CLIFF NOTARIUS WAS LOOKING CLOSELY at the blueprint for a green makeover of his street as city officials explained their plan for installing something called "green infrastructure" in his leafy neighborhood in northwest Washington, D.C. Speaking at the local community center, they were saying their plans offered a new, better way to control the stormwater that runs in sheets during big storms down his street and into a city drain.

The drain is part of a stormwater system that the city started building in the late 1800s to solve a recurring problem: the flooding of downtown streets. But over time, their solution — a big system of underground drainage pipes — created another problem. As stormwater gushes into drains, it carries motor oil and sediment and excess nutrients like nitrogen and phosphorus through the pipes and empties these pollutants into the city's rivers. Those rivers flow into the Chesapeake Bay, where the pollutants lead to poor water quality, low oxygen zones, and disappearing seagrasses.

Notarius learned that green infrastructure in his neighborhood offered a way of handling stormwater that could help reduce water pollution and make his neighborhood look prettier. The city would help homeowners pay the cost of installing native plants on their lawns in "rain gardens" designed to soak up stormwater. Workers would install strips of porous concrete paving along the neighborhood's streets and alleys to capture more water. And along Notarius's block, they would build "bioretention bump-out" boxes — rain gardens that extend out from street curbs into parking lanes. Water running into the gutter would enter through one end of the garden and percolate slowly into the ground.

Each of these techniques could reduce stormwater volume, at least according to scientific studies conducted elsewhere. This project, however, would be one of the first in Washington in which the city measured results from rain gardens and bump-outs installed across an entire neighborhood.

Notarius liked the idea of a greener neighborhood. After all, he considers himself environmentally conscious; he has a sticker in his window announcing that his house's electricity is generated from wind power.

A new solution, however, can create a new problem. Notarius said he was concerned about the pair of 40-foot-long bump-outs planned for his block: in his residential neighborhood they would eliminate parking spaces. "On a typical day, it's wall-to-wall cars here," he said, "and it's hard to find a place to park when you come home." Green infrastructure, it seems, wasn't only about the greenery.

A green approach to stormwater management was coming not just to Notarius's neighborhood but also to other urban locations around the Chesapeake region. Collectively, the plantings and bump-outs represent an ongoing experiment to explore two key questions: What does it take to install a meaningful amount of green infrastructure across a city or suburb? And can this greening reduce stormwater flow enough to help improve water quality in local waterways and the Chesapeake Bay?

For Washington's city leaders, reducing the stormwater flow has been a regulatory imperative for more than decade. Since 2001, the U.S. Environmental Protection Agency (EPA) has issued a series of directives mandating that reduction and requiring other measures to improve local water quality. The EPA set limits to cap the amounts of various contaminants — like nutrients and sediments, but also metals and coliform bacteria — in the Potomac and Anacostia Rivers and Rock Creek. Those caps were intended to advance a requirement under the federal Clean Water Act to make waterways fishable and swimmable. Reaching that goal meant a lot of work — the Anacostia is listed as one of the dirtiest rivers in America.

Then in 2010 the EPA gave Washington additional marching orders that underscored the need for reductions in stormwater flow. The city would have to join surrounding states to reduce excess nutrients and sediments flowing into local waterways and reaching the Chesapeake Bay. These jurisdictions would have to hit a set of pollution limits called the Chesapeake Bay Total Maximum Daily Load or TMDL. Reducing nutrients and sediments was a priority because they fueled summertime bursts of algae that in turn fed the creation of the Bay's dead zones. And the sediments were clouding the estuary's water and cutting out light essential to the growth of the aquatic vegetation that plays a vital role in keeping the waterway healthy.

Using green infrastructure to cut down the flow of stormwater would play an important role in reducing these loads. Not the only role — a large proportion of the nutrients and sediments that Washington was sending to the estuary was being discharged by the regional sewer plant that serves the city, the Blue Plains Advanced Wastewater Treatment Plant.

But urban stormwater emerged as an important priority because it is the fastest-growing source of nutrients and sediments in the Chesapeake's watershed. Overall, stormwater accounts for about 16 percent of the nitrogen, 16 percent of the phosphorus, and 25 percent of the sediment.

In Washington, cutting stormwater flows would also require dealing with problems posed by the city's "gray infrastructure" — two vast networks of underground concrete pipes. One of these networks is called a "combined sewage system" because it carries a mix of stormwater and raw sewage. About one-third of the city's developed land is drained by this network, which carries waste from flushed toilets and stormwater through street drains to the Blue Plains plant. Perched on the east bank of the Potomac River, it is the largest such plant in the world, but during large rainstorms, this combined flow exceeds even this plant's treatment capacity. When that happens, the flow backs up, and the combined-sewage- pipe network is designed to discharge it from 53 outfalls directly into the Potomac and Anacostia Rivers and Rock Creek.

Washington's water and sewer authority, called DC Water, responded to this problem with a big, gray-infrastructure fix. To settle legal challenges by the EPA and environmental advocacy groups, the water authority agreed in 2005 to build a series of underground tunnels crossing the city to store the backed-up flow during most storms until the Blue Plains plant can treat it. Construction is underway and is to be completed by 2030. (See Digging Deep to Improve Water Quality)

"That vast tunnel will probably be here 500 years from now," says George Hawkins, general manager of DC Water. "This is like the Roman aqueducts, we're building stuff that's going to be here forever."

In 2015, DC Water completed a separate, large improvement to the sewage system designed to further improve water quality. To comply with EPA regulations, DC Water finished a major upgrade to the Blue Plains plant, adding new treatment technology designed to cut its nitrogen discharge by nearly half.

Together, those two projects — the tunnels and the treatment-plant upgrade — are expected to accomplish most of the reduction in nitrogen, phosphorus, and sediment that Washington is responsible for achieving under the EPA's Total Maximum Daily Load to improve water quality in the Chesapeake Bay.

But the city government also faces additional EPA requirements to reduce the volume flowing through yet another network of stormwater-drainage pipes. This one is called the MS4 network, short for the Municipal Separate Storm Sewer System. These pipes carry only stormwater and serve the two-thirds of city land not covered by the combined sewage system. The MS4 network presents its own set of environmental problems. The water in this network does not flow through the city's treatment plant but empties, mostly untreated, from more than 400 outfalls into local rivers.

To reduce that flow and improve water quality, the EPA required the city in 2011 to install green infrastructure, like rain gardens, across the parts of the city drained by the MS4 system. The agency included that condition in the city's five-year permit under the Clean Water Act to operate the MS4 system. City planners and the agency agreed that a key way to reduce stormwater was to convert some of the city's hard, paved surfaces — rooftops, streets, and parking lots — into pockets that function like sponges, soaking up rainwater, instead of acting as chutes leading to storm drains. Reducing the flow of stormwater would also reduce the amount of nutrients and sediment reaching water bodies. And the plants in the new gardens and bump-out boxes would also take up nitrogen.

Green infrastructure would play a role in Washington's plan to meet its Total Maximum Daily Load target for the Bay. The plan calls for taking steps to reduce by 11 percent the amount of nitrogen that the MS4 stormwater pipes send to the Bay. Phosphorus must be reduced by 27 percent and sediment by 26 percent.

Gray infrastructure, green infrastructure — "I think it is a really interesting contrast, two night-and-day approaches to dealing with the same issue," says Glenn Moglen, an expert in urban hydrology at Virginia Tech. The big tunnel approach is expensive but reduces stormwater and nutrient flows by predictable amounts. The green approach can be more cost effective. But because it relies on smaller-scale practices spread more widely over the city, the success of green infrastructure depends on a lot of intangibles, including the performance of the rain gardens, how well they are maintained over time, and how many people choose to install them, Moglen says.

"The green infrastructure is aesthetically more appealing, and it has the potential for engaging the community," he says. But "so much depends on human behavior and human investment in these things for the long term.”

One of the open questions about green infrastructure is whether enough of it can be built fast enough to really make a dent in the city's stormwater problem.

Under Washington's MS4 stormwater permit, the city is required to ensure that about 400 acres of hard, "impervious" surfaces like streets and rooftops are converted to green infrastructure by 2016. Four hundred acres is a pretty big surface, equivalent to a parking lot slightly larger than the area covered by the entire National Mall. To help meet that requirement, the city changed its stormwater-management rules in 2013 in a way that encourages the construction of green infrastructure.

The new rules require new construction projects and renovations over a certain size (5,000 square feet) to include design features that soak in up to 1.2 inches of rain before it can run off-site, away from the building. The figure of 1.2 inches was chosen because 90 percent of all rainstorms in an average year dump that much or less in the Washington region. To comply, developers have to install measures like sidewalk rain gardens or rooftop plantings called green roofs. Some of the water captured by these features seeps into the ground, some evaporates.

Making plans is one thing. But as of 2015, Washington was behind in its progress toward turning 400 acres of hard, impervious surfaces into greenscaped land. From 2011 to 2015, the city recorded only about 100 acres (about four million square feet) in this category. The slower-than-anticipated pace reflects that this part of the city's green infrastructure plan relies heavily on the construction and renovation of commercial buildings. These large projects trigger the city's stormwater rule about controlling the first 1.2 inches of rainfall. But economic factors slowed the speed of construction for several years after the 2008 recession, says Steve Saari, a city official who helps to lead the green infrastructure efforts for the District Department of Energy and Environment.

Saari and his colleagues expect that pace will pick up. The city's plans depend on it — the city estimates that these large projects will provide most of the reduction in stormwater volumes in the MS4 permit area, as developers work to comply with the city's rule.

On other fronts, the city has made more progress. Another regulatory requirement in D.C.'s MS4 permit is to install 350,000 square feet of green roofs. Grasses and plants are grown in beds of soil atop buildings with flat roofs. Rainwater soaks into the soil and the plants take up the water for growth and return water to the atmosphere through a biological process called transpiration.

To encourage homeowners and developers to build green roofs, the city funds a program that subsidizes construction at a rate of up to $15 per square foot. Between 2011 and 2015 owners installed nearly 900,000 square feet of green roofs on more than 150 buildings in the District. And in 2014 alone, more green roofs were built in Washington than in any other American city.

Strips of permeable paving were installed by the city in this Washington residential neighborhood (above top and left) to let rainwater soak into the ground. In urban areas, more stormwater runs off paving and roofs into storm drains than in less developed, unpaved areas (graphic, above bottom), where more rainwater seeps slowly or "infiltrates" into the ground and can also be taken up by plants. Photographs, Jeffrey Brainard (above top); D.C. Department of Energy and Environment (left); graphic (above bottom), D.C. Water

 

While Washington city officials are required to reduce stormwater, they have also embraced the effort and the green projects needed to achieve it, arguing that the changes make the city more environmentally sustainable and a more attractive place to live. For example, in 2006 the city revised its building code to require construction of energy-efficient buildings known as LEED certified. The code promotes the construction of green roofs because, besides diverting water from the sewer system, they can help cool down buildings, lowering air-conditioning costs.

In addition to energy savings, green infrastructure offers other benefits. A study in Portland, Oregon, for example, found that property values increased after the city carried out an extensive green infrastructure plan and built more than 700 bioretention features. And a green infrastructure project in one section of Baltimore measured increased satisfaction among city residents who live in greened areas (see Whatever Happened to Watershed 263?).

The visible, social benefits of green infrastructure are among the reasons that George Hawkins, the general manager of DC Water, pushed to include a green-infrastructure component in the mostly gray-infrastructure plan for building the big tunnels to solve Washington's problem of combined sewage overflow. The 2005 settlement that resulted in the tunnel-construction plan was amended in 2015 to include construction of green infrastructure within 500 acres of city street right of way and public land.

"Green infrastructure works all the time and gray infrastructure only works when the storm is big enough to cause the overflow," Hawkins says. "All the rest of the time it's under there, empty, doing nothing."

How many pollutants green infrastructure removes and under what conditions are questions that water scientists and engineers have studied for years.

Allen P. Davis at the University of Maryland, College Park has spent 20 years researching the long-term performance of rain gardens and similar bioretention stormwater control measures. A professor of civil and environmental engineering, he has been a pioneer in running field and laboratory experiments to test their performance. His research has focused on experimental rain gardens on the university campus and at locations in the surrounding Prince George's County. He helped the county develop one of the first manuals on maintaining green infrastructure so that it continues to work as designed over time.

"Ten years ago, we were just trying to see what these things did — bioretention rain gardens were holes in the ground," he says. "And now I think we're beyond that. The science has evolved where they're not just black boxes. Now we have a pretty good fundamental understanding where we can say, if we want to get a lot of water removal, a lot of nitrogen removal, you design it this way."

The promising news is that his studies — and others by researchers at institutions like North Carolina State University, the University of New Hampshire, and Villanova University — have documented that properly designed bioretention cells can reduce both the volume of stormwater and the amount of excess nutrients that harm water quality. Davis says that the evidence indicates that, under ideal conditions, bioretention cells can remove up to 60 percent of total nitrogen in the stormwater they capture and up to 75 percent of the phosphorus.

Other scientists have found a similarly broad range of performance in green roofs. The amount of research on this topic has picked up in recent years, and studies are finding that green roofs can reduce stormwater runoff by amounts ranging from 30 to 86 percent, according to a review published in 2014 by Roger Babcock at the University of Hawaii. Green roofs also reduced peak flows, when stormwater gushes fastest.

But performance was uneven, varying with factors like the species of plants grown, the depth of the soil laid on roofs, and the amount of rainfall in each storm. As rainfall increased and the green roof's soil became saturated, the volume of stormwater retained decreased. More research is needed to predict consistently how much stormwater green roofs can reduce and under what conditions, Babcock wrote.

Washington's plans for spreading green infrastructure across the city is based on extrapolating from scientific studies like these to hit targets for reducing stormwater volume and nutrients. But plenty of challenges remain both for researchers and for the city managers who would try to apply the scientific findings in practice. The limited number of field studies of rain gardens done to date may not apply consistently in other locations with somewhat different soil types, say Davis and other researchers.

What's more, without proper design, rain gardens can temporarily release more phosphorus than they retain because of the relatively high phosphorus content in some of the soil types used as the plant bed. And the top layer of soil in rain gardens can become clogged with sediments, requiring maintenance to restore their ability to soak up water. In sum, there's little information about the performance of bioretention areas over time, Davis says.

And there is another source of unpredictability about the long-term performance — estimates that changing climate will bring more intense storms that dump more rain over longer periods on city streets and rain gardens. "Ideally, if you get a nice gentle rain once a week, these things will work perfectly," Davis says. But as bioretention areas become saturated, he says, their performance drops off.

Determining the value of green infrastructure for reducing stormwater volume and nutrient loads will require scientists to go beyond controlled laboratory studies and collect data about the effects of green infrastructure installed across a large swath of an urban area, Davis says. "I'm not sure anyone has been able to do that yet because we haven't yet been able to green up a large-enough area to make an impact," he says. "Somewhere along the line, we have to see [reductions in nutrients] show up in the streams and the rivers and, ultimately, the Bay."

Steve Saari and his colleagues in Washington’s government also wanted evidence that the city’s green-infrastructure measures were reducing stormwater flows. So they tried a test of green infrastructure on a relatively small scale, two neighborhoods in northwest Washington called Chevy Chase, D.C., and Petworth. The areas measured only 14 and 13 acres, respectively. But they were chosen in part because each neighborhood study area drained into a single stormwater drain. That would allow the city to install green infrastructure and then measure the effects on stormwater flows.

Starting in 2012, both neighborhoods received a mix of green infrastructure funded by the city as part of an existing city-wide program called RiverSmart Washington. Workers installed strips of permeable paving and bioretention bump-outs along city streets. The city dug more than 60 test holes to ensure that these installations along public streets would be located in areas where the soil would drain quickly enough to make a difference.

The city also gave individual homeowners incentives of up to $5,000 each to subsidize the cost of installing green infrastructure on their private property — measures like planting rain gardens and trees. By the time the work ended in 2014, city officials obtained what they considered to be a high participation rate among the residents there — about half of the property owners in both neighborhoods installed at least one of these features. That gave Saari and his colleagues confidence that they should see measurable results.

In 2015, city officials presented some preliminary findings. Their monitoring showed a significant decline in the volume of stormwater in Chevy Chase. Further analysis will quantify exactly how much. Monitoring work in Petworth is not yet complete. The results so far were a useful and welcome confirmation that the collective efforts are making a difference, Saari says.

Not that every step of these projects was easy. Residents of the Chevy Chase neighborhood were upset when they were told that they couldn’t drive on the permeable pavement installed along streets and alleys because the new concrete took several weeks to cure completely. Still, Saari says, an important lesson from the project was figuring out what kinds of permeable paving are practical and cost-effective to install in neighborhood alleys.

Another important lesson concerned parking — often a sensitive topic in D.C. and other cities. At public hearings, Chevy Chase residents complained that the bump-outs would take up parking in the neighborhood or be placed directly in front of some houses. Complaints like that weren’t always predictable. The project managers received more complaints from the Chevy Chase neighbors about losing parking spaces than they did from those in Petworth — even though Chevy Chase had more available parking spaces, about 180 spaces for 120 cars, by the city’s count.

Saari says that the original blueprint for the Chevy Chase project located the bioretention bump-outs where they would collect the most stormwater. But to address the neighbors’ concerns, city planners decided in the end to remove the bump-outs from Notarius’s block. They relocated others away from the fronts of houses and closer to street corners.

“We sited our bump-outs in areas where people don’t park as much,” Saari says. “That’s kind of common sense, but it was a lesson we needed to learn.” As Washington pushes for more green infrastructure to manage stormwater, it will probably need to balance engineering practicalities with social realities like these. Success may depend on how people feel about how green infrastructure actually looks and how it works in their neighborhoods.