The Bay watershed leads the nation in stream restoration, with more than 4,700 projects undertaken since 1990 at a cost of more than $400 million, according to a new study.

Those projects span more than 2,200 miles of rivers and streams, and include everything from planting streamside forest buffers to driving bulldozers through streambeds to reconfigure unstable channels.

That makes the watershed valuable as a potential “testing ground” to determine what restoration approaches are most effective, said the study published in the June issue of the journal, Frontiers in Ecology and the Environment.

Before that can happen, though, a greater effort is needed to measure project success. And in that category, the scientists said the Bay watershed lags. They could find monitoring records for only 5.4 percent of stream restoration projects. The national average was slightly better—10 percent.

As a result, while stream restoration efforts “fit nicely” with the Bay Program’s overall goals of improving water quality and habitat, the scientists said it’s uncertain whether those benefits are being realized.

“We really don’t know how effective what we have been doing is,” said Margaret Palmer, an entomology professor at the University of Maryland and a co-author of the paper. “Streams are where rivers are born, and where coastal zones are born, and if our streams are broken, our Bay is going to be broken.”

Palmer and researchers at other universities started a project funded by the National Science Foundation, the Packard Foundation and the C.S. Mott Foundation called the National River Restoration Science Synthesis project. Its aim is to create the first-ever nationwide database of stream restoration projects which could be used to evaluate how different techniques fare in different types of geographic settings.

But the research team discovered that was no simple task, either in the Bay or in other regions of the nation. Little effort has been made to assemble stream restoration data in centralized locations, and researchers often found themselves digging through agency file cabinets to find project reports.

Ultimately, they assembled a database of 37,000 projects nationwide, including 4,700 in the Bay watershed.

Nationwide, the research team recently reported in the journal Science that $14 billion to $15 billion had been spent on river and stream restoration in the continental United States since 1990, or about $1 billion a year.

That figure is probably an underestimate, the scientists said, because project records they reviewed often did not include matching funds or in-kind contributions such as labor. Also, it did not include the costs of such hugely expensive restoration projects such as those involving the Missouri River, or the Kissimmee River in Florida.

But the team found that a comprehensive evaluation of progress was not possible because project reports often contained little evidence of follow-up evaluation. When monitoring did take place, it was often superficial, such as checking to see whether trees that were planted survived, or if physical changes made to the stream were still in place—not whether the stream’s actual health had improved. As a result, it was impossible to identify when and if projects had environmental benefits or—perhaps more importantly—why some projects fail.

“There are a lot of good people out there doing this work, but they are mostly storing their information in their heads or on their personal bookshelves, where they have access to it and learn from it but others do not even know it exists,” Palmer said. “I can’t help thinking what we could do if that information was more widely available and it was analyzed in a systematic way to identify common elements of ecological success in restoration.”

The average cost of a stream project in the Bay watershed is $86,700, according to the report. If the 25 most expensive projects were excluded, each of which cost $1 million or more, the cost drops to about $41,000 per project. That was far less than the national average of $360,800 per project. Part of the reason for the difference is that a large number of projects within the watershed are streambank reforestation efforts, which tend to be less costly.

Many projects in the Bay region were also small—the median size was 1,500 feet of stream length. The paper raised concerns that numerous small projects may not be effective unless they are coordinated within an overall watershed plan.

The scientists noted that long-term monitoring is important for the watershed because the Bay Program assumes that practices such as planting riparian forest buffers will result in water quality benefits and nutrient reductions based on relatively few studies. Also, the benefits of such restoration projects may not be realized for a decade or more.

Monitoring to ensure that those benefits are, in fact, realized would help not only the Bay, but other areas of the nation. Because so many restoration projects have been undertaken in the Bay region, the authors said it could serve as a “testing ground” to determine what techniques are the most effective.

The researchers suggested that the Bay Program develop a centralized tracking system to catalog and analyze the results of stream restoration projects. They also suggested that projects should be coordinated within each subwatershed, rather than be completed piecemeal.

Such a database may help to identify the locations where certain types of restoration efforts will be most effective. Some practices, for instance, may work best in the Coastal Plain, while others are typically more effective in a Piedmont stream.

Not every project needs to be monitored, Palmer said, but a program based on a scientifically credible design that examines different types of projects in different types of settings would yield useful insights. “I agree that we can’t monitor everything—perhaps we only monitor 5 or 10 percent of all projects. But these would be randomly selected, and rigorously monitored using well-designed methods,” she said. “The costs would not be prohibitive. In fact, monitoring is a trivial cost compared to channel reconfiguration costs or floodplain reconnection.”

Investing in monitoring may even help save money, Palmer said, because it could help show whether costly projects are more beneficial than less expensive efforts.

“I don’t want to paint a picture that we are wasting a bunch of money,” Palmer said. “I want to say basically that we want to use the limited funds that we have most effectively.”

Federal agencies involved in the Bay Program have been discussing whether there are ways to coordinate stream restoration work and develop a set of monitoring sites, said Al Todd, watershed program leader with the U.S. Forest Service’s Northeastern Area, who has been heavily involved in stream restoration efforts in the Bay.

“I think it is fair to say that there is more work going on than there is a thorough study of the benefits coming from it,” Todd said. But that doesn’t mean the projects are not working, he added, noting that most restoration techniques have their basis in studies suggesting they provide certain benefits.

“Whether you get those benefits to their optimum potential in every site obviously has a lot to do with all the variety that we see out there in nature,” Todd said. “But it would certainly behoove us to have greater monitoring going on.”

The problem, he said, is that most funding programs do not encourage monitoring. “The people who give out money, whether it is Congress or state legislatures, want to see results,” he said. “They want to see new trees in the ground, new acres of wetlands, new oyster reefs built. That is how they are measuring progress.”

Further, a multitude of state and federal agencies, as well as nonprofit organizations, are involved with stream restoration projects aimed at achieving different goals. That makes it difficult to set common monitoring objectives or to organize a centralized data collection system. “There is probably a lot more monitoring going on out there than we think,” Todd said, “but it is not clearly coordinated.”

Another problem, he said, is that the vast majority of restoration sites in the region are on private property. Many owners are willing to have people come onto their land to do stream work, but are less likely to allow continued access to strangers for years to come to conduct monitoring.

The fact that the researchers found so much money being spent on restoration, Todd said, was not surprising and showed “the huge backlog of need” that exists for stream restoration after centuries of degradation. “In stream restoration, we are in a position where the practice is outpacing our knowledge,” he said. “If we are investing that much in stream and riparian restoration, then we probably should be investing a lot more in monitoring and evaluation.”