July 2004 04/17/WQPC-WWF EPA/600/R-04/125 Environmental Technology Verification Report Stormwater Source Area Treatment Device The Stormwater Management StormFilter® Using ZPG Filter Media Prepared by NSF International Under a Cooperative Agreement with U.S. Environmental Protection Agency ETV ETV ET ------- THE ENVIRONMENTAL TECHNOLOGY VERIFICATION PROGRAM U.S. Environmental Protection Agency NSF International ETV Joint Verification Statement TECHNOLOGY TYPE: APPLICATION: TECHNOLOGY NAME: TEST LOCATION: COMPANY: ADDRESS: WEB SITE: EMAIL: STORMWATER TREATMENT TECHNOLOGY SUSPENDED SOLIDS AND ROADWAY POLLUTANT TREATMENT THE STORMWATER MANAGEMENT STORMFILTER® USING ZPG FILTER MEDIA MILWAUKEE, WISCONSIN STORMWATER MANAGEMENT, INC. 12021-B NE Airport Way Portland, Oregon 97220 http://www.stormwaterinc.com mail@ stormwaterinc.com PHONE: (800)548-4667 FAX: (503)240-9553 NSF International (NSF), in cooperation with the EPA, operates the Water Quality Protection Center (WQPC), one of six centers under ETV. The WQPC recently evaluated the performance of the Stormwater Management StormFilter® (StormFilter) using ZPG filter media manufactured by Stormwater Management, Inc. (SMI). The system was installed at the "Riverwalk" site in Milwaukee, Wisconsin. Earth Tech, Inc. and the United States Geologic Survey (USGS) performed the testing. The U.S. Environmental Protection Agency (EPA) has created the Environmental Technology Verification (ETV) Program to facilitate the deployment of innovative or improved environmental technologies through performance verification and dissemination of information. The goal of the ETV program is to further environmental protection by accelerating the acceptance and use of improved and more cost-effective technologies. ETV seeks to achieve this goal by providing high quality, peer- reviewed data on technology performance to those involved in the design, distribution, permitting, purchase, and use of environmental technologies. ETV works in partnership with recognized standards and testing organizations; stakeholder groups, which consist of buyers, vendor organizations, and permitters; and with the full participation of individual technology developers. The program evaluates the performance of innovative technologies by developing test plans that are responsive to the needs of stakeholders, conducting field or laboratory tests (as appropriate), collecting and analyzing data, and preparing peer-reviewed reports. All evaluations are conducted in accordance with rigorous quality assurance protocols to ensure that data of known and adequate quality are generated and that the results are defensible. 04/17/WQPC-WWF The accompanying notice is an integral part of this verification statement. VS-i July 2004 ------- TECHNOLOGY DESCRIPTION The following description of the StormFilter was provided by the vendor and does not represent verified information. The StormFilter installed at the Riverwalk site consists of an inlet bay, flow spreader, cartridge bay, overflow baffle, and outlet bay, housed in a 12 foot by 6 foot pre-cast concrete vault. The inlet bay serves as a grit chamber and provides for flow transition into the cartridge bay. The flow spreader traps floatables, oil, and surface scum. This StormFilter was designed to treat stormwater with a maximum flow rate of 0.29 cubic feet per second (cfs). Flows greater than the maximum flow rate would pass the overflow baffle to the discharge pipe, bypassing the filter media. The StormFilter contains filter cartridges filled with ZPG filter media (a mixture of zeolite, perlite, and granular activated carbon), which are designed to remove sediments, metals, and stormwater pollutants from wet weather runoff. Water in the cartridge bay infiltrates the filter media into a tube in the center of the filter cartridge. When the center tube fills, a float valve opens and a check valve on top of the filter cartridge closes, creating a siphon that draws water through the filter media. The filtered water drains into a manifold under the filter cartridges and to the outlet bay, where it exits the system through the discharge pipe. The system resets when the cartridge bay is drained and the siphon is broken. The vendor claims that the treatment system can remove 50 to 85 percent of the suspended solids in stormwater, along with removal of total phosphorus, total and dissolved zinc, and total and dissolved copper in ranges from 20 to 60 percent. VERIFICATION TESTING DESCRIPTION Methods and Procedures The test methods and procedures used during the study are described in the Test Plan for Verification of Stormwater Management, Inc. StormFilter® Treatment System Using ZPG Media, "Riverwalk Site, " Milwaukee, Wisconsin (NSF International and Earth Tech, March 2004) (VTP). The StormFilter treats runoff collected from a 0.19-acre portion of the eastbound highway surface of Interstate 794. Milwaukee receives an average of nearly 33 inches of precipitation, approximately 31 percent of which occurs during the summer months. Verification testing consisted of collecting data during a minimum of 15 qualified events that met the following criteria: • The total rainfall depth for the event, measured at the site, was 0.2 inches (5 mm) or greater (snow fall and snow melt events do not qualify); • Flow through the treatment device was successfully measured and recorded over the duration of the runoff period; • A flow-proportional composite sample was successfully collected for both the influent and effluent over the duration of the runoff event; • Each composite sample was comprised of a minimum of five aliquots, including at least two aliquots on the rising limb of the runoff hydrograph, at least one aliquot near the peak, and at least two aliquots on the falling limb of the runoff hydrograph; and • There was a minimum of six hours between qualified sampling events. Automated sample monitoring and collection devices were installed and programmed to collect composite samples from the influent, the treated effluent, and the untreated bypass during qualified flow events. In addition to the flow and analytical data, operation and maintenance (O&M) data were recorded. Samples were analyzed for the following parameters: 04/17/WQPC-WWF The accompanying notice is an integral part of this verification statement. July 2004 VS-ii ------- Sediments Metals • total suspended solids (TSS) • total dissolved solids (TDS) • suspended sediment concentration (SSC) • particle size analysis VERIFICATION OF PERFORMANCE total and dissolved cadmium, lead, copper and zinc Nutrients Water Quality Parameters • total and • chemical oxygen dissolved demand (COD) phosphorus • dissolved chloride • total calcium and magnesium Verification testing of the StormFilter lasted approximately 16 months, and coincided with testing conducted by USGS and the Wisconsin Department of Natural Resources. A total of 20 storm events were sampled. Conditions during certain storm events prevented sampling for some parameters. However, samples were successfully taken and analyzed for all parameters for at least 15 of the 20 total storm events. Test Results The precipitation data for the 20 rain events are summarized in Table 1. Table 1. Rainfall Data Summary Event Start Number Date 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 6/21/02 7/8/02 8/21/02 9/2/02 9/18/02 9/29/02 12/18/02 4/19/03 5/4/03 5/30/03 6/8/03 6/27/03 7/4/03 7/8/03 9/12/03 9/14/03 9/22/03 10/14/03 10/24/03 1 1/4/03 Start Time 6:54 21:16 20:08 5:24 5:25 0:49 1:18 5:39 21:21 18:55 3:26 17:30 7:25 9:49 15:33 5:22 2:28 1:03 16:46 16:14 Rainfall Amount (inches) 0.52 1.5 1.7 1.2 0.37 0.74 0.37 0.55 0.90 0.54 0.62 0.57 0.53 0.33 0.22 0.47 0.27 0.25 0.71 0.60 Rainfall Duration (hnmin) 0:23 2:04 15:59 3:24 4:54 7:54 3:47 10:00 11:44 4:06 11:09 13:25 40:43 3:37 1:55 6:35 2:09 2:07 15:07 2:09 Peak Runoff Discharge Volume Rate (ft3)1 (gpm)1 420 1,610 1,620 1,180 350 730 300 340 540 320 450 460 550 260 150 340 270 220 410 560 447 651 671 164 136 70.9 61.0 96.9 73.2 83.9 140 107 143 62.8 21.5 264 104 56.5 75.8 906 04/17/WQPC-WWF 1 Runoff volume and peak discharge volume was measured at the outlet monitoring point. The accompanying notice is an integral part of this verification statement. VS-iii July 2004 ------- The monitoring results were evaluated using event mean concentration (EMC) and sum of loads (SOL) comparisons. The EMC or efficiency ratio comparison evaluates treatment efficiency on a percentage basis by dividing the effluent concentration by the influent concentration and multiplying the quotient by 100. The efficiency ratio was calculated for each analytical parameter and each individual storm event. The SOL comparison evaluates the treatment efficiency on a percentage basis by comparing the sum of the influent and effluent loads (the product of multiplying the parameter concentration by the precipitation volume) for all 15 storm events. The calculation is made by subtracting the quotient of the total effluent load divided by the total influent load from one, and multiplying by 100. SOL results can be summarized on an overall basis since the loading calculation takes into account both the concentration and volume of runoff from each event. The analytical data ranges, EMC range, and SOL reduction values are shown in Table 2. Table 2. Analytical Data, EMC Range, and SOL Reduction Results Parameter1 TSS ssc TDS Total phosphorus Dissolved phosphorus Total magnesium Total calcium Total copper Total lead Total zinc Dissolved copper Dissolved zinc COD Dissolved chloride Units mg/L mg/L mg/L mg/L as P mg/L as P mg/L mg/L ug/L ug/L ug/L ug/L ug/L mg/L mg/L Inlet Range 29 -780 51-5,600 <50 0.05 0.01 4.0 9.4 15 <31 -600 -0.63 -0.20 -174 -430 -440 -280 77-1,400 <5 26 18 3.2 -58 -360 -320 -470 Outlet EMC Range Range (percent) 20- 12- <50- 0.03 0.01 1.1 4.0 7.0- <31 28- <5 16- 17- 3.3- -380 -370 4,2002 -0.30 -0.19 -26 -68 -140 -94 -540 -42 -160 -190 2,6002 -33 3- -600 0- -35 53- 26- 8.3 33- 20- -47 -86 -91 -740 -95 99 - 10 -70 -38 -96 -93 -96 -91 -89 -64 -56 -47 -24 SOL Reduction (percent) 46 92 -1702 38 6 85 79 59 64 64 16 17 16 -2422 1 Total and dissolved cadmium and dissolved lead concentrations were below method detection limits for every storm event. 2 Dissolved chloride and TDS results were heavily influenced by a December storm event when road salt was applied to melt snow and ice. Based on the SOL evaluation method, the TSS reductions nearly met the vendor's performance claim, while SSC reductions exceeded the vendor's performance claim of 50 to 85 percent solids reduction. The StormFilter also met or exceeded the performance claim for total and dissolved phosphorus, total copper, and total zinc. The StormFilter did not meet the performance claim for dissolved copper or dissolved zinc, both of which were 20 to 40 percent reduction, and had no performance claims for any other parameters. The TDS and dissolved chloride values were heavily influenced by a single event (December 18, 2002), where high TDS and dissolved chloride concentrations were detected in the effluent. The event was likely influenced by application of road salt on the freeway. When this event is omitted from the SOL calculation, the SOL value is -37 percent for TDS and -31 percent for dissolved chloride. 04/17/WQPC-WWF The accompanying notice is an integral part of this verification statement. July 2004 VS-iv ------- Particle size distribution analysis was conducted on samples when adequate sample volume was collected. The analysis identified that the runoff entering the StormFilter contained a large proportion of coarse sediment. The effluent contained a larger proportion of fine sediment, which passed through the pores within the filter cartridges. For example, 20 percent of the sediment in the inlet samples was less than 62.5 urn in size, while 78 percent of the sediment in the outlet samples was less than 62.5 um in size. System Operation The StormFilter was installed prior to verification testing, so verification of installation procedures on the system was not documented. The StormFilter was cleaned and equipped with new filter cartridges prior to the start of verification. During the verification period, two inspections were conducted as recommended by the manufacturer. Based on visual observations, the inspectors concluded that a major maintenance event, consisting of cleaning the vault and replacing the filter cartridges, was not required. After the verification was complete, a major maintenance event was conducted, and approximately 570 pounds (dry weight) of sediment was removed from the StormFilter's sediment collection chamber. Quality Assurance/Quality Control NSF personnel completed a technical systems audit during testing to ensure that the testing was in compliance with the test plan. NSF also completed a data quality audit of at least 10 percent of the test data to ensure that the reported data represented the data generated during testing. In addition to QA/QC audits performed by NSF, EPA personnel conducted an audit of NSF's QA Management Program. Original signed by Original Signed by Lawrence W. Reiter, Ph. D. September 21, 2004 Gordon E. Bellen September 23, 2004 Lawrence W. Reiter, Ph. D. Date Gordon E. Bellen Date Acting Director Vice President National Risk Management Laboratory Research Office of Research and Development NSF International United States Environmental Protection Agency NOTICE: Verifications are based on an evaluation of technology performance under specific, predetermined criteria and the appropriate quality assurance procedures. EPA and NSF make no expressed or implied warranties as to the performance of the technology and do not certify that a technology will always operate as verified. The end user is solely responsible for complying with any and all applicable federal, state, and local requirements. Mention of corporate names, trade names, or commercial products does not constitute endorsement or recommendation for use of specific products. This report is not an NSF Certification of the specific product mentioned herein. Availability of Supporting Documents Copies of the ETV Verification Protocol, Stormwater Source Area Treatment Technologies Draft 4.1, March 2002, the verification statement, and the verification report (NSF Report Number 04/17/WQPC-WWF) are available from: ETV Water Quality Protection Center Program Manager (hard copy) NSF International P.O. Box 130140 Ann Arbor, Michigan 48113-0140 NSF website: http://www.nsf.org/etv (electronic copy) EPA website: http://www.epa.gov/etv (electronic copy) Appendices are not included in the verification report, but are available from NSF upon request. 04/17/WQPC-WWF The accompanying notice is an integral part of this verification statement. July 2004 VS-v ------- Environmental Technology Verification Report Stormwater Source Area Treatment Device The Stormwater Management StormFilter® Using ZPG Filter Media Prepared for: NSF International Ann Arbor, MI 48105 Prepared by Earth Tech Inc. Madison, Wisconsin With assistance from: United States Geologic Survey (Wisconsin Division) Wisconsin Department of Natural Resources Under a cooperative agreement with the U.S. Environmental Protection Agency Raymond Frederick, Project Officer ETV Water Quality Protection Center National Risk Management Research Laboratory Water Supply and Water Resources Division U.S. Environmental Protection Agency Edison, New Jersey July 2004 ------- Notice The U.S. Environmental Protection Agency (EPA) through its Office of Research and Development has financially supported and collaborated with NSF International (NSF) under a Cooperative Agreement. The Water Quality Protection Center (WQPC), operating under the Environmental Technology Verification (ETV) Program, supported this verification effort. This document has been peer reviewed and reviewed by NSF and EPA and recommended for public release. Mention of trade names or commercial products does not constitute endorsement or recommendation by the EPA for use. ------- Foreword The following is the final report on an Environmental Technology Verification (ETV) test performed for NSF International (NSF) and the United States Environmental Protection Agency (EPA). The verification test for The Stormwater Management StormFilter® using ZPG Media was conducted at a testing site in downtown Milwaukee, Wisconsin, maintained by Wisconsin Department of Transportation (WisDOT). The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions leading to a compatible balance between human activities and the ability of natural systems to support and nurture life. To meet this mandate, EPA's research program is providing data and technical support for solving environmental problems today and building a science knowledge base necessary to manage our ecological resources wisely, understand how pollutants affect our health, and prevent or reduce environmental risks in the future. The National Risk Management Research Laboratory (NRMRL) is the Agency's center for investigation of technological and management approaches for preventing and reducing risks from pollution that threaten human health and the environment. The focus of the Laboratory's research program is on methods and their cost-effectiveness for prevention and control of pollution to air, land, water, and subsurface resources; protection of water quality in public water systems; remediation of contaminated sites, sediments and ground water; prevention and control of indoor air pollution; and restoration of ecosystems. NRMRL collaborates with both public and private sector partners to foster technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's research provides solutions to environmental problems by: developing and promoting technologies that protect and improve the environment; advancing scientific and engineering information to support regulatory and policy decisions; and providing the technical support and information transfer to ensure implementation of environmental regulations and strategies at the national, state, and community levels. This publication has been produced as part of the Laboratory's strategic long-term research plan. It is published and made available by EPA's Office of Research and Development to assist the user community and to link researchers with their clients. Lawrence W. Reiter, Acting Director National Risk Management Research Laboratory 11 ------- Contents Verification Statement VS-i Notice i Foreword ii Contents iii Figures iv Tables iv Abbreviations and Acronyms vi Chapter 1 Introduction 1 1.1 ETV Purpose and Program Operation 1 1.2 Testing Participants and Responsibilities 1 1.2.1 U.S. Environmental Protection Agency 2 1.2.2 Verification Organization 2 1.2.3 Testing Organization 3 1.2.4 Analytical Laboratories 4 1.2.5 Vendor 4 1.2.6 Verification Testing Site 4 Chapter 2 Technology Description 6 2.1 Treatment System Description 6 2.2 Filtration Process 7 2.3 Technology Application and Limitations 8 2.4 Performance Claim 8 Chapters Test Site Description 9 3.1 Location and Land Use 9 3.2 Contaminant Sources and Site Maintenance 10 3.3 Stormwater Conveyance System 11 3.4 Water Quality /Water Resources 11 3.5 Local Meteorological Conditions 11 Chapter 4 Sampling Procedures and Analytical Methods 12 4.1 Sampling Locations 12 4.1.1 Site 1 -Influent 12 4.1.2 Site 2 - Treated Effluent 12 4.1.3 Other Monitoring Locations 13 4.2 Monitoring Equipment 14 4.3 Contaminant Constituents Analyzed 15 4.4 Sampling Schedule 16 4.5 Field Procedures for Sample Handling and Preservation 18 Chapter 5 Monitoring Results and Discussion 20 5.1 Monitoring Results: Performance Parameters 20 5.1.1 Concentration Efficiency Ratio 20 5.1.2 Sum of Loads 27 5.2 Particle Size Distribution 33 Chapter 6 QA/QC Results and Summary 35 6.1 Laboratory/Analytical Data QA/QC 35 6.1.1 Bias (Field Blanks) 35 6.1.2 Replicates (Precision) 36 in ------- 6.1.3 Accuracy 38 6.1.4 Representativeness 40 6.1.5 Completeness 40 6.2 Flow Measurement Calibration 41 6.2.1 Inlet- Outlet Volume Comparison 41 6.2.2 Gauge Height Calibration 44 6.2.3 Point Velocity Correction 44 6.2.4 Correction for Missing Velocity Data 44 Chapter 7 Operations and Maintenance Activities 47 7.1 System Operation and Maintenance 47 7.1.1 Major Maintenance Procedure 48 Chapter 8 References 49 Glossary 50 Appendices 52 A Verification Test Plan 52 B Event Hydrographs and Rain Distribution 52 C Analytical Data Reports 52 Figures Figure 2-1. Schematic drawing of atypical StormFilter system 6 Figure 2-2. Schematic drawing of a StormFilter cartridge 7 Figure 3-1. Location of test site 9 Figure 3-2. Drainage area detail 10 Figure 3-3. StormFilter drainage area condition 10 Figure 4-1. View of monitoring station 12 Figure 4-2. View of ISCO samplers 13 Figure 4-3. View of datalogger 13 Figure 4-4. View of rain gauge 14 Figure 6-1. Calibration curves used to correct flow measurements 42 Figure 6-2. Event 2 example hydrograph showing period of missing velocity data 45 Tables Table 2-1. StormFilter Performance Claims 8 Table 4-1. Field Monitoring Equipment 14 Table 4-2. Constituent List for Water Quality Monitoring 15 Table 4-3. Summary of Events Monitored for Verification Testing 17 Table 4-4. Rainfall Summary for Monitored Events 18 Table 5-1. Monitoring Results and Efficiency Ratios for Sediment Parameters 21 Table 5-2. Monitoring Results and Efficiency Ratios for Nutrient Parameters 23 Table 5-3. Monitoring Results and Efficiency Ratios for Metals 24 Table 5-4. Monitoring Results and Efficiency Ratios for Water Quality Parameters 26 Table 5-5. Sediment Sum of Loads Efficiencies Calculated Using Various Flow Volumes 28 Table 5-6. Sediment Sum of Loads Results 29 IV ------- Table 5-7. Nutrient Sum of Loads Results 30 Table 5-8. Metals Sum of Loads Results 31 Table 5-9. Water Quality Parameter Sum of Loads Results 32 Table 5-10. Particle Size Distribution Analysis Results 34 Table 6-1. Field Blank Analytical Data Summary 35 Table 6-2. Field Duplicate Sample Relative Percent Difference Data Summary 37 Table 6-3. Laboratory Duplicate Sample Relative Percent Difference Data Summary 38 Table 6-4. Laboratory MS/MSD Data Summary 39 Table 6-5. Laboratory Control Sample Data Summary 39 Table 6-6. Comparison of Inlet and Outlet Event Runoff Volumes 43 Table 6-7. Gauge Corrections for Flow Measurements at the Inlet 44 Table 6-8. Missing Sample Aliquots Due to Missing Inlet Velocity Data 46 Table 7-1. Operation and Maintenance During Verification Testing 47 ------- Abbreviations and Acronyms ASTM BMP cfs COD EMC EPA ETV ft2 ft3 g gal gpm in kg L Ib LOD LOQ NRMRL mg/L NSF NIST O&M QA QAPP QC SMI ssc SOL SOP IDS TO TP TSS USGS VA vo VTP WDNR WQPC WisDOT WSLH ZPG American Society for Testing and Materials Best Management Practice Cubic feet per second Chemical oxygen demand Event mean concentration U.S. Environmental Protection Agency Environmental Technology Verification Square feet Cubic feet Gram Gallon Gallon per minute Inch Kilogram Liters Pound Limit of detection Limit of quantification National Risk Management Research Laboratory Microgram per liter (ppb) Micron Milligram per liter NSF International, formerly known as National Sanitation Foundation National Institute of Standards and Technology Operations and maintenance Quality assurance Quality Assurance Project Plan Quality control Stormwater Management, Inc. Suspended sediment concentration Sum of loads Standard Operating Procedure Total dissolved solids Testing Organization Total phosphorus Total suspended solids United States Geological Survey Visual accumulator Verification Organization (NSF) Verification test plan Wisconsin Department of Natural Resources Water Quality Protection Center Wisconsin Department of Transportation Wisconsin State Laboratory of Hygiene ZPG media, a mixture of zeolite, perlite, and granular activated carbon VI ------- Chapter 1 Introduction 1.1 ETV Purpose and Program Operation The U.S. Environmental Protection Agency (EPA) has created the Environmental Technology Verification (ETV) Program to facilitate the deployment of innovative or improved environmental technologies through performance verification and dissemination of information. The goal of the ETV program is to further environmental protection by substantially accelerating the acceptance and use of improved and more cost-effective technologies. ETV seeks to achieve this goal by providing high quality, peer reviewed data on technology performance to those involved in the design, distribution, permitting, purchase, and use of environmental technologies. ETV works in partnership with recognized standards and testing organizations; stakeholders groups, which consist of buyers, vendor organizations, and permitters; and with the full participation of individual technology developers. The program evaluates the performance of innovative technologies by developing test plans that are responsive to the needs of stakeholders, conducting field or laboratory (as appropriate) testing, collecting and analyzing data, and preparing peer reviewed reports. All evaluations are conducted in accordance with rigorous quality assurance protocols to ensure that data of known and adequate quality are generated and that the results are defensible. NSF International (NSF), in cooperation with the EPA, operates the Water Quality Protection Center (WQPC). The WQPC evaluated the performance of The Stormwater Management StormFilter® using ZPG Filter Media (StormFilter), a Stormwater treatment device designed to remove suspended solids, metals, and other Stormwater pollutants from wet weather runoff. It is important to note that verification of the equipment does not mean that the equipment is "certified" by NSF or "accepted" by EPA. Rather, it recognizes that the performance of the equipment has been determined and verified by these organizations for those conditions tested by the Testing Organization (TO). 1.2 Testing Participants and Responsibilities The ETV testing of the StormFilter was a cooperative effort among the following participants: • U.S. Environmental Protection Agency • NSF International • U.S. Geologic Survey (USGS) • Wisconsin Department of Transportation (WisDOT) • Wisconsin Department of Natural Resources (WDNR) • Wisconsin State Laboratory of Hygiene (WSLH) • USGS Sediment Laboratory • Earth Tech, Inc. • Stormwater Management, Inc. (SMI) ------- The following is a brief description of each ETV participant and their roles and responsibilities. 7.2.7 U.S. Environmental Protection Agency The EPA Office of Research and Development, through the Urban Watershed Branch, Water Supply and Water Resources Division, National Risk Management Research Laboratory (NRMRL), provides administrative, technical, and quality assurance guidance and oversight on all ETV Water Quality Protection Center activities. In addition, EPA provides financial support for operation of the Center and partial support for the cost of testing for this verification. The key EPA contact for this program is: Mr. Ray Frederick, ETV WQPC Project Officer (732)321-6627 email: Frederick.Ray@epamail.epa.gov U.S. EPA, NRMRL Urban Watershed Management Research Laboratory 2890 Woodbridge Avenue (MS-104) Edison, New Jersey 08837-3679 7.2.2 Verification Organization NSF is the verification organization (VO) administering the WQPC in partnership with EPA. NSF is a not-for-profit testing and certification organization dedicated to public health, safety, and protection of the environment. Founded in 1946 and located in Ann Arbor, Michigan, NSF has been instrumental in development of consensus standards for the protection of public health and the environment. NSF also provides testing and certification services to ensure that products bearing the NSF name, logo and/or mark meet those standards. NSF personnel provided technical oversight of the verification process. NSF also provided review of the verification test plan (VTP) and this verification report. NSF's responsibilities as the VO include: • Review and comment on the VTP; • Review quality systems of all parties involved with the TO, and qualify the TO; • Oversee TO activities related to the technology evaluation and associated laboratory testing; • Conduct an on-site audit of test procedures; • Provide quality assurance/quality control (QA/QC) review and support for the TO; • Oversee the development of the verification report and verification statement; and, • Coordinate with EPA to approve the verification report and verification statement. ------- Key contacts at NSF are: Mr. Thomas Stevens, Program Manager Mr. Patrick Davison, Project Coordinator (734) 769-5347 (734) 913-5719 email: stevenst@nsf.org email: davison@nsf.org NSF International 789 North Dixboro Road Ann Arbor, Michigan 48105 (734)769-8010 1.2.3 Testing Organization The TO for the verification testing was Earth Tech, Inc. of Madison, Wisconsin (Earth Tech), which was assisted by the U.S. Geological Service (USGS), located in Middleton, Wisconsin. USGS provided testing equipment, helped to define field procedures, conducted the field testing, coordinated with the analytical laboratories, and conducted initial data analyses. The TO provided all needed logistical support, established a communications network, and scheduled and coordinated activities of all participants. The TO was responsible for ensuring that the testing location and conditions allowed for the verification testing to meet its stated objectives. The TO prepared the VTP; oversaw the testing; and managed, evaluated, interpreted and reported on the data generated by the testing, as well as evaluated and reported on the performance of the technology. TO employees set test conditions, and measured and recorded data during the testing. The TO's Project Manager provided project oversight. The key personnel and contacts for the TO are: Earth Tech, Inc.: Mr. Jim Bachhuber P.H. (608)828-8121 email: jim_bachhuber@earthtech. com Earth Tech, Inc. 1210 Fourier Drive Madison, Wisconsin 53717 United States Geologic Survey: Ms. Judy Horwatich (608) 821-3874 email: jawierl@usgs.gov ------- USGS 8505 Research Way Middleton, Wisconsin 53562 1.2.4 Analytical Laboratories The Wisconsin State Laboratory of Hygiene (WSLH), located in Madison, Wisconsin, analyzed the stormwater samples for the parameters identified in the VTP, except for suspended sediment concentration and particle size. The USGS Sediment Laboratory, located in Iowa City, Iowa, performed the suspended sediment concentration separations and particle size analyses. The key analytical laboratory contacts are: Mr. George Bowman Ms. Pam Smith (608) 224-6279 (319) 358-3602 email: gtb@mail.slh.wisc.edu email: pksmith@usgs.gov WSLH USGS Sediment Laboratory 2601 Agriculture Drive Federal Building Room 269 Madison, Wisconsin 53718 400 South Clinton Street Iowa City, Iowa 52240 7.2.5 Vendor Stormwater Management, Inc. (SMI) of Portland, Oregon, is the vendor of the StormFilter, and was responsible for supplying a field-ready system. SMI was also responsible for providing technical support, and was available during the tests to provide technical assistance as needed. The key contact for SMI is: Mr. James Lenhart, P.E. (800) 548-5667 email: jiml@stormwaterinc.com Stormwater Management, Inc. 12021-B NE Airport Way Portland, Oregon 97220 1.2.6 Verification Testing Site The StormFilter was installed in a parking lot under Interstate 794 on the east side of the Milwaukee River in downtown Milwaukee, Wisconsin. The StormFilter treated storm water collected from the decking of Interstate 794. The unit was installed in cooperation with the Wisconsin Department of Transportation (WisDOT), which is the current owner/operator of the system. ------- The key contact for WisDOT is: Mr. Robert Pearson (608) 266-7980 email: robert.pearson@dot.state.wi.us Bureau of Environment Wisconsin Department of Transportation 4802 Sheboygan Avenue, Room 451 Madison, Wisconsin 53707 ------- Chapter 2 Technology Description The following technology description data was supplied by the vendor and does not represent verified information. 2.1 Treatment System Description ® The Stormwater Management StormFilter using ZPG Media (StormFilter) is designed to remove sediments, metals, and other roadway pollutants from storm water. The StormFilter device under test was designed to treat storm water with a maximum flow rate of 0.29 cubic feet per second (cfs). The unit consisted of an inlet bay, flow spreader, cartridge bay, an overflow baffle, and outlet bay, all housed in a 12 ft by 6 ft pre-cast concrete vault. A 2 ft by 6 ft inlet bay served as a grit chamber and provided for flow transition into the 7.4 ft by 6 ft cartridge bay. The flow spreader provided for the trapping of floatables, oil, and surface scum. The unit also included nine filter cartridges filled with ZPG filter media (a mixture of zeolite, perlite, and granular activated carbon), installed inline with the storm drain lines. The cartridge bay provided for sediment storage of 0.87 cubic yards. A schematic of the StormFilter and a detail of the filter cartridge are shown in Figures 2-1 and 2-2. ACCESS DOORS LADDER FLOW SPREADER OUTLET PIPE FLOW SF'READER ENERGY DISSIPATOR HIGH FLOW BYPASS StormFiKerCARiRIDGE Figure 2-1. Schematic drawing of a typical StormFilter system. Additional equipment specifications, test site descriptions, testing requirements, sampling procedures, and analytical methods were detailed in the Test Plan for the Verification of the StormFilter* Treatment System using ZPG Media, "Riverwalk" Site, Version 4.3. The verification test plan (VTP) is included in Appendix A. ------- 2.2 Filtration Process The filtration process works by percolating storm water through a series of filter cartridges filled with ZPG media, which is a mixture of zeolite, perlite, and granular activated carbon. The filter media traps particulates and adsorbs materials such as suspended solids and petroleum hydrocarbons. The media will also trap pollutants such as phosphorus, nitrogen, and metals that commonly bind to sediment particulates. A diagram identifying the filter cartridge components is shown in Figure 2-2. CHECK WIVE RLTER MEDIA CENTER TUBE FLOAT SEAT SCRUBBING REGULATORS UNDER-DRAiN MANIFOLD FLOAT HOOD OUTER SCREEN QPT(Of4ALSECGNDARY FILTER MEDIA FILTERED WATER UNDER-DRAIN MANIFOLD . CAST INTO VAULT FLXXJR VAULT FLOOR Figure 2-2. Schematic drawing of a StormFilter cartridge. Storm water enters the cartridge bay through the flow spreader, where it ponds. Air in the cartridge is displaced by the water and purged from beneath the filter hood through the one-way check valve located on top of the cap. The water infiltrates through the filter media and into the center tube. Once the center tube fills with water, a float valve opens and the water in the center tube flows into the under-drain manifold, located beneath the filter cartridge. This causes the check valve to close, initiating a siphon that draws storm water through the filter. The siphon continues until the water surface elevation drops to the elevation of the hood's scrubbing regulators. When the water drains, the float valve closes and the system resets. ------- The StormFilter is equipped with an overflow baffle designed to bypass flows and prevent catch basin backup and surface flooding. The bypass flow is discharged through the outlet pipe along with the treated water. 2.3 Technology Application and Limitations StormFilter Treatment Systems are flexible in terms of the flow it can treat. By varying the holding tank size, and number of filter cartridges, the treatment capacity can be modified to accommodate runoff from various size watersheds. The filtration systems can be designed to receive runoff from all rainstorm events, or they can be designed with a high flow bypass system. The StormFilter installed at the Riverwalk site was designed to receive all the runoff from the drainage area. 2.4 Performance Claim SMI recognizes that stormwater treatment is a function of influent concentration and particle size distribution in the case of sediment removal. The performance claims for the StormFilter unit installed at the Riverwalk site are summarized in Table 2-1. SMI does not provide any additional removal claims for constituents other than those specified in Table 2-1. Table 2-1. StormFilter Performance Claims Removal Efficiency Range Constituent (Percent) Total suspended solids (TSS) 50 - 85 Total phosphorus 30-45 Dissolved phosphorus Negligible Total zinc 30-60 Dissolved zinc 20 - 40 Total copper 30-60 Dissolved copper 20-40 ------- Chapter 3 Test Site Description 3.1 Location and Land Use The StormFilter system is located in a municipal parking lot beneath an elevated freeway (1-794) and just east of the Milwaukee River, in downtown Milwaukee Wisconsin. The parking lot is located is just west of Water Street, between Clybourn Street and St. Paul Avenue. Figure 3-1 shows the location of the test site, and Figure 3-2 details the drainage area. Figure 3-1. Location of test site. The StormFilter receives runoff from 0.187 acres of the eastbound highway surface of Interstate 794. Surface inlets on the highway collect the runoff and convey the water to the treatment device via downspouts from the deck surface to beneath the parking lot below the highway deck, as shown in Figure 3-3. The drainage area determination was based on the following information and assumptions: 1. WisDOT design plans for Interstate 794 dated 1966 (scale: 1 inch equals 20 feet) and rehabilitation plans dated 1994; 2. The assumption that resurfacing the deck did not change the basic slope or relative drainage area to each inlet; and 3. The assumption that adjacent storm drains were capable of capturing all the flow in their respective drainage areas, forming a hydrologic barrier. The drainage site is not impacted by surrounding land uses due to its elevated highway decking. ------- StormFilter Drainage Area 1-794 Eastbound Lanes Figure 3-2. Drainage area detail. Figure 3-3. StormFilter drainage area condition. 3.2 Contaminant Sources and Site Maintenance The main pollutant sources within the drainage area are created by vehicular traffic, atmospheric deposition, and, winter salt applications that are applied as conditions require. 10 ------- The storm sewer catch basins do not have sumps. Conventional (mechanical) street sweeping is done on a monthly basis in the summer months (June through August). There are no other stormwater best management practices (BMPs) within the drainage area. 3.3 Stormwater Conveyance System The entire drainage area is served by a storm sewer collection system. Before installation of the StormFilter system, the drainage area discharged storm water directly to the Milwaukee River through the system under the parking lot. The highway deck is about 15 feet above the parking lot. Thus, the storm sewer conveyance system drops vertically through an 8-inch pipe to a point below the parking lot surface, then travels about 6.5 feet horizontally to the inlet monitoring (flow and quality) site, and another two feet to the StormFilter. The StormFilter outlet is connected to an 8-inch pipe that discharges without further treatment to the Milwaukee River. 3.4 Water Quality/Water Resources Stormwater from the site is discharged directly to the Milwaukee River, just upstream of the mouth to Milwaukee Harbor, and then into Lake Michigan. The river and harbor have had a history of severe water quality impacts from various sources including contaminated river sediments, urban non-point source runoff, rural non-point sources (higher upstream in the watershed), and point source discharges. The water quality in the river suffers from low dissolved oxygen, high nutrient, metals, bacteria levels, and toxic contamination. Most of the urban communities within the Milwaukee River watershed, including the City of Milwaukee, are under the State of Wisconsin stormwater permitting program (NR 216). This program meets or exceeds the requirements of EPA's Phase I stormwater regulations. 3.5 Local Meteorological Conditions The VTP (Appendix A) includes summary temperature and precipitation data from the National Weather Service station from the Mitchell Field Airport in Milwaukee. The statistical rainfalls for a series of recurrence and duration precipitation events are presented in the VTP (Hull et al., 1992). The climate of Milwaukee, and in Wisconsin in general, is typically continental with some modification by Lakes Michigan and Superior. Milwaukee experiences cold snowy winters, and warm to hot summers. Average annual precipitation is approximately 33 inches, with an average annual snowfall of 50.3 inches. 11 ------- Chapter 4 Sampling Procedures and Analytical Methods Descriptions of the sampling locations and methods used during verification testing are summarized in this section. Additional detail may be found in the VTP (Appendix A). 4.1 Sampling Locations Two locations in the test site storm sewer system were selected as sampling and monitoring sites to determine the treatment capability of the StormFilter. 4.1.1 Site 1 - Influent This sampling and monitoring site was selected to characterize the untreated stormwater from the entire drainage area. A velocity/stage meter and sampler suction tubing were located in the influent pipe, upstream from the StormFilter so that potential backwater effects of the treatment device would not affect the velocity measurements. The monitoring station (Figure 4-1) and test equipment (Figure 4-2 and 4-3) are shown below. Figure 4-1. View of monitoring station. 4.1.2 Site 2 - Treated Effluent This sampling and monitoring site was selected to characterize the stormwater treated by the StormFilter. A velocity/stage meter and sampler suction tubing, connected to the automated sampling equipment, were located in an eight-inch diameter plastic pipe downstream from the StormFilter. 12 ------- Figure 4-2. View of ISCO samplers. Figure 4-3. View of datalogger. 4.1.3 Other Monitoring Locations In addition to the two sampling and monitoring sites, a water-level recording device was installed in the StormFilter vault. The data from this device were used to verify the occurrence of bypass conditions. A rain gauge was located adjacent to the drainage area to monitor the depth of precipitation from storm events. The data were used to characterize the events to determine if they met the requirements for a qualified storm event. The rain gauge is shown in Figure 4-4. 13 ------- Figure 4-4. View of rain gauge. 4.2 Monitoring Equipment The specific equipment used for monitoring flow, sampling water quality, and measuring rainfall is listed in Table 4-1. Table 4-1. Field Monitoring Equipment Equipment Sitel Site 2 Rain Gauge StormFilter Vault Water Quality Sampler Velocity Measurement Stage Meter Datalogger Rain Gauge ISCO 3700 refrigerated automatic sampler (4, 10 L sample bottles) Marsh-McBirney Velocity Meter Model 270 Marsh-McBirney Velocity Meter Model 270 ISCO 3700 refrigerated automatic sampler (4, 10 L sample bottles) Marsh-McBirney Velocity Meter Model 270 Marsh-McBirney Velocity Meter Model 270 Campbell Scientific Campbell Scientific Inc. CR10X datalogger Inc. CR10X datalogger Rain-O- Matic Campbell Scientific Inc. SWD1 Campbell Scientific Inc. CR10X datalogger 14 ------- 4.3 Contaminant Constituents Analyzed The list of constituents analyzed in the stormwater samples is shown in Table 4-2. The vendor's performance claim addresses reductions of sediments, nutrients (total phosphorus) and heavy metals from the runoff water. Table 4-2. Constituent List for Water Quality Monitoring Parameter Total dissolved solids (TDS) Total suspended solids (TSS) Total phosphorus Suspended sediment concentration (SSC) Total calcium Total copper Dissolved copper Total magnesium Dissolved zinc Total zinc Dissolved phosphorus Dissolved cadmium Total cadmium Total lead Dissolved lead Dissolved chloride Chemical oxygen demand (COD) Sand-silt split Five point sedigraph Sand fractionation Reporting Units mg/L mg/L mg/L as P mg/L mg/L Mg/L Mg/L mg/L Mg/L Mg/L mg/L as P Mg/L Mg/L Mg/L Mg/L mg/L mg/L NA NA NA Limit of Limit of Detection Quantification Method1 50 2 0.005 0.1 0.2 1 1 0.2 16 16 0.005 6 6 31 31 0.6 9 NA NA NA 167 7 0.016 0.5 0.7 3 3 0.7 50 50 0.016 20 20 100 100 2 28 NA NA NA SM 2540C EPA 160.2 EPA 365.1 ASTMD3977-97 EPA 200.7 SM3113B SM3113B EPA 200.7 EPA 200.7 EPA 200.7 EPA 365.1 EPA 200.7 EPA 200.7 EPA 200.7 EPA 200.7 EPA 325. 2 ASTMD1252-88(B) Fishman et al. Fishman et al. Fishman et al. 1 EPA: EPA Methods and Guidance for the Analysis of Water procedures; SM: Standard Methods for the Examination of Water and Wastewater (19th edition) procedures; ASTM: American Society of Testing and Materials procedures; Fishman et al.: Approved Inorganic and Organic Methods for the Analysis of Water and Fluvial Sediment procedures. 15 ------- 4.4 Sampling Schedule USGS personnel installed the monitoring equipment under a contract with the WDNR. The monitoring equipment was installed in the December of 2001. In March through May 2002, several trial events were monitored and the equipment tested and calibrated. Verification testing began in June 2002, and ended in November 2003. Table 4-3 summarizes the sample collection data from the storm events. These storm events met the requirements of a "qualified event," as defined in the VTP: 1. The total rainfall depth for the event, measured at the site rain gauge, was 0.2 inches (5 mm) or greater (snow fall and snow melt events did not qualify). 2. Flow through the treatment device was successfully measured and recorded over the duration of the runoff period. 3. A flow-proportional composite sample was successfully collected for both the influent and effluent over the duration of the runoff event. 4. Each composite sample collected was comprised of a minimum of five aliquots, including at least two aliquots on the rising limb of the runoff hydrograph, at least one aliquot near the peak, and at least two aliquots on the falling limb of the runoff hydrograph. 5. There was a minimum of six hours between qualified sampling events. Table 4-4 summarizes the storm data for the qualified events. Detailed information on each storm's runoff hydrograph and the rain depth distribution over the event period are included in Appendix B. The sample collection starting times for the influent and effluent samples, as well as the number of sample aliquots collected, varied from event to event. The influent sampler was activated when the influent velocity meter sensed flow in the pipe. The effluent sampler was activated when the filtration process discharged treated effluent. Twenty events are reported in this verification, as shown in Tables 4-3 and 4-4. At the onset of the monitoring program, the site was not monitored under the ETV program. Both TSS and SSC were being analyzed, but due to budgetary concerns, TSS was discontinued and not sampled for five events (events 3 through 7). Once the monitoring program was entered into the ETV program, the TSS parameter was reinstated, and the monitoring program was extended so that TSS and SSC data was collected for 15 events. The extension of the verification program resulted in the collection of flow data for 20 events and analytical data for other parameters for 15 or more events. 16 ------- Table 4-3. Summary of Events Monitored for Verification Testing Inlet Sampling Point (Site 1) Outlet Sampling Point (Site 2) Event Start Start End End No. of Start Start End End No. of Number Date Time Date Time Aliquots Date Time Date Time Aliquots 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 6/21/02 7/8/02 8/21/02 9/2/02 9/18/02 9/29/02 12/18/02 4/19/03 5/4/03 5/30/03 6/8/03 6/27/03 7/4/03 7/8/03 9/12/03 9/14/03 9/22/03 10/14/03 10/24/03 1 1/4/03 6:54 21:21 20:12 5:25 5:31 2:52 1:19 5:56 21:28 19:00 3:30 17:32 7:27 9:52 15:35 5:34 2:29 1:11 16:59 15:58 6/21/02 7/8/02 8/22/02 9/2/02 9/18/02 9/29/02 12/18/02 4/19/03 5/5/03 5/30/03 6/8/03 6/28/03 7/6/03 7/8/03 9/12/03 9/14/03 9/22/03 10/14/03 10/24/03 1 1/4/03 7:40 23:41 12:37 9:48 10:25 9:27 6:02 15:55 7:18 23:22 14:55 11:01 9:47 13:45 17:31 12:05 4:54 3:21 21:49 19:20 7 29 30 21 10 9 18 18 23 13 14 18 19 8 8 15 8 15 20 10 6/21/02 7/8/02 8/21/02 9/2/02 9/18/02 9/29/02 12/18/02 4/19/03 5/4/03 5/30/03 6/8/03 6/27/03 7/4/03 7/8/03 9/12/03 9/14/03 9/22/03 10/14/03 10/24/03 1 1/4/03 6:57 21:24 20:27 5:30 5:54 3:19 1:44 6:04 21:35 19:05 3:32 17:43 7:30 9:59 16:12 6:11 2:36 1:25 17:10 16:18 6/21/02 7/8/02 8/22/02 9/2/02 9/18/02 9/29/02 12/18/02 4/19/03 5/5/03 5/30/03 6/8/03 6/28/03 7/6/03 7/8/03 9/12/03 9/14/03 9/22/03 10/14/03 10/24/03 1 1/4/03 7:34 23:26 12:21 9:12 10:49 9:33 6:05 15:57 7:18 23:59 15:10 11:34 10:26 14:06 18:23 12:10 4:35 3:34 22:19 19:48 7 29 16 24 8 16 9 15 26 15 20 22 26 11 7 11 13 10 20 14 17 ------- Table 4-4. Rainfall Summary for Monitored Events Event Start Number Date 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 6/21/02 7/8/02 8/21/02 9/2/02 9/18/02 9/29/02 12/18/02 4/19/03 5/4/03 5/30/03 6/8/03 6/27/03 7/4/03 7/8/03 9/12/03 9/14/03 9/22/03 10/14/03 10/24/03 1 1/4/03 Start Time 6:54 21:16 20:08 5:24 5:25 0:49 1:18 5:39 21:21 18:55 3:26 17:30 7:25 9:49 15:33 5:22 2:28 1:03 16:46 16:14 End Date 6/21/02 7/8/02 8/22/02 9/2/02 9/18/02 9/29/02 12/18/02 4/19/03 5/5/03 5/30/03 6/8/03 6/28/03 7/6/03 7/8/03 9/12/03 9/14/03 9/22/03 10/14/03 10/24/03 1 1/4/03 End Time 7:17 23:20 12:07 8:48 10:19 8:43 5:05 15:39 9:05 23:01 14:35 10:55 10:08 13:26 17:28 11:57 4:37 3:10 11:53 18:23 Rainfall Amount (inches) 0.52 1.5 1.7 1.2 0.37 0.74 0.37 0.55 0.90 0.54 0.62 0.57 0.53 0.33 0.22 0.47 0.27 0.25 0.71 0.60 Rainfall Duration (hr:min) 0:23 2:04 15:59 3:24 4:54 7:54 3:47 10:00 11:44 4:06 11:09 13:25 40:43 3:37 1:55 6:35 2:09 2:07 15:07 2:09 Peak Runoff Discharge Volume Rate (ft3)1 (gpm)1 420 1,610 1,620 1,180 350 730 300 340 540 320 450 460 550 260 150 340 270 220 410 560 447 651 671 164 136 70.9 61.0 96.9 73.2 83.9 140 107 143 62.8 21.5 264 104 56.5 75.8 906 1 Runoff volume and peak discharge volume measured at the outlet monitoring point. 4.5 Field Procedures for Sample Handling and Preservation Data gathered by the on-site datalogger were accessible to USGS personnel by means of a modem and phone-line hookup. USGS personnel collected samples and performed a system inspection after storm events. Water samples were collected with ISCO automatic samplers programmed to collect one-liter aliquots during each sample cycle. A peristaltic pump on the sampler pumped water from the sampling location through Teflon™-lined sample tubing to the pump head where water passed through approximately three feet of silicone tubing and into one of four 10-liter sample collection bottles. Samples were capped and removed from the sampler after the event by the WisDOT or USGS personnel depending upon the schedule of the staff. The samples were forwarded to USGS personnel if the WisDOT personnel collected them. The samples were then transported to the USGS field office in Madison, Wisconsin, where they were split into multiple 18 ------- aliquots using a 20-liter Teflon-lined churn splitter. When more than 20 liters (two 10-liter sample collection bottles) of sample were collected by the autosamplers, the contents of the two full sample containers would be poured into the churn, a portion of the sample in the churn would be discarded, and a proportional volume from the third sample container would be poured into the churn. The analytical laboratories provided sample bottles. Samples were preserved per method requirements and analyzed within the holding times allowed by the methods. Particle size and SSC samples were shipped to the USGS sediment laboratory in Iowa City, Iowa (after event 2, SSC samples were analyzed at WSLH). All other samples were hand-delivered to WSLH. The samples were maintained in the custody of the sample collectors, delivered directly to the laboratory, and relinquished to the laboratory sample custodian(s). Custody was maintained according to the laboratory's sample handling procedures. To establish the necessary documentation to trace sample possession from the time of collection, field forms and lab forms (see Appendix B of the VTP) were completed and accompanied each sample. 19 ------- Chapter 5 Monitoring Results and Discussion The monitoring results related to contaminant reduction over the events are reported in two formats: 1. Efficiency ratio comparison, which evaluates the effectiveness of the system on an event mean concentration (EMC) basis. 2. Sum of loads (SOL) comparison, which evaluates the effectiveness of the system on a constituent mass (concentration times volume) basis. The StormFilter is designed to remove suspended solids from wet-weather flows. The VTP required that a suite of analytical parameters, including solids, metals, and nutrients, be evaluated because of the vendor's performance claim. 5.1 Monitoring Results: Performance Parameters 5.1.1 Concentration Efficiency Ratio The concentration efficiency ratio reflects the treatment capability of the device using the event mean concentration (EMC) data obtained for each runoff event. The concentration efficiency ratios are calculated by: Efficiency ratio = 100 x (l-[EMCeffluent/EMCinfiuent]) (5-1) The influent and effluent sample concentrations and calculated efficiency ratios are summarized by analytical parameter categories: sediments (TSS, SSC, and TDS); nutrients (total and dissolved phosphorus); metals (total and dissolved copper, total and dissolved zinc, total lead and total cadmium); and water quality parameters (COD, dissolved chloride, total calcium and total magnesium). The water quality parameters were not specified in the vendors' performance claim and were monitored for other reasons outside the scope of the ETV program. Sediments: The influent and effluent sample concentrations and calculated efficiency ratios for sediment parameters are summarized in Table 5-1. As discussed in Section 4.4, TSS analysis was not conducted on the samples collected from events 3 through 7. The TSS inlet concentrations ranged from 29 to 780 mg/L the outlet concentrations ranged from 20 to 380 mg/L, and the efficiency ratio ranged from -33 to 95 percent. The SSC inlet concentrations ranged 51 to 5,600 mg/L, the outlet concentrations ranged from 12 to 370 mg/L, and the efficiency ratio ranged from 3 to 99 percent. 20 ------- Table 5-1. Monitoring Results and Efficiency Ratios for Sediment Parameters Event No. I1 21 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Rainfall (in) 0.52 1.5 1.7 1.2 0.37 0.74 0.37 0.55 0.90 0.54 0.62 0.57 0.53 0.33 0.22 0.47 0.27 0.25 0.71 0.60 Inlet (mg/L) 71 51 NA NA NA NA NA 780 73 110 60 77 29 57 700 50 37 35 67 55 TSS Outlet Reduction Inlet (mg/L) (Percent) (mg/L) 83 28 NA NA NA NA NA 380 34 70 40 46 30 24 36 49 31 20 36 73 -17 45 - - - - - 51 53 36 33 40 -3 58 95 2 16 43 46 -33 370 310 65 320 120 140 770 5,600 830 1,300 420 370 51 74 3,800 410 480 410 420 100 ssc Outlet Reduction Inlet (mg/L) (Percent) (mg/L) 63 20 19 13 43 12 130 370 34 68 40 47 32 23 29 49 21 21 33 97 83 94 71 96 64 91 83 93 96 95 90 87 37 69 99 88 96 95 92 3 <50 <50 <50 39 NA <50 600 520 78 66 <50 90 60 82 210 <50 50 50 <50 <50 IDS Outlet (mg/L) <50 <50 <50 38 NA <50 4,200 720 90 130 76 160 110 110 190 60 80 74 60 <50 Reduction (Percent) - - - 3 - - -600 -38 -15 -91 - -80 -83 -34 10 - -60 -48 - - 1 SSC analyzed at USGS Sediment Laboratory; all other parameters analyzed at WSLH NA: Not Analyzed 21 ------- The results show a large difference between inlet TSS and SSC concentrations. In each event where both parameters are analyzed, inlet SSC concentrations range from 30 percent to almost 1,200 percent higher than the equivalent TSS concentration. Both the TSS and SSC analytical parameters measure sediment concentrations in water; however, the TSS analytical procedure requires the analyst to draw an aliquot from the sample container, while the SSC procedure requires use of the entire contents of the sample container. If a sample contains a high concentration of settleable (large particle size) solids, acquiring a representative aliquot from the sample container is very difficult. Therefore a disproportionate amount of the settled solids may be left in the container, and the reported TSS concentration would be lower than SSC. The highest concentrations of influent TDS concentrations were observed from events 7 and 8. These two events occurred during the winter (12/18/02 and 4/19/03 respectively) and were likely influenced by road salting operations. This explanation is supported by the high chloride concentrations observed in the inlet samples for these two events (see Table 5-4). Nutrients: The inlet and outlet sample concentrations and calculated efficiency ratios are summarized in Table 5-2. The total phosphorus inlet concentration ranged from 0.05 mg/L to 0.63 mg/L, and the dissolved phosphorus inlet concentration ranged from 0.014 mg/L to 0.20 mg/L. Reductions in total phosphorus EMCs ranged from 0 to 70 percent. Dissolved phosphorus EMCs ranged from -35 to 38 percent. Most of the inlet and outlet dissolved phosphorus concentrations were close to the 0.005 mg/L (as P) detection limit, with little, if any, differences between inlet and outlet concentrations. Metals: The inlet and outlet sample concentrations and calculated efficiency ratios are summarized in Table 5-3. Reductions in metal EMCs followed a similar pattern as the phosphorus results, in that the total fraction all showed higher concentrations and greater EMC reductions than the dissolved faction. The total copper inlet concentration ranged from 15 to 440 ng/L, and the EMC reduction ranged from 8 to 96 percent. The total zinc inlet concentration ranged from 77 to 1,400 |ig/L, and the EMC reduction ranged from 20 to 89 percent. Total zinc and total copper inlet concentrations exhibited field precision, as measured by a statistical analysis of field duplicate samples, that was outside a range identified as acceptable in the test plan. This is explained in greater detail in Section 6.1.2. The dissolved copper inlet concentration ranged from less than 5 to 58 |ig/L, and the EMC reduction ranged from -47 to 64 percent. The dissolved zinc inlet concentration ranged from 26 to 360 |ig/L, and the EMC reduction ranged from -86 to 56 percent. The total and dissolved cadmium and dissolved lead concentrations in both the inlet and outlet samples were below detection limits for every sampled storm event. Total lead concentrations were below detection limits in both the inlet and outlet samples for nine of the sampled events, while the EMC ranged from 33 to 91 percent for the seven events where total lead was detected in the inlet sample. Water quality parameters: inlet and outlet sample concentrations and calculated efficiency ratios for water quality parameters are summarized in Table 5-4. High dissolved chloride concentrations in both the inlet and outlet were observed for events 7 and 8 (12/18/02 and 4/19/03). The likely source of the chloride is the winter application of road salt to the highway. Aside from these two events, dissolved chloride concentrations in the inlet and outlet samples were relatively low, and the StormFilter system did not remove dissolved chloride. 22 ------- Table 5-2. Monitoring Results and Efficiency Ratios for Nutrient Parameters Total Phosphorus Dissolved Phosphorus Inlet Outlet Reduction Inlet Outlet Reduction Event No.1 (mg/L as P) (mg/L as P) (Percent) (mg/L as P) (mg/L as P) (Percent) 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 0.14 0.11 0.05 0.10 0.14 0.10 0.33 0.50 0.17 0.20 0.19 0.24 0.16 0.63 0.10 0.15 0.15 0.10 0.08 0.04 0.05 0.10 0.03 0.20 0.29 0.08 0.14 0.08 0.19 0.11 0.30 0.10 0.10 0.10 29 27 20 50 29 70 39 42 53 30 58 21 31 52 0 33 33 0.041 0.041 0.014 0.030 0.059 0.021 0.035 0.027 0.057 0.045 0.023 0.061 0.048 0.20 0.020 0.043 0.040 0.039 0.037 0.013 0.032 0.046 0.021 0.029 0.017 0.043 0.028 0.028 0.059 0.049 0.19 0.027 0.054 0.046 4.9 9.8 7.1 -6.7 22 0.0 17 37 25 38 -22 3.3 -2.1 5.0 -35 -26 -15 1 Phosphorus parameters were not analyzed during events 13, 19 or 20. 23 ------- Table 5-3. Monitoring Results and Efficiency Ratios for Metals Total Copper Event No.1 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 Inlet2 (Hi/L) 41 34 15 29 130 16 130 280 44 79 36 48 36 330 32 440 46 Dissolved Copper Outlet Reduction Inlet (ug/L) (Percent) (ug/L) 28 19 10 10 30 7 78 140 20 42 23 44 29 69 21 18 15 32 44 33 66 77 56 40 50 55 47 36 8 19 79 34 96 67 <5 10 6.1 7.7 21 5.0 14 28 11 17 18 20 13 58 5.5 9.0 50 Outlet Reduction Inlet2 (ug/L) (Percent) (ug/L) <5 8.8 5.4 7.0 14 4.5 20 27 8.7 15 7.6 23 15 42 6.2 11 18 - 12 11 9 33 10 -47 3 24 10 58 -13 -14 27 -13 -17 64 220 200 180 200 680 77 390 1,400 230 240 120 200 230 1,400 180 650 300 Total Zinc Dissolved Zinc Outlet Reduction Inlet (ug/L) (Percent) (ug/L) 140 76 39 56 110 28 300 540 91 140 84 160 79 210 110 69 66 36 62 78 72 84 64 23 61 60 42 30 20 66 85 39 89 78 60 59 27 49 87 26 59 110 64 67 37 81 57 360 26 42 46 Outlet Reduction (ug/L) (Percent) 34 51 20 43 51 16 110 84 45 70 32 96 42 160 30 47 42 43 14 26 12 41 38 -86 24 30 -4 14 -19 26 56 -15 -12 9 1 Metals parameters were not analyzed during events 13, 19 or 20. 2 Total copper and total lead inlet data exhibited precision (field duplicates) outside the targeted goal of 25 percent (see discussion in Section 6.1.2). 24 ------- Table 5-3 (cont'd). Total Cadmium Dissolved Cadmium Total Lead Dissolved Lead Event No.1 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 Inlet (ng/L) <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 Outlet Reduction (jig/L) (percent) NA <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 Inlet (ng/L) <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 Outlet Reduction (jig/L) (percent) NA <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 <6 Inlet (ng/L) <31 <31 <31 <31 <31 <31 130 190 <31 53 33 <31 <31 280 140 110 <31 Outlet (ng/L) NA <31 <31 <31 <31 <31 72 <31 <31 32 <31 <31 <31 37 94 53 <31 Reduction Inlet (percent) (jig/L) <31 <31 <31 <31 <31 <31 45 <31 91 <31 <31 40 <31 52 <31 <31 <31 87 <31 33 <31 52 <31 <31 Outlet Reduction (jig/L) (percent) NA <31 <31 <31 <31 <31 <31 <31 <31 <31 <31 <31 <31 <31 <31 <31 <31 1 Metals parameters were not analyzed during events 13, 19 or 20. NA: Not analyzed 25 ------- Table 5-4. Monitoring Results and Efficiency Ratios for Water Quality Parameters Event No.1 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 Chemical Oxvsen Demand Dissolved Chloride Inlet (mg/L) 42 39 18 29 80 28 68 320 53 67 41 85 63 300 38 48 51 Total Calcium Outlet Reduction Inlet Outlet Reduction Inlet (mg/L) (Percent) (mg/L) (mg/L) (Percent) (mg/L) 37 25 24 24 78 17 130 190 38 61 36 81 53 160 34 72 50 12 36 -33 17 2.5 39 -91 41 28 9.0 12 4.7 16 47 11 -50 2.0 5.8 4.6 4.5 3.2 NA 3.6 310 470 25 14 9.4 17 20 34 6.1 9 5.4 5.2 4.6 3.4 3.3 NA 4.0 2,600 660 31 32 17 35 22 35 9.7 16 NA 10 0 24 -3 - -11 -740 -40 -24 -130 -81 -110 -10 -3 -59 -78 - 42 28 9.7 55 17 9.4 130 430 62 40 37 29 12 230 41 73 60 Total Magnesium Outlet Reduction Inlet Outlet Reduction (mg/L) (Percent) (mg/L) (mg/L) (Percent) 15 6 4.4 4.4 9.7 4 48 68 11 17 9.6 17 8.9 16 8.8 8.3 7 64 79 55 92 43 57 63 84 82 58 74 41 26 93 79 89 88 21 14 4.2 26 7.3 4.0 56 174 28 18 18 11 4.9 120 20 36 22 5.8 1.9 1.6 1.4 3.2 1.1 8.5 26 2.8 4.8 3.0 4.2 2.3 4.4 3.7 2.5 1.9 72 86 62 95 56 73 85 85 90 73 83 62 53 96 82 93 91 1 Parameters were not analyzed during events 13, 19 or 20. NA: Not Analyzed 26 ------- 5.1.2 Sum of Loads The sum of loads (SOL) is the sum of the percent load reduction efficiencies for all the events, and provides a measure of the overall performance efficiency for the events sampled during the monitoring period. The load reduction efficiency is calculated using the following equation: % Load Reduction Efficiency = 100x(l-(A/B)) (5-2) where: A = Sum of Effluent Load = (Effluent EMCi)(Flow Volumei) + (Effluent EMC2)(Flow Volume2) + (Effluent EMCn)(Flow Volumen) B = Sum of Influent Load = (Influent EMCi)(Flow Volumei) + (Effluent EMC2)(Flow Volume2) + (Effluent EMCn)( Flow Volumen) n= number of qualified sampling events Flow calibration: Before the flow and concentration results could be used for calculating the inlet and outlet sediment loads, the flow rate calculations were modified based on calibration of the flow meters, correction to the velocity data, and corrections for the gauge heights. A discussion describing these calibration procedures is in Chapter 6. These modifications made significant changes to the volumes used for the inlet and outlet of the StormFilter. After these adjustments were made to the velocity and flow measurements, the event volumes at the inlet and outlet sites were compared. Low variability was observed between the inlet and outlet volumes for each storm. Differences between the volumes were 15 percent or less for 17 of the 20 storms. The average difference between the inlet and outlet volumes was 11 percent. There was not a trend as to whether the inlet or outlet flow volumes were larger. Although the volumes were close, the differences could still influence the SOL calculations. With perfect measurements, the inlet and outlet volumes should be exactly the same, since there is no place the water could be lost in the treatment system. It was decided that the outlet volumes would best represent the flows at both the outlet and inlet. The outlet volumes are considered more accurate because the inlet experienced most of the missing velocity data (see Section 6.2). If the missing velocity data was the result of higher solids concentrations and/or much higher velocities at the inlet, these characteristics could make the inlet flow measurements less reliable than the outlet measurements. Air entrapment caused by high velocities over the top of the velocity probe could also cause a disturbance in the probe's electromagnetic signal. To demonstrate the impact of using the volume calculations at each site, all three possible combinations for the sediment results are presented below: using outlet volumes to calculate loads at both sites; using inlet volumes to calculate loads at each site, and using the respective inlet and outlet volumes to calculate loads at each site. Table 5-5 demonstrates that using the different load calculation methods had little impact on the resulting SOL calculations for the sediment parameters. For this reason, the loads for the remaining parameters (metals, nutrients, and other parameters) are calculated only using the outlet volumes for each site. 27 ------- Table 5-5. Sediment Sum of Loads Efficiencies Calculated Using Various Flow Volumes Flow Location Inlet only Outlet only Inlet and Outlet Load Reduction Efficiency (Percent)1 TSS SSC TDS 47 92 -45 46 92 -46 50 93 -38 1 Load reduction efficiencies were calculated without data from events 3 through 7, when no TSS samples were collected (see Section 4.4). Sediment: Table 5-6 summarizes results for the SOL calculations analysis using three approaches: all events reported and all parameters; results for SSC samples for those events with data from TSS, TDS and SSC parameters (does not include events 3 through 7); and results for TDS samples for all events except for an apparent outlier (event 7, likely influenced by application of road salt). These results show no significant difference between the SOL reductions of SSC. By eliminating event 7 from the TDS SOL calculations, the SOL reduction improves from -170 percent to -37 percent. The SOL analyses indicate a TSS reduction of 47 to 50 percent, and SSC reduction of 92 to 93 percent. The TSS load reduction nearly meets SMFs performance claim of 50 to 85 percent TSS reduction, while SSC reduction exceeds the performance claim. The large discrepancy in TSS versus SSC is likely due to the large particle sizes found in the runoff (see Section 5.2) and the methodology difference between the two analytical procedures. Analytical procedures for TSS require an aliquot to be removed from the sample container. When larger sediment particles are in the sample container, it is unlikely (even when the container is stirred) that the larger particles will be evenly distributed throughout the container, making the aliquot not representative of the sediment in the sample. SSC analytical procedures require the entire volume of sample to be analyzed for sediment volume, eliminating this issue. Nutrients: The SOL data for nutrients are summarized in Table 5-7. The total phosphorus load reduction of 38 percent met SMFs performance claim of 30 to 45 percent reduction. Additionally, the dissolved phosphorus load reduction of six percent also met SMFs performance claim of negligible dissolved phosphorus removal. 28 ------- Table 5-6. Sediment Sum of Loads Results TSS „ „ Runoff Inlet Inlet Volume (ft3) (mg/L) (Ib) 1* 420 2* 1,610 3 1,620 4 1,180 5 350 6 730 7 300 8 340 9 540 10 320 11 450 12 460 13 550 14 260 15 150 16 340 17 270 18 220 19 410 20 560 71 1.9 51 5.2 NA NA NA NA NA 780 17 73 2.5 110 2.3 60 1.7 77 2.2 29 1.0 57 0.9 700 6.6 50 1.1 37 0.6 35 0.5 67 1.7 55 1.9 Outlet (mg/L) 83 28 NA NA NA NA NA 380 34 70 40 46 30 24 36 49 31 20 36 73 Total (all events monitored) 47 Load Reduction Efficiency SSC Total (omitting events Load Reduction Efficiency TDS Total (omitting event Load Reduction Efficiency (Percent) 3-7) (Percent) 7) (Percent) Outlet (Ib) 2.2 2.8 - - - - - 8.1 1.2 1.4 1.1 1.3 1.0 0.4 0.3 1.0 0.5 0.3 0.9 2.6 25 46 Inlet (mg/L) 370 310 65 320 120 140 770 5,600 820 1,300 420 370 51 74 3,800 400 480 410 420 100 SSC Inlet (Ib) 9.8 32 6.6 24 2.6 6.3 14 120 28 26 12 11 1.8 1.2 35 8.7 8.2 5.7 11 3.6 370 314 Outlet (mg/L) 63 20 19 13 43 12 130 370 34 68 40 47 32 23 29 49 21 21 33 97 Outlet (Ib) 1.7 2.0 1.9 1.0 0.9 0.6 2.4 8.0 1.2 1.4 1.1 1.4 1.1 0.4 0.3 1.0 0.4 0.3 0.9 3.4 31 92 24 92 Inlet (mg/L) <50 <50 <50 39 NA <50 600 520 78 66 <50 90 60 82 210 <50 50 50 <50 <50 TDS Inlet (Ib) 0.7 2.5 2.5 2.9 - 1.1 11 11 2.6 1.3 0.7 2.6 2.1 1.3 2.0 0.5 0.8 0.7 0.6 0.9 48 37 Outlet (mg/L) <50 <50 <50 38 NA <50 4,200 720 90 130 76 160 110 110 190 60 80 74 60 <50 Outlet (Ib) 0.7 2.5 2.5 2.8 - 1.1 79 15 3.1 2.5 2.1 4.7 3.8 .8 .8 .3 .4 .0 .5 0.9 130 -170 51 -37 * SSC Analyzed at USGS Sediment Laboratory NA Not Analyzed Italicized numbers represent results where one-half the method detection limit was substituted for values below detection limits. 29 ------- Table 5-7. Nutrient Sum of Loads Results Event No. Total Phosphorus (g) Inlet Outlet Dissolved Phosphorus (g) Inlet Outlet 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 Total: Load Reduction (Percent): 1.7 4.8 2.1 3.3 1.4 2.0 2.8 4.8 2.6 1.8 2.5 3.0 1.2 2.6 1.0 1.2 0.91 40 Efficiency 1.2 3.6 1.7 1.6 1.0 0.67 1.7 2.8 1.2 1.3 1.0 2.5 0.79 1.2 0.91 0.74 0.60 24 38 0.49 1.87 0.64 1.00 0.59 0.44 0.30 0.26 0.88 0.41 0.29 0.79 0.35 0.83 0.19 0.33 0.24 9.9 0.47 1.68 0.60 1.06 0.46 0.44 0.25 0.16 0.66 0.25 0.36 0.77 0.36 0.80 0.26 0.41 0.28 9.3 6 Metals: The SOL data for metals are summarized in Table 5-8. The total zinc (64 percent) and total copper (60 percent) load reductions met or exceeded the 30 to 60 percent performance claim for these constituents. Total zinc and total copper inlet concentrations exhibited field precision, as measured by a statistical analysis of field duplicate samples, that was outside a range identified as acceptable in the test plan. This is explained in greater detail in Section 6.1.2. The dissolved zinc (17 percent) and dissolved copper (16 percent) load reduction were lower than the 20 to 40 percent performance claim for these constituents. The dissolved zinc and copper influent concentrations were relatively low for most events. Load reduction for dissolved zinc with influent concentrations greater than 100 |ig/L was 42 percent and load reduction dissolved copper with influent concentrations greater than 50 |ig/L was 50 percent. There were no performance claims reported for total lead or total cadmium. 30 ------- Table 5-8. Metals Sum of Loads Results Event Total Copper (g) No- Inlet1 Outlet Dissolved Copper (g) Inlet Outlet Total Zinc (g) Inlet1 Outlet Dissolved Zinc (g) Total Lead (g) Inlet Outlet Inlet Outlet 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 Total: 4.9 16 6.9 9.6 13 3.3 11 26 6.8 7.2 4.6 6.2 2.6 14 3.1 33 2.8 171 3.4 8.6 4.6 3.3 3.0 1.5 6.7 13 3.1 3.8 2.9 5.7 2.1 2.9 2.0 1.4 0.9 69 - 0.37 0.24 0.21 0.18 0.09 0.12 0.36 0.23 0.22 0.26 0.27 0.10 0.30 0.06 0.06 0.30 3.4 - 0.32 0.21 0.19 0.12 0.08 0.18 0.35 0.18 0.19 0.11 0.31 0.12 0.21 0.07 0.07 0.11 2.8 27 92 81 66 68 16 34 130 36 22 15 26 17 57 18 49 18 771 17 35 18 19 11 5.8 26 51 14 13 11 21 5.8 8.9 10 5.2 4.0 274 0.73 2.17 1.1 1.3 0.76 0.46 0.52 1.4 1.4 0.85 0.54 1.1 0.45 1.8 0.29 0.27 0.27 15 0.41 1.9 0.79 1.2 0.45 0.28 0.97 1.1 0.96 0.89 0.47 1.3 0.33 0.82 0.33 0.31 0.25 12 - - - - - - - 1.1 0.63 2.5 0.20 - 0.67 0.41 0.49 0.23 - - 1.4 0.19 1.5 1.0 0.72 0.34 8.5 3.0 Load Reduction Efficiency (Percent): 59 16 64 17 64 2 Total copper and total lead inlet data exhibited precision (field duplicates) outside the targeted goal of 25 percent (see discussion in Section 6.1.2). Italicized numbers represent results where one-half the method detection limit was substituted for values below detection limits. Note: total and dissolved cadmium and dissolved lead SOL calculations were not conducted because all values were below detection limits. 31 ------- Water quality parameters: The SOL data for water quality parameters are summarized in Table 5-9. The StormFilter system achieved a 16 percent load reduction for COD, a 79 percent load reduction for total calcium, and an 85 percent load reduction for total magnesium. The negative load reduction (-242 percent) for dissolved chloride was influenced by high effluent concentrations during events 7 and 8 (December 2002 and April 2003). These events were likely biased by earlier applications of road salt for deicing. SMI did not make any performance claims for these parameters. Table 5-9. Water Quality Parameter Sum of Loads Results Event No. 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 Total: COD (Ib) Inlet Outlet 1.1 3.9 1.8 2.1 1.8 1.3 1.3 6.7 1.8 1.3 1.2 2.4 1.0 2.8 0.8 0.8 0.7 33 Load Reduction Efficiency (Percent): 1.0 2.5 2.5 1.8 1.7 0.8 2.5 4.0 1.3 1.2 1.0 2.3 0.9 1.5 0.7 1.2 0.7 28 16 Dissolved Chloride Total and Reduction Efficiency (omitting events 7 and 8) Dissolved Chloride (Ib) Inlet Outlet 0.15 0.46 0.46 0.24 NA 0.17 5.93 9.9 0.86 0.29 0.27 0.48 0.32 0.32 0.13 0.15 0.07 20 4.4 0.14 0.46 0.35 0.24 NA 0.18 49 14 1.1 0.65 0.49 1.00 0.36 0.32 0.21 0.27 NA 69 -240 5.7 -31 Total Calcium (Ib) Inlet Outlet 1.1 2.8 0.99 4.0 0.38 0.43 2.5 9.2 2.1 0.8 1.1 0.84 0.20 2.2 0.86 1.2 0.81 31.5 0. 0. 0. 0. 0. 0. 0. 1 0. 0. 0. 0. 0. 0. 0. 0. 0. 39 61 45 32 22 18 90 .4 36 33 27 50 15 15 19 14 10 6.70 Total Magnesium (Ib) Inlet Outlet 0. 56 1.4 0. 1 0. 0. 1 3 0. 0. 0. 0. 0. 1 0. 0. 0. 43 .9 16 18 .1 .7 94 36 51 32 08 .1 42 61 30 14.1 79 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 2 15 19 16 10 07 05 16 55 10 10 08 12 04 04 08 04 03 .1 85 NA: not analyzed 32 ------- 5.2 Particle Size Distribution Particle size distribution analysis was conducted on selected events. Three types of analyses were conducted. The ability of the lab to conduct the specific analysis depended on the available sample volume, the sediment concentration, and the particle sizes in the sample. The ISCO samplers did not always collect an adequate volume of sample to conduct the full suite of particle size analyses. 1. A "sand/silt split" analysis determined the percentage of sediment (by weight) larger than 62 |im (defined as sand) and less than 62 jim (defined as silt). This analysis was performed on the outlet samples of events 3 4, 6, 15, and 16. 2. A Visual Accumulator (VA) tube analysis (Fishman et al., 1994) defined the percent of sediment (by weight) sized less than 1000, 500, 250, 125, and 62 jim. The analyses were conducted on the inlet and outlet samples of events 1, 2, and 9, and on the inlet samples of events 4, 6, 15, and 16. 3. A pipette analysis (Fishman et al., 1994) was conducted to further define the silt portion of a sample as the percent of sediment (by weight) sized less than 31, 16, 8, 4, and 2 jim. This analysis was conducted on the inlet and outlet samples of events 7 and 8. The particle size distribution results are summarized in Table 5-10. In each event where particle size analysis was conducted, the outlet samples had a higher percentage of particles in the silt category (<62.5 um) than the equivalent inlet sample. This is a result of the filtering mechanism of the StormFilter removing a higher percentage of the larger sediment particles. 33 ------- Table 5-10. Particle Size Distribution Analysis Results Percent Less Than Particle Size Event No, 1 2 3 4 6 7 8 9 15 16 Location Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet1 Inlet Outlet <1000 80 100 52 100 100 71 93 90 90 92 100 90 72 <500 64 100 45 100 73 52 93 61 77 81 81 75 44 <250 36 98 25 100 42 17 58 47 49 34 57 23 23 <125 22 93 12 96 32 9 39 42 34 19 50 4 15 <62.5 <31 18 91 12 88 32 82 8 92 32 91 40 38 100 97 30 26 100 96 15 44 4 13 92 <16 <8 <4 <2 33 25 16 10 96 86 78 66 20 14 11 8 86 66 55 48 1 No data reported due to laboratory error. 34 ------- Chapter 6 QA/QC Results and Summary The Quality Assurance Project Plan (QAPP) in the VTP identified critical measurements and established several QA/QC objectives. The verification test procedures and data collection followed the QAPP. QA/QC summary results are reported in this section, and the full laboratory QA/QC results and supporting documents are presented in Appendix C. 6.1 Laboratory/Analytical Data QA/QC 6,1.1 Bias (Field Blanks) Field blanks were collected at both the inlet and outlet samplers on three separate occasions to evaluate the potential for sample contamination through the entire sampling process, including automatic sampler, sample-collection bottles, splitters, and filtering devices. "Milli-Q" reagent water was pumped through the automatic sampler, and collected samples were processed and analyzed in the same manner as event samples. The first field blank was collected on 04/02/02 (before the first event was sampled), allowing the USGS to review the results early in the monitoring schedule. The second and third field blanks were collected on 11/11/02 (between events 6 and 7) and 6/30/03 (between events 12 and 13), respectively. Results for the field blanks are shown in Table 6-1. All but nine analyses were below the limits of detection (LOD), and all detects were below the limit of quantification (LOQ). These results show a good level of contaminant control in the field procedures was achieved. Table 6-1. Field Blank Analytical Data Summary Parameter TSS ssc TDS COD Dissolved copper Total copper Dissolved zinc Total zinc Dissolved phosphorus Total phosphorus Dissolved chloride Total calcium Total magnesium Units mg/L mg/L mg/L mg/L Mg/L Mg/L Mg/L Mg/L mg/L mg/L mg/L mg/L mg/L Blank 1 (4/2/2002) Inlet Outlet <2 ~ <50 <9 <5 <5 <16 <16 ~ <0.005 3.3 0.7 <0.2 <2 ~ <50 <9 <5 <5 <16 <16 ~ <0.005 <0.6 <0.2 <0.2 Blank 2 (11/11/2002) Inlet Outlet ~ ~ <50 <9 <1 <1 <16 <16 <0.005 0.025 <0.6 <0.2 <0.2 ~ ~ <50 <9 <1 <1 <16 <16 <0.005 <0.005 <0.6 <0.2 <0.2 Blank 3 (6/30/2003) Inlet Outlet <2 <2 <50 12 1.7 2 <16 <16 <0.005 <0.005 0.8 0.2 <0.2 <2 <2 <50 14 2.3 2 <16 <16 <0.005 <0.005 <0.6 <0.2 <0.2 LOD 2 2 50 9 1 1 16 16 0.005 0.005 2 0.2 0.2 LOQ 7 7 167 28 3 3 50 50 0.016 0.016 3.3 0.7 0.7 35 ------- 6.1.2 Replicates (Precision) Precision measurements were performed by the collection and analysis of duplicate samples. The relative percent difference (RPD) recorded from the sample analyses was calculated to evaluate precision. RPD is calculated using the following formula: \X\- X2\ %RPD = - x 100% where: xi = Concentration of compound in sample x_2 = Concentration of compound in duplicate x = Mean value of xi and X2 Field precision: Field duplicates were collected to monitor the overall precision of the sample collection procedures. Duplicate inlet and outlet samples were collected during five different storm events to evaluate precision in the sampling process and analysis. The duplicate samples were processed, delivered to the laboratory, and analyzed in the same manner as the regular samples. Summaries of the field duplicate data are presented in Table 6-2. Overall, the results show good field precision. Below is a discussion on the results from selected parameters. TSS and SSC: Most results were within targeted limits. Outlet samples (lower concentrations and smaller particle sizes) showed higher precision. The SSC inlet sampling had two occurrences of percent RPD exceeding the limit. The poorer precision for the inlet samples could be due to the sample handling and splitting procedures, or sampling handling for analysis, or a combination of factors. Tests conducted by Horowitz, et al. (2001) on the sample splitting capabilities of a churn splitter showed the bias and the precision of the splits is compromised with increasing sediment concentrations and particle size. The tests identified the upper particle size limits for the churn splitter is between 250 and 500 microns (Horowitz, et al, 2001). According to the data summarized in Table 5-10, 63 percent of the particles in inlet samples were greater than 250 microns. Dissolved constituents (sediment phosphorus, and metals): These parameters consistently had very low RPD (very high precision). This supports the idea that the sample splitting operation may be the source of higher RPD in the high particulate samples. Total metals: The total zinc and total copper data generally had the highest discrepancies (highest RPD, or lowest precision). Similar to the particulate sediment results, the highest RPDs occurred in the inlet samples, which had higher particulate concentrations. The total calcium and total magnesium data showed higher precision. Total phosphorus: This parameter was consistently below or near the acceptable RPD value of 30 percent. Again, the highest discrepancies occurred at the inlet analyses, with very good duplicate agreement at the outlet samples. 36 ------- Table 6-2. Field Duplicate Sample Relative Percent Difference Data Summary Parameter Unit TSS mg/L SSC mg/L TDS mg/L Dissolved ug/L copper Total ug/L copper Dissolved ug/L zinc Total ug/L zinc Dissolved mg/L phosphorus Total mg/L phosphorus Total mg/L calcium Total mg/L magnesium 9/19/2002 Rep Rep RPD la Ib (Pet) Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet - - 500 39 <50 <50 8.9 6.8 140 17 35 22 134 61 0.03 0.027 0.16 0.067 16 6.1 7.8 2.5 - - 680 39 52 <50 9.5 8.4 35 18 31 22 328 63 0.031 0.026 0.11 0.065 20 6.2 10 2.5 - - 30 0 NA 0 7 21 120 6 12 0 84 3 3 4 37 3 23 2 26 0 4/19/2003 Rep Rep RPD 2a 2b (Pet) 780 380 5,600 370 520 720 28 27 280 140 110 84 1,400 540 0.027 0.017 0.50 0.29 430 68 170 26 840 380 4,900 370 520 730 28 26 370 140 120 91 2,200 540 0.025 0.016 0.56 0.30 480 68 200 26 7 0 14 0 0 1 0 5 29 0 6 8 46 0 8 6 10 3 9 0 14 0 6/27/2003 Rep Rep RPD 3a 3b (Pet) 77 46 370 47 90 162 20 23 48 44 81 96 200 160 0.061 0.059 0.235 0.19 29 17 11 4.2 96 47 210 48 86 160 21 23 52 46 77 92 320 160 0.063 0.058 0.32 0.19 32 18 12 4.2 22 2 54 2 5 1 6 0 8 4 5 4 48 0 3 2 31 0 9 2 3 0 9/12/2003 Rep Rep RPD 4a 4b (Pet) 700 36 3,800 29 210 190 58 42 330 69 360 160 1,400 220 0.20 0.19 0.63 0.30 230 16 120 4.4 820 31 2,400 32 220 190 59 41 260 68 350 150 1,700 210 0.21 0.19 0.58 0.29 220 16 110 4.2 16 15 44 10 6 0 2 2 25 1 1 3 21 3 3 0 7 4 7 0 9 5 10/14/2003 Rep Rep RPD 5a 5b (Pet) 35 20 410 21 50 74 50 18 46 15 46 42 300 66 0.040 0.046 0.15 0.098 60 7.0 22 1.9 44 25 310 22 <50 58 170 19 130 15 47 43 280 67 0.039 0.046 0.11 0.098 62 7.1 27 2.0 23 22 29 5 0 24 108 6 97 0 2 2 5 2 3 0 35 0 4 1 20 5 Single dash indicates no sample processed for event 37 ------- Laboratory precision: The WSLH analyzed duplicate samples from aliquots drawn from the same sample container as part of their QA/QC program. Summaries of the field duplicate data are presented in Table 6-3. Table 6-3. Laboratory Duplicate Sample Relative Percent Difference Data Summary Average Maximum Minimum Std. Dev. Objective Parameter1 Count2 (percent) (percent) (percent) (percent) (percent) Total calcium Dissolved chloride Dissolved copper Total copper Total magnesium TSS Dissolved phosphorus TDS Total phosphorus Dissolved zinc Total zinc 19 21 12 21 19 16 18 18 20 17 18 1.7 0.69 2.1 1.8 1.2 1.3 1.3 o o J.J 1.4 1.5 1.7 4.6 2.4 8.7 4.6 3.6 3.5 1.6 12 6.4 5.6 3.8 0.19 0.03 0.03 0.09 0.01 0 0 0 0 0.09 0 1.2 0.60 2.9 1.5 1.2 1.1 0.51 o o J.J 1.6 1.4 1.2 25 25 25 25 25 30 30 30 30 25 25 1 Laboratory precision may also be evaluated based on absolute difference between duplicate measurements when concentrations are low. For data quality objective purposes, the absolute difference may not be larger than twice the method detection limit. 2 Analyses where both samples were below detection limits were omitted from this evaluation. The data show that laboratory precision was maintained throughout the course of the verification project. The field and analytical precision data combined suggest that the solids load and larger particle sizes in the inlet samples are the likely cause of poor precision, and apart from the field sample splitting procedures for inlet samples, the verification program maintained high precision. 6.1.3 Accuracy Method accuracy was determined and monitored using a combination of matrix spike/matrix spike duplicates (MS/MSD) and laboratory control samples (known concentration in blank water). The MS/MSD data are evaluated by calculating the deviation from perfect recovery (100 percent), while laboratory control data are evaluated by calculating the absolute value of deviation from the laboratory control concentration. Accuracy was in control throughout the verification test. Tables 6-4 and 6-5 summarize the matrix spikes and lab control sample recovery data, respectively. 38 ------- Table 6-4. Laboratory MS/MSD Data Summary Parameter Total calcium COD Dissolved chloride Total copper Dissolved copper Total magnesium Dissolved phosphorus Total phosphorus Total zinc Dissolved zinc Count 22 20 21 22 14 22 19 19 22 19 Average (percent) 96.5 97.9 101 101 98.5 97.5 102 102 94.9 97.9 Maximum (percent) 113 119 108 116 113 102 106 109 101 114 Minimum (percent) 90.8 79.4 97.3 91.3 90.8 93.0 96.9 97.3 91.0 91.8 Std. Dev. (percent) 5.1 10.3 2.4 7.7 6.1 2.5 2.3 3.2 2.6 5.0 Range (Pet) 85- 75- 90- 80- 85- 85- 90- 90- 85- 85- 115 125 110 120 115 115 110 110 115 115 The balance used for solids (TSS, TDS, and total solids) analyses was calibrated routinely with weights that were NIST traceable. The laboratory maintained calibration records. The temperature of the drying oven was also monitored using a thermometer that was calibrated with an NIST traceable thermometer. Table 6-5. Laboratory Control Sample Data Summary Parameter Total calcium COD Dissolved chloride Total copper Dissolved copper Total magnesium SSC TSS Dissolved phosphorus TDS Total phosphorus Total zinc Dissolved zinc Count 18 20 48 21 36 18 13 12 6 18 24 19 9 Mean (percent) 97 101 100 99 102 98 99 99 101 106 101 97 99 Maximum (percent) 105 107 110 106 110 103 108 120 102 122 108 103 102 Minimum (percent) 93 923 92 91 94 94 87 86 100 94 96 94 97 Std. Dev. (percent) 2.8 3.4 2.8 4.5 3.5 1.9 6.2 9.9 0.5 7.1 2.3 2.1 1.8 39 ------- 6.1.4 Representativeness The field procedures were designed to ensure that representative samples were collected of both influent and effluent stormwater. Field duplicate samples and supervisor oversight provided assurance that procedures were being followed. The challenge in sampling stormwater is obtaining representative samples. The data indicated that while individual sample variability might occur, the long-term trend in the data was representative of the concentrations in the stormwater, and redundant methods of evaluating key constituent loadings in the stormwater were utilized to compensate for the variability of the laboratory data. The laboratories used standard analytical methods, with written SOPs for each method, to provide a consistent approach to all analyses. Sample handling, storage, and analytical methodology were reviewed to verify that standard procedures were being followed. The use of standard methodology, supported by proper quality control information and audits, ensured that the analytical data were representative of actual stormwater conditions. Regarding flow (velocity and stage) measurements, representativeness is achieved in three ways: 1. The meter was installed by experienced USGS field monitoring personnel familiar with the equipment, in accordance with the manufacturer's instructions; 2. The meter's individual area and velocity measurements were converted to a representation of the flow area using manufacturer's conversion procedures (see Chapter 9 of Marsh-McBirney's O&M Manual in Appendix A of the VTP); 3. The flow calculated from the velocity/stage measurements was calibrated using the procedure described in Section 6.2 To obtain representativeness of the sub-samples (aliquots) necessary to analyze the various parameters from the event sample, a churn splitter was used. As noted in Radtke, et al. (1999), the churn splitter is the industry standard for splitting water samples into sub-samples. However, inconsistencies were noted in the sub-samples, especially when the sample contained high concentrations of large-sized sediments. The even distribution of the larger particulates becomes problematic, even with the agitation action of the churn within the splitter (Horowitz, et al, 2001). The issue of the potential for uneven distribution of particulates has not been fully resolved to date. 6.1.5 Completeness The flow data and analytical records for the verification study are 100 percent complete. There were instances of velocity "dropouts" during some events. A discussion of the calibration procedures for flow data (velocity and stage measurements), including how velocity dropouts were addressed, is provided in Section 6.2. 40 ------- 6.2 Flow Measurement Calibration Flow meters at the inlet and outlet of the StormFilter were calibrated on April 20, 2003 and November 8, 2003 using similar procedures. A truck-mounted three-inch Parshall flume was used to calibrate the flow meter at the inlet and outlet pipes. Three 5-horsepower pumps were used to supply water from the Milwaukee River to the flume. Water was pumped into a chamber box before the flume approach to minimize turbulence. The discharge point of the flume was connected to the clean-out access on the storm inlet downspout. Connecting to the access point created some head for flow before it entered the StormFilter system's inlet pipe. Four different pumping rates produced different flow rates, ranging from 0.02 to 0.55 cfs, into the pipe. Even though a large flume was used, its capacity was only sufficient to fill the pipe to about three quarters full. A plot of flume versus flow meter flow rates was created for both the inlet and the outlet, as shown in Figure 6-1. These plots were used to adjust the recorded flow rates. The correction reduced the inlet and outlet flows by 16 percent and 17 percent, respectively. 6.2.1 Inlet- Outlet Volume Comparison This StormFilter configuration did not have an external bypass mechanism, so the calculated influent and effluent event volumes should ideally be the same, and a comparison of the calculated influent and effluent volumes can be used to ensure both flow monitors worked properly. The StormFilter unit does retain a certain amount of water between events, but since this retained volume is constant between events, the net runoff volume into the unit should equal the net runoff volume exiting the unit.. Good agreement was observed between the inlet and outlet volumes for each storm. Differences between the inlet and outlet volumes were 15 percent or less for 17 of the 20 storms. The average difference between the volumes was 11 percent. There was not a trend as to which volume was larger for each storm. Table 6-6 summarizes the volume comparisons for each event. 41 ------- M — ^O, E o> | iZ 0.8 -| n 7 0 fi n *=> n 4 0 3 n 9 0 1 n Riverwalk South Inlet Calibration 04-20-03 y = 0.7789X ,^**^ R2 = 0.9968 ^^- ^^^^ ^^^^^ ^^^>^^ 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Calculated Flow (cfs) (a) April 20, 2003 tf\ <+}. O ***** 5 _o o> E .3 "• 0.8 -, 0 7 n R n 5 n 4 n ^ 0 7 n 1 n c Riverwalk South Inlet Calibration 11-08-03 y = 0.9002x R2 = 0.9722 ^^-^** ^^^^"""^ A* ^^^***^ ^^^^ ^^^*~^ ^^^*^^ —^^+ **^ J 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Calculated Flow (els) (b) November 8, 2003 Figure 6-1. Calibration curves used to correct flow measurements. 42 ------- Table 6-6. Comparison of Inlet and Outlet Event Runoff Volumes Event No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Event Volumes1 Inlet Outlet Difference (ft3) (ft3) (percent) 290 1,700 1,600 1,000 390 730 270 400 610 340 500 420 530 290 160 350 220 210 410 680 420 1,600 1,600 1,200 350 730 300 340 540 320 450 460 550 260 150 340 270 220 410 560 -45 6 0 -20 10 0 -11 15 11 6 10 -10 -4 10 6 O -23 -5 0 18 1 Corrected for point vs. area coefficient, flow calibration, and velocity dropouts. The outlet volumes were considered most accurate because the inlet site experienced the majority of the missing velocity data. Possible reasons for the missing data points could be higher solids concentrations interferes with the velocity meter's capabilities, higher flow velocities at the inlet, or air entrapment at the inlet creating a disturbance in the probe's electromagnetic signal. Because of the more complete velocity data coverage at the outlet site, the outlet volumes were used for the SOL calculations (although SOL calculations for the sediment data are presented for inlet only, outlet only, and inlet and outlet). Section 6.2.4 discusses the corrections applied for the velocity dropout conditions in greater detail. 43 ------- 6.2.2 Gauge Height Calibration Static gauge height measurements were made at the inlet and outlet pipes by constricting the pipe to a steady-state water level. An inflatable ball was used to block the pipe. Water level readings from a measuring tape inside the pipe were compared to the water surface level recorded by the flow meters (located within the inlet and outlet pipes, as described in Section 4). Gauge heights were checked four times during the project. A gauge height correction curve with three gauge height points—bottom, middle, and top (approximately 0.0 ft, 0.3 ft, and 0.6 ft above the invert pipe elevation)—was developed for each pipe, as shown in Table 6-7. Most of the correction factors for the inlet lowered the recorded gauge height by approximately five percent. Corrections for the outlet pipe were also small (less than ±0.05). Table 6-7. Gauge Corrections for Flow Measurements at the Inlet Gauge Height Point 1 Date Gauge Correction Height (ft) (unitless) Gauge Height Point 2 Gauge Correction Height (ft) (unitless) Gauge Height Point 3 Gauge Correction Height (ft) (unitless) 4/01/02 4/11/03 4/11/03 8/14/03 8/14/03 11/8/03 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.002 0.015 -0.005 -0.005 0.318 0.318 0.350 0.250 0.350 0.350 -0.035 -0.035 0.002 0.025 -0.005 -0.005 0.636 0.635 0.635 0.500 0.635 0.635 -0.036 -0.036 0.002 0.033 -0.005 -0.005 6.2.3 Point Velocity Correction Equations have been developed by the flow monitoring equipment manufacturer to correct for velocity measurements recorded at a single point. A point velocity can be different than the average velocity over the entire depth of the water in the pipe. The equation for the flow equipment lowered all the measured velocities by approximately 10 percent. 6.2.4 Correction for Missing Velocity Data For reasons that are not completely understood, the velocity readings at the inlet and outlet pipes would occasionally drop to zero. This occurred at the inlet meter during five events (events 2, 3, 6, 10, and 14) and at the outlet meter during one event (event 2). The missing velocity data for events 2, 3, 6, 10, and 14 amounted to 35, 15, 7, 10, and 6 percent of the total event data, respectively, based on storm flow volume. 44 ------- The velocity dropout occurrences were corrected in the following manner, as demonstrated with the inlet velocity data from event 2. The meter failed to record approximately eight minutes of the 135 minutes of runoff during one of the flow peaks (see Figure 6-2). Since the gauge heights were available during the missing velocity period, the gauge heights could be used to calculate the missing velocity data. This was done by creating regression relationships between gauge height and velocity. C, - o 2- 0-1 2- 0-1 Timet Figure 6-2. Event 2 example hydrograph showing period of missing velocity data. By filling in the missing velocity data, the increases in volumes at the inlets for the five storms ranged from 6 to 35 percent, with an average increase of 15 percent. The criterion for a qualified event includes successfully recording flow data throughout the duration of the event (see Section 4.4). An important part of deciding whether to qualify or reject an event is determining the amount of missing data from the event. The velocity measurements trigger the data logger to collect samples, so no samples would be collected when the velocity meter recorded zero velocity. It is possible to use the estimated flow data to determine the number of samples that should have been collected when the velocity dropped to zero, as shown in Table 6-8. The VTP included a completeness goal of 85 percent, which was used as the criteria for determining whether sufficient data was collected from a particular event. A number of storms were eliminated from the verification of the StormFilter, because they were missing more than 15 percent of the aliquots. Some storms also had some missing velocity data near the end of the hydrograph. It appears that zero velocity was recorded when the water did not cover the velocity probe. A gauge height was still available for this part of most storms. A gauge height relationship with flow was estimated for these very low flows and the relationship was used to estimate the missing volume. This added a small amount of volume to each storm. 45 ------- Table 6-8. Missing Sample Aliquots Due to Missing Inlet Velocity Data Event Number of No. Missing Aliquots 2 O 4 10 17 4 O 4 1 1 Total Aliquots Collected Missing Aliquots and Missing for Storm (Percent) 33 33 25 14 9 12 9 16 7 11 In spite of the missing aliquots, each composite sample collected was comprised of a minimum of five aliquots, including at least two aliquots on the rising limb of the runoff hydrograph, at least one aliquot near the peak, and at least two aliquots on the falling limb of the runoff hydrograph, and therefore met the qualified event criteria as stated in the protocol 46 ------- Chapter 7 Operations and Maintenance Activities 7.1 System Operation and Maintenance SMI recommends initially scheduling one minor inspection and one major maintenance activity per year at the for a typical installation. A minor maintenance activity and inspection consists of visually inspecting the unit and removing trash and debris. During this activity, the need for major maintenance should be determined. A major maintenance consists of pumping accumulated sediment and water from the vault and replacing the filter cartridges. SMI indicates that the sedimentation rate is the primary factor for determining maintenance frequency, and that a maintenance schedule should be based on site-specific sedimentation conditions. The TO followed the manufacturer's guidelines for maintenance on the StormFilter system during the verification testing. Installation of the StormFilter was completed in December 2001. In the spring of 2002, the system was placed into operation and adjustments to the system were completed, ETV monitoring of the system began in June, 2003. Table 7-1. Operation and Maintenance During Verification Testing Date Activity Personnel Time/Cost June, 19, 2002 (Major maintenance) November 7, 2002 (Minor maintenance) April 24, 2003 (Minor maintenance) StormFilter unit was cleaned of accumulated sediment and filter cartridges were replaced. StormFilter visual inspection by WisDOT. Reported observing the following: 1) 0.20 ft of standing water in the filter vault; 2) no measurable accumulation of sediment in tank bottom; 3) less than 5 percent of water surface area contained floating debris (scum, leaves, cigarette butts; pieces of Styrofoam, etc.) 4) observed a slight oil sheen. USGS assessed need for major maintenance. Concluded major maintenance not required at the time based on following observations: 1) TSS from a 4/4/03 event showed good reductions (Inlet: 736 mg/1; Outlet: 31 mg/1). Note: this was not an ETV qualified event. 2) the tank calibration plot from 4/18/03 showed discharge from device through the filters at a gage height of 1.25; 3) observed filter media; and color was not black, but a light gray. Earth Tech, USGS; WDNR; SMI; total of 3 staff days. WisDOT: 2 staff hours 4 staff hours. 47 ------- Table 7-1 (cont'd). Date Activity Personnel Time/Cost January 27, 2004 Post-monitoring clean out. The procedure is Staff time: 40 hours (Maj or maintenance) summarized in Section 7.1.1. Lab costg (drying & weighing canisters): $1,200.00 7.1.1 Major Maintenance Procedure As noted in Table 7-1, major maintenance, consisting of removing the sediments collected in the StormFilter and replacing the filter cartridges, was conducted after the final storm event. During the major maintenance event, water collected in the StormFilter was pumped into a 400-gallon tank, and the settled sediments were collected, dried and weighed, and the filter cartridges were replaced. The following procedures were undertaken during the major maintenance event. Inlet Bay Cleaning Procedure 1. Removed plastic flow diverter 2. Removed sediment slurry with trash pump into 400-gallon cleaning tank 3. Removed plastic manifold and shoveled heavy sediment into 9 5-gallon buckets (mostly sand sized particles) Canister Bay Cleaning Procedure 1. Removed as much of wet slurry as possible to 400-gallon cleaning tank with trash pump 2. Removed heavy sediment into 5-gallon bucket and dumped into 400-gallon tank 3. Removed canisters with boom truck and capped outlet 4. Removed sediment from under canisters 5. Replaced old canisters with pre-weighed clean canisters (ZPG media) 400-Gallon Cleaning Tank 1. Tank had about 150 gallons of water and sediment (water was left to settle sediment) 2. Used lab pump to decant liquid off the top. Filled about 4 buckets and rest went to sanitary sewer (about 130 gallons) 3. Used an ash shovel connected to a doll to scoop up the organics and sediment into 5- gallon buckets 4. Tap water was used to rinse out remainder of sediment in tank (put into buckets) The wet slurry collected from the StormFilter was transported off-site for drying. The dry weight of the solids collected in the StormFitler was approximately 570 pounds. SMI recommends that the cartridge filter media be characterized and disposed of in accordance with applicable regulations, and that the remaining cartridge components be shipped back to SMFs Portland, Oregon facility for cleaning and reuse. 48 ------- Chapter 8 References 1. APHA, AWWA, and WEF. Standard Methods for the Examination of Water and Wastewater, 19th ed. Washington, DC, 1995. 2. Horowitz; AJ; Hayes, T.S.; Gray; J.R.; Capel, P.D. Selected Laboratory Evaluations of the Whole-Water Sample-Splitting Capabilities of A Prototype Fourteen-Liter Teflon* Churn Splitter, U.S. Geological Survey Open-File Report 01-386, 2001. 3. Huff, F. A., Angel, J. R. Rainfall Frequency Atlas of the Midwest, Midwestern Climate Center, National Oceanic and Atmospheric Administration, and Illinois State Water Survey, Illinois Department of Energy and Natural Resources. Bulletin 71, 1992. 4. Fishman, M. J., Raese, J. W., Gerlitz, C. N., Husband, R. A., U.S. Geological Survey. Approved Inorganic and Organic Methods for the Analysis of Water and Fluvial Sediment, 1954-94, USGS OFR 94-351, 1994. 5. NSF International and Earth Tech, Inc. Test Plan for the Verification of Stormwater Management, Inc. StormFilter* Treatment System Using ZPG Filter Media, "Riverwalk Site"Milwaukee, Wisconsin. March 22, 2004. 6. NSF International. ETV Verification Protocol Stormwater Source Area Treatment Technologies. U.S. EPA Environmental Technology Verification Program; EPA/NSF Wet- weather Flow Technologies Pilot. March 2002 (v. 4.1). 7. Radtke, D.B. et al., National Field Manual for the Collection of Water-Quality Data, Raw Samples 5.1. U.S. Geological Survey Techniques of Water-Resources Investigations Book 9, Chapter A5, pp 24-26, 1999. 8. United States Environmental Protection Agency. Methods and Guidance for Analysis of Water, EPA 821-C-99-008, Office of Water, Washington, DC, 1999. 49 ------- Glossary Accuracy - a measure of the closeness of an individual measurement or the average of a number of measurements to the true value and includes random error and systematic error. Bias - the systematic or persistent distortion of a measurement process that causes errors in one direction. Comparability - a qualitative term that expresses confidence that two data sets can contribute to a common analysis and interpolation. Completeness - a quantitative term that expresses confidence that all necessary data have been included. Precision - a measure of the agreement between replicate measurements of the same property made under similar conditions. Protocol - a written document that clearly states the objectives, goals, scope and procedures for the study. A protocol shall be used for reference during Vendor participation in the verification testing program. Quality Assurance Project Plan - a written document that describes the implementation of quality assurance and quality control activities during the life cycle of the project. Residuals - the waste streams, excluding final effluent, which are retained by or discharged from the technology. Representativeness - a measure of the degree to which data accurately and precisely represent a characteristic of a population parameter at a sampling point, a process condition, or environmental condition. Wet-Weather Flows Stakeholder Advisory Group - a group of individuals consisting of any or all of the following: buyers and users of in drain removal and other technologies, developers and Vendors, consulting engineers, the finance and export communities, and permit writers and regulators. Standard Operating Procedure - a written document containing specific procedures and protocols to ensure that quality assurance requirements are maintained. Technology Panel - a group of individuals with expertise and knowledge of stormwater treatment technologies. Testing Organization - an independent organization qualified by the Verification Organization to conduct studies and testing of mercury amalgam removal technologies in accordance with protocols and Test Plans. Vendor - a business that assembles or sells treatment equipment. 50 ------- Verification - to establish evidence on the performance of in drain treatment technologies under specific conditions, following a predetermined study protocol(s) and Test Plan(s). Verification Organization - an organization qualified by EPA to verify environmental technologies and to issue Verification Statements and Verification Reports. Verification Report - a written document containing all raw and analyzed data, all QA/QC data sheets, descriptions of all collected data, a detailed description of all procedures and methods used in the verification testing, and all QA/QC results. The Test Plan(s) shall be included as part of this document. Verification Statement - a document that summarizes the Verification Report reviewed and approved and signed by EPA and NSF. Verification Test Plan - A written document prepared to describe the procedures for conducting a test or study according to the verification protocol requirements for the application of in drain treatment technology. At a minimum, the Test Plan shall include detailed instructions for sample and data collection, sample handling and preservation, precision, accuracy, goals, and quality assurance and quality control requirements relevant to the technology and application. 51 ------- Appendices A Verification Test Plan B Event Hydrographs and Rain Distribution C Analytical Data Reports 52 ------- |