Cleaner Farms with Bacteria

Scientists are borrowing bacteria’s biological powers to clean up agricultural runoff, as Emma Hiolski finds in the field. Illustrated by Alex Boersma and Mona Luo.

Illustration: Alex Boersma

Ross Clark remembers the first time he had to clear out a trough of fetid, bacteria-infested water. He pulled on his waders and maneuvered into the reeking gully. The trough—a critical precursor to a groundbreaking experiment—was a massive trench filled with a brew of water and woodchips. The woodchips, however, had blocked the water from flowing out through a drain at the end of the trench. Clark shoveled and raked the woodchips to clear the drain—an exhausting job that left him covered in small cuts, which quickly became infected.

The second time the drain clogged and the trench overflowed, Clark gritted his teeth and cleared the drain again. He kept hoping each time would be the last. But for more than a year, he and his team have patched and unclogged the trench too many times to count, he says.

They’ve persevered, though, largely due to Clark’s unwavering commitment to cleaning up the waterways on California’s Central Coast. The trench that he and his team labored to clear—home to bacteria that helps clean fertilizer out of the water when it runs off from farm soil into waterways—was a prolonged dress rehearsal for a larger production to come.

If we stick with it and it does something to improve water quality, it’s worth it.

Clark, director of the Central Coast Wetlands Group in Moss Landing, has spent the last 20 years working to improve regional water quality. He and a large team have built and tested two full-scale trenches, or bioreactors, that house water-cleaning bacteria. In November 2016, the team finished constructing its most ambitious project yet: A set of 12 parallel trenches filled with wet woodchips, designed to filter runoff from about 600 acres of farmland in Castroville.

Across the nation, excess fertilizer from crops pollutes local waterways, causing algae and other plants to grow and clog lakes and estuaries. Some algae can also produce harmful toxins, poisoning humans and animals.

Clark’s goal is to trap excess nutrients from fertilizer before they wash into Elkhorn Slough and out to the Monterey Bay National Marine Sanctuary. With multiple channels instead of just one, this new bioreactor is also an outdoor lab. The team can experiment with various water treatment methods and discover scientific solutions for cleaning up the messy boundaries between fertile farms and our fragile water supply.

From farms to fish kills

Fertilizers are the plant equivalent of daily multivitamins. They provide essential chemicals that plants use to grow and produce flowers, fruit and seeds. Just as adding fertilizer to a flower pot yields a lusher plant with more flowers, agricultural fertilizers boost crop growth and improve fruit and vegetable production. The most prevalent nutrient in fertilizers is nitrogen, which plants need to build DNA and proteins and to photosynthesize.

Farmers apply fertilizer to the ground, but the nitrogen easily hitches a ride in water draining out of fields in a chemical form called nitrate. Nitrate courses through creeks and streams to lakes and coastal waters, causing mischief along the way. Microscopic plants called algae take advantage of the extra nutrients and multiply rapidly to form thick, soupy swirls in the water. These blooms can clog waterways and prevent swimming and boating. Some algae produce harmful toxins that poison people, pets and wildlife. When nitrate runs low, the algae die off; as they decay, the decomposition sucks oxygen out of the water. Fish and other aquatic animals suffocate in these oxygen-deprived “dead zones.”

Perhaps the best-known example of a dead zone forms each spring and summer in the Gulf of Mexico off the coast of the southern United States. Fertilizer from millions of acres of Midwestern farmland drains into the Mississippi River, delivering a massive load of nutrients to the Gulf. The resulting bloom creates a patch of oxygen-deprived water about the size of Connecticut.

To combat the nutrient pollution problem, agricultural engineers around the world have turned to bacteria for help. Under the right conditions, certain types of bacteria can breathe in nitrate, cleaning it out of the water and releasing it as harmless nitrogen gas. Scientists and engineers provide everything these bacteria need to survive and consume nitrate inside special trenches, called bioreactors. Woodchips provide food for the bacteria to eat. Water from farm fields provides nitrate and other nutrients. As long as water flows through steadily, it’s an all-you-can-eat buffet for the microbes.

Bacteria breathe nitrate and convert it to harmless nitrogen gas. Animation: Mona Luo

This type of cleanup happens naturally in marshes and swampy environments, where oxygen content is low and bacteria thrive in plant roots and soils. However, growers can’t afford to cede productive farmland to create entire wetlands for cleaning their drainage water.

That’s where bioreactors come in. Each one is essentially a “wetland-in-a-box,” according to Clark—a cleanup crew that takes up little area.

In the trenches

Clark’s exposure to fertilizer runoff problems began during the 1990s. He worked for the California Coastal Commission, “spending incredible amounts of time and money documenting [nutrient] pollution,” he says. As a former kelp forest ecologist, he had a keen sense of the ecosystem-wide impacts of polluted waters. But he was unsatisfied by measuring problems with no progress toward workable solutions, so he left the commission. He became director of the Central Coast Wetlands Group in 2010 to provide science-based solutions for improving water quality, he says. Part of his approach includes developing bioreactors, which have been used elsewhere in the U.S. and around the world.

In the Midwest, bioreactor use has been going strong for the past 10 years, says Laura Christianson, an agricultural engineer at the University of Illinois, Urbana. There, a network of underground pipes drains water out of the top soil layer so crops don’t drown. These drainage systems also carry away much of the nutrients from fertilizer. Farmers can piggyback on the system by installing bioreactors at the edges of their fields so bacteria can clean the water before it discharges into streams and local waterways.

The typical bioreactor set-up is a narrow trench filled with woodchips. For an 80-acre field, the trench would need to be 15 feet wide, 80 to 100 feet long, and 3 to 4 feet deep, according to Christianson. This requires major earthwork equipment, “like digging a basement for a house,” she says.

Clark and his team have adapted this model to fit farming in California. The Central Coast region contains incredibly fertile land, and the moderate climate allows farmers to grow year-round. Constant growing and crop rotation means farmers apply fertilizer multiple times each year, but bacteria aren’t as active at cooler winter temperatures.

“We’ve collected enough data to know that [bioreactors] just aren’t performing in this kind of weather,” says Fred Watson, an ecologist and water quality researcher at California State University, Monterey Bay.



Average fertilizer use in California counties from 1987-2006, adjusted for county land area. Hover over counties for more detail. Source: USGS. Graphic: Emma Hiolski

With help from Watson, Clark has spent the last 20 years converting flood-prone farmland back into wetlands and building bioreactors to treat agriculture runoff. As in the Midwest, many farmers in California also drain their fields. Drainage ditches and pumping stations move the water into storm channels and out to Monterey Bay. Clark and Watson have integrated their projects with this system by pumping water into bioreactors before letting it drain into storm channels. To keep their new 12-channel bioreactor from flooding, they got creative with their earthworks, building dirt mesas to raise the trenches above sea level.

They also used a piggybacking approach to select a location for the new bioreactor. They diverted one of the pumps in Castroville so that runoff now splashes into a newly built retention pond. From there, Clark’s team pumps the water into the new bioreactor. Twelve parallel channels, each five feet wide, fit into an 80-foot long by 60-foot wide pit. Each trough, separated by black pond liner draped over chain link fencing, is its own bioreactor. Winter rains churned the surrounding soil into thick, sucking mud, and wildflowers and mustard plants bloom around the edges. Farm trucks rumble by on the adjacent dirt road, and power lines emit a faint sizzle overhead. Here, Clark’s team will search for the best method to clean up agricultural drainage water.

Ambitious design

At the new Castroville bioreactor site, Watson perches on a makeshift chair of concrete blocks to share more of the scientific story behind the project. Over the sound of early morning birdsong and distant highway traffic, he explains that he and Clark designed this bioreactor with multiple channels so they could analyze three different ways to treat the water: standard woodchip bioreactor, greenhouse bioreactor, and a plant-based bioreactor. As water from the retention pond flows through the channels—a process that takes about one day—the researchers will track how well each treatment method removes nitrate.

How the bioreactor works: (1) Water carries fertilizer from fields into a drainage ditch. (2) Nitrate-rich water from the ditch is pumped into a retention pond, then into the bioreactor. (3) As the water flows through the bioreactor channels, bacteria clean excess nitrate from the water. (4) The freshly cleaned water drains into the adjacent wetland. Note: This graphic shows a simplified, single-field scenario; the bioreactor actually cleans runoff from multiple farms that drain into a shared ditch. Animation: Alex Boersma

Three channels will house standard woodchip bioreactors. The scientists can use data from these channels to determine how the other experimental treatments compare. Three other channels will remain empty to serve as a baseline, non-treatment comparison.

In another three channels, Watson is testing a greenhouse-style setup to protect the bacteria during the chilly winter growing season, when the weather is too cold for them to stay active. To help keep the bacteria happy year-round, the team installed household insulation under the pond liner of the greenhouse bioreactors, and covered the channels with plastic tarp to capture the sun’s warmth.

“I’ve become obsessed, really, with figuring out a way to passively heat these things,” says Watson, clad in a thick fleece cap and gloves to ward off the bite of the January air. “That’s the big research question we’re about to get answers to.”

Finally, in the last batch of three channels, Clark and Watson are testing something new: growing plants instead of bacteria. They are testing whether a nutrient-loving aquatic plant can remove excess nitrate. If the team can find a plant that thrives in the nutrient-rich fresh water of the bioreactor, Clark hopes that farmers could one day use their bioreactors to both clean up drainage water and generate a sellable product, like biofuel.

Overall, Clark says the project’s goal is to identify a low-cost, effective water treatment method, provide data to state and county water regulators on which methods work best, and work to establish an incentive for growers to install their own bioreactors.

In this podcast, Emma Hiolski describes how scientists engineered a clean water project to deter nuisance geese. Illustration by Alex Boersma.

A global problem

Clark’s bioreactor setup is unique because it allows the team to compare different water treatment methods to one another as they each clean the same body of water. Rather than saying a bioreactor in the Central Valley operating at 110°F is superior to one on the coast operating at 54°F, Clark says his aim is to test different designs on the same water under the same environmental conditions. This approach, especially trying out plants, hasn’t been taken with full-scale bioreactors before.

Christianson, of the University of Illinois, is excited that bioreactor technology is gaining traction on the West Coast. Insulating the bioreactors is a great idea, she says—though the idea wouldn’t necessarily translate to the Midwest, where below-ground bioreactors are naturally insulated by the soil.

These experiments and water quality–improvement efforts are critical to help manage another longstanding problem caused by nutrient pollution: groundwater contamination. For people who might drink such water, this is the biggest risk posed by nitrate pollution, says Bridget Hoover, director of the Monterey Bay National Marine Sanctuary’s Water Quality Program.

Nitrates make it harder for blood to carry oxygen effectively, causing symptoms such as lower blood pressure, higher heart rate, headaches, abdominal cramps, and vomiting, according to the U.S. Centers for Disease Control. Infants younger than six months are less able to resist the insidious impacts of nitrates, which can lead to “blue baby syndrome.” To counteract the risks to its drinking water, the City of Santa Maria, 170 miles south of Castroville, also constructed a bioreactor to filter agriculture runoff that drains into the city’s water supply.

Though nutrient pollution is a pervasive and widespread problem, there hasn’t been much progress in addressing it. The main challenge, according to UC Santa Cruz hydrologist Andy Fisher, is that nutrients come from such a mosaic of sources that it’s impossible to pinpoint polluting targets for cleanup.

Bacteria use woodchips as a food source. Illustration: Mona Luo

“You can’t nail down who put it into the environment, so how do you decide who’s going to pay for getting it out?” he says.

Another challenge is that there is little incentive for farmers to spend time and money on methods that aren’t approved or acknowledged by regional and national water managers, says Dale Huss, a managing partner of Sea Mist Farms and vice president of artichoke production for Ocean Mist Farms in Castroville. Huss donated use of 18 acres of land he rents from the PG&E utility company, with its permission, for the bioreactor and an adjacent wetland restoration.

“To me, it just seems like the right thing to do,” he says. “It’s good for agriculture, it’s good for the environment, it’s good for the company, and it’s good to be seen working with folks on the environmental side.”

Because growers are reluctant to risk land and money on something regulators don’t give credit for, one of Clark’s goals is to provide data to back up bioreactor installations. If water-quality agencies have evidence that certain methods remove nitrates effectively, he reasons, farmers wouldn’t need to bear the burden of installing a bioreactor and proving it works before receiving credit for it.

Full steam ahead

A few weeks before St. Patrick’s Day, Clark’s team is measuring how fast water flows through the bioreactor channels. In an unintentional homage to the Chicago River’s annual celebration of the holiday, they use a bright green, non-toxic dye to visualize the water’s movement through the bioreactors. It spreads slowly through the channels and out the drain into the waterway below, swirling gently like cream into coffee. Eventually the water in the channels matches the color of the fresh grass springing up around the bioreactor.

These bioreactors also feature screen-shielded drains to prevent woodchip clogs—and to avoid the necessity of humans wading into the trenches to clear them. The team is also carefully patching leaks in the pond liner with a heat gun. With most of the troubleshooting done, and after years of permits, legal logistics, securing funding, planning and construction, Clark and company are looking forward to adding the woodchips and gearing up for the next phase of data collection. “Let’s get it done,” Clark says.

The Castroville bioreactor was finally filled and functioning in June 2017 after months of preparation. In the center: woodchip bioreactor, then (left to right) plant bioreactor, control channel, greenhouse bioreactor. Photo: Jason Adelaars

Though the road has been long, Watson is also hopeful that everything will work. Once the system is up and running, his goal is to get nitrate levels in water from the Castroville Ditch as close to the drinking water regulatory limit as possible.

“We’ve got plenty of money in the budget,” he says, “and plenty of students to help out.”

This rare project will remain in place for years to come, so scientists and students can study and hone bioreactor engineering. “Even if these projects take forever and cost a lot of money,” says Hoover, “if we just stick with it and in the end it does something, something to improve water quality, it’s worth it.”

© 2017 Emma Hiolski / UC Santa Cruz Science Communication Program

Emma Hiolski

Emma Hiolski


B.S. (marine biology, biochemistry) Eckerd College

Ph.D. (microbiology & environmental toxicology) University of California, Santa Cruz

Internship: Chemical & Engineering News, Washington, D.C.

I grew up in Chicago near a Great Lake, but I was enchanted by the distant, exotic world of marine biology. My siren was the memoir Dolphin Chronicles, which called me to the warm, salty waters of Florida’s Gulf Coast for college. Before I knew it, I was also studying biochemistry and even some neuroscience.

Though these disparate topics cohered during my graduate research—I studied how marine toxins affect the brain—I felt too specialized. I had too much fun learning about my peers’ work. My department’s science writing course helped me recognize that I could share my excitement and always continue to learn. Coincidentally, the memoir that inspired me was written by SciCom grad Carol Howard, who had laid a path I followed to my own calling more than 30 years later.

Emma’s website

Alex Boersma

Alex Boersma


B.A. (geology) Vassar College

Internship: Goldbogen Lab, Hopkins Marine Station, Pacific Grove, CA

Alex is an artist and paleobiologist from Toronto, Ontario. After studying geology and studio art at Vassar College in New York, she worked at the Smithsonian National Museum of Natural History, conducting research on whale evolution in the Pyenson lab, Paleobiology department. Articles on her research have appeared in multiple news outlets, including the New York Times, BBC, Nature and the Washington Post.

From her background in science academia, Alex has experienced first-hand how important good visuals are to communicating research and engaging a wide audience in scientific findings. She aims to create scientific illustrations that can function not only as an accompaniment to scientific writing, but also stand alone as works of art. Her focus is on using visual storytelling to engage the public in scientific ideas and the natural world. Working in a variety of mediums, both traditional and digital, Alex also has a particular love of printmaking and incorporating unconventional mediums in science illustration.

Alex’s website

Mona Luo

Mona Luo


B.A. (studio art and biology) University of California, Santa Barbara

Internships: Zoological Museum of China (Beijing), Institute of Parasitology (Ceske Budejovice, Czech Republic)

Mona is a science illustrator with a love for the underappreciated and the overlooked. Her favorite subjects include obscure processes and complex lifecycles, with a particular soft spot for parasites.

Mona’s website