NJCAT TECHNOLOGY VERIFICATION. Up-Flo Filter Hydro International

Transcription

1 NJCAT TECHNOLOGY VERIFICATION Up-Flo Filter Hydro International November, 008

2 TABLE OF CONTENTS List of Figures iii List of Tables iv. Introduction. NJCAT Program. Technology Verification Report. Technology Description.. Technology Status.. Specific Applicability 9.. Range of Contaminant Characteristics 0.. Range of Site Characteristics 0.. Material Overview, Handling and Safety. Project Description. Key Contacts. Evaluation of the Applicant. Corporate History. Organization and Management. Operating Experience with Proposed Technology. Patents 9. Technical Resources, Staff and Capital Equipment 0. Treatment System Description. System Description. Filter Media. Technical Performance Claim. Treatment System Performance. Laboratory Studies System Description Test Program 8 i

3 Procedure 8. Verification Procedures 9 One-Module Testing 9 Six-Module Testing. Inspection and Maintenance.. First-Year Inspection and Maintenance.. Routine Inspection and Maintenance.. Maintenance Procedures.. Solids Disposal 7.. Damage Due to Lack of Maintenance 7. Technical Evaluation Analysis 9. Verification of Performance Claim 9. Limitations 9.. Factors Causing Under-Performance 9.. Pollutant Transformation and Release 9.. Sensitivity to Heavy Loading 0.. Bypass Flow 0.. Mosquitoes 7. Net Environmental Benefit 8. References Appendix A - Analytical Results for One-Module System Appendix B - Analytical Results for Six-Module System 7 ii

4 List of Figures Figure - Up-Flo Filter Components 7 Figure - Filter Module Components 8 Figure - Outlet Module and Filtered Draindown 9 Figure - Organizational Chart for Hydro International Figure - Results from SBIR Field Testing Controlled Tests 7 Figure - Results from SBIR Field Testing Actual Rainfall Events 8 Figure - Available Up-Flo Filter Systems Figure - Potential Energy or Driving Head (0-inches from top of the media) Figure - Particle Size Distribution of Sil-Co-Sil 0 Figure - Up-Flo Filter System as Tested in the Laboratory Figure - Schematic of Test Setup 7 Figure - Removal Efficiency vs. Flow rate for One-Module System 0 Figure. Removal Efficiency vs. Flow rate for Six-Module System Figure -: Up-Flo Filter Maintenance Photographs (New Zealand) 8 iii

5 List of Tables Table. Sil-Co-Sil 0 removal for one-module system 9 Table. Averages of results from one-module system tests 0 Table. Up-Flo Filter: One-module system removal efficiency per NJDEP weighting Table. Sil-Co-Sil 0 removal for six-module system Table. Averages of results from six-module system tests Table. Up- Flo Filter : Six-module system removal efficiency per NJDEP weighting iv

6 . Introduction. New Jersey Corporation for Advanced Technology (NJCAT) Program NJCAT is a not-for-profit corporation to promote in New Jersey the retention and growth of technology-based businesses in emerging fields such as environmental and energy technologies. NJCAT provides innovators with the regulatory, commercial, technological and financial assistance required to bring their ideas to market successfully. Specifically, NJCAT functions to: Advance policy strategies and regulatory mechanisms to promote technology commercialization; Identify, evaluate, and recommend specific technologies for which the regulatory and commercialization process should be facilitated; Facilitate funding and commercial relationships/alliances to bring new technologies to market and new business to the state; and Assist in the identification of markets and applications for commercialized technologies. The technology verification program specifically encourages collaboration between vendors and users of technology. Through this program, teams of academic and business professionals are formed to implement a comprehensive evaluation of vendor specific performance claims. Thus, suppliers have the competitive edge of an independent third party confirmation of claims. Pursuant to N.J.S.A. :D- et seq. (Energy and Environmental Technology Verification Program) the New Jersey Department of Environmental Protection (NJDEP) and NJCAT have established a Performance Partnership Agreement (PPA) whereby NJCAT performs the technology verification review and NJDEP certifies the net beneficial environmental effect of the technology. In addition, NJDEP/NJCAT work in conjunction to develop expedited or more efficient timeframes for review and decision-making of permits or approvals associated with the verified/certified technology. The PPA also requires that: The NJDEP shall enter into reciprocal environmental technology agreements concerning the evaluation and verification protocols with the United States Environmental Protection Agency, other local required or national environmental agencies, entities or groups in other states and New Jersey for the purpose of encouraging and permitting the reciprocal acceptance of technology data and information concerning the evaluation and verification of energy and environmental technologies; and The NJDEP shall work closely with the State Treasurer to include in State bid specifications, as deemed appropriate by the State Treasurer, any technology verified under the Energy and Environment Technology Verification Program.

7 . Technology Verification Report On April, 007, Hydro International (9 Hutchins Drive, Portland, ME 00) submitted a formal request for participation in the NJCAT Technology Verification Program. The technology proposed The Up-Flo Filter is an upward flow stormwater filtration system that includes pretreatment prior to filtration in one device. It is an effective flow-through treatment system for removal of gross pollutants (floating debris, free oils and settleable solids) during the pretreatment stages. Fine suspended solids, particulate matter and pollutants (phosphorus and heavy metals) associated with the fine sediment fraction are removed as flow passes through the filtration and adsorption media. The request (after pre-screening by NJCAT staff personnel in accordance with the technology assessment guidelines) was accepted into the verification program. This verification report covers the evaluation based upon the performance claim of the vendor, Hydro International (see Section ). The verification report differs from typical NJCAT verification reports in that final verification of the Up-Flo Filter (and subsequent NJDEP certification of the technology) awaits completed field testing that meets the full requirements of the Technology Acceptance and Reciprocity Partnership (TARP) Stormwater Best Management Practice Tier II Protocol for Interstate Reciprocity for stormwater treatment technology. This verification report is intended to evaluate the Up-Flo Filter performance claim for the technology based primarily on carefully conducted laboratory studies. The performance claim is expected to be modified and expanded following completion of the TARP required field-testing. Future addendums to this verification report are also expected as different filtration media are evaluated and data submitted to NJCAT.. Technology Description.. Technology Status: general description including elements of innovation/uniqueness/ competitive advantage. In 990, Congress established deadlines and priorities for EPA to require permits for discharges of stormwater that is not mixed or contaminated with household or industrial wastewater. Phase I regulations established that a NPDES (National Pollutant Discharge Elimination System) permit is required for stormwater discharge from municipalities with a separate storm sewer system that serves a population greater than 00,000 and certain defined industrial activities. To receive a NPDES permit, the municipality or specific industry has to develop a stormwater management plan and identify Best Management Practices for stormwater treatment and discharge. Best Management Practices (BMPs) are measures, systems, processes or controls that reduce pollutants at the source to prevent the pollution of stormwater runoff discharge from the site. Phase II stormwater discharges include discharges from classes of smaller municipalities than those specifically classified as Phase I discharge. Impervious structures like rooftops, driveways, sidewalks, and roads not only reduce infiltration of rainfall and runoff but also degrade runoff quality. The most important hydraulic factors that affect urban runoff volume are directly connected impervious surfaces such as paved streets, driveways, parking areas draining to curb-and-gutter drainage systems, and roofs draining

8 directly to a storm or combined sewer. The pollutant characteristics and loads emanating from urban runoff are very diverse which requires a treatment train approach when incorporating BMPs into a Stormwater Management Plan. Hydro International has developed a modular upflow filtration system that includes pretreatment to mitigate pollution from stormwater generated from impervious surfaces. Upflow filters are filtration devices that utilize an upward flow path through filtration and adsorption media to separate suspended particulate matter and associated pollutants out of a liquid. Flow in an upward direction counters gravitational forces to fluidize the media, allowing the entire depth of the media bed to be utilized. Fluidizing the media bed, combined with pretreatment, increases hydraulic loading rates and filter life compared to traditional downflow filters where filtration depth is typically limited to the first - inches. General The Up-Flo Filter is a passive subsurface filtration system that can be retrofitted into an existing stormdrain manhole or supplied as a complete system housed in a -ft diameter manhole or precast vault. The vaulted systems are designed to house multiple platforms each having one to six Filter Modules. It is designed with a treatment train concept that incorporates gravitational separation of floating and settling materials, screening, and filtration of stormwater flows. A filtered Draindown prevents media from remaining saturated and becoming anaerobic. Inspection and maintenance is necessary to maintain the design filtration rate as discussed in Section.. A siphon-activated bypass conveys flows larger than the design filtration rate for on-line installations. The major components of a typical -module configuration in a -ft manhole are illustrated in Figure -. Inlet Grate Concrete Bypass Siphon / Floatables Baffle Filter Module Media Pack (see Figure -) Angled Screen Bypass Weir Outlet Module and Draindown (see Figure -) Conveyance Slot Outlet Pipe Sump Figure -: Up-Flo Filter Components 7

9 Operation of the Up-Flo Filter is initiated during a rainfall event when stormwater is conveyed into the chamber from a surface inlet or directly from the drainage system s pipe network. As flow enters the chamber, internal components act as baffles to force gross debris and sediment to settle into the sump and floating debris to rise to the surface. Depending on the runoff rate entering the chamber, a water column builds above the top of the media until it reaches the Bypass Weir elevation (Refer to Figure -). This water column provides the driving head to force flow upward through an Angled Screen and Media Pack into a Conveyance Slot where filtered flow is discharged into the Outlet Module. The Media Pack includes a bottom layer of flow Distribution Media, two filter bags with media, and a top layer of flow Distribution Media (Figure -). The Distribution Media is a polyethylene fiber web filtration media used to support the media bags and evenly disperse the flow across the entire surface of the media. The Angled Screens are designed to minimize the chance of ragging and blinding as they are situated below the filter modules, sheltering them from the direct path of the influent. The angle is designed to release any pollutants that may temporarily lodge on the screens once flow subsides and draindown initiates. Conveyance Slot (to Outlet Module) Lid with Integral Media Restraint Filter Media Flow Distribution Media Angled Screen Upward Flow Direction (to Outlet Module) Figure -: Filter Module Components The driving head or water column above the top of the Media Bags imposes an upward pressure on the media. This causes the individual particles in the filter media to shift from a state where they are compressed together by gravitational forces to a state where they achieve very minute separation from each other. As they become suspended in an upward-flowing column of water, the media matrix becomes fluidized. This allows fines to be trapped throughout the entire depth of the media bed, rather than just the first few inches, thus increasing the length of filter runs. High hydraulic loading rates are maintained without compromising effluent control by 8

10 restraining the media within the Media Packs. Treated flow exits the Filter Module(s) into the Outlet Module via a conveyance channel located above the media. Flow in excess of the design filtration capacity discharges over the Bypass Weir. The Outlet Module has a hood to act as a Floatables Baffle preventing the escape of buoyant debris and trash. It also siphons excess flows into the outlet once the air in the outlet chute is displaced, increasing the maximum discharge rate for extreme events. After a storm event, the water column drops to the top of the Media Bags at which point there is no longer any head to drive flow. The Up-Flo Filter employs a patented siphon-activated Draindown (Figure -) that allows the water level in the chamber to drop below the filter media between events preventing the media from becoming anaerobic. The siphon draws the remaining water through the Draindown (which incorporates a screen and filtration media) prior to discharging into the outlet. During the drain-down mode of operation, a light backwashing effect occurs that washes captured pollutants off the surface of the filter bag, helping to prevent blinding and prolonging media life. By draining the water out of the media, the weight of the filter bags is reduced for easier removal during maintenance operations. Outlet Module Filter Modules Conveyance Slots Filtered Draindown Figure -: Outlet Module and Filtered Draindown.. Specific Applicability In general, the Up-Flo Filter is an effective filtration system, particularly suited to urban environments such as those with high percentages of impervious surfaces and thereby limited to subsurface treatment options. Different media are available that can target very fine sediment fractions at low concentrations. Therefore, the system is effective for targeting pollutants such as 9

11 heavy metals, phosphorus, polycyclic aromatic hydrocarbons (PAHs), and toxic organic compounds that are associated with the very fine sediment fraction. Because of its modular construction and ease with which its media can be easily exchanged, specific dissolved pollutants of concern can be targeted with customized media mixes as they become available... Range of Contaminant Characteristics The Up-Flo Filter will capture and retain gross pollutants like trash, debris and free oils during pretreatment and is capable of filtering very fine suspended solids and particulate matter for a wide range of sediment concentrations and hydraulic loading rates. Laboratory and field research has demonstrated that the Up-Flo Filter removes over 9% of the total suspended solids above 0 µm in particle size and about 0% of particles in the 0. to 0 µm range (Khambhammettu, et al., 00). Research by Pitt et al., (999) and Morquecho et al., (00), have shown strong associations between small particle sizes and other pollutants such as heavy metals and phosphorus (TP). Field testing of the Up-Flo Filter substantiated this earlier research. Field tests showed TP removal by filtration ranged from % to 78% for the TP mass associated with particle sizes ranging from 0.- micron to 0 microns. Similarly, heavy metals (iron, chromium, copper and lead) removal by filtration ranged from 0% to 00% (non-detect)... Range of Site Characteristics In general, stormwater filtering systems are best applied to impervious source areas where pollutant loads in stormwater are expected to be highest. Performance data from field testing of slow-rate filters ( gpm/ft ) in Virginia, Texas, Washington, Wisconsin, Oregon, and California show that TSS removal efficiency significantly improves when the influent concentration rises above approximately 0 mg/l (Minton, 00). Additionally, impervious drainage areas associated with critical source areas are likely to contain pollutant loadings of hydrocarbons, toxic trace metals, nutrients, pathogens, and/or other toxicants and pollutants that are greater than the loadings of normal runoff (Bannerman et al., 99; Pitt et al., 99; Claytor and Schueler, 99). The Up-Flo Filter is intended for treating the runoff from source areas having higher pollutant loads before it mixes with runoff from less polluting areas. Stormwater runoff from specific problem sources or hotspots include; auto recyclers/junkyards, commercial garden nurseries, parking lots, vehicle fueling and maintenance stations, bus or truck (fleet) storage areas, industrial rooftops, marinas, outdoor transfer facilities, public works storage areas, and vehicle and equipment-washing/steam-cleaning facilities. As pretreatment and filtration are accomplished in one device and the modular components are designed for retrofit applications, the Up-Flo Filter is designed to be installed throughout a catchment area to target potential hot spots. 0

12 Treating runoff near source areas is also an important element of Low Impact Development (LID). The success of LID depends on the use of numerous small BMPs designed to replicate the natural hydrology of a site. Breaking up the drainage in this way also results in much greater overall control of the runoff during smaller storms and for the first flush of each storm when the most pollution transport typically occurs. Since the Up-Flo Filter becomes an integral component of a site s drainage network, standard drainage engineering evaluation of its impact to the system must be considered. Land use, site hydrology, the storm water management plan for the site, and local regulatory requirements all play a role in the number of modules and thus chamber design. Hydro International works with consultants, contractors and agency reviewers to ensure that design and installation requirements are met for each installation... Material Overview, Handling and Safety Site preparation, filter unit delivery, manhole or vault construction, and filter module installation are all general construction practice. There is no handling of hazardous material. Field personnel should take precautions while handling and installing the Up-Flo Filter. Field personnel should use appropriate safety equipment, including hardhat and steel-toe boots. Personnel who operate field equipment during the installation process should have appropriate training, supervision, and experience. The Up-Flo Filter is considered a confined space such that confined space training is needed to enter the manholes and vaults. Entry also requires the use of a gas detector for safety. Standard OSHA confined space entry procedures should be followed (9 CFR 90.). Only persons who are certified by OSHA to make confined space entries should enter an Up-Flo Filter System. Safe and legal disposal of pollutants is the responsibility of the maintenance contractor. Solids recovered from the Up-Flo Filter can typically be land-filled. It is possible that there may be some specific land use activities that create hazardous solids, which will be captured in the system. Such material would have to be handled and disposed of in accordance with hazardous waste management requirements.. Project Description This project included the evaluation of assembled reports, academic research, company manuals and technical product literature, as well as independent field and laboratory reports to verify that the Up-Flo Filter meets the performance claim stated in Section.

13 . Key Contacts Rhea Weinberg Brekke Executive Director New Jersey Corporation for Advanced Technology c/o New Jersey Eco Complex 00 Florence Columbus Road Bordentown, NJ ext. 7 rwbrekke@njcat.org Richard S. Magee, Sc.D., P.E., BCEE Technical Director New Jersey Corporation for Advanced Technology Vultee Drive Florham Park, NJ rsmagee@rcn.com Hsin-Neng Hsieh, Ph.D., P.E. Professor New Jersey Institute of Technology Civil & Environmental Engineering Department University Heights Newark, NJ hiseh@adm.njit.edu John MacKinnon Regulatory Specialist 9 Hutchins Drive Portland ME 00 Phone: (07) 7-00 ext. 0 Fax: (07) 7- jmackinnon@hil-tech.com Sandy Blick Program Coordinator Stormwater Management Rule Implementation Division of Watershed Management NJ Department of Environmental Protection 0 East State Street Trenton, NJ Sandra.Blick@dep.state.nj.us David Scott Stormwater Product Manager 9 Hutchins Drive Portland ME 00 Phone: (07) 7-00 ext. Fax: (07) 7- dscott@hil-tech.com. Evaluation of the Applicant. Corporate History Hydro International, founded in 980, evolved from pioneering work undertaken by Bernard Smisson dating back to the early 90s when Mr. Smisson built and tested the very first fullscale vortex type combined sewer overflow (CSO) unit at Bristol in the U.K. This first generation separator was found to be effective in retaining 70% of the pollution load (Smisson, 97).

14 Smisson s pioneering work was followed by the development in the 970s of the USEPA Swirl Concentrator - a second generation of Hydrodynamic Vortex Separators (HDVS), by the American Water Works Association and EPA, with Mr. Smisson acting as a consultant (Sullivan et al., 97, 98). A third generation of HDVS was subsequently developed in the UK in the early 980s, with Bernard Smisson s assistance, to overcome identified shortcomings with the EPA Swirl Concentrator, particularly to reduce shoaling of solids on the base, to reduce head loss at high flows and to further improve performance. This configuration was subsequently patented and commercialized with the trade name Storm King Overflow. Hydro International was initially formed to promote the hydrodynamic vortex separator and vortex flow control technology around the world. Its main business at present is the design and supply of products for the control and treatment of stormwater and wastewater and currently supplies 0 different types of product for stormwater and wastewater management. These vary in size from small flow control devices to large-scale treatment systems, incorporating devices weighing up to tonnes, to control stream flow discharges from entire towns. Hydro uses thirdparty engineering and fabrication companies, based mainly in Europe and North America, to manufacture its products which it sells predominantly in Europe and North America. Hydro also sells some of its smaller and more transportable products into parts of Asia. Originally, all its products were based around Hydro International s patented vortex technology, which required only a flow of water to operate. The company has since developed or licensed other products some of which require an external power source. It also supplies geo-plastic water storage, filtration and drainage devices some in the form of containers and others which can extend across hundreds of square feet just below the surface of a construction or landscaped site. Hydro International was floated on the London Stock Exchange in 99 and moved to the AIM market in 00 (seeing this as a more appropriate market for a company of its size). Hydro currently has 0 employees in five offices in the UK, Ireland and the US. It also has a small office in Poland. The company has a four () part growth strategy comprised of the following: Organic growth through Core Products in Core Markets; Geographic expansion into new territories; Development of new products to address identified markets; and Acquisition of Intellectual Property or companies for products that compliment the Group s existing portfolio. Hydro was founded on innovation, with the company s first products having been developed by the proprietors. Recognizing the significance of R&D/product development, the company has remained committed to this activity and currently has comprehensive R&D facilities in both the UK and US. Through this, the product portfolio has expanded significantly, with the company s products covered by more than 70 patents. In summer 00, Hydro International was looking to expand its stormwater product portfolio to include a filtration device. Key US market sectors, such as Washington State and the Chesapeake Bay area had re-written their stormwater treatment regulations such that filtration devices were required to attain the permitted quality of treated stormwater effluent. Prior to 00, Dr. Robert Pitt, a professor at the University of Alabama, was awarded a Small Business Innovation Research (SBIR) Grant from the US EPA to investigate a design for an

15 upflow stormwater filter in 999 (Pitt et al., 999). Soon after, under the small company name of Stormtrain L.L.C., Dr. Pitt patented the design developed under the SBIR program as the Upflow Stormwater Filter, an upflow filter that includes a sump, screen, drain-down mechanism, high flow bypass, floatables trap and secondary media chamber. Hydro International purchased the Intellectual Property rights to the Upflow Stormwater Filter from Stormtrain L.L.C. in December 00. Hydro International then redesigned the Upflow Stormwater Filter to maximize its commercial viability, while retaining the patented features and other beneficial treatment aspects, and delivered the new filtration device to market in 00 under the product name Up-Flo Filter.. Organization and Management The Hydro International plc board consists of executive directors and two non-executive directors. One of the directors serves as the chairman of the board. The Group has a number of operating subsidiary companies (mainly in Europe and the US) together with an export sales function responsible for export sales and supporting a number of licensees in different countries (see Figure -). Figure -: Organizational Chart for Hydro International Decisions with regards to policy, direction and strategy are taken at the plc board level with the operating subsidiaries being responsible for running their respective business units to agreed operational budgets and targets. The company is ISO accredited and has a Group-wide Innovation Team responsible for functions such as R&D, IT and Products. Outside its core markets, Hydro has established relationships with distributors in Australia, Egypt, Japan, South Korea, Malaysia, New Zealand and Singapore. Supplying these more remote areas has required Hydro to forge new links with foreign engineering companies so that some of its less transportable products can be sourced locally.

16 . Operating Experience with the Proposed Technology The effectiveness of upflow filtration technology for high-rate treatment of stormwater has been proven through comprehensive full-scale testing under controlled conditions in the laboratory (Andoh et al., 007 and Glennon et al., 00) and under a range of controlled and actual storm conditions in the field at a site in Tuscaloosa, Alabama by researchers at the University of Alabama (Pitt and Khambhammettu, 00). The results of testing and evaluations leading ultimately to the development and commercialization of the Up-Flo Filter are fully documented as part of the SBIR (Small Business Innovative Research) and SBIR research funded by the US EPA (US Infrastructure 00; Pitt et al., 00). The SBIR research was a three-phase program including bench-scale testing, pilot-scale testing, and a 0-month field evaluation of a full-scale single-module prototype. After the three phases of research and development were completed in 00, Hydro International became involved and spent two years commercializing and evaluating the initial prototype system. A six-module system that kept all the functionality of the prototype was developed so that it could be installed into a standard -ft diameter manhole or onto a -ft square platform for vaulted configurations. Since 00, more than 0 commercialized systems have been installed, particularly in New Zealand with the Up-Flo Filter being shown to be most effective in reducing the pollutants that are highly associated with particulate matter. There is also a six-module Up-Flo Filter at the Stormwater Center at the University of New Hampshire and a single-module system at Penn State Harrisburg Environmental Engineering Laboratory in Middletown, PA for independent performance evaluation. Following more than a year of field evaluations and completion of the ETV program at Penn State, the commercialized system s components were further refined and evaluated in the laboratory at Hydro International for NJDEP interim certification. Verification studies for the interim certification included tests conducted on the commercially available fullscale -ft diameter Up-Flo filtration system. The tested system allowed evaluation of one to six filter modules. The SBIR project was designed to develop and demonstrate the effectiveness of upflow filtration for the treatment of stormwater runoff. This research was conducted in three phases: bench-scale tests to evaluate the potential of upflow filters to remove solids and dissolved pollutants, pilot-scale tests to evaluate pollutant removal using real stormwater runoff and flowrate testing of a prototype device. Test media in the laboratory and the field included two types of sand, a peat moss-sand mixture and a compost-sand mixture (US Infrastructure, 00). The bench scale testing for solids removal demonstrated that upflow filters operated in the flow rate range of rapid sand filters ( 0 gpm/ft ) and were capable of removing over 80% of the solids found in the influent water. Testing of these filters for dissolved pollutant removal showed that the organic filter media provided the best removals for most dissolved pollutants (metals and organics). The pilot-scale testing was performed using pre-settled stormwater runoff and the results were compared to prior down-flow filter results obtained by the same research team.

17 Pollutant removals were similar to down-flow filters with the benefits of upflow filtration primarily seen in the delay of measurable headloss/pressure buildup in the filters. Substantially longer filter lifetimes were seen with the upflow filters compared to down-flow filters with the same media (US Infrastructure, 00). The SBIR testing was designed to investigate pollutant removal efficiencies of a full-scale gpm prototype. Testing was conducted in two parts, controlled flow tests and testing during actual rain events over a 0-month period. The test site was a retrofitted catch basin located in the parking lot of the Tuscaloosa City Hall, Alabama (Pitt and Khambhammettu, 00; Khambhammettu, 00). The catch basin received runoff from a 0.9-acre drainage area comprised of parking, roofs, and adjacent storage areas. To subject the prototype to a wide range of flow conditions, the prototype was approximately ¼ the optimal size, assuming a gpm design filtration rate. The filtration rate required for treating 90% of the annual flows at the test site was estimated to be 00 gpm while the average runoff flow for the observed rain events was gpm. The controlled testing closely replicated real storm conditions in terms of variable flow, variable TSS influent concentrations, and the wide range of particle sizes that are more characteristic of actual stormwater runoff. During controlled testing, equal weight fractions of Sil-Co-Sil 0, Sil-Co-Sil 0, coarse sand, and fine sand were mixed to prepare a test sediment. The test sediment particle size ranged from 0.-micron to,000 microns. For the CPZ Mix TM filter media used during the tests, the following four influent sediment concentrations were tested: 00 mg/l, 0 mg/l, 00 mg/l, and 0 mg/l. At each of the four concentrations, three separate tests were conducted at high, medium, and low flow rates. The high flow rate was approximately 7 gpm, the medium flow rate was approximately gpm, and the low flow rate was approximately gpm. The influent water source was provided by a nearby fire hydrant. Tests were run by manually feeding the test sediment into the influent water over the whole period of each experiment. Samples were collected using a dipper grab sampler every minute for the 0-minute duration of each test and composited in a churn sample splitter. Using the churn splitter, three samples of,000 ml each were collected for laboratory analysis. The test results from the 00 mg/l influent concentration experiments showed average TSS removals of 8%, 90%, and 9% for the high, medium, and low flow rates, respectively. The test results from the 0 mg/l influent concentration experiments showed average TSS removals of 8%, 88%, and 98% for the high, medium, and low flow rates, respectively. The test results from the 00 mg/l influent concentration experiments showed average TSS removals of 8%, 9%, and 99% for the high, medium, and low flow rates, respectively. The test results from the 0 mg/l influent concentration experiments showed average TSS removals of 97%, 8%, and 99.7% for high, medium, and low flow rates, respectively. The test results show that, with only one exception for medium flow at the 0 mg/l influent concentration, the TSS removals improved with decreasing flow at each influent concentration (see Figure -).

18 Suspended Solids (mg/l) High Flow 00 Mid Flow 00 Low Flow 00 High Flow 0 Mid Flow 0 Low Flow 0 High Flow 00 Mid Flow 00 Low Flow 00 High Flow 0 Mid Flow 0 0 Influent Conc. Effluent Conc. Low Flow 0 Figure -: Results from SBIR Field Testing Controlled Tests During actual rain event testing, an automatic programmable sampler was used to simultaneously collect subsamples from the Up-Flo Filter prototype influent stream and effluent stream. Thirty-one separate rain events occurred during the 0-month monitoring period from February nd to November st, 00. Twenty-four samples were analyzed from ten of those rain events. Figure - shows the paired average influent and effluent suspended solids (TSS) concentrations for the 0 storm events monitored. Seven out of the ten events had influent TSS concentrations below 0 mg/l which is generally low. The highest observed average effluent TSS concentration was 8 mg/l associated with an influent that averaged 77 mg/l for Event 8. The results reported for TSS removal performance were influenced by the inability of the automatic samplers to collect particles in suspended sediment larger than approximately 0 microns. Larger particles were either too large for the sample line intake, or they had a settling velocity approaching the sample velocity in the intake line. Typically, TSS analysis includes measurement of particles less than 00 microns, so the TSS samples collected by the automatic samplers during the rain events likely underrepresented the concentration of TSS in stormwater runoff. The measured effluent TSS concentrations during the actual storm events are generally in line with those observed for the controlled tests with similar patterns and trends observed (i.e. higher influent TSS resulting in higher percentage removals). An assessment of overall average influent and effluent concentrations on the basis of an equal weighting for each event, irrespective of volume of flow treated, results in an overall influent and effluent TSS of 7. mg/l and 9. mg/l respectively. This equates to an average overall TSS removal of 7% for the actual storm events monitored. Mass balance assessments taking account of sediments captured in the sump showed removals of over 80% for solids. 7

19 Paired Average Sample Concentrations (CPZ Mix ) - Actual Storms 00 TSS Concentration (mg/l) Event Event Event Event Event Event Event 7 Event 8 Event 9 Event 0 0 Influent Effluent Figure -: Results from SBIR Field Testing Actual Rainfall Events Following the SBIR testing, Hydro International commercialized and tested the prototype to allow multiple modules that could be connected together and fit into different shaped stormdrain chambers. The initial tests were conducted on a full-scale single module design fit into a -ft square chamber having a -ft sump and were designed to evaluate effluent control and hydraulics for four different media mixes: Filter Sand, Hydro International s CPZ Mix, CPS Mix and Perlite. The test results reported elsewhere (Glennon et al., 00 and Andoh et al., 007) were based on using Sil-Co-Sil 0 (d 0 of microns) as the test sediment (Hydro International, 00). Reported efficiency was greater than 9% at a flow rate of gpm using CPZ Mix, and 9% at a flow rate of gpm using Filter Sand. Additional full-scale lab studies conducted at the Hydro International test facility in Portland, ME tested TSS removal efficiency using Sil-Co-Sil 0 as the test sediment (Hydro International, 007). Sil-Co-Sil 0 has a d 0 = microns and a particle size gradation with 00% of the particles smaller than microns and 7% of particles smaller than microns in diameter. Influent concentrations were maintained in the 0 00 mg/l range. Percent removal of TSS in the influent was tested using CPZ Mix, Filter Sand, CPS Mix, and Perlite for filter media. All of the tests were conducted at an operating head of 0 inches. Tests were run by collecting simultaneous grab samples from the influent and effluent streams; at least nine influent and effluent sample sets were collected during testing on each of the filter media. TSS in the influent and effluent samples was analyzed using the SSC Test Method Filtration in ASTM, 999, D The test results showed average TSS removals of 87% at a flow rate of gpm using CPZ Mix, 9% at a flow rate of gpm using Filter Sand, 88% at a flow rate of 0 gpm using CPS 8

20 Mix TM, and 88% at a flow rate of 8 gpm using Perlite. Dixon s Q tests showed 99.9% confidence that there were no outliers amongst the influent and effluent sample sets. ANOVA analyses showed over 99.9% confidence that the effluent sample set was significantly different from the influent sample set. Although the lab tests were not meant to replicate conditions found in the field, they did demonstrate that the Up-Flo Filter consistently removed more than 80% of TSS from an influent stream containing a TSS gradation similar to that typically found in the field. Hydro International has also conducted tests to determine the effectiveness of Perlite as the filtration media in the Up-Flo Filter for removal of oils not associated with removal of fine sediments or removed due to buoyancy forces (Hydro International, 00). A scaled filter module, designed to eliminate all factors other than absorption onto Perlite for the removal of oil, was constructed for the test. The tests were performed at an equivalent hydraulic loading rate to a single full-scale Up-Flo Filter loading rate of gpm. The tests were conducted using oil concentrations that are more representative of an oil spill than of typical stormwater runoff, since even high-use sites rarely have oil and grease concentrations that exceed mg/l. Investigating the Perlite s capacity to control oil spills was the primary test objective. Prior to breakthrough conditions, the test showed that the Perlite achieved oil removal efficiencies greater than 90%. After breakthrough, the removal efficiencies were still greater than 0%. These test results, along with visual evidence that free oil collects on the water surface in the chamber that houses the filter modules indicate that the Up-Flo can provide reasonable protection against oil spills. The floatable trash retention capability of the Up-Flo Filter was measured at 0 gpm (0. cfs), 00 gpm (0. cfs), and 0 gpm ( cfs) (Hydro International, 00). To simulate a range of bypass overflow intensities, each of the flow rates tested was well above the gpm filtration rate of one filter module. For each trial, flows were piped into the Up-Flo Filter at a constant flow rate. The water level in the Up-Flo test tank would rise until it reached the height of the inner weir of the bypass siphon, where the water level would reach steady state. The chamber of the Up-Flo Filter was preloaded with 0 gallons of floatable trash and debris; including plastic containers, styrofoam packing peanuts, paper coffee cups, lids and straws. Due to the trap created in the chamber by the module components, no floatable materials were present at the outlet while the water level was rising and once it exceeded the bypass weir elevation. The Up-Flo Filter is completing verification reporting from the Environmental Technology Verification (ETV) program at the Penn State Harrisburg Environmental Engineering Laboratory and the final report is scheduled for release in Fall 008. Further field evaluations of the Up- Flo Filter and other media mixes are currently underway at the Stormwater Center at the University of New Hampshire and by the University of Alabama.. Patents Robert E. Pitt, David A. Woelkers, and Jeffrey K. Suhr hold the USA patent in reference to Up- Flo Filter: Upflow Surface Water Runoff Filtration System, Patent No. 7,00,00 B. 9

21 . Technical Resources, Staff and Capital Equipment For over years, Hydro International has been working in partnership with their customers to ensure successful solutions throughout the design and installation process and has developed considerable expertise in the implementation of sustainable drainage systems. These systems include treatment, storage and flow controls. Technical assistance is provided by an engineering staff at Hydro International s U.S. headquarters in Portland, Maine in addition to local Hydro International representatives in the State of New Jersey. Custom sizing and drawings are available for a given project. Hydro International maintains a full-scale test facility in Portland, Maine as described in Section.. To ensure results are accurate and unbiased, Hydro International utilizes full-scale, state-ofthe-art testing technology both in-house and through independent centers of excellence including the following: Academic Institutions Federal and State Regulatory Agencies Research Institutions Accredited Analytical Laboratories Consulting Engineers Municipalities In addition to field testing and external validation, Hydro International has developed considerable expertise in Computational Fluid Dynamics (CFD) simulation. This ability to mathematically model flow fields and assess device characteristics is enabling rapid prototyping, thereby shortening product development cycles and improving the quality of outputs. Hydro International promotes the benefits of sustainable strategies to the wider water environmental community. In addition to contributing to industry events, the company hosts educational conferences, which encourage knowledge sharing, dialogue and provides networking opportunities between environmental regulators, municipalities, engineers and academic institutions. 0

22 . Treatment System Description. System Description The Up-Flo Filter operates on simple fluid hydraulics. It is self-activating, has no moving parts, no external power requirements and is fabricated with durable non-corroding components. No measures are required to operate the unit and maintenance is limited to periodic inspections, and typically annual sediment and floatables removal and Media Pack and Draindown replacement. The Up-Flo Filter is a flow-through treatment system with standard inlet/outlet pipes or catch basin with an inlet grate that is designed to filter a wide range of flows and runoff volume. The modular design of the Up-Flo Filter allows the engineer to determine the number of modules required based on a treatment flow and known filtration rate per module. Depending on the number of modules required, one to six Filter Modules are supplied for a standard -ft manhole, or greater than six Filter Modules are supplied in vaults that house multiple rings of one to six Filter Modules. The following illustrations in Figure - show the flexibility of the modular design. Figure -: Available Up-Flo Filter Systems Hydraulic testing for available media establishes the actual flow rate for varying driving head allowing customized sizing when necessary. By adjusting the height of the outlet module s bypass weir, the driving head acting on the media can increase or decrease the hydraulic loading rate per Filter Module. A standard height Bypass Weir is set to provide a driving head of 0- inches, measured from the top of the media (see Figure -). Depending on the media, the target flow rate per module at 0 inches of driving head is 0- gpm. Its siphon-activated Bypass is capable of discharging up to cfs for a standard height -ft diameter manhole.

23 Bypass Weir to Outlet Invert Bypass Weir to top of media -inches 0-inches Head Figure -: Potential Energy or Driving Head (0-inches from top of the media). Filter Media The treatment media is the heart of the Up-Flo Filter. Depending on the media type and hydraulic loading rate, the Up-Flo Filter media utilizes physical filtration, ion exchange and sorption to remove pollutants from stormwater. As described earlier, each Media Pack contains two layers of flow Distributing Media and two Media Bags that retain the media. A combination of different media can be utilized to maximize the removal of site-specific pollutants. Four filter media have been developed for use, namely, Filter Sand, Perlite, CPZ and CPS. The filtration rate through an upflow filter is dependent upon the surface area and depth of the filtration bed, as well as the particle size, density, porosity and expansion and fluidization of the filtration media. Hydraulic characterization is necessary for different media as the filtration rate through the media will not be the same for a given driving head acting on the media. Hydraulic testing was conducted to determine the filtration rate of the Up-Flo Filter at varying levels of driving head for four media types (Hydro International, 007). The test results showed flow rates of gpm using CPZ Mix TM, gpm using Filter Sand, 0 gpm using CPS Mix TM, and 8 gpm using Perlite at 0 inches of driving head. Performance evaluations of the media are discussed in Section..

24 The four media types that have been hydraulically characterized are: This proprietary carbon-peat-zeolite mix is Hydro International s custom media blend. The proportions and gradations of activated carbon, peat and manganese-coated zeolite have been adjusted for enhanced pollutant removal while allowing a high hydraulic throughput. The activated carbon targets TSS, synthetic organics, pesticides, herbicides and metals. Peat is effective at extracting free-phase or dissolved hydrocarbons, organics and heavy metals. The zeolite is effective at removing TSS, while the manganese coating provides enhanced metals removal. Filter sand is an excellent traditional filter media. The gradation is carefully selected to optimize both removal efficiencies and hydraulic throughput. The filter sand is effective at removing TSS and its associated pollutants, such as nutrients and metals, from stormwater runoff. This proprietary carbon-peat-sand mix is a cold-climate alternative to CPZ Mix, as road salt dissolved in stormwater runoff will recharge the zeolite media. During this recharging process, pollutants previously sorbed to the zeolite surface will be washed off into the effluent hence the use of sand in this mix. The gradations and proportions of activated carbon and peat in the CPS Mix are the same as in the CPZ Mix, but the zeolite is replaced by traditional, non-reactive filter sand. The fine grade filter sand provides the same TSS removal efficiency as the zeolite, but lacks the enhanced metals removal. Perlite is a naturally occurring expanded volcanic ash with a high surface area per unit volume. Perlite is effective at capturing TSS and oils & grease while allowing for a high hydraulic throughput.

25 . Technical Performance Claim Claim The Up-Flo Filter equipped with Hydro-Filter-Sand (HFS ) with one to six. ft Filter Modules and installed in a -foot diameter manhole (minimum.9 ratio of sump crosssectional area to filter surface area) with a -foot sump, having a design hydraulic loading rate of 0 gpm per Filter Module (8. gpm/ft ) and an operating head of 0 inches above the filter ( inch above the outlet invert elevation), has been shown to have greater than 80% TSS removal efficiency, measured as suspended solids concentration (SSC), for Sil-Co-Sil 0, a manufactured silica product with a d 0 of microns, with influent sediment concentrations ranging from 90. to mg/l in laboratory studies using simulated stormwater. Hydro International is conducting additional laboratory and field studies to support sediment, phosphorus, and metals removal claims for other filter media including CPZ media, CPS media, and Perlite. An addendum to this report with additional claims will be provided once this data is submitted and evaluated.. Treatment System Performance A number of tests were performed at the Hydro International test facility to evaluate the performance of the Up-Flo Filter. The testing was conducted on a one-module filter system and a six-module filter system fitted into a standard -ft diameter free-standing reinforced concrete manhole. The filter media used for the tests was HFS.. Laboratory Studies Overall, 0 experiments were conducted at the Hydro International Laboratory Test Facility to assess the performance of the Up-Flo Filter. The feed material was Sil-Co-Sil 0 as specified for stormwater filtration devices under the NJDEP test protocol. Figure - shows a plot of Sil- Co-Sil 0, which has a d 0 of microns.

26 Figure - Particle Size Distribution of Sil-Co-Sil 0 System Description Testing was conducted in the -ft diameter free-standing concrete manhole with a -ft sump. The Filter Modules were mounted in the manhole, feet from the base of the unit. The internal polypropylene components and Type 0 stainless steel support frame were the same as installed in commercial units. The set-up simulates a catch-basin insert application with stormwater runoff flowing into the chamber from an overhead grate. A -inch, non-variable Flygt pump delivered flows at a constant rate of 8 gpm (.0 cfs) from a,000-gallon clean water reservoir to a pipe network that included two butterfly valves for throttling the flow rate to the Up-Flo Filter from 0-0 gpm. The test manhole has a -inch outlet pipe that discharges into a large discharge tank on the floor of the lab. Two -inch Flygt pumps send clean water from the discharge tank back into the clean water reservoir. Sedimentladen water discharging into the discharge tank during performance testing was pumped outside for treatment and disposal. Photos of the system tested are shown below in Figure - and a schematic of the test set-up is shown in Figure -.

27 Figure -: Up-Flo Filter System as Tested in the Laboratory

28 Figure - Schematic of Test Setup The manhole configuration Up-Flo Filter is equipped with between one to six Filter Modules for testing. Each module holds two () Media Bags with each bag having a filter depth of -in. Separate testing was conducted prior to this exercise to verify the capacity of the siphonactivated Bypass normally located on top of the Outlet Module using a higher rate pump. The results showed that the Bypass has a capacity of 0 gpm ( cfs). The filtration capacity of a Filter Module depends on the filter media housed within the module and the driving head (height of water) acting on the filter media. This test evaluated the performance of a one-module set up with a filtration capacity of approximately 0 gpm at 0 inches of driving head above the filter and a six module system with a filtration capacity of approximately 0 gpm at 0 inches of head above the filter. A Hershey VP-80 butterfly valve and a Hershey VP-8 butterfly valve were used to step the flow down to the desired influent flow rate for each test. The flow rate was calibrated using the Volumetric Time-To-Fill Method. After the valves had been set to their desired notches, the time to fill the discharge tank to the 0 cubic foot mark was recorded. The resulting flow rate was calculated by dividing the volume by the time. 7

29 Test Program The goal of the test program was to determine the hydraulic loading rate per Filter Module that will meet the 80% TSS removal target set by the NJDEP. A series of tests were conducted on the one-module filter system and another series of tests were conducted on the six-module filter system. Three tests each at five different flow rates were performed on each filter system. The first test was designed to simulate 7-0 mg/l influent TSS, the second test was designed to simulate 0-0 mg/l influent TSS, and the third test was designed to simulate 0-0 mg/l influent TSS of stormwater being treated by the Up-Flo Filter. The five different flow rates correspond to the percentage of design flow used in the calculations for weighted TSS removal efficiency that was developed by NJDEP for evaluating stormwater treatment systems. A total of six discrete influent samples and six discrete effluent samples were collected during each test. The samples were sent to a commercial laboratory (Northeast Laboratory Services) for analysis using ASTM 977 Method B. After receipt and review of the analytical results from Northeast Laboratory Services, event mean concentrations (EMCs) for the influent and effluent streams, respectively, were calculated using the discrete sample results. Procedure The -inch submersible pump in the,000-gallon clean water reservoir is turned on to pump flows to the Up-Flo Filter manhole. The flows are adjusted using the two butterfly valves until flow to the manhole is in steady state at the required flow rate. The required mass of Sil-Co-Sil 0 to obtain the concentration for the target flow rate is weighed out and mixed with water to form a slurry in a -gallon barrel. The slurry is kept in suspension by continually stirring the mix over the course of the test. A Watson Marlow peristaltic pump connected to the slurry in the barrel is turned on to dose the influent flow at a constant rate through a -inch stand pipe located a few feet upstream of the Up-Flo Filter manhole (refer to Figure -). When the feed sediment enters the chamber, the stop watch is started and testing is deemed to have started. The first influent sample is taken to minutes after the introduction of the slurry depending on the flow rate. Five other influent samples are taken at one-minute intervals for the one-module filter system and at half-minute intervals for the six-module system. The first effluent sample is taken after a timed interval corresponding to the time required to displace one test volume in the manhole following the first influent sample. Five more effluent samples are taken at the same interval following the corresponding influent samples. After the last effluent sample, the peristaltic pump is turned off and the submersible pump in the reservoir is also turned off. The water level in the chamber is monitored during the duration of a test to ensure there is no significant change in head over the duration of the test. All of the tests were run with the filtered Draindown in place; the reported removal rates are a product of flow through both the Filter Modules and the Draindown. Effluent samples were not collected during the Draindown time after the tests were run. 8

30 . Verification Procedures The Up-Flo Filter was verified under two different filter configurations. The first configuration consisted of the one-module filter system and the second configuration consisted of the sixmodule system. These two configurations represent the range for which a -ft diameter manhole filter system can operate. One-Module Testing Fifteen tests were performed using the one-module configuration. The flow rates for the tests ranged from. to.7 gpm. The results for the range of flows tested are shown in Table.. The head required to drive flow through the system was fairly consistent for the three tests run at a given flow rate. Filter performance was generally dependent on flow with better removals achieved at lower flow rates. Observed pollutant removal efficiencies ranged from 7-9%. Table. Sil-Co-Sil 0 removal for one-module system Test Target Flow Rate (gpm) Actual Flow rate (gpm) Target TSS Influent Range (mg/l) Actual TSS Influent (mg/l) Effluent (mg/l) Removal Rate (%) Head (in) See Appendix A for the Northeast Laboratory Services analytical results from the six discrete influent and effluent samples for each of the tests. 9

31 Averages for actual flow, removal rate, and head for each of the targeted flow rates are shown in Table.. Table. Averages of results from one-module system tests Target Flow Rate (gpm) Actual Average Flow Rate (gpm) Average Removal (%) Average Head (in) * * Design hydraulic loading rate, or, 00% operating rate The plot in Figure. shows the removal rates for the Up-Flo Filter in the one-module configuration for the tests averaged by flow rate. The equation for the line of fit for the plot is used to calculate the efficiency of the Up-Flo Filter. Based on the weighting developed by NJDEP for stormwater treatment systems, the removal rates for the Up-Flo Filter using a onemodule system is shown in Table.. With a design hydraulic loading rate of 0 gpm per module, the calculation shows that for the Sil-Co-Sil 0 using Hydro Filter Sand, the Up-Flo Filter exceeds the 80% removal target set by NJDEP. Removal Efficiency (%) y = -0.9x + 9. R = Flow Rate (gpm) Figure - Removal Efficiency vs. Flow rate for One-Module System 0

32 Table. Up-Flo Filter: One-module system removal efficiency per NJDEP weighting Operating rate Weight Factor Design Flow (gpm) Removal (%) Weighted removal (%) Treatment efficiency 8

33 Six-Module Testing The results of the six-module tests are shown in Table.. The tests had flows ranging from 0 to gpm with target influent concentrations of 7-0 mg/l, 0-0 mg/l, and 0-0 mg/l. Three of the tests were randomly selected and repeated on July 0, 008. The test set-up and sampling were witnessed and documented by the University of New Hampshire (UNH) Stormwater Center (Houle, 008). Removal rates ranged from 7 to 89% with the highest removal occurring at the lower flow rates. Table. Sil-Co-Sil 0 removal for six-module system Test Target Flow Rate Actual Flow rate Target TSS Influent Range Influent (mg/l) Effluent (mg/l) Removal Rate (%) Head (in) (gpm) (gpm) (mg/l) * * * *Independently witnessed by UNH Stormwater Center See Appendix B for the Northeast Laboratory Services analytical results from the six discrete influent and effluent samples for each of the tests and the UNH witnessed tests.

34 Averages for actual flow, removal rate, and head for each of the targeted flow rates are shown in Table.. Table. Averages of results from six-module system tests Target Flow Rate (gpm) Actual Average Flow Rate (gpm) Average Removal (%) Average Head (in) * *Design hydraulic loading rate, or, 00% operating rate The plot in Figure. shows the removal rates for the Up-Flo Filter in the six-module configuration for the tests averaged by flow rate. The equation for the line of fit for the plot is used to calculate the efficiency of the Up-Flo Filter. Based on the weighting developed by NJDEP for stormwater treatment systems, the removal rates for the Up-Flo Filter using a sixmodule system is shown in Table.. With a design hydraulic loading rate of 0 gpm per module, the calculation shows that for the Sil-Co-Sil 0 using the Hydro Filter Sand, the Up- Flo Filter exceeds the 80% removal target set by NJDEP. Removal Efficiency (%) y = -0.07x R = Flow Rate (gpm) Figure - Removal Efficiency vs. Flow Rate for Six-Module System

35 Table. Up-Flo Filter : Six-module system removal efficiency per NJDEP weighting Operating Weight Design Flow Removal Weighted removal rate Factor (gpm) (%) (%) Treatment efficiency 8 Testing performed in the Hydro International Laboratory Test Facility with Sil-Co-Sil 0 demonstrated that the Up-Flo Filter operating at the design hydraulic loading rate of 0 gpm per module (8. gpm/ft ) for a one-module system and for a six-module system is capable of reducing fine sediment loads in excess of 80%.. Inspection and Maintenance Maintenance activities can be categorized by those that can be performed from outside the Up- Flo vessel and those that are performed inside the vessel. Maintenance performed from outside the vessel includes removal of floatables and oils that have accumulated on the water surface and removal of sediment from the sump. Maintenance performed inside the vessel includes removal and replacement of Media Bags, flow Distribution Media and Draindown. A vactor truck is required for removal of oils, water, sediment, and to enter the vessel for performing inside maintenance. There is the need to follow OSHA Confined Space Entry procedures when entering the Up-Flo vessel. Operation and maintenance procedures are described in the O&M manual (Hydro International, 007)... First-Year Inspection and Maintenance The frequency of inspection and maintenance can be determined in the field after installation. Based on site characteristics such as contributing area, types of surfaces (e.g., paved and/or landscaped), site activities (e.g., short-term or long-term parking), and site maintenance (e.g., sanding and sweeping), inspection and maintenance should be conducted at intervals of no more than six months during the first year of operation. Maintenance personnel should observe and record pollutant accumulations during the first year of service in order to benchmark the maintenance intervals that will later be established for the site. Pollutant accumulations should be measured or monitored using the following procedures: Measurement of sediment depth in the sump. Minimum 8 should separate the Draindown inlet from stored sediment in the sump in order to minimize sediment entrainment into the Draindown. A simple probe, such as the Sludge-Judge, can be used

36 to determine the depth of the solids in the sump. In a typical -foot diameter manhole installation, the sediment depth should be no more than. Measurement of sediment accumulation in Media Bags. Minimum filtration rate is generally reached when the Media Bags have accumulated approximately 0 lbs of sediment (Hydro Int. asserts that this is a conservative estimate based on prior ETV and Tuscaloosa, AL. testing. [Hydro International, 008]). Determining the amount of accumulated sediment will be accomplished by removing both of the Media Bags from one of the Media Packs and weighing the bags separately. Since a new Media Bag weighs approximately 0 lbs wet, the difference in weight will approximately equal the weight of solids that have accumulated in the bag. A spent Media Bag weighs approximately 0 lbs wet. Measurement of oil layer on water surface. Since water in the Up-Flo vessel drains down to an elevation below the bottom of the Filter Modules when the system is idle, the amount of accumulated oil must be minimized so that oil is not entrained into the Media Pack when stormwater begins to fill the vessel at the start of a storm event. Oil accumulation should be limited to. or less. Probes can be used to measure oil thickness. Monitoring for Draindown clogging. The Draindown is designed to lower the water level in the Up-Flo vessel to an elevation below the bottom of the Filter Modules between storm events. If the water level is above the bottom of the filter module one to two days after a storm event the Draindown has likely become clogged with sediment. Monitoring for slime and debris covering the flow Distribution Media or angled screens. After removal of the Media Bags, the bottom flow Distribution Media should be removed and inspected to determine if it is coated with slime or debris. Similarly, the angled screen should be inspected for blockages and ragging. Monitoring for floatables on the water surface. Similar to oil, the amount of accumulated floatables must be minimized to prevent trash and loose debris from becoming trapped on the angled screens when stormwater begins to fill the Up-Flo vessel at the start of a storm event. Floatables should not be allowed to completely cover the surface of the water. The solids loading rate in the sump will be calculated by measuring the sediment depth in the sump and dividing the depth by the correlating interval of time since the oil was removed or the sump was last cleaned. Similarly, starting with fresh Media Bags, the solids loading rate in the Media Packs will be calculated by weighing the Media Bags and dividing the weights by the correlating interval of time since they were installed. The wet weight of the heaviest bag will be used to determine the loading rate. As previously mentioned, a spent Media Bag weighs approximately 0 lbs wet. The spent Media Bag weight estimate was based on calculations of sediment loading in an Up-Flo Filter that was run to exhaustion during laboratory testing. The rate of oil accumulation will be calculated by measuring the thickness of the oil layer and dividing the thickness by the correlating interval of time since the sump was last cleaned. Ordinarily, oil thickness will not be measurable unless a spill has occurred. Consequently, any oil will typically be removed along with water when cleaning the sump.

37 Monitoring the Draindown for clogging, monitoring the flow Distribution Media and angled screens for slime and debris, and monitoring the accumulation of floatables will provide an estimate of how long the Up-Flo Filter can operate before its performance can become impaired by one of these factors... Routine Inspection and Maintenance After completion of the first year of operation, the established inspection and maintenance intervals will keep pollutant loadings within their respective limits. Removal of oils and floatables, replacement of the Draindown, replacement of flow Distribution Media, and cleaning of angled screens will occur at the same frequency as cleaning of the sump and replacement of Media Bags unless the first year of operation indicates otherwise. Keeping to the established maintenance intervals will keep treatment flow rates at, or above, the design flow rate. Typically, annual maintenance is adequate. In addition to scheduled maintenance, occasional checks for Up-Flo Filter clogging can be performed by removing the manhole cover during a storm, monitoring the water level in the manhole or vault, and determining whether the filter is in bypass. A properly-sized filter (on-line or off-line) that is in bypass during a storm that is producing runoff at, or below, the filter s design filtration rate needs maintenance... Maintenance Procedures The access port located at the top of the manhole or vault provides access to the Up-Flo vessel for maintenance personnel to enter the vessel and remove and replace Media Packs. The same access would be used for maintenance personnel working from the surface to net or skim debris and floatables or to vactor out sediment, oil, and water. Unless the Up-Flo Filter has been installed in a very shallow unit, it is necessary to have personnel with OSHA-confined space entry training performing the maintenance that occurs inside the vessel. Maintenance activities include inspection, floatables removal, oil removal, sediment removal, Media Pack replacement, and Draindown replacement. Maintenance intervals are determined from monitoring the Up-Flo Filter during its first year of operation. Depending on the site, some maintenance activities may have to be performed on a more frequent basis than others. In the case of floatables removal, a vactor truck is not required. Otherwise, a vactor truck is normally required for oil removal, removal of sediment from the sump, and to dewater the vessel for replacement of the Media Packs and Draindown. All inspection and maintenance activities would be recorded in an Inspection and Maintenance Log. Completion of all the maintenance activities for a typical -foot diameter manhole installation takes less than one hour. Approximately 0 gallons of water and up to 0. yd of sediment could be removed in the process. In an installation equipped with six Filter Modules, Media Bags ( bags per module) would be removed and replaced. Assuming a spent Media Bag weight of 0 lbs, up to 00 lbs of spent Media Bags would be removed. All consumables, including

38 Media Bags and flow Distribution Media, are supplied by Hydro International. Good housekeeping practices upstream of the Up-Flo Filter can significantly extend Media Bag life. For example, sweeping paved surfaces, collecting leaves and grass trimmings, and protecting bare ground from the elements will reduce loading to the system. Media Packs should not be installed in the Filter Modules until construction activities are complete and site stabilization is effective. Maintenance activities performed on an Up-Flo Filter installed in New Zealand are illustrated in Figure -. The photographs also show the condition of the flow Distribution Media, Media Bags, and filter media (i.e., sand) after they have become spent. More than 00 Up-Flo Filters of various sizes have been commissioned in New Zealand since 00. The filters were permitted by the Auckland Regional Council and installed to treat runoff from a variety of land uses. Many of these have been in operation for more than a year and have gone through at least one maintenance cycle, including replacement of filter Media Bags. Because maintenance of the Up- Flo Filter is relatively simple, the New Zealand experience has involved systems being maintained by a servicing team who are a retired couple... Solids Disposal Sediment, floatables, gross debris, and spent Media Bags can generally be disposed of at the local landfill in accordance with local regulations. The toxicity of the residues produced will depend on the activities in the contributing drainage area, and testing of the residues may be required if they are considered potentially hazardous. Sump water can generally be disposed of at a licensed water treatment facility but the local sewer authority should be contacted for permission prior to discharging the liquid. Significant accumulations of oil removed separately from sump water should be transported to a licensed hazardous waste treatment facility for treatment or disposal. In all cases, local regulators should be contacted about disposal requirements... Damage Due to Lack of Maintenance Delayed maintenance would result in clogged filters and/or blinded angled screens. In that situation, the Up-Flo Filter would go into bypass and there would be no treatment of the incoming stormwater. Because the siphonic Bypass can easily convey all of the flow to the Outlet Module, there would be no lasting damage to the system. Replacement of the Media Bags and removal of sediment from the sump would restore the Up-Flo Filter to its original treatment efficiency. Establishing and adhering to a regular maintenance schedule ensures optimal performance of the system. 7

39 Figure -: Up-Flo Filter Maintenance Photographs (New Zealand) 8