Mercury, to put it plainly, is bad stuff. In people, it causes exhaustion, headaches, memory loss, even brain damage; in birds it appears to have similar nervous system impacts and can harm their eggs; in the environment, there’s almost no way to get rid of it. Mercury in the San Francisco Bay is especially sinister: It’s not just a silent, toxic reminder of past environmental sins, but may actually be thwarting our attempts to atone for them. Wetland restoration projects, probably the best thing we’ve come up with for cleaning up our bay and mitigating past errors, often make mercury levels worse. But why that is, where it’s happening, and what we can do about it, are still perplexing.
Which is why marine scientists Katie Harrold and Aroon Melwani are standing waist-deep in the bay near Arrowhead Marsh, dragging a ridiculously awkward twenty-foot-wide seine net through the wetland in pursuit of tiny, bottom-of-the-food-chain fish. Frigid bay water laps at their rubber waders as they shuffle forward by inches to draw in the net. The bay-bottom mud is several inches deep, gooey like custard, clingy like glue, and stinky like the rotten cesspool of decay that it is. There is grumbling about the cold, and envious muttering about another scientist on the project who hurt his back and can’t participate in net-dragging excursions.
Once they’ve walked the net closed, Melwani and Harrold haul it out of the water and over to the bank, where both plop down in inch-deep muck and start unfolding the twine. At the bottom is their quarry, a giant blob of mud and inch-long flopping things, some of which are the crucial “biosentinel” fish they need to tell them how much mercury has entered the food chain in this corner of the bay. Using their bare fingers, Melwani and Harrold start grabbing fish, storing the useful ones in Ziploc bags and chucking the rest back into the water. Of course, there are not enough fish in the first round for all the testing they want to do, so they’ll have to drag the net another two or three times, at which point Harrold looks up at me, her face and glasses splattered, her hands dripping brown, and says, “So, you want to take a turn with the net?”
About a year ago, the US Environmental Protection Agency approved an updated plan for water quality objectives
in San Francisco Bay. One of the plan’s conclusions: It’s going to take time to get mercury levels in sediment where we want them, something on the order of 125 years. San Francisco Bay Water Board engineer Richard Looker presented this model to a group of scientists at a meeting in October, titling his PowerPoint slide “@@@!!! This is going to take a long time!”
The slide got an appreciative laugh from the scientists, although there are plenty of reasons to think that maybe it won’t be quite that hard. (“That 125 years is a really, really rough estimate,” Looker told me later. “People had wanted to know, and that was the best I could do.”) Ridding the bay of mercury won’t be easy, though. “There’s just a lot of mercury in the system,” Looker says. “The only way you lose mercury is it gets buried, it goes out the Golden Gate, or it evaporates. It doesn’t just degrade like a chemical, it has to be physically removed.” Even worse, the question of how mercury ends up in fish—and where it comes from in the first place—is still murky as bay mud.
These are the basics: Mercury gets into the environment in a number of different ways, from natural sources like volcanoes to urban sources like coal-fired power plants or stormwater runoff. In Northern California, mercury was historically used in gold mining, and many of the watersheds in the Sierra and Delta show high levels of mercury from California’s Gold Rush days. Most of the mercury used in gold mining came from mines in the New Almaden area of San Jose, where for decades it has washed down the Guadalupe River and into the southern part of the San Francisco Bay, which still has the highest mercury levels of any part of the bay.
But all this stuff in the air and water is regular ol’ mercury, and one of the odd parts of the puzzle is that regular ol’ mercury isn’t harmful to people or wildlife when it’s in the water. The damaging stuff, the kind that gets into the food chain, is a specific form of mercury called methylmercury. A confounding thing is that there isn’t necessarily a consistent relationship between the amount of mercury in the water, air, and sediment, and the amount of methylmercury in the food chain in that area. It takes a certain kind of bacteria to convert mercury into methylmercury (a process called methylation), and those bacteria are found in areas with lots of vegetation and living organisms. Which is confounding thing number two: In Northern California’s aquatic environments, those areas tend to be wetlands.
Ironically, these areas, which environmentalists have long struggled to save, can create and amplify methylmercury;
in some cases restoring wetlands may resolve one eco-hazard while accidentally worsening another. Even stranger, some areas methylate more than others, which has led to delays and confusion about whether to attempt restorations.
This nasty surprise became known in the mid-1990s, just as scientists and managers in the Bay Area were starting
to map out the West Coast’s largest wetland restoration project. The bay has lost eighty percent of its historic tidal wetlands, and getting them back is a huge priority for scientists and environmentalists. Wetlands are environmentally imperative for a huge variety of reasons: They sequester carbon, filter pollutants from the water, provide crucial habitat for birds and fish, control sediment, and are an important buffer against sea level rise.
The ambitious plan to have 100,000 acres of tidal wetlands ringing the San Francisco Bay by 2030—an almost unequivocally beneficial project—has had to confront the reality that no amount of restoration can return the bay to what it was. The omnipresent mercury that lurks in the environment, waiting to be methylated by restored wetlands, is one reminder that our attempts at restoration could just be making other problems worse.
“Wetlands are in virtually every respect a good thing, in terms of the benefits they provide to the environment and society,” Looker says. “It kills me that there’s a blemish on the reputation of wetlands. But they’re just doing what they do, and mercury methylation is a consequence.” If there was a king among Northern California’s mercury-
in-the-food-chain scientists, it might be Darell Slotton. A research ecologist at UC Davis, Slotton has defined the scope of the mercury problem in the bay and Delta and pioneered a key methodology for studying it. Starting with research in gold-mining-legacy Sierra watersheds, Slotton has spent years identifying the right kinds of small critters, called biosentinels, to study methylmercury patterns in the food chain.
For the most part, it’s useful to focus on animals that have very well-defined, local home areas (so you can identify where the mercury enters the system), wide distribution (so you can compare areas), short life spans (so you can better identify when the mercury enters the system), and regular habits (so you can test seasonally to see if methylmercury is more of a problem at certain times of year). Slotton works mostly with small fish, particularly certain smelts and Mississippi silversides.
After identifying rough patterns of methylmercury in the Sierra and Coast Range in the 1990s, Slotton and other
researchers funded by the CALFED Bay-Delta Program program turned their attention to untangling the wetland methylation scene. The Delta has had major restoration projects along its main channels, and as he set out Slotton expected to find higher levels of methylmercury in animals around those projects.
Yet when the results came back, the highest levels of methylmercury were actually in the periphery of the Delta. The restoration projects in the central and south Delta and Napa-Sonoma Marsh had relatively low levels. “It’s kind of the opposite of what we expected to find,” Slotton says. “There are certain kinds of wetlands that disproportionately
pump out methylmercury in certain conditions. But it looks like a lot of what’s out there being restored is fortuitously not a problem area.”
In some ways, that was great news. But because nobody quite understood why areas could behave so differently,
this mixed result made life very complicated for people who were trying to bring back the wetlands. A typical conversation between a scientist and a restoration manager around the turn of the century might have gone something like this:
Scientist: Not all restored wetlands amplify methylmercury.
Restoration Manager: Great!
Scientist: But some do.
Manager: Ah. Which ones?
Scientist: Uh, no one is quite sure.
So how did the people overseeing this gigantic wetlands restoration project know which areas to rescue?
That was Steve Ritchie’s mercury headache. Ritchie was the guy who, informed of the mercury-wetland connection
in the early 2000s, had to help the US Fish and Wildlife Service decide whether to proceed with wetland restoration projects in the South Bay’s salt ponds. Ritchie already knew that the salt ponds and restored wetlands ringing the area south of the Dumbarton Bridge have some of the highest total mercury concentrations in water and sediment of anywhere in the bay. He didn’t know what effect those new wetlands would have on methylation.
Ritchie convened a scientific working group and asked for a yes/no answer: “‘Should we do this at all?’ People said, ‘Yes, but look for ways to be cautious,’” Ritchie recalls.
“The consensus was that habitat is so much of a limiting factor for species that the risk is worth it.” In other words, they’d chance it for the sake of restoring badly-needed wildlife habitat. Especially risky was the area around the Alviso Slough, at the mouth of the Guadalupe River, where mercury had washed down the river and been trapped in the subsided salt ponds for decades. Ritchie asked a group led by Letitia Grenier at the San Francisco Estuary Institute to take a close look at salt ponds and nearby restored wetlands in the Alviso area, and to predict what might happen if the rest of the salt ponds were restored.
Grenier picked three biosentinels: birds, fish, and flies. Between 2006 and 2008, she and her colleagues collected hundreds of each, stalking through the marsh to collect tiny long-jawed mudsucker fish and brine flies (which are, as Grenier puts it, “sacrificed” for the cause), or jab needles into sparrows (which give blood and then are released). The South Bay hadn’t been studied in this way before, so it took a while to identify the best sentinel species and how to catch them. “The fun thing about all of this is you have to learn how to do it,” Grenier said. “You’re like a hunter-gatherer.”
The results were again encouraging: It appeared that animals in the salt ponds had higher levels of methylmercury
in their blood or tissue than animals in the restored wetlands nearby. Although the results from 2008 aren’t available yet, Grenier’s conclusion at the end of 2007 was that turning those particular salt ponds into wetlands would reduce the amount of mercury in the food chain. Ritchie decided to go ahead with a cautious, and reversible,
restoration plan.
“There’s nothing here that says you shouldn’t move forward, but whatever happens you’ve got to monitor it,” Grenier says. However, she says, it’s imperative to start now. “Given that the loss of our wetlands is happening, and sea level rise is happening, if we’re going to make marshes, we’ve got to start letting them accrete tomorrow, not in five or ten years as we dither about mercury.”
The apparent randomness of which areas are prone to methylation could have led to a great deal of dithering among experts about which areas to restore. Fortunately, over the last several years Slotton and others have isolated a pattern that appears to explain why some wetlands amplify mercury concentrations and others don’t. They started their research by getting better seasonal data, increasing the number of sites where they were collecting fish several times a year from three to 26. They noticed that their data showed huge seasonal spikes in mercury at some sites.
They started working backward, asking, “Well, what happened before this spike?” Slotton says. His answer: “Oh, the San Joaquin river jumped its banks and flooded a bunch of farmland. Six weeks later, we see mercury spiking in the little fish fifty miles downstream. You can link just about every one of the mercury spike events we’ve identified
to this issue of occasional flooding.”
Slotton says that there’s basic chemistry at work in the wet-and-dry cycles: When sediments containing mercury dry out, it changes the way the mercury is bound to the sediment, so that in the next flood it’s much more readily
converted into toxic methylmercury. In other words, seasonally flooded wetlands and floodplains seemed to be generating the most methylmercury.
In 2006 and 2007, Slotton added sites where he could test that hypothesis, and the two-year span of the project turned out to be perfect: In 2006 the region recorded historically high rainfall and severe flooding, and the following year there was a drought. The difference between the fish downstream of those areas during those two years was dramatic.
In July of 2006, at some sites downstream of major seasonal flooding, Slotton was capturing two-to-three-inch fish with mercury concentrations “that would be of concern in a thousand-pound swordfish,” he says. A year later, in the drought period, he captured the same kind of fish in the same place and found a tenfold decrease. “That was it,” Slotton says. “This is the most compelling evidence we’ve had to date.” Regardless of where the marsh is located—scientists have looked at similar data from Napa Marsh and sites in the South Bay—it now seems clear that episodic flooding plays a major role in creating new pulses of methylmercury, and moving them into the food chain.
Slotton’s finding was exciting for managers in the Delta region, where much of the wetland is in so-called “managed marshes”—diked-off areas where the flow of water in and out is controlled by private hunting clubs to make the habitat more friendly for migratory birds, particularly ducks. The results of the research suggest that the clubs could change the way they move water around, even slightly, and have a big effect on methylmercury concentrations. (Not that this would be easy—historically, the hunting clubs, scientists, and environmentalists have not necessarily communicated well with each other.)
But that’s the Delta. In the San Francisco Bay, where most of the marshes slated for restoration are tidal—not diked off, and not human-controlled—the question of where the mercury actually gets into the fish and birds of the bay is still a mystery. Isolating those sources would be the first key to controlling the problem. The second key would be figuring out which sources are more likely to affect wildlife. “If we can answer that second question, that gives us a tool to prioritize our actions,” Looker says. “If we find that industrial mercury is most likely to enter the food web based on its chemical form and where it is discharged, that may give us reason to have additional controls on industrial projects.”
If scientists can figure out what’s causing the mercury to get into the food web, and identify the methylation
hotspots, they might be able to reduce the bay’s level of harmful methylmercury even if its total mercury concentration remains high. That’s why Slotton and Ben Greenfield, a scientist at the San Francisco Estuary Institute, are currently conducting a huge survey of small fish from more than 100 sites around the San Francisco Bay, designed to give even more exact answers about how and where mercury gets into fish. And that’s why Katie Harrold and Aroon Melwani—and I, apparently—are standing in the bay collecting fish.
These topsmelt and gobies may hold clues that will help Greenfield and Slotton find a clear pattern for how methylmercury enters the bay. “What would be really useful is if we suddenly found that all the mercury in the bay was coming from one pipe, and we could just close that one pipe and all our problems would be solved,” Greenfield says, although it probably won’t be that easy. “The more likely outcome is it’s going to be a little bit of this and a little bit of that. We’re trying to figure out what is the this and that.” When they do, it may mean much greater control over the mercury side effects of wetland restoration projects, which is why collecting those small fish seems so imperative.
I accept Harrold’s offer of a turn at the net, don chest waders and rubber boots, and wade out into the water. Negotiating a heavy net full of mud and fish (and one Tecate can) while standing in bay mud that’s up to your shins is no easy task. As Melwani walks toward me to close up the net, one of my boots goes calf-deep into the mud. I yank, and my balance goes backward, but my foot doesn’t move. Then I’m teetering, taking my hand off the pole that holds up the net, wobbling. Finally, there’s nothing else to do but subside quietly backward, in my most dignified manner, into the muddy shallows.
There is a lovely squelching sound as I land on my back, and my hand, propped out for balance, sinks in up to the wrist. Melwani looks at me sympathetically, continuing to hang onto his end of the net; he is much more practiced at this than I am. “Sometimes,” he says, with his very best kindergarten-teacher voice, “we all have to sit down.” As I sit there in the mud, thinking about spending the next three days in a hot shower, I’m reminded again that with mercury science, nothing comes easy.