Bioretetion has been shown to effectively remove heavy metals from stormwater runoff, which makes it a very useful design tool. However, there are two common concerns related to long term use of bioretention for heavy metals removal:
(1) When will breakthrough occur, i.e. when will the binding sites be saturated so that dissolved heavy metals start leaching out of the bioretention systems?
(2) When will heavy metal concentrations in the bioretention soil reach toxic levels?
Based on research to date we believe the benefits of bioretention systems typically far outweigh the concerns.
Sources of heavy metals in stormwater runoff
Heavy metals from normal pavement and vehicle wear are washed from pavements into stormwater runoff when it rains or snows. Unless they are intercepted, for example, through Stormwater Control Measures (SCMs), stormwater runoff carries these heavy metals into receiving surface waters. If allowed to reach excessively high concentrations in downstream surface water bodies, these heavy metals threaten the survival of aquatic organisms at all levels of the food chain.
Heavy metals can also be toxic to humans, and most can be lethal. At sub-lethal levels, they can negatively impact the central nervous system, lungs, kidneys, liver, blood composition, urinary system, and reproductive systems. Long term exposure can also cause cancer. For more detailed information on human health effects of each individual heavy metal, see OSHA’s website.
Heavy metals of concern typically include copper, chromium, mercury, nickel, zinc, lead, arsenic, and cadmium. Primary sources of these heavy metals in stormwater runoff include vehicle and pavement wear and maintenance, buildings, and atmospheric deposition. Building sidings are major sources of copper, zinc, lead, and cadmium; and atmospheric deposition contributes copper, cadmium, and lead (Davis et al 2001). Arsenic is commonly an ingredient in road salts. Table 1 below lists sources of the heavy metals of primary concern in roadway runoff in more detail.
|Aluminum||Natural as well as anthropogenic sources such as aluminum works industries|
|Cadmium||Tire wear, brake pads, combustion of soil, insecticides are also other sources|
|Chromium||Corrosion of welded metal plating, moving engine parts, brake lining wear|
|Cobalt||Wastes from tire and vehicle appliance manufacturing|
|Copper||Metal plating, bearing and bushing wear, moving engine parts, brake lining wear, fungicides and insecticides|
|Iron||Auto body rust, steel roadway structures, moving engine parts, corrosion of vehicular bodies.|
|Lead||Leaded gasoline, tire wear|
|Nickel||Diesel fuel and gasoline, lubricating oil, metal plating, bushing wear, brake lining wear, asphalt paving|
|Zinc||Tire wear, motor oil, grease|
Gunawardena et al 2013 found that atmospheric zinc deposition was correlated with traffic volume, and copper, cadmium, nickel, and lead, were correlated with traffic congestion.
Heavy metals removal mechanisms in bioretention
Many studies have shown that bioretention SCMs, including trees and soils, as in a Silva Cell system, are very effective at removing heavy metals from stormwater runoff. Mechanisms for heavy metal removal from stormwater runoff in bioretention SCM sinclude sedimentation, filtration, adsorption and vegetation uptake.
Sedimentation and filtration effectively remove particulate forms of heavy metals. Particulate heavy metals are therefore expected to be removed from stormwater runoff as long as the bioretention cell’s soil is not clogged (i.e. as long as it drains adequately).
Dissolved heavy metals are removed from stormwater by sorption onto soil organic matter and clay. Once the sorption capacity of a soil i ssaturated, dissolved heavy metals will “breakthrough”and dissolved heavy metals will be discharged to receiving waters. Because dissolved heavy metals are more bioavailable than particulate bound heavy metals, they may be more detrimental to aquatic ecology (Kominkove and Nabelkova 2007 in Hatt et al 2011), so long term retention of dissolved heavy metals is especially important (Hatt et al 2011).
In many studies, most heavy metals were found primarily in particulate forms in stormwater runoff (e.g. literature review in Morquecho, R. 2005), with the exception of zinc, which is found primarily in dissolved form. However, whether heavy metals are in dissolved or particulate can vary depending on many factors, including, pH, temperature, and the presence of binding sites (e.g.literature review in Morquecho 2005).
Summary of Research Related to Efficiency of Bioretention for Heavy Metals Removal
Both traditional bioretention and trees and soil below paving, as in a Silva Cell system, have been found to have high heavy metal removal rates (e.g. Davis et al 2003, Page et al 2014).
Organic matter provides binding sites for heavy metals, so it is not surprising that organic matter has been found to improve a bioretention cell’s dissolved heavy metal removal lifespan. Morgan et al (2011) compared columns with varying amounts of compost, ranging from no compost to 50% compost (by bulk volume) and found that increasing organic matter content increased time to breakthrough for dissolved cadmium and zinc (i.e. bioretention lifespan for dissolved cadmium and zinc removal). More specifically, they found that“Increasing the compost fraction from 0% to 10% more than doubles the expected lifespan for 10% breakthrough in 15 cm [6 inches] of filter media removing cadmium and zinc” (brackets added). Time to breakthrough continued to increase significantly from 10% to 30%, and from 30% to 50% compost. Copper removal also increased with increasing compost fraction. They concluded that “Based on the field study results, organic matter is the most important constituent when considering removal of dissolved toxic metals in a bioretention facility.”
Li and Davis (2008) found that metals are captured in the top 10 cm [4 inches] of the bioretention soil, and Jones and Davis (2013) also found that metals were most concentrated near the inflow point and in the top 3-12 cm [1.2 to 4.7 inches] of the bioretentionc ell.
Research Related To Lifespan of Bioretention For Heavy Metals Removal
So, given all this, when will breakthrough occur, ie. when will the binding sites be saturated so that dissolved heavymetals start leaching out of the bioretention systems, and when will heavy metal concentrations in the bioretention soil reach regulatory threshholds?
The answer is “it depends.” The following factors typically increase the time before breakthrough occurs and the time it will take for heavy metals to build up to toxic levels:
1) lower influent heavy metals concentrations (influent heavy metal concentrations vary a lot from site to site)
2) greater content of soil organic matter and other heavy metal binding sites
3) smaller contributing watershed (resulting in lower influent heavy metals concentrations, all other things being equal)
4) pre-treatment to capture sediments and associated heavy metals
That being said, studies generally indicate that at “typical” concentrations of heavy metals, with organic matter content typical for bioretention soils, and with typical contributing watershed size, the amount of time it takes for heavy metals to build up to toxic levels in bioretention soil, or to “break through,” is generally at least as long as typical useful bioretention and pavement lifespans.
When will heavy metal concentrations in the bioretention soil reach toxic levels?
Since dissolved heavy metals are more bioavailable than particulate bound heavy metals, long term retention of dissolved heavy metals is particularly important. Of all the heavy metals of concern, zinc has been shown to be the most mobile, and it is therefore typically the first heavy metal to break through. Based on column experiments using synthetic stormwater to investigate the removal and retention of cadmium, copper and zinc, Morgan et al (2011) found that, “at stormwater concentrations of zinc and cadmium, 15cm [6inches] of filter media composed of 30% compost and 70% sand will last 95 years until breakthrough, when the effluent concentration is 10% of the influent concentration” (brackets added). This 95 year lifespan is significantly longer than a typical useful pavement or bioretention lifespan, so based on this study, a typical bioretention system should not experience breakthrough of heavy metals. Moreover, with breakthrough defined as “when the effluent concentration is 10% of the influent concentration”, 90% of the heavy metals are still retained in the bioretention system at breakthrough.
When will heavy metal concentrations in the bioretention soil reach regulatory thresholds?
Several field studies indicate that at typical stormwater concentrations, concentrations of heavy metals in bioretention soils are not expected to reach regulatory thresholds for at least a few decades (see summaries below). The time span for heavy metals to accumulate to concentrations that exceed regulatory thresholds could be significantly increased with pre-treatment to capture suspended solids.
Even if high levels of heavy metals are captured in bioretention cells, this is typically still preferable to the alternative of no treatment: piping the heavy metals to surface water bodies. Heavy metals are more easily cleaned out of bioretention cells than out of surface waters. Where bioretention cells are designed under pavement, for example, with Silva Cells, it is more difficult to access the soil if/when soil replacement is desired, BUT, at least as importantly, the risk of human ingestion of bioretention soil under pavementi s also much smaller than the risk with open bioretention cells (i.e. not under pavement). These heavy metal concentrations are not typically harmful to trees.
Results from representative studies estimating bioretention lifespan for heavy metals removal based on soil concentrations are summarized below.
The Toronto and Region Conservation Authority (2008) studied a 3 year old demonstration project as well as 7 older pervious interlocking concrete paver (PICP) sites (ranging from 4 to 17 years old) and 5 bioswales (ranging from 2 to >18 years old). Soil quality results (copper, zinc, lead, iron) from older PICP and bioswale sites indicated that land fill disposal or remediation of the underlying soils would typically not be required when the pavers or swales need to be replaced. The concentrations of metals in the bioswale cores were below background concentrations for agricultural soils. Metal concentrations in soils at the demonstration site in 2005 and 2007 were similar, indicating little if any accumulation of metals in surface soils over a two year period.
Jones and Davis (2013) assessed accumulated lead, copper, and zinc media samples from a 4-year-old bioretention cell. They found that “After 4 years of operation, total metal concentrations are well below the regulatory cleanup thresholds stipulated by the Maryland Department of the Environment (MDE) and the U.S.EPA. Significant capacity for metal accumulation, at least on the order of several decades, is estimated to remain … The media sequential extraction results indicate that accumulated metal will remain largely sequestered within the cell rather than become mobilized into the environment. [italics added]”
Davis et al (2003) studied the effect of constructed bioretention boxes and 2 field scale bioretention cells on copper, lead and zinc removal. United States hazardous waste disposal legislation does not yet include regulations for BMP sediment and soil. However, several states are beginning to adopt the US EPA Part 503 biosolids regulations to regulate stormwater BMP sediment disposal. These regulations limit cumulative metal loadings allowed through the application of wastewater biosolids. Several researchers have used these regulations to evaluate metal accumulation in bioretention soils. Davis et al (2003) found that “after 20 years, cadmium, lead and zinc accumulations reach or exceed regulatory limits for biosolids application (U.S. EPA,1993). The time required for metal accumulations to reach these limits are 20, 77, 16, and 16 years for cadmium, copper, lead, and zinc, respectively.”
They also evaluated the media with respect to hazardous waste classification criteria based on allowable toxicity characteristic leaching procedure (TCLP) concentrations.The EPA developed the Toxicity Characteristic Leaching Procedure (TCLP) to determine whether a waste may be accepted into a typical municipal landfill (as defined by RCRA Subtitle D) based on its potential to leach dangerous concentrations of toxic chemicals into ground water. If TCLP analytical results are above the TCLP D-list maximum contamination levels (MCLs), the waste cannot be accepted at a typical municipal landfill and must be taken to a hazardous waste disposal facility. Only cadmium and lead have TCLP limits. Results from the bioretention media study by Davis et al (2003) showed that “After 20 years, accumulated levels of these two metals approach, but do not exceed, the allowable TCLP extraction levels. The time required to reach the TCLP limits are 50 years for cadmium and 26 years for lead. This example, however, assumes that all metals are readily available and will be extracted in the TCLP; it is, thus,conservative. Greater metal accumulation could occur before TCLP limits are reached.”
Li and Davis(2008) studied lead, copper, and zinc removal in a bioretention cell 3 years after installation and again 4 years after installation and found that “the captured metals exhibited a strong association with the media, suggesting that they are not washed out by subsequent wet weather flows.” Lead was the limiting metal in bioretention accumulation and exceeded regulatory values for heavy metals in residential soils (based on mean of values of 30 states in the US, typically based on child soil exposure at home). This lead was tightly bound to the soil and not likely to be washed out.
Research has shown that on an average site, breakthrough is not likely to occur within the useful lifespan of pavement.
Time to breakthrough can be increased by increasing soil organic matter to provide more binding sites in the soil. Time before heavy metal concentrations in the bioretention soil reaches regulatory thresholds will vary a lot depending on influent concentrations. Several field studies indicate that at typical stormwater concentrations, concentrations of heavy metals in bioretention soils are not expected to reach regulatory thresholds for at least a few decades. The timespan for heavy metals to accumulate to regulatory thresholds could be significantly increased with pretreatment to capture suspended solids. Even if toxic metals do accumulate to regulatory thresholds, covering bioretention soil with pavement, such as in a Tree in Soil under Suspended Pavement, will minimize the chance of human ingestion of the soil.
Davis, A. P., Sholouhian, M., and Ni, S. (2001). Loadings of lead, copper, cadmium, and zinc in urban runoff from specific sources. Chemosphere, 44(5), 997–1009.
Davis, A. P. , Shokouhian, M., Sharma, H., Minami, C., and Winogradoff, D. (2003). Water quality improvement through bioretention: Lead, copper, and zinc removal. Water Environment Research, 75(1), 73–82.
Hatt, B.E., Steinel, A., Deletic, A., and Fletcher, T.D. (2011). “Retention of heavy metals by stormwater filtration systems: Breakthrough analysis.” Water, Science, and Technology. 64(9), 1913-1919.
Li, H. and Davis, A.P. (2008). “Heavy metal capture and accumulation in bioretention media.” Environmental Science & Technology. 42, 5247-5253.
Jones, P. and Davis, A. (2013). ”Spatial Accumulation and Strength of Affiliation of Heavy Metals in Bioretention Media.” J. Environ. Eng., 139(4), 479–487.
Kominkova, D. and J. Nabelkova. (2007). Effect of urban drainage on bioavailability of heavy metals in recipient. Water Science and Technology 56(9), 43–50.
Morgan, J.G., K.A. Paus, R.M. Hozalski and J.S. Gulliver. (2011). Sorption and Release of Dissolved Pollutants Via Bioretention Media. SAFL Project Report No. 559, September 2011. http://purl.umn.edu/116560.
Morquecho, R. 2005. Pollutant Associations With Particulates In Stormwater. Thesis. Doctor of Philosophy in the Department of Civil and Environmental Engineering in the Graduate School of The University of Alabama.
Page, J. L., R. J. Winston, and W.F. Hunt, III. (2014). Field Monitoring of Two Silva Cell™ Installations in Wilmington, North Carolina: Final Monitoring Report.
Torno, H.C. (1994). Storm Water NPDES Related Monitoring Needs. Proceedings of an Engineering Foundation Conference. ASCE, New York, 10017-2398.
Toronto and Region Conservation. (2008). Performance Evaluation of Permeable Pavement and a Bioretention Swale, Seneca College, King City, Ontario.
Nathalie Shanstrom is a sustainable landscape architect with The Kestrel Design Group.