• facebook
  • twitter
  • YouTube
  • RSS feed
  • 651-201-6000
  • 800-967-2474
  • 711 TTY

NodeFire Save Document
Home > Protecting Our Lands & Waters > Clean Water Fund > Clean Water Research Program > Denitrification

Analyzing and optimizing denitrification in agricultural surface waters

Principal Investigator: Jessica Kozarek 
Organization(s): University of Minnesota, St. Anthony Falls Laboratory
Sponsor: Clean Water Fund
Award Amount: $396,935
Start Date: 7/1/2013 | End Date: 6/30/2017
Project Manager(s): Heidi Peterson

FINAL REPORT is available in the Minnesota Water Research Digital Library

Background Information

Denitrification is a natural, microbially mediated process that reduces nitrate-nitrogen (NO3-) to dinitrogen gas (N2). Denitrification is the only process which permanently removes nitrogen from aquatic systems.

Nitrate-nitrogen is naturally occurring in the environment while human activities such as, sewage disposal, livestock production, and crop fertilization can elevate levels of nitrate-nitrogen in lakes, rivers, streams and groundwater. At elevated levels, nitrate-nitrogen can have negative effects on human health and aquatic ecosystems. Denitrification converts nitrate-nitrogen to dinitrogen gas, a harmless gas that makes up 81% of the earth’s atmosphere.

About this Project

This research project examined areas of the landscape best suited for denitrification. The idea was to identify landscape features with high potential for denitrification and use those areas to “treat” excess nitrate-nitrogen from agricultural fields. Denitrification is influenced by many environmental variables and the rate of denitrification varies considerably across the landscape. A combination of laboratory and outdoor experiments were used to evaluate different conditions.

Key Findings-Hot Spots and Hot Moments

In general, differences in denitrification rates, microbial communities and the abundance of denitrifying genes were found in the non-floodzone, floodzone, and in-channel areas of the study. Understanding these differences is key to developing best management practices to optimize nitrate removal in agricultural surface waters.


  • A series of crop-specific field days, winter workshops and grower meetings
  • Articles in leading agricultural magazines and newspapers
  • University of Minnesota-Extension articles distributed to the > 1,500 subscribers of Minnesota Crop News blog
  • A University of Minnesota-Extension fact sheet

Events will include:

Relatively stable denitrification rates across sites and seasons.

  • Denitrification rates had little response to environmental variables
  • Bacteria were low in abundance and diversity


  • When compared to non-floodzone sites, the periodic flooding in the floodzone lead to an increase in potential denitrification
  • Single flood events lead to hot moments (enhanced denitrification).
  • Denitrification rates and bacterial abundance and diversity respond to flooding history.
  • Predictive equations combining environmental variables failed to account for the flooding history and will likely misrepresent the relative denitrification rates


  • Denitrification rates varied by season and were dependent on nitrate concentration in the water.
  • Denitrification rates varied throughout the channel and were dependent on organic matter concentration.
  • Bacteria were low in abundance and diversity in sandy, low organic matter sediment. Disruptions of mobile sediment lead to decreased denitrification rates.
  • Predictive equations combining environmental variables described the denitrification rates within channel sediments.

Future Research

Floodplain areas are often recognized as potential biogeochemical hotspots, but questions remain as to how to optimize denitrification in these areas. Specifically, data from our project were able to illustrate a dynamic response in denitrification in response to flooding. The effect of flooding on denitrification varied based on nitrate concentrations and organic matter content, but for situations with high organic matter and nitrate concentrations, the following responses were observed:

  • Length of denitrification activity was enhanced (days) after a short-term flood (hours) with no corresponding increase in microbial denitrifier abundances.
  • Denitrification activity was enhanced after a long-term flood (week) and remained enhanced
  • The microbial community responded in the floodplain during a wet year, becoming more similar to the channel community.

To accurately predict floodplain denitrification, more research is required on denitrification response in floodplain areas as a function of flooding history (duration and frequency).


From this research, several factors were found that increased denitrification and could be applied towards nitrate-nitrogen remediation strategies.

Target areas with low-organic sediment and amend the sediment with a stable carbon source
Field and laboratory results showed that organic sediments had much greater denitrification rates than low organic sediments.

Modify ditch geometry
Traditional agricultural ditches are trapezoidal with steep sides. Results from this study showed that denitrification rates at a site that was periodically under water (inundated) in a ditch with a constructed floodplain had higher rates of denitrification than at a location that was periodically inundated in a trapezoidal ditch. Also, results showed that periodic inundation increased denitrification rates. Therefore, remediation strategies that create sites that periodically flood, and especially strategies that include constructed floodplains, have the potential to increase nitrate removal in agricultural ditches.

Two-stage ditch
One potential example of modified ditch geometry is a constructed two-stage ditch. These practices have the added benefit of slower flows, more reactive surface area, and paired nitrification-denitrification, all leading to enhanced denitrification rates. Also, with slower flows, more sediment will settle, reducing suspended solid and phosphorus loading, and potentially increasing sediment organic matter.

Control ditch water level
Controlling water level would allow for slower flows through the ditch, reducing shear stress and allowing for more sediment water contact time. This would also allow for controlled pulse flows of ditch water. As seen from the Outdoor StreamLab (OSL) results, short-term pulse flows have the potential of increasing denitrification rates, especially in time periods with high water nitrate conditions.

Even with these remediation strategies, it is important to note that nitrate concentrations in ditch water is often so high that these practices alone are not enough to decrease nitrate levels to an acceptable level through denitrification. For this reason, a combination of practices should be used to reduce high nitrate loads in the agricultural Midwest. Headwaters need to be targeted since headwaters, with their shallower and slower flows, have a higher potential for denitrification than higher order streams. However, we do not have enough kilometers of headwater to remove nitrate through denitrification, emphasizing that on-field practices are necessary if we are to solve the nitrate problem.  

MDA Contact

Margaret Wagner
Supervisor, Clean Water Technical Unit