Lawrence Owens, California State University Fresno, Visalia, California, USA
Walter J. Ochs, Water Management Consultant, Virginia, USA
Physical, chemical and biological treatment processes
Treatment in constructed wetlands
Selection of treatment process
Methods of treatment
A treatment example
The first portion of this chapter concerns the direct treatment of agricultural drainage water to improve its quality. The general process of selecting a treatment process is described, followed by a description of treatment methodologies.
The first steps in the selection of any treatment process for improving drainage water quality are to thoroughly define the problem and to determine what the treatment process is to achieve. In most cases, either regulatory requirements or the desire to re-use the water will be the driving force in defining the treatment issue(s) to be selected for a particular drainage water. A thorough knowledge and understanding of these water quality criteria is required prior to selecting any particular treatment process. Most of the treatments discussed will not reduce the salt concentration in the water and some may result in increased salt concentration.
Historical data of the chemical constituents in the drainage water are used for decision making. Once the framework that governs the selection of the treatment process has been identified (e.g., regulatory requirements, re-use, or both), then the available data must be evaluated as to its adequacy for the preliminary selection of possible treatment options. In general, a more complete data record, in terms of both the length of record and the constituents monitored, will make the task of treatment selection easier. As a minimum, one would need the ranges of expected flow volume and concentrations of the constituents of concern. In addition, knowledge of any seasonal variations of flow or constituents would be highly desirable. If data are not available, a portion of the project budget should be allotted to obtain this information in order to avoid costly mistakes at a later stage in the process. It is also necessary, when using historical data, to establish the reliability of the data. This can be done by reviewing the analytical procedures used, by comparing data with an independent source, or by comparing historical values with more recent data.
The next step in developing a successful treatment programme is to identify the selection criteria. This will require input from all parties involved in the project including the funding source(s), users (farmers, industries or municipalities), water districts, regulatory agencies and the public. Criteria that will affect process selection include: capital cost, O&M costs, land requirements, level of treatment required and regulatory constraints. These criteria need to be defined as closely as possible in the earliest stages of process selection. However, it is important to note that there will undoubtedly be modifications to the criteria as the project progresses.
Once the selection criteria have been defined, possible treatment options can be identified. A good way to begin is with one or more 'brainstorming' sessions where any possible options are first identified without any suggestion being rejected. It is important to have input from individuals with experience and expertise in a variety of areas, so as not to 'zero-in' on a particular type of technology initially. This list of all possible options should then be evaluated against the identified selection criteria and ranked according to how each option meets the criteria. It is also appropriate to evaluate the known advantages and disadvantages of each option. From these evaluations, there may be one clear option to pursue or there may be several options which meet the selection criteria. If there are several options, then additional, more stringent criteria should be used to narrow the field down to two or three preliminary selections.
Unless the preliminary option or options are standard processes in use with drainage water or have been successfully used in applications with water very similar to the water in question, it is highly advisable to conduct pilot-scale tests. In general, the larger the pilot system, the better it will indicate the efficiency of the process when built at full scale. The single most important issue in pilot-scale operation is to use the water which will actually be treated by the full-scale process. Drainage water is such a complex solution that synthetic mixes, while useful for initial and fundamental laboratory studies, are poor substitutes for evaluating actual treatment processes. Conditions in the pilot-scale system should approximate as closely as possible the operating conditions of a full-scale system. Hydraulic retention time (residence time), flow velocities, applicable loading rates and removal efficiencies of target constituents are typical parameters of concern to be measured and evaluated in the pilot testing. If the pilot tests do not show that the treatment process can achieve the desired results, then the process must be modified and retested or other options must be investigated.
Once pilot testing is complete and there is at least one process that has been shown to be effective, then the best option can be selected for full-scale implementation. The options should be judged against the selection criteria developed early in the process. At this point, other factors concerning the processes should also be evaluated such as: complexity; number and expertise of personnel required to operate and maintain the system; waste products produced and their disposal; expandability of the system; ability of the system to remove the target constituents to below the currently desired levels (e.g., in case regulations become more stringent); and ability of the system to remove constituents which are not currently of concern (e.g., in case new constituents become regulated). The consensus of as many of the involved parties as possible is desirable in the final selection of the best treatment option. Achieving consensus at this point will avoid conflict in the future.
While a detailed description of each type of process is beyond the scope of these Guidelines, an overview will be given of the most common treatment methods which have application in treating agricultural drainage water. More detailed descriptions and design details of common treatment processes can be found in water and wastewater treatment texts, such as those by Montgomery (1985) and Metcalf and Eddy (1991).
Treatment approaches can be divided into three general types: physical, chemical and biological. Many processes exhibit both physical and chemical aspects and so are sometimes called physical/chemical or physicochemical treatment.
Several physical processes aim to remove suspended particulate matter. While subsurface drainage water itself is usually low in suspended particles, these processes might be used in an overall treatment process for the removal of particulates formed in other stages of the treatment, such as removal of bacteria from a biological system or removal of precipitates formed in a chemical treatment process. Particle removal unit processes include sedimentation, flotation, centrifugation and filtration. Filtration further includes granular media beds, vacuum filters, belt presses and filter presses.
Adsorption is the process of removing soluble contaminants by attachment to a solid. A common example is the removal of soluble organic compounds via adsorption onto granular activated carbon (GAC). GAC is useful for its ability to remove a wide range of contaminants. Certainly, if pesticides were a concern for the drainage water being examined, the use of GAC adsorption would be a leading candidate for treatment.
Another possible treatment for removing volatile compounds from water is air stripping. In a conventional countercurrent air stripping operation, the contaminated water is distributed at the top of a tall reactor vessel that is packed with materials or structures with a high surface area and void ratio. As the water advances downward, clean air is introduced at the bottom of the reactor and moves upward. As the water and air make contact, volatile compounds are transferred from the liquid phase to the gas phase according to gas transfer theory.
If removal of salts and production of a high quality water is the treatment objective, membrane processes or distillation (discussed below) will be leading process candidates. The separation of salts and organic compounds can be accomplished by using a selectively permeable membrane. Membrane processes are also finding use in the water treatment industry for removing particulates and microbial contaminants such as Giardia and Cryptosporidium (Jacangelo et al., 1991). Membrane processes can be divided into three main categories: dialysis, electrodialysis and reverse osmosis. Each of these processes requires some type of driving force energy to separate the contaminants from the clean water. For dialysis, the driving force is the difference in concentrations of the contaminant across the membrane (a concentration gradient). For electrodialysis, the driving force is an electrical potential. For reverse osmosis, the driving force is applied pressure. The use of membrane processes for salt removal has generally been considered too expensive for drainage water, but new developments in membrane technology may make this an attractive option for treating at least a portion of the flow to reduce the total salt concentration. The use of a membrane process such as microfiltration for particle removal may also be a cost-effective alternative as part of an overall treatment system.
Distillation is a thermal process used for salt removal. Heat is used to vaporize the water, leaving the salts behind. The water vapour is condensed to a high quality water. Distillation is energy intensive and has largely been replaced by reverse osmosis for desalination applications.
Coagulation and flocculation
Coagulation and flocculation are used to remove particles of all types from water. The particles might be colloids present in the drainage water that are too small to remove by gravity settling or filtration, or they might be colloidal precipitates formed during a treatment process. Coagulation is the process by which these small particles are destabilized and the initial aggregation of the destabilized particles into larger particles called floes. Coagulation is accomplished by the addition of a coagulant, which can be either an inorganic metallic salt such as alum (aluminium sulphate) or ferric chloride, or a high molecular weight organic polymer. The coagulant serves to neutralize interparticle charge repulsion and to enmesh the particles into an aggregated floe. Flocculation is a slow mixing of the particles to bring them into contact with one another to form even larger particles. The objective is to produce a large fast-settling floe.
Certain compounds, especially metals, can be removed by changing their solubility to cause their precipitation. Many metals can be precipitated as a metal hydroxide by increasing pH with lime or caustic soda to achieve the pH of minimum solubility. These would include Cr, Ni, Cu, Fe, Pb and Hg, and other elements such as As. The pH of minimum solubility varies according to the metal in question. Precipitation can also occur by the formation of insoluble compounds through adding certain chemicals, or from chemical species formed during another treatment step. Such is the case with sulphides and carbonates, which can be formed during biological treatment and combine with metals and some cations to form precipitates. Precipitation can lead to a slight net decrease in TDS.
Ion exchange involves the chemical exchange of ions (charged dissolved molecules or atoms) in solution with ions on a solid phase. The solid phase, usually a synthetic organic resin, is chosen to specifically adsorb the constituent(s) of interest. To maintain the electrical charge balance, the resin must release an equal amount of charge into solution. For example, in water softening, a cation exchange resin initially holds Na+, but releases them as it adsorbs Ca2+ and Mg2+ from the water stream moving through the resin. Anion exchange resins are also available. For either type of resin, there is a preferential adsorption of either the cations or the anions if more than one type is present. For example, calcium is more strongly adsorbed than magnesium, and sulphate is more strongly adsorbed than nitrate. Certain resins have been developed to maximize their affinity for a specific ion (e.g., borate).
Advanced oxidation processes
One major disadvantage of both GAC adsorption and air stripping for organic compounds is that they only transfer the contaminant from one phase (water) to another phase (carbon or air, respectively). Advanced oxidation processes (AOPs) are capable of compound destruction, or more accurately, the mineralization of chlorinated organic compounds to non-toxic constituents such as carbon dioxide, water and chlorides. AOPs rely on the production of highly reactive radicals to break down the organic compounds. AOPs are commonly based on the use of hydrogen peroxide (H2O2) or ozone (O3) in combination with ultraviolet (UV) light to cause radical formation. Another type of AOP uses photoactive metal catalysts and UV light to generate the radicals (Suri et al., 1993). Their high cost is the primary disadvantage of most AOPs.
Biological treatment can be a useful tool in drainage water treatment for the removal of both organic and inorganic contaminants. Biological treatment usually refers to the use of bacteria in engineered reactor systems for effecting the removal or change of certain constituents, such as organic compounds, trace elements and nutrients. Algae have also been used and natural wetlands systems can be used in some cases to replace conventional reactors. The bacterial reactions involved can be divided into two major categories according to the use of oxygen by the bacteria. In aerobic systems, O2 is provided and used by the bacteria to biochemically oxidize organic compounds to carbon dioxide and water, and possibly to oxidize reduced compounds before their release to the environment. Aerobic systems are usually odour free. In an aerobic system, oxygen is the electron acceptor and organic carbon sources are usually the electron donors in the biochemical reactions that take place. In an anaerobic system, oxygen is excluded and the bacteria utilize compounds other than molecular oxygen for the completion of metabolic processes.
Biological reactor types can be broadly divided into two types: suspended growth and attached growth. In suspended growth systems, the bacteria are grown and maintained in suspension by mixing of the liquid contents. In attached growth systems, the bacteria grow in a thin layer (a biofilm) on an inert support, such as plastic media or fluidized sand. Both aerobic and anaerobic systems can have either suspended or attached growth depending upon the reactor configuration. For a discussion of specific reactor types, see Metcalf and Eddy (1991).
An example of drainage water treatment in the San Joaquin Valley of California will illustrate the combination of several of the processes discussed above. Work conducted by Squires of EPOC AG (1987), Macy et al. (1993) and Owens et al. (1995) shows the successful use of anaerobic biological treatment for the removal of Se from agricultural drainage water. In the process, an oxidized, soluble form of Se (selenate) is biochemically reduced to the insoluble elemental Se form. The process occurs through anaerobic respiration, with an added carbon source such as methanol or acetate acting as the electron donor and the selenate acting as the electron acceptor. In the process, nitrate is also removed by reduction to nitrogen gas (here the nitrogen in the nitrate is serving as the electron acceptor).
The particulate elemental Se can then be removed by membrane microfiltration (EPOC AG, 1987) and also by slow sand media filtration (Dhaliwal, 1992). Experiments using ferric chloride as a coagulant (Salamor et al., 1996) show improved particulate Se removal. As an added benefit, the iron hydroxide precipitate adsorbs soluble selenite, which is an intermediate formed during the biochemical reduction of selenate.
Flow-through wetland functions
Planning and design of flow-through wetlands
Hydraulic and geohydraulic characteristics
Soils and geologic characteristics
Implementation, monitoring and management
The protection of wetlands is one area of public concern where agricultural and environmental values need to be further harmonized, shared and promoted. Ecologically sensitive wetlands need to be distinguished from poorly drained, wet, depressional cropland. The latter is farmland whose productivity can only be improved through better drainage. Wetlands which are transitional zones between terrestrial and aquatic systems serve valuable ecological functions.
The water table is usually near the surface, and such wetlands support fish, wildlife, waterfowl and aquatic plants. They also trap sediment and pollutants, and cycle nutrients. Wetlands are of hydrologic value in that they serve as flood control systems, recharge groundwater, and maintain instream flows. They also have economic value if they support recreation and other economic functions. A challenge is to ensure that agricultural and drainage activities do not negate the environmental value of wetlands. Cropland and wetlands must be viewed as part of a viable and sustainable ecosystem.
Constructed wetlands for domestic and industrial wastewater treatment are gaining acceptance in many countries. They can also be beneficial in providing treatment of agricultural drainage wastewater under many circumstances (Post and Ochs, 1995). Flow-through wetlands are one option for the management of agricultural drainage water. They can be described as constructed wetlands or natural depressional areas prepared to facilitate the movement of surface waters through specially selected vegetation. Flow-through wetlands are an alternative to evaporation ponds, and should be considered. Where the water quality is appropriate, they are preferable to evaporation ponds. Evaporation tends to concentrate the ions contained in the drainage water, and some of these substances can become toxic in the ecosystem. Concentration of the ions reduces the opportunity for beneficial use of the drainage water in downstream areas by degrading receiving waterways or water bodies. Wetlands can reduce some pollutants such as N, P and sediment, and improve water quality. Wetlands consume water and will reduce the total available water for downstream users even though the general water quality may be improved. Water use by wetlands is approximately equal to potential ET rates.
The primary values of flow-through wetlands for drainage water quality improvement from irrigated areas are to:
i. reduce the evaporation opportunity time which will minimize the direct concentration of salts and minor constituents which may become toxic when concentrated;
ii. provide physical filtration and sedimentation of soil particles and attached contaminants; and
iii. provide vegetation to remove excess N, P, K and organic wastes.
Secondary values in the use of flow-through wetlands are for wildlife habitat, recharge of groundwater aquifers, wind erosion control, and fish or shellfish production by:
i. providing breeding, nesting, feeding and cover habitats for invertebrates, insects amphibians, reptiles, birds and mammals;
ii. providing water detention that will facilitate infiltration of surface waters through the soil profile with opportunity for degradation of chemicals and vegetative uptake of pollutants;
iii. reducing wind erosion in the vicinity of wetlands, thus preventing wind blown contaminants in the air; and
iv. providing potential for fish or shellfish production if the pollutants present will not cause health problems to consumers.
Site selection, data collection and the operational needs are of critical importance when planning and designing flow-through wetlands for drainage water management. Wetlands are often created to provide waterfowl habitat. However, consideration must also be given to the physical benefits achievable if the wetland or series of wetlands are developed to improve the quality of drainage waters. The topographic, soil and geologic characteristics will have a great impact on the effectiveness of pollution control benefits from the completed system. The vegetative materials selected for growth within the wetland will determine the success of the effort. The vegetation must be tolerant to the extremes in salinity and pollution anticipated and must be selected to maximize the uptake of nutrients and pollutants. Normally, a series of wetlands with different vegetative and soil materials will facilitate operational management and improve system effectiveness. Considerations for harvesting the vegetation and management of the vegetation may require interior dikes and gate structures (Post and Ochs, 1995). To provide practical removal of some pollutants, harvesting of the vegetation will be required and economic considerations for disposal or marketing of the vegetation should be evaluated. Plants are being developed that have a superior capability to remove pollutants from soil, such as penny cress and other metal 'scavengers'. These plants can be grown periodically to rejuvenate the wetland and enhance the sustainability of treatment benefits. At the end of the growing season, the plants can be cut, dried and burned. Metals can be extracted from the ashes for recycling.
Site selection involves consideration of hydraulic and geohydraulic characteristics of the soil and subsoil to ensure the success of the system. Flow-through wetlands must be located where the drainage waters to be managed can physically be delivered. The topography must be suitable to allow the construction of dikes and outlet facilities. Proper soil must also be available nearby for embankments and lining the wetted perimeter of embankment areas. The site must have depressional characteristics that are large enough to allow the creation of water bodies with a depth range that will facilitate a vigorous growth of the selected vegetative materials.
Water depths in the wetlands must be controlled to provide the appropriate environment for growing the specific vegetation desired. The water depth and flow regime must also consider vector control and health considerations related to water borne diseases (see Chapter 7). The sizing of each wetland should be related to the opportunity time required for the uptake of the pollutant or pollutants that each vegetative wetland area is designed to reduce. The sizing of the constructed wetlands should also consider the volume of drainage water that will be transported to the wetland system for improving its quality. The water discharged from the wetlands must have a suitable outlet.
Effective constructed wetlands should permit vegetative materials to proliferate. Soils of the wetland should have a permeability which is sufficiently low to minimize percolation. Wetlands should be located in areas that are not porous, as retention of water is the primary characteristic required. If the site has sections of porous soil, then borrow pits with clays or other less porous material may be required to facilitate the lining of these highly permeable areas. As pollutants are removed in the initial wetlands, drainage water is discharged through lower wetlands in the system and it may recharge the groundwater system in the lower areas. Barriers to infiltration thus become an important feature in the design of systems in the upper wetlands and are not necessary in the lower wetlands. The accumulation of dead plant material, suspended solids, algae and other clogging material on the bottom can significantly reduce seepage rates.
Cavernous geologic areas should be avoided unless special precautions for seepage control and dike stability are taken. The expected quality of infiltrated waters from a constructed wetland system and the groundwater must be analysed to ensure that groundwater degradation does not occur.
As far as is both possible and practical, the vegetative types must be carefully selected to facilitate the reduction of pollutants. Normally, a mix of vegetation will optimize the uptake of nutrients. Consultation with plant material experts is essential for the success of wetland systems designed to remove pollutants. Field trials and laboratory testing of the vegetation will often be necessary prior to its introduction. This is particularly true if the vegetation is new to the region. Harvest considerations must also be made when selecting vegetation. The harvested biomass may have some value that will reduce the operational cost of the system.
Grass and reed types of vegetation are normally the most effective for improving water quality. However, some tree species may add diversity and habitat value to the wetland area if included in the site plans. A good source of information on vegetative design for constructed wetlands, as well as other planning and design parameters, can be found in the wetland restoration chapter of the Soil Conservation Service (1992) publication.
The establishing of the proper vegetation type will probably require a temporary water management plan to facilitate early vegetative establishment. As vegetation selection guidelines have not been developed for every climatic region, or for each pollutant, field trials and tests are strongly recommended prior to constructing wetlands.
Pollutants generally move with the water flow, thus slow, continuous, uniform flow will provide the optimum pollutant uptake opportunity for the vegetation. The flow pattern should be designed to minimize evaporation in order to reduce the risk of developing water toxic to the ecosystem and to avoid channelling from inlet to outlet.
The monitoring of water quality parameters is of utmost importance. For this to be done properly, it is important to establish the baseline situation. This means that an effort will be needed to determine the discharge quantity and quality of the water to be treated throughout the year prior to passage through the wetlands. Monitoring will be required to properly manage the wetland and to determine the effectiveness of individual wetlands.
Improved agriculture technology and irrigation water management will probably change the future composition and quantity of the discharged water. Thus, an interactive management programme should be developed that is effective for the conditions of the designed flow-through wetland.
Institutional responsibility is important to the success of wetland systems for water quality and habitat protection. If the water quality control system is to remain effective, it will require numerous adjustments and proper maintenance. It must be managed in accordance with the designs for the primary function of water quality improvement. However, secondary functions should not be ignored and often require some special management to achieve the benefits desired. The institution responsible for the wetland needs to have the financial means to operate, monitor and maintain the system in a sustainable manner. The one constituent that wetlands cannot remove is salt.