Principles of Induced Infiltration and Artificial Recharge

The concept of capturing surface water sources by pumping a well has blossomed into a pragmatic method of developing large-capacity water wells.  J.G. Ferris, E.M. Burt, G.J. Strammel, and E.G. Crosthwaite discussed how both underground and surface water sources are affected by this technique (1954, Ground Water Resources of Southwestern Oakland Co., Michigan, Michigan Department of Conservation Progress Report 16, 54-58).  Their discussion is reproduced here.

Induced infiltration from underground sources

"The regional lowering of groundwater levels in the vicinity of a well development disturbs the initial condition of hydrodynamic equilibrium between the aquifer which is being developed and the overlying and underlying aquifers or confining beds. … Initially, the water in the water-table (unconfined) aquifer stands at a higher level than the piezometric surface (potentiometric surface) in the underlying formations of shale and sandstone.  As a consequence of the higher head, groundwater, may percolate from the sand and gravel formation into the underlying shale and sandstone…".

"When groundwater is pumped from the water-table aquifer at rates sufficient to lower water levels below the piezometric surface of the artesian formation, leakage gradients will be reversed and groundwater now percolates upward from the artesian formation and enters the water-table aquifer if the confining bed is permeable enough. … In areas of extensive development, where a large regional lowering of groundwater level may occur, vertical leakage may be induced from underlying or overlying formations at appreciable rates and over a quite large area.  Under these conditions, as development continues and increases in magnitude, the contribution of groundwater from adjacent formations also increases.  If the groundwaters contained in overlying or underlying formations are of different chemical composition from those in the aquifer in which development occurs, there will result noticeable changes in the quality of groundwater pumped from the aquifer developed.  The rate of vertical leakage is proportional to the rate of pumping from wells in the aquifer of development.  Thus, it would appear that changes in chemical quality will be most marked in the wells or in the areas where withdrawals are greatest. Although this generalization may be appropriate for many wells, for some wells it maybe modified by other conditions such as differences in the degree of penetration of the pumped wells, differences in the vertical permeability from area to area, and the affects of differences in density of the several groundwaters involved."

Induced infiltration from surface sources

"Where an aquifer is intersected by a perennial stream, large groundwater supplies may be developed by the installation of wells or other types of subsurface intakes, which parallel the course of the stream and are at sufficient depth below the stream to permit the development of adequate gradients from the stream to the subsurface intake.  As groundwater is withdrawn, water levels are drawn down in the vicinity of the intake.  As pumping continues, the cone of depression deepens and the area of interception expands.  When the piezometric head in the reservoir, adjacent to the underlying stream, is lowered below stream stage, water from the stream moves down gradient into the aquifer and toward the center of withdrawal…"

"The interception and diversion of surface water by induced infiltration takes advantage of the slow sand filter provided by Nature’s reservoir, as contrasted to the more widely used and more costly procedure of direct surface intake, which requires the construction of extensive filter beds. Of course, Nature did not provide adequate filtering media everywhere adjacent to the streams and in many places the only choice is to construct filter plants.  Unfortunately, man’s awakening to the availability and practical usefulness of groundwater reservoirs as natural filters has been rather slow, and even at this late date is still not complete.  As a consequence, it is not uncommon to find monumental filter plants constructed by man with their foundations seated in natural filter media that dwarf man’s limited efforts."

"In using a groundwater reservoir as a filter medium, careful consideration must be given to the proper location of the underground intakes in order to take full advantage of local hydrogeologic conditions and at the same time obtain adequate filtration.  At first consideration, it may appear advisable to locate the subsurface intake as close to the stream bed as possible, because this would develop the maximum gradient and result in peak performance.  However, such a procedure would reduce to a minimum the volume of filter material between the stream and the underground intake and thereby minimize filtration benefits.  It would also limit benefits gained in modulating the temperature of inflow water as it moves through the earth.  Although the filtration and temperature benefits increase as the underground intake is located at progressively greater distances from the stream, the increased distance reduces the gradient from the stream to the intake and thereby reduces the hydraulic performance of the intake structure."

"Many of the larger groundwater installations developed in this country are of the induced infiltration type.  It has been indicated by Jeffords [1945, Recharge to water-bearing formations along the Ohio Valley: Am. Water Works Assoc. Jour., vol. 37, pp. 144-145], by Kazmann [1948, River infiltration as a source of groundwater supply: Am. Soc. Civil Eng. Trans., vol. 113, pp. 404-420], and by Rorabaugh [1951, ‘Stream-bed percolation in development of water supplies,’ Union Geol. Geophys. Internat., Assoc. Internat. d’Hydrologie Scientifique, Assemblee Gen. de Bruxelles 1951, Extract du Tome II, pp. 165-174] from observation of such systems, that essentially complete filtration can be effected and many other advantages gained.  Among these advantages are the elimination of short-term fluctuations in turbidity, temperature and other phases of water quality that are so characteristic of many surface streams.  Inasmuch as groundwater movement from the stream bed to the intake structure would be generally of the order of a few feet to perhaps tens of feet per day, the time of travel from the stream to the groundwater intake may be on the order of several days to as much as several months for intakes within a few hundred feet of the stream.  As a consequence of these long times of contact with the reservoir media, the infiltrated stream water absorbs heat from or delivers heat to the reservoir particles.  Thus, in effect, the groundwater reservoir serves not only as a filtering medium, but also as a heat exchanger."

"In periods of flood when the stream is very turbid, the method of direct-surface intake requires continual vigilance on the part of the filter plant operator, but where a subsurface intake is used as in the induced infiltration method, much of this burden is eliminated.  Where chlorination or water softening is necessary, operation schedules can be planned from weeks to months in advance.  Under continued operation and with a moderately- to highly-permeable connection to the stream, the average quality of water produced by an induced infiltration system approximates the average quality of the surface source, except that its variations in quality and temperature are greatly reduced as a consequence of the lengthy times in transit through the aquifer."

Natural reservoirs of groundwater have great value, but they can be overdeveloped if natural and artificial discharge exceed natural recharge.  To overcome this difficulty, some areas have attempted to increase the natural recharge by artificially replenishing the groundwater reservoir.  Two general methods have been developed: recharge by surface application andrecharge through wells.  Both of these techniques require a thorough knowledge of both the geology and hydrology of the area.

Artificial recharge basics

Artificial recharge may be defined as the practice of artificially increasing the amount of water that enters a groundwater reservoir.  Artificial recharge has application in waste disposal, secondary oil recovery, and land subsidence problems, as well as water supply problems.

Specific purposes for which artificial recharge is practiced are, as listed in W.C. Walton’s book, Groundwater Resource Evaluation (1970, McGraw-Hill), to:

  1. Conserve and dispose of runoff and flood waters
  2. Supplement the quantity of groundwater available
  3. Reduce or eliminate decline in the water level of groundwater reservoirs
  4. Reduce or balance salt water intrusion
  5. Store water to reduce costs of pumping and piping
  6. Store water in off-seasons for use during the growing seasons
  7. Conserve energy in geothermal applications
  8. Remove suspended solids by filtration through the ground.

An artificial recharge installation may serve more than one purpose.  In certain areas, for example, artificial recharge not only adds water to the available groundwater supply, but also is a means of disposing of storm water runoff.  In another instance, artificial recharge to control salt water intrusion is also increasing the available supply of fresh water and alleviating a ground subsidence condition that has been in progress for years.

The advantages of groundwater storage compared to surface storage are no losses by evaporation, reduced construction cost in preparing the surface reservoir, and seasonal availability of water, e.g., increasing water in a depleted aquifer, usually accomplished during the off-season.

The largest potential reservoir for the storage of potable water is in the unsaturated zone.  Use of this space for the storage and retrieval of potable water is a multifaceted problem that requires application of the best talent from the scientific community.

Although artificial recharge is a potential means of solving some water supply problems, each application must be evaluated to determine if it is physically and economically feasible.  Geologic and hydrologic conditions that may affect the recharge must be evaluated in each location where artificial recharge is to be used.  The recharge water must also be analyzed to determine if it is chemically compatible with the groundwater and whether it requires pretreatment to avoid clogging the aquifer.  Finally, the most suitable method of recharge for the application must be selected and its cost determined.

R.G. Kazmann (1965, Modern Hydrology, Harper and Row) makes some important points when he writes, "Let us recognize that progress in the utilization of aquifers as storage reservoirs is slow because of certain disadvantages … both physical and sociological.  The physical difficulties include the need for the cost of providing clean, nonturbid water for recharge purposes and the necessity and cost of pumping the stored water out of the aquifer.  The sociological difficulties include the problem of obtaining title to a perennial water supply to be used for recharge purposes and the associated problem of establishing property rights to the stored water once it has been placed in an aquifer."

Artificial recharge has many similarities to liquid-waste disposal through deep wells.  In both, the problem is to place liquid in a permeable lithologic unit at an economic rate and to predict movement, chemical reactions, and physical changes that take place while the liquid is in the reservoir.  Differences between the two operations are principally in the type of fluid injected and the ultimate objective.  In artificial recharge, the objective is to store and retrieve water of good quality; in waste disposal, the objective is to permanently store water of objectionable quality.  In both artificial recharge and liquid-waste storage, the nature of the storage must be known, particularly that of the unsaturated zone.  The techniques of investigation for recharge and waste disposal are generally the same.

Water commonly is recharged by surface spreading through basins or by induced recharge from adjacent streams and lakes or through injection wells.  Research in recharge through basins has been dominated by mathematical models based on idealized conditions and empirical relations, derived by experimental sequencing of recharge operations, and operational controls in the pretreatment of recharge water.  Recharge by injection wells has been undertaken in a variety of hydrologic environments.  In Israel, efforts have been directed toward the analyses of diffusion and dispersion of the injected water.  Much research in the United States has been directed toward the movement of bacteria and organic matter through an aquifer and toward the chemical modeling of changes in recharged water as it moves.

Much more research is needed on the basic properties of aquifers, particularly in the unsaturated zone, and on all aspects of recharge water quality.  Research and the use of data produced are increasingly the responsibility of interdisciplinary teams that consider the geologic, hydraulic, and economic aspects of the system.

avra_valley_projectThe Avra Valley Recharge Project, pictured at right, is designed to store Colorado River water underground for future use.  Photo courtesy Central Arizona Project.

Water may be recharged by releasing it into basins formed by excavation or by the construction of containment dikes or small dams.  Horizontal dimensions of such basins vary from a few meters to several hundred meters.  The most common system consists of individual basins fed by pumped water from nearby surface water sources.  Silt-free water avoids the problem of sealing basins during flooding.  Even so, most basins require periodic scraping of the bottom surface when dry to preserve a percolation surface.

In California, basins have been successfully built and operated in abandoned stream channels.  In alluvial plains, basins may parallel existing channels with water being led into the upper basin by canal.  As the first basin fills, it spills into the second. This is repeated through the entire chain of basins. From the lowest basin, excess water is returned to the main channel.  By this method, spreading is accomplished on what otherwise might be considered waste land and permits water contact over 75 to 80 percent of the gross area.

Basins, because of their general feasibility and ease of maintenance, are the most favored method of artificial recharge from the surface.

Ditches or furrows, which are shallow, flat-bottomed, and closely spaced to obtain maximum water contact area, are another alternative. Gradients of major feeder ditches should be sufficient to carry suspended material through the system since deposition of fine-grained material clogs soil surface openings.  Water spreading in a natural stream channel may use any of the methods described.  Whatever method of surface application is adopted, the primary purpose is to extend the time and the area over which water is recharged.

In irrigated areas, water is sometimes spread by irrigating with excess water during nonirrigating seasons.  The method requires no additional cost for land preparation as the distribution system is already installed.  Even keeping irrigation canals full will contribute to recharge by seepage from the canals.  Where a large portion of the water supply is pumped, the method has the advantage of raising the water table and consequently reducing power costs.  Disadvantages include additional energy cost, evaporation losses, and leaching of soil nutrients.

Economical surface recharge depends upon maintenance of a high infiltration rate.  Typical infiltration rate curves, however, show a pronounced decrease with time.  The initial decrease is attributed to dispersion and swelling of soil particles after wetting.  The subsequent increase occurs as entrained soil air is eliminated by solution in passing water, while the final gradual decrease results from microbial growths clogging the soil pores.

Generally, recharge rates decrease as the mean particle size of soil on a spreading area decreases.  Efforts to maintain soil infiltration rates include additions of organic matter and chemicals to the soil and vegetation cover.

Alternating wet and dry periods on a basin generally will furnish a greater total recharge than does continuous spreading, in spite of the fact that water is in contact with the soil for as little as one-half the total time.  This occurs because soil porosity increases during vegetation growth periods.  Soils must be aerated to allow vegetation growth.

Studies of small ponds have confirmed that infiltration rates are directly proportional to the hydraulic head and to the permeability of material surrounding the ponded water.

Clogging due to artificial recharge in laboratory-simulated, unconsolidated aquifers displays two patterns.  The first, resulting from recharge with turbid water containing an effective microbial inhibitor, shows clogging throughout the aquifers ranging in depth from 48 cm to 123 cm.  The rate of clogging at different depths was dependent on the size distribution of the porous media.  The second pattern, resulting from recharge with nonturbid water and no effective microbial inhibitor, shows clogging only in the top few centimeters.

Recharge through pits and tubular wells

Water spreading cannot be effective in areas where subsurface strata restrict the downward passage of water.  In areas where the impervious layer is close to the ground surface, recharging can be conducted by specially designed wells.  If these penetrate to more permeable substrata, water can percolate directly into an aquifer.  Generally, recharge wells cost more to construct and may recharge smaller volumes of water than do spreading areas, but may be the only practical recharge method in these circumstances.  This method should be applied only when good quality water is used for recharge.

A recharge well (injection well, inverted well, diffusion well, or disposal well) may be defined as a well that admits water from the surface to underground formations.  Its flow is the reverse of a pumping well, but its construction should be the same.  Well recharging is practical in many geologic environments where aquifers must be recharged, and where economy of surface space, such as in urban areas, is an important consideration.

If water is admitted into a well, a cone of recharge will be formed which is similar in shape, but is the reverse of a cone of depression surrounding a pumping well.

By comparing the discharge equations for pumping and recharge wells, it might be anticipated that the recharge capacity would equal the pumping capacity of a well if the recharge cone has dimensions equivalent to the cone of depression.  Field measurements, however, rarely support this reasoning.  Recharge rates seldom equal pumping rates.  The discrepancy lies in the fact that pumping and recharge differ by more than a simple change of flow direction.  A properly designed recharge well will recharge as much as the pumping capacity.  The problem lies in recharge water quality and turbidity.

Any silt carried by water into a recharge well is filtered out and tends to clog the aquifer surrounding the well.  Similarly, recharge water may also contain bacteria that can form growths on the well screen and the surrounding formation, thereby reducing the effective flow area.  Chemical constituents of the recharge water may differ sufficiently from the normal groundwater to cause undesirable chemical reactions, i.e., ion exchange in aquifers containing sizable fractions of silt and clay.  These factors all act to reduce recharge rates and, as a result, well recharging has been limited to areas where local conditions and experience have shown the practicality of the method.

In order for a recharge project to be successful, field conditions must provide for appropriate storage, movement, and use of recharge water.  The California Department of Water Resources has listed the following physical requirements for recharging.

  1. Geology. The basin must be suitable from the standpoint of storage capacity and transmissibility of aquifers.
  2. Water. Adequate recharge water must be available.
  3. Infiltration. Recharge rates must be maintained at adequate levels.
  4. Drainage. Where a water table is near ground surface, adequate storage capacity in the basin for recharging must be provided.
  5. Water quality. Recharge water must be chemically compatible with existing groundwater and have a suitable temperature.
  6. Recovery efficiency. Pumping lifts must not be excessive, installed pumping capacity must be efficiently used, and quality of water recovered must be satisfactory.

"Recently attention has been directed toward plans for recharging through wells in industrial areas in which the groundwater supplies have been depleted by heavy pumping, largely for cooling purposes, and in which the public supplies are obtained from surface sources.  The plan involves recharging with water from the public supply in winter when the surface water is cold, and thus increasing the supply of cool groundwater in the summer when the surface water is too warm to be satisfactory for cooling purposes.  This plan is promising for many cities in which winter and early spring are the seasons in which there is excess surface water and low water consumption, leaving unused facilities available to furnish clear, sterile water for recharging at low cost.  To the extent that cold water from the public supply is used in winter instead of groundwater, the production wells can be used for recharging.  This plan was put into practice in Louisville during the World War II emergency on a small scale but with definitely beneficial results (Guyton, WE, Stuart, WT., and Maxey, G.B., 1944, Progress report on groundwater resources of the Louisville area, Ky. Mimeographed report issued by the U.S. Geol. Survey in cooperation with the Kentucky Geol. Division and city of Louisville)."

The above information is excerpted in large part from Chapter 17 of the 1999 NGWA Press publication, Ground Water Hydrology for Water Well Contractors.