Conservation Drainage Network

Conservation Drainage

by

Global engineering design review of wood-based denitrifying bioreactors

Summary prepared by: Laura Christianson

Christianson, L., C. Díaz-García, R. Moghaddam & C. Rosen (2026) Global engineering design review of wood-based denitrifying bioreactors, Environmental Technology Reviews, 15:1, 484-503. https://doi.org/10.1080/21622515.2026.2691809

Major Findings:

This paper reviewed full-size denitrifying bioreactors used for a variety of nitrate-treatment applications around the world (see map). Performance of all bioreactors is underpinned by basic engineering concepts, many of which were summarized here to better position bioreactor research and implementation to accelerate in coming decades.

Results of this review documented:

  • Tile drainage, wastewater, shallow groundwater, and stream/ditch uses of denitrifying bioreactors each have unique implementation opportunities and challenges.
  • Only a few bioreactors have been built to treat drainage from whole catchments, but catchment- and field-scale bioreactors provided similar nitrate-nitrogen loss reductions of approximately 3.6 lb N/ac/y. The catchment-scale bioreactors, which treated on the order of 270 acres, provided greater N load reductions of 660 lb N/y compared to field-scale bioreactors, which treated more like 35 acres and removed 120 lb N/y (all median values).
  • Woodchip bioreactors treat nitrate in tile drainage water effectively across latitudes, despite cooler drainage water in northern locations.
  • Decreases in performance with increasing bioreactor age are most likely hydraulics-related although more monitoring of bioreactors > 4 years old is recommended.

Figure 1. Map of the 353 compiled wood-based bioreactor site-years; including 136 site-years for field-scale tile drainage treatment, 23 for catchment-scale tile drainage treatment, 13 for stream or ditch water treatment; 45 for wastewater or non-agricultural drainage treatment; 51 for denitrifying walls, and 85 for pilot-scale bioreactors.

Contributions to the conservation drainage community

After roughly a quarter century of woodchip bioreactor research, much is known about this flexible and robust technology. Treatment of nitrate in tile drainage has been the primary application to date, so current understanding of bioreactor design, construction, and management can be heavily credited to the agricultural drainage sector. However, to continue to build momentum for future bioreactor advances, lessons should be borrowed from other sectors and from around the globe. Looking forward, more long-term monitoring is recommended especially for evaluation of: aging bioreactors; treatment of larger drainage areas and high flows; designs with varying aspect ratios and features; and pairing bioreactors with other technologies (e.g. automation, water storage).

by

Controlled drainage may benefit soybean yields in dry years, but reductions are possible in wet years

Summary prepared by Jeff Strock and  Jane Frankenberger

Strock, J.S., Youssef, M.A., Moursi, H., Nelson, K., Ranaivoson, A.Z. and Poole, C.A., 2025. A synthesis of soybean yield response to controlled drainage under varying precipitation patterns in the US Midwest. Agricultural Water Management, 318, p.109707. https://doi.org/10.1016/j.agwat.2025.109707

Major Findings:

This synthesis of soybean yield comparisons under paired controlled drainage (CD) and free drainage (FD) fields at 11 Midwestern sites (Figure 1) found that while CD consistently benefits water quality, its effect on soybean yield is limited and variable.   Yields were similar under FD and CD in most years, and across all 31 site-years there was no significant overall difference in soybean yield between FD (average yield 58 bu ac⁻¹) and CD (average yield: 58 bu ac⁻¹) (Figure 2).

Figure 1. Soybean yield study locations (Strock et al., 2025)

Figure 2: Yields under controlled drainage (CD) were very similar to yields under free drainage (FD) in most years (Strock et al., 2025)

The authors analyzed those observations with yield differences of ±4% or more, and found that the main factors determining yield response were the timing of wet and dry periods, drainage system design, and management of water levels. Specifically, during mild to moderate drought years CD increased yields by 4–8%, while during wet growing seasons CD reduced yields by 4–9%. They also concluded that intensive management or automation of water control structure outlets is needed to avoid potential yield reductions.

Contributions to the conservation drainage community

The authors developed practical guidelines for management of the drainage outlet to maximize benefits during dry periods and minimize harm during wet periods, summarized as follows:

  • Planting to V5: The drainage system should be maintained in FD mode to maintain adequate water level for emergence and well-aerated rootzone for healthy vegetative growth.
  • June (R1, beginning flowering): A shift to CD during this time of the year would capture soil moisture in the soil profile, but the drainage systems should remain in FD mode if wet conditions persist throughout June.
  • July (R2, full flowering to R4, full pod): The drainage system should be operated in CD mode since dry stress is likely during this time.
  • August and September (R5, beginning seed to R6, full seed): Timely adjustments of the outlet depth with CD are particularly needed during this period to minimize  waterlogging during critical periods of seed development.
  • September and October (R7 to R8, maturity): Once soybean has matured, the drainage system can be changed from CD to FD in preparation for harvest. Managing the drainage system in FD mode improves trafficability for fall field operations.
  • When fall field operations have been completed, the system should be shifted into CD mode to minimize nutrient loss that may occur through drainage water flow.

Publicly available data

This is the latest paper using data from the Transforming Drainage database, which brought together research data spanning 39 experimental sites across the tile-drained U.S. Midwest. Data can be downloaded from the repository at the USDA National Agricultural Library Ag Data Commons, or at Drainagedata.org. Site Summaries are also available with maps, description of the data in the database, and citations for further information for each experimental site.

by

How Long-Term Subsurface Drainage Changes Soil Properties

Summary prepared by: Sushant Mehan, Department of Agricultural & Biosystems Engineering, South Dakota State University

Frankl, A. L., Sherbine, K. T., Strock, J. S., Fernández, F. G., Cates, A. M., & Pease, L. A. (2023). Comparing the short- and long-term impacts of subsurface drainage installation on soil physical and biological properties. Journal of Soil and Water Conservation, 78(6), 457–465. https://doi.org/10.2489/jswc.2023.00147

Major Findings:

Researchers evaluated seven agricultural fields in northwest Minnesota with different drainage histories:

  • >15 years since drainage installation
  • <5 years since drainage installation
  • One undrained reference field

Soils were sampled at 0–15 cm and 15–30 cm depths during the 2021 growing season. Measurements included:

  • Saturated hydraulic conductivity (Kfs)
  • Bulk density
  • Aggregate stability
  • Soil biological indicators:
    • Potentially mineralizable carbon (PMC)
    • Water-extractable organic carbon (WEOC)

Results showed:

  • Fields drained for more than 15 years had significantly higher saturated hydraulic conductivity, indicating faster water movement through the soil profile (Figure 1). This increase is likely due to the long-term development of preferential flow paths, such as macropores and cracks, that connect the soil surface to subsurface drains.
  • At depths below 15 cm, fields with older drainage systems showed higher biologically active carbon and nitrogen pools, including greater potentially mineralizable carbon, water-extractable organic carbon, and water-extractable organic nitrogen (Figure 2). These indicators suggest enhanced microbial activity in deeper soil layers over time.
  • Despite expectations, bulk density and aggregate stability showed little difference between newer and older drainage systems. Tillage practices appeared to have a stronger influence on these properties than drainage age alone.

Figure 1. Fields drained for more than 15 years showed faster water movement through soil compared to recently drained fields, indicating long-term development of preferential flow pathways. (Taken from Frankl et al. (2023)).

Contributions to the conservation drainage community

Drainage impacts are dynamic, not one-time changes. Older drainage systems may: (i) Move water and nutrients more efficiently; (ii) Increase connectivity between surface soils and drains; and (iii) Influence long-term nutrient loss pathways. Soil health indicators can improve below the surface, even if surface properties appear unchanged. Considering drainage age alongside management practices may improve decisions related to nutrient management, drainage water recycling, controlled drainage, and edge-of-field practices.

Subsurface drainage systems continue to reshape soil physical and biological processes for decades after installation. Tracking soil health and hydrologic indicators over time provides a more complete picture of drainage impacts and supports better long-term water quality management.

Figure 1. Soils in fields with long-term subsurface drainage showed greater biologically active carbon and nitrogen pools, particularly below 15 cm, indicating increased microbial activity over time. (Taken from Frankl et al. (2023)).

by

Design approach tradeoffs and annual performance prediction for denitrifying bioreactors treating subsurface drainage

Summary prepared by: Laura Christianson

Christianson, L.E., C.H. Hay, R.D. Christianson, B.M. Maxwell, and R. Cooke. 2026. Design approach tradeoffs and annual performance prediction for denitrifying bioreactors treating subsurface drainage. Journal of Environmental Management, 397, 128245. https://doi.org/10.1016/j.jenvman.2025.128245

Major Findings:

This paper describes monitoring at ten woodchip bioreactors that were built between 2016 and 2021 in Illinois. A total of 36 site-years of bioreactor performance were assessed. These results showed:

  • The ten bioreactors removed nitrate from the tile drainage water in all years, including both wet and dry years, which reiterates this practice’s dependability.
  • Across the dataset, the bioreactors removed 107 pounds of nitrate-nitrogen (N) per year which equated to reducing the annual N loss by 3.7 lb N/ac (median values; Figure 1a and 1b).
  • The bioreactors generally treated the majority of the annual drainage flow from the field and removed approximately half of the nitrate that entered (medians: 60% and 51% respectively; Figure 1c). This resulted in a median edge-of-field N removal efficiency of 17%.

Figure 1. Summary performance of 36 woodchip bioreactor site-years from Illinois, USA monitored at ten individual bioreactors. Boxes represent the 25th-75th percentiles, stems encompass the 10th-90th percentiles, and dots show the outliers. Inside each box, the solid line represents the median and dashed line represents the mean.

This study showed two notable tradeoffs in how bioreactors work and how we think about their design:

  • A given bioreactor was able to treat more water less well, or less water better. In scientific terms, annual bioreactor hydraulic retention time (or, the average length of time a parcel of water stayed inside the bioreactor for treatment) was negatively correlated with the annual percentage of flow captured but positively correlated with annual nitrate removal efficiency for the treated water. As you hold water longer inside a bioreactor, you achieve a greater nitrate removal efficiency for that water but then you treat a relatively lower fraction of water from the field because more flow will bypass the bioreactor.
  • Site-years with the highest edge-of-field N removal efficiencies (37–72 %) often removed less than 100 lb N annually. Site-years that removed the most pounds of N (for example, >220 lb N annually) exhibited a lower range of edge-of field removal efficiencies (7–36 %). This was because efficiency-based performance metrics such as % removal are limited in that they are, by definition, relative to inflow values. Across the dataset, as annual N loading from the drainage areas increased, for example in a wetter year or at a larger drainage area, the edge-of-field N removal efficiency decreased whereas annual N load removed by the bioreactors increased.

Contributions to the conservation drainage community

This study supported previous studies by showing that bioreactors effectively remove nitrate from tile drainage water.  These bioreactors worked in wet and dry years and at relatively small and large drainage areas (17 to 72 acres).

This study also highlighted a dichotomy in how we often think about bioreactor performance. Achieving a high nitrate removal efficiency (that is, a high % removal) was inconsistent with removing a high mass load of nitrate (that is, removing lots of pounds of nitrate). Design approaches need to better balance removing high nitrate loads for greater watershed benefit with appropriate site-level effectiveness targets.

This study points to the need for additional bioreactor design research to better tune design aims and performance objectives. In summary, (1) bioreactors work and (2) more research is needed to expand their application for greater water quality benefit at scale.

by

Effectiveness of saturated buffers on water pollutant reduction from agricultural drainage

Summary prepared by: Gabriel Johnson, Department of Natural Resources Ecology and Management, Iowa State University

Johnson, G., Christianson, L., Christianson, R., Davis, M., Díaz-García, C., Groh, T., … & Rogovska, N. (2023). Effectiveness of saturated buffers on water pollutant reduction from agricultural drainage. Journal of Natural Resources and Agricultural Ecosystems, 1(1), 49-62. https://doi.org/10.13031/jnrae.15516

Major Findings:

This paper reviewed published studies (2014-2022) on the performance and cost-effectiveness of saturated buffers in tile-drained landscapes. Results of this review documented:

  • Saturated buffers reduced annual nitrate losses by an average of 46% ± 24% at the edge of the field, corresponding to a rate of 8.4 ± 5.3 lb N/ac/y. Recent research shows that their performance may be even greater (Ghane 2025).
  • Within drainage diverted through the saturated buffer, 82% ± 22% of nitrate was removed on average.
  • Nitrogen reductions occur primarily through the process of denitrification, but plant uptake and microbial immobilization also play a role in the nitrogen dynamics within saturated buffers. In limited study, nitrous oxide losses were found to be nearly equal to traditional (unsaturated) buffers (8.4 lb N) and much less than adjacent cropland (32.4 lb N).
  • Cost-effectiveness for nitrate removal averaged $2.13 ± $1.27/lb N (ranging from $0.54 to $4.17/lb N), corresponding to $10 to $27/ac treated per year.

Figure 1. Saturated buffer operation diagram. Tile flows are routed through the control structure, with the diverted flow saturating buffer soils. Interactions between nitrate (red diamonds) and microbes contribute to denitrification and conversion to nitrogen gas (purple circles).

Contributions to the conservation drainage community

This review clearly demonstrates the value of saturated buffers as an effective practice in reducing nitrate-nitrogen loads from subsurface drainage discharge. A summary analysis of nitrogen removal performance as well as cost-effectiveness is provided.

An outline of future opportunities to advance saturated buffer research, performance and adoption is also provided. Key research areas include further exploring buffer nutrient and carbon cycling across varying contexts, sensitivity of performance to site suitability factors, and the influence of vegetation management in terms of nutrient budgets. Key areas for practice implementation and adoption include potential strategies to optimize designs and management for enhanced performance (e.g., alternative distribution pipe arrangements, optimized buffer widths, automated control structure management), as well as improved and streamlined processes to identify suitable sites for implementation. Future results from siting and design efforts should help inform existing conservation program standards (NRCS Conservation Practice Standards), while also identifying new, alternative saturated buffer applications (e.g., grass waterways, contour buffer strips). Also, innovative implementation models, such as the Batch and Build concept, can capitalize on efficiencies of scale, reducing implementation costs and improving cost-effectiveness.

Finally, to further support saturated buffer research, monitoring recommendations are outlined. These recommendations included recommending the use of 4-chamber control structures when monitoring flow rates across weir stoplogs, more intensive water sampling frequencies (particularly for phosphorus) to better understand concentration and load reductions, and the development of more robust methods to determine nutrient concentrations as discharge moves through the buffer to the adjacent stream.

by

Saturated buffer paired-field study reveals new insights into its water quality performance

Summary prepared by: Ehsan Ghane, Department of Biosystems and Agricultural Engineering, Michigan State University

Ghane, E., AbdalAal, Y., Tehrani, A. (2025). Paired-field evaluation of a saturated buffer reveals significant water-quality benefit through upstream weir management. Agricultural Water Management. 318, 109664. https://doi.org/10.1016/j.agwat.2025.109664

Ghane, E. (2025). Unveiling the true potential of saturated buffers (E3535). Michigan State University Extension Bulletin. www.canr.msu.edu/drainage

Major Findings:

We conducted a paired-field study at an on-farm site in Michigan, USA, comparing a saturated buffer (SB) system to a free drainage control field. Results showed:

  • The saturated buffer system significantly reduced nitrate-N load by 54.5% (11.1 kg/ha) annually compared to free drainage.
  • The upstream weir of the system, functioning as controlled drainage, was the primary driver behind the annual load reduction, accounting for most of the flow and nitrate-N load reduction.
  • High-level weir management (<35 cm weir depth below ground surface at structure) was more than twice as effective as low-level weir management (>35 cm weir depth below ground surface at structure), removing 0.076 kg/ha nitrate-N daily.

Contributions to the conservation drainage community

This study showed that in saturated buffer applications where a three-chamber structure is used, the upstream weir acts as controlled drainage reducing both flow and nitrate-N load in conjunction with the saturated buffer. Actual performance of saturated buffers can easily be underestimated when not acknowledging the functionality of the upstream weir. Using similarly structured paired-field studies in the future may help others explicitly quantify the influence of the entire saturated buffer system, which includes the upstream weir management as well as the flow through the buffer.

It was also clear during this study that even though the water-managed zone upstream of the control structure was only 6.6% of the total drainage area due upstream field slopes, the upstream weir management functioning as controlled drainage reduced nitrate-N load by more than 50% compared to the free drainage field. This suggests that prior assumptions around site suitability for controlled drainage requiring flat fields with slopes generally less than 1% may be overly restrictive, particularly when paired with complementary practices like saturated buffers.