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

-------