Overview and management strategies for bacteria in stormwater

Dog waste sign

In water quality monitoring, specific disease-producing (pathogenic) organisms are not easily identified. Testing for them is difficult, expensive, and time-consuming. Instead, fecal coliforms and Escherichia coli (E. coli), two closely related bacteria groups, can indicate the presence of pathogens. Fecal coliform and E. coli found in Minnesota rivers and streams may come from human, pet, livestock, and wildlife waste and are more common in heavily populated or farmed areas. Bacteria may reach surface water through malfunctioning or illicit septic system connections, urban stormwater, manure spills or runoff, and more (Minnesota Pollution Control Agency website, accessed January 25, 2018).

This page provides information on bacteria in urban stormwater, including a discussion of sources of bacteria and management strategies for minimizing bacteria loading from urban stormwater runoff to surface water. Note that the focus is on bacteria because bacteria are used as a surrogate for assessing potential contamination by pathogenic microorganisms. A short section on this page specifically discusses pathogens and their relationship to indicator bacteria.

Source and concentrations of bacteria in urban stormwater

Ultimately the source of bacteria in urban stormwater is animal waste. Identifying the specific source is more challenging and likely varies with location and land use. Typical sources include domestic pets and wildlife, particularly birds. Sources of bacteria to receiving waters include urban stormwater runoff, leaking sewer lines, sewer overflows, septic systems, landfills, marinas and pumpout facilities, poorly operating packing plants, and other illicit discharges.

Some general observations from the literature are summarized below.

• Fecal coliform levels are considerably lower (about 90 percent lower) in runoff that occurs in winter compared to summer months (Selvakumar and Borst, 2006)
• Bacteria levels can increase sharply during snowmelt events
• Bacteria concentrations in runoff increase as percent imperviousness increases up to about 20 percent impervious, but are unaffected by further increases in impervious surfaces
• Residential lawns, driveways, and streets are the major source areas for bacteria, while rooftops and parking lots are usually smaller source areas. Irrigated lawns, in particular, are high contributors
• Sartor and Gaboury (1984) reported nearly 92 percent of the bacteria originated from streets in the residential-institutional land-use site, whereas only about 33 and 19 percent of the bacteria originated from streets in the industrial and commercial land-use sites
• Bannerman et al. (1993) reported that 78 percent of the fecal coliform bacteria load for one of the same residential land-use study subbasins studied by Waschbusch et al. (1999) originated from streets
• In areas with pet waste ordinances and education, pet wastes are likely to be a minor contributor to bacteria (Burnhart)

Land use	Median (MPN/ml)	Maximum (MPN/ml)
Commercial	6900	350000
Industrial	9700	290000
Residential	20000	600000
Rooftop	1250
Open space	4500
Forested	< 100

Sewage typically contains fecal coliform concentrations in excess of one million most probable number per 100 milliliter (MPN/100 ml). This is about two orders of magnitude greater than urban stormwater concentrations. General indicators of bacteria sources include the following.

• Higher concentrations in baseflow compared to stormwater runoff indicate sewage or other illicit discharges
• Elevated concentrations in winter indicate sewage or other illicit discharges
• Elevated concentrations during dry weather conditions indicate sewage or other illicit discharges or runoff from lawn irrigation
• Very high concentrations indicate sewage or other illicit discharges

A complicating factor is that bacteria can survive and grow both within the storm sewer system and within receiving waters. Growth within the storm sewer systems includes both the surface and subsurface conveyances. For example, coliform bacteria have been found to survive and grow in moist soils and leaf piles. A recent study in Minneapolis indicated that catch basins are an important source, largely as a result of growth within the catch basin.

The following table provides a summary of data from the literature. Maximum concentrations are included to illustrate the tremendous variability that may occur in bacteria concentrations. The values represent a compilation of data from several sources (see references at the bottom of this page).

Relationship between bacteria and pathogens in stormwater

Although pathogens are the primary concern in stormwater, they are difficult and expensive to sample. Thus, as stated above, indicator bacteria are used as a surrogate for the presence of pathogens.

Pathogens are typically found in stormwater runoff, but at relatively low concentrations. Geometric mean densities for some pathogens in urban stormwater runoff and urban streams are summarized below.

• Enterovirus: 6.9 and 170 plaque-forming units per 10 liters
• Salmonella sp.: 30 and 5.7 most probable number per 10 liters
• Pseudomonas aeruginosa: 2000 and 590 most probable number per 10 liters
• Staphylococcus aureus: 120 and 12 most probable number per 100 milliliters

Epidemiological studies show mixed results, ranging from no observed impacts to significant impacts for people exposed to water receiving urban stormwater runoff.

Wu et al. (2011) studied the correlation between several microbial indicators and pathogens in water samples. Some conclusions include the following.

• Indicator organisms are possibly correlated with pathogens if sufficient data are available
• Indicator organisms cannot signal the presence of pathogenic contamination for a given water sample
• Specific coliphages are better indicators for viral pathogens
• C. perfringens, total and fecal coliforms are likely useful indicators for all three biotypes of pathogens
• E. coli and enterococci did not show any greater likelihood of correlating with pathogens than other indicators
• The presence of E. coli and enterococci in water generally indicates fecal contamination and thus a health risk, regardless of whether or not specific pathogens are observed

Meeting bacteria water quality targets

Information on this page can be used to help meet water quality targets. Water quality targets are established for various purposes including meeting Clean Water Act (CWA) requirements, meeting local water quality goals or requirements, and meeting non-regulatory targets. CWA requirements include antidegradation, TMDL limits, and NPDES permit requirements. Each of these are described below.

Information: Note that information presented in the Stormwater Manual can be used to meet NPDES permit requirements. This includes information on all BMPs discussed in the Manual unless otherwise noted. Check with MPCA's Stormwater Program for applicability of information not contained in the Manual, including BMPs and BMP credits.

Antidegradation

Water quality standards include an antidegradation policy and implementation method. The water quality standards regulation requires States and Tribes to establish a three-tiered antidegradation program to protect existing water quality and water uses in receiving waters (see [1]).

There are no specific antidegradation requirements applicable to bacteria. Compliance with Minimum Control Measure (MCM) 5 of the MS4 permit constitutes compliance with antidegradation requirements. The permit requires no net increase in discharges of volume, total phosphorus (TP) and total suspended solids (TSS) for new development, while a reduction in these are required for redevelopment projects covered under the permit. Practices that infiltrate or capture and reuse stormwater runoff are typically used to meet these permit requirements because they help meet the volume requirements. Reductions in volume and/or TSS loads would typically be associated with reductions in bacteria loads.

Total Maximum Daily Loads (TMDLs)

Map showing 2018 U.S. EPA-approved stream and river stretches impaired for either e coli or fecal coliform.

The 2018 impaired water list includes 7 beach impairments for e coli, 304 river or stream stretches impaired for e coli, and 53 river or stream stretches impaired for fecal coliforms. Click here to link to MPCA's impaired waters website. A map illustrating U.S. EPA-approved listings for e coli and fecal coliforms is shown on the right.

The MS4 permit requires permittees to demonstrate progress toward meeting applicable Wasteload Allocations in U.S. EPA-approved TMDLs. General information on meeting TMDL requirements in NPDES permits is found here, while reporting requirements are found here. Below is additional information that may be useful.

• Forms and guidance for TMDLs: includes information on completing the TMDL Annual Report form and other guidance documents.
• Information on modeling, including Available stormwater models and selecting a model and Detailed information on specific models.
• Information on pollutant removal by BMPs. Pollutant removal information is limited to structural BMPs. Each BMP included in the manual also has a section on credits, which provides useful information for determining pollutant removal for TP for that BMP.

Permittees with required reductions in bacteria loading should consider implementing a treatment train approach, which is discussed in greater detail below.

Stormwater management for bacteria

Watershed scale stormwater management approach using a multi-BMP approach to managing the quantity and quality of stormwater runoff. The BMP sequence starts with pollution prevention and progresses through source control, on-site treatment, and regional treatment before the runoff water is discharged to a receiving water. On-site and regional practices treat stormwater runoff and can be incorporated into a stormwater treatment train.

Management of urban stormwater to control or reduce bacteria concentrations and loading should focus on identifying the most important sources and employing specific practices to address those sources. If significant reductions in bacteria loading are required or desired, a treatment train approach should be utilized. The treatment train approach for bacteria focuses on implementing the following hierarchy of practices:

• pollution prevention and source control
• pretreatment for structural BMPs
• infiltration
• settling
• filtration

Construction stormwater

Limited data exists on bacteria loads associated with construction site runoff. The Minnesota Construction Stormwater General Permit does not have specific requirements for bacteria, but requires control of sediment. Sawyer et al. (2010) found Escherichia coli (E. coli) concentrations from construction site runoff (mean = 771 Most Probable Number per 100 milliliters (MPN/100 ml) were consistently and significantly higher than water quality criteria established by the US Environmental Protection Agency for recreational waters. Results, however, were variable. Basin discharges showed significantly higher bacterial concentrations (mean = 1,368 MPN/ 100 ml) than flows coming directly from construction sites. Within sediment basins, both mean water column (877 MPN/100 ml) and mean sediment (188,828 MPN/100 ml) E. coli densities were higher than recommended EPA criteria, with mean concentrations in sediments significantly exceeding the corresponding overlying water column.

Practices that control sediment will reduce bacterial loads. For information on construction site best management practices, link here.

Education and outreach

Education and outreach activities may include brochures, posters, websites, event attendance, utility bill inserts, television advertisements, articles in homeowner association newsletters and other approaches that reach citizens and promote behavioral changes. Campaigns to reduce water waste by reducing over-irrigation help communities to meet both conservation and water quality objectives (Urban Water Resources Research Council, 2014).

Repair of aging infrastructure and correcting illicit connections

The Urban Water Resources Research Council (2014) states "Aging and leaking sanitary sewer and stormwater conveyance pipes can introduce pollutants to the MS4 through SSOs caused by blockages, line breaks, or other sewer defects; exfiltration of sewage from sanitary sewers; and infiltration of groundwater when the MS4 lies below the water table (Sercu et al. 2011). Upgrading, repairing, or slip-lining faulty sanitary sewer pipes will reduce pollutant loads by eliminating the leaks in those pipes. Additionally, upgrading or repairing storm drain pipes can minimize the infiltration of contaminated groundwater into the MS4 (Geosyntec 2012)." More detailed information on the effects and management of aging infrastructure and illicit discharges can be found in Section 7.2 (pages 145-148) of the Urban Water Resources Research Council report.

Pollution prevention

These practices reduce the amount of bacteria generated, remove bacteria prior to it being entrained in runoff. These are summarized below for residential, municipal, and industrial sources.

Prevention practices for residential areas

The following table summarizes residential prevention practices that are effective at reducing bacteria concentrations. The table indicates the relative effectiveness of each practice and provides a short description of the practice. Bacteria removal efficiencies are not established for these BMPs.

Residential pollution prevention methods effective for controlling or reducing bacteria. Source: modified from the Center for Watershed Protection.
Link to this table

Practice	Relative effectiveness	Method	Image1
Litter and Animal Waste Control	High	Properly dispose of pet waste and litter in a timely manner and according to local ordinance requirements.
Yard Waste Management	Low	Prevent yard waste from entering storm sewer systems and water bodies by either composting or using curbside pickup services and avoiding accumulation of yard waste on impervious surfaces; keep grass clippings and leaves out of the street.
Septic Tank Maintenance	High
Exposed Soil Repair	Low	Use native vegetation or grass to cover and stabilize exposed soil on lawns to prevent sediment wash off.
Native Landscaping	Low	Reduce turf areas by planting native species to reduce and filter pollutant-laden runoff and prevent the spread of invasive, non-native plant species into the storm sewer system.
Healthy Lawns	Low	Maintain thick grass planted in organic-rich soil to a height of at least 3 inches to prevent soil erosion, filter stormwater contaminants, and absorb airborne pollutants; limit or eliminate chemical use and water and repair lawn as needed
Proper lawn irrigation and watering2	Medium-high	Over-watering lawns has been shown to be an important source of bacteria to streets and sidewalks. Implement appropriate watering practices to avoid runoff from pervious surfaces where animals (e.g. birds, mammals) are active.	3

¹ Photo credits
²For tips on proper watering, see this page
³Image courtesy Chesapeake Lawn Service

Prevention practices for municipalities

The following table summarizes municipal prevention practices that are effective at reducing bacteria concentrations. The table indicates the relative effectiveness of each practice and provides a short description of the practice. Bacteria removal efficiencies are not established for these BMPs.

Municipal pollution prevention methods effective for controlling or reducing bacteria. (Source: modified from the Center for Watershed Protection).
Link to this table

Practice	Effectiveness	Method	Image1
Dumpster and Landfill Management	High	Ensure that contaminated material is contained to prevent solid and/or liquid waste from being washed into storm sewer systems or water bodies.
Sanitary Sewer System Maintenance	High	Regularly inspect and flush sanitary pipes to ensure that there are no leaks in the system and that the system is properly functioning.
Litter and Animal Waste Control	High	Mandate litter and pet waste cleanup within the community and control waste-generating wildlife, such as geese; provide waste containers for litter and pet waste in public areas.
Public Educations	Moderate	Label storm drains to indicate that no dumping is allowed and institute pollution prevention programs to educate and implement needed community practices.
Staff and Employee Educations	Moderate	Provide internal training for staff and provide direction to hired employees or volunteers regarding pollution prevention techniques to be used during work activites.

¹ Photo credits

Prevention practices for industrial sources

The following table summarizes industrial prevention practices that are effective at reducing bacteria concentrations. The table indicates the relative effectiveness of each practice and provides a short description of the practice. Bacteria removal efficiencies are not established for these BMPs.

Industrial & commercial pollution prevention practices for bacteria.
Link to this table

Practice	Effectiveness	Method	Image
Dumpster and Landfill Management	High	Ensure that contaminated material is contained to prevent solid and/or liquid waste from being washed into storm sewer systems or water bodies.
Sanitary Sewer System Maintenance	High	Regularly inspect and flush sanitary pipes to ensure that there are no leaks in the system and that the system is properly functioning.

Street sweeping

Several articles in the literature present results from street sweeping studies. Examples include the following.

• Zarriello et al. (2002) estimated street sweeping reduced bacteria loads to the Lower Charles River in Massachusetts by 7 to 17 percent. Removal rates were 5 percent for wet vacuum, 20 percent for regenerative air, and 50 percent for dry vacuum sweepers.
• Selbig and Bannerman (2007) discuss changes in debris loading for regenerative-air, vacuum-assist, high-frequency broom, and low-frequency broom sweeping practices. However, bacteria is not discussed.
• Law et al. (2008) found for a given set of assumptions and sweeping frequencies, it is expected that the range in pollutant removal rates from street sweeping for total solids was 3 to 8 percent, with the lower end representing monthly street sweeping by a mechanical street sweeper and the upper end the pollutant removal efficiencies using regenerative air/vacuum street sweeper at weekly frequencies. However, bacteria is not discussed.
• Sutherland (2011) provides a comprehensive summary of street sweeping, including information on effectiveness of different sweepers and factors affecting the performance of street sweeping. However, bacteria is not discussed.

Bird and mammal control

Birds are an important source of bacteria, particularly if roof runoff or runoff from irrigated lawns is routed to the stormwater conveyance system. Table 7.2 (page 152) of the Urban Water resources Research Report (2014) provides a summary of methods for controlling waterfowl populations. Table 7.3 (page 156) in the same report provides a summary of control practices for pigeons.

For mammals, the Urban Water resources Research Report (2014, see Section 7.8, page 157)) includes the following recommendations.

• Develop a wildlife management plan, working with city wildlife conservation staff and/or state division of wildlife
• Modify habitat and reduce urban food sources
• Install storm drain inlet/outlet controls through grates and trash rack
• Clean out storm drains to remove animal waste. When storm drains are power-washed (“jetted”), it is important the discharge be collected by a vacuum truck, otherwise, pollutants are simply flushed into the receiving water.
• Relocate wildlife by trapping

Pretreatment

Pretreatment is needed to protect infiltration and filtration BMPs from the build-up of trash, gross solids, and particulate matter. When the velocity of stormwater decreases, sediment and solids drop out. If pretreatment is not provided, this process will occur in the infiltration or filtration cell, resulting in long-term clogging and poor aesthetics. Therefore, pretreatment is a required part of the design for infiltration and filtration BMPs.

There are three typical methods for pretreatment: vegetated filter strips (VFS), forebays, and vegetated swales. Vegetated swales and filter strips are typically not effective at reducing bacteria loads. Properly constructed and maintained forebays can effectively reduce bacteria loading as a result of sediment retention.

Maintenance

Maintenance of the storm sewer system and structural BMPs removes trash and other debris that may contain bacteria and pathogens and removes sediments that can act as sources for microbial growth. Maintenance should be targeted in areas where bacteria loads are likely or known to be greatest, such as residential areas. Maintenance includes more frequent cleaning and use of more efficient technology, such as vacuum cleaning. Some maintenance activities include the following.

• Storm sewer cleaning
• Catch basin cleaning
• Maintenance of structural BMPs

Infiltration

Infiltration practices are structural Best Management Practices (BMPs) designed to capture stormwater runoff and allow the captured water to infiltrate into soils underlying the BMP. Infiltration BMPs are designed to capture a particular amount of runoff. For example, the construction stormwater permit requires that post-construction BMPs capture the first inch of runoff from new impervious surfaces, assuming there are no constraints to infiltration. BMPs designed to meet the construction stormwater permit are required to infiltrate captured water within 48 hours, with 24 hours recommended when discharges are to a trout stream.

Bacteria removal is assumed to be 100 percent for all water that infiltrates. Any water bypassing the BMP does not receive treatment. Examples of infiltration BMPs, with links to appropriate sections in the Manual, include the following.

• Infiltration (infiltration basin, infiltration trench, dry well, underground infiltration)
• Bioinfiltration (rain garden or bioretention with no underdrain)
• Permeable pavement
• Tree trench/tree box
• Dry swales with check dams

Additional BMPs that result in infiltration include stormwater and rainwater harvest with irrigation and impervious surface disconnection. For these BMPs infiltration typically occurs into turf or other vegetated areas. Disconnection of impervious surface does not qualify for credits for meeting the Construction Stormwater permit. Harvest BMPs do qualify for credit because they capture an instantaneous volume of water.

The links above take you to the main page for each BMP. Each BMP section has a page on pollutant credits. These credit pages provide information on runoff volume and pollutant removal for the BMP, including credits that can be applied to meet a performance goal such as a Total Maximum Daily Load (TMDL).

In soils where there are constraints on infiltration, BMPs may be designed with underdrains. Unless the BMP is lined, some water will infiltrate through the bottom and sides of the BMP. Bacteria removal for the portion of captured runoff that infiltrates is 100 percent. Water draining to the underdrain undergoes some treatment. These BMPs are discussed in more detail in the filtration section below.

Settling practices

Information: Note that we refer to constructed ponds and constructed wetlands in this section. Natural ponds and wetlands are not stormwater treatment practices and are therefore not included in this discussion.

If prevention, source control and infiltration practices cannot fully meet protection or restoration targets for stormwater, settling and filtration practices may be used. Settling practices include constructed stormwater ponds, including variants, and constructed stormwater wetlands, including variants. Manufactured devices and forebays are both settling practices but are primarily used for pretreatment.

Information on design, construction, operation and maintenance, credits, and other characteristics of these BMPs can be found on the main pages for constructed stormwater ponds and constructed stormwater wetlands.

Filtration practices

Filtration practices are typically used when infiltration practices are not feasible, such as areas with low infiltration soils or shallow bedrock (see section on infiltration constraints. Filtration practices include bioretention with underdrains, media filters, and swales. Vegetated filter strips are often used as a pretreatment practice.

Information on design, construction, operation and maintenance, credits, and other characteristics of these BMPs can be found on the main pages for media filters and swales, green roofs, and bioretention.

References

• Bannerman, R.T., Owens, D.W., Dodds, R.B., and Hornewer, N.J.. 1993. Sources of pollutants in Wisconsin stormwater. Water Science Technology, v. 28, no. 3-5, p. 241–259.
• Burnhart, Matt, (undated). Sources of bacteria in Wisconsin stormwater: Madison, WI. Wisconsin Department of Natural Resources. 34 p.
• Law, N.L., DiBlasi, K., and U. Ghosh 2008. Deriving Reliable Pollutant Removal Rates for Municipal Street Sweeping and Storm Drain Cleanout Programs in the Chesapeake Bay Basin. Center for Watershed Protection.
• McLellan, S.L., E. Jensen Hollis. 2006. Bacteria Sources and Fate Report. Bacteria Source, Transport and Fate Study - Phase 1, Volume 3.122 p.
• Sartor, J.D., and Gaboury, D.R.. 1984. Street sweeping as a pollution control measure—Lessons learned over the past ten years. Science of the Total Environment. v. 33, p. 171–183.
• Sawyer, C.B., J.C. Hayes, and W.R. English. 2010. Characterization of Escherichia Coli for Sediment Basin Systems at Construction Sites. World Environmental and Water Resources Congress 2010, May 16-20, 2010 | Providence, Rhode Island, United States.
• Schueler, T. 2000. Microbes and Urban Watersheds: Concentrations, Sources, & Pathways. Watershed Protection Techniques. 3(1): 554-565
• Selbig, W.R. and R. T. Bannerman. 2007. Evaluation of Street Sweeping as a Stormwater-Quality-Management Tool in Three Residential Basins in Madison, Wisconsin. USGS Scientific Investigations Report 2007–5156.
• A. Selvakumar, and M. Borst. 2006. Variation of microorganism concentrations in urban stormwater runoff with land use and seasons. J Water Health, 4, 109-124.
• Sutherland, R. 2011. Street Sweeping 101. Stormwater. January-February 2011.
• Tiefenthaler, L., E. D. Stein, and K.C. Schiff. 2011. Levels and patterns of fecal indicator bacteria in stormwater runoff from homogenous land use sites and urban watersheds. Journal of Water and Health. Vol 09.2:279-290.
• United States Geological Survey. 1998. Urban Stormwater Quality, Event-Mean Concentrations, and Estimates of Stormwater Pollutant Loads, Dallas-Fort Worth Area, Texas, 1992–93. Water-Resources Investigations Report 98-4158.
• Urban Water Research Council. 2014. Pathogens in Urban Stormwater Systems. UWRRC Technical Committee Report. 289 p.
• Waschbusch, R.J., Selbig, W.R., and Bannerman, R.T.. 1999. Sources of phosphorus from two urban residential basins in Madison, Wisconsin, 1994–95. U.S. Geological Survey Water-Resources Investigations Report 99-4021, 47 p.
• Wu, J., S. C. Long, D. Das, and S. M. Dorner. 2011. Are microbial indicators and pathogens correlated? A statistical analysis of 40 years of research. Journal of Water and Health. 09.2:265-278.
• Zarriello, P. J., R. F. Breault, and P. K. Weiskel. 2002. Potential Effects of Structural Controls and Street Sweeping on Stormwater Loads to the Lower Charles River, Massachusetts. Water-Resources Investigations Report 02-4220. 50 p.