SWS Contract Report 452 DEVELOPMENT OF DESIGN CRITERIA FOR SIDESTREAM ELEVATED POOL AERATION STATIONS by Thomas A. Butts Prepared for and in cooperation with The Metropolitan Sanitary District of Greater Chicago September 1988 DEVELOPMENT OF DESIGN CRITERIA FOR SIDESTREAM ELEVATED POOL AERATION STATIONS by Thomas A. Butts Illinois Department of Energy and Natural Resources State Water Survey Division Water Quality Section P.O. Box 697 Peoria, Illinois 61652 Prepared for and in cooperation with The Metropolitan Sanitary District of Greater Chicago September 1988 CONTENTS Page Introduction 1 Historical perspective 1 Future considerations 2 Study objectives and scope . . . . 3 Acknowledgments 4 Weir aeration theory and concepts 5 Basic concepts 5 Flow considerations 6 Weir geometry 7 Basic structures 7 Tailwater pool design 8 Weir-face design 8 Receiving pool length 9 Water quality considerations'. 9 Methods and procedures 11 Pilot plant design 12 Operating procedures 15 Data reduction 16 Results 17 Effects of modes of operation 18 Design regression equations 19 Discussion 20 Summary and conclusions 24 References 25 Figures and tables 29 Appendices Appendix A. Chicago Pump Company hydraulic chart for discharge over rectangular sharp-crested weirs 51 Appendix B. Photographs showing various modes of operation ... 53 Appendix C. ANOVA results for b-values and r-values for flow ranges grouped by weir geometric designs 61 Appendix D. Data used in the stepwise regression analysis. ... 71 DEVELOPMENT OF DESIGN CRITERIA FOR SIDESTREAM ELEVATED POOL AERATION STATIONS by Thomas A. Butts INTRODUCTION The Chicago metropolitan area is located along the headwaters of the Illinois Waterway as shown on figure 1. The Illinois Waterway is special among the many streams and rivers within Illinois: it drains 43 percent of the state and small portions of Wisconsin and Indiana. During dry weather, its headwaters consist principally of treated Chicago area wastewaters diluted with flow diverted from Lake Michigan at the three locations shown on figure 2. The waterway is no longer a free-flowing stream; it consists of eight navigational pools extending over 327 miles between the Mississippi River and Lake Michigan (figure 1). Chicago-area treated wastewater flows are derived from approximately 5.1 million people and a large, mixed industrial base. The Metropolitan Sanitary District of Greater Chicago (MSD) operates treatment facilities that discharge an average of 1400 million gallons per day of secondary and tertiary treated sewage into 70.5 miles of constructed channels and "improved" natural water courses, as shown in figure 2. Historical Perspective Prior to 1900 most Chicago-area wastes were discharged to Lake Michigan via either the Chicago River or the Calumet River systems, which are shown on figure 2. In 1871, a deep cut was made between the Chicago River and the Illinois and Michigan (I & M) Canal as a means of flushing a significant portion of the wastes down the canal and eventually to the Illinois River at LaSalle-Peru, where the canal intersects the river. In most respects, this attempt to relieve the Chicago area of unsanitary water conditions was unsuccessful. Consequently, plans were soon formulated to dig what was to become known as the Chicago Sanitary and Ship Canal. This canal was to be bigger, deeper, and more hydraulically efficient than the old I & M canal. It was eventually completed, and on January 17, 1900, the first Lake Michigan water was released into the high-capacity Sanitary and Ship Canal. The Sanitary and Ship Canal is designed to handle a maximum flow of 10,000 cfs. However, in 1913, the United States filed the first of a long succession of suits designed to limit total diversion well below this. This suit requested a diversion limitation of 4167 cfs, and a U.S. Supreme Court decision was rendered in 1925 upholding this request. This constraint prompted the MSD to eventually construct three major treatment facilities to prevent downstream water quality deterioration. The locations of these plants are shown on figure 2. Since 1925, through legal suits and subsequent court decisions, several changes have occurred in the amount of diversion allowed and the way it is to be administered and controlled. In 1930, the annual average diversion was set at 1500 cfs exclusive of municipal water supply needs. Between 1 January 31, 1957, and June 12, 1967, flows up to 8500 cfs were allowed. However, after June 12, 1967, total diversion was limited to 3200 cfs on an average annual basis, with a 5-year accounting period used. Present policy dictates that: 1) The regulation of discretionary diversion (direct wastewater dilution needs) and stormwater runoff flow is the responsibility of the Illinois Department of Transportation, Division of Water Resources (DWR). Prior to 1967, the MSD was responsible for regulating these activities. 2) A 40-year accounting period is to be used for computing the 3200 cfs average annual diversion, as opposed to the previously set 5-year period. 3) Discretionary diversion is set at a maximum of 320 cfs on an average annual basis. 4) The accounting year runs from October through September. Previously it ran from March through February. The new period coincides with the U.S. Geological Survey standard "water year." DWR encourages municipalities and subdivisions to fully utilize the flexibility of the 40-year averaging period. Diversion for navigation- related operations is limited to 255 cfs, including 130 cfs (40-year period) for lockages, 30 cfs (40-year period) for lock leakages, and 95 cfs (5-year period) for navigational makeup. The 320 cfs direct discretionary diversion allotment that is presently being used for water quality enhancement during summer months is scheduled to be reduced to 101 cfs on October 1, 2000. By this date, Phase I of the MSD Tunnel and Reservoir Project (TARP) and instream aeration projects are projected to be completed and will provide improvements in water quality. Future Considerations The MSD treatment plants are well operated and produce good effluents. Butts et al. (1983) conducted a computer model study of the effects of effluents from the three major plants (figure 2) on downstream dissolved oxygen (DO) and ammonia concentrations. The results showed that upgrading of only the Calumet treatment plant is warranted, and that if this were to be done, only modest improvements in downstream DOs would be realized during dry, warm-weather conditions. The minimum DO standard set for the Chicago-area watercourses shown in figure 2 is 4.0 mg/1 (Illinois Pollution Control Board, 1986). This is probably an unrealistically high standard for present application in that waterway physical characteristics limit the assimilation of even low levels of oxygen-consuming wastes. During low flows, long residence times occur, allowing much dissolved oxygen to be used or depleted through the stabilization of dissolved biochemical oxygen demand (BOD) and sediment oxygen demand (SOD). The deep, slow-moving water facilitates sediment deposition and the creation of very high SODs. Discretionary diversion from Lake Michigan during critical low-flow periods helps reduce the severity of the DO depletion somewhat since high-DO water is introduced into the system and detention times are reduced. However, several reaches, particularly 2 along the Calumet Sag Canal and the Sanitary and Ship Canal, periodically exhibit DO concentrations below 2.0 mg/1. To help alleviate these severe DO depletions, the MSD presented testimony to the Illinois Pollution Control Board (IPCB) in the early 1970s regarding the use of instream aeration in conjunction with wastewater treatment. The IPCB accepted the MSD's proposal, and two instream aeration stations were constructed in 1979 and 1980. These two stations are located at Devon and Webster Avenues on the North Shore Channel (figure 2) and use compressed air, distributed via bottom diffusers, to transfer oxygen to the water column during critical periods. Construction costs along with routine operation and maintenance costs associated with the establishment and running of these stations have been higher than anticipated. However, the results achieved from the two operating systems indicate that significant localized improvements in DO can be realized by using some form of instream aeration. Consequently, the instream aeration concept need not and should not be abandoned. The use of sidestream elevated pool aeration stations has been included in recently proposed regulatory changes now before the IPCB. These changes involve revising the effluent standards relative to the MSD's three major treatment facilities, upgrading two sections of waterway to a "General Use" designation, and revising the waterway DO standard in the Cal-Sag Channel from 4.0 mg/1 to 3.0 mg/1. On the basis of the criteria established by these proposed regulatory changes, a system of sidestream elevated pool aeration (SEPA) stations has been proposed as an alternative to the previously constructed compressed air systems. This system will allow low-DO water to be withdrawn from the waterway by means of spiral pumps. The water will be routed through an elevated pool and passed over a weir or series of weirs to be aerated before being routed back to the waterway downstream of the intake point. The concept, as originally envisioned, is presented in figures 3 and 4. The DO concentration of the sidestream will be raised sufficiently to provide an overall DO concentration of 3.0 mg/1 or higher in the receiving stream. The SEPA stations will operate only during critical periods when the DO concentrations in the waterway fall below 3.0 mg/1. Computer model studies conducted by the MSD, which have simulated DO conditions throughout the waterway system, indicate that SEPA stations are needed at five critical locations along the Cal-Sag Channel (figure 5). Each station will have the capability of withdrawing approximately 50% of the total flow from the waterway. Study Objectives and Scope The artificial reaeration of large streams through use of the SEPA concept has never been attempted before. Even design criteria for the reaeration of wastewater effluents and small streams based on weir aeration concepts are limited. Additionally, some of the limited published information is misleading and/or incorrect, and if not properly screened and/or utilized, it could provide unexpectedly poor results in the final analysis of any major large-scale SEPA-like project. 3 The Water Quality Section (WQS) of the Illinois State Water Survey (ISWS) has performed numerous weir and spillway aeration studies during the last decade. These studies have involved both controlled and uncontrolled field studies and controlled laboratory studies. Much practical information has been gleaned from this work. A good understanding of basic weir-aeration theory has been achieved, and concepts have evolved that can be helpful in designing efficient weir-aeration systems. The primary purpose of this endeavor was to perform a full-scale, on-site weir aeration pilot study to verify selected design criteria previously established by the ISWS and others. The verified data were to be used to develop practical engineering design procedures and equations for use in designing economical and efficient SEPA stations. Specific input by the WQS of ISWS consisted of: 1. Designing the weir system to be studied. Schematic diagrams of the basic designs to be evaluated were provided. 2. Providing information on desirable sampling methods and sampling frequencies, based on data generated during previous WQS weir-aeration studies. 3. Providing computer services for data storage, data reduction, and data analyses. The WQS has developed efficient computer programs designed specifically for handling the type and magnitude of data generated during this study. 4. Disseminating data and information to the MSP on a timely basis so they could proceed with preliminary and final SEPA station design. This report constitutes a formalization of previously disseminated data and information. The MSD was responsible for developing detailed construction plans of the full-scale weir and for constructing it on-site. MSD personnel conducted the daily experiments and periodically sent the results to the WQS laboratory in Peoria for review and comments. ISWS personnel inspected the weir setup before the initial startup and once during an actual experimental run. Acknowledgments This study was partially funded by a grant from the Metropolitan Sanitary District of Greater Chicago, Frank Dalton, General Superintendent. The research was conducted as part of the work of the Water Quality Section of the Illinois State Water Survey, Richard G. Semonin, Chief. Grant money was administered through the MSD's Research and Development Laboratory, Cecil Lue-Hing, Director. David R. Zenz, Coordinator of Research at the Research and Development Laboratory, was responsible for initiating the project at the urging of Bill Macaitis, Assistant Chief Engineer in the MSD Collection Facilities Division, who envisioned the SEPA station concept as a practical alternative to diffused-air, instream aeration. 4 Special recognition is given to Bernard Sawyer and Gilbert Elenbogen of MSD's Research and Development Laboratory, who made the study possible through their careful supervision of the daily experimental runs. Their dedication to the project was exemplary. Dana Shackleford and James Marsa of the WQS were responsible for much of the data handling and most of the computer outputs. Word processing of the report manuscript was performed by Linda Johnson. WEIR AERATION THEORY AND CONCEPTS The fact that aeration occurs at weirs, dams, spillways, and waterfalls is readily apparent as evidenced by the white-water turbulence that normally appears below such structures. However, the mechanism by which this aeration occurs has never been clearly or fully defined, especially in terms of practical engineering design concepts and parameters. Basic Concepts Gameson (1957), in some original weir and dam aeration work, proposed the use of an equation involving both theoretical and rational concepts that relate water fall height, water temperature, structural geometry, and water quality to a factor defined as the deficit ratio, r. The definition of r is: where Cg is the DO saturation concentration at a given temperature and C A and Cg are, respectively, the DO concentrations above and below the dam or flow-release structure. Although equation 1 is simple, it serves to illustrate two principles important to dam aeration concepts. First, it demonstrates that the upstream DO concentration dictates the rate of oxygen exchange at any weir or dam. Second, for a given set of water and temperature conditions, higher ratios reflect higher aeration efficiencies. Relative to the first concept, Gameson (1957) and Gameson et al. (1958) found in laboratory experiments that the ratio is independent of above-dam DO concentrations of Cg + 10 mg/1. However, data collected by Barrett et al. (1960) indicate that this independence may be reduced to Cg ± 4 mg/1 for full-sized field structures. The original dam aeration formula (Gameson, 1957; Gameson et al., 1958) relating temperature, water quality, dam cross-sectional design, and differential water levels to the deficit ratio has been modified and refined and appears in the following form (Water Research Centre, 1973): where a is the water quality factor; b is the weir, spillway, or gate coefficient; h is the static head loss at the dam (i.e., upstream and 5 downstream water surface elevation difference) in meters; and T is the water temperature in °C. This equation can be used to model the relative and absolute efficiencies of a weir spillway or flow-release structure by determining specific values of 'b.' Every spillway or gate has a specific coefficient, but generalized categories can be developed in reference to a standard. The standard weir (b = 1.0) is by definition a sharp-crested weir with the flow free-falling into a receiving pool having a depth equal to or greater than 0.1h + 6 cm. An idealized step weir (a series of sharp-crested weirs) has a b-value of 1.9 (Water Research Centre, 1973); however, actual field-measured values are usually lower. The formula was developed by British researchers from data collected at many relatively low-head channel dams and weirs transecting small streams. Good reproducibility can be achieved when h does not exceed 3 to 4 meters, the maximum height of the dams at which data collections were made during development of the formula. In addition, close examination of the equation reveals that the factor (h)(l - O.llh) mathematically restrains the use of the equation to heights of 4.55 meters or less. The water quality factor (a) has to be evaluated experimentally in the field or estimated from published criteria. Refinements of Gameson's (1957) early categorization of a-values are: grossly polluted water, a = 0.65; moderately polluted, a = 1.0; slightly polluted, a = 1.6; and clean water, a = 1.8. These values are based on a minimal amount of field and laboratory data. Their direct applications are subjective, and since considerable numerical latitude exists between values, significant errors can result. Flow Considerations Conspicuously missing from equation 2 is a flow-related variable. Considerable divergence of opinion currently exists concerning the effect flow rate changes have on weir or dam aeration. Some researchers have concluded that weir aeration is affected very little over a wide range of flows, while others have presented data directly contradicting this. Barrett et al. (1960) conclude, as a result of numerous field studies of channel dams and weir installations, that the omission of a flow-rate factor in equation 2 does not appear to greatly affect its validity. British researchers (Water Research Centre, 1973) indicate that a 3.5-fold change in flow produced no significant change in "r" for a step-weir in a river. Butts and Evans (1980) concluded that a two-fold range in flow did not affect aeration in small-scale laboratory experiments using a V-notch weir and water falls up to one meter. In contrast, Apted and Novak (1973) state that, "Contrary to previous thought, the oxygen uptake is seen to be dependent upon the discharge. This is a result of an increase in jet penetration with an increase in discharge (as shown by Hausler [1961])." By plotting flow rates versus deficit ratios, Apted and Novak show that an inverse relationship exists between flow and oxygen uptake. Tebbutt (1972) indicated that the reaeration efficiency of a cascade decreases slightly with increased flow in laboratory-scale experiments. 6 Several other researchers, according to Nakasone (1987), have shown that aeration efficiency increases with discharge up to a certain point and then decreases with additional increases in discharge. This contention is supported somewhat by field observations reported by Butts and Evans (1978); they concluded that the aeration capacities of channel dams and river spillways are very sensitive to flow. Full-scale installations appear to be most efficient toward the high end of the low-flow spectrum, with efficiencies subsequently dropping off slowly but gradually at flows below and above the optimum. Mastropietro (1968) presents dam reaeration formulas and computational methodologies that incorporate flow as the most important independent variable. Tebbutt et al. (1977) conducted laboratory, step-weir aeration studies and concluded that the main factors controlling reaeration performance are flow per unit width and total fall, with secondary influences being related to individual step height and the overall slope of the step-weir system. Butts and Adkins (1987) expanded upon Butts and Evans' (1980) work using a V-notch free-falling weir. They found that increasing the flow range 10-fold from the previous 2-fold experimental range produced small but statistically significant differences in the reaeration rate as measured by "r." However, for actual field conditions, for which unit flow rates were much higher, Butts and Adkins (1987) found that flow rate ranks only sixth out of a total of 13 independent variables statistically equated to the deficit ratios determined for the flow-release gates at the Starved Rock dam on the Illinois River. Additional experimental work is needed to better define the role flow rate plays in full-scale dam and/or weir aeration efficiencies. Weir Geometry Structural configuration (geometric design) of a weir or spillway is a second factor affecting aeration efficiencies that has evoked some contradictory conclusions in the literature. The dam or weir coefficient "b" in equation 2 is an all-inclusive factor taking geometric factors into account. Certain basic weir designs have been clearly established as superior aerators. Basic Structures British researchers (Water Research Centre, 1973) have defined a standard weir as a sharp-crested, free-falling weir having a b-value of unity as referenced to equation 2. Spillways and weirs that are not free-falling, such as Ogee spillways and sloping-face structures, usually produce b-values less than 1.0, whereas stepped weirs (as opposed to cascades) produce b-values greater than 1.0. A cascade is defined as a spillway consisting of a large number of small steps with no significant receiving depth below each step. For a weir to be considered a step weir, each step must be followed by a receiving pool; the flow cannot merely splash onto a flat surface as it passes to successively lower levels. The theoretical b-value for a 4-step weir having a total fall of 2 m and an r-value of 1.304 can be shown to be 2.89 (Water Research Centre, 1973). Cascades usually produce smaller b-values than those for sharp-crested, free-falling weirs with deep receiving pools. Jarvis (Water Research Centre, 1973) studied 13 cascades and reported that 7