The box culvert that runs under Route 353 in Maryland’s Wicomico County doesn’t look like much, but the water running through them is not only heading to the Chesapeake, but possibly leading to a cleaner Bay as well.

During the past four years, the water passing through the culverts has carried an average of 25 percent less nitrogen than it did before.

The drop came after farmers in the watershed immediately upstream agreed to dramatically alter their farming practices. A nearby watershed, where typical farming practices were maintained, saw no nitrogen reductions.

The study, conducted in the rural headwaters of the Pocomoke River, where long chicken houses punctuate the flat fields, is one of the few which have documented watershed-scale nutrient reductions from an agricultural practice.

But it also suggests that achieving measurable nutrient reductions from agricultural lands may require more widespread and aggressive action than has generally been practiced in the Bay watershed and elsewhere.

The project originated in the early 1990s when the Wicomico Soil Conservation District proposed using federal farm money to measure the effectiveness of nutrient management and animal waste storage on two small watersheds along the Maryland– Delaware border.

“They wanted to know what worked,” said John McCoy, who has overseen the project since 1994 for the Maryland Department of Natural Resources’ Watershed Restoration Division. “They wanted to demonstrate that what they were doing on the ground was actually making a difference.”

As part of the project, the district asked the DNR to monitor water quality in the streams leaving two subwatersheds of Green Run, a tributary in the Pocomoke headwaters.

The project began in 1994, but three years later, a fish kill blamed on pfiesteria in the lower Pocomoke led to the passage of a state law requiring nutrient management on all farmland. Nutrient management plans seek to optimize the application rate of fertilizer with the crop needs for particular fields.

With the change in state law, conservation district officials decided to take the Green Run study a step further.
One watershed continued as before. The other watershed, covering 1,779 acres, was subjected to stepped-up actions. Beginning in 1998, all of the wastes from 1,400,000 chickens per year—more than 5,800 tons over the past four years—was trucked out of the watershed. Instead of using poultry waste as fertilizer, the farmers were paid to apply chemical fertilizer on their 848 acres of crops. Each fall, after the corn and soybeans were harvested, cover crops were planted to absorb excess nutrients left in the ground.

During the past four years, average nitrogen concentrations in water leaving the test watershed declined 25 percent. Nitrogen concentrations were unchanged in the control watershed.

Phosphorus runoff was unchanged from the test watersheds, probably because it was at very high concentrations in the soil before the study even started, leading to continued high levels in the stream.

Nonetheless, the nitrogen results of the study are significant. While research conducted on individual fields has shown various nutrient control techniques can reduce agricultural runoff, few studies have shown an impact at larger scales.

The reason for the success on Green Run, officials say, may be that participation in the program was 100 percent, and all farmers were making significant changes in the way they operated.

“We offered some pretty generous incentive payments to get their cooperation and participation,” said Mike Sigrist, Wicomico County district conservationist with the federal Natural Resources Conservation Service, a part of the U.S. Department of Agriculture. “But we know that with the existing management practices that we have, we can affect water quality.”

Such large-scale participation coupled with large nutrient control efforts are probably necessary to show results, McCoy said. Other watershed studies, with lower levels of implications and less dramatic nutrient changes by farmers, may not have been adequate to show up in water quality monitoring.

“You can’t just tweak the system,” he said. “We’re talking about a landscape issue, where you have to treat as much of a landscape as you can.”

The reason is that simply reducing a pound of fertilizer applied to a field does not necessarily mean there will be one less pound of nitrogen entering the stream.

Nutrients—especially nitrogen—on a farm field are naturally lost to many sources. The majority of the nitrogen goes into crops. The rest is stored in the soil, denitrified (converted into inert nitrogen gas) by natural processes, moved into groundwater or absorbed into such nitrogen “sinks” as wetlands, buffers or cover crops.

Further, the amount absorbed by plants—and the amount running off the land — is highly affected by other factors, such as the amount of rainfall, sunshine and temperature. The better the crops grow, the more nutrients they absorb; when they fare poorly, more nutrients are left to sink into groundwater or run off into the stream.

That bottom line: Because nitrogen is lost to many sources, it may take several pounds of nitrogen reduction to reduce nitrogen entering the stream by one pound.

In the Green Run watershed, the total amount of nitrogen going onto the ground from all sources—fertilizer, manure, atmospheric deposition and septic systems—is huge, averaging 241,000 pounds per year before the experiment, or more than 275 pounds per acre of cropland. But much of that is removed by crops: Nitrogen removal by a field of corn could range from 112 to 140 pounds per acre. Most of the rest was stored in the soil, denitrified or absorbed by nitrogen sinks, such as wetlands.

Before the experiment, monitoring showed the amount leaving in the stream averaged 21 pounds per acre. In the last four years, that was reduced to an average of 14 pounds per acre.

Achieving that reduction required a massive change on the land: The amount of nitrogen applied in the watershed was cut in half to 116,000 pounds, and cover crops were planted on all fields to absorb leftover nitrogen in the fall.

The big reduction in nitrogen application was the result of the switch from chicken manure to chemical fertilizers. Much of the nitrogen in chicken manure is not in forms readily available to the crops, meaning roughly twice as much has to be applied. “You can manage more precisely with inorganic fertilizer than with manure,” McCoy said.

What the study doesn’t show is whether the reductions were primarily caused by planting cover crops, or by switching from chicken manure to chemical fertilizer—or a combination of the two actions.

That has big implications. If poultry waste is the main problem, it could signal that alternative methods of disposal may be needed—whether burning, repackaging as a marketable fertilizer or something else—to protect water quality.

If cover crops are more important, it may mean more money is needed to provide incentives for farmers to plant them. Cover crops, usually grass, are an added expense that provides no income. “One of the things we are trying to do with these guys is keep them in business,” McCoy said.

Of course, McCoy and Sigrist noted, it may turn out that both practices are important. The answer may come soon.

This year, farmers on the test watershed began reapplying chicken waste to the cropland. McCoy and Sigrist hope to get funding to pay them to continue planting cover crops this fall. Monitoring will show whether the level of nitrogen entering the stream changes as a result.

Ken Staver, a researcher with the University of Maryland’s Wye Research and Education Center who is doing soil studies on both watersheds, said both practices probably contributed to the reductions.

Studies by Staver and colleagues have shown that cover crops alone can reduce nitrogen export by as much as 50 percent, but that requires “aggressive” use of cover crops, he said, such as planting them immediately after harvest—something that normally doesn’t happen on a farm.

“I love to see evidence that cover crops are making a big difference,” Staver said. “But here, I suspect it has something to do with both of them. With both of those practices, you would expect to have an effect.”

Despite the exact cause, Staver said, it was important that the project showed significant nutrient reductions on a watershed scale. “They did have reductions,” he said. “Not many have shown that before.”