Introduction To Planning of Desalination Projects: Nikolay Voutchkov, PE, BCEE
Introduction To Planning of Desalination Projects: Nikolay Voutchkov, PE, BCEE
Introduction To Planning of Desalination Projects: Nikolay Voutchkov, PE, BCEE
Introduction to Planning of
Desalination Projects
by
1. INTRODUCTION
The purpose of project planning is to define the size, location and scope of the desalination
project and chart a roadmap for project implementation. The first step of project planning is to
determine the area that the desalination plant must serve, identify the type of water use and
assess the water demand and quality requirements of each water customer over the useful life of
the desalination project, typically 25 to 30 years.
Once the project size and service area are determined, the next step of the planning process is to
define the project. This requires identifying the most viable plant site location, intake and
discharge type and configurations, characterization of source water quality and selection of the
treatment process.
The selection of the most cost-effective and environmentally sound desalination project location
and configuration is based on a thorough evaluation of a number of alternatives for key
desalination project components such as saline source water intake, concentrate discharge,
pretreatment, RO system, post-treatment and product water delivery system.
Once the project scope and schedule are defined, the project environmental impact assessment
must be prepared. Project entitlements such as legal rights to land use, water rights for source
water collection; easements for project-related infrastructure; rights-of-way; electric power
supply agreement; and environmental permits, licenses and other regulatory, legal and
contractual documentation, need to be obtained. This process can take several months to one of
more years, depending on the governance environment and project size and complexity.
2. MAIN CONSIDERATIONS
The boundary of the desalination plant service area may also be influenced by the ability of the
water authority to supply lower cost water to the same area from one or more other sources, or to
increase the level of water conservation and/or water reuse in this area in order to balance water
demand and supply.
Other important factors associated with the size of the service area of the desalination plant are
the costs of water production and of water delivery. Usually, a larger desalination project service
area will result in a larger size plant, which in turn will yield cost savings from economies of
scale. On the other hand, delivering water to a larger service area may require the construction
of additional costly fresh water conveyance and distribution infrastructure, which can negate the
savings associated with the construction of one or more larger plants.
Due to the variables above, the optimum size and boundaries of the desalination plant service
area have to be established based on a life-cycle cost benefit analysis which balances the cost
savings associated with building larger desalination plant and expenditures associated with
delivering the desalinated water to the final users. .
Usually, desalination is one of the most-costly sources of water supply available for a given
service area. Desalination plant capacity is often determined based on the fresh water flow
which this water supply alternative can provide during periods of prolonged drought as
compared to other water supply resources, and on the incremental costs of new water supplies.
A second factor for the selection of optimum desalination plant size is the economy-of-scale
benefit of building one or more large desalination plants supplying fresh water for the entire
service area vs. installing a number of smaller facilities located closer to the main water users
within the service area.
Supply analysis usually considers annual and daily water supply patterns, water distribution
system hydraulic limitations, water quality and quantity requirements of large users in the service
area, and future water demand projections. This analysis also includes requirements and costs
for conveyance, connection to existing infrastructure and associated capacity limits, hydraulic
distribution requirements (i.e. size of piping and equipment, and operating pressure of the water
distribution system at the point/s of delivery of desalinated water) and system conveyance
capacity limitations and potential solutions.
The plant site footprint requirements in Table 1 are based on a comparative review of over 40
desalination projects worldwide. However, sometimes environmental and zoning regulations,
physical constraints and/or soil conditions associated with a particular site may yield smaller or
larger plant site requirements.
In some cases, when the available site is located in a densely populated area or land costs are at
premium, the desalination plant can be located at a fraction of the site footprint shown in Table
1.
Development of more compact plant layout requires some of the main treatment equipment and
systems to be installed below ground or in multi-story buildings, which is usually more costly
than housing all plant treatment facilities in single-story buildings at or above site grade.
The most viable site location for a given project is usually determined through a cost-benefit
analysis of several alternative sites within the plant service area. Potential sites must meet the
following requirements as a minimum:
• adequate site size, configuration and footprint;
• accessibility from existing main roads, highways, etc.;
• proximity (preferably within 8 km/5 miles) to points of delivery of desalinated water and
to source of electrical supply;
• proximity (preferably within 1.0 km/0.6 miles) from to the source of saline water and the
location/s of concentrate discharge;
• suitable land planning and zoning requirements;
• minimal soil or ground water contamination, vegetation, debris, and existing surface and
underground structures and utilities;
• location of the site outside of environmentally sensitive areas such as wildlife sanctuaries,
migratory bird stopover sites, and natural habitats of endangered species;
• location of the plant intake and outfall away from areas of high biological significance and
sensitivity such as coral reefs; kelp/seagrass beds; aquatic animal sanctuaries; water habitat
restauration zones; and highly productive coastal wetlands.
• reasonable costs for obtaining entitlements associated with the use of and access to the
desalination plant site (i.e. less than 0.5 % of the total plant construction costs);
• adequate distance (at least 30 m/100 ft or more) away from residential dwellings, hotels,
hospitals, schools, places of worship, and other developments that are sensitive to increased
levels of noise and traffic during plant construction and operations.
After potential sites are identified, a number of engineering and environmental review activities
are typically completed for each site:
• geological reconnaissance survey;
• land zoning and real estate ownership and cost survey;
• site flooding analysis – the site should be located above the 100-year flood line;
• traffic/access survey;
• survey of existing above and underground utilities and structures (subsurface utility
survey);
• biological and archeological surveys;
• evaluation of near- and offshore marine resources with a focus on the type, environmental
sensitivity, and location of aquatic species inhabiting the desalination plant intake and
discharge areas;
The engineering information collected from the above site studies and investigations is typically
compiled into project site alternatives product water delivery routes that are then ranked based on
their merits and potential disadvantages. The most viable site is selected on the grounds of a
comprehensive site evaluation and cost-benefit analysis of project alternatives.
Desalination plant intakes are configured and designed to collect saline source water of adequate
quantity and quality reliably and sustainably and to facilitate cost-effective production of
desalinated water with minimum impact to the terrestrial and aquatic environments in the
vicinity of the project location. Intakes are thus important components of the plant, and their type
and location have a measurable impact on the quality, cost, and potential environmental impacts
of plant operations.
Currently, there are two categories of widely used desalination plant source water collection
facilities: open intakes and subsurface intakes (wells and infiltration galleries).
Open intakes collect saline source water directly from a surface water body (brackish river or
lake, the ocean, etc.) via on-shore or off-shore inlet structure and pipeline interconnecting this
structure to the desalination plant. Subsurface intakes, such as vertical and horizontal wells,
slant wells, and infiltration galleries are typically used to collect saline water from brackish
aquifers for BWRO desalination and from near- or off-shore coastal aquifers for SWRO
desalination. Both open and subsurface intakes have functional and capacity constraints and
their operation can potentially have environmental impacts.
Once the location of an adequate source water aquifer is identified, this aquifer needs to be
characterized in terms of transmissivity, thickness, water quality and potential interconnection
with other aquifers. The productivity of the target source water aquifer and the projected
capacity of the individual extraction wells is determined based on a number of aquifer type, size,
water quality and sustainable yield studies including:
• hydrogeological survey which includes the collection of aquifer formation deposits for
visual classification and grain-size distribution analysis as well as for groundwater table
level determination;
• installation and operation of test and observation (monitoring) wells to determine the safe
aquifer water extraction yield;
• collection of samples for groundwater quality/contamination analysis; and
• hydrogeological modeling of well yield, radius of influence and water quality changes over
time as well as groundwater movement and interference with existing fresh and brackish
water collection fields.
If the target brackish source water aquifer is already in use, water quality information from
existing wells can be applied for projecting of the source water quality of the new wells.
However, this water quality information alone may not be adequate to predict changes of aquifer
water quality as a result of the increased rate of extraction from this aquifer.
Extracting additional volume of water from a given aquifer will always result in the modification
of the natural groundwater movement regime in the aquifer. Some brackish groundwater
aquifers are density stratified and as lower salinity water is extracted from the top portion of the
aquifer, a higher salinity groundwater propagates upwards and increases the desalination plant
feed water salinity over time. In addition, for semi-confined aquifers, ground water from an
adjacent aquifer can move to the main water extraction aquifer for a given project.
If the water quality of the adjacent aquifer is very different from that of the main aquifer, the
overall plant source water quality may change over time. Therefore, it is of critical importance
to complete predictive plant source water quality modeling as a part of the project planning
process.
The prime criteria for selecting the most suitable location for the BWRO project source water
aquifer are its safe yield capacity and its proximity to the desalination plant site. The presence of
potential sources of subsurface or surface contamination which can propagate and contaminate
the plant source water (i.e. intake well proximity to unlined sanitary or hazardous waste landfills;
leaking fuel oil storage tanks; cemeteries leaking highly toxic preserving solutions such as
formaldehyde; industrial or military sites of known ground water or surface water contamination,
etc.) means the site is not suitable.
Another issue of importance is the proximity of the desalination plant intake wells to existing
fresh water supply wells and the potential for the desalination plant well operation to result in the
decrease in the fresh water well production capacity.
Surface Water Intakes. At present, less the 10% of the brackish water desalination plants
worldwide have surface water intakes. Such intakes are typically located in the confluence of
river and an ocean or sea. One of the largest desalination plants with such intake at present is the
Beckton desalination plant near London, United Kingdom (Figure 1). The plant has capacity of
150,000 m3/day (40 MGD) and is operated by Thames Water. The criteria for selection of
location and configuration are similar to these of open seawater intakes.
Productivity of the Coastal Aquifer. The capacity of the coastal aquifer and the quality of
water such aquifer can yield are the two most important factors which define the size of the
seawater desalination plant and often, its location. Hydrogeological study is a critical component
of the SWRO desalination project planning process and aims to determine the aquifer’s safe
yield and sustainable water quality.
Beaches and shallow bays which have low transmissivity, contain large quantity of silted beach
deposits, and are poorly flushed are unsuitable for installation of beach well intakes. Beach
wells and near shore open intakes use the same seawater as a source. In desalination plants with
open intakes the solids in the source seawater are removed in the plant pretreatment filtration
system. With beach well intakes, the same amount of solids is retained on the ocean floor while
the filtered water is slowly conveyed through the beach sub-terrain formation until it reaches the
well collectors.
The wave action near the ocean floor allows the solids separated from the beach well source
water to be dissipated in the ocean. If the bay area is not well flushed, then these solids will
accumulate on the ocean floor and will ultimately reduce the well capacity and negatively impact
the source water quality.
Useful Life of Beach Well Intakes. Depending on the site specific conditions in the vicinity of
the subsurface intake collection area, beach wells may have a shorter useful life compared to
open ocean intakes. The useful life of open ocean intakes is typically between 30 and 100 years,
depending on the quality and type of materials of construction, and configuration.
Without major refurbishment, beach wells will typically operate at design capacity for a period
10 to 20 years. Over time, the beach well yield may diminish due to naturally occurring scaling
of the well collectors caused by chemical precipitates or/and bacterial growth. The rate of well
yield decrease with time is difficult to predict and requires specialized expertise and detailed
studies. Therefore, beach well intakes are usually designed with 20 to 25% reserve/standby well
capacity, which adds to their capital costs and the size of the impacted beach shore area.
Beach erosion can also significantly impact the useful life of the intake wells. As seen on Figure
2, if the well intake area is exposed to high rate of beach erosion, wells may loose soil support,
productivity and structural integrity. Therefore, beach erosion may shorten the useful life of the
beach wells significantly, and may increase the overall life-cycle water production cost.
Due to its significant potential impact on the intake system operation and costs, beach erosion in
the vicinity of the intake location has to be thoroughly investigated and considered prior to
selection. If the selected beach site has a high potential for erosion, then anti-erosion measures
need to be implemented. It is preferable to use deep open intake or install the intake wells inland
in an area outside of the zone of active beach erosion.
Beach erosion may also impact the useful life and integrity of open intakes. Therefore, in coastal
zones exposed to active beach erosion the first several hundred to one thousand meters of the
intake pipeline closest to the shore are typically installed under the ocean floor at depth beyond
the zone of active beach erosion.
The useful life of a well-designed and operated seawater desalination plant is 25 to 30 years. As
beach wells often have a shorter useful life span than that of the desalination plant, in the worst-
case scenario, two sets of beach wells may be required over the useful life of the SWRO plant.
The need for replacement of some or all of the original beach wells after the first 10 to 20 years
of operation will also magnify the shoreline impacts of the beach wells and increase the overall
cost of water production. Therefore, the potential difference between the useful life of beach
wells and open intakes has to be reflected in the life-cycle cost comparison associated with
intake selection.
Source Water Pretreatment Requirements. Seawater beach wells typically yield better intake
water quality than open intakes in terms of turbidity, algal content and silt density index. These
are the key parameters associated with the selection, sizing, complexity and costs of the
desalination plant pretreatment system. It should not be assumed that pretreatment will not be
required however: as this is only the case for very site specific favorable hydrogeological
conditions (i.e., the wells are located in well flushed ocean bottom or shore; are sited away from
surface fresh water influence; and are collecting seawater from a coastal aquifer of uniformly
porous structure, such as limestone or dolomite).
Long term operational experience at numerous small seawater desalination plants in the
Caribbean and several medium-size plants in Malta which have well intakes located in limestone
and other favorable rock formations, indicates that such plants can successfully operate with
minimal pretreatment (typically bag or cartridge filters and/or sand strainers) ahead of SWRO
pretreatment. However, most seawater desalination plants using subsurface intakes have to
include an additional granular or membrane filtration step prior to RO membrane salt separation
to be able to process source water collected by subsurface intakes.
Existing experience with the use of beach wells for seawater desalination in California and at the
largest beach-well seawater desalination plant on the West Coast in Salina Cruz, Mexico indicate
that some desalination plants using beach wells may face a costly problem – high concentrations
of manganese and/or iron in the intake water. Unless removed ahead of the reverse osmosis
membrane system, iron and manganese may quickly foul the cartridge filters and SWRO
membranes, and render the desalination plant inoperable.
The treatment of beach well water which naturally contains high concentrations of iron and/or
manganese requires chemical conditioning and installation of conservatively designed
“greensand” pretreatment filters ahead of the SWRO system. This pretreatment requirement
may significantly reduce the cost benefits of the use of beach wells as compared to an open
intake. Open seawater intakes rarely have iron and manganese source water quality related
problems as open ocean water does not contain these compounds in significant quantities that
can cause RO membrane fouling.
An example of a beach well desalination plant which faced an elevated source water iron
problem is the 4,500 m3/day (1.2 MGD) Morro Bay SWRO facility located in Northern
California, USA. The plant source water is supplied by five beach wells with production
capacity of 1,100 m3/day to 1,900 m3/day (0.3 to 0.5 MGD), each. The beach well intake water
has iron concentration of 5 to 17 mg/L. For comparison, open intake seawater typically, has
several orders of magnitude lower iron concentration.
The Morro Bay facility was originally designed without pretreatment filters, which resulted in
plugging of the RO cartridge filters within half-an-hour of starting operations during an attempt
to run the plant in 1996. The high-iron concentration problem was resolved by the installation of
pretreatment filter designed for a surface loading rate of 6.1 m3/ m2.h (2.5 gpm/sq ft). For
comparison, a typical open-intake desalination plant is designed for pretreatment loading rates of
10 to 13.5 m3/ m2.h (4.0 to 5.5 gpm/sq ft) and therefore, will require less pretreatment filtration
capacity.
The largest existing Pacific-coast seawater desalination plant in Salina Cruz, Mexico has also
faced iron and manganese challenges, which have been resolved by the installation of
pretreatment filters and chemical conditioning of the beach well water. The existing experience
shows that the costs for pretreatment of seawater with high iron/manganese content collected by
a beach well intake are typically comparable or higher than these for pretreatment of seawater
collected using an open-ocean intake.
Source Water Quality Variations. Open ocean intakes provide relatively consistent seawater
quality in terms of total dissolved solids concentration. The intake source water TDS
concentration data collected for the development of the Huntington Beach seawater desalination
project in Southern California, USA indicate that the open intake salinity varied within 10% of
its average value of 33.5 ppt.
Although in general beach wells produce source water of consistent salinity, they can also yield
water of unpredictably variable TDS concentration with swings exceeding over 30% of the
average. For example, the TDS concentration of the two operational wells at the Salina Cruz
water treatment plant vary in a wide range – for well No. 2 between 16.8 and 21.8 ppt, and for
well No. 3 between 17.8 and 19.8 ppt. The wide range of source salinity concentration in this
case is explained by fresh groundwater influence.
A similar trend was observed at the Moro Bay SWRO plant in California. During the plant’s
initial operation in 1992, the well water TDS was approximately 26,000 mg/L. In December
2001, the TDS of the intake water was 6,300 mg/L. The December 2002 data for the same plant
indicate intake salinity of 22,000 mg/L.
The wide range of intake salinity in systems using beach wells over time requires the installation
of variable frequency drives on the high pressure pump motors for efficient power use control,
which ultimately increases the construction cost of such system and complicates its operation.
One important issue to consider when assessing the viability of using beach wells is whether
intake well salinity can change unpredictably over time when the well operation is influenced by
fresh water inflow to the well source water aquifer. This uncertainty of intake water quality
increases the risk of uncontrollable increase in unit cost of water production over time and has to
be considered when comparing the overall life-cycle costs of the desalination plant operations.
Therefore, the beach well intake water quality must be characterized by installing a set of test
wells and collecting water quality samples under a variety of operational conditions. Thorough
year-round water quality characterization is of high importance for beach wells which source
water may be influenced by fresh groundwater aquifers with seasonal fluctuation of water
quality.
Emerging Contaminants. Usually open ocean intakes are considered less viable source of water
for desalination plants in areas located close to wastewater discharges or industrial and port
areas. However, the open intake seawater is typically free of endocrine-disruptor or carcinogenic
type of compounds such as: Methyl tert-butyl ether (MTBE), N-Nitrosodimethylamine (NDMA)
and 1,4-dioxane. Long-term water quality data collected for the development of the Huntington
Beach and Carlsbad SWRO projects in Southern California and a number of other desalination
plants worldwide, confirm this observation.
Beach well water may contain difficult to treat compounds especially when they are influenced
by contaminated groundwater. Example is the Morro Bay SWRO plant, where beach well intake
water was contaminated by MTBE caused by a leak from an underground gasoline tank. MTBE
is a gasoline additive. Similar problems were observed at the California’s Santa Catalina Island
500 m3/day (0.132 MGD) seawater desalination plant that uses beach well intake.
Although beach wells have proven to be quite cost-competitive for plants of capacity smaller
than 4,000 m3/day (1 MGD), open surface ocean intakes have found significantly wider
application for large SWRO desalination plants. At present, worldwide there are less than a
dozen operational SWRO facilities with capacity larger than 20,000 m3/day (5.3 MGD) using
beach well intakes. The largest SWRO facility with beach wells is the 54,000 m3/day (14.3
MGD) Pembroke plant in Malta. This plant has been in operation since 1991.
The largest SWRO plant in North America which obtains source water from beach wells is the
15,000 m3/day (3.8 MGD) water supply facility for the Pemex Salina Cruz refinery in Mexico.
This plant also has the largest existing seawater intake wells – three Ranney-type radial
collectors with capacity of 15,000 m3/day (3.8 MGD), each.
Surface Water Intakes for SWRO Plants. Open intakes typically include an inlet structure
(forebay) with coarse bar screens, a source water conveyance pipeline or channel connecting the
inlet structure to an onshore concrete screen chamber and mechanical fine screens in the
chamber.
Depending on the location of the inlet structure, the intakes can be on-shore or off-shore.
Offshore intakes with vertical inlet structures are the most commonly used for seawater
desalination projects. The off-shore inlet structure is usually a vertical concrete or steel well
(vault) or pipe located at or above the ocean floor and submerged below the water surface (see
Figure 3).
The open intake inlet system may include passive wedgewire screens (see Figure 4). The use of
such screens eliminates the need for coarse and fine screens on shore. Wedgewire screens are
cylindrical metal screens with trapezoidal-shaped “wedgewire” slots with openings of 0.5 to 10
mm (typical size – 3 mm). They combine very low flow-through velocities, small slot size, and
naturally occurring high screen surface sweeping velocities to minimize impingement and
entrainment. These screens are designed to be placed in a water body where significant
prevailing ambient cross flow current velocities higher than 0.3 m/s (1.0 fps) are naturally
available . This high cross-flow velocity allows organisms that will otherwise be impinged on
the wedgewire intake, to be carried away from the screens with the flow.
An integral part of a typical wedgewire screen system is an air-burst back-flush system, which
directs a charge of compressed air to each screen unit to blow-off debris back into the water
body, where they are carried away from the screen unit by the ambient cross-flow currents.
Collocated intake is a type of open intake for desalination plants co-sited with existing power
generation stations using seawater for once-through cooling purposes. Intake and/or discharge of
collocated desalination plants are typically directly connected to the discharge outfall of a coastal
power plant. This warmer cooling water discharged by the power plant is less viscous than the
ambient ocean water, which reduces the energy needed for desalination by membrane separation.
In addition, the use of collocated intakes in most cases eliminates the need for construction of
separate intake and outfall for the desalination plant, which reduces the overall project capital
expenditures.
Open intakes have some of the same challenges associated with the siting and construction of
beach wells including beach erosion and impacts from high-magnitude earthquakes and storms.
In addition, since open intakes collect water directly from the water column, the source water
could contain large quantities of debris, algae, silt, hydrocarbons, and other contaminants of
anthropogenic or natural origin. Since RO membranes are easily fouled by such contaminants,
use of open intakes usually requires elaborate seawater pretreatment as compared to the
construction of beach wells.
Considerations for Selection of SWRO Plant Intake Type. At present, open-ocean intakes are
the most widely used type of intake technology worldwide, and can be installed in practically
any location and built to any size. While open intakes are suitable for all sizes of desalination
plants, their cost effectiveness depends on a number of location-related factors such as plant size,
depth and geology of the ocean floor, water quality contamination on their performance (i.e.
wastewater and storm water outfalls, ship channel traffic and large industrial port activities) and
ease of installation.
Subsurface intakes have found somewhat limited application, and mainly for relatively small
plants where only a small number of beach wells is required. This is due in part to the difficulty
in finding favorable hydrogeological conditions in the vicinity of the desalination plant site. Land
availability is a further constraint often encountered in densely populated coastal areas where
large desalination plants are needed.
Both open and subsurface intakes offer different advantages and disadvantages in terms of costs,
construction complexity, environmental impacts, operational considerations, and subsequent
source water pretreatment and concentrate disposal needs. The selection of the most suitable
intake system for the site-specific conditions of a given desalination project should be based on
life-cycle cost-benefit analysis and environmental impact assessment including all key project
components i.e. intake, pretreatment, membrane salt separation, post-treatment, and concentrate
disposal.
Intake selection should be based on a reasonable balance between the costs and environmental
impacts associated with production of water. While thorough feasibility evaluation of intake
alternatives is warranted, this evaluation should be initiated with pre-screening for fatal flaws
based on site specific studies for the selected intake location. If the pre-screening shows that
certain intake alternatives have one or more fatal flaws that preclude their use, such intake
systems should be removed from the evaluation process.
Selection of saline water source and thorough analysis of its water quality are of critical
importance for the successful planning, implementation and long-term operation of desalination
projects. Source water quality has impact on all key treatment processes of desalination plants
and their construction and operation costs.
Typically, the content of total dissolved solids, as well as the concentration of main ions
(sodium, calcium, magnesium, bromide, boron, chlorides, sulfates, carbonates, and bicarbonates)
in the source water are of prime importance for planning of both brackish and seawater
desalination plants. These parameters, along with water temperature and pH, drive the design
and configuration of the reverse osmosis system of most desalination projects since usually, over
60% of plant construction costs and O&M expenditures are associated with the RO facilities. If
brackish water of adequate quality and yield is available in the service area, the construction of
brackish rather than seawater desalination plant will usually be less costly, and is almost always
preferred. Mostly, such choice is limited by the availability of suitable, plentiful brackish water
sources especially for larger desalination projects. Compared to the ocean, which is practically
“limitless” source of drinking water, confined groundwater aquifers have volume and recharge
rate constrains, which limit the safe yield of the wells used for groundwater collection and the
associated desalination plant capacity.
Groundwater collected via subsurface intakes can contain high concentrations of dissolved and
colloidal iron and manganese in reduced form, colloidal silica, nitrates, ammonia, cyanide and
radionuclides and may have a very low level of oxygen. The concentrations of these source
water quality constituents must be evaluated in detail as they have significant impact on
desalination project planning and design.
Furthermore, some brackish groundwater sources may have an elevated concentration of natural
organic matter that causes discoloration of the source water or may contain odorous gases such
as hydrogen sulfide. Presence of such contaminants at high levels typically requires additional
treatment to produce desalinated water of drinking quality.
On the other hand, surface saline water sources (including brackish lake or river waters, or open
ocean seawater) can periodically be exposed to algal blooms or contain floating oil, grease and
hydrocarbons as well as man-made pollutants originating from surface wastewater discharges.
In addition surface source waters can have high levels of suspended solids and nutrients which
typically originate from surface water runoff and/or anthropogenic contamination. Such source
water quality contaminants must be taken into consideration in project planning and can impact
on plant design and costs.
Since water quality can vary significantly over time, source water characterization should
encompass both typical conditions and events which result in extremely low of high values of the
water quality parameters discussed above. These include heavy storms and ship traffic, dredging
of the intake area, algal blooms, seasonal changes in the direction of underwater currents and
near-shore winds, and periodic industrial and municipal wastewater discharges which fluctuate in
volume and quality daily and seasonally.
Source water quality has a measurable impact on the cost of production of desalinated water. In
general, both construction and O&M costs increase with the increase in the total dissolved solids
concentration of the source water and decrease in water temperature. Source seawater TDS
concentration is directly related to the SWRO system design feed pressure, the overall plant
design recovery and configuration. The use of lower salinity source water typically allows
reducing the costs associated with construction and operation of the RO system and at the same
time to increase plant recovery.
The consistency of the desalination source water quality is often equally important for a
successful desalination plant design and operation, as is the level of TDS in the source water.
For example, construction of a plant intake near the confluence of river into the ocean can reduce
the overall source water salinity and therefore, can decrease the plant’s total energy use.
However, if the river water carries heavy loads of turbidity, organics, nutrients, and man-made
pollutants the removal of the contaminants contributed by the river water may require a more
elaborate desalination pretreatment, which in turn may negate the cost savings from use of lower
salinity water.
Concentrations of TDS (300 to 500 mg/L), chlorides (150 to 240 mg/L) and sodium (90 to 180
mg/L) in the permeate generated by single-pass SWRO systems are typically within US EPA
regulatory requirements and WHO drinking water quality guidelines. However, if such water is
intended to be used for irrigation of crops sensitive to salinity (i.e. avocados, strawberries) and/or
ornamental plants (some species of palm trees, flowers or grasses), the introduction of this
desalinated water into the distribution system may pose potential challenges unless the
desalinated water is diluted by other water sources in the distribution system to below 250 mg/L
of TDS, 120 mg/L of chlorides and 80 mg/L of sodium, respectively.
Alternatively, seawater treatment by two-pass RO system can produce water quality suitable for
all municipal, agricultural and horticultural uses. Similarly, treatment of brackish water with
typical brackish RO membranes in single-pass RO systems can meet the product water quality
targets for TDS, sodium and chloride listed above.
In addition, the ratio of sodium (Na+), calcium (Ca2+) and magnesium (Mg2+) ions in irrigation
water, referred to as Sodium Adsorption Ratio (SAR), can also impact on some crops.
Excessively high SAR value (i.e. high level of sodium and low content of calcium and
magnesium) contributes to soil dispersion and structural breakdown, which in turn results in
filling up of the soil pores with finer soil particles, and ultimately in reduced water and nutrient
infiltration rates, and reduced crop yield.
The permeate produced from seawater by single-pass RO systems is relatively high in sodium
and very low in calcium and magnesium as compared to traditional water supply sources. As a
result, the SAR value of this permeate is usually unacceptably high (8 to 12 meq/L) for direct
agricultural irrigation of most crops. However, RO permeate post-treatment including calcium
addition and as needed second RO pass treatment, allows reducing the SAR of the desalinated
water to acceptable levels of 4 to 6 meq/L or less which makes it suitable for agricultural
irrigation
While still under debate, there are recommendations from the World Health Organization and the
agricultural community to establish a minimum limit for magnesium in public water supply
based on its benefits to human health and agriculture. Typically, desalinated water contains
magnesium levels of less than 2 mg/L while mineral supplementation to enhance human health
protection and nutrient value for agricultural application is recommended at levels of 5 to 10
mg/L and 15 to 20 mg/L, respectively.
The levels of boron and bromide in desalinated water are usually an order of magnitude higher
than these in conventional fresh water sources. For example, typical river water has boron
concentration of 0.05 to 0.20 mg/L, while source seawater boron levels are usually between 4.0
and 6.0 mg/L, and the boron content of desalinated water treated by a single-pass SWRO system
is usually between 0.7 and 1.5 mg/L. Two-pass SWRO systems typically produce water of
boron levels between 0.3 and 0.5 mg/L. Both of these ranges are well within the boron limit of
2.4 mg/L recommended in the 2011 WHO drinking water guidelines.
The US EPA and all other states except California have not established drinking water regulatory
requirements for boron. The California Department of Public Health has a boron action level of
1 mg/L. Some countries such as Israel, Cyprus, Qatar, Bahrain, and UAE have boron limits in
the drinking water of 0.5 mg/L or less. The low boron limits in these countries are driven by the
use of a large portion of the desalinated water for irrigation of citrus fruits (oranges, lemons,
lime, etc.) and by the need to improve the quality of traditional water sources which have
naturally high boron levels. Boron concentrations above 0.5 mg/L are known to have a negative
impact on citrus fruit yield, size and color.
Bromide levels in fresh water sources are usually between 0.05 and 0.30 mg/L, while source
seawater has bromide concentration of 55 to 85 mg/L. The content of bromide in permeate
produced by single pass SWRO systems is typically between 0.6 and 0.9 mg/L. Two-pass
SWRO systems can produce bromide levels in a range of 0.2 to 0.4 mg/L.
Drinking water can exhibit unpleasant taste and odor changes if desalinated water with bromide
concentration of 0.4 mg/L or more is blended with other water sources that contain phenols.
Bromide concentration of desalinated seawater may also have a significant effect on the finished
water quality if this water is disinfected using chloramines rather than chlorine, or if it is
ozonated.
Disinfection of desalinated water with chlorine only (in the form of chlorine gas or sodium
hypochlorite) creates very stable chlorine residual that shows minimal decay over long periods of
time (60 days or more). Therefore, when desalinated water is used as the main source of water
supply in a given service area, chlorination (rather than chloramination) is the most commonly
applied disinfection method.
Ozonation of desalinated water with bromide concentration of 0.4 mg/L or more may result in
the formation of an unacceptably high concentration of bromate which exceeds most drinking
water regulations worldwide, which stipulate maximum bromate limit of 10 µg/L.
Specific class of organic compounds that are of concern for the quality of desalinated water
originating from surface water sources are low-molecular weight toxins such as domoic acid and
saxitoxin, often referred to as algal toxins. Such toxins are generated during algal blooms when
the concentration of algae in the seawater increases several hundred times.
Algal bloom events are accompanied with an overall deterioration of source seawater quality,
including water discoloration, oxygen depletion and elevated content of organics released from
algal cell decay. Certain algae, such as Pseudo-nitzshia serata have red pigmentation and their
excessive growth during algal blooms results in reddish discoloration of the seawater - red tide -
as illustrated in Figure 5.
While a number of other organic toxins such as yessotoxin, okadaic acid, brevetoxin, microcystin
and nodularin, are generated during algal blooms, domoic acid and saxitoxin are two algal toxins
of specific interest as their molecular weight and size are comparable with the average molecule
rejection size (molecular cut-off) of the SWRO membranes and theoretically some of them might
pass through the membranes. Domoic acid is of particular concern because it concentrates up to
several hundred times in shellfish and when ingested with the contaminated shellfish it can cause
amnesic shellfish poisoning.
A common practice of health departments of coastal United States, Australia, and other
countries worldwide is to monitor concentration of domoic acid in shellfish tissue and to issue
advisories for temporary discontinuation of shellfish harvesting when the concentration of
domoic acid in the tissue exceeds 80 µg/L.
Despite the small molecular weight of some of the algal toxins, reverse osmosis membranes can
completely reject such organic compounds and thereby can produce safe drinking water even
when the source seawater is exposed to heavy algal blooms. Rejection of domoic acid and
saxitoxin by SWRO membranes have been studied at the West Basin Municipal Water District
and Carlsbad pilot SWRO plants in Southern California, US in 2005 during a 50-year red-tide
algal bloom. The test results at both facilities indicate that the two algal toxins are completely
rejected by the SWRO membranes and permeate produced by these membranes is safe for
human consumption.
Pathogens. While SWRO membranes are not an absolute barrier for microbial contaminants,
they are typically expected to achieve 4 to 6-logs of pathogen removal or more. Pretreatment
filtration system upstream of the RO desalination membranes typically provides an additional 2
to 4-logs of pathogen removal.
16
14
12
10
Log Removal
0
CMF UF SWRO CMF- UF- CMF UF SWRO CMF- UF- CMF UF SWRO CMF- UF-
SWRO SWRO SWRO SWRO SWRO SWRO
MS-2 PRD1 Fr
Table 2 summarizes the minimum and maximum reduction requirements and the credits given to
typical treatment processes employed in seawater and brackish water desalination. Analysis of
Table 2 indicates that a typical SWRO desalination plant including conventional pretreatment,
followed by RO membrane system and chlorine disinfection can be assigned a total of 6-log
virus removal credit, 5-log giardia removal credit, and 4-log Cryptosporidium removal credit,
which matches closely the maximum log reduction requirements that may be imposed on a
desalination project even under worst-case scenario source water quality.
0.5 - 1-
Viruses 4 6 1 2 2 0 2
1(4) 4(4)
While SWRO membranes can consistently provide over 4 logs (99.99%) of pathogen rejection,
due to the lack of standard procedures for RO membrane integrity testing at present, often
SWRO membrane systems are typically credited with only 2-log pathogen removal by the
regulatory agencies involved in public health protection. The two-log removal credit of SWRO
systems is assigned based on the continuous monitoring of the actual membrane TDS log
removal (measured as conductivity log removal).
Since SWRO membrane systems typically remove at least two logs (99%) of the source water
salinity, TDS removal in this case is used as a conservative surrogate measure of pathogen
removal. As the desalination industry evolves, it is anticipated that alternative membrane
integrity test procedures will be developed in the future, which will allow assignment of
significantly higher pathogen removal credit to SWRO membranes, reflective of their actual
ability to provide very high levels of pathogen removal.
Organics. Similar to SWRO membranes, brackish water RO membranes are also capable of
removing over 90% of most organics contained in the source water. Due to their higher
molecular weight cutoff, brackish RO membranes typically have lower rejection of organic
compounds characterized by small molecular weight.
Pathogens. Brackish water RO membranes can provide over 4 logs of pathogen rejection.
However, similar to SWRO membranes they are often credited with only one or two logs of
pathogen removal due to the lack of standard on-line testing method that allows continuous
monitoring of their actual pathogen removal and integrity.
Organic content of desalinated water is usually an order-of-magnitude lower than that of most
fresh surface water sources, and thus has a significantly lower organics-related DBP formation
potential than traditional fresh water supplies. While RO membranes reject most organics in the
source water, the process is not as efficient in removing a DBPs, which are formed when
chlorine is used for source water pretreatment.
Brackish water membranes are less efficient in terms of DBP removal than seawater RO
membranes. In addition, because the BWRO permeate is more often blended with source water,
this blend may require enhanced post-treatment to reduce DBPs.
Blending of low-DBP desalinated seawater with surface water of high DBP content can reduce
the overall DBP concentration of the drinking water. However, as indicated previously, when
desalinated water has a high content of some unwanted minerals, such as bromide, boron,
sodium, and chlorides, mixing this water with drinking water originating from other sources may
have a negative impact on the blended water quality. The compatibility of the blended water
sources must be taken into consideration.
Potential differences in bromide and TOC levels in the blended waters may have an effect on the
DBP concentration of the blend. The type of disinfection used for the various water sources may
impact DBP formation and chlorine residual stability.
Before blending, desalinated water usually has significantly lower levels of calcium and
magnesium ions as well as low alkalinity concentration as compared with fresh surface water
sources. Blending of desalinated water with drinking water of high hardness and high alkalinity
may be sufficient to provide the needed chemical stability if the blended water meets target water
quality requirements for corrosion control.
reduce boron concentrations may be necessary to achieve suitable levels if reclaimed water of
elevated boron content is to be used for irrigation of sensitive plants and crops.
Low Alkalinity Impact on WWTP Nitrification. Often, desalinated water has lower alkalinity
compared to other water sources. In such cases, introduction of desalinated water to the
distribution system will lower the alkalinity of the influent of the wastewater treatment system
processing such water. Wastewater alkalinity concentration is very important if the treatment
plant has biological nitrification system, which consumes 7.14 mg of alkalinity (as CaCO3) for
every milligram of nitrified ammonia (as N) contained in the wastewater.
While alkalinity is added to the desalinated product water at a dosage of 40 to 100 mg/L for
corrosion protection, such dosage is often inadequate to sustain the WWTP nitrification process,
even though wastewater alkalinity is typically 100 to 150 mg/L higher than that of the drinking
water. Possible solutions include increasing the desalinated product water alkalinity, employing
biological denitrification in the activated sludge system of the wastewater plant, and directly
increasing alkalinity of the wastewater treatment plant influent by feeding strongly basic
conditioning chemical, such as sodium or calcium hydroxide.
As discussed previously, two other key factors that influence the selection of the target
desalination plant product water quality are: (1) the content of specific minerals (i.e., sodium,
chloride, boron and bromides) in the water, and (2) the overall water production costs. These
two factors are inter-related – production of higher quality desalinated water is possible at
incrementally (15 to 50 %) higher cost.
In most municipal applications, desalination plants are designed to produce water of quality
compliant with drinking water regulations, especially if this water is the main source of supply
for the service area. However, many utilities and municipalities worldwide use desalinated water
as a supplemental source of water supply only and this water is blended in the distribution
system with other existing water supplies.
The quality of the desalinated water can be adjusted to a target level more easily than the quality
of other traditional water sources, and desalinated water is often produced at quality higher than
that of the other sources and subsequently blended to improve the final product water. While
this water quality improvement approach results in higher production cost of desalinated water, it
is often the most cost effective overall strategy for improvement of water quality in the entire
distribution system.
Another approach to determine the target desalination project product water quality is to try to
match it as close as possible with the quality of the other water resources delivered to the same
service area. This approach, while usually more costly, simplifies the decision making-process
in terms of potential modifications which will have to be made to the existing distribution system
operations and water quality.
In addition to the municipal uses discussed above, the target desalinated water quality may be
driven to even higher levels of salt removal by the needs of some industrial applications,
especially these where ultrapure water quality is necessary. Such applications may need the
enhanced removal of sodium, silica, specific ions, oxygen, and other water quality constituents
which will require RO permeate treatment through one or more additional water quality
polishing processes such as ion exchange, activated carbon adsorption, advanced oxidation, etc.
Such water quality polishing steps can sometimes double desalinated water costs as compared to
expenditures associated with producing drinking water for potable use.
6. PLANT DISCHARGE
Typically, both brackish and seawater RO desalination plants generate three key waste streams:
• concentrate (brine) which usually has 1.5 to 5 times higher salinity than the saline source
water;
• spent filter backwash water from the plant pretreatment facilities which has the same
salinity as the source water; and
• spent chemicals and flush water from periodic RO membrane cleaning which usually are
of lower salinity than the source water.
Of these three streams concentrate is by far the largest in volume with potential negative
environmental impacts. Therefore, engineering practitioners sometimes refer to desalination
plant discharge as concentrate discharge, although they are not truly synonymous.
Management of concentrate and other waste streams associated with production of desalinated
water is one of the important project planning factors that determine plant location, size and
treatment processes. For a potential desalination plant site to be feasible, it has to be located
within a reasonable distance (typically 0.5 to 10.0 km/0.3 to 6.1 miles) away from a suitable
site/s for concentrate disposal. For a plant discharge disposal site to be suitable, it will have to
have physical configuration and receiving capacity that allow for the concentrate and, if possible,
other plant waste streams to be continuously disposed in an environmentally safe manner for the
entire duration of the useful life of the project.
The most common method for disposal of concentrate from seawater desalination plants is
surface water discharge via ocean outfall. Common concentrate discharge alternatives for
BWRO desalination plants are: discharge to sanitary sewer, deep well injection and evaporation
ponds. Other concentrate management methods which are not as widely practiced include spray
irrigation, zero-liquid discharge (ZLD) by concentrate evaporation and salt crystallization, and
beneficial use of concentrate/ocean brine mining. Such methods are either very costly (e.g.,
ZLD) or seasonal in nature (spray irrigation and some methods for beneficial reuse). Depending
on the size of the project, especially for larger inland brackish water desalination projects, it may
not be possible to apply a single method for concentrate disposal, and often such projects rely on
multiple disposal alternatives.
Desalination project planning activities associated with concentrate disposal include water
quality characterization of concentrate and other waste streams generated by the desalination
plant, development of feasible alternatives for management of the desalination plant waste
streams and selection of the most viable desalination plant discharge management alternative
based on environmental impact and life-cycle cost analyses.
6.1. Concentrate
Concentrate is the largest waste discharge stream in desalination plants. The volume of
concentrate generated by seawater desalination plants is significant as a typical SWRO
separation process converts only 40 to 55% of the source water into desalinated freshwater. The
remaining source water is rejected as concentrate. Seawater concentrate contains over 99% of all
source seawater salts and dissolved constituents and its mineral content is between 1.5 and 2.0
times higher than that of the source seawater.
BWRO plants usually convert 70 to 90% of the source water into fresh water and generate
relatively smaller volumes of concentrate compared to SWRO plants of the same production
capacity. However, the mineral content of brackish water concentrate per unit volume is
typically 2.5 to 6.5 times higher than that of the source water.
Concentrate water quality is largely determined by the quality of source water and the design of
the desalination plant and therefore can be projected based on characterization of the source
water quality. Open ocean seawater quality is usually very consistent and over 98% of seawater
concentrate salinity is attributed to five dissolved minerals: sodium, chloride, sulfate, magnesium
and calcium. The characterization of seawater concentrate focuses on the measurement of the
concentration of these minerals, the total content of dissolved solids, conductivity, pH,
temperature, turbidity, silt density index, total suspended solids and oxygen content, and
concentration of organic and inorganic contaminants defined by the regulatory requirements
pertinent to the discharge area.
Water quality of the concentrate generated by SWRO desalination plants with subsurface (e.g..,
well) intakes is dependent on whether the coastal source water aquifer is influenced by
contaminants present in surrounding aquifers. For example, alluvial aquifers often contain
elevated concentrations of colloidal iron and manganese and have very low levels of oxygen,
which may have a dramatic impact on the source and product water quality and on the plant
concentrate. Therefore, such aquifers should also be characterized thoroughly during the
planning phase of the desalination project.
Water quality of brackish water desalination processes may vary significantly between locations
and may contain additional constituents, such as colloidal iron, manganese, silica, nitrates,
phosphates, arsenic, cyanide, ammonia, and organics. BWRO concentrate may be dominated by
sodium or calcium cations and chloride, sulfate, or bicarbonate anions. Groundwater based
concentrate frequently has high levels of carbon dioxide (CO2) and hydrogen sulfide (H2S),
which require degasification prior to discharge. Low oxygen levels in concentrate resulting from
desalination of groundwater sources may also require aeration or other means to increase
dissolved oxygen prior to discharge. Therefore, these water quality parameters will have to be
included in the source water quality characterization.
In addition to the above water quality parameters, desalination plant concentrate should also be
analyzed for acute and chronic whole effluent toxicity (WET). This determines the potential
synergistic environmental impacts of various contaminants contained in the concentrate. Whole
effluent toxicity of the concentrate is difficult to predict based on chemical characterization of
the saline source water only. While acute and chronic WET thresholds of the concentrate can
typically be correlated with the level of salinity in the source water, some aquatic species can
also be impacted by the ion makeup of the concentrate (i.e. the relative ratios of ions such as
calcium, magnesium, sodium, etc.). Therefore, the most reliable and thorough characterization
of the concentrate water quality is achieved by pilot testing of desalination systems with
configuration, design and operational conditions similar to these planned for the full scale
desalination plant. The concentrate generated from this pilot plant is then analyzed for all
government regulated discharge water quality parameters, including WET.
The amount of solids contained in the spent filter backwash water are dependent on the source
water quality and type of pretreatment system employed (granular or membrane filters).
Typically, membrane-based pretreatment systems produce larger volumes of backwash water
(1.5 to 2 times), but require less, if any, coagulant compared to granular filters, which tend to
generate waste streams with a higher proportion of solid constituents. Depending on the
pretreatment system, the waste stream may contain iron salts used as coagulants, in addition to
suspended solids (for example debris, silt, shell particles) naturally occurring in the source water.
Often spent pretreatment filter backwash water (with or without treatment) is blended and
discharged along with the concentrate. The blended plant discharge may contain elevated
turbidity, total suspended solids, color, organic content, iron and manganese, and biochemical
oxygen demand. The concentration of each contaminant of concern in the blend can be
calculated as a flow-weighted average of the concentrations of the same contaminant in the
individual waste streams. Alternatively, if a desalination pilot plant is available, the water
quality of the mixed plant discharge can be determined by direct sampling and laboratory
analysis.