.  If the Minamata Convention actually kicks in and we reduce emissions, would we expect to see an effect on rice grains in these new developing paddies around the world?  Pulling that all together is difficult because we all know mercury is a legacy pollutant.”

Just because we stop emissions doesn’t mean it’s still not in the soil.  “So pulling these pieces together to understand how changes in mercury deposition and changes in rice locations might affect grain concentrations, we’re working with folks now just basically sharing our data because very few people do this kind of ecosystem-level work.”

Finally, I just want to plug the California Rice Commission website.  They have been documenting the wildlife visitation there for the last two winters, particularly during the droughts, and it’s just amazing and really high professional-quality photos of wildlife in case you’re interested.”


QUESTION: Mercury is evil, but there are all sorts of other nasty stuff associated with agriculture.    Where does mercury fit on the sort of scale of competing with the other stuff for evil-ness?

ANSWER: Wow. The axis of evilness  – I don’t know if I could actually speak to that.  I will say that it’s the most regulated contaminant in the bay, but another big contaminant in the bay is PCBs.  So depending on which hat you wear, you might say PCBs is a bigger fish concern or mercury.  With agriculture, I’m not sure which are the, you know, real – the real concerning – if you’re thinking pyrethroids or some sort – like, actually I’m wondering what you’re thinking of the bad …

QUESTION: Did the rice people use a lot of insecticide or … ?

ANSWER: Not so much now because we’re going for a higher-end market that doesn’t really want the contaminants, or the pollutants, from that.  But that said, you know, copper sulfate is what you use to kill algae; and zinc sulfate is what you use to kill fungi.  So those are two things that are still being added in these fields, but they don’t tend to accumulate in the rice in a way of concern.  So I think the bigger issues for agriculture in terms of toxics is organic pesticides and things like that.  And I don’t know how that would fit into a rice story.

QUESTION:  What’s the original source of concentration in coal, to begin with? And second, do we see any pattern geographically or in geological time in coal deposits? Do they differ wildly?  Is Paleozoic coal worse than Mesozoic coal or such?

ANSWER: Two things. Regarding coal, there is a fascinating level of research on that.  I don’t know if there’s a paleo story, but I do know that we now can look at natural abundance isotopes, and if you look at mercury in the atmosphere, you can know whether it came from coal from China or coal from South America or coal from the United States.   You can actually trace this stuff now because of isotopes. So maybe there’s a story there too for age.”

How it got there – most of these fossil fuel deposits are basically wetlands or old algal accumulation spots – where old organic matter accumulates.  And so metals are really common in coal.  – Yeah, I would think that if that’s the mechanism, that then deposits would vary very greatly depending on geography and … depositional environment.  And so there’s much cleaner coal and much worse coal.”

For more information …

Daily emailsSign up for daily email service and you’ll never miss a post!

Sign up for daily emails and get all the Notebook’s aggregated and original water news content delivered to your email box by 9AM. Breaking news alerts, too. Sign me up!

Print Friendly, PDF & Email