Overview Description of the CSO Control Program

Utilizing the approximately two years of sampling data at each basin, a detailed evaluation of nine basins was undertaken to examine the performance of the facilities relative to the three goals summarized in Table 2. The River Rouge CSO basin did not become operational until 2002. It is currently being evaluated to determine its performance.

Goal 1: Ability to Compare Data for All Basins

The first goal or criteria for success was the ability to compare data from all Phase I CSO basins. The design approaches and assumptions used in the design of the basins varied widely among the basin designers. In order to draw conclusions about the performance of the demonstration CSO basins, the actual basin detention time at each facility given the same storm event and conditions needed to be determined. Knowing the actual detention time versus a design detention time provides a means of comparing the relative sizes of these facilities and of comparing their size to the demonstration and presumptive sizing criteria. Then an evaluation can be made as to the effectiveness of the residence time on settling and disinfection.

The CSO basin evaluation has resulted in the following findings:

1. The amount of runoff that is generated by a storm event is greatly influenced by antecedent moisture conditions. Antecedent conditions relate to the amount of rainfall that occurred in the days prior to the event being evaluated.
2. The amount of runoff that is generated by a rainfall event is greatly influenced by seasonal variation in the infiltration capacity of the soil and interception by foliage. A greater response is seen during winter and early spring conditions when infiltration capacity is lower or non-existent due to frozen ground conditions and due to lower interception by foliage (tree cover).
3. The amount of runoff that is generated by a rainfall event is greatly influenced by seasonal variation in the peak hour intensity of the rainfall events, which was also found to vary seasonally, with higher peak intensities occurring during May through October.

Thus, a design event cannot be adequately defined without specifying these items.

These results illustrate the value of having site-specific monitoring data to develop the parameters used for sizing the basins. Critical in this sizing analysis are a consideration of the seasonal variation in rainfall intensities and the impact of antecedent moisture conditions. Having sufficient data to estimate flows that would be generated by a design event is difficult, but it is vital to ensure that the facility will be sized properly.

Because the actual detention times have been determined to be greater than the original design intended for three of the four basins analyzed, it is important to look at the actual detention time versus the design detention time in making an assessment of how effective a CSO basin is in meeting the regulatory requirements. The additional detention time provided by these basins could be providing additional treatment effectiveness. This benefit needs to be considered in answering the question of "What is the proper size for a basin to meet regulatory requirements?"

Goal 2: Protect Public Health and Eliminate Raw Sewage

The objective of Goal 2 is two-fold: a) the protection of public health and b) elimination of raw sewage. Success in achieving Goal 2 is demonstrated using measurements and observations of CSO basin effluent.

Protect Public Health

The evaluation of the first part of Goal 2, protection of public health, is based on disinfection of effluent and determining if the Michigan fecal coliform limit of 400 cts/100 ml is met by the effluent while minimizing total residual chlorine (TRC). A higher chlorine dose results in a greater treatment of the fecal coliform, but it also tends to result in a higher TRC. At the start of the project, the goal for CSO basin effluent TRC was 1 mg/L or lower. Individual effluent samples collected during a range of events were tested for fecal coliform concentrations and discrete measurements of TRC were also made. For each facility and storm, an event geometric mean for effluent fecal coliform and an event mean TRC value were calculated. Success is judged to be achieved if the event geometric mean fecal coliform across the range of monitored storm events is in compliance with the limit.

The results of the basin evaluations indicate the facilities can consistently meet the fecal coliform limit when operated to maintain effluent TRC concentrations of 1 mg/L or greater. For discharge events with effluent TRC concentrations less than 1 mg/L, the fecal coliform limit was usually not met. As a result of the evaluation and a consensus by one of the Rouge Project work groups, the target range for effluent TRC was revised from 1.0 to 1.5 mg/L.

Eliminate Raw Sewage

The second part of Goal 2, determining if the basin eliminates raw sewage, is based on the ability to remove sanitary trash and identifiable sanitary solids. The basis for this criterion is that the discharge can only be considered treated and no longer raw if it does not have the visual appearance of raw sewage (in addition to being disinfected). Removal of sanitary materials includes three process treatment components: screening, skimming and settling of the flow. Screens installed at each facility contribute to the removal of larger debris from the flow. Baffle walls installed in the vicinity of influent, intermediate or discharge weirs help to facilitate the removal of floating materials. These materials are captured in the basin for removal during dewatering or flushing. Sedimentation allows for the removal of heavier objects and solids. The ability of the facilities to achieve removal of these objects is typically related to the surface overflow rate and the weir loading rate.

Elimination of raw sewage was evaluated by visual observations and a netting study. Visual observations were performed during daylight hours at the four CSO basins where it was feasible, and were taken as representative of the other basins. Mesh nets with small rectangular openings of 6 mm were placed over a portion of the basin outfall grating at 4 basins to determine if any sanitary trash were being discharged from the basin to the river. No evidence of sanitary trash was seen in the effluent of those basins observed, except for a small amount on one occasion. This event was not viewed as a concern as it was caused by abnormal basin operation following a power failure.

It is very important to note that based on the evaluations performed on the effluent discharge from the basins, MDEQ has concluded that all nine of the Rouge Phase I CSO basins evaluated are currently meeting the Goal 2 criteria of the elimination of raw sewage and the protection of public health. The tenth facility is still being evaluated.

Goal 3: Achieve State Water Quality Standards

The objective of Goal 3 is to achieve state water quality standards in the receiving stream at times of discharge. Success in achieving Goal 2 (protect public health and eliminate raw sewage) is demonstrated using measurements of basin effluent, whereas the success in achieving Goal 3 is demonstrated using measurements in the receiving stream at times of discharge. The four criteria of success for Goal 3 as outlined in Table 3 are 1) the dissolved oxygen (DO) standard, 2) the physical characteristics standard, 3) the total residual chlorine (TRC) standard, and 4) the health of the biological community (as a surrogate for toxic materials and other pollutants).

Dissolved Oxygen Standard

The state's DO standard for the Rouge River is a minimum of 5 mg/L at all times. To achieve Goal 3 for the DO standard required measuring DO in the river both upstream and downstream to show that the basin does not cause the DO to go below the standard of 5 mg/l. This criterion was evaluated by continuous water quality monitoring stations and predictive modeling to determine the magnitude and location of transitory dissolved oxygen sags caused by effluent from the facilities.

All available monitoring data and modeling results have led to the following findings:

1. The DO standard is being achieved downstream of all three Oakland County CSO basins at times of discharge, except for DO sags that were not caused by the basin effluent.
2. At times of basin discharge, the DO standard is being achieved from the Redford basin down to a location where other uncontrolled CSOs impact the river. DO did sometimes drop below the standard during wet weather events, but since this pattern was nearly identical upstream and downstream of the Redford basin, the cause of the violation was not attributed to the basin.

Physical Characteristics

The state's physical characteristics standard prohibits unnatural physical properties (turbidity, color, oil films, floating solids, foams, settable solids, suspended solids, deposits) in quantities which are, or may become injurious to any designated use of the receiving stream.

The physical characteristics criterion was evaluated by comparing effluent and instream total suspended solids, and by periodic visual observations of the effluent plumes, primarily to check for the presence of oil films or high turbidity. Also, effluent nets on the CSO discharges were used to determine if any sanitary trash or identifiable sanitary solids were present in the effluent. The documented data for the physical properties considered indicate there are no unnatural physical properties in quantities that are considered injurious to any designated use at times of basin discharge.

Total Residual Chlorine

As required by the state water quality standards, instream TRC concentrations must not exceed the final acute value of 0.038 mg/L, unless a mixing zone demonstration is performed and accepted by the state. This standard is difficult to meet given that effluent TRC must be at least 1 mg/L for adequate reduction of fecal coliform bacteria. The TRC standard was evaluated by measuring instream TRC concentrations downstream of CSO basin discharges. Due to difficulties in safely accessing the river during high flow conditions, these measurements were limited to one facility where a boat could be used to perform instream sampling. While instream TRC concentrations in excess of 0.038 mg/L were measured near shore at several downstream transects, the concentrations were below detection for two-thirds of the river cross-section. While the results show there is a zone for fish passage, a formal mixing zone demonstration was not performed for the facility.

In summary, the evaluation performed to date has shown that instream TRC measurements have exceeded the state standard, but there has been no conclusion regarding the specific environmental impact caused by chlorine residuals in the Rouge River. As the next step in the evaluation process, the NPDES permit for the three Oakland County CSO basins requires that a TRC Mixing Zone/Plume Definition Study be performed for each basin. The purpose of the study is to determine the water quality impacts in the Rouge River from TRC in the treated basin effluent. The permit required a work plan for the study to be submitted in September 2003 with the study report to be submitted by March 2006. If the study indicates that effluent limitations or additional controls for the discharge of TRC are needed to meet water quality standards at times of discharge, the MDEQ may propose to modify the NPDES permits for these three basins to include appropriate effluent limitations and/or controls for the discharge of TRC.

Biological Assessment

Finally, the health of the biological community was assessed in the vicinity of several Rouge Phase I CSO basin discharges as a surrogate for assessing the presence of toxic materials and other pollutants in the basin effluent. Macroinvertebrate samples were collected upstream and downstream of the CSO basins in Redford and Oakland County (four basins total) to investigate the compliance of these basins with Goal 3. Collection of organisms and calculation of community scores followed the Great Lakes and Environmental Assessment Section (GLEAS) Procedure 51 (MDNR, 1991). As a result of this assessment, it was agreed by MDEQ that the water quality standards for toxic substances are presumed to be achieved upstream and downstream of the Oakland County and Redford CSO basins, even at times of discharge, as there is no measurable effect of these discharges on the health of the biological community.

Summary

It is very important to note that evaluations relative to Goal 3, achieving state water quality standards, have been completed at three of the Rouge Phase I CSO basins and MDEQ has certified that these basins are meeting water quality standards at times of discharge, except for meeting the yet-to-be-evaluated total residual chlorine standard. This certification is expected for at least another three basins in the year 2004. Additional details of the in-stream evaluation process can be found in the paper titled "Evaluation of In-Stream Impacts of CSO Control Facilities" Trend analysis results clearly demonstrate that DO concentrations are improving in the Rouge River Watershed during both wet and dry weather conditions. Eight of nine locations show a statistically significant improving trend for the mean DO with the annual average improvement ranging up to 0.53 mg/L per year. The ninth location is still being evaluated and is influenced by many uncontrolled CSO outfalls.

Lessons Learned on Evaluation Question 1

Lessons learned relative to evaluating compliance of the Rouge Phase I CSO basins with regulatory requirements include the following:

1. Regulatory requirements can be ambiguous; thus, developing a well defined, technical evaluation process was critical to determine compliance with regulatory requirements. Use of collaborative work groups that involved all the stakeholders provided important consensus on defining criteria for success.
2. The process for evaluating the impacts of the basins has led to constructive decisions. Working with a large group trying to achieve consensus has taken more time, but was well worth the effort. MDEQ agreed that implementation of further controls is dependent on this evaluation.
3. A design event cannot be adequately defined without specifying antecedent conditions, season, and rainfall intensity. These factors all impact the amount of wet weather response that will be generated for the same size storm event. Ideally, continuous simulations that account for these factors by using rainfall records over a number of years should be used to predict runoff rates and volumes. These simulation results would better predict the performance of the basins in terms of frequency and volumes of CSO captured or overflowed.
4. The Rouge Phase I CSO basins provide retention and treatment and the frequency and magnitude of discharges has been considerably reduced. Therefore, the basins only discharge treated effluent in large events when assimilative capacity of the river is the greatest.
5. Other sources of water pollution are significant, but the evaluation of CSO impacts can be reasonably isolated from other non-point source pollution impacts to draw conclusions on the effectiveness of the CSO basins in achieving the water quality standards.
6. A two-year time frame for monitoring and/or 10 overflow events at each facility is generally sufficient to characterize effluent quality, demonstrate ability to meet effluent permit limits, and demonstrate ability to meet water quality standards at times of discharge. In some cases, an additional 3 to 6 months of study is being performed to further examine some issues specific to certain basins.
7. Treated effluent from the CSO control facilities studied so far seems to be relatively good quality. Permit effluent limits for fecal coliform can be met consistently when effluent TRC is greater than 1 mg/L. Water quality standards are being met at some facilities, but detailed evaluations are ongoing for the other basins, particularly for DO.
   8. Instream TRC from basin effluent exceeds state standards at times, but further evaluation is needed to determine if there is any adverse environmental impact.

Overall, the evaluation of the Phase 1 CSO control facilities is providing valuable technical information for future phases of CSO control in the Rouge watershed and for communities embarking on CSO control in other watersheds. The development of an evaluation process provided an innovative forum for stakeholders to collaboratively establish objectives for CSO control within the goals of urban watershed restoration. Since wet weather control is expensive, having a well defined, technical evaluation process to determine compliance with regulatory requirements is important.

Evaluation Question 2: What Treatment and Hydraulic Processes Are Most Effective?

Treatment and Hydraulic Processes Used

A variety of design features were employed to explore what treatment and hydraulic processes are most effective in reducing CSO loads to the river. The Rouge facilities all included influent or effluent screens, skimming baffles, storage and settling compartments within the basins that can capture small events or work in a flow-through mode for large events, and disinfection using sodium hypochlorite. The facilities also included a variety of innovative technologies: swirl concentrator, first flush compartments, compartments in parallel, compartments in series, decanting outlets and shunt channels

A summary of the basin configuration and size for the nine Rouge Phase I CSO basins in operation is provided in Table 4. Six of the facilities have the ability to capture the first flush; that is, to route flows to a second flow-through tank after the first tank/compartment is full. This prevents the discharge of any loads that may be captured with the first flush, which has been found to have higher concentrations of pollutants. One facility has a swirl concentrator, and three facilities have a shunt channel to avoid potential for resuspending solids after the facility is full. For screening, most of the facilities used a bar spacing of 0.75 inches. Two basins used screen bar spacing of 0.5 inches and one used 1.5 inches.

Table 4
Rouge Phase I CSO Basin Configuration and Size

For achieving disinfection, the basins were designed around dosing concentrations of 10 to 11.6 mg/L and assuming the sodium hypochlorite is stored at a 5 to 6 percent solution. Design dosing rates ranged from 1,400 to 12,600 gal/hr.

Results of Evaluation Question 2

The primary treatment process that is most effective is simply the size of the overall basin that translates into volume captured by the basin. The larger the basin, the lower the volume and frequency of overflow will be.

The evaluation of the different design features and capacities of the Rouge River CSO basins provide a basis for comparing the respective operating performances of the different components and different basis of sizing. As more operating experience is gained, the benefits of first flush tanks, tanks in series versus tanks in parallel, and shunt channels will be able to be better quantified.

Decanting of stored and treated CSOs is a potential procedure that is being considered. Decanting refers to the practice of screening, chlorinating, and settling CSO in the basin, then releasing a portion of the treated effluent to the receiving water. Tests are being performed and submitted to the Michigan Department of Environmental Quality for review. In one test in January 1999, the treated effluent had Carbonaceous Biochemical Oxygen Demand (CBOD) and TSS less than 20 mg/L, and the bacteria and TRC standards were also met. This practice has the advantage of enhancing flows in the river and avoiding the cost of treating the flow again at the wastewater treatment plant.

Based on the two years of experience, it is clear that more design attention should be paid to how the basins can be operated in low flow conditions. Much of the design effort for any CSO control facility focuses on performance under peak flow conditions. However, there are usually dozens of events each year that are smaller events with low flows. In fact, low flows accounted for over 80 percent of the events in the Rouge watershed. Supervisory control and data acquisition (SCADA) and weather radar information is critical to monitoring flows and storms to know how to initially respond to small events and low flow rates. Also special operating procedures are required to accurately monitor low flows. The multi-path ultrasonic flow meters that are used at the Rouge Phase I CSO basins are good for high flows, but they have not proven to be accurate under low flow collection conditions, especially when the water level within the collection system and CSO basin rises. At low flows, it is particularly difficult to accurately determine the proper rate of chlorine addition. Continuing operating experience is required to establish standard operating procedures for low flow conditions.

Better estimates of the effectiveness of the mixing of the sodium hypochlorite with the CSO influent is needed to better assess the impact of mixing to disinfection effectiveness. A standard methodology for evaluation of the mixing process would be helpful for evaluation of this aspect of the basin processes.

At this time, additional investigation is underway to "stress test" the swirl concentrator in the Redford CSO facility and to determine the performance of this unit separate from the detention basins that are also part of the facility. These tests will improve operating protocols for determining how the swirl concentrator should be used in conjunction with the basins.

Lessons Learned on Evaluation Question 2

The Rouge Project has evaluated CSO discharge data in the Rouge watershed and influent and effluent quality data from existing CSO basins in Michigan. Analysis was performed using results from 390 CSO storm water runoff events in the Rouge River watershed and from 44 events at CSO basins outside of the watershed. The following lessons learned were developed from this analysis:

1. Samples collected at CSO discharges confirm that pollutant concentrations decrease over the course of the discharge event. This phenomenon has been identified for CSO discharges for years and is referred to as a first flush phenomenon, with the first flush generally defined as the first 30 minutes of a storm water runoff event.
2. Samples collected of the influent to the facilities indicate that smaller events have greater pollutant concentrations than larger events. In addition, the smaller events, which are captured by the facilities for subsequent discharge back to the collection system after the storm event, account for over 80 percent of the events.
3. Because the facilities can contain the smaller events with the higher pollutant concentrations, the annual pollutant load captured is much greater than the annual volume captured. The use of first flush compartments to capture the first portion of larger events also contributed to this result.
4. The magnitude of CSO pollutant load reduction is related more to capture of concentrated portions of the flow than removal by settling. Thus, use of first flush compartments and flow bypass (shunting) were included in the design of some of the facilities. No significant removal of BOD, ammonia or phosphorus (other than related to volumetric capture) was apparent in the facilities.
5. CSO pollutant quality during larger wet weather events (those that would activate a treatment facility) was generally quite dilute for the following parameters: BOD (series), ammonia and phosphorus.
6. Effluent quality was generally low in BOD ( < 30 mg/L).
7. Significant reductions in TSS were noted for flow through processes even with loading rates as high as 7,900 gpd/sf.
8. Removal of visible sanitary trash was accomplished through a combination of settling/baffling system and screens. It was unresolved whether fine screens were necessary to achieve this performance objective in facilities where settling and skimming were provided. Moreover, the following was not determined:
   1. Maximum screen size that would work when settling and skimming are provided.
   2. Screen size necessary when degree of settling and skimming is reduced due to either a condition of a smaller basin facility or use of a flow through facility, such as a tunnel. A tunnel or a smaller basin that operates in a flow-through mode would not offer the settling velocity characteristics provided by the basin facilities evaluated for this document.
9. Effective disinfection of discharges was accomplished with sodium hypochlorite. Minimum detention times at peak overflow conditions rarely fell below 30 minutes. The relationship between detention time and reduction in bacteria colonies could not be determined; however, for the events monitored, the results suggested that the detention time provided was not the controlling factor.
10. First flush capture compartments assisted in the effective disinfection of discharges with less variability in results.
11. The wide range of flow rates with various levels of solids though the facility must be carefully considered in the design of the hydraulics and mechanics of moving flows. For instance, an underflow pump was installed without consideration of need to handle solids and grits, causing problems with the operation of this part of the flow stream through the facility.

Evaluation Question 3: What is Needed for Operational Effectiveness?

In addition to the effectiveness of the design of the various treatment and hydraulic processes employed at the Rouge Phase I CSO basins, the effectiveness of the operation of these processes is also critical to the successful performance of the basins. The process features of each facility include retention (storage), screening, and disinfection of the discharge with sodium hypochlorite. In addition, equipment is provided to clean the facility following an event, monitor flow rate and volume, collect samples, and return flow to the interceptor system (dewatering). The facilities have computer control systems with varying degrees of sophistication.

Operating data collected since June 1997 provides important information on design features and on various aspects of operating CSO facilities. The Rouge CSO control facilities have been activated on average once or twice per month. The Hubbell-Southfield facility is activated most frequently, because the overflow point that it serves has the largest drainage area and smallest relative capacity for transmission of wet weather flows downstream for treatment. Elsewhere, the storage capacity and dewatering capacities have been exceeded on average only 2 to 4 times per year, depending on the facility. When storage capacity is exceeded the basins provide settling, screening, and disinfection of the discharge.

The success of a facility's operation is dependent on the ability of the facilities to reliably meet permit and other regulatory performance measures. Of particular interest is the relationship between the water quality benefits of certain design features and the practices in operating these features. The following discusses the operation and performance of disinfection, screening, dewatering, and flushing. Training of staff, use of mobile staffing, control systems, and monitoring are additional aspects of operating a facility effectively that are also discussed.

Staffing

The 10 Rouge Phase I CSO retention treatment basins include three operated by the City of Detroit Water and Sewerage Department (DWSD), three operated by the Oakland County Drain Commissioner, and four operated by the Wayne County Department of Environment. The local communities assist Wayne County at two of these basins.

The Wayne County and the DWSD basins have no full-time staff except for at the Redford basin, which has one full-time staff person. The Oakland County basins have one full time staff person per basin. All basins rely on mobile crews and have SCADA technology for operational control. Existing staff persons with pump station and wastewater sampling experience were selected to form the core group of CSO basin operating, sampling and maintenance group.

Each operating agency has a somewhat different staffing plan. The basins operated by Wayne County use a 3-person team per basin, and each team is mobile. The basins operated by the Oakland County Drain Commissioner have a mobile supervisor plus one person stationed full-time on day shift at each basin. The DWSD facilities use a 3-person team per basin, and each team is mobile. The basins are within 4 to 13 miles of home base for all three operating agencies.

In October 2000, the operators participated in an "Operator Forum" to provide feedback on the equipment and operational characteristics of their facilities.. A number of their comments are included in the discussions on the various operations provided in the following sections. The comments relate to experience at nine of the basins, as the River Rouge basin was not operational at the time of the forum.

Operations Evaluated:

Monitoring Program

Detailed plans for performance monitoring were developed in collaboration between the owners of the facilities, the MDEQ, and representatives of the Rouge Project. The purpose of the monitoring plan was to assess the relative effectiveness of the different basin sizes, storage configurations, treatment technologies and operational practices used in the facilities. The evaluation program was complex, and it was a significant factor in the operation and maintenance requirements for the basins in the first two years of operation. Details of the full monitoring program can be found in "CSO Basins: Getting..." Comments received from the operators on flow monitoring and sampling are detailed in the above report.

Disinfection

All facilities are designed to meet NPDES effluent limits for fecal coliform of 400 cts / 100 ml and an effluent goal for total residual chlorine (TRC) of 1 mg/L. Operational control is based on monitoring the influent flow and effluent TRC.

Disinfection is achieved using liquid sodium hypochlorite. Details of the disinfection process used can be found in "CSO Basins: Getting..."

There are several operational problems that have arisen with the chlorination systems. For example, first, there has been more rapid degradation of the strength of the stored sodium hypochlorite strength between storm events than anticipated. Second, there is the tendency for the flow meters to under-record flow rates at low flows as the water level is rising. As a result, this causes too little hypochlorite to be applied. Third, there was a problem getting accurate measurements of effluent TRC. The report "CSO Basins: Getting..." contains comments received from operators on the disinfection system and its operation.

Screening

All facilities were designed to provide screening of the CSO basin flows. Successful removal of sanitary material is defined by MDEQ's criteria as the removal of sanitary trash with a size of 4 mm or greater. Subsequent studies of the effluent of the facilities have shown that the sanitary trash removal goals are met for the facilities evaluated.

The size of screenings collected ranged from sanitary trash to tires. The Wayne County and Detroit facilities rely on influent mechanically cleaned bar screens with spacing that varies from 0.5-inch to 1.5-inch. The Oakland County facilities use effluent static upflow screens with 0.75-inch spacing. Following the conclusion of a wet weather event, the material that has accumulated on the screen is sent to the interceptor with the dewatered flow. The dewatering wet wells for the Oakland County facilities also have bar screens. Operators try to force the screenings through to the interceptor. The retained screenings are scraped and disposed to a landfill.

In general, the operators have been relatively satisfied with the screening equipment and operation of the screens. The report "CSO Basins: Getting..." contains comments received from operators on the screening system and its operation.

Dewatering

The basins are dewatered after an event, when capacity is available in the receiving interceptor. The Wayne County facilities dewater by gravity, with the rate controlled by a control valve. A meter provides information in the downstream interceptor. The Oakland County basins are dewatered by pumping. A SCADA system is used to monitor levels in the receiving interceptors as well as total discharge rate at their point of connection to the DWSD system to ensure that peak contract capacity is not exceeded. The Detroit basins are pumped, though the top 9-feet of the Hubbell-Southfield facility is dewatered by gravity. Level indicators are provided in the downstream interceptor. Significant maintenance issues include the need to visit sites regularly, reliability of the SCADA system, and maintaining the dewatering pumps and the sump pumps.

The time required for basin dewatering ranges from 5 to 45 hours. The time required depends on the volume stored, the interceptor capacity available and in some cases also depends on the dewatering pumping capacity.

Flushing

Eight of the Rouge Phase I CSO basins use tipping buckets to remove settled debris after an event. The ninth and largest facility, Hubbell Southfield Basin, uses flushing nozzles. In some locations, there are auxiliary hoses to clean debris from wet wells and channels.
In general the tipping buckets are proving to be adequate. In some cases, where there is a long flushing length to clean, the process needs to be repeated several times and manual cleaning of some corners and walls is required.

The flushing nozzle system installed in the one facility has been a disappointment. The pressure maintained in the system has not been adequate to scour deposited solids from the floor. These nozzles are located near the ceiling of the facility. Older CSO facilities operated by Wayne County and Oakland County also use flushing nozzle systems. Both of these agencies are satisfied with the performance of these systems. The Wayne County facility has the nozzles mounted near the floor, and the Oakland County system has jets that are intended to flush debris to a center trough, where the bulk of the material is removed manually.

The report "CSO Basins: Getting..." contains comments received from operators on the flushing process and its operation.

Control Systems

The CSO facilities have varying degrees of control systems. The control systems help to determine the status of various components of the facility and the adjacent collection system, as well as various degrees of remote monitoring and operation.
The ability to remotely monitor the facilities is critical as this helps to determine when staff needs to be mobilized. At this time, remote operation of the facilities during an event is not considered feasible, regardless of the sophistication of the control system. Experience to date demonstrates that staff must be present during events for proper operation.
The report "CSO Basins: Getting..." contains comments received from operators on the control system and its operation.

Lessons Learned on Evaluation Question 3

Operational lessons learned from the Rouge Phase I CSO basins include the following:

1. 1. Protocols for items such as chemical feed, flow monitoring and sampling are often different for smaller wet weather events. As the vast majority of events will be small, the facility design should also recognize the operational differences for small storms.
2. Sodium hypochlorite disinfection systems need to address proper materials for long life, system flushing, testing of stored solution, and flow pacing of chemical addition.
3. Complex hydraulics, intermittent events and highly variable flow rates made the use of high tech flow metering equipment problematic. The best flow measurement results were achieved by metering with Parshall flumes and by calculating flows from pump capacity and run times.
4. Process flow control needs to be able to respond to highly variable flow rates. For example, a static weir for flow splitting is preferable to meter/ valve arrangements. Additionally, gate-closing action may not be quick enough for wet weather flow variations.
5. Regardless of level of automation, facilities generally require staff presence during wet weather events.
6. Start-up and testing of any CSO basin will likely take at least two years to fine tune, depending on the frequency of overflow events. Because most overflows from the combined sewer system were contained in the basins, there were very few events (e.g., 2 to 4) each year through which operators were able to develop experience with the flow-through operation of the facility.

Evaluation Question 4: What is the Proper Size for CSO Basins To Comply With Regulatory Requirements?

The Rouge Project was initiated in part to answer this very question. Since wet weather control is