Arboretum wetlands releasing rather than treating harmful nutrients

East of Curtis Prairie at the UW-Madison Arboretum sit four long parallel swales – low-lying, wet stretches of land – and several ponds. Here, on a day in early May, sounds of traffic from the Beltline Highway mixed with birdsong from nearby woods. Three turkeys wandered along the swales.

Before new growth from the spring filled in, remnants of dead cattails from previous years stood out prominently. Alive, they piled up in the wetlands, dominating other plant life, especially in the easternmost swale.

The UW-Madison planned these wetlands as a stormwater treatment facility, which will take in water to filter this spring. But before they were put into use, a three-year research project collaborated on by six UW-Madison researchers with various backgrounds -- including ecology, hydrology and bioengineering – found that these constructed wetlands actually were discharging harmful nutrients into runoff because of a combination of long pooling period and the abundant cattails taking over the land.

Cattails store large amounts of phosphorus. When they shed leaves and biodegrade, they release that phosphorus, which threatens water quality. In a situation of constant pooling, invasive cattails were able to thrive and crowd out other native plants in the Arboretum, and, in the process brought phosphorus, from the topsoil above ground.

Phosphorous has been implicated in algal bloom and fish death.

Yet, the university still plans to use the wetlands, despite these findings.

The purpose of the wetlands had been “predetermined,” said Rhonda James, a Senior Landscape Architect of the Campus Planning and Landscape Architecture Staff at UW-Madison. The wetlands are part of a larger stormwater treatment system designed to remove harmful chemicals and slow down the water flow, James said.

Joy Zedler, a professor of botany and restoration ecology at UW-Madison who worked on the study, said that allowing more water into the site could cause a more serious cattail invasion and the release of even more harmful nutrients.

“The design (of the facility) is based on a model that failed to account for the effects of cattails,” Zedler said. “If doctors found that an obsolete medicine was doing more harm than good, would they continue prescribing it because the label said it works?”

During 2008 and 2009, contractors dug four parallel swales at the arboretum and seeded the land with 27 native species. But few of them survived through the next year. Instead, cattails invaded and prospered.

After vegetation established and grew dense, researchers directed stormwater into the four swales and assessed the improvement of water quality.

The westernmost channel (Swale 0) was not a focus of the study. The other three wetlands, Swales I, II, and III, which were designed to be identical, however, turned out to be quite different.

The clay layers beneath three swales differ in thickness. While the two swales in the middle (I and II) dried quickly after rainfall, the easternmost wetland (Swale III) held water continuously. The longer the water pooled in each area, the more cattails grew and edged out other plants.

Photos by Stephanie Prellwitz. Map by Google.

But the most surprising result was that the quality of water treatment was different in the three swales. Scientists measured the concentration of suspended solids – small nutrient particles suspended in the water but not truly dissolved – and the nutrients that flowed into and out of each swale, and they calculated each wetlands’ ability to filter different substances.

Results indicated that the channel with fewest cattails, Swale II, was most effective in terms of removing suspended solids and nitrogen. The easternmost swale (III), with a longer pooling period and abundant cattails, actually worsened the water quality. All three swales discharged phosphorus, and Swale III also released nitrogen and suspended solids.

The wetlands, which were supposed to absorb and remove phosphorus –- in the runoff, turned out to be a source of the nutrient.

The study suggested that a layer of 6-inch nutrient-rich topsoil, which was added to the swales during construction, may have been a source of extra phosphorus. 

Adding topsoil “is a common engineering practice,” said Stephanie Prellwitz, a former graduate student in biological system engineering at the University of Wisconsin and one of the researchers on the project.

“Normally it is not an issue to add six inches of topsoil, but in the wetland, the practice needs to be reconsidered,” Prellwitz said.

In Swale III, because of continuous water retention, topsoil lacking in oxygen and phosphorus became soluble and mobile, and therefore, more likely to be taken up by plants, explained Zedler.

The giant biomass created by cattails, said Prellwitz, is like a “tea bag.” Water runs through the leaves and leached out the nutrients in them.

However, the situation could be worsened, Zedler said.

During the research, only a small amount of water was allowed to run through the wetland, to make sure that each swale behaved separately, so researchers could compare each condition, explained James.

The University planned to run more water through the swales this spring. Zedler said that might result in a replication of Swale III, which had more standing water, more cattails and more nutrients released into the runoff.

The situation with this wetland reminds Zedler of a similar case, an eight-acre stormwater pond near Fish Hatchery Road. Invasive cattails dominated in the shallow shoreline surrounding the pond, and crowded out other native species.

Other scientists in the group, however, thought that it might not be proper to over-interpret the study’s results.

Before entering the swales, the stormwater was “pretreated” in two ponds – forebay and retention pond, said Prellwitz, where water was slowed down and solid particles in the water began to settle down. Therefore, many nutrients, including phosphorous, that are attached to these particles also settled down to the bottom of the ponds.

This could limit the wetlands' effectiveness at cleaning the water, as some processes require the presence of specific levels of nutrients.

“It is very difficult to predict ahead of time whether it will act as a source or a sink for nutrients,” wrote Prellwitz in later e-mails, “Based on our experience and other research in this area, the performance of treatment wetlands can be extremely variable because of its complexity.”

“[Zedler] thinks we are not changing fast enough,” said James. “But I think [the study’s implications] can still be applied to future projects.”

James said that “cattails are a poor ground cover for stormwater facilities.” She said her office is now working to identify ways to discourage cattails, although most discussions are still at the preliminary phase. 

In the engineering world, said Prellwitz, “we sort of got into the habit of designing these systems without looking at actually how they are performing.” She thought an important lesson from this study, is “not to assume that the water is being cleaned,” and instead, “be better at measuring the performance of these systems.”