The Stormwater Management Stormfilter® for the Removal of SIL-CO-SIL 106, a Standardized Silica Product: ZPGTM Stormfilter Cartridge at 28 L/Min (7.5 Gpm)

Level 3 Stormwater Treatment Engineering Report Northwest Container Services–Tacoma Facility 1801 Taylor Way Tacoma, Washington 98409 Pierce County

Site Operator: Northwest Container Services, Inc Permit Number: WAR126969 Permit Type: Industrial Stormwater General Permit

Prepared by: PBS Engineering and Environmental Inc. 415 W 6th Street, Suite 601 Vancouver, Washington 98660

May 15, 2018 PBS Project 17635.007

415 W 6 TH STREET, SUITE 601 VANCOUVER, WA 98660 3 60. 695. 3 4 8 8 M A I N 866.727.0140 FAX PBS USA .COM Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

TABLE OF CONTENTS

Industrial Stormwater General Permit S8.D.3.a References ...... iii General Information ...... iv 1 INTRODUCTION ...... 1 2 FACILITY ASSESSMENT ...... 2 2.1 Facility Description ...... 2 2.2 Surface Water Drainage ...... 2 3 TREATMENT ALTERNATIVE EVALUATION...... 3 3.1 Water Quality Characterization ...... 3 3.2 Hydrologic Analysis ...... 3 3.3 Treatment Alternatives ...... 4 3.4 Treatment Alternatives Expected Performance ...... 4 3.5 Preliminary Cost Estimates ...... 5 4 PROPOSED STORMWATER TREATMENT IMPROVEMENTS ...... 7 4.1 Selected Treatment BMP and Sizing Calculations ...... 7 4.2 Treatment Process and Operation ...... 7 4.3 Use of Chemicals in the Treatment Process ...... 7 4.4 Expected Treatment Performance...... 7 5 CERTIFICATION BY A LICENSED PROFESSIONAL ENGINEER ...... 9

May 15, 2018 i PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

SUPPORTING DATA TABLES Table 1: Water Quality Characterization of NW Container Tacoma Stormwater (Q2 2013–Q2 2018) Table 2. Selected Design Flow Rates for Basin 1 Table 3. Summary of Projected Pollutant Reduction for Alternative 1 Table 4. Summary of Projected Pollutant Reduction for Alternative 2 Table 5: Treatment Technology Estimated Preliminary Cost Estimates Table 6. Summary of Projected Pollutant Reduction for the Proposed Basin 1 Treatment System

FIGURES Figure 1. Vicinity Map Figure 2. Site Map Figure 3. Plan View: Proposed Basin 1 Treatment System Figure 4. Flow Diagram: Proposed Basin 1 Treatment System

APPENDICES Appendix A: Conceptual Designs for Basin 1 Alternatives Appendix B: Western Washington Hydrology Model Results Appendix C: Contech Provided Studies

May 15, 2018 ii PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

Industrial Stormwater General Permit S8.D.3.a References

S8.D.3.a The Engineering Report must include: Report Section

i. Brief summary of the treatment alternatives considered and why the proposed option was Sections 3.3, 3.4, and selected. Include cost estimates of ongoing operation and maintenance, including disposal of 3.5 any spent media;

ii. The basic design data, including characterization of stormwater influent, and sizing Sections 3.1, 3.2, and calculations of the treatment units; 4.1

iii. A description of the treatment process and operation, including a flow diagram; Section 4.2

iv. The amount and kind of chemicals used in the treatment process, if any. Note: use of Section 4.3 stormwater treatment chemicals requires submittal of Request for Chemical Treatment Form;

v. Results to be expected from the treatment process including the predicted stormwater Sections 3.4 and 4.4 discharge characteristics;

vi. A statement, expressing sound engineering justification through the use of pilot plant data, results from similar installations, and/or scientific evidence that the proposed treatment is Section 4.4 and 5 reasonably expected to meet the permit benchmarks; and

vii. Certification by a licensed professional engineer. Section 5

May 15, 2018 iii PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

General Information

Name of Facility: Northwest Container Services – Tacoma Facility Location of Facility: 1801 Taylor Way Tacoma, Washington 98409

Type of Facility: Intermodal Container Transfer Station Primary SIC Code 4731 (Arrangement of Transportation of Freight and Cargo) Primary NAICS 488510 (Freight Transportation Arrangement)

Type of Permit: Industrial Stormwater – General Permit Permit Number: WAR126969 County: Pierce County

Site Area: 305,216 square feet (7.01 acres) Impervious Area: 284,149 square feet (6.52 acres)

Chief Official: Gary Cardwell Title: Division Vice President

Site Contact Name: Bob Sherwood Site Contact Title: District Manager Telephone Number: 253.272.3134 Fax Number: 253.838.3745

May 15, 2018 iv PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

1 INTRODUCTION PBS Engineering and Environmental Inc. (PBS) has prepared this engineering report to present the selection and design of stormwater best management practices (BMPs) at the Northwest Container Services Tacoma facility (Facility) in Tacoma, Washington. The Facility is operated by Northwest Container Services, Inc. (NWCS). The Facility operates under coverage of an Industrial Stormwater General Permit (ISGP) issued by the Washington Department of Ecology (DOE) on May 16, 2013, and reissued on December 3, 2014.

In 2017, the Facility triggered a Level 3 corrective action for total zinc per ISGP Section 8.D at sample location OF1. It also triggered a Level 2 corrective action for turbidity per ISGP Section S8.C at OF1; however, NW Container elected to install a Level 3 corrective action in lieu of a Level 2 corrective action at OF1 as well. Therefore, this engineering report was written to address stormwater treatment measures that will address both total zinc and turbidity for Drainage Basin 1. This engineering report is consistent with the requirements set forth in S8.D.3.a of the Level 3 corrective action condition of the Permit.

May 15, 2018 1 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

2 FACILITY ASSESSMENT Included in this section is a summary of the Facility operations as well as a description of the existing surface water drainage.

2.1 Facility Description The Facility is located at 1801 Taylor Way (47o 16’ 25” N latitude; 122o 23’ 10” W longitude) in Tacoma, Washington. The Facility is located within the Port of Tacoma. Land use surrounding the Facility is predominately industrial, with industrial properties bordering the site to the east and west, Taylor Way to the south, and the Hylebos Waterway to the north. Figure 1 shows the general location of the Facility in relation to surrounding properties, transportation routes, surface waters, and other relevant features. The Facility is approximately 7.0 acres in size, and the majority of the site is impervious. Impervious surfaces consist of pavement and building rooftops. The only pervious surface is a small stretch of gravel ballast rock around the railroad tracks along the western property boundary, which occupies approximately seven percent of the facility area.

The Northwest Container Services Tacoma facility is an intermodal container facility. The Facility is open Monday through Friday from 8:00 am to 5:00 pm and is operated by NW Container. Containers are moved between trucks and trains and vice versa. A small number of containers are routinely stored at the site, including a stock of refrigerated containers (i.e., “reefers”).

The Facility operates forklifts and container reach stackers (i.e., “stackers”) for moving containers between trucks and trains. The Facility consists of a main yard alongside a double set of railroad tracks. The main yard is used to move and store containers, perform maintenance, store maintenance supplies, park vehicles, and to house office buildings. A small entry gatehouse is located near the southeast corner of the site.

2.2 Surface Water Drainage The NW Container Tacoma yard consists of one drainage basin: Basin 1. Stormwater runoff from the site enters the stormwater system through a series of 19 catch basins located throughout the yard. Stormwater is conveyed by gravity through the site drainage system to a single consolidated point (manhole) in the northeast quadrant of the site. From here, stormwater flows into a BaySaver BayFilter underground stormwater treatment structure (BayFilter vault). The treatment structure consists of 14 stormwater Enhanced Media Cartridges (EMC). The current treatment BMPs used at the facility are catch basin inserts with secondary filter inserts with zeolite pouches, the BayFilter vault, and a zeolite pouch in the outlet chamber of the BayFilter vault.

The manhole directly downstream (north) of the BayFilter vault serves as the sample point (OF1) for Basin 1. The outlet of the BayFilter vault gravity flows to the site’s stormwater outfall to the Hylebos Waterway, which ultimately discharges into Commencement Bay of the Puget Sound. Figure 2 details the general configuration of the stormwater system, catch basin IDs, drainage basin boundaries and the location of the existing treatment system.

May 15, 2018 2 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

3 TREATMENT ALTERNATIVE EVALUATION This section presents the treatment alternatives evaluation for the Level 3 corrective action. Included in this section is a summary of the treatment alternatives considered and their corresponding preliminary cost estimates as required in S8.D.3.a.i of the Permit. This section also presents the hydrologic analysis and water quality characterization used to size and select the treatment alternatives as required by S8.D.3.a.ii of the Permit.

3.1 Water Quality Characterization Stormwater samples have been collected from the Facility’s sample point and analyzed for permit-required analytes from the second quarter of 2013 to present. Table 1 presents average concentrations of the parameters analyzed as required by the Permit.

Table 1: Water Quality Characterization of NW Container Tacoma Stormwater (Q2 2013–Q2 2018) Basin Total Zinc (µg/L) Turbidity (NTU)

1 224 16 Note: Bold results indicate the pollutant concentration exceeded the statewide benchmark

3.2 Hydrologic Analysis The hydrologic analysis performed was in accordance with the criteria and guidelines set forth by the City of Tacoma’s Stormwater Management Manual and the Stormwater Management Manual for Western Washington (SWMMWW). Section 3.1.2 of Volume 5 of the City of Tacoma’s Stormwater Management Manual requires that treatment facilities not preceded by an equalization or storage basin shall be sized to receive and treat the water quality design “flow rate at or below 91 percent of the runoff volume, as estimated by [Western Washington Hydrology Model]” (WWHM). Section 4.1.2 of Volume 5 of the SWMMWW requires that stormwater treatment facilities are sized to treat at least 91 percent of the runoff volume as estimated by an approved continuous runoff model. The WWHM is a continuous simulation hydrologic model developed and approved by Ecology and was used to size the stormwater treatment systems in both alternatives for the Facility. The WWHM used 60 years of precipitation data from the National Weather Service’s Tacoma – S 36th St rain gauge station (A2143) in Pierce County, Washington. Precipitation data from this rain gauge were imported into the model to represent the local historical rainfall.

The WWHM evaluates both pre- and post-development scenarios where changes in the contributing pervious and impervious areas from the predevelopment scenario are compared to the post development scenario. None of the treatment alternatives considered were expected to change the existing pervious and impervious conditions of the facility; therefore, the impervious and pervious contributing areas in pre- and post- development scenarios were the same.

The water quality analysis tool of the WWHM was used to estimate the design flow rate or water quality flow rate that corresponds to treating 91 percent of the runoff volume. The water quality analysis tool used the simulated precipitation data and a surface area characterization to estimate the water quality flow rate. Water quality flow rates were estimated for offline and online facilities. An offline facility is sized to receive and treat the water quality design flow rate to the applicable performance goal, and the higher incremental portion of flow rates are bypassed around the treatment facility. Online facilities are sized to convey flow rates in excess of the design flow rate provided that a net pollutant reduction is maintained. The WWHM-estimated online and offline water quality flow rates for the two flow scenarios for Basin 1 are presented in Table 2.

May 15, 2018 3 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

Table 2. Selected Design Flow Rates for Basin 1 Offline Water Quality Online Water Quality Selected Design Flow Rate (gpm) Flow Rate (gpm) Flow Rate Offline water 248.7 432.3 quality flow rate Note: gpm = gallons per minute

The offline water quality flow rate was used to size both alternatives because the bypassed flow will include peaks of large storm events that are not representative of typical stormwater discharges from the site. The WWHM output results are included in Appendix B.

3.3 Treatment Alternatives A total of two alternatives were considered for the Facility. Alternative 1 evaluated a two-stage Contech StormFilter cartridge vault treatment system to treat the stormwater flow, whereas Alternative 2 evaluated an enhanced media-bed filtration system to treat the stormwater flow. Appendix A provides conceptual designs for the proposed treatment alternatives. The exact location of vaults and tanks in either alternative have yet to be determined.

3.4 Treatment Alternatives Expected Performance The treatment technologies selected for the Facility have either been assessed for performance through the Technology Assessment Protocol – Ecology (TAPE) program or have been implemented at similar industrial facilities and are successfully meeting ISGP benchmarks. The TAPE program provides a peer-reviewed certification process for emerging stormwater treatment technologies. As part of the TAPE certification process, laboratory and field tests are performed on these technologies and the findings from these experiments are made available to the public. Findings evaluated to determine a technology’s ability to meet treatment performance goals are outlined in Volume 5 of the SWMMWW.

Ecology will certify a treatment technology for Pilot Use Level Designation (PULD) if it successfully meets one or more performance goals during laboratory tests and Conditional Use Level Designation (CULD) if it meets one or more performance goals during both laboratory and field tests. Once a technology receives a PULD and CULD, Ecology allows the technology to be installed and operated in the state of Washington where the technology can receive a final General Use Level Designation (GULD) certification based on performance data of the full-scale system in operation. A summary of the designations Ecology has provided for the technologies considered are listed below: • GULD for Contech StormFilter as basic treatment for TSS. • CULD for proprietary media filtration system as basic treatment for TSS and enhanced treatment for dissolved zinc.

TSS removal data was available for the Contech StormFilter cartridge treatment system from the TAPE program, as well as TSS and dissolved zinc removal data for the proprietary media filtration system. No total zinc removal data was available for either of the above treatment technologies from the TAPE program. In order to calculate expected treatment system capabilities, manufacturer-provided historic treatment system performance data from similar facilities was used to determine average expected treatment system pollutant removal for the StormFilter system.

In some instances, either only TSS or only dissolved zinc removal data was available. For the technologies evaluated in this study, treatment performance for TSS versus turbidity is generally expected to be similar.

May 15, 2018 4 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

However, the two parameters are not equivalent and in PBS’ experience treatment performance for turbidity is more difficult to accurately quantify due to the variety of water quality characteristics that can influence this measurement. Therefore, where only TSS removal data was available, PBS used a correlation factor of 1.5 to estimate turbidity removal. For example, for a predicted TSS removal of 50 percent, an estimated turbidity removal of 33 percent was used.

Treatment performance for total zinc versus dissolved zinc is also expected to be similar, although a portion of the total zinc can be expected to be removed as TSS and turbidity is removed. Dissolved or soluble metals speciation analyses was performed, and most results indicate that the dissolved fraction is 70 to 100% of the total concentration. However, to be conservative in this estimate, PBS used a correlation factor of 1.5 to estimate total zinc removal when only dissolved zinc removal data was available.

Table 3 and Table 4 summarize the projected pollutant concentration and net total pollutant removal for Alternative 1 and Alternative 2, respectively.

Table 3. Summary of Projected Pollutant Reduction for Alternative 1

Parameter Average Influent Projected Pollutant Projected Pollutant (unit of measure) Concentration Removal (%) Concentration

Total Zinc 224 82.4% 39.5 (µg/L) Turbidity 16 76.0% 3.8 (NTU) Note: Bold results indicate the pollutant concentration exceeded the statewide benchmark

Table 4. Summary of Projected Pollutant Reduction for Alternative 2

Parameter Average Influent Projected Pollutant Projected Pollutant (unit of measure) Concentration Removal (%) Concentration

Total Zinc 224 51.0% 109.8 (µg/L) Turbidity 16 65.3% 5.6 (NTU) Note: Bold results indicate the pollutant concentration exceeded the statewide benchmark

3.5 Preliminary Cost Estimates Table 5 provides the estimated preliminary capital and annual O&M costs for each alternative. Capital costs include both construction and non-construction related costs. Engineering, contractor selection support, permitting, construction-period engineering support, and contingency funds are included in the non- construction related costs. O&M costs include labor that may be needed to operate the treatment system, purchase of any required consumables (including replacement media), and disposal of solids generated by or accumulated within the treatment system.

May 15, 2018 5 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

Table 5: Treatment Technology Preliminary Cost Estimates

Non- Construction Total Capital Annual Alternative Construction Costs Costs2 O&M Costs3 Costs1

1. Treatment: Two-stage Contech $315,000 $178,000 $495,000 $27,000 StormFilter

2. Treatment: Enhanced Media $353,000 $196,000 $550,000 $27,900 Filtration System

Notes: 1. Non-construction costs include construction management, general and administrative expenses, contractor profit, overhead, mechanical and electrical work, engineering, and permitting. 2. Capital costs were rounded up to the nearest $5,000. 3. Annual operation and maintenance costs were rounded to the nearest $100.

Alternative 2 is the most expensive option when considering both capital and annual O&M costs, and the treatment system would be aboveground. The equipment cost for the enhanced media filtration system is the primary factor in this alternative’s high cost.

NW Container and PBS identified Alternative 1 as the preferred alternative. While both alternatives have similar O&M costs, the projected O&M costs for Alternative 1 are lower. Other advantages include that it is simple to operate, does not require the use of chemicals, and will be installed underground to preserve operational space for the Facility.

May 15, 2018 6 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

4 PROPOSED STORMWATER TREATMENT IMPROVEMENTS This section presents the design of the preferred treatment alternative selected for the Level 3 corrective action.

4.1 Selected Treatment BMP and Sizing Calculations Alternative 1 was selected as the treatment BMP for the Level 3 corrective action in Basin 1. Stormwater will be treated in a subgrade, two-stage Contech StormFilter filtration system. Using the offline flow rate of 248.7 gpm, two 8-foot by 16-foot Contech StormFilter vaults will be installed in series. The type of media, size, and number of treatment cartridges has yet to be determined at this time.

4.2 Treatment Process and Operation Currently, stormwater collected throughout Basin 1 flows to an existing manhole that discharges to the existing BayFilter vault. In the proposed treatment system, the outlet of the existing BayFilter vault will be converted into a lift station and routed into the Contech treatment system. The existing BayFilter cartridges and manifold piping will be removed, but the vault will remain to house the pumps as well as provide settling and storage volume. However, this may be insufficient and subsequent design may include other solids removal pretreatment technology, which will be determined in detailed engineering design. A new manhole will be installed downstream of the treatment system that will act as the new sample point (OF1) for Basin 1.

The invert elevation of the inlet piping into the existing BayFilter vault is equal to the invert elevation of the outlet piping, providing no hydraulic drop through the treatment system. The Contech StormFilter vault will require hydraulic drop through the vault to ensure solids are adequately settled in the pretreatment chamber and metals and solids are removed in the treatment system. The required amount of hydraulic drop through the system depends on the design cartridge height. Detailed engineering design is still needed to determine the required cartridge size and necessary hydraulic drop to implement the treatment system; although it is expected that stormwater will need to be pumped to supply the necessary elevation drop.

Figure 3 provides the plan view of the proposed stormwater improvements for Basin 1. Figure 4 provides the treatment process flow diagram for the proposed stormwater improvements in Basin 1.

4.3 Use of Chemicals in the Treatment Process The ISGP requires that the amount and kind of any chemicals used in the proposed treatment process are described in the engineering report for the Level 3 corrective action. The proposed treatment BMP for the Facility does not use any treatment chemicals.

4.4 Expected Treatment Performance Contech provided results from a pilot study performed in a laboratory as well as field tests performed on a full-scale system installed in Milwaukee, Wisconsin. In the pilot study, seven different simulations were performed in a controlled, laboratory environment to determine TSS and turbidity removal capabilities. Results indicated that the StormFilter cartridge using ZPG media provided a mean TSS reduction of 87 percent and a 51 percent mean decrease in turbidity.1 The field tests were performed at the “Riverwalk” site in Milwaukee, Wisconsin, to determine the sediment, metals, and nutrients removal capabilities of StormFilter cartridges using ZPG media.2 NSF International (NSF) teamed with the U.S. Environmental Protection Agency

1 Evaluation of the Stormwater Management StormFilter® for the removal of SIL-CO-SIL 106, a standardized silica product: ZPGTM StormFilter cartridge at 28 L/min (7.5 gpm). Contech Stormwater Solutions Product Evaluation. Publication #PE- E062. April 11, 2006. 2 Environmental Technology Verification Report, Stormwater Source Area Treatment Device, The Stormwater Management StormFilter® using ZPG Filter Media. NSF International, under a cooperative agreement with U.S. Environmental Protection Agency. Publication #04/17/WQPC-WWF. July 2004.

May 15, 2018 7 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

(EPA) to evaluate the performance of the StormFilter using ZPG filter media. A total of 20 precipitation events were sampled over the course of the study. Of those, 17 events produced zinc removal data. The average zinc influent concentration was 406 micrograms per liter (µg/L), and the average effluent concentration was 135 µg/L. The corresponding average percent removal was 58 percent. The influent total zinc concentrations from this facility are similar to the range of concentrations historically observed at NW Container. Therefore, the range of removal is expected to be similar as well. Appendix C provides the results from these studies.

Table 6 presents the projected pollutant reductions for the Basin 1 proposed treatment system. Using the total zinc and turbidity removals presented in the TAPE results as well as the Contech-provided studies, the proposed treatment systems are projected to reduce pollutant concentrations to below the benchmarks.

Table 6. Summary of Projected Pollutant Reduction for the Proposed Basin 1 Treatment System

Parameter Average Influent Net Pollutant Projected Pollutant (unit of measure) Concentration Removal (%) Concentration Total Zinc 224 82.4% 39.5 (µg/L) Turbidity 16 76.0% 3.8 (NTU) Note: Bold results indicate the pollutant concentration exceeded the statewide benchmark

May 15, 2018 8 PBS Project 17635.007

Level 3 Stormwater Treatment Engineering Report Northwest Container Services – Tacoma Facility Northwest Container Services, Inc. Tacoma, Washington

5 CERTIFICATION BY A LICENSED PROFESSIONAL ENGINEER The undersigned registered professional engineer (PE) is familiar with the current engineering report requirements set forth in the Industrial Stormwater General Permit issued by the State of Washington Department of Ecology to satisfy the requirements of a Level 3 corrective action. The PE attests that the necessary treatment BMPs that are suited to remove turbidity and total zinc from stormwater runoff with the goal of attaining the benchmark values specified under Section 5 of the Permit, were selected in consultation with the PE.

Name: Sean Hanrahan, PE

Registration Number: 50726

State: Washington

Title, Company: Environmental Engineer PBS Engineering and Environmental Inc.

Date: 05/15/2018

May 15, 2018 9 PBS Project 17635.007

FIGURES Figure 1. Vicinity Map Figure 2. Site Map Figure 3. Plan View : Proposed Drainage Basin 1 Treatment System Figure 4. Flow Diagram : Proposed Drainage Basin 1 Treatment System

5/10/2018 3:56:53 PM

PROJECT LOCATION CAD Plot Date/Time: Taylor Ford User: VICINITY MAP Layout Tab:

SOURCE: USGS TACOMA NORTH WA QUADRANGLE 1994, PHOTO REVISED 1993.

SEATTLE SITE OLYMPIA SCALE: 1" = 2,000'

VANCOUVER 0' 1,000' 2,000' 4,000' WASHINGTON PREPARED FOR: NORTHWEST CONTAINER SERVICES, INC

L:\Projects\17000\17600-17699\17635_NWCS-Tacoma\DWG\SPCC_SWPPP\17635.004_FIG 1.dwg MAY 2018 VICINITY MAP 17635.005 1801 TAYOR WAY FIGURE TACOMA, WASHINGTON 1 Filename: Vancouver, WA 98660 314 W 15th Street 360.695.3488 pbsusa.com PBS Engineering and Environmental Inc. 5/10/2018 3:48:26 PM * A-2 A-3 A-4 D-1

CAD Plot Date/Time: A-1

Taylor Ford TAYLOR WAY * G-1

User: G-1

T-1 T

SITE MAP * * * Layout Tab:

HYLEBOS WATERWAY

LEGEND TANK LEGEND

A-1

A-2 EXISTING SITE MAP NORTHWEST CONTAINER SERVICES, INC 1801 TAYLOR WAY, TACOMA, WASHINGTON A-3 PROJECT A-4 17635.004 * D-1 DATE T G-1 APPROXIMATE SCALE: 1" = 80' MAY 2018 T-1

L:\Projects\17000\17600-17699\17635_NWCS-Tacoma\Stormwater Treatment\Reports\Engineering Report\Figure 2.dwg FIGURE 0 40' 80' 160' 2 Full Size Sheet Format Is 11x17; If Printed Size Is Not 11x17, Then This Sheet Format Has Been Modified & Indicated Drawing Scale Is Not Accurate. PREPARED FOR: NORTHWEST CONTAINER SERVICES, INC Filename: 5/15/2018 10:55:04 AM Vancouver, WA 98660 314 W 15th Street 360.695.3488 pbsusa.com PBS Engineering and Environmental Inc. CAD Plot Date/Time: Taylor Ford User:

PLAN VIEW * A-2 A-3 A-4 D-1

A-1 Layout Tab: TAYLOR WAY * G-1

G-1

T-1 T * * *

HYLEBOS WATERWAY

LEGEND TANK LEGEND

A-1

A-2 PROPOSED BASIN 1 TREATMENT SYSTEM PLAN VIEW NORTHWEST CONTAINER SERVICES, INC 1801 TAYLOR WAY, TACOMA, WASHINGTON A-3 PROJECT A-4 17635.005 * D-1 DATE T G-1 APPROXIMATE SCALE: 1" = 80' MAY 2018 T-1

L:\Projects\17000\17600-17699\17635_NWCS-Tacoma\Stormwater Treatment\Reports\Engineering Report\AppA_Conceptual Designs\Conceptual Designs.dwg FIGURE 0 40' 80' 160' 3 Full Size Sheet Format Is 11x17; If Printed Size Is Not 11x17, Then This Sheet Format Has Been Modified & Indicated Drawing Scale Is Not Accurate. PREPARED FOR: NORTHWEST CONTAINER SERVICES, INC Filename: Filename: L:\Projects\17000\17600-17699\17635_NWCS-Tacoma\Stormwater Treatment\Reports\Engineering Report\AppA_Conceptual Designs\Conceptual Designs.dwg Layout Tab: FLOW DIAGRAM User: Taylor Ford CAD Plot Date/Time: 5/15/2018 11:06:46 AM Full SizeSheet FormatIs11x17; IfPrintedSizeIsNot 11x17, ThenThisSheet FormatHasBeenModified &IndicatedDrawing ScaleIsNotAccurate. SITE STORMWATERRUNOFF STORMFILTER STORMFILTER EXISTING CONTECH CONTECH OUTFALL STATION SAMPLE BASINS CATCH POINT PUMP (OF1) PREPARED FOR:NORTHWEST CONTAINER SERVICES, INC PROJECT

17635.005 PBS Engineering and MAY 2018 PROPOSED BASIN 1 TREATMENT SYSTEM FLOW DIAGRAM FIGURE Environmental Inc. DATE

4 314 W 15th Street Vancouver, WA 98660 NORTHWEST CONTAINER SERVICES, INC 360.695.3488 1801 TAYLOR WAY, TACOMA, WASHINGTON pbsusa.com

APPENDIX A Conceptual Designs for Basin 1 Alternatives Alternative 1 Alternative 2

5/15/2018 10:58:02 AM Vancouver, WA 98660 314 W 15th Street 360.695.3488 pbsusa.com PBS Engineering and Environmental Inc. CAD Plot Date/Time: Taylor Ford User:

ALTERNATIVE 1 * A-2 A-3 A-4 D-1

A-1 Layout Tab: TAYLOR WAY * G-1

G-1

T-1 T * * *

HYLEBOS WATERWAY

LEGEND TANK LEGEND

A-1

A-2 ALTERNATIVE 1 CONCEPTUAL DESIGN NORTHWEST CONTAINER SERVICES, INC 1801 TAYLOR WAY, TACOMA, WASHINGTON A-3 PROJECT A-4 17635.005 * D-1 DATE T G-1 APPROXIMATE SCALE: 1" = 80' OCTOBER 2017 T-1

L:\Projects\17000\17600-17699\17635_NWCS-Tacoma\Stormwater Treatment\Reports\Engineering Report\AppA_Conceptual Designs\Conceptual Designs.dwg ALTERNATIVE 0 40' 80' 160' 1 Full Size Sheet Format Is 11x17; If Printed Size Is Not 11x17, Then This Sheet Format Has Been Modified & Indicated Drawing Scale Is Not Accurate. PREPARED FOR: NORTHWEST CONTAINER SERVICES, INC Filename: 5/10/2018 2:53:59 PM Vancouver, WA 98660 314 W 15th Street 360.695.3488 pbsusa.com PBS Engineering and Environmental Inc. CAD Plot Date/Time: Taylor Ford User:

ALTERNATIVE 2 * A-2 A-3 A-4 D-1

A-1 Layout Tab: TAYLOR WAY * G-1

G-1

T-1 T * * *

HYLEBOS WATERWAY

LEGEND TANK LEGEND

A-1

A-2 ALTERNATIVE 2 CONCEPTUAL DESIGN NORTHWEST CONTAINER SERVICES, INC 1801 TAYLOR WAY, TACOMA, WASHINGTON A-3 PROJECT A-4 17635.005 * D-1 DATE T G-1 APPROXIMATE SCALE: 1" = 80' OCTOBER 2017 T-1

L:\Projects\17000\17600-17699\17635_NWCS-Tacoma\Stormwater Treatment\Reports\Engineering Report\AppA_Conceptual Designs\Conceptual Designs.dwg ALTERNATIVE 0 40' 80' 160' 2 Full Size Sheet Format Is 11x17; If Printed Size Is Not 11x17, Then This Sheet Format Has Been Modified & Indicated Drawing Scale Is Not Accurate. PREPARED FOR: NORTHWEST CONTAINER SERVICES, INC Filename:

APPENDIX B Western Washington Hydrology Model Results

WWHM2012 PROJECT REPORT ______

Project Name: 17635.004_Flow Calcs Site Name: Site Address: City : Report Date: 5/15/2018 Gage : Data Start : 10/01/1901 Data End : 09/30/2059 Precip Scale: 1.00 Version Date: 2018/03/08 Version : 4.2.14 ______

Low Flow Threshold for POC 1 : 50 Percent of the 2 Year ______

High Flow Threshold for POC 1: 50 year ______

PREDEVELOPED LAND USE

Name : Basin 1 Bypass: No

GroundWater: No

Pervious Land Use acre C, Lawn, Mod .49

Pervious Total 0.49

Impervious Land Use acre DRIVEWAYS FLAT 6.52

Impervious Total 6.52

Basin Total 7.01

______

Element Flows To: Surface Interflow Groundwater

______

MITIGATED LAND USE

Name : Basin 1 Bypass: No

GroundWater: No Pervious Land Use acre C, Lawn, Mod .49

Pervious Total 0.49

Impervious Land Use acre DRIVEWAYS FLAT 6.52

Impervious Total 6.52

Basin Total 7.01

______

Element Flows To: Surface Interflow Groundwater

______

ANALYSIS RESULTS

Stream Protection Duration

______

Predeveloped Landuse Totals for POC #1 Total Pervious Area:0.49 Total Impervious Area:6.52 ______

Mitigated Landuse Totals for POC #1 Total Pervious Area:0.49 Total Impervious Area:6.52 ______

Flow Frequency Return Periods for Predeveloped. POC #1 Return Period Flow(cfs) 2 year 2.307285 5 year 3.105531 10 year 3.686916 25 year 4.48396 50 year 5.124592 100 year 5.806715

Flow Frequency Return Periods for Mitigated. POC #1 Return Period Flow(cfs) 2 year 0 5 year 0 10 year 0 25 year 0 50 year 0 100 year 0 ______

Stream Protection Duration Annual Peaks for Predeveloped and Mitigated. POC #1 Year Predeveloped Mitigated 1902 2.703 2.703 1903 2.997 2.997 1904 3.488 3.488 1905 1.529 1.529 1906 1.702 1.702 1907 2.321 2.321 1908 1.889 1.889 1909 2.307 2.307 1910 2.221 2.221 1911 2.509 2.509 1912 4.274 4.274 1913 1.788 1.788 1914 7.641 7.641 1915 1.550 1.550 1916 2.879 2.879 1917 1.087 1.087 1918 2.304 2.304 1919 1.424 1.424 1920 1.911 1.911 1921 1.633 1.633 1922 2.588 2.588 1923 1.784 1.784 1924 3.331 3.331 1925 1.400 1.400 1926 2.710 2.710 1927 2.210 2.210 1928 1.654 1.654 1929 3.313 3.313 1930 3.428 3.428 1931 1.664 1.664 1932 1.799 1.799 1933 1.776 1.776 1934 2.948 2.948 1935 1.520 1.520 1936 2.148 2.148 1937 3.163 3.163 1938 1.552 1.552 1939 1.949 1.949 1940 3.435 3.435 1941 3.391 3.391 1942 2.602 2.602 1943 2.549 2.549 1944 3.695 3.695 1945 2.764 2.764 1946 2.175 2.175 1947 1.671 1.671 1948 2.308 2.308 1949 3.541 3.541 1950 2.001 2.001 1951 3.028 3.028 1952 3.523 3.523 1953 3.249 3.249 1954 1.879 1.879 1955 1.730 1.730 1956 1.705 1.705 1957 1.856 1.856 1958 2.340 2.340 1959 2.350 2.350 1960 1.826 1.826 1961 5.278 5.278 1962 2.247 2.247 1963 1.658 1.658 1964 4.925 4.925 1965 2.191 2.191 1966 1.818 1.818 1967 2.585 2.585 1968 2.150 2.150 1969 1.942 1.942 1970 2.230 2.230 1971 2.164 2.164 1972 7.172 7.172 1973 4.066 4.066 1974 2.975 2.975 1975 3.148 3.148 1976 3.321 3.321 1977 1.397 1.397 1978 2.416 2.416 1979 2.493 2.493 1980 2.449 2.449 1981 2.300 2.300 1982 1.876 1.876 1983 2.567 2.567 1984 2.554 2.554 1985 2.933 2.933 1986 1.471 1.471 1987 2.550 2.550 1988 1.533 1.533 1989 1.400 1.400 1990 1.864 1.864 1991 2.786 2.786 1992 2.611 2.611 1993 2.985 2.985 1994 2.070 2.070 1995 1.599 1.599 1996 2.164 2.164 1997 1.924 1.924 1998 2.308 2.308 1999 2.462 2.462 2000 2.186 2.186 2001 1.735 1.735 2002 3.262 3.262 2003 1.857 1.857 2004 2.773 2.773 2005 5.298 5.298 2006 2.476 2.476 2007 2.797 2.797 2008 2.295 2.295 2009 1.739 1.739 2010 2.246 2.246 2011 2.349 2.349 2012 2.199 2.199 2013 2.092 2.092 2014 1.989 1.989 2015 3.448 3.448 2016 2.089 2.089 2017 3.369 3.369 2018 2.071 2.071 2019 3.078 3.078 2020 2.482 2.482 2021 2.079 2.079 2022 3.513 3.513 2023 4.311 4.311 2024 4.768 4.768 2025 2.244 2.244 2026 2.469 2.469 2027 2.750 2.750 2028 1.076 1.076 2029 1.790 1.790 2030 3.543 3.543 2031 1.119 1.119 2032 1.885 1.885 2033 2.367 2.367 2034 1.853 1.853 2035 2.337 2.337 2036 1.851 1.851 2037 2.489 2.489 2038 2.423 2.423 2039 4.750 4.750 2040 1.872 1.872 2041 2.379 2.379 2042 2.721 2.721 2043 3.010 3.010 2044 2.084 2.084 2045 1.690 1.690 2046 1.873 1.873 2047 2.290 2.290 2048 1.888 1.888 2049 2.803 2.803 2050 2.114 2.114 2051 3.009 3.009 2052 2.246 2.246 2053 1.910 1.910 2054 3.917 3.917 2055 2.319 2.319 2056 2.998 2.998 2057 1.471 1.471 2058 2.817 2.817 2059 3.513 3.513 ______

Stream Protection Duration Ranked Annual Peaks for Predeveloped and Mitigated. POC #1 Rank Predeveloped Mitigated 1 7.6405 7.6405 2 7.1721 7.1721 3 5.2984 5.2984 4 5.2777 5.2777 5 4.9251 4.9251 6 4.7679 4.7679 7 4.7500 4.7500 8 4.3112 4.3112 9 4.2740 4.2740 10 4.0664 4.0664 11 3.9165 3.9165 12 3.6947 3.6947 13 3.5430 3.5430 14 3.5408 3.5408 15 3.5235 3.5235 16 3.5131 3.5131 17 3.5127 3.5127 18 3.4881 3.4881 19 3.4485 3.4485 20 3.4347 3.4347 21 3.4278 3.4278 22 3.3913 3.3913 23 3.3694 3.3694 24 3.3310 3.3310 25 3.3215 3.3215 26 3.3134 3.3134 27 3.2620 3.2620 28 3.2488 3.2488 29 3.1630 3.1630 30 3.1478 3.1478 31 3.0778 3.0778 32 3.0277 3.0277 33 3.0097 3.0097 34 3.0094 3.0094 35 2.9980 2.9980 36 2.9974 2.9974 37 2.9850 2.9850 38 2.9746 2.9746 39 2.9481 2.9481 40 2.9333 2.9333 41 2.8792 2.8792 42 2.8168 2.8168 43 2.8032 2.8032 44 2.7972 2.7972 45 2.7863 2.7863 46 2.7734 2.7734 47 2.7644 2.7644 48 2.7500 2.7500 49 2.7211 2.7211 50 2.7105 2.7105 51 2.7030 2.7030 52 2.6108 2.6108 53 2.6023 2.6023 54 2.5877 2.5877 55 2.5846 2.5846 56 2.5673 2.5673 57 2.5545 2.5545 58 2.5497 2.5497 59 2.5487 2.5487 60 2.5094 2.5094 61 2.4928 2.4928 62 2.4893 2.4893 63 2.4816 2.4816 64 2.4758 2.4758 65 2.4687 2.4687 66 2.4617 2.4617 67 2.4487 2.4487 68 2.4232 2.4232 69 2.4158 2.4158 70 2.3786 2.3786 71 2.3671 2.3671 72 2.3502 2.3502 73 2.3489 2.3489 74 2.3396 2.3396 75 2.3375 2.3375 76 2.3210 2.3210 77 2.3195 2.3195 78 2.3078 2.3078 79 2.3077 2.3077 80 2.3069 2.3069 81 2.3038 2.3038 82 2.3002 2.3002 83 2.2951 2.2951 84 2.2899 2.2899 85 2.2473 2.2473 86 2.2462 2.2462 87 2.2459 2.2459 88 2.2435 2.2435 89 2.2302 2.2302 90 2.2205 2.2205 91 2.2101 2.2101 92 2.1994 2.1994 93 2.1914 2.1914 94 2.1855 2.1855 95 2.1750 2.1750 96 2.1643 2.1643 97 2.1640 2.1640 98 2.1502 2.1502 99 2.1478 2.1478 100 2.1142 2.1142 101 2.0924 2.0924 102 2.0888 2.0888 103 2.0839 2.0839 104 2.0791 2.0791 105 2.0705 2.0705 106 2.0704 2.0704 107 2.0012 2.0012 108 1.9889 1.9889 109 1.9488 1.9488 110 1.9424 1.9424 111 1.9244 1.9244 112 1.9107 1.9107 113 1.9099 1.9099 114 1.8889 1.8889 115 1.8883 1.8883 116 1.8849 1.8849 117 1.8794 1.8794 118 1.8763 1.8763 119 1.8726 1.8726 120 1.8717 1.8717 121 1.8640 1.8640 122 1.8568 1.8568 123 1.8559 1.8559 124 1.8533 1.8533 125 1.8513 1.8513 126 1.8263 1.8263 127 1.8181 1.8181 128 1.7987 1.7987 129 1.7897 1.7897 130 1.7881 1.7881 131 1.7836 1.7836 132 1.7761 1.7761 133 1.7391 1.7391 134 1.7354 1.7354 135 1.7299 1.7299 136 1.7055 1.7055 137 1.7018 1.7018 138 1.6897 1.6897 139 1.6709 1.6709 140 1.6644 1.6644 141 1.6583 1.6583 142 1.6545 1.6545 143 1.6328 1.6328 144 1.5989 1.5989 145 1.5523 1.5523 146 1.5505 1.5505 147 1.5327 1.5327 148 1.5287 1.5287 149 1.5196 1.5196 150 1.4713 1.4713 151 1.4708 1.4708 152 1.4239 1.4239 153 1.4004 1.4004 154 1.4003 1.4003 155 1.3968 1.3968 156 1.1186 1.1186 157 1.0865 1.0865 158 1.0761 1.0761 ______

Stream Protection Duration POC #1 The Facility PASSED

The Facility PASSED.

Flow(cfs) Predev Mit Percentage Pass/Fail 1.1536 4724 4724 100 Pass 1.1938 4186 4186 100 Pass 1.2339 3668 3668 100 Pass 1.2740 3247 3247 100 Pass 1.3141 2891 2891 100 Pass 1.3542 2593 2593 100 Pass 1.3943 2338 2338 100 Pass 1.4344 2090 2090 100 Pass 1.4745 1900 1900 100 Pass 1.5146 1695 1695 100 Pass 1.5547 1516 1516 100 Pass 1.5949 1379 1379 100 Pass 1.6350 1249 1249 100 Pass 1.6751 1123 1123 100 Pass 1.7152 1030 1030 100 Pass 1.7553 940 940 100 Pass 1.7954 852 852 100 Pass 1.8355 780 780 100 Pass 1.8756 715 715 100 Pass 1.9157 643 643 100 Pass 1.9559 589 589 100 Pass 1.9960 536 536 100 Pass 2.0361 494 494 100 Pass 2.0762 451 451 100 Pass 2.1163 416 416 100 Pass 2.1564 375 375 100 Pass 2.1965 341 341 100 Pass 2.2366 319 319 100 Pass 2.2767 289 289 100 Pass 2.3168 265 265 100 Pass 2.3570 239 239 100 Pass 2.3971 220 220 100 Pass 2.4372 195 195 100 Pass 2.4773 187 187 100 Pass 2.5174 169 169 100 Pass 2.5575 157 157 100 Pass 2.5976 142 142 100 Pass 2.6377 132 132 100 Pass 2.6778 125 125 100 Pass 2.7180 119 119 100 Pass 2.7581 113 113 100 Pass 2.7982 104 104 100 Pass 2.8383 95 95 100 Pass 2.8784 90 90 100 Pass 2.9185 84 84 100 Pass 2.9586 81 81 100 Pass 2.9987 70 70 100 Pass 3.0388 65 65 100 Pass 3.0790 62 62 100 Pass 3.1191 61 61 100 Pass 3.1592 58 58 100 Pass 3.1993 57 57 100 Pass 3.2394 55 55 100 Pass 3.2795 52 52 100 Pass 3.3196 50 50 100 Pass 3.3597 47 47 100 Pass 3.3998 42 42 100 Pass 3.4399 40 40 100 Pass 3.4801 38 38 100 Pass 3.5202 34 34 100 Pass 3.5603 30 30 100 Pass 3.6004 29 29 100 Pass 3.6405 28 28 100 Pass 3.6806 28 28 100 Pass 3.7207 27 27 100 Pass 3.7608 27 27 100 Pass 3.8009 26 26 100 Pass 3.8411 26 26 100 Pass 3.8812 25 25 100 Pass 3.9213 24 24 100 Pass 3.9614 23 23 100 Pass 4.0015 22 22 100 Pass 4.0416 20 20 100 Pass 4.0817 19 19 100 Pass 4.1218 18 18 100 Pass 4.1619 18 18 100 Pass 4.2020 17 17 100 Pass 4.2422 16 16 100 Pass 4.2823 15 15 100 Pass 4.3224 14 14 100 Pass 4.3625 14 14 100 Pass 4.4026 14 14 100 Pass 4.4427 14 14 100 Pass 4.4828 14 14 100 Pass 4.5229 13 13 100 Pass 4.5630 13 13 100 Pass 4.6032 13 13 100 Pass 4.6433 13 13 100 Pass 4.6834 12 12 100 Pass 4.7235 12 12 100 Pass 4.7636 11 11 100 Pass 4.8037 10 10 100 Pass 4.8438 10 10 100 Pass 4.8839 10 10 100 Pass 4.9240 10 10 100 Pass 4.9641 9 9 100 Pass 5.0043 9 9 100 Pass 5.0444 9 9 100 Pass 5.0845 8 8 100 Pass 5.1246 7 7 100 Pass ______

______

Water Quality BMP Flow and Volume for POC #1 On-line facility volume: 0.7139 acre-feet On-line facility target flow: 0.9632 cfs. Adjusted for 15 min: 0.9632 cfs. Off-line facility target flow: 0.5541 cfs. Adjusted for 15 min: 0.5541 cfs. ______

LID Report

LID Technique Used for Total Volume Volume Infiltration Cumulative Percent Water Quality Percent Comment Treatment? Needs Through Volume Volume Volume Water Quality Treatment Facility (ac-ft.) Infiltration Infiltrated Treated (ac-ft) (ac-ft) Credit Total Volume Infiltrated 0.00 0.00 0.00 0.00 0.00 0% No Treat. Credit Compliance with LID Standard 8 Duration Analysis Result = Passed

______

Perlnd and Implnd Changes No changes have been made. ______

This program and accompanying documentation are provided 'as-is' without warranty of any kind. The entire risk regarding the performance and results of this program is assumed by End User. Clear Creek Solutions Inc. and the governmental licensee or sublicensees disclaim all warranties, either expressed or implied, including but not limited to implied warranties of program and accompanying documentation. In no event shall Clear Creek Solutions Inc. be liable for any damages whatsoever (including without limitation to damages for loss of business profits, loss of business information, business interruption, and the like) arising out of the use of, or inability to use this program even if Clear Creek Solutions Inc. or their authorized representatives have been advised of the possibility of such damages. Software Copyright © by : Clear Creek Solutions, Inc. 2005-2018; All Rights Reserved.

APPENDIX C Contech Provided Studies Evaluation of the Stormwater Management StormFilter for the Removal of SIL-CO-SIL® 106, a standardized silica product: ZPGTM StormFilter cartridge at 28 L/min (7.5 gpm) Environmental Technology Verification Report, Stormwater Source Area Treatment Device, The Stormwater Management StormFilter® Using ZPG Filter Media

Product Evaluation

Evaluation of the Stormwater Management StormFilter® for the removal of SIL-CO-SIL® 106, a standardized silica product: ZPG™ StormFilter cartridge at 28 L/min (7.5 gpm)

Overview A Stormwater Management StormFilter® (StormFilter) ZPG™ cartridge was tested to assess its ability to remove total suspended solids (TSS) and decrease turbidity from simulated stormwater. Under controlled conditions, 7 runoff simulations (sims) were performed using influent TSS with a silt texture (20% sand, 80% silt, 0% clay), variable event mean concentrations (EMCs) between 0 and 300 mg/L, and a filtration rate of 28 L/min (7.5 gpm) (100% design, per cartridge, operating rate for this configuration). The mean TSS (silt) removal efficiency for this StormFilter cartridge configuration was determined using regression statistics and found to be 87% (P=0.05: L1=86%, L2=89%) over the range of influent EMCs tested. Turbidity data was also collected and indicated that this StormFilter cartridge configuration was capable of a 51% (P=0.05: L1=47%, L2=55%) mean decrease in turbidity.

Introduction The goal of testing the ZPG™ StormFilter cartridge was to determine its TSS and turbidity removal performance given a standardized commercial product as the contaminant surrogate. Utilizing a standardized contaminant surrogate eliminates contaminant characteristics as a variable, thereby providing opportunities to compare StormFilter performance with that of other StormFilter configurations or treatment systems tested using the same contaminant surrogate. To assure the comparability of this experiment with other StormFilter performance evaluations, the methodology used for this experiment is identical to that used in previous cartridge-scale StormFilter evaluations for solids removal (Stormwater360, 2002; SMI, 2002a).

Procedure Media A StormFilter ZPG™ cartridge was used for this experiment. This specific type of cartridge contains ZPG™ multipurpose media, a proprietary blend of organic and inorganic media (as per Stormwater360 product specifications). ZPG™ media is effective in the removal of solids, metals and organic chemicals. Prior to testing, the ZPG™ StormFilter cartridge used for testing was flushed so as to remove the residual dust within the media left over from the cartridge production process, as well as to allow the media to approach a typical, wet operating condition. Individual, ~400-L, tap water flushes were performed according to the operation segment of the procedure section. Flushing was ceased after eight flushes, at which point the effluent TSS EMC had decreased to 8.8 mg/L from an initial value of 218 mg/L.

Contaminant A commercial ground silica product, SIL-CO-SIL® 106 (SCS 106), was used as the surrogate for TSS. This product is manufactured by the US Silica Company∗ and the sample used for testing originated from the Mill Creek, OK plant. SCS 106 has a uniform specific gravity of 2.65 and is specified by the State of Washington Department of Ecology (WADOE) for the laboratory evaluation of stormwater treatment technologies (WADOE, 2002) for TSS removal. An average particle size distribution is shown in Figure 1, revealing a silt texture (USDA scale) consisting of 20% sand, 80% silt, and 0% clay-sized particles (Stormwater360, 2002). Based upon a 400-L influent volume, target TSS EMCs were determined for each planned contaminated simulation and associated masses of contaminant were placed in 1-L HDPE bottles of tap water--one bottle of concentrate per planned contaminated simulation. Target TSS EMCs were distributed between 0 and 300 mg/L. The order in which they were used was randomly selected using random number techniques so as not to bias the performance results. The SCS 106 concentrates were given the opportunity to hydrate prior to experimentation so as to promote the disintegration of any aggregate particles that may have been present. The concentrates were then left out at room temperature and periodically shaken to encourage the dissolution of any aggregates.

CLAY SILT SAND 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 % Finer (by mass) 0.2 0.1 0.0 1 10 100

Particle Size (um) Figure 1. Particle size distribution for SCS 106. Sand/silt/clay fractions according to USDA definitions are approximately 20%, 80%, and 0% for SCS 106, indicating that the texture corresponds to a silt material. Test Apparatus The typical precast StormFilter system is composed of three bays: the inlet bay, the filtration bay, and the outlet bay. Stormwater first enters the inlet bay of the StormFilter vault through the inlet pipe. Stormwater in the inlet bay is then directed through the flow spreader, which traps some floatables, oils, and surface scum, and over the energy dissipator into the filtration bay where treatment takes place. Once in the filtration bay, the stormwater begins to

∗U.S. Silica Company, P.O. Box 187, Berkeley Springs, WV 25411; (800) 243-7500; www.u-s-silica.com

pond and percolate horizontally through the media contained in the StormFilter cartridges. After passing through the media, the treated water in each cartridge collects in the cartridge’s center tube from where it is directed into the outlet bay by an under-drain manifold. The treated water in the outlet bay is then discharged through the single outlet pipe to a collection pipe or to an open channel drainage way. The test apparatus used for this experiment simulates the filtration bay component of a full-scale StormFilter system, including the energy dissipator. Since the design of full-scale StormFilter systems varies, and since the operation of a full-scale system in the laboratory environment would require very large volumes of water, the use of the most common components among all of the possible designs, the StormFilter cartridge and the associated volume of filtration bay area, were selected so as to provide a very conservative estimate of StormFilter performance. Unlike chemical removal testing, suspended solids removal testing is challenging due to the relatively large, dense, insoluble nature of the contaminant. Care must be taken to maintain the suspension of solids within the influent and effluent reservoirs, maintain the suspension of solids within the conveyance system, avoid the fouling of flow metering devices, avoid the destruction of individual solids by the pumping system, and avoid the destruction of the pumping system by the solids. Mixer Flow Meter (Recirculation) Delivery Manifold

Influent Tank Test Tank StormFilter Cartridge

Mixer

3-way (Recirculation)

Ball Valve Energy Effluent Dissipator Pump Tank

Under Drain Manifold

Pump

Figure 2. Schematic diagram of the cartridge-scale test apparatus. Arrows indicate flow pathways. Dashed arrows indicate recirculation pathways employed during influent and effluent sampling. The apparatus used for this experiment was carefully designed to meet these challenges. Figure 2 demonstrates the layout of the test apparatus. Influent and effluent storage is provided by individual 950-L (250 gallon), conical bottom polyethylene tanks (Chem- Tainer). The conical bottom design ensures full drainage of the tanks, in addition to the movement of all solids out of the tanks. Four, evenly-spaced, vertically-oriented baffles, measuring 91 x 8 x 1-cm (36 x 3 x 0.5-in) (L x W x Thickness), affixed to the sidewalls of the influent and effluent tank prevent a mixer-induced vortex. Suspension of solids within the tanks is maintained by individual, 1/2-hp, electric propeller mixers with stainless steel mixing assemblies (J.L. Wingert, B-3-TE-PRP/316). The propeller design maximizes the vertical circulation of solids within the tank and ensures the homogeneity of the mixture. Magnetic drive pumps (Little Giant, TE-6-MD-HC) are used to transfer the influent, and also to re-circulate

water through the underlying manifolds of both tanks during sampling so as to eliminate any possibility of sediment accumulation in the manifolds. Influent is carried from the influent tank by the magnetic drive pump plumbed with 25- mm (1-in) PVC hose into a PVC intake manifold below the influent tank and discharging into a delivery manifold of 25-mm PVC pipe. Despite the associated head loss, 25-mm diameter hose and pipe are used to ensure high flow velocities that maintain the suspension of solids during transfer. A 25-mm, 3-way, side-control, ball valve used for flow control assures very high flow velocities in the intake manifold, allows some degree of re-circulation back into the reservoir, and allows the high power pump to operate relatively unrestricted. Discharge from the delivery manifold into the 56 x 56 x 62-cm (22 x 22 x 24.5-in) (L x W x H) polypropylene StormFilter cartridge test tank is by discharge into the tank-mounted energy dissipater, which consists of a vertical length of 76-mm (3-in) PVC pipe with an open bottom and multiple 3-mm (0.125-in) wide horizontal slots along its entire length. The energy dissipater is a typical component of a StormFilter system and is used to minimize the re-suspension of settled material within the test tank by restricting turbulence to the region within the dissipater. Discharge from the StormFilter cartridge test tank into the effluent tank is through free discharge from the under-drain manifold component of the test tank positioned over the top of the effluent tank. Flow into the StormFilter cartridge test tank is controlled by the 3-way ball valve placed between the pump and the delivery manifold, and flow is monitored with a paddle-wheel type electronic flow meter (GF Signet, Rotor-X Low Flow) coupled with a flow transmitter with totalizer (GF Signet, Processpro). Operation The operational procedure consisted of performing multiple runoff simulations (sims) using the same StormFilter cartridge test tank and apparatus described in the Test Apparatus section above. Sims proceeded as follows. The influent tank was filled with ~400-L of tap water, and the predetermined contaminant concentrate was added to the influent tank. The influent tank was then mixed thoroughly with the mechanical mixer while influent was re-circulated through the underlying manifold and allowed to equilibrate for 5 to 10 minutes before sampling. Following influent sample collection, a portion of flow was redirected to the test tank energy dissipator via the delivery manifold through adjustment of the 3-way valve. Flow rate was controlled through periodic adjustment of the 3-way valve so as to maintain a constant flow rate reading of 28 L/min ± 2 L/min (7.5 gpm ± 0.5 gpm). Mixing and re-circulation of the effluent reservoir was started towards the end of a sim to allow effluent equilibration prior to sample collection. The influent pump was operated until as much of the influent had been pumped from the influent reservoir and underlying manifold as was possible, at which point the influent pump was shut down and the StormFilter cartridge test tank was allowed to drain. Once the float valve within the StormFilter cartridge closed, effluent was sampled and the total sim volume reported by the totalizer was recorded. Sampling Composite samples of influent and effluent were collected for TSS and turbidity analysis. One set of samples was collected for TSS analysis by North Creek Analytical (NCA), Beaverton, OR, and an additional set was collected for internal turbidity analysis. For this document, a set is defined as a collection of influent and effluent sample pairs corresponding to a specific sim. Sample handling was performed in accordance with standard handling techniques. All samples to be tested for TSS were promptly refrigerated following collection. Samples were shipped to the laboratory in coolers, accompanied by ice-packs and chain-of-custody documentation for analysis within seven days. NCA performed TSS analysis according to

ASTM method D3977, which is essentially the same as the “whole-sample” variation of EPA method 160.2 (SMI, 2002b). Samples were extracted with a 1-L PE, 1.2-m ladle using a sweeping motion across and through the center of the reservoir. Six 1-L grab samples were collected in an 8-L churn sample splitter (Bel-Art Products) for composite sample extraction according to manufacturer instructions. Care was taken to transfer all solids from the ladle through quick emptying of the ladle while using a swirling motion. The churn splitter was used to dispense approximately 250- mL of composite sample into 250-mL (8-oz) HDPE bottles for TSS analysis and an additional 500-mL composite sample was dispensed to a 1-L (32-oz) HDPE bottle for turbidity analysis. The sampling ladle and churn splitter were subject to a high-pressure wash between uses. Internal Analysis Turbidity, a measure of the light-dispersing characteristics of a fluid, was measured using a bench-top turbidimeter (LaMotte Model 2020). The sample was swirled in its bottle immediately before pouring a subsample to the turbidimeter tube. The tube was wiped clean of moisture using lint-free wipes and then swirled, taking care to prevent bubbles in the sample and to maintain a clean tube surface, prior to insertion into the turbidimeter. The turbidimeter tube was rinsed with deionized water between each use. Results TSS removal and turbidity results are shown in Table 1. The discrete efficiencies, efficiencies of individual pairs of associated influent and effluent TSS EMCs, suggest an increase with increasing influent TSS EMC. A similar trend is evident for the generally increasing turbidity reduction contrasted to increasing average influent turbidity.

Table 1. Summary of influent and effluent TSS EMCs and turbidity along with TSS removal and turbidity decrease results shown according to increasing influent TSS EMC.

Discrete TSS Average Average Discrete Influent Effluent Sim Removal Influent Effluent Turbidity TSS EMC TSS EMC Sim Volume Efficiency Turbidity Turbidity Decrease (mg/L) (mg/L) (L) (%) (NTU) (NTU) (%) ND (4.00) 7.09 addition 0.45 2.3 addition 7 401 25.4 14.2 44.1 4.1 5.4 addition 4 398 49.1 17.0 65.4 8.8 7.7 12.5 6 397 107 21.1 80.3 17 10.2 40.0 1 393 144 28.2 80.4 25 15 40.0 2 396 188 33.2 82.3 35 19 45.7 5 393 292 45.5 84.4 53 29 45.3 3 389

Discussion Quality Control For TSS analysis, Method Blank and Duplicate quality control samples are typically used to measure bias and precision. Method Blank results as reported by the analytical laboratory were non-detect (<4 mg/L) for the four sets of analyses that comprised the data set shown in Table 1. Unfortunately, since the “whole-sample” nature of ASTM method D3977 involves the use of the entire sample volume, none of the sample volume is left over for traditional Duplicate analysis. Thus dedicated Duplicate samples were collected for 2 of the 14 TSS analyses (14% duplicates) and are displayed in Table 2. The results of the Method Blank and Duplicate analyses demonstrate an acceptable level of bias and precision according to SMI (2002c).

Table 2. Summary of Quality Control results.

Official Duplicate Relative Sim Influent/Effluent Result Result Percent (I or E) (mg/L) (mg/L) Difference (%) 2 I 144 143 0.7 2 E 28.2 29.0 2.8

TSS and Turbidity Removal Performance Evaluation The graphed results of the external TSS analysis, displayed in Figure 3, show a regressed removal efficiency of 87% (P=0.05: L1=86%, L2=89%), which is calculated by subtracting the regression coefficient (slope) from 1. Based upon an analysis of variance (ANOVA), the regression is significant at the P<0.001 level (<0.1% probability of no correlation between influent and effluent TSS EMC’s). Coupled with y-intercept and regression coefficients that are both significant at the P<0.001 levels, this signals a good fit of the data points to the regression equation, which is visually supported by the tight 95% confidence intervals. At P<0.001, the confidence in the TSS EMC removal performance estimate is within the 0.05 limit considered by SMI (2002d) to indicate a valid estimate.

200 ANOVA Source of Variation df SS MS F Explained 1 987.3 987.3 303.8*** Unexplained 5 16.25 3.2495 Total 6 1003.5 150 SIGNIFICANCE OF COEFFICIENTS Coeff. Std. Error t y0=9.193 1.078 8.527*** a=0.1259 0.0072 17.43*** 100 * = 0.01 < P < 0.05 ** = 0.001 < P < 0.01 ***= P < 0.001

Effluent TSS EMC (mg/L) TSS Effluent 50

Regression Equation: y = 0.13x + 9.19 r2 = 0.984 0 0 100 200 300 Influent TSS EMC (mg/L) Figure 3. Regression analysis applied to the TSS data associated with the estimation of the SCS 106 TSS removal efficiency of the ZPG™ StormFilter cartridge at 28 L/min. The solid line is the regression. The dotted lines signify the lower and upper 95% confidence intervals. ANOVA indicates a significant (P<0.001) linear relationship between influent and effluent TSS EMC. The decrease in turbidity associated with the ZPG™ cartridge test is less than the reduction of TSS. The mean turbidity reduction, shown in Figure 4, was observed to be 51% (P=0.05: L1=47%, L2=55%) based upon regression analysis that is significant at the P<0.001 level. The y-intercept and regression coefficients, significant at the P<0.01 and P<0.001 levels, respectively, provide ample confidence in the observed relationship.

35 ANOVA Source of Variation df SS MS F 30 Explained 1 500.6 500.6 893.0*** Unexplained 5 2.803 0.561 Total 6 503.4

25 SIGNIFICANCE OF COEFFICIENTS Coeff. Std. Error t y0=2.675 0.4378 6.111** a=0.4874 0.0163 29.88*** 20 * = 0.01 < P < 0.05 ** = 0.001 < P < 0.01 15 ***= P < 0.001

10 Effluent Turbidity (NTU) Turbidity Effluent

5 Regression Equation: y = 0.49x + 2.68 r2 = 0.994 0 0 102030405060

Influent Turbidity (NTU) Figure 4. SCS 106 turbidity reduction by the ZPG™ StormFilter cartridge at 28 L/min. The solid line is the regression. The dotted lines signify the upper and lower 95% confidence intervals. ANOVA indicates a significant (P<0.001) linear relationship between influent and effluent turbidity.

TSS Removal Performance with Regard to Particle Size Based upon the particle size distribution presented in Figure 1, SCS 106 consists primarily of silt-sized silica particles (80% by mass between 2 and 50 microns). Combined with the TSS removal estimate of 87% (by mass) demonstrated in Figure 3, some qualitative inferences concerning the particle size specific removal efficiency of the system can be made. Assuming that larger particles are preferentially removed over smaller particles, it could be said that the system under review removed particles down to the 6 micron level since, conservatively, 87% (by mass) of SCS 106 is composed of silica particles larger than 6 microns. Since it is likely that some particles smaller than 6 microns were retained and some particles larger than 6 microns were lost by the system, the efficiency of the system under review with regard to particle size is probably best represented by a size range. With this in mind, a better qualitative statement with regard to the particle size removal efficiency of the system under review would be that it is capable of removing silica particles in the vicinity of 10 microns.

Conclusions The tests utilizing SCS 106 as a contaminant generated results for the assessment of the silt TSS and turbidity removal efficiency of the ZPG™ StormFilter cartridge. The use of a standardized contaminant surrogate allows the results from laboratory evaluations of the TSS removal performance of stormwater treatment systems to be easily compared. In summary:

1. A ZPG™ StormFilter cartridge test unit, operating at 28 L/min, and subject to TSS with a silt texture (20% sand, 80% silt, and 0% clay by mass) originating from SCS 106 provides a mean TSS removal efficiency of 87% (P=0.05: L1=86%, L2=89%); 2. A ZPG™ StormFilter cartridge test unit, operating at 28 L/min, and subject to TSS with a silt texture (20% sand, 80% silt, and 0% clay by mass) originating from SCS 106 provides a mean turbidity reduction of 51% (P=0.05: L1=47%, L2=55%); 3. A ZPG™ StormFilter cartridge test unit, operating at 28 L/min is effective on silica particles down to the 10 micron size range;

It is important to emphasize that these conclusions reflect laboratory-based testing performed under controlled conditions. Field conditions are notoriously variable with regard to TSS characteristics and sampling methods, and comparison of this experiment to field-derived data will be accordingly affected. Laboratory studies are beneficial for the evaluation of system performance potential as part of the product development or system comparison process.

Stormwater360, Stormwater Management Inc, and Vortechnics Inc. are now CONTECH Stormwater Solutions Inc.

References Stormwater360. (2002). Evaluation of the Stormwater Management StormFilter® cartridge for the removal of SIL-CO-SIL 106, a synthetically graded sand material: Coarse/fine perlite StormFilter cartridge at 28 L/min (7.5 gpm). (Report No. PD-02-003.1). Portland, Oregon: Author.

Stormwater Management Inc (SMI). (2002a). Evaluation of the Stormwater Management StormFilter® cartridge for the removal of SIL-CO-SIL 106, a synthetically graded sand material: Coarse perlite StormFilter cartridge at 28 L/min (7.5 gpm). (Report No. PD-02-002.1). Portland, Oregon: Author.

Stormwater Management Inc. (2002b). Influence of analytical method, data summarization method, and particle size on total suspended solids (TSS) removal efficiency (Report No. PD- 02-006.1). Portland, Oregon: Author.

Stormwater Management Inc (SMI). (2002c). Stormwater Management StormFilter Quality Assurance Project Plan. Portland, Oregon: Author.

Stormwater Management Inc. (2002d). Performance Summarization Guidelines (SMI PD-02- 001.0). Portland, OR: Author.

State of Washington Department of Ecology (WADOE). (2002, October). Guidance for Evaluating Emerging Stormwater Treatment Technologies: Technology Assessment Protocol— Ecology (WADOE Publication No. 02-10-037). Retrieved November 11, 2002, from: http://www.ecy.wa.gov/programs/wq/stormwater/newtech/02-10-037%20TAPE.pdf

Revision Summary PE-E062 Document rebranded.

PE-E061 Document number changed; document rebranded; no substantial changes.

PD-04-006.0 Original

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

THE ENVIRONMENTAL TECHNOLOGY VERIFICATION PROGRAM

U.S. Environmental Protection Agency NSF International

ETV Joint Verification Statement

TECHNOLOGY TYPE: STORMWATER TREATMENT TECHNOLOGY APPLICATION: SUSPENDED SOLIDS AND ROADWAY POLLUTANT TREATMENT TECHNOLOGY NAME: THE STORMWATER MANAGEMENT STORMFILTER® USING ZPG FILTER MEDIA TEST LOCATION: MILWAUKEE, WISCONSIN COMPANY: STORMWATER MANAGEMENT, INC. ADDRESS: 12021-B NE Airport Way PHONE: (800) 548-4667 Portland, Oregon 97220 FAX: (503) 240-9553 WEB SITE: http://www.stormwaterinc.com EMAIL: mail@ stormwaterinc.com

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. July 2004 VS-i 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 Nutrients Water Quality Parameters • total suspended solids (TSS) • total and • total and • chemical oxygen • total dissolved solids (TDS) dissolved dissolved demand (COD) • suspended sediment cadmium, lead, phosphorus • dissolved chloride concentration (SSC) copper and zinc • total calcium and • particle size analysis magnesium VERIFICATION OF PERFORMANCE 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 Peak Rainfall Rainfall Runoff Discharge Event Start Start Amount Duration Volume Rate Number Date Time (inches) (hr:min) (ft3)1 (gpm)1 1 6/21/02 6:54 0.52 0:23 420 447 2 7/8/02 21:16 1.5 2:04 1,610 651 3 8/21/02 20:08 1.7 15:59 1,620 671 4 9/2/02 5:24 1.2 3:24 1,180 164 5 9/18/02 5:25 0.37 4:54 350 136 6 9/29/02 0:49 0.74 7:54 730 70.9 7 12/18/02 1:18 0.37 3:47 300 61.0 8 4/19/03 5:39 0.55 10:00 340 96.9 9 5/4/03 21:21 0.90 11:44 540 73.2 10 5/30/03 18:55 0.54 4:06 320 83.9 11 6/8/03 3:26 0.62 11:09 450 140 12 6/27/03 17:30 0.57 13:25 460 107 13 7/4/03 7:25 0.53 40:43 550 143 14 7/8/03 9:49 0.33 3:37 260 62.8 15 9/12/03 15:33 0.22 1:55 150 21.5 16 9/14/03 5:22 0.47 6:35 340 264 17 9/22/03 2:28 0.27 2:09 270 104 18 10/14/03 1:03 0.25 2:07 220 56.5 19 10/24/03 16:46 0.71 15:07 410 75.8 20 11/4/03 16:14 0.60 2:09 560 906 1 Runoff volume and peak discharge volume was measured at the outlet monitoring point.

04/17/WQPC-WWF The accompanying notice is an integral part of this verification statement. July 2004 VS-iii

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

SOL

Inlet Outlet EMC Range Reduction Parameter1 Units Range Range (percent) (percent) TSS mg/L 29 – 780 20 – 380 -33 – 95 46 SSC mg/L 51 – 5,600 12 – 370 3 – 99 92 TDS mg/L <50 – 600 <50 – 4,2002 -600 – 10 -1702 Total phosphorus mg/L as P 0.05 – 0.63 0.03 – 0.30 0 – 70 38 Dissolved phosphorus mg/L as P 0.01 – 0.20 0.01 – 0.19 -35 – 38 6 Total magnesium mg/L 4.0 – 174 1.1 – 26 53 – 96 85 Total calcium mg/L 9.4 – 430 4.0 – 68 26 – 93 79 Total copper µg/L 15 – 440 7.0 – 140 8.3 – 96 59 Total lead µg/L <31 – 280 <31 – 94 33 – 91 64 Total zinc µg/L 77 – 1,400 28 – 540 20 – 89 64 Dissolved copper µg/L <5 – 58 <5 – 42 -47 – 64 16 Dissolved zinc µg/L 26 – 360 16 – 160 -86 – 56 17 COD mg/L 18 – 320 17 – 190 -91 – 47 16 Dissolved chloride mg/L 3.2 – 470 3.3 – 2,6002 -740 – 24 -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 µm in size, while 78 percent of the sediment in the outlet samples was less than 62.5 µm 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.

i 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

ii 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 Chapter 3 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

iii 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 a typical 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

v Abbreviations and Acronyms

ASTM American Society for Testing and Materials BMP Best Management Practice cfs Cubic feet per second COD Chemical oxygen demand EMC Event mean concentration EPA U.S. Environmental Protection Agency ETV Environmental Technology Verification ft2 Square feet ft3 Cubic feet g Gram gal Gallon gpm Gallon per minute in Inch kg Kilogram L Liters lb Pound LOD Limit of detection LOQ Limit of quantification NRMRL National Risk Management Research Laboratory µg/L Microgram per liter (ppb) µm Micron mg/L Milligram per liter NSF NSF International, formerly known as National Sanitation Foundation NIST National Institute of Standards and Technology O&M Operations and maintenance QA Quality assurance QAPP Quality Assurance Project Plan QC Quality control SMI Stormwater Management, Inc. SSC Suspended sediment concentration SOL Sum of loads SOP Standard Operating Procedure TDS Total dissolved solids TO Testing Organization TP Total phosphorus TSS Total suspended solids USGS United States Geological Survey VA Visual accumulator VO Verification Organization (NSF) VTP Verification test plan WDNR Wisconsin Department of Natural Resources WQPC Water Quality Protection Center WisDOT Wisconsin Department of Transportation WSLH Wisconsin State Laboratory of Hygiene ZPG 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)

1 The following is a brief description of each ETV participant and their roles and responsibilities.

1.2.1 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: [email protected]

U.S. EPA, NRMRL Urban Watershed Management Research Laboratory 2890 Woodbridge Avenue (MS-104) Edison, New Jersey 08837-3679

1.2.2 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.

2 Key contacts at NSF are:

Mr. Thomas Stevens, Program Manager Mr. Patrick Davison, Project Coordinator (734) 769-5347 (734) 913-5719 email: [email protected] email: [email protected]

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: [email protected]

Earth Tech, Inc. 1210 Fourier Drive Madison, Wisconsin 53717

United States Geologic Survey:

Ms. Judy Horwatich (608) 821-3874 email: [email protected]

3 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: [email protected] email: [email protected]

WSLH USGS Sediment Laboratory 2601 Agriculture Drive Federal Building Room 269 Madison, Wisconsin 53718 400 South Clinton Street Iowa City, Iowa 52240

1.2.5 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: [email protected]

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.

4 The key contact for WisDOT is:

Mr. Robert Pearson (608) 266-7980 email: [email protected]

Bureau of Environment Wisconsin Department of Transportation 4802 Sheboygan Avenue, Room 451 Madison, Wisconsin 53707

5 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 stormwater. 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.

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.

6 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.

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.

7 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 (I-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.

Test Site

I-794 an g

Milwaukee I-43 River Milwaukee Harbor – Lake Michi Milwaukee Harbor – Lake

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.

Milwaukee River

StormFilter Drainage Area

I-794 Eastbound Lanes

Figure 3-2. Drainage area detail.

Inlet to StormFilter

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.

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.

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.

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.

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

Rain StormFilter Equipment Site 1 Site 2 Gauge Vault Water ISCO 3700 refrigerated ISCO 3700 refrigerated Quality automatic sampler (4, automatic sampler (4, Sampler 10 L sample bottles) 10 L sample bottles) Velocity Marsh-McBirney Marsh-McBirney Measurement Velocity Meter Model Velocity Meter Model 270 270 Stage Meter Marsh-McBirney Marsh-McBirney Campbell Velocity Meter Model Velocity Meter Model Scientific Inc. 270 270 SWD1 Datalogger Campbell Scientific Campbell Scientific Campbell Inc. CR10X datalogger Inc. CR10X datalogger Scientific Inc. CR10X datalogger Rain Gauge Rain-O- Matic

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

Reporting Limit of Limit of Parameter Units Detection Quantification Method1 Total dissolved solids (TDS) mg/L 50 167 SM 2540C Total suspended solids (TSS) mg/L 2 7 EPA 160.2 Total phosphorus mg/L as P 0.005 0.016 EPA 365.1 Suspended sediment mg/L 0.1 0.5 ASTM D3977-97 concentration (SSC) Total calcium mg/L 0.2 0.7 EPA 200.7 Total copper µg/L 1 3 SM 3113B Dissolved copper µg/L 1 3 SM 3113B Total magnesium mg/L 0.2 0.7 EPA 200.7 Dissolved zinc µg/L 16 50 EPA 200.7 Total zinc µg/L 16 50 EPA 200.7 Dissolved phosphorus mg/L as P 0.005 0.016 EPA 365.1 Dissolved cadmium µg/L 6 20 EPA 200.7 Total cadmium µg/L 6 20 EPA 200.7 Total lead µg/L 31 100 EPA 200.7 Dissolved lead µg/L 31 100 EPA 200.7 Dissolved chloride mg/L 0.6 2 EPA 325.2 Chemical oxygen demand mg/L 9 28 ASTM D1252-88(B) (COD) Sand-silt split NA NA NA Fishman et al. Five point sedigraph NA NA NA Fishman et al. Sand fractionation NA NA NA 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.

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.

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 6/21/02 6:54 6/21/02 7:40 7 6/21/02 6:57 6/21/02 7:34 7 2 7/8/02 21:21 7/8/02 23:41 29 7/8/02 21:24 7/8/02 23:26 29 3 8/21/02 20:12 8/22/02 12:37 30 8/21/02 20:27 8/22/02 12:21 16 4 9/2/02 5:25 9/2/02 9:48 21 9/2/02 5:30 9/2/02 9:12 24 5 9/18/02 5:31 9/18/02 10:25 10 9/18/02 5:54 9/18/02 10:49 8 6 9/29/02 2:52 9/29/02 9:27 9 9/29/02 3:19 9/29/02 9:33 16 7 12/18/02 1:19 12/18/02 6:02 18 12/18/02 1:44 12/18/02 6:05 9 8 4/19/03 5:56 4/19/03 15:55 18 4/19/03 6:04 4/19/03 15:57 15 9 5/4/03 21:28 5/5/03 7:18 23 5/4/03 21:35 5/5/03 7:18 26 10 5/30/03 19:00 5/30/03 23:22 13 5/30/03 19:05 5/30/03 23:59 15 11 6/8/03 3:30 6/8/03 14:55 14 6/8/03 3:32 6/8/03 15:10 20 12 6/27/03 17:32 6/28/03 11:01 18 6/27/03 17:43 6/28/03 11:34 22 13 7/4/03 7:27 7/6/03 9:47 19 7/4/03 7:30 7/6/03 10:26 26 14 7/8/03 9:52 7/8/03 13:45 8 7/8/03 9:59 7/8/03 14:06 11 15 9/12/03 15:35 9/12/03 17:31 8 9/12/03 16:12 9/12/03 18:23 7 16 9/14/03 5:34 9/14/03 12:05 15 9/14/03 6:11 9/14/03 12:10 11 17 9/22/03 2:29 9/22/03 4:54 8 9/22/03 2:36 9/22/03 4:35 13 18 10/14/03 1:11 10/14/03 3:21 15 10/14/03 1:25 10/14/03 3:34 10 19 10/24/03 16:59 10/24/03 21:49 20 10/24/03 17:10 10/24/03 22:19 20 20 11/4/03 15:58 11/4/03 19:20 10 11/4/03 16:18 11/4/03 19:48 14

Table 4-4. Rainfall Summary for Monitored Events

Peak Rainfall Rainfall Runoff Discharge Event Start Start End Amount Duration Volume Rate Number Date Time End Date Time (inches) (hr:min) (ft3)1 (gpm)1 1 6/21/02 6:54 6/21/02 7:17 0.52 0:23 420 447 2 7/8/02 21:16 7/8/02 23:20 1.5 2:04 1,610 651 3 8/21/02 20:08 8/22/02 12:07 1.7 15:59 1,620 671 4 9/2/02 5:24 9/2/02 8:48 1.2 3:24 1,180 164 5 9/18/02 5:25 9/18/02 10:19 0.37 4:54 350 136 6 9/29/02 0:49 9/29/02 8:43 0.74 7:54 730 70.9 7 12/18/02 1:18 12/18/02 5:05 0.37 3:47 300 61.0 8 4/19/03 5:39 4/19/03 15:39 0.55 10:00 340 96.9 9 5/4/03 21:21 5/5/03 9:05 0.90 11:44 540 73.2 10 5/30/03 18:55 5/30/03 23:01 0.54 4:06 320 83.9 11 6/8/03 3:26 6/8/03 14:35 0.62 11:09 450 140 12 6/27/03 17:30 6/28/03 10:55 0.57 13:25 460 107 13 7/4/03 7:25 7/6/03 10:08 0.53 40:43 550 143 14 7/8/03 9:49 7/8/03 13:26 0.33 3:37 260 62.8 15 9/12/03 15:33 9/12/03 17:28 0.22 1:55 150 21.5 16 9/14/03 5:22 9/14/03 11:57 0.47 6:35 340 264 17 9/22/03 2:28 9/22/03 4:37 0.27 2:09 270 104 18 10/14/03 1:03 10/14/03 3:10 0.25 2:07 220 56.5 19 10/24/03 16:46 10/24/03 11:53 0.71 15:07 410 75.8 20 11/4/03 16:14 11/4/03 18:23 0.60 2:09 560 906

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

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.

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 × (1-[EMCeffluent/EMCinfluent]) (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.

Table 5-1. Monitoring Results and Efficiency Ratios for Sediment Parameters

TSS SSC TDS Event Rainfall Inlet Outlet Reduction Inlet Outlet Reduction Inlet Outlet Reduction No. (in) (mg/L) (mg/L) (Percent) (mg/L) (mg/L) (Percent) (mg/L) (mg/L) (Percent) 11 0.52 71 83 -17 370 63 83 <50 <50 - 21 1.5 51 28 45 310 20 94 <50 <50 - 3 1.7 NA NA - 65 19 71 <50 <50 - 4 1.2 NA NA - 320 13 96 39 38 3 5 0.37 NA NA - 120 43 64 NA NA - 6 0.74 NA NA - 140 12 91 <50 <50 - 7 0.37 NA NA - 770 130 83 600 4,200 -600 8 0.55 780 380 51 5,600 370 93 520 720 -38 9 0.90 73 34 53 830 34 96 78 90 -15 10 0.54 110 70 36 1,300 68 95 66 130 -91 11 0.62 60 40 33 420 40 90 <50 76 - 12 0.57 77 46 40 370 47 87 90 160 -80 13 0.53 29 30 -3 51 32 37 60 110 -83 14 0.33 57 24 58 74 23 69 82 110 -34 15 0.22 700 36 95 3,800 29 99 210 190 10 16 0.47 50 49 2 410 49 88 <50 60 - 17 0.27 37 31 16 480 21 96 50 80 -60 18 0.25 35 20 43 410 21 95 50 74 -48 19 0.71 67 36 46 420 33 92 <50 60 - 20 0.60 55 73 -33 100 97 3 <50 <50 -

1 SSC analyzed at USGS Sediment Laboratory; all other parameters analyzed at WSLH NA: Not Analyzed

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 µg/L, and the EMC reduction ranged from 8 to 96 percent. The total zinc inlet concentration ranged from 77 to 1,400 µg/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 µg/L, and the EMC reduction ranged from –47 to 64 percent. The dissolved zinc inlet concentration ranged from 26 to 360 µg/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.

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 0.14 0.10 29 0.041 0.039 4.9 2 0.11 0.08 27 0.041 0.037 9.8 3 0.05 0.04 20 0.014 0.013 7.1 4 0.10 0.05 50 0.030 0.032 -6.7 5 0.14 0.10 29 0.059 0.046 22 6 0.10 0.03 70 0.021 0.021 0.0 7 0.33 0.20 39 0.035 0.029 17 8 0.50 0.29 42 0.027 0.017 37 9 0.17 0.08 53 0.057 0.043 25 10 0.20 0.14 30 0.045 0.028 38 11 0.19 0.08 58 0.023 0.028 -22 12 0.24 0.19 21 0.061 0.059 3.3 14 0.16 0.11 31 0.048 0.049 -2.1 15 0.63 0.30 52 0.20 0.19 5.0 16 0.10 0.10 0 0.020 0.027 -35 17 0.15 0.10 33 0.043 0.054 -26 18 0.15 0.10 33 0.040 0.046 -15

1 Phosphorus parameters were not analyzed during events 13, 19 or 20.

Table 5-3. Monitoring Results and Efficiency Ratios for Metals

Total Copper Dissolved Copper Total Zinc Dissolved Zinc Event Inlet2 Outlet Reduction Inlet Outlet Reduction Inlet2 Outlet Reduction Inlet Outlet Reduction No.1 (µg/L) (µg/L) (Percent) (µg/L) (µg/L) (Percent) (µg/L) (µg/L) (Percent) (µg/L) (µg/L) (Percent) 1 41 28 32 <5 <5 - 220 140 36 60 34 43 2 34 19 44 10 8.8 12 200 76 62 59 51 14 3 15 10 33 6.1 5.4 11 180 39 78 27 20 26 4 29 10 66 7.7 7.0 9 200 56 72 49 43 12 5 130 30 77 21 14 33 680 110 84 87 51 41 6 16 7 56 5.0 4.5 10 77 28 64 26 16 38 7 130 78 40 14 20 -47 390 300 23 59 110 -86 8 280 140 50 28 27 3 1,400 540 61 110 84 24 9 44 20 55 11 8.7 24 230 91 60 64 45 30 10 79 42 47 17 15 10 240 140 42 67 70 -4 11 36 23 36 18 7.6 58 120 84 30 37 32 14 12 48 44 8 20 23 -13 200 160 20 81 96 -19 14 36 29 19 13 15 -14 230 79 66 57 42 26 15 330 69 79 58 42 27 1,400 210 85 360 160 56 16 32 21 34 5.5 6.2 -13 180 110 39 26 30 -15 17 440 18 96 9.0 11 -17 650 69 89 42 47 -12 18 46 15 67 50 18 64 300 66 78 46 42 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).

Table 5-3 (cont’d).

Total Cadmium Dissolved Cadmium Total Lead Dissolved Lead Event Inlet Outlet Reduction Inlet Outlet Reduction Inlet Outlet Reduction Inlet Outlet Reduction No.1 (µg/L) (µg/L) (percent) (µg/L) (µg/L) (percent) (µg/L) (µg/L) (percent) (µg/L) (µg/L) (percent) 1 <6 NA - <6 NA - <31 NA - <31 NA - 2 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 3 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 4 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 5 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 6 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 7 <6 <6 - <6 <6 - 130 72 45 <31 <31 - 8 <6 <6 - <6 <6 - 190 <31 91 <31 <31 - 9 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 10 <6 <6 - <6 <6 - 53 32 40 <31 <31 - 11 <6 <6 - <6 <6 - 33 <31 52 <31 <31 - 12 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 14 <6 <6 - <6 <6 - <31 <31 - <31 <31 - 15 <6 <6 - <6 <6 - 280 37 87 <31 <31 - 16 <6 <6 - <6 <6 - 140 94 33 <31 <31 - 17 <6 <6 - <6 <6 - 110 53 52 <31 <31 - 18 <6 <6 - <6 <6 - <31 <31 - <31 <31 -

1 Metals parameters were not analyzed during events 13, 19 or 20. NA: Not analyzed

Table 5-4. Monitoring Results and Efficiency Ratios for Water Quality Parameters

Chemical Oxygen Demand Dissolved Chloride Total Calcium Total Magnesium Event No.1 Inlet Outlet Reduction Inlet Outlet Reduction Inlet Outlet Reduction Inlet Outlet Reduction (mg/L) (mg/L) (Percent) (mg/L) (mg/L) (Percent) (mg/L) (mg/L) (Percent) (mg/L) (mg/L) (Percent) 1 42 37 12 5.8 5.2 10 42 15 64 21 5.8 72 2 39 25 36 4.6 4.6 0 28 6 79 14 1.9 86 3 18 24 -33 4.5 3.4 24 9.7 4.4 55 4.2 1.6 62 4 29 24 17 3.2 3.3 -3 55 4.4 92 26 1.4 95 5 80 78 2.5 NA NA - 17 9.7 43 7.3 3.2 56 6 28 17 39 3.6 4.0 -11 9.4 4 57 4.0 1.1 73 7 68 130 -91 310 2,600 -740 130 48 63 56 8.5 85 8 320 190 41 470 660 -40 430 68 84 174 26 85 9 53 38 28 25 31 -24 62 11 82 28 2.8 90 10 67 61 9.0 14 32 -130 40 17 58 18 4.8 73 11 41 36 12 9.4 17 -81 37 9.6 74 18 3.0 83 12 85 81 4.7 17 35 -110 29 17 41 11 4.2 62 14 63 53 16 20 22 -10 12 8.9 26 4.9 2.3 53 15 300 160 47 34 35 -3 230 16 93 120 4.4 96 16 38 34 11 6.1 9.7 -59 41 8.8 79 20 3.7 82 17 48 72 -50 9 16 -78 73 8.3 89 36 2.5 93 18 51 50 2.0 5.4 NA - 60 7 88 22 1.9 91

1 Parameters were not analyzed during events 13, 19 or 20. NA: Not Analyzed

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 = 100×(1-(A/B)) (5-2)

where:

A = Sum of Effluent Load = (Effluent EMC1)(Flow Volume1) + (Effluent EMC2)(Flow Volume2) + (Effluent EMCn)(Flow Volumen)

B = Sum of Influent Load = (Influent EMC1)(Flow Volume1) + (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.

Table 5-5. Sediment Sum of Loads Efficiencies Calculated Using Various Flow Volumes

Flow Load Reduction Efficiency (Percent)1 Location TSS SSC TDS Inlet only 47 92 -45 Outlet only 46 92 -46 Inlet and Outlet 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 SMI’s 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 SMI’s performance claim of 30 to 45 percent reduction. Additionally, the dissolved phosphorus load reduction of six percent also met SMI’s performance claim of negligible dissolved phosphorus removal.

Table 5-6. Sediment Sum of Loads Results

TSS SSC TDS Runoff Inlet Inlet Outlet Outlet Inlet Inlet Outlet Outlet Inlet Inlet Outlet Outlet Event No Volume (ft3) (mg/L) (lb) (mg/L) (lb) (mg/L) (lb) (mg/L) (lb) (mg/L) (lb) (mg/L) (lb) 1* 420 71 1.9 83 2.2 370 9.8 63 1.7 <50 0.7 <50 0.7 2* 1,610 51 5.2 28 2.8 310 32 20 2.0 <50 2.5 <50 2.5 3 1,620 NA - NA - 65 6.6 19 1.9 <50 2.5 <50 2.5 4 1,180 NA - NA - 320 24 13 1.0 39 2.9 38 2.8 5 350 NA - NA - 120 2.6 43 0.9 NA - NA - 6 730 NA - NA - 140 6.3 12 0.6 <50 1.1 <50 1.1 7 300 NA - NA - 770 14 130 2.4 600 11 4,200 79 8 340 780 17 380 8.1 5,600 120 370 8.0 520 11 720 15 9 540 73 2.5 34 1.2 820 28 34 1.2 78 2.6 90 3.1 10 320 110 2.3 70 1.4 1,300 26 68 1.4 66 1.3 130 2.5 11 450 60 1.7 40 1.1 420 12 40 1.1 <50 0.7 76 2.1 12 460 77 2.2 46 1.3 370 11 47 1.4 90 2.6 160 4.7 13 550 29 1.0 30 1.0 51 1.8 32 1.1 60 2.1 110 3.8 14 260 57 0.9 24 0.4 74 1.2 23 0.4 82 1.3 110 1.8 15 150 700 6.6 36 0.3 3,800 35 29 0.3 210 2.0 190 1.8 16 340 50 1.1 49 1.0 400 8.7 49 1.0 <50 0.5 60 1.3 17 270 37 0.6 31 0.5 480 8.2 21 0.4 50 0.8 80 1.4 18 220 35 0.5 20 0.3 410 5.7 21 0.3 50 0.7 74 1.0 19 410 67 1.7 36 0.9 420 11 33 0.9 <50 0.6 60 1.5 20 560 55 1.9 73 2.6 100 3.6 97 3.4 <50 0.9 <50 0.9 Total (all events monitored) 47 25 370 31 48 130 Load Reduction Efficiency (Percent) 46 92 -170 SSC Total (omitting events 3-7 ) 314 24 Load Reduction Efficiency (Percent) 92 TDS Total (omitting event 7) 37 51 Load Reduction Efficiency (Percent) -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.

Table 5-7. Nutrient Sum of Loads Results

Total Phosphorus Dissolved Phosphorus (g) (g) Event No. Inlet Outlet Inlet Outlet 1 1.7 1.2 0.49 0.47 2 4.8 3.6 1.87 1.68 3 2.1 1.7 0.64 0.60 4 3.3 1.6 1.00 1.06 5 1.4 1.0 0.59 0.46 6 2.0 0.67 0.44 0.44 7 2.8 1.7 0.30 0.25 8 4.8 2.8 0.26 0.16 9 2.6 1.2 0.88 0.66 10 1.8 1.3 0.41 0.25 11 2.5 1.0 0.29 0.36 12 3.0 2.5 0.79 0.77 14 1.2 0.79 0.35 0.36 15 2.6 1.2 0.83 0.80 16 1.0 0.91 0.19 0.26 17 1.2 0.74 0.33 0.41 18 0.91 0.60 0.24 0.28 Total: 40 24 9.9 9.3 Load Reduction Efficiency 38 6 (Percent):

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 µg/L was 42 percent and load reduction dissolved copper with influent concentrations greater than 50 µg/L was 50 percent. There were no performance claims reported for total lead or total cadmium.

Table 5-8. Metals Sum of Loads Results

Event Total Copper (g) Dissolved Copper (g) Total Zinc (g) Dissolved Zinc (g) Total Lead (g) No. Inlet1 Outlet Inlet Outlet Inlet1 Outlet Inlet Outlet Inlet Outlet 1 4.9 3.4 - - 27 17 0.73 0.41 - - 2 16 8.6 0.37 0.32 92 35 2.17 1.9 - - 3 6.9 4.6 0.24 0.21 81 18 1.1 0.79 - - 4 9.6 3.3 0.21 0.19 66 19 1.3 1.2 - - 5 13 3.0 0.18 0.12 68 11 0.76 0.45 - - 6 3.3 1.5 0.09 0.08 16 5.8 0.46 0.28 - - 7 11 6.7 0.12 0.18 34 26 0.52 0.97 - - 8 26 13 0.36 0.35 130 51 1.4 1.1 1.1 0.63 9 6.8 3.1 0.23 0.18 36 14 1.4 0.96 2.5 0.20 10 7.2 3.8 0.22 0.19 22 13 0.85 0.89 - - 11 4.6 2.9 0.26 0.11 15 11 0.54 0.47 0.67 0.41 12 6.2 5.7 0.27 0.31 26 21 1.1 1.3 0.49 0.23 14 2.6 2.1 0.10 0.12 17 5.8 0.45 0.33 - - 15 14 2.9 0.30 0.21 57 8.9 1.8 0.82 - - 16 3.1 2.0 0.06 0.07 18 10 0.29 0.33 1.4 0.19 17 33 1.4 0.06 0.07 49 5.2 0.27 0.31 1.5 1.0 18 2.8 0.9 0.30 0.11 18 4.0 0.27 0.25 0.72 0.34 Total: 171 69 3.4 2.8 771 274 15 12 8.5 3.0 Load Reduction Efficiency 59 16 64 17 64 (Percent):

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.

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

COD Dissolved Chloride Total Calcium Total Magnesium Event (lb) (lb) (lb) (lb) No. Inlet Outlet Inlet Outlet Inlet Outlet Inlet Outlet 1 1.1 1.0 0.15 0.14 1.1 0.39 0.56 0.15 2 3.9 2.5 0.46 0.46 2.8 0.61 1.4 0.19 3 1.8 2.5 0.46 0.35 0.99 0.45 0.43 0.16 4 2.1 1.8 0.24 0.24 4.0 0.32 1.9 0.10 5 1.8 1.7 NA NA 0.38 0.22 0.16 0.07 6 1.3 0.8 0.17 0.18 0.43 0.18 0.18 0.05 7 1.3 2.5 5.93 49 2.5 0.90 1.1 0.16 8 6.7 4.0 9.9 14 9.2 1.4 3.7 0.55 9 1.8 1.3 0.86 1.1 2.1 0.36 0.94 0.10 10 1.3 1.2 0.29 0.65 0.8 0.33 0.36 0.10 11 1.2 1.0 0.27 0.49 1.1 0.27 0.51 0.08 12 2.4 2.3 0.48 1.00 0.84 0.50 0.32 0.12 14 1.0 0.9 0.32 0.36 0.20 0.15 0.08 0.04 15 2.8 1.5 0.32 0.32 2.2 0.15 1.1 0.04 16 0.8 0.7 0.13 0.21 0.86 0.19 0.42 0.08 17 0.8 1.2 0.15 0.27 1.2 0.14 0.61 0.04 18 0.7 0.7 0.07 NA 0.81 0.10 0.30 0.03 Total: 33 28 20 69 31.5 6.70 14.1 2.1 Load Reduction Efficiency 16 -240 79 85 (Percent): Dissolved Chloride Total and 4.4 5.7 Reduction Efficiency (omitting events 7 and 8) -31

NA: not analyzed

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 µm (defined as sand) and less than 62 µm (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 µm. 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 µm. 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 µm) 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.

Table 5-10. Particle Size Distribution Analysis Results

Percent Less Than Particle Size (µm) Event No, Location <1000 <500<250 <125 <62.5 <31 <16 <8 <4 <2 1 Inlet 80 64 36 22 18 Outlet 100 100 98 93 91 2 Inlet 52 45 25 12 12 Outlet 100 100 100 96 88 3 Inlet 100 73 42 32 32 Outlet 82 4 Inlet 71 52 17 9 8 Outlet 92 6 Inlet 93 93 58 39 32 Outlet 91 7 Inlet 90 61 47 42 40 38 33 25 16 10 Outlet 100 97 96 86 78 66 8 Inlet 90 77 49 34 30 26 20 14 11 8 Outlet 100 96 86 66 55 48 9 Inlet 92 81 34 19 15 Outlet 100 81 57 50 44 15 Inlet 90 75 23 4 4 Outlet1 16 Inlet 72 44 23 15 13 Outlet 92

1 No data reported due to laboratory error.

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

Blank 1 Blank 2 Blank 3

Parameter Units (4/2/2002) (11/11/2002) (6/30/2003) Inlet Outlet Inlet Outlet Inlet Outlet LOD LOQ TSS mg/L <2 <2 -- -- <2 <2 2 7 SSC mg/L ------<2 <2 2 7 TDS mg/L <50 <50 <50 <50 <50 <50 50 167 COD mg/L <9 <9 <9 <9 12 14 9 28 Dissolved copper µg/L <5 <5 <1 <1 1.7 2.3 1 3 Total copper µg/L <5 <5 <1 <1 2 2 1 3 Dissolved zinc µg/L <16 <16 <16 <16 <16 <16 16 50 Total zinc µg/L <16 <16 <16 <16 <16 <16 16 50 Dissolved phosphorus mg/L -- -- <0.005<0.005 <0.005 <0.005 0.005 0.016 Total phosphorus mg/L <0.005 <0.005 0.025 <0.005 <0.005 <0.005 0.005 0.016 Dissolved chloride mg/L 3.3 <0.6 <0.6 <0.6 0.8 <0.6 2 3.3 Total calcium mg/L 0.7 <0.2 <0.2 <0.2 0.2 <0.2 0.2 0.7 Total magnesium mg/L <0.2 <0.2 <0.2 <0.2 <0.2 <0.2 0.2 0.7

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:

 xx12−  %RPD =   × 100% (6-1)  x 

where: x1 = Concentration of compound in sample

x2 = Concentration of compound in duplicate xxx= Mean value of 12 and

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.

Table 6-2. Field Duplicate Sample Relative Percent Difference Data Summary

9/19/2002 4/19/2003 6/27/2003 9/12/2003 10/14/2003 Rep Rep RPD Rep Rep RPD Rep Rep RPD Rep Rep RPD Rep Rep RPD Parameter Unit 1a 1b (Pct) 2a 2b (Pct) 3a 3b (Pct) 4a 4b (Pct) 5a 5b (Pct) TSS mg/L Inlet - - - 780 840 7 77 96 22 700 820 16 35 44 23 Outlet - - - 380 380 0 46 47 2 36 31 15 20 25 22 SSC mg/L Inlet 500 680 30 5,600 4,900 14 370 210 54 3,800 2,400 44 410 310 29 Outlet 39 39 0 370 370 0 47 48 2 29 32 10 21 22 5 TDS mg/L Inlet <50 52 NA 520 520 0 90 86 5 210 220 6 50 <50 0 Outlet <50 <50 0 720 730 1 162 160 1 190 190 0 74 58 24 Dissolved µg/L Inlet 8.9 9.5 7 28 28 0 20 21 6 58 59 2 50 170 108 copper Outlet 6.8 8.4 21 27 26 5 23 23 0 42 41 2 18 19 6 Total µg/L Inlet 140 35 120 280 370 29 48 52 8 330 260 25 46 130 97 copper Outlet 17 18 6 140 140 0 44 46 4 69 68 1 15 15 0 Dissolved µg/L Inlet 35 31 12 110 120 6 81 77 5 360 350 1 46 47 2 zinc Outlet 22 22 0 84 91 8 96 92 4 160 150 3 42 43 2 Total µg/L Inlet 134 328 84 1,400 2,200 46 200 320 48 1,400 1,700 21 300 280 5 zinc Outlet 61 63 3 540 540 0 160 160 0 220 210 3 66 67 2 Dissolved mg/L Inlet 0.03 0.031 3 0.027 0.025 8 0.061 0.063 3 0.20 0.21 3 0.040 0.039 3 phosphorus Outlet 0.027 0.026 4 0.017 0.016 6 0.059 0.058 2 0.19 0.19 0 0.046 0.046 0 Total mg/L Inlet 0.16 0.11 37 0.50 0.56 10 0.235 0.32 31 0.63 0.58 7 0.15 0.11 35 phosphorus Outlet 0.067 0.065 3 0.29 0.30 3 0.19 0.19 0 0.30 0.29 4 0.098 0.098 0 Total mg/L Inlet 16 20 23 430 480 9 29 32 9 230 220 7 60 62 4 calcium Outlet 6.1 6.2 2 68 68 0 17 18 2 16 16 0 7.0 7.1 1 Total mg/L Inlet 7.8 10 26 170 200 14 11 12 3 120 110 9 22 27 20 magnesium Outlet 2.5 2.5 0 26 26 0 4.2 4.2 0 4.4 4.2 5 1.9 2.0 5 Single dash indicates no sample processed for event

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 19 1.7 4.6 0.19 1.2 25 Dissolved chloride 21 0.69 2.4 0.03 0.60 25 Dissolved copper 12 2.1 8.7 0.03 2.9 25 Total copper 21 1.8 4.6 0.09 1.5 25 Total magnesium 19 1.2 3.6 0.01 1.2 25 TSS 16 1.3 3.5 0 1.1 30 Dissolved phosphorus 18 1.3 1.6 0 0.51 30 TDS 18 3.3 12 0 3.3 30 Total phosphorus 20 1.4 6.4 0 1.6 30 Dissolved zinc 17 1.5 5.6 0.09 1.4 25 Total zinc 18 1.7 3.8 0 1.2 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.

Table 6-4. Laboratory MS/MSD Data Summary

Average Maximum Minimum Std. Dev. Range Parameter Count (percent) (percent) (percent) (percent) (Pct) Total calcium 22 96.5 113 90.8 5.1 85 – 115 COD 20 97.9 119 79.4 10.3 75 – 125 Dissolved chloride 21 101 108 97.3 2.4 90 – 110 Total copper 22 101 116 91.3 7.7 80 – 120 Dissolved copper 14 98.5 113 90.8 6.1 85 – 115 Total magnesium 22 97.5 102 93.0 2.5 85 – 115 Dissolved phosphorus 19 102 106 96.9 2.3 90 – 110 Total phosphorus 19 102 109 97.3 3.2 90 – 110 Total zinc 22 94.9 101 91.0 2.6 85 – 115 Dissolved zinc 19 97.9 114 91.8 5.0 85 – 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

Mean Maximum Minimum Std. Dev. Parameter Count (percent) (percent) (percent) (percent) Total calcium 18 97 105 93 2.8 COD 20 101 107 923 3.4 Dissolved chloride 48 100 110 92 2.8 Total copper 21 99 106 91 4.5 Dissolved copper 36 102 110 94 3.5 Total magnesium 18 98 103 94 1.9 SSC 13 99 108 87 6.2 TSS 12 99 120 86 9.9 Dissolved phosphorus 6 101 102 100 0.5 TDS 18 106 122 94 7.1 Total phosphorus 24 101 108 96 2.3 Total zinc 19 97 103 94 2.1 Dissolved zinc 9 99 102 97 1.8

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.

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.

Riverwalk South Inlet Calibration 04-20-03

0.8 0.7 0.6 y = 0.7789x 0.5 R2 = 0.9968 0.4 0.3

Flume Flow (cfs) 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Calculated Flow (cfs)

(a) April 20, 2003

Riverwalk South Inlet Calibration 11-08-03

0.8 0.7 y = 0.9002x 0.6 R2 = 0.9722 0.5 0.4 0.3 0.2 Flume Flow (cfs) 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Calculated Flow (cfs)

(b) November 8, 2003 Figure 6-1. Calibration curves used to correct flow measurements.

Table 6-6. Comparison of Inlet and Outlet Event Runoff Volumes

Event Volumes1 Event Inlet Outlet Difference No. (ft3) (ft3) (percent) 1 290 420 -45 2 1,700 1,600 6 3 1,600 1,600 0 4 1,000 1,200 -20 5 390 350 10 6 730 730 0 7 270 300 -11 8 400 340 15 9 610 540 11 10 340 320 6 11 500 450 10 12 420 460 -10 13 530 550 -4 14 290 260 10 15 160 150 6 16 350 340 3 17 220 270 -23 18 210 220 -5 19 410 410 0 20 680 560 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.

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 Gauge Height Point 2 Gauge Height Point 3 Date Gauge Correction Gauge Correction Gauge Correction Height (ft) (unitless) Height (ft) (unitless) Height (ft) (unitless) 4/01/02 0.0 0.0 0.318 -0.035 0.636 -0.036 4/11/03 0.0 0.0 0.318 -0.035 0.635 -0.036 4/11/03 0.0 0.002 0.350 0.002 0.635 0.002 8/14/03 0.0 0.015 0.250 0.025 0.500 0.033 8/14/03 0.0 -0.005 0.350 -0.005 0.635 -0.005 11/8/03 0.0 -0.005 0.350 -0.005 0.635 -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.

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.

Velocity Dropout Velocity Flow

Stage

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.

Table 6-8. Missing Sample Aliquots Due to Missing Inlet Velocity Data

Event Number of Total Aliquots Collected Missing Aliquots No. Missing Aliquots and Missing for Storm (Percent) 2 4 33 12 3 3 33 9 4 4 25 16 10 1 14 7 17 1 9 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

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 StormFilter unit was cleaned of accumulated Earth Tech, USGS; (Major maintenance) sediment and filter cartridges were replaced. WDNR; SMI; total of 3 staff days. November 7, 2002 StormFilter visual inspection by WisDOT. WisDOT: 2 staff (Minor maintenance) Reported observing the following: 1) 0.20 ft of hours 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. April 24, 2003 USGS assessed need for major maintenance. 4 staff hours. (Minor 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/l; Outlet: 31 mg/l). 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.

Table 7-1 (cont’d).

Date Activity Personnel Time/Cost January 27, 2004 Post-monitoring clean out. The procedure is Staff time: 40 hours (Major maintenance) summarized in Section 7.1.1. Lab costs (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 SMI’s Portland, Oregon facility for cleaning and reuse.

Chapter 8 References

1. APHA, AWWA, and WEF. Standard Methods for the Examination of Water and Wastewater, 19th ed. Washington, DC, 1995.

2. Horowitz; A.J; 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.

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.

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.

Appendices

A Verification Test Plan B Event Hydrographs and Rain Distribution C Analytical Data Reports