The runoff of fertilizers from farm fields affects water quality throughout the world. In the United States, fertilizer runoff sends large amounts of nitrate—a form of nitrogen—into watersheds such as the Chesapeake Bay and the Gulf of Mexico. There, the nitrate can overstimulate the growth of vegetation and encourage algal blooms, leading to low levels of oxygen in the water, or hypoxia, and causing the deaths of marine animals.
Complicated engineering solutions to the problem, such as building more wastewater treatment plants, may sound good, but Matthew C. Reid, Civil and Environmental Engineering, is interested in harnessing the power of nature instead. “Traditional water infrastructure solutions to pollution are not very efficient,” he says. “They use a lot of energy. Their life cycle costs can be significant, especially when we’re talking about agricultural impacts from nonpoint sources.”
It’s not feasible to build a water treatment facility for each farm, Reid points out. “We really have to be inspired by the environment,” he says. “What can we learn from wetlands, for example, which are known to be effective at acting like sinks for pollutants or nutrients in the environment?”
Woodchip Bioreactors to the Rescue
Reid is exploring the feasibility of using woodchip bioreactors to convert polluting nitrates present in farm runoff into harmless nitrogen gas. A bioreactor is actually a trench dug into the earth and filled with woodchips. Farm drainage water is routed through it, and bacteria living on the chips consume carbon from the wood. As part of their metabolic cycle, the bacteria essentially breathe nitrate, converting it into nitrogen gas—a process known as denitrification. “If all goes well, you end up with effluent that has a lower concentration of nitrate in it,” Reid says.
Bioavailability of carbon is a key to the success of denitrification. Woodchips are mostly made of carbon, but much of it is lignin, a recalcitrant form of carbon that is hard to break down, especially in low-oxygen conditions that may occur in a flooded bioreactor, Reid explains. Decomposers such as fungi can break down the wood into a form that is more accessible for the denitrifying bacteria, but the fungi are thought to need oxygen to survive.
Oxygen: A Key Ingredient
To explore the impact of oxygen in the bioreactor, Reid and Philip McGuire, PhD ’21 Civil and Environmental Engineering, joined with M. Todd Walter, Biological and Environmental Engineering, to carry out research funded by a grant from the National Science Foundation. The researchers collected woodchips from the bioreactor connected to Cornell’s Homer C. Thompson Vegetable Research Farm in Freeville, New York. Using a laboratory flow-through system, they exposed the woodchips to the air for a few days and then reflooded them with water.
“We really have to be inspired by the environment. What can we learn from wetlands, for example, which are known to be effective at acting like sinks for pollutants?”
“We wondered if the oxygen would facilitate the breakdown of the wood through the action of fungi or other organisms,” Reid says. “If so, that would convert the nonbiodegradable forms of carbon to a more bioavailable form, so that when the bioreactor is reflooded, it’s able to remove nitrate better and faster. And that’s exactly what we found.”
While this was a positive discovery, Reid and McGuire worried that there might be a tradeoff: More oxygen introduced into the system might lead to more production of nitrous oxide, a greenhouse gas that is an intermediate in the reduction of nitrate to nitrogen gas. Instead, they found that less nitrous oxide was produced. “We think it’s because there was more carbon—more food—for the microbes to consume, which allowed the bacteria to consume the nitrous oxide faster, leading to lower net nitrous oxide accumulation,” Reid says.
The Fungi-Woodchip-Bacteria Connection
Reid and McGuire have also joined with James P. Shapleigh, Microbiology, in collaboration with the Department of Energy’s Joint Genome Institute, to sequence the DNA and RNA of microbes in the bioreactor in an effort to understand more about its microbial community. Preliminary sequencing has revealed that fungi are indeed present in the bioreactor, which is an unexpected finding since fungi are not typically associated with flooded, anaerobic settings, Reid says. This outcome strengthens the researchers’ hypothesis that fungi are helping to break down the woodchips, liberating some of the more soluble sugars from the wood for the nitrate-metabolizing bacteria to feed on.
Woodchips are just one of many possible materials for the bioreactor, and finding the right material is a delicate matter. “We use woodchips, but there are times when we’ll also use corncobs or other residue from agricultural production,” Reid says. “Having the right balance of material to release carbon at the right rate is important because it has to degrade fast enough to provide enough carbon for the denitrifiers but slow enough that it doesn’t have to be replaced too often. Also, you don’t want to release too much carbon because that can lead to hypoxia in water—the very problem that we are trying to solve.”
Fine-Tuning Bioreactor Processes
The positive effects on denitrification that occur when the bioreactor is drained and then reflooded point to the benefits of being able to fine-tune the process as needed, Reid says. He is now collaborating with Nils Napp, Electrical and Computer Engineering, and Scott Steinschneider, Biological and Environmental Engineering, to pioneer an automated system to optimize the bioreactor’s performance. With funding from the Cornell Institute for Digital Agriculture, the researchers will install an automated valve in the bioreactor at the Thompson Research Farm.
“A nitrate sensor in the effluent will measure the nitrate concentration every 15 minutes,” Reid explains. “If it becomes too high, the sensor will send a signal to the valve to open and drain the bioreactor. It will stay open for a specific period of time and then close and the bioreactor can reflood. This should improve the release of carbon and the removal of nitrate.”
The researchers are also planning to add a supplemental pump with sensors that can measure the flow of water coming into the bioreactor. When the flow is high—for example, during a storm—the sensors will trigger the addition of supplemental carbon into the system. “When the flow becomes very high, it can be difficult for the carbon derived from the wood to be sufficient to reduce all that nitrate,” Reid says. “In that case, adding some sugar water can provide some supplemental carbons.”
Reid continues, “The bioreactor is most effective if we have a steady, low level of flow. But with climate change, we are increasingly getting short, high-flow storm events that are very challenging. If we can build in some level of active, autonomous control, the bioreactor can be more adaptive to different situations. It needs to be sturdy, reliable, and able to remain out in the field for several months without requiring maintenance. If we want farmers to take up this technology, it has to be self-sufficient.”