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
------- |