During floods, water moves massive amounts of carbon laterally through ecosystems, according to a study that could have implications for climate change and water quality.
The discovery runs counter to how many people see the carbon cycle as vertical—CO2 moving up and down between soil, plants, and the atmosphere.
Researchers based their findings, which appear in Geophysical Research Letters, on analysis of more than 1,000 watersheds, covering about 75 percent of the contiguous US.
Carbon in the environment, specifically dissolved organic carbon or DOC, is a master variable that influences many of our planet’s fundamental processes, such as water chemistry, greenhouse gas emissions, and pollutant transport across land and water, says lead author Jay Zarnetske, an earth and environmental scientist at Michigan State University.
MORE SAMPLING, NOT LESS
“When water flows through ecosystems, it picks up organic carbon from plants and soils, and in many cases, water determines whether the ecosystem is a net carbon source or sink,” he says. “The massive amount of carbon that leaks out of ecosystems as DOC is about as big as the net amount of carbon taken up from the atmosphere each year. So accurate accounting is crucial when managing the ‘carbon bank account.'”
DOC in rivers is like making tea, Zarnetske says. “You start with relatively clear water. This tea then gets flushed to streams during floods, often turning the water brown.”
The new work suggests a better way to account for the carbon leaving ecosystems as DOC by including data from flood events. Citing logistical and safety concerns, scientists typically give rivers wide berth during floods.
As a result, they know less about DOC behavior during floods. When water is flowing fast and brown, though, is when the most carbon is being transported out of most watersheds. In other words, this is a time when more sampling is needed.
The scientists were surprised to learn that floods readily flush carbon from landscapes in diverse ecosystems across North America, spanning from Michigan forests to the Sonoran Desert. They initially thought the DOC would be diluted by floods in many parts of the country.
Floods, however, lead to the release of large amounts of DOC—or stronger tea, metaphorically speaking—from almost all environments in a relatively short time.
“We knew that DOC went up during floods in some areas, but we were surprised to see the same pattern in the vast majority of watersheds all across the country,” Zarnetske says. “Deserts don’t have as much DOC as deciduous forests, but when you have an event like a flash flood, the process is the same, and the torrents of water are chock full of carbon.”
Have you read?
Another important confirmation was the significant role wetlands play in watersheds. The DOC flushing behavior across the US was primarily related to the acreage of wetlands in a watershed. Wetlands act as buffers or storage zones for DOC in watersheds. If floodwaters rise, water and DOC in the wetlands closest to the river can rapidly spill over.
Consequently, where natural wetlands are located within the watershed is important. Draining natural wetlands and “trading them” for another nearby swamp or building an artificial wetland might look good on paper, but it’s going to affect an area’s ability to store and release carbon, Zarnetske says.
“Wetlands are major controls for carbon balance and water quality, and they’re also some of the most vulnerable landscapes,” he says. “If you move them, you’re changing a region’s plumbing and the chemistry.”
For the current study, scientists used data from across the US, but they didn’t wade in a single stream or swamp. Their results came from scads of data that state and federal government agencies, primarily the US Geological Survey, collected over decades. While collecting this long-term data may not seem as exciting as conducting new experiments, the historic data are valuable and their value only grows with time, Zarnetske says.
“It’s not flashy, but it’s powerful data,” he says. “These data were being collected long before we knew of computers and methods powerful enough to analyze it all.”