This paper will present the development of a clearwell design that would meet the disinfection goals while also providing the plant with much needed operational storage. Various site constraints and design criteria that needed to be addressed while meeting the dual requirements for disinfection and operational storage will be discussed. These will include:

• varying demands throughout the year
• physical site limitations
• maximizing use of existing clearwell storage
• minimizing headloss conditions to reduce excavation costs
• limiting chlorine residual levels to avoid exacerbating finished water taste and odour
• ceasing prechlorination during periods of high source water organics levels
• ensuring disinfection byproducts would not exceed the established goal
• additional operational storage to provide the operator adequate time to react to variables

The new AEP Guidelines for disinfection which now focus on removal of protozoan cysts and viruses will be presented as they apply to the Glenmore WTP.

On the basis of the required operational storage and disinfection CT, the size, arrangement and T₁₀ for the clearwell will be discussed and how the varying demand during different seasons of the year was taken into account in arriving at the final sizing and depth of the clearwell.

The opportunity to reduce construction costs and the analyses that lead to the final design will also be presented.

KEYWORDS
Clearwell, Disinfection, Chlorine, Giardia, CT, Water Quality

INTRODUCTION
The Glenmore Water Treatment Plant, constructed in three phases in 1933, 1957 and 1965, is a conventional treatment plant comprising coagulation, hydraulic flocculation, crossflow sedimentation, filtration and disinfection. It obtains its water from the Elbow River impounded by the Glenmore Dam.

In 1992, the City of Calgary commenced the Glenmore Water Treatment Plant Upgrade Study based on goals identified in a scoping study completed the previous year. The study, which included a comprehensive pilot plant evaluation of several unit processes, chemical coagulants, disinfectants and operating procedures, resulted in a number of recommendations contained in a four phase upgrading plan. Phase 1 provides basic upgrading of the existing plant to improve its reliability and to enable it to meet current and foreseeable regulatory requirements related to water quality. Phase 2 will expand the plant's capacity by 200 ML/d, or more as deemed appropriate. Phases 3 and 4 will address alternative disinfection and waste management respectively when these issues arise.

Construction on Phase 1 improvements commenced in 1996 with the addition of rapid mix facilities ahead of the flocculators, replacement of filter media in all twenty four filters and piping additions to allow a filter-to-waste stage in the filter backwashing procedure. All three projects are designed to assist in meeting future water quality goals. They do not, however, constitute a departure from previous treatment processes and would have been required in the normal course of time without regard to changing regulatory requirements.

A major change in regulatory requirements for disinfection of potable water supplies is contained in the Standards and Guidelines for Municipal Waterworks, Wastewater and Storm Drainage Systems recently introduced by Alberta Environmental Protection (AEP). The disinfection guideline, based on the Surface Water Treatment Rule (SWTR) and the interim Enhanced Surface Water Treatment Rule (ESWTR), both promulgated by the United States Environmental Protection Agency (USEPA), embodies the concept of the product of disinfectant residual concentration and contact time, commonly termed CT. Compliance with the new guideline will require the City to construct a large clearwell for filtered water storage on the Glenmore Water Treatment Plant site. The clearwell and its accompanying treated water pump station represent a substantial capital investment and significant change to how the plant has previously been operated.

WATER QUALITY
General
The object of the Phase 1 improvements, of which the proposed clearwell is an integral part, is to ensure continued high quality water production including the removal and/or inactivation of cysts, pathogenic bacteria and viruses and the reduction in levels of turbidity and organic matter. These water quality parameters and regulations governing them, must be taken into account in developing design concepts for the clearwell and associated pump station.

Alberta Environmental Protection (AEP) Guidelines
Potable water quality has to meet the requirements of the Standards and Guidelines for Municipal Waterworks, Wastewater and Storm Drainage Systems, recently updated by AEP. These guidelines require, as a minimum, that surface water be treated by chemically assisted rapid sand filtration and disinfection. As defined by the guidelines, conventional treatment meeting this requirement includes chemical mixing, coagulation, flocculation, solids separation and filtration; the unit processes employed at the Glenmore Water Treatment Plant. Disinfection may be achieved by chlorine, chlorine dioxide, chloramines, ozone or a combination of one or more of these disinfectants. The primary aim of such treatment is the production of water that is free of pathogenic organisms, bacteria and viruses. Secondary aims are palatability and aesthetic appearance.

Treatment of surface water in accordance with AEP guidelines will be considered satisfactory based on specific levels of Giardia removal/inactivation directly related to known concentrations of cysts in raw water. Relatively unpolluted raw water with a cyst concentration less than 1 per 100 L will require 3 log reduction/inactivation, while raw water with a cyst concentration greater than 100 per 100 L will require in excess of 5 log removal. Table 1 reproduces a similar table contained in the AEP guidelines.

This level of Giardia removal/inactivation is higher than that previously required by most regulatory agencies in North America, where 3 log removal was normally mandated regardless of the background level of cysts in the raw water. It reflects the growing concern over the presence of Giardia and Cryptosporidium in raw water, particularly to which cattle and other animals have ready access. No easy and reproducible method of detection and enumeration presently exists for Cryptosporidium oocysts. Furthermore they are more difficult to remove than are Giardia cysts. AEP, following the lead set by the USEPA in its interim ESWTR, has established the higher removal/inactivation levels for Giardia, believing they will provide greater protection also against Cryptosporidium. There has been some discussion about establishing specific Cryptosporidium levels in the final ESWTR, to be promulgated in the year 2000. In view of the difficulties being encountered in collecting relevant information on this organism, including methods of detection and enumeration, this is not likely. AEP has stated that Giardia removal/inactivation will remain the sole criterion for determining satisfactory protection against cyst contamination for the foreseeable future.

AEP recognizes that Giardia cysts are removed in a well operated conventional water treatment plant. Accordingly, a credit for 2.5 log reduction of Giardia will be given for such a plant, excluding disinfection, if it meets the following turbidity reduction limits:

• When source turbidity is greater than or equal to 2.5 NTU, the filtered water turbidity shall be less than 0.5 NTU in at least 95% of the measurements made each calendar month.

However, the filtered water turbidity may exceed 0.5 NTU but not exceed an upper limit of 1 NTU in more than 5% of the measurements made each calendar month. The filtered water turbidity shall not exceed 2 NTU at any time.

• When source turbidity is less than 2.5 NTU, the filters shall achieve either:
  • An 80% reduction in source turbidity based on an average of the daily turbidity reductions measured in a calendar month, or
  • a filtered water turbidity of less than or equal to 0.1 NTU.

Continuous monitoring of actual cyst or cyst-sized particles will not be a prerequisite for this credit. Although the AEP guidelines do allow for greater than 2.5 log Giardia reduction credit to be granted for a filtration system (AEP Standards and Guidelines 2.2.4 and 4.1.5.1 (7)) this would require the use of particle counting or microscopic particulate analysis data. The collection and use of such data would be onerous and expensive but could be worthy of future consideration.

Protozoans
The actual Giardia log reduction/inactivation required is related to the geometric mean of the cyst concentration in the raw water. It will be subject to negotiation between the City and AEP when the City renews its licence to take water from the Elbow River. Based on the City's records of cyst concentration in the Glenmore Reservoir water, Figure 1 and on discussions with AEP representatives, the Glenmore Water Treatment Plant will be required to achieve in the order of 4 - 4.5 log reduction/inactivation of Giardia. Of this amount, the plant should be granted a 2.5 log removal credit for chemically assisted filtration, leaving 1.5 - 2 log inactivation (4 - 4.5 log minus 2.5 log) to be achieved by disinfection.

It is worth noting that the stringent water quality requirements, particularly the need to remove protozoans such as Giardia and Cryptosporidium, are a direct consequence of contamination of the watershed upstream of the Glenmore Dam. The cost of complying with these requirements, both in the capital construction cost and in the plant operating cost, is high. This high cost of treating water to ensure that it is safe for public use mandates serious consideration of watershed protection measures.

Turbidity
In accordance with the AEP guidelines, credit for Giardia cyst removal in the treatment process will be solely dependent on achieving the turbidity reduction required by these guidelines. Experience has shown that, under most conditions, the Glenmore plant can achieve the turbidity reduction levels required, albeit, at times, at less than the plant's design capacity. The plant improvements already completed under Phase 1 are expected to allow the turbidity reduction levels to be achieved under all raw water conditions at a production rate of 300 ML/d.

Natural Organic Matter
Natural organic matter (NOM) in the raw water as indicated by the level of total organic carbon (TOC) varies seasonally, with a higher level in spring and summer particularly during run-off. During these periods of higher TOC about half is in soluble form and is relatively unaffected by chemical coagulation, passing straight through the pretreatment and filtration stages. The particulate fraction of the TOC is removed in the treatment process. Because trihalomethanes (THM's), which can be carcinogenic in prolonged dosages, are formed when organic matter is chlorinated it is prudent to delay chlorination until after filtration, i.e. until much of the TOC has been removed. The proposed clearwell will provide contact time for post filtration primary disinfection. Until that time, however, contact time provided by the sedimentation and filtration stages is essential for effective disinfection with chlorine.

Notwithstanding prechlorination, the level of total trihalomethane (TTHM) formed at the Glenmore plant is well below the Guidelines for Canadian Drinking Water Quality (GCGWQ) interim maximum acceptable concentration (IMAC) of 100 g/L. There is however, little room for complacency in this fact for two reasons; the level of TOC is likely to continue increasing due to agricultural and other activities in the watershed, and future TTHM allowable limits will probably be further reduced, reflecting the public's growing concerns about potentially harmful disinfection byproducts in treated water. The first factor, once again points to the need for watershed protection while the second reinforces the need for continuing vigilance in water treatment operations and further research into alternative methods of disinfection.

WATER QUANTITY
Plant Capacity
The Glenmore Water Treatment Plant was designed with a maximum capacity of 640 ML/d. It has been unable to achieve this production rate and meet acceptable water quality standards however, for the past number of years. This is due in part, to the increasingly stringent water quality standards compared to those prevailing at the time the plant was built. The plant's nominal capacity of 640 ML/d cannot be achieved even when raw water conditions are good. Under poor raw water conditions the plant production rate must be severely curtailed in order to comply with current water quality goals. The Phase 1 improvements are intended to ensure that the plant will be capable of meeting present and foreseeable future quality requirements under all raw water conditions at a capacity of 300 ML/d. The 300 ML/d rate appears to be the limit for particulate removal under the most difficult to treat conditions for the existing sedimentation and filtration components. Much higher rates of production will be possible under more favourable raw water conditions, and for this reason, the Phase 1 plant improvements will be hydraulically designed for 500 ML/d.

Construction Staging
Improvements to the chemical mixing and filtration components in Phase 1 have been completed. Final design of the remaining components of the improvements, the clearwell and pump station, will be completed this year. Construction of Phase 2 improvements, which will add a further 200-300 ML/d to the treatment capacity, will be undertaken as water demand dictates and may be completed in two or more stages. Precisely when these improvements will be needed is of particular importance because the timing could affect some aspects of the clearwell and pump station design; specifically the elevations of these two structures.

Clearwell & Pump Station Elevations
The top water level for the Phase 1 clearwell was established to allow gravity flow into it from the future Phase 2 filters. The pump station elevations were determined, in turn, by the clearwell elevations and other hydraulic considerations. Provision for gravity flow from Phase 2 requires that the clearwell floor elevation be approximately 2.0 m lower than it would need to be if the clearwell were to be used only for Phase 1, i.e. for the existing plant. Ozonation facilities, should they be constructed as part of the Phase 2 improvements, would necessitate construction of an intermediate pump station between the ozone contact chamber and the filters.

Should the Phase 2 improvements not be required within the next 10-15 years, there would be little justification in building a clearwell at the lower elevation needed to accommodate gravity flow from this future expansion. The increased excavation costs are significant and, even without a rigorous economic analysis of sunk costs versus future operating costs, suggest that intermediate pumping, when required, would be a wiser choice. The likelihood of ozonation in the future with its need for associated pumping facilities supports this alternative. For these reasons, projections of population growth and water demands were revisited during the Clearwell and Pump Station Concept Review of January 28, 29, 1998. The purpose of this last review was not so much to establish the size of treatment components as to determine when they were most likely to be needed.

Water Demand Projections
Water supply projects, including intakes, pump stations and treatment plants, have traditionally been designed to serve maximum day demands; distribution system reservoirs being provided to accommodate hourly demand fluctuations, fire demands and other emergency requirements. These water demands are estimated using projected population numbers, per capita consumption and daily and hourly peaking factors, among other things. The projected peak day and average day water demands for 1997 - 2019, using the City's population projections and per capita water demands are shown in Figure 1. Superimposed on the peak day water demand line is the Glenmore Water Treatment Plant capacity, including Phases 1, 2a and 2b, and the Bearspaw Water Treatment Plant capacity.

Figure 1. Water Demand Forecasts and Proposed Plant Phasing

Phase 2 Implementation
Figure 1 shows, with the Glenmore plant producing a minimum of 300 ML/d and assuming full production from the Bearspaw plant, there should be little difficulty in meeting most peak day water demands for the next 10-12 years, perhaps longer. Moreover, peak day demands tend to occur during prolonged dry periods when the most difficult raw water conditions have passed. At these times production from Glenmore could be increased above 300 ML/d. There is always the possibility however, of peak day demands occurring earlier in the year, when such an increased production rate might not be possible.

As stated earlier, water treatment plants have traditionally been designed to meet peak day demands. These peaks in the case of most larger North American communities, including Calgary, are caused mainly by lawn watering. It is a matter for debate whether we should continue to spend large sums on capital projects dictated by the need to provide water for unrestricted lawn irrigation. Communities with limited water resources, such as those relying on groundwater, often curtail lawn watering during summer months as a matter of course. The City of Calgary has had good response in the past when it has appealed for water use restraint during potential shortages. Provided such appeals are not too frequently made, it is likely that any shortfall in production during peak demand times could be offset by a reduction in lawn watering.

It would appear that the Phase 2 expansion to the Glenmore plant is most likely to be required after the year 2012 or thereabouts. That being the case, the clearwell elevations should be established to enable gravity flow from the existing plant without regard to Phase 2.

Water Demand Forecasts and Proposed Plant Phasing

DISINFECTION
Introduction
The City of Calgary employs chlorine for disinfection at both its water treatment plants, Glenmore and Bearspaw. At Glenmore sufficient chlorine is added to the raw water prior to coagulation to maintain a free residual throughout the treatment process.

Additional chlorine is dosed after filtration to ensure a free residual of 0.6-0.7 mg/L entering the distribution system. Prior to 1992, free residual entering the distribution was maintained at a lower level, 0.2-0.3 mg/L, to avoid chlorinous taste complaints. This lower level however, was insufficient to guarantee a free chlorine residual throughout the system and therefore increased to its present level.

An increased number of taste and odour complaints noted in 1992 were largely due to raising the chlorine level in the distribution system; a common experience whenever a change of this nature is made to the disinfection procedure. There was a noticeable decrease in taste and odour complaints in 1993 and subsequent years, indicating widespread acceptance of the slight increase in chlorine taste at the tap.

Disinfection Options
The more stringent Giardia removal/inactivation requirements of the new rule will directly affect the operation of the Glenmore Water Treatment Plant and largely determine the size of the clearwell to be provided under the Phase 1 improvements.

Two possible technologies exist to meet disinfection goals, namely chlorination or ozonation. Chlorination is a proven process already practised at the plant and familiar to operators. In conjunction with the proposed, new clearwell it can meet the disinfection goals in a cost effective manner.

Ozone is a highly effective disinfectant and requires only short contact times, usually in the order of 10 minutes. Consequently, its use would obviate the need for a large clearwell. However, its use presents the following significant challenges:

Ozone should be applied prior to filtration so any ozone byproducts such as aldehydes and Assimilable Organic Carbon (AOC) are absorbed by biological activity within the filters. If applied upstream of clarification, the ozone system would need to be larger, as raw water exhibits a significantly higher ozone demand than clarified water. If applied downstream of filtration there would be an increased potential for bacterial regrowth within the distribution system. Chlorination would still be required for secondary or residual disinfection, and there is unlikely to be any substantial reduction in the annual consumption of chlorine.

There is no tolerance within the existing hydraulic grade to accommodate ozonation unless intermediate pumping is also employed. Furthermore, as a Cryptosporidium goal was unlikely in the foreseeable future, installing ozone now in anticipation of this possible, future goal offers few advantages. Lastly, net present value of capital and operating costs for intermediate pumping and ozonation are unlikely to be lower than those of the chlorination option. Consequently, it was concluded that the use of chlorination in conjunction with a new clearwell is the correct approach towards meeting current disinfection goals.

Trihalomethanes & Prechlorination
As discussed previously, any treatment strategy should ensure the removal of as much organic matter as possible prior to the addition of chlorine, to minimize the formation of trihalomethanes (THM's). The worst condition for THM generation occurs in the spring when higher levels of particulate and soluble NOM are experienced. With prechlorination, this can lead to significantly elevated THM concentrations.

Consequently, it was decided the clearwell should be of sufficient size to meet disinfection goals using post-filtration chlorination only during the spring and for flows up to 300 ML/d. By contrast, during winter and summer, NOM levels are much lower, in a more soluble form and relatively unaffected by chemical coagulation, so prechlorination should cause only minimal increases to THM concentrations in the treated water compared to the use of post filtration chlorination. Prechlorination, to some extent, was therefore acceptable during the winter when lower organic levels would result in lower THM concentrations. Sufficient contact time would be provided jointly within the pretreatment process and new clearwell to meet disinfection goals for flows up to 300 ML/d.

Clearwell Size
Both 40 and 50 ML clearwell sizes have been considered. 50 ML is the maximum practical size for the structure considering the site available, and it makes most efficient use of the site. Tables 2 and 3 indicate the available disinfection capacity of 40 and 50 ML contact capacity clearwells for various seasons, based on the following:

• A 2 log Giardia inactivation goal for the disinfection system.
• Clearwell T ₁₀ /T = 0.7.
• "C" = 0.7 mg/L of chlorine.
• A pre-chlorination T ₁₀ /T = 0.45 within the existing cross flow clarifiers and 0.7 in the filters.

Table 2 indicates that for a 40 ML clearwell:

• Prechlorination must be used for all flows above 170 ML/d in the winter and 225 ML/d in spring. To meet summer peak flows of 500 ML/d either "C" could be increased to 0.77 mg/L or a low level of pre-chlorination applied.

Table 3 indicates that for a 50 ML clearwell:

• Prechlorination must be used in the winter for all flows above 215 ML/d but "C" can be maintained at or below 0.7 mg/L.
• There is just sufficient capacity to meet a spring design flow of 300 ML/d without prechlorination, if "C" is raised to about 0.77 mg/L.
• There is sufficient capacity to meet summer peak flows of 500 ML/d with "C" at or below 0.7 mg/L and without prechlorination.

From these results it is concluded that a 50 ML contact capacity clearwell should be installed. A smaller capacity would fail to meet the disinfection system goals, particularly during periods of elevated organics in the spring.

OPERATIONAL STORAGE
In addition to providing the necessary contact capacity, adequate operational storage should also be provided in the clearwell. As the plant is manned continuously, it was established that operational treated water storage on site should provide a minimum of 30 minutes at the average design flow of 300 ML/d (6.2 ML). This would afford the operator sufficient time to react to an upset condition.

The existing clearwell located beneath the filters has a capacity of approximately 11.4 ML. This equates to a nominal 55 minutes of storage at 300 ML/d. The locations of the filter outlets and the difficulty of providing acceptable baffling make it impractical to utilize the existing clearwell for contact storage. However, it still offers valuable balancing storage and as much use as possible should be made of this asset.

Initially, it was proposed the Phase 2 plant hydraulic grade would dictate the minimum water level in the new clearwell. At a flow of 500 ML/d this would be 1091.69 m, only slightly above the existing clearwell floor elevation of 1091.20 m. The existing clearwell level would fluctuate in synchrony with the 0.5 m balancing level variations available in the new clearwell and so offered slightly less than 2.0 ML of additional balancing storage. See Figure 2. Consequently, the existing clearwell would be always nearly empty and little use would be made of the possible 11.4 ML of storage.

Consideration was given to providing a level control valve on the outlet of the existing clearwell that would allow the existing clearwell level to rise above the hydraulic gradient and provide approximately 9.5 ML of additional balancing storage as shown in Figure 2. The outlet valve would modulate based on new clearwell level. If the new clearwell level fell, the valve would open allowing balancing volume within the existing clearwell to be accessed.

However, there were some serious disadvantages to this. To fully utilize the existing clearwell storage would require leaving the new clearwell at its initial design elevation. This would be a very expensive way to provide operational storage. Also, placing complete reliance on a control valve on such a critical conduit was considered an unacceptable risk.

So the arrangement finally selected was to elevate the new clearwell as high as the hydraulic profile from the existing clearwell would safely permit. The new clearwell would have an extra 0.5 m water depth to provide approximately 4.5 ML operational storage which together with 2.0 ML storage for the same 0.5 m depth in the existing clearwell, would meet the required minimum.

Figure 2. Clearwell Elevation Options

SUMMARY
The new AEP Guidelines mandate a disinfection strategy based on the CT concept for removal/inactivation of viruses and protozoan cysts. For the raw water cyst concentrations at the Glenmore Water Treatment Plant, the governing goal is a 4.5 log removal/inactivation of Giardia cysts, with a 2.5 log credit for removal through the filtration process.

Based on sound economics, the decision to continue with chlorine for primary and secondary disinfection will require a 50 ML contact capacity clearwell to achieve the required 2.0 log inactivation of Giardia. Although arguments could be made for a bigger clearwell - to completely avoid prechlorination and reduce TTHM's - site constraints present significant costs for anything larger. Furthermore, by nature of the NOM being almost completely soluble for most of the year, moving the primary chlorine injection point downstream of the filters will have little impact on annual TTHM's.

The approach has therefore been to design the clearwell to meet the entire CT requirement during spring - when NOM is high and contains significant particulates that can be removed by filtration - with a small, 10% increase in the chlorine dose. During summer, CT can be met with the current, 0.7 mg/L, chlorine residual. For the winter, when NOM is typically very low, prechlorination will be used to supplement the available CT through the clearwell to achieve the overall CT requirements.

Costs were a significant factor in selecting operational storage. To maximize use of the existing 11.3 ML clearwell would require locating the new clearwell CT storage below the existing, and providing flow control between the two. This was considered too expensive - to locate the new clearwell so deep; and too risky to place complete reliance on a control valve for all operational storage. A total of almost 7 ML of operational storage will be provided - 2 ML in the existing clearwell and a little less than 5 ML in the new. This will offer over 30 minutes storage at a design flow of 300 ML/d; sufficient for the operator to react to a plant upset.

The strategy adopted for the design of this clearwell has been to ensure compliance with the disinfection regulations, while minimizing current capital expenditure. By utilizing the existing facilities, the clearwell size has been optimized. Its construction cost has been reduced by elevating it to the maximum hydraulic grade. It is recognized that the Phase 2 plant will need to include intermediate pumping to achieve this hydraulic grade, but in view of the anticipated 15 year horizon for Phase 2, capital expenditure now could not be justified.