Water quality characterization of bridge deck runoff in NC

Science Center Objects

There is evidence that bridge deck runoff has a relatively high loading of a variety of constituents such as nutrients, solids, pesticides, metals, and polycyclic aromatic hydrocarbons (PAHs). Information on the quality of bridge deck runoff in North Carolina is, however, lacking. Stormwater permits are designed to reduce nonpoint source loadings of anthropogenically derived constituents to surface waters. Permits for bridges in NC, however, must be based on data collected from studies which were conducted 10 – 20 years ago and in other parts of the U.S. As a result permit requirements for bridges in NC may be unnecessarily conservative or inadequate for protecting receiving water quality.

The primary objective of this investigation is to identify the loading of selected constituents in stormwater runoff from representative bridges across North Carolina. Working collaboratively, NCDOT, DWQ and USGS identified other study objectives which could provide information valuable in helping understand the effects of bridge deck runoff on receiving water quality and in managing stormwater runoff from bridges.

 

Background

I25 bridge from bottom

Bridge deck runoff collection system on US-64 bridge over Hiwassee River in Cherokee County, North Carolina.

(Public domain.)

On July 1, 2008, the North Carolina General Assembly passed House Bill 2436, Session Law 2008-107, Stormwater Runoff from Bridges Section 25.18.(a,b,c).  This bill requires the North Carolina Department of Transportation (NCDOT) to study 50 bridges to (1) quantify the constituents in stormwater runoff from bridges across the state, (2) evaluate the treatment practices that can be used to reduce constituent loadings to surface waters from bridges, and (3) determine the effectiveness of the evaluated treatment practices.

There is evidence that bridge deck runoff has a relatively high loading of a variety of constituents such as nutrients, solids, pesticides, metals, and polycyclic aromatic hydrocarbons (PAHs). Information on the quality of bridge deck runoff in North Carolina is, however, lacking.  Stormwater permits are designed to reduce nonpoint source loadings of anthropogenically derived constituents to surface waters.  Permits for bridges in NC, however, must be based on data collected from studies which were conducted 10 – 20 years ago and in other parts of the U.S.  As a result permit requirements for bridges in NC may be unnecessarily conservative or inadequate for protecting receiving water quality.  Added to this need for more detailed, current, and local bridge deck runoff data are the requirements of House Bill 2436.

Objective

The primary objective of this investigation is to identify the loading of selected constituents in stormwater runoff from representative bridges across North Carolina.  Working collaboratively, NCDOT and USGS identified other study objectives which could provide information valuable in helping understand the effects of bridge deck runoff on receiving water quality and in managing stormwater runoff from bridges.

The objectives for this investigation, are as follows:

  • Characterize stormwater runoff quality and quantity from selected representative bridges in North Carolina.
  • In order to better understand the effects of stormwater runoff from bridges on receiving waters, (a) determine if the chemistry of bed sediments upstream and downstream from selected bridges differs substantially; (b) measure stream water quality upstream from selected bridges in order to compare bridge deck stormwater concentrations and loads to stream constituent concentrations and loads; and (c) estimate the length of the mixing zone at the bridge deck study sites under a range of flow conditions, where the mixing zone is defined here as the stream reach required for a point source of stormwater entering the stream from the bank to became fully mixed across the stream.

Scope

This investigation will measure bridge deck runoff from 15 bridges across NC.  Bridges will represent a range of physiographic and climatic conditions, a range of average daily traffic (ADT), and a range in size.  Runoff from both concrete deck and asphalt deck bridges will be sampled.  The goal is to sample 12 runoff events at each bridge during the study.  Samples will be analyzed for a wide range of constituents, including nutrients, major and trace metals, oil and grease, and semivolatile organic compounds.

Bridges at which runoff will be sampled are fitted with a collection system so that all bridge runoff flows through a single pipe, thereby facilitating sampling.  Discharge from the collection system flows across a grass swale or through a pond before entering the stream.

Bottom sediment quality will be measured at 30 sites, 15 of which will be the bridge deck runoff monitoring sites, and 15 of which will be at bridges in which runoff discharges from scuppers (essentially a series of pipes along the curb to drain the bridge) directly into the stream.  This type of drainage system is much more common across NC than the collection system.   Samples at each bridge will be collected once at a location upstream from the bridge and at a second location downstream from the bridge.  Bed sediment will be analyzed for nutrients, major and trace metals, and semivolatile organic compounds.  Both total and total recoverable concentrations of inorganic elements will be measured.

Four streams at bridge deck runoff sites will be sampled intensively in order to estimate annual loadings of suspended sediment, nutrients and other commonly detected constituents.  Stream concentrations and loads will be compared to bridge deck runoff concentrations and loadings at these sites in order to provide insight into the relative contribution of bridge deck runoff to total stream quality.

Background

On July 1, 2008, the North Carolina General Assembly passed House Bill 2436, Session Law 2008-107, Stormwater Runoff from Bridges Section 25.18.(a,b,c). This bill requires the North Carolina Department of Transportation (NCDOT) to study 50 bridges to (1) quantify the constituents in stormwater runoff from bridges across the state, (2) evaluate the treatment practices that can be used to reduce constituent loadings to surface waters from bridges, and (3) determine the effectiveness of the evaluated treatment practices.

There is evidence that bridge deck runoff has a relatively high loading of a variety of constituents such as nutrients, solids, pesticides, metals, and polycyclic aromatic hydrocarbons (PAHs). Information on the quality of bridge deck runoff in North Carolina is, however, lacking. Stormwater permits are designed to reduce nonpoint source loadings of anthropogenically derived constituents to surface waters. Permits for bridges in NC, however, must be based on data collected from studies which were conducted 10 - 20 years ago and in other parts of the U.S. As a result permit requirements for bridges in NC may be unnecessarily conservative or inadequate for protecting receiving water quality. Added to this need for more detailed, current, and local bridge deck runoff data are the requirements of House Bill 2436.

Objective

The primary objective of this investigation is to identify the loading of selected constituents in stormwater runoff from representative bridges across North Carolina. Working collaboratively, NCDOT and USGS identified other study objectives which could provide information valuable in helping understand the effects of bridge deck runoff on receiving water quality and in managing stormwater runoff from bridges.

The objectives for this investigation, are as follows:

  • Characterize stormwater runoff quality and quantity from selected representative bridges in North Carolina.
  • In order to better understand the effects of stormwater runoff from bridges on receiving waters, (a) determine if the chemistry of bed sediments upstream and downstream from selected bridges differs substantially; (b) measure stream water quality upstream from selected bridges in order to compare bridge deck stormwater concentrations and loads to stream constituent concentrations and loads; and (c) estimate the length of the mixing zone at the bridge deck study sites under a range of flow conditions, where the mixing zone is defined here as the stream reach required for a point source of stormwater entering the stream from the bank to became fully mixed across the stream.

Scope

This investigation will measure bridge deck runoff from 15 bridges across NC. Bridges will represent a range of physiographic and climatic conditions, a range of average daily traffic (ADT), and a range in size. Runoff from both concrete deck and asphalt deck bridges will be sampled. The goal is to sample 12 runoff events at each bridge during the study. Samples will be analyzed for a wide range of constituents, including nutrients, major and trace metals, oil and grease, and semivolatile organic compounds.

Bridges at which runoff will be sampled are fitted with a collection system so that all bridge runoff flows through a single pipe, thereby facilitating sampling. Discharge from the collection system flows across a grass swale or through a pond before entering the stream.

Bottom sediment quality will be measured at 30 sites, 15 of which will be the bridge deck runoff monitoring sites, and 15 of which will be at bridges in which runoff discharges from scuppers (essentially a series of pipes along the curb to drain the bridge) directly into the stream. This type of drainage system is much more common across NC than the collection system. Samples at each bridge will be collected once at a location upstream from the bridge and at a second location downstream from the bridge. Bed sediment will be analyzed for nutrients, major and trace metals, and semivolatile organic compounds. Both total and total recoverable concentrations of inorganic elements will be measured.

Four streams at bridge deck runoff sites will be sampled intensively in order to estimate annual loadings of suspended sediment, nutrients and other commonly detected constituents. Stream concentrations and loads will be compared to bridge deck runoff concentrations and loadings at these sites in order to provide insight into the relative contribution of bridge deck runoff to total stream quality.

Approach

The specific tasks to meet the study objectives are as follows:

  • Quantify bridge deck runoff volume and quality.
  • Quantify stream water quality upstream of selected bridge deck monitoring sites.
  • Compare streambed sediment quality at locations upstream and downstream from selected NCDOT bridges.
Bridge deck runoff monitoring equipment

Bridge deck runoff monitoring station at US74 bridge over Smith Creek near Wilmington, New Hanover County, North Carolina

(Public domain.)

Task 1. Quantify bridge deck runoff volume and quality

Bridge deck runoff volume and water quality and will be monitored at 15 bridge sites in North Carolina (five in the Blue Ridge, seven in the Piedmont, and three in the Coastal Plain; table 2). Runoff samples will be collected by using an automated sampler, flow will be measured continuously in the collection system discharge pipe by using acoustic velocimetry, and a raingage will be installed at each site.

All pertinent bridge characteristics will be provided to USGS by NCDOT from the NCDOT bridge maintenance database. NCDOT also will provide continuous daily traffic counts at each of the 15 sites during the study. Basins upstream from the bridge sampling site will be characterized by compiling information on physical features (area, slope, etc.), population, land use, point sources, and upstream bridge crossings.

Close up pipe outflow

Close up pipe outflow, US74 bridge over Smith Creek near Wilmington, New Hanover County, North Carolina

(Public domain.)

Bridge Deck Runoff Constituents

A broad range of constituent groups will be measured in the bridge deck runoff (table 1). These constituents were measured in at least 20 percent of the 218 highway runoff studies summarized in the National Highway Runoff Data and Methodology Synthesis (Granato, 2003) and include physical properties, solids, nutrients, major elements, trace metals, and PAHs. The number of analytes may be reduced if initial analyses consistently show concentrations that are below the laboratory reporting level. Total bridge deck stormwater loads will be based on total recoverable constituents because 1) there are fewer sample processing steps and therefore fewer opportunities for contamination, 2) North Carolina water quality criteria currently are based on total recoverable concentrations, and 3) most historical data are for total recoverable concentrations. All chemical analyses of water samples will be conducted at the USGS National Water Quality Laboratory (NWQL) in Denver, Colorado, with the exception of suspended sediment concentrations, which will be determined at the USGS Kentucky Sediment Laboratory.

Table 1. Water-quality constituents to be analyzed for bridge deck runoff samples.

Constituent group Analytes
Water samples (12 events at all bridge deck stormwater sites)
Physical Specific conductance, pH
Solids Total solids, total dissolved solids, total suspended solids, and suspended sediment concentration
Nutrients Total Kjeldahl nitrogen, nitrate+nitrite, ammonia, total phosphorus, ortho-phosphorus
Major elements Calcium, magnesium, sodium, potassium, chloride, sulfate, bromide, and alkalinity
Trace metals Total recoverable and dissolved: aluminum, arsenic, cadmium, chromium, copper, iron, lead, manganese, mercury, nickel, selenium, and zinc
Organic compounds Total and dissolved organic carbon, oil and grease, total petroleum hydrocarbons, 56 semivolatile organic compounds including PAHs and pthalates
Fluvial sediment (4 samples collected at one bridge deck site)
Trace metals Total-digestion: aluminum, arsenic, cadmium, chromium, copper, iron, lead, manganese, mercury, nickel, selenium, and zinc

Task 2. Quantify constituent concentrations and loads upstream from bridges

Constituent concentrations represent conditions to which aquatic organisms are exposed. Prevailing water-quality standards also are expressed as concentrations. Therefore, comparison of constituent concentrations in bridge runoff and receiving waters is warranted. In addition, an evaluation of how much the mass of material washed off the bridge deck adds to the instream load is a measure of the relative effect of the bridge on downstream stream water quality. To make this comparison, measured bridge deck constituent loads of suspended sediment, nitrogen, phosphorus, total organic carbon, total dissolved solids, sodium, chloride, and other commonly detected constituents will be compared to instream loads at selected sites.

Four bridge deck monitoring sites that are co-located with USGS stream gages (table 5) will be sampled monthly for one year to characterize instream concentrations and flux. The monthly samples will be supplemented with 6 high-flow events sampled at each site using automated samplers. The four sites represent a gradient of daily traffic counts: Swannanoa River (traffic count is in the 90th percentile of the 15 bridge sites), Little River (75 percentile), Mountain Creek (35th percentile), and Black River (5th percentile). A new, temporary streamflow gage was installed at the Swannanoa River site; USGS gages already exist at the remaining three sites.

Task 3. Compare streambed sediment quality upstream and downstream from bridges

Concentrations of material attached to sediment particles represent water-quality conditions on a time scale from weeks to years, depending on recent hydraulics at the site, whereas a single water sample represents water quality on a time scale from minutes to hours. Bed sediments provide habitat for aquatic organisms and an interface for ground-water and surface-water interactions; thus, they are an important component of stream ecosystems. Streambed-quality data will aid in the interpretation of benthic macroinvertebrate communities at sites near bridges, which are being assessed separately from this study.

Many constituents associated with road runoff preferentially adsorb to particulates, and thus are found in higher concentrations in sediment than in overlying water. These hydrophobic constituents may include several trace metals, nutrients, and persistent organic compounds such as polycyclic aromatic hydrocarbons (PAHs).

Streambed Sediment Constituents

Fine-grained sediments will be analyzed for major and trace elements, total and organic carbon, total nitrogen, phosphorus, and sulfur, and semivolatile organic compounds including PAHs and pthalates (table 2). Both total-digestion and total recoverable concentrations of inorganic elements will be measured. As mentioned previously, total-digestion concentrations reflect the underlying mineralogy of the sediment, while total recoverable concentrations reflect metals that are adsorbed or lightly bound to the mineral matrix. Sediment quality guidelines are variously based on total-digestion or total-recoverable concentrations.

Table 2. Constituents to be analyzed in bed sediment

Constituent group Analytes
Fine-grained fraction (diameter <63 microns)
Nutrients and carbon Total nitrogen, total phosphorus, total carbon and inorganic carbon
Major elements Calcium, magnesium, sodium, potassium, sulfur
Trace metals Total-digestion and total recoverable aluminum, arsenic, cadmium, chromium, copper, iron, lead, manganese, mercury, nickel, selenium, and zinc
Organics Total organic carbon and semivolatile compounds including polycyclic aromatic hydrocarbons and pthalates

 

View Bridge Stormwater Study Sites map

Table 3. Real-time streamgage and precipitation stations co-located with the bridge deck runoff monitoring sites.

Site Name                                                                                           Site Type                      USGS Station #       Period of record

Raingage at Town Creek near Wilmington, NC, Bridge No 90061  Precipitation 340813077591645 April 2009  to April 2010
Raingage at Smith Creek near Wilmington, NC, Bridge No 640131 Precipitation 341528077550745 April 2009  to April 2010
Raingage at Black River near Tomahawk, NC, Bridge No. 810014 Precipitation 344516078172145 April 2009  to April 2010
Raingage at Mountain Creek near Bahama, NC, Bridge No. 310005 Precipitation 360908078540745 May 2009 to April 2010
Rainage at Little River near Orange Factory, NC, Bridge No. 310064  Precipitation 360829078550945 May 2009 to April 2010
Raingage at Perry Creek near Raleigh, NC, Bridge No. 910124  Precipitation 355247078325045 April 2009  to April 2010
Raingage at Mango Creek near Raleigh, NC, Bridge No. 911102  Precipitation 354703078304845 March 2009  to April 2010
Raingage at Mallard Creek near Charlotte, NC, Bridge No. 590296 Precipitation 351911080450545 April 2009  to April 2010
Raingage at Swift Creek at Garner, NC, Bridge No. 910255 Precipitation 354217078392245 May 2009 to March 2010
Raingage at Middle Creek near Fuquay-Varina, NC, Bridge No. 910273 Precipitation 353633078411045 May 2009 to March 2010
Raingage at Swannanoa River near Black Mountain, NC, Bridge No. 100494  Precipitation 353708082182145 March 2009  to April 2010
Raingage at Big Ivy Creek near Mars Hill, NC, Bridge No. 100734 Precipitation 354728082321945 July 2009 to April 2010
Raingage at Dillingham Creek at Barnardsville, NC, Bridge No. 100145 Precipitation 354607082260945 May 2009 to April 2010
Raingage at Bolyston Creek at Mills River, NC, Bridge No. 440008 Precipitation 352231082325645 May 2009 to April 2010
Raingage at Flat Creek near Weaverville, NC, Bridge No. 100250 Precipitation 354306082372645 July 2009 to April 2010
Black River near Tomahawk, NC Streamflow and Water Quality 02106500 October 1951 - present
Little River at SR 1461 near Orange Factory, NC Streamflow and Water Quality 0208521324 September 1987 - present
Mountain Creek at SR 1617 near Bahama, NC Streamflow  0208524090 October 1994 - present
Mountain Creek at SR 1616 near Bahama, NC Water Quality 0208524088 May 2009 - April 2010
Swannanoa River near Black Mountain, NC Streamflow and Water Quality 03448800 March 2009 - present

Results of the bridge deck stormwater quality monitoring, stream water quality sampling, streambed sediment sampling, and empirical mixing calculations of bridge stormwater constituents with receiving waters will be used to characterize and document the bridge deck runoff and effects of bridge deck runoff on receiving water.

An online USGS Scientific Investigations Report, documenting the data collection and approach used to characterize the bridge deck runoff and determine the impact of bridge deck runoff on receiving waters of the state, has been published and is available online at http://pubs.usgs.gov/sir/2011/5180/.

Related References

Dupuis, T.V., 2002, Assessing the impacts of bridge deck runoff contaminants in receiving waters: National Cooperative Highway Research Program Report 474, Transportation Research Board, National Research Council.

Granato, G.E., 2003, National Highway Runoff Water-Quality Data and Methodology Synthesis, Volume III-Availability and documentation of published information of synthesis of regional or national highway-runoff quality data: Federal Highway Administration Publication (FHWA-EP-03-056), 71p.

Malina, J.F, Barrett, M.E., Jackson, A., Kramer, T., 2005, Characterization of stromwater runoff from a bridge deck and approach highway, effects on receiving water quality: Center for Transportation Research, University of Texas at Austin, Federal Highway Administration Publication (FHWA/TX-06/0-4543-1), 88 p. http://www.utexas.edu/research/ctr/pdf_reports/0_4543_1.pdf

URS Corporation, 2010, Stormwater runoff from bridges: Final report to Joint Legislation Transportation Oversight Committee, In fulfillment of Session Law 2008-107 for NC Department of Transportation, 262 p. https://connect.ncdot.gov/resources/hydro/Stormwater%20Resources/Stormwater%20Runoff%20from%20Bridges%20-%20May%202012.pdf

Wagner, C.R., Fitzgerald, S.A., Sherrell, R.D., Harned, D.A., Staub, E.L., Pointer, B.H., and Wehmeyer, L.L., 2011, Characterization of stormwater runoff from bridges in North Carolina and the effects of bridge deck runoff on receiving streams: U.S. Geological Survey Scientific Investigations Report 2011-5180, 95 p. + 8 appendix tables, available online at http://pubs.usgs.gov/sir/2011/5180/.