Lessons Learned from Bioretention Research

UPDATES: January 2026 | Volume 21, Issue 1

Written by John Gulliver & Andy Erickson 

After more than 20 years of bioretention (rain garden) research, we thought that we would take a step back, to see what we have learned over the years.  Many of the lessons are about the details of our research techniques, but there are a few that we believe any stormwater professional would appreciate.  These are in the following categories:

• Bioretention should take runoff from impervious surfaces in order to treat it.
• There are two very different treatment practices in the bioretention category, with different outcomes and concerns.
• Compost can release nutrients
• Plants maintain or increase soil porosity
• Infiltration rates vary greatly in one bioretention practice
• Uncertainties in design need to be addressed

First, what are bioretention practices?  The U.S. Environmental Protection Agency states: “Bioretention practices, such as rain gardens, are landscaped depressions that treat on-site stormwater discharge from impervious surfaces such as roofs, driveways, sidewalks, parking lots and compacted lawns. They are used to collect stormwater and filter it through a mixture of soil, sand and/or gravel.” (https://www.epa.gov/system/files/documents/2021-11/bmp-bioretention-rain-gardens.pdf) This definition is inclusive of bioswales and most infiltration basins, and we will treat it as such (Figure 1).

Impervious Surfaces and Bioretention. Bioretention practices have been developed to treat water that comes off impervious surfaces, to avoid runoff directly into the receiving storm sewer or water body.  We have seen situations where a bioretention practice is put in areas with a large grassed area, where it typically would receive little runoff and be of minimal use.

Biofiltration versus Bioinfiltration. While looking similar from the surface, these are two different types of bioretention. Biofiltration involves filtration through engineered media and quick return to stormwater system via a perforated pipe underdrain. Many biofiltration practices also have the ability to infiltrate runoff, but the relatively short residence time of biofiltration to an underdrain typically means that most of the water will be sent back into the sewer system (Erickson et al., 2013) unless the underdrain is elevated within the storage bed. Bioinfiltration involves filtration through engineered media and then through porous media of vadose zone and surface aquifers (Figure 2). 

Compost in Bioretention Practices. Compost captures metals, bacteria, and many organic chemicals (Paus et al., 2014; LeFevre et al., 2015), but it also can release dissolved nutrients such as phosphate, nitrogen, and carbon compounds (Erickson, et al., 2022). This is not typically a concern in a bioinfiltration practice, because of the long residence time and nutrient uptake and sorption in the natural soil.  In a biofiltration practice, however, nutrient release is a concern because the underdrain typically will export a higher concentration of nutrients than the runoff that enters the biofiltration practice (Figure 3). If you are concerned about nutrient pollution in your receiving water body, then cap those underdrains. You can go out and measure surface water level 48 to 72 hours after a storm to see if your bioretention practices are draining properly. If one is concerned with nutrients, biofilters need to be designed with a reduction of compost.  A new biofilter media needs to be developed so that the compost can be reduced as far as possible.

Plants maintain or increase soil porosity. Plants not only transpire water to get ready for the next storm, but their roots also affect soil structure in a positive manner (Figure 4).  This results in many bioretention practices in which infiltration rates are constant or increase over time (Paus et al., 2013).

Infiltration rates vary greatly in one bioretention practice. Due to variable compaction of the soil and the presence of plants that open up the soil, we have found that infiltration rates can vary by 2 orders of magnitude (i.e., 100x) in one infiltration practice (Asleson, et al. 2009; Ahmed et al., 2015).  To assess the drainage of an infiltration practice, you need 20 measurements at different locations to determine the overall average infiltration rate to within a factor of two, and 10 measurements to determine the infiltration rate within a factor of three (Ahmed et al. 2015).  A better means of assessing bioretention infiltration is to fill the practice with a water truck or from a fire hydrant to see how it drains, with costs that are similar to the infiltration rate measurements (Figure 5).  You can also install pressure probes and soil moisture probes to assess infiltration under natural rainfalls, with costs that are approximately double the first two (Gulliver and Erickson 2024).

Uncertainties in design that need to be addressed. The primary uncertainty in the design of bioretention practices is whether the practice will drain properly into the underlying soils. A methodology using permeameters below the ground surface and at the future elevation of the native soil, needs to be developed for bioretention practice.  Soil texture can be assessed with soil cores, but that doesn’t say anything about soil structure.  A permeameter can estimate the infiltration rate at the depth to which the bioretention practice will be dug (Figure 6). There is no protocol developed for the use of these permeameters to estimate infiltration rates in future bioretention practices.

Conclusions

There is a need to differentiate between biofiltration and bioinfiltration systems. Bioretention systems have been with us for over 30 years, and are sensitive to leaching because they are filtering water and returning it to the stormwater system, but there has not been sufficient research performed on these practices.  Compost was chosen as a bioretention media by default.  It is only recently that we have become aware of the ramifications of this decision for biofiltration systems.  We need to investigate new media for biofiltration systems if we are concerned with nutrient leaching.

References

• Ahmed, F., J.S. Gulliver and J.L. Nieber (2015). Field Infiltration Measurements in Grassed Swales, Journal of Hydrology, 530, 604–611.
• Asleson, B.C., R.S. Nestingen, J.S. Gulliver, R.M. Hozalski, and J.L. Nieber (2009). Assessment of Rain Gardens by Visual Inspection and Controlled Testing,  Journal of the American Water Resources Association, 45(4), 1019-1031.
• Erickson, A.J., P.T. Weiss and J.S. Gulliver, (2013). Optimizing Stormwater Treatment Practices: A Handbook of Assessment and Maintenance, Springer, New York, NY.
• Erickson, A.J., J.L. Kozarek, K.A. Kramarczuk, and L. Lewis (2022). “Biofiltration Media Optimization: Final Report.” Project Report No. 603, St. Anthony Falls Laboratory, University of Minnesota, Minneapolis, MN. https://hdl.handle.net/11299/253917.
• Gulliver, J.S. and Erickson, A.J. (2024). Measuring Infiltration within Green Stormwater Infrastructure (GSI) Filtration and Infiltration Basins: Years 1 – 3, Proj. Rep. 606, St. Anthony Falls Laboratory, University of Minnesota, Minneapolis, MN. https://hdl.handle.net/11299/263948
• LeFevre, G.H., K.H. Paus, P. Natarajan; J.S. Gulliver, P.J. Novak, and R.M. Hozalski. (2015). "A Review of Dissolved Pollutants in Urban Stormwater and their Removal and Fate in Bioretention Cells." Journal of Environmental Engineering, 141 (1). http://dx.doi.org/10.1061/(ASCE)EE.1943-7870.0000876.
• Paus, K.H., J. Morgan, J.S. Gulliver, T. Leiknes and R.M. Hozalski (2013). Assessment of the Hydraulic and Toxic Removal Capacities of Bioretention Cells after 2 to 8 Years of Service, Water, Soil and Air Pollution, 225 (1803).
• Paus, K.H., J. Morgan, J. S. Gulliver, and R. M. Hozalski, (2014). Effects of Bioretention Media Compost Fraction on Toxic Metals Removal, Hydraulic Conductivity, and Phosphorous Release, Journal of Environmental Engineering, 140(10), 04014033.