A civic hydrology park emerges on Duke University’s campus

By Kofi Boone, ASLA

Having lived in Durham, North Carolina, for more than a decade, I’ve come to realize that it’s almost impossible to discuss Durham without referencing Duke University, and vice versa. Duke is a private university, and its West Campus, although in the city, stands apart and within Duke Forest, a vast patch of woods created through a component of a century-old Olmsted Brothers master plan. The campus landscapes cultivated by Duke offer a stark experiential contrast to the eclectic environmental qualities of a rapidly suburbanizing region. Duke’s campus is a big draw for wedding receptions, picnics, walking and biking, and the occasional respite from nearby urban life. Durhamites regularly use the campus as an extended city park system. I’ve visited Duke’s landscapes many times with family and students in search of memorable settings in an educational environment.

Duke Pond, one of the newest campus landscapes, has been an increasingly popular attraction. On a recent visit to Duke Pond with my daughter, she waded into shallow water to scoop up a tadpole and said, “This place is kinda scruffy, but I like it!” When I relayed this story to Warren T. Byrd Jr., FASLA, the renowned landscape architect who concluded his career at Nelson Byrd Woltz Landscape Architects with this project, he laughed. He was thrilled that younger generations felt comfortable engaging the landscape directly. Enabling the informal discovery of ecology was what he had in mind. On a campus populated with works by many leading landscape architects, most of them manicured and tightly controlled, the pond offers an example of a different aesthetic as well as the roles landscape can play in exciting the next generation about environmental stewardship.

In 2007, North Carolina was in the grip of one of the worst droughts in the state’s history. Governor Michael Easley ordered a ban on nonessential watering in all 100 counties in the state. In Durham, the sting of the ban was felt by home and business owners who watched their lawns and garden beds shrivel. But the order did not appear to restrict water use at Duke University, in Durham. Duke, one of the city’s largest employers, did not stop nonessential water use. Duke’s irrigation of its athletic fields during the drought made local and national news.

Over decades, the campus grew in a suburban pattern that largely disregarded the 1924 master plan’s intent. Hough remarked that the 2000 master plan was focused on “fixing the issues with suburban planning, promoting infill, and developing new processes to strengthen and protect the landscape.” Part of that process included learning from leading researchers how campus planning and design was affecting water quality. Curtis Richardson, a researcher at Duke’s Nicholas School of the Environment, and his colleagues established an effort called the Stream and Wetland Assessment Management Park, or SWAMP, and monitored the water quality of the stream that snakes through the West Campus. They found high levels of nitrogen and phosphorus, primarily owing to runoff from Duke’s campus spaces. There was a need for new processes to improve water quality.

There was also a need for strategies to increase water supply. Chilled water was one of the largest water uses on campus, more than 10 times higher than irrigation, at 30 percent of total campus usage. This meant Duke purchased potable water from Durham for $80 million per year. This water was crucial to the function of Duke Hospital, which is on the West Campus. “The hospital is a Level 1 trauma center,” says James Caldwell, a civil engineer at McAdams. A requirement for this type of center is access to a two-week supply of water in case of disaster. The economic and procedural costs of not addressing Duke’s water consumption patterns became the imperatives pushing for what we now know as the Duke Pond.

Caldwell and his company conducted a stormwater impact analysis to model the behavior of stormwater on campus and to determine sites for stormwater harvesting. The site of a new chiller plant was selected for construction first and would serve half of Duke’s West Campus, including Duke University Hospital. The pond site, then wooded, could yield 100 million gallons per year from a 265-acre drainage area. This would be enough water to cut potable water consumption in half. The chiller plant’s design and construction were state of the art, with a sophisticated regulation system from Siemens. But in the early stages of the project, the stormwater harvest pond was to be inaccessible to the public and hidden in the woods.

Hough saw an opportunity to reinforce the principles of the master plan through the project. A site was chosen for the pond, 18 acres near a major campus gateway, which could help reinforce the university’s values. “The engineering planning was fine,” Hough says. “But they were not concerned with placemaking.” He went back to the 1924 master plan and saw that two lakes were shown in the campus vision. One was created as a part of the Sarah P. Duke Gardens, but the second was never built. Hough made a series of persuasive presentations to university officials, including the facilities planning team that was developing the new chiller plant and the pond, that convinced them not only to focus on reclaiming water but also to reclaim the design and planning intent hinted at nearly 80 years earlier.

“It wouldn’t have happened without Mark,” says Byrd. He recalled completing the Dell at the University of Virginia and touring the site with Hough. The Dell in some ways served as a precedent for Duke Pond. Although its primary role was stormwater control, the Dell included design elements to reinforce campus and regional character. Juxtaposing formal elements (site organization, walls) with informal elements (native planting beds, irregular water edges) resonated with the vocabulary of Duke’s campus landscape. Byrd had ties with Duke, having worked with his wife, Susan Nelson, ASLA, on a Duke Gardens visitor center in the late 1990s. At Hough’s invitation, Byrd and his office joined with the engineering and facilities team to do a rapid site analysis and begin a conceptual design process.

Many constraints informed the approach, among them the existing underground infrastructure. Byrd described the process as “back and forth between sketching and data—maximum water volume versus user experience and saving trees. That was a real give and take,” he said. I was surprised at the extent of tree removal when I viewed the before and after construction aerials. However, the context of the larger canopy of Duke’s campus and the careful considerations given to revegetation to protect water quality were also evident in the design process. And out of that give and take, a design concept emerged. The team reframed the project as a civic hydrology park attempting to balance utility and placemaking. Duke University tripled the project’s budget, to nearly $12 million, to incorporate important placemaking elements. If successful, the Duke Pond design would pay for itself in water savings in around 15 years.

Image courtesy Nelson Byrd Woltz Landscape Architects.Byrd and the team created a geometric site plan with an oval forebay and amphitheater-style terraces, which were conceived as an outdoor classroom. You enter the site at Circuit Drive, the major prospect, with a view over the entire site. From there you go down a series of steps and terraces slightly off-center from the forebay, following a culvert that serves as the water source for the pond.

Duke stone, the unique 12-color blend indigenous to a nearby vein of Carolina’s slate belt, reappears throughout the site, including in the culvert structure. The stone clads many of Duke’s buildings and was selected by the Duke family at the conception of West Campus to symbolize the campus buildings’ “coming out of the ground.” The occurrence of the stone at the pond lends unity and solidity to the landscape. The forebay serves a utilitarian function; it captures and slows the first flush of stormwater, including trash and pollutants. But it is also an elegant form. A bowed bridge, built with milled wood reclaimed from trees felled on the site, marks the boundary of the forebay, and a deck extends into the water from the terraces. The entire composition and the stone-and-timber palette offer a legible and contemporary recollection of the sense of place evoked by regional mill ponds from a century ago. The design formalizes on its other edge across the water as well. The gateway image from Towerview Drive is a stone-lined sluiceway in a bosque of trees with the dam holding the pond behind it. The view is striking from the street and the pond, although it is hard to reach on foot. Because the pond occupies a campus gateway, this area offers a memorable transition between the campus and the city.

An important challenge for the design team was the two to four feet of water-level fluctuation required as the pond filled or was drawn down to serve the chiller plant. “The pond has held as much as 90 million gallons in August,” and could be drawn down at any moment, Caldwell says. Evan Grimm, a Nelson Byrd Woltz senior associate, remarked, “We didn’t want a brown ring” around the pond when water was tapped for the chiller plant. To address this factor, the design team emulated wetlands and stream edges by transforming the perimeter of the pond into a series of fall shelves and wetland benches.

The variation in water depths from edge to center allowed the use of a wide variety of vegetation to address both the water quality and appearance of the pond. Grimm says it was experimental at the time and resulted in a diverse planting palette of many nitrogen- and phosphorus-fixing plants. I enjoyed experiencing the green ring of plants on my visits to the pond. It offered the most complex mixes of color and texture within the site, and it looked compelling across the seasons. You can see it best when walking along the southern edge of the pond beyond the bridge to a small deck, pier, and shelter that anchor the largest pond expanse. Cypress trees, well-known for handling water-level fluctuations, march from the shore to the lower benches. Clumps of bee balm and other native perennials are dense and tightly knit. Insects swarm the flowering plants, and fish and amphibians swim among the roots and stems. On the whole, the native plantings along the pond edge are well established. SWAMP continues to monitor water quality and reported a 68 percent reduction in nitrogen and phosphorus since the park opened, largely owing to the pollutant-filtering plants.

A third challenge for the design was providing a transition from the height of the mature existing pine forest to the scale of a pedestrian walking along the looped path. The construction of the pond required the removal of hundreds of trees, mostly pine and some southern hardwoods. Grimm described how the steep slopes (created by the pond’s excavation) and the tall remaining trees created an imposing vertical wall of existing vegetation. Native fast-growing shade trees and flowering trees were planted in strategic locations to address the disparity. Tulip poplars, maples, and other trees are growing rapidly, especially between the main path and the pond edge. What was once a more uniform pine forest now has a more dynamic range of planting, reflecting more of the diversity you might expect at a pond edge. However, there are some maintenance issues. Invasive shrubs and ground covers are sneaking in the zone between the edges of the site and the main path. Hough mentioned that establishing the vertical transition envisioned in the plan will take additional maintenance approaches to succeed. The maintenance issues and lack of shade in some areas seem to be growing pains and not systemic issues with the design. Hough acknowledges these short-term challenges, and with attention to them, as the landscape matures, they should become less apparent.

The main path is made of gravel, and it offers some moments of interest, but it also suffers from erosion in steep areas. There are signs that the place is accessed on many sides by a wide array of people. Several desire lines have developed from bus stops and other places off-site. These incursions are good things—evidence that the place has been discovered and is being integrated more fully into the city’s pedestrian system. However, there are still questions about the connection points between the formal and the informal surfaces used by walkers. Responding to these emerging volunteer path networks will be critical to enhancing the experience of walking through the site.

The transformative effect that a landscape mind-set had on what could have been a purely utilitarian project cannot be overstated. It permeates each aspect of the Duke Pond process. Duke University is a wealthy institution, and one could question the transferability of the process of the pond’s creation. However, with a future phase coming, which includes a second chiller plant and a similar pond, the collaborations forged in this project could set a benchmark for institutional partnership in the pursuit of water conservation. When the future phase is complete, Duke’s potable water consumption from the City of Durham will go to zero. And the landscape expectations of similar work will begin from a higher place.

Project Credits

CLIENT DUKE UNIVERSITY, DURHAM, NORTH CAROLINA. LANDSCAPE ARCHITECT NELSON BYRD WOLTZ, CHARLOTTESVILLE, VIRGINIA. CIVIL ENGINEER MCADAMS COMPANY, DURHAM, NORTH CAROLINA. GENERAL CONTRACTOR LECHASE CONSTRUCTION, DURHAM, NORTH CAROLINA. GRADING AND UTILITY CONTRACTOR MID-ATLANTIC INFRASTRUCTURE SYSTEMS, WINSTON-SALEM, NORTH CAROLINA. GEOTECHNICAL GEOTECHNOLOGIES, RALEIGH, NORTH CAROLINA. STRUCTURAL ENGINEER LHC, RALEIGH, NORTH CAROLINA. MECHANICAL ENGINEER AFFILIATED ENGINEERS, CHAPEL HILL, NORTH CAROLINA.

Kofi Boone, ASLA, is a professor of landscape architecture at North Carolina State University in the College of Design and a member of the Landscape Architecture Foundation’s board of directors.