Features Contamination Water Issues
Agriculture and nitrates

A decade of research by one of Canada’s leading ground water scientists
has revealed some surprising results, including a relatively simple
method for remediating public drinking water supplies impacted by
long-term fertilizer use.

A decade of research by one of Canada’s leading ground water scientists has revealed some surprising results, including a relatively simple method for remediating public drinking water supplies impacted by long-term fertilizer use. It all happened in the southern-Ontario city of Woodstock.

Fertilizer use on farmland can lead to excess nitrates in ground water-sourced, public drinking supplies. A decade-long Canadian study has come up with solutions and astonishing results.  Advertisement

Dr. David Rudolph is a professor in the Department of Earth and Environmental Sciences at the University of Waterloo, in Waterloo, Ont. An expert on regional-scale ground water protection and management, his research focuses on quantifying the effects of beneficial management practices (BMPs) on ground water quality.

The National Groundwater Association (NGWA) named Rudolph the 2013 Darcy Lecturer, an honor given to one preeminent ground water professional each year to share the results of his or her work with others. Rudolph delivered one of the last of these lectures in December at the NGWA Groundwater Expo in Nashville.

The findings he shared with attendees are extraordinary – not just because they challenge conventional wisdom regarding fertilizer use. The results of his team’s research also offer concrete solutions, for both the long and short term, to address the pervasive problem of excess nitrates in public drinking water supplies.

Nitrates in drinking water
Across the globe, agricultural land use is the largest nonpoint source threat to surface and ground water quality. In addition to increases in pathogenic microbes in surface and ground water from cattle manure, heavy fertilizer use since the mid-20th century has caused spikes in nutrient concentrations in public drinking water supplies, in many cases well above levels safe for human consumption.

At the same time, food and agricultural water demands are only growing. The United Nations forecasts food demand will rise 70 per cent by 2050, with agricultural water demand jumping 50 per cent over that period. And with global food production capacity relying heavily on fertilizer use, the problem of nitrates in drinking water isn’t going anywhere.

“How do we maintain the quality of ground water in the face of managing the agricultural landscape for optimal food production?” asks Rudolph. “Our challenge is to sustainably manage ground water to meet these demands.”

In Woodstock, where Rudolph has conducted research since 2003, the city accesses hundred-foot deep wells in glacial drift to withdraw up to 300,000 gallons of drinking water per day from each well. Over the past 30 years, the city has seen a progressive increase in nitrate concentrations in those wells.

In the 1990s, concentrations topped 10 mg/L nitrate as nitrogen, the maximum acceptable concentration in drinking water established by the Ontario Ministry of the Environment (as well as other regulating bodies like the U.S. Environmental Protection Agency). Elevated levels of nitrate in drinking water are associated with gastric problems and blue baby syndrome, the latter being a blood disorder that can be fatal to infants.

Sorting out where and how nitrates travel through ground water is extremely difficult due to the fact that the source of those nitrates – agricultural fertilizers – are spread over vast tracts of farmland.

“The classic of all nonpoint source problems is determining the capture zone for a well,” says Rudolph. “It’s always moving. Recharge is very dynamic.”

Additional factors like varying fertilizer application rates and complex subsurface hydrology make the problem even more complicated. Vertical transport dynamics in the vadose zone create a time-lagged flux of nutrients from the surface to the saturated zone as precipitation events move nitrates to the water table in multiple stages over several years. Even with significant recharge events, fertilizer stops moving within an hour, creating a strong component of time lag when trying to connect land use with ground water quality.

Woodstock’s experiment
With nitrates rising in municipal wells and few tools to combat the problem, the city of Woodstock braced itself for the cost of building a multi-million dollar water treatment facility to protect residents.

Taking a chance, they instead chose to adopt a beneficial management plan (BMP) focused on managing land-use to protect ground water quality and supply. Part of this strategy included a long-term monitoring program conducted by Rudolph and his team to assess the results of the city’s efforts.

“The city purchased land around the wells, very carefully managing fertilization there,” says Rudolph. “They wanted to rely on the BMP as an alternative to building an expensive treatment plant, while also maintaining production of that land,” he adds.

One goal of the BMP was to find the optimal amount of fertilizer to both produce a good crop and minimize nitrate leaching to the subsurface. All told, fertilizer application was reduced by roughly 50 per cent. The BMP also included crop rotation practices, such as using nitrogen fixers like legumes and red clover as cover crops.

In the past, there were little quantitative data for evaluating the impact of BMPs on ground water quality, making the case for reducing fertilizer use a hard sell to farmers.

“The changes we make at the surface may take years to show up in the water table,” says Rudolph, adding, “There are very few studies I can easily show the agricultural community to demonstrate these approaches work.”

Rudolph aimed to change that with this project. His research, drawing on data gathered from over 130 monitoring wells, has three important goals: 1) quantify changes in nitrate leaching in the field pre- and post-BMP; 2) predict the magnitude of nitrate concentration reduction in the city wells; and 3) estimate the time needed to achieve those reductions. The research has an eye toward constructing datasets acceptable to the agricultural community and keeping uncertainty within easily understood limits.

Unexpected results
The magnitude of the BMP’s impact was anything but ambiguous. Rudolph and his team observed a drop in nitrate concentrations beneath the root zone from 20 mg/L to less than 8 mg/L and a 60 per cent decrease in total nitrate mass loading in the subsurface. These numbers are roughly proportional to the 50 per cent reduction in fertilizer application implemented as part of the BMP.

Most astonishing of all was the fact that farmers saw no reduction in agricultural yields. In fact, yields actually increased by 5 per cent.

Rudolph’s team also used monitoring well data to calculate rates of recharge and nitrate leaching, which were then used to create computational models showing that by 2020, nitrate concentrations in city wells would decline roughly 20 per cent to levels safe for human consumption. 

“We thought the reaction would be great,” he says, referring to the demonstrated effectiveness of the BMP, “but to the public, it was a bit of a shock to have to wait up to 10 years to see those reductions.”

Given that it would take nearly a decade for the effects of the BMP to propagate to the wells, it looked like Woodstock would have to shell out millions to build a water treatment plant after all.

But Rudolph wasn’t ready to give up just yet. Instead, he and his team set out to test an innovative method to reduce legacy nitrate levels in ground water, one that would leverage naturally occurring biological processes.

Under anaerobic conditions, bacteria consume nitrate during respiration through a process called denitrification. The process requires an electron donor, such as organic matter, which is oxidized to reduce the nitrate, an electron acceptor.

“Could we get that to occur in the field while we were waiting for the newly freshened plume to arrive at the wellhead?” asked Rudolph. This approach led him to an in-situ denitrification method developed by Dr. Rick Devlin at the University of Kansas.

This method, called cross-injection, called for injection and extraction wells installed just upgradient of the municipal wells. Injection wells received acetate, a weak acid and an electron donor, in hopes of spurring bacteria into consuming excess nitrate in the subsurface.

It worked. Within a week of oxygen consumption in the aquifer, nitrate levels dropped down to 4 mg/L.

“Without treatment, approximately 1 ton of nitrate per 5 meter width of the aquifer moves toward those production wells. By treating them with acetate, we were able to reduce that by half.”

Ultimately, the experiment saved millions for the city of Woodstock and has implications for any government seeking to balance the demands of food production and ground water protection.

Rudolph points to areas like California’s Central Valley, where elevated nitrates in ground water threaten one in 10 drinking water production wells. Simple, cost-effective strategies, like the one Rudolph’s team used, could be keys to protecting public health without having to construct expensive water treatment facilities.

Rudolph’s ultimate goal is for these results to inform public policy on responsible land-use practices, which is something that can only be achieved with buy-in from key stakeholders like farmers themselves.

“The reception has been positive,” notes Rudolph. “They’re willing to listen, but adoption is slow.”

It’s no surprise given the traditional reliance on fertilizers as insurance against low crop yields. He’s convinced, however, that once farmers understand the tangible benefits of BMPs, including less time and money spent fertilizing crops, change will be possible.  As one farmer excitedly told Rudolph, “That gives me an extra two weeks in Florida.”