The treated water quality given in the feasibility study needs to be updated taking into consideration the Lebanese standards and the latest amendments in various drinking water guideline standards i.e., WHO and EU. The feasibility study was completed in 1994; however in 1998 the European Council issued the new drinking water directive, 98/83/EC which provides a more detailed treated water consent under three different categories; microbiological parameters, chemical parameters and indicator parameters. Recently the third edition of the World Health Organization (WHO), Guidelines for Drinking Water Quality was released in 2008.
The previously recommended treated water quality targets will have to be revisited prior the start of the project. A combination of the national standards and that of EU and WHO standards are recommended for the Awali treatment scheme to derive a more comprehensive water quality than the one previously defined. Each parameter will have to be evaluated one by one on the basis of their effect on public health, implication on water transmission and public acceptability. The recommended treated water quality targets are compared to standards in Table 3 -32.
Table 3 32 Drinking Water Standards
Recommended in 1994 feasibility study
Lebanese national standards (guiding limits) (MoE Decision 52/1/1996
Lebanese national standards (max acceptable)
400 mS/cm @20 degrees
Per 100 mL
Water Treatment Process Scheme
The proposed treatment process has been reevaluated by Montgomery Watson as part of updating the feasibility report. The major goal was to assess its ability for fulfilling the new drinking water guidelines and standards issued by the European Council and also the World Health Organization (WHO). When doing this, the variable raw water quality due to seasonal changes has also been taken into account to provide a process scheme that will be capable of treating two different raw water characters to the desired effluent quality. The selection of appropriate chemicals that will be easy to handle, readily available and will have the minimum impact to the process in terms of sludge production and alkalinity consumption were considered as the key points in selecting these chemicals.
The updated and proposed new treatment scheme will comprise the following unit operations and processes:
Collection of supernatant from thickening and dewatering;
Dirty backwash collection.
The process flow diagram of the updated proposed treatment scheme can be seen in Figure 3 -7 and Figure 3 -8. Two options are presented which are related to the bypass of coagulation/flocculation/settling phases when the raw water quality is good and will not need settling. In this case as mentioned above direct filtration will be enough to treat the water to comply with the treated water quality. As it can be seen from the process flow diagram for the first option, the raw water bypasses all the coagulation/flocculation and settling tanks and flows into the filters. In this case, an inline static mixer is provided and the coagulant, flocculant can be dosed to this section of the plant before going to the sand filters. In this option, raw water passes through ozonation for pre-oxidation and pre-disinfection. The static mixer will aid in coagulation and flocculation. Alternatively, the chemicals can also be dosed to the outlet of the cascade aeration structure where the turbulence is high.
In the second option, the raw water flows through ozonation, coagulation and flocculation but bypasses settling and flows directly to the sand filters. In this way during direct filtration the main flash and slow mixing units will still be used for coagulation and flocculation. Both of the above mentioned options avoid the use of secondary (intermediate mixing) mixing facilities to be utilized only during direct filtration. The arrangement to be implemented should be decided by the contractor. The Awali treatment plant will have two parallel streams.
The unit operations and processes and the justifications for the updated process scheme are discussed separately in this section. The tentative dimensions given in the tables are for 3.09 m3/sec net inlet flow capacity which includes 3% water losses for backwashing. Exact amount of water losses to be taken into account for the inlet flow should be defined by the contractor.
Figure 3 7 Proposed Treatment Process (Option1)
Figure 3 8 Proposed treatment Process (Option2)
Screening has been foreseen to avoid small/large objects such as grasses, leaves, plastic debris etc flowing to the treatment plant which can enter the raw water source in many different ways. Absence of screens can lead to problems. It is essential to install coarse screens to protect the downstream processes. Therefore bar screens have been foreseen with 40-50 mm opening widths. At this stage automatically operated screens have been foreseen, however the type and operation mode of the screen will have to be finalized by the contractor.
Aeration involves bringing air in contact with water to transfer volatile substances from the liquid into the gaseous phase and to dissolve beneficial gases into the water. The purposes of aeration in water treatment are:
to reduce the concentration of taste and odor causing substances such as H2S or other volatile organic compounds;
Oxidation of iron and manganese;
Addition of oxygen to the raw water which can be deficient in dissolved oxygen.
The available data does not suggest the necessity of aeration at first glance, however it should be noted that the data is very limited and does not extensively cover the seasonal changes in the raw water quality. Cascade aeration is proposed which does not involve any mechanical parts and the oxygen transfer is solely related to the fall of the water over a number of steps. It will be a concrete structure having three steps and each step having a 50-cm fall (Table 3 -33).
The water is supplied via long tunnels so especially during the summer times the raw water could be deficient in oxygen which can have implications on taste and odor issues. Therefore it is sensible to construct a cascade aeration system for the Awali process scheme. One cascade aeration system can be constructed to serve the system requirements; this should be checked by the contractor.
Table 3 33 Proposed Specifications of Cascade Aeration System
Ozone is a powerful oxidant and has many uses in water treatment, including oxidation of organic chemicals. Ozone can also be used as a primary disinfectant. Ozone gas (O3) is formed by passing dry air or oxygen through a high-voltage electric field. When used for the pretreatment of raw surface water, ozone prevents the formation of THMs and other chlorinated derivatives. Ozone is more widely used as the pre-oxidant in water treatment systems because it has a number of benefits by improving clarification effectiveness (turbidity, color, residual micro-algae, OM, THM precursors) and in most cases by reducing the coagulant demand. In many plants it has been proven that with pre-ozonation:
Coagulation and flocculation are enhanced and performance of sedimentation and filtration processes is improved;
Odors are not created or intensified by formation of complexes;
Chlorine demand is reduced and in turn lowers chlorine dosage and so THM formation potential;
Complex taste, odor and color problems are effectively reduced or eliminated;
Organic impurities are rapidly oxidized;
Effective pre-disinfection is achieved over a wide range of temperature and pH;
Removal of pesticides and herbicides;
Removes iron and manganese.
Ozone is now widely used in most of the conventional potable water treatment plants and the dosage is very small which will be in the range of 0.5 to 2 mg/L. It is suspected that the dosage will not exceed 0.7-1.0 mg/L for the Awali scheme.
Available data does not necessitate the use of a pre-oxidation step at first glance, however it is noteworthy to mention that the current available analysis results are very limited and do not cover a year round seasonal sampling analysis. The quality of both water sources during wet weather events is very important to conclude on the necessity of the pre-oxidation step using ozone. Nevertheless, the following are the specific benefits of having pre-ozonation at the Awali plant:
Will improve the performance of coagulation and flocculation and even reduce the coagulant demand;
Will oxidize the organics;
Will eliminate problems with taste, odour and color;
Will eliminate the risk disinfection byproducts (THM and other);
Will remove the small amount of pesticides and herbicides detected in the raw water which have strict treated water quality targets.
Intermediate ozonation has also been foreseen in the feasibility report as a future item to aid in the removal of pesticides, herbicides, to enhance coagulation, flocculation and reduce the disinfection by products such as THMs and chlorinated compounds. Taste and odor issues caused by disinfectants and DBPs are best controlled through careful operation of the disinfection process. In principle, they can be avoided by using ozone.
By the use of the process scheme given in Figure 3 -7 and Figure 3 -8 the water will be pre-ozonated both during settling and direct filtration.
Concrete covered tanks will be used to provide the required ozone contact time which will be determined by the contractor stage.
In the absence of a detailed raw water sampling and analysis campaign, it is prudent to include the preozonation for both oxidation and pre-disinfection purposes for the Awali treatment plant. However the final decision will be taken once the detailed analysis results are available to ascertain the present raw water quality. The possibility of a lower water quality both for the Awali and lake Karaoun sources should not be ruled out as over the years, residential and commercial developments, agricultural activity and industrialization may have affected the water sources and to be sure, up to date sampling and analysis should be conducted.
Table 3 34 Proposed Specification for Pre-oxidation and Disinfection.
Upstream of settling tank
Generators (one standby)
Capacity of each generator
Ozone contact basin
Coagulation is a chemical treatment process used to destabilize colloidal particles. In this process chemicals are added to the water that either break down the stabilizing forces, enhance the destabilizing forces, or both. Typically aluminum and iron salts are used as coagulants i.e., aluminium sulphate, ferric chloride, ferrous sulphate etc. As mentioned earlier, although several options have been discussed for the main coagulant in the feasibility study, ferric chloride was chosen as the chemical to be used for coagulation. After reevaluation of the current and best practice and considering the advantages and disadvantages of these chemicals, anhydrous aluminum sulphate is proposed as the main coagulant.
Although both chemicals are used in water treatment processes depending on price, availability, handling etc, aluminum sulphate has a wider usage. There are several advantages of aluminum sulphate compared to iron salts which can be listed as:
Availability, as anhydrous aluminum sulphate is very easy to obtain (Subject to availability on the local market);
Price, as anhydrous aluminum sulphate is cheaper compared to especially ferric chloride (Subject to availability on the local market);
Will consume less alkalinity of natural water compared to ferric chloride (ferric based salts will consume 0.75-0.92 mg/L CaCO3 alkalinity per 1 mg of salt dosed; alum will consume 0.50 mg/L CaCO3 alkalinity per 1 mg of salt dosed);
Will produce less inorganic sludge compared to ferric chloride (ferric based salts will produce 0.54- 0.66 mg insoluble precipitate per 1 mg of salt dosed; alum will produce 0.26 mg insoluble precipitate per 1 mg of salt dosed).
Apart from its advantages, handling of aluminum sulphate is slightly more complex which requires the use of silos and feeding screws to prepare 6%-10% w/w dosing solution in flash mixing tanks. Furthermore, the optimum working pH of aluminum sulphate is 6.0-7.4 which will require the dosing of an acid (pH adjustment chemical) to reduce the pH of the incoming raw water to the desirable range and also meet the 0.05 mg/L Al effluent consent. Since the raw water pH is in the range of 6.9-8.0, the amount of acid to be dosed will not be significant. On the other hand, due to the addition of certain chemicals, the natural alkalinity of the water will be consumed and a final alkalinity buffering and pH correction will have to be done anyway. Ferric chloride has the disadvantage and possibility of leaving a residual red color in the treated water if the process is not well controlled. This has been experienced in some plants. Furthermore, iron promotes the growth of iron bacteria if ferrous sulphate is used. This may cause rust-colored deposits on the walls of tanks, pipes and channels and carry-over of deposits into the water.
However, it is very important to once again mention that the final selection of the coagulation chemical will be done following jar tests conducted on samples taken from both water sources also taking into consideration the availability of the chemicals in the local Lebanese market. It is also known from previous experience that alum is a more cost effective chemical than ferric.
Coagulation will be carried out in concrete flash mixing tanks by the use of rapid mixers to provide thorough dispersion and mixing of the chemical. The success of this unit operation depends on this.
Mechanical in-tank coagulation is necessary especially due to low raw water temperatures. It is proposed to use two tanks in series where the first tank will receive the pH adjustment chemical and the coagulant.
Table 3 35 Proposed Specifications for Coagulation
In tank mixing
4 (2 in series in each stream)
Turbine flash mixer
pH adjustment chemical
The coagulation process chemically modifies the colloidal particles so that the stabilizing forces are reduced. To ensure that a maximum amount of turbidity is removed, mixing conditions and energy input must be properly provided after rapid mixing to allow aggregation of destabilized particles. The coagulated water must be gently stirred to promote the growth of flocs which can be removed by sedimentation or filtration. The typical floc size is in the range of 0.1-2.0 mm. Jar tests will have to be conducted to determine the correct type of polyelectrolyte i.e. anionic, non-ionic or cationic. Mostly anionic types of polymers are used depending on the nature of the colloids. Metallic oxides are generally positively charged. However most surface waters carry negatively charged colloids which may require the use of cationic polymer. As mentioned previously, jar tests will be conducted to determine the type of polymer to be used during settling and direct filtration. Coagulated water and the polymer will be mixed in concrete flocculation tanks equipped with slow paddle type stirrers. To enhance the growth of flocs tapered velocity gradient will be applied using three tanks in series.
Table 3 36 Proposed Specifications for Flocculation
In tank mixing
6 (3 in series in each stream)
Paddle type slow stirrer
Energy Gradient (first compartment)
Energy Gradient (second compartment)
Energy Gradient (third compartment)
Power (first compartment)
Power (second compartment)
Power (third compartment)
Due to land constraints, lamella type plate settlers will be used. This type of settlers consists of banks of small plates inclined at 45o to 60o angles from horizontal. The lamella plate settlers provide enhanced solids removal because, 1) the settling distance that a particle falls to enter the sludge zone is reduced(thus, the surface loading rate is reduced in the basin), 2) Laminar flow is achieved through the plates(thus nearly ideal settling conditions are encountered), 3) Density currents, temperature currents andwave action do not hinder the sedimentation process.
The sludge will be automatically removed using motorized valves and pumps. Sludge will be pumped to sludge thickeners. Mechanical scrapers will be used to scrape the settled sludge to the hoppers.
Table 3 37 Proposed Specifications for Sedimentation
Lamella Plate settler
2 (one in each stream)
Specific loading rate
Footprint loading rate
3.26 x 1.25
Number of lamella per unit
Number of rows per unit
5 – 5.5
On top of lamella stacks
Automatic via pumps and motorized valves
Further removal of colloidal particles is required to meet stringent public health standards. The filtration process used in water treatment involves passing of the flow through a bed of granular media such as sand, anthracite or activated carbon. As the water passes through the media, the suspended particles are entrapped in the pore spaces of the media and thus removed from the liquid stream.
Rapid sand filters have also been foreseen in the updated process scheme as done in the feasibility study.
Dual media sand filters are recommended with sand and anthracite which will facilitate further removal of organics and also eliminate taste and odour problems. The filter media can also be replaced with granular activated carbon if the necessity arises. Filters will have a combined air and water backwash (CAW) sequence to provide the most efficient way of removing entrapped solids and this will be fully automatic.
There are two options for the dirty backwash water disposal and handling as proposed by the designer: 1) it can be discharged to the thickeners or 2) it can be directly sent to the wadi. The latter must be further investigated and confirmedwith the local regulatory authorities. The dirty backwash water tank will be equipped with a submersiblemixer to keep its content in suspension.
Post chlorination will be carried out using gas chlorine. The contact tank will have a detention time of 30 minutes and will be baffled to provide plug flow conditions. Two tanks have been foreseen for ease of operation and maintenance. Average chlorine dose of 3.5 mg/L and a maximum dose of 5 mg/L is expected and the capacity of the gas chlorination system has been based on this dosage.
The preliminary dimensions of each chlorine contact tank will be as follows; length=25m, width =12.5m and the water depth will be 4.25m.
Treated Water Reservoir
A treated water reservoir has been foreseen to store the treated water up to a maximum of 1 hour. The same criteria have been taken into account in the feasibility study. Two tanks each having a capacity of 5750 m3 will be sufficient to satisfy this requirement. The tanks can be isolated with penstocks. The preliminary dimensions of each reservoir will be as follows: length=33m, width=33m and the water depth will be 5m. The necessity and capacity of the treated water reservoir will be reevaluated by the contractor.
Post pH Adjustment
Chemical dosing will consume the natural alkalinity of the raw water and hence decrease the pH as the water passes through the treatment steps. In the updated process, due to the recommendation of alum, acid will be dosed to decrease the pH of the raw water to the desired level for optimum coagulation. This will further reduce the buffer capacity of the water and decrease the pH. Therefore the final treated effluent has to have the necessary alkalinity buffer and also has to be in the pH range as given in the treated water quality.
Certain chemicals can be used for this purpose such as hydrated lime, quick lime or caustic soda. In the feasibility study lime has been chosen because of its price. Nothing has been mentioned about lime being more readily available than caustic. However as outlined in the same report there are a lot of implications associated with using lime and many plants in the world have considered caustic for pH adjustment. Below are some of the negative aspects of using lime in water treatment plants:
It is very difficult to handle. In large treatment plants it can only be stored in silos which result in arching. It is also very dusty;
There are many impurities;
It may further increase the turbidity of the water due to these impurities. (this is especially crucial if dosed into treated water);
They require sophisticated storage and handling equipments;
Capital cost of lime storage and dosing facilities is high and so are the maintenance costs.
Furthermore when dissolved in water, Ca2+ contained in lime may form CaCO3 and CaSO4 precipitates which can settle in pipes, reservoirs and cause scaling. Therefore due to the reasons stated above, it can be concluded that technically, caustic soda is a preferred chemical to be used for post pH adjustment. However, the final selection of the post pH adjustment chemical will be left to the final design stage as there have been some reports that caustic is not available in the local market but can be imported. This needs to be further investigated.
Ammoniation is the process where monochloramines are formed by the addition of ammonium into the treated water. Ammonia converts free chlorine residual to chloramines. In this form, chlorine is less reactive, lasts longer and has fewer tendencies to combine with organic compounds thus reducing taste and odor and THM formation. However dosage of liquid ammonia has to be carefully adjusted as excessive amounts can lead to the formation of disinfection by products. Ammoniation process was foreseen as an optional item in the feasibility study however it is recommended to include this as part of the process scheme. It is expected that the dosage will be in the range of 0.5-1.0 mg/L.
Total quantities of chemical storage will depend of the selection of coagulant and flocculent and their relevant dosing rates decided by the contractor. It is very hard to be definitive about this at this stage but the designer has provided in the following table a guidance on expected quantities.
Table 3 39 Chemical Storage
No. Days Storage
(40% solution, 1450kg/m3)
2 tanks X 135 m3
3 tanks X 135 m3
2 tanks X 7 m3
3 tanks X 7 m3
Anionic Polymer (Dry, 900kg/m3)
(25% solution, 1250kg/m3)
2 tanks X 190 m3
3 tanks X 190m3
(Liquid gas, one tone cylinder)
(liquid 25% solution, 0.95 tonnes/m3)
2 tanks X 135 m3
3 tanks X 12m3
(Dry, 4% soluble)
One m3 day tank
One m3 day tank
Note * Quantities based on potassium Permanganate
Coagulation sludge is produced by flocculating and settling natural turbidity. Alum and iron salts will react with alkalinity and form precipitates of alum and iron hydroxides. Settled sludge contains these hydroxide precipitates and also turbidity causing organic and inorganic compounds. The sludge produced from water treatment facilities is stable, because mostly there is no organic matter to undergo active decomposition or promote an anaerobic condition. As a result, the sludge is often allowed to accumulate in sedimentation basins and holding/thickening tanks for days. Basic characteristics of coagulation sludge which needs to be taken into account during the design are as follows:
The solids concentration, thickening, density and de-waterability of the produced sludge are highly dependent on the raw water quality;
Treatment of high turbidity surface water will result in sludge that is more concentrated and less difficult to dewater, sludge produced from the treatment of low turbidity surface water will be difficult to process;
Coagulation sludge from water containing high algae, will result in light and low solids concentration;
Sludge that have a high proportion of metal hydroxides are easily dewatered because metal hydroxides have water molecules in their structure that can separate the floc and other particles;
Addition of polymer, lime will increase the solids concentration;
Alum sludge is a voluminous gelatinous sludge with poor compressibility. It will generally concentrate to 0.5-3% solids in the sedimentation basin;
Sludge concentration in the settling tanks depends on how they are operated. Typical concentration varies between 0.5-3% however this will increase if the sludge is allowed to accumulate for some time;
Density of sludge depends on the moisture content. Normally for surface water sludge, the density of dry sludge is in the range of 1200-1520 kg/m3.
In the 1994 feasibility study, several options have been proposed for the management of solids which can be listed as:
Option A - Marine disposal;
Option B - Disposal at the Ras Damour Power Station;
Option C - Disposal at a nearby cement plant;
Option D - Disposal to restore local quarries;
Option E - Disposal to landfill;
Option F - Disposal to agricultural land; and
Option G - Return of raw sludge to Joun.
It was concluded in the 1998 EIA report that disposal of sludge to the marine environment (Option A) would be an unacceptable long-term proposal, would give rise to low levels of water treatment chemicals being present in the marine environment and would have a high capital cost. Amongst all the alternatives given above, Option D was selected as the most viable option which is the disposal of sludge to restore local quarries (e .g. the small quarry west of the Ouardaniye WTW) with possible future use of the sludge as a construction material represented the best option for sludge disposal, provided that care will be taken to avoid groundwater contamination. In the longer term, it is also stated that the sludge may be disposed of to an engineered landfill or a wastewater treatment plant, once these are constructed. The selected option therefore represents a flexible approach to sludge disposal.
In light of the proposed sludge management strategies given in the previous EIA study (for the short and long term) the onsite sludge treatment has to be provided. The following steps are recommended for the treatment of sludge:
Sludge holding /thickening;
Sludge dewatering (with polymer aid);
Sludge liquor collection (supernatant tank).
As discussed earlier, alum sludge has a slimy, voluminous character which makes it difficult to process.
Therefore the sizing of the units and design criteria must be carefully selected.
Sludge thickening can be carried out in circular gravity thickeners which will also serve the purpose of sludge holding and storage. Thickened sludge can then be dewatered using belt press or centrifuges.
Centrifuges have been preliminarily proposed during the feasibility study however they are costly and energy intensive. Alternatively, belt presses are proposed at this stage as they are very low in energy consumption and the sludge to be processed is fairly stable and will not potentially emit any malodors. In any case belt presses can be completely covered. Cationic polymer is recommended to aid in increasing the solids concentration and dewater ability however the selection of the type of polymer will be done at time of construction by the contractor. A supernatant tank has been foreseen with a reasonable detention time to collect sludge liquors resulting from thickening and dewatering processes. Sludge yield figures for treating 250,000 m3/s and 500,000 m3/s (twice the expected capacity under the current project) are given in Table 3 -40. It depends on chemical type and dosage and whether direct filtration or coagulation/flocculation and settling are done.
Table 3 40 Sludge Yield
Maximum 72 hours
For treating 250,000 m3/s
For treating 500,000 m3/s
The yield assumes ferric sulphate is dosed as the principal coagulant and that the clarification produces a 0.3% so0lids sludge.
Dry solids in the sludge cake will be 230*4*12*0.97= 10,708 kg/d (10.8 tons/d). Dry solids will not change unless the solids capture of the machine changes or more solids are produced from the liquid process due to higher turbidity, higher chemical dosage…etc. It is the wet sludge amount that will change with respect to dewatered sludge concentration and cake density. So for average conditions, the dry solids in the sludge cake will be approximately 11 tons/d and the wet sludge will be in the range of 58 m3/d to 73 m3/d dependant on the cake concentration (12-18%) for a density of 1200 kg/m3.
The conceptual sizes of the sludge treatment units are shown in Table 3 -41
Table 3 41 Conceptual Design Parameters of Sludge Treatment Units