1985

R E V I E W : H Y D R O L O G Y

Flow and Storage in Groundwater SystemsWilliam M. Alley,1* Richard W. Healy,2 James W. LaBaugh,1 Thomas E. Reilly1

The dynamic nature of groundwater is not readily apparent, except where dischargeis focused at springs or where recharge enters sinkholes. Yet groundwater flow andstorage are continually changing in response to human and climatic stresses. Wisedevelopment of groundwater resources requires a more complete understanding ofthese changes in flow and storage and of their effects on the terrestrial environmentand on numerous surface-water features and their biota.

Groundwater is a crucial source of freshwater throughout the world. Morethan 1.5 billion people worldwide (1)

and more than 50% of the population of theUnited States (2) rely on groundwater fortheir primary source of drinking water.Groundwater is an essential part of the hy-drologic cycle (Fig. 1) and is important insustaining streams, lakes, wetlands, andaquatic communities.

During the past 50 years, groundwaterdepletion has spread from isolated pockets tolarge areas in many countries throughout theworld. Prominent examples include the HighPlains of the central United States, wheremore than half the groundwater in storage hasbeen depleted in some areas, and the NorthChina Plain, where depletion of shallow aqui-fers is forcing development of deep, slowlyreplenished aquifers with wells now reachingmore than 1000 m (3). Groundwater deple-tion may be the single largest threat to irri-gated agriculture, exceeding even the buildupof salts in soil (3). In arid regions, much ofthe groundwater removed from storage todaywas recharged during wetter conditions in thelast ice age, causing further concerns aboutpresent withdrawal rates. Global groundwaterdepletion has been appreciable enough tocontribute to sea-level rise during the pastcentury as a result of water pumped fromwells that returns to the sea either by runoffor by evapotranspiration followed by precip-itation (4).

Many unfamiliar with its dynamic natureview groundwater as a static reservoir. Evenspecialists may overlook its linkages acrossthe biosphere and consider it an isolated partof the environment (5). Yet, as discussedbelow in general terms and through exam-ples, the dynamic aspects of groundwaterflow systems, their recharge, and interactions

with surface water and the land surface arenumerous and extend over many differenttime scales.

Dynamics of Groundwater FlowSystemsA groundwater system comprises the subsurfacewater, the geologic media containing the water,flow boundaries, and sources (such as recharge)and sinks (such as springs, interaquifer flow, orwells). Water flows through and is stored withinthe system. Under natural conditions, the traveltime of water from areas of recharge to areas ofdischarge can range from less than a day to more

than a million years (6). Water stored within thesystem can range in age (7) from recent precip-itation to water trapped in the sediments as theywere deposited in geologic time.

The variability of aquifer response timesis illustrated by the time required for the

hydraulic head (water levels) in a groundwa-ter system to approach equilibrium after somehydraulic perturbation, such as well pumpingor a change in recharge rate. This can beestimated for confined groundwater systems(8) as

T* 5 SsLc2/K (1)

where T* is the hydraulic response time (T)for the basin, Ss is specific storage (L21), Lc issome characteristic length (L) of the basin,and K is hydraulic conductivity (L/T). Thehydraulic conductivity, a measure of perme-ability, can range over 12 orders of magni-tude (8), and the distance between boundariesof groundwater systems can range frommeters to hundreds of kilometers. Using Eq.1, hydraulic response times calculated fortwo idealized systems (9) are 0.1 day (144min) for horizontal flow in a confined stream-aquifer system and 4.0 3 107 days (110,000years) for vertical flow in a thick regionallow-permeability unit.

The time of travel through the systemdepends on the spatial and temporal gradi-ents of hydraulic head, hydraulic conduc-tivity, and porosity of the system. The timeof travel through a system is different fromthe hydraulic response time to approach

1U.S. Geological Survey, 411 National Center, Reston,VA 20192, USA. 2U.S. Geological Survey, Post OfficeBox 25046, Mail Stop 413, Denver Federal Center,Lakewood, CO 80225, USA.

*To whom correspondence should be addressed. E-mail: [email protected]

Fig. 1. Global pools and fluxes of water on Earth, showing the magnitude of groundwater storagerelative to other major water storages and fluxes. [Reproduced from (82) with permission from thepublisher, Elsevier Science (USA)]

S C I E N C E ’ S C O M P A S S ● R E V I E W

www.sciencemag.org SCIENCE VOL 296 14 JUNE 2002 1985

equilibrium. For example, it was calculatedabove that the hydraulic head in the confinedstream-aquifer system responded to a perturba-tion in less than a day; however, the time re-quired for water to move through the entirewidth of the system is on the order of 30,000days (82 years) under natural conditions (10).Fractured-rock systems in bedrock usually havesmaller effective porosities than unconsolidated

porous media systems such as sands and grav-els, and flow velocities through fractured-rocksystems can be relatively fast (11). For exam-ple, travel times of water over distances ofseveral kilometers have been estimated at lessthan a year for municipal wells completed infractured dolomite in Wisconsin (12). Seasonalvariations in recharge and pumping affect thevariability in travel times in such cases. In

more sluggish groundwater systems,such as the Bangkok Basin in Thailand(13), long-term climate and geologicchange need to be considered in under-standing the movement of groundwaterover tens of thousands of years. The long-term movement of groundwater also influ-ences virtually all geologic processes (14,15), including diagenesis, ore mineraliza-tion, and petroleum accumulation.

The time of travel of water is impor-tant in determining the movement ofcontaminants within a groundwater sys-tem. The large extent of groundwatercontamination worldwide from surfacesources reflects the fact that shallowgroundwater ages are typically a few dec-ades or less. Hydraulic gradients causedby large-capacity wells can further re-duce the travel times of contaminants towells (16).

Water withdrawn from a groundwa-ter system initially comes from storage.Over time, the effects of the withdraw-al are propagated through the system asheads decrease at greater distancesfrom the point of withdrawal. Ulti-mately, the effect of the withdrawalreaches a boundary (such as a stream)where either increased recharge to thegroundwater system or decreased dis-charge from the system occurs. The com-mon assumption that the rate of groundwa-ter withdrawal is “safe” or “sustainable” ifit does not exceed the natural rate ofrecharge is not correct, because it ignoresthese changes in discharge from and re-charge to the groundwater system (17, 18).The sources of water supplying pumpagefrom 10 major regional aquifer systems inthe United States are shown in Fig. 2.These illustrate the variability of aquiferresponse to long-term pumping and theextent to which changes in recharge anddischarge can exceed changes in storage.

Computer models of flow and solutetransport have been integral tools forevaluation of groundwater resourcesfor many years; they have been appliedto a wide range of problems, from localcontamination to the origin of largemineral bodies from continental-scalefluid migrations (19). The predictivecapability of models permits hypothe-sis testing, which enhances our under-standing of current conditions, as well

as forecasting of aquifer response to futureclimatic or anthropogenic stresses. Recentlinkages of groundwater flow models withland surface–atmosphere models (20) andof transport models with geochemical reac-tion models (21) have extended the types ofproblems that can be addressed. Automaticcalibration schemes and uncertainty analy-sis (22) have enhanced model application,

Fig. 2. Sources of water that supply withdrawals from major aquifer systems in the United States arehighly variable, as shown by these results from model simulations for various periods (83). The Floridanand Edwards-Trinity aquifer systems, which equilibrate rapidly after pumping, were simulated assteady-state with no long-term change in storage. In contrast, the Southern High Plains (with mostnatural discharge occurring far from pumping wells) and the deeply buried Great Plains aquifer systemhave had substantial changes in groundwater storage. The distinction between changes in recharge andchanges in discharge is a function of how the system was defined (i.e., a gain to one system may resultin a loss from an adjoining system). For example, groundwater withdrawals from confined aquifers(Northern Atlantic Coastal Plain, Gulf Coastal Plain) can cause flow to be diverted (recharged) into thedeeper regional flow regime that would otherwise discharge to streams in the outcrop areas or causevertical leakage across confining units. Groundwater recharge in a region can be increased as a result ofhuman modifications, such as return flow of excess irrigation water (California Central Valley). Note thatthe areal extent of the Southeastern Coastal Plain aquifer system overlaps the areal extents of theFloridan and Gulf Coastal Plain aquifer systems.

S C I E N C E ’ S C O M P A S S

14 JUNE 2002 VOL 296 SCIENCE www.sciencemag.org1986

and new computer visualization tools haveadvanced our understanding of the effectsof variability in aquifer properties ongroundwater flow patterns.

Accuracy of model predictions is con-strained by the correctness of the model (i.e.,proper representation of relevant processes)and uncertainty in model parameters. Thelatter uncertainty is due to the limited accu-racy with which parameter values can bemeasured and, more important, to the sub-stantial heterogeneity inherent in aquifercharacteristics. The inability to describe andrepresent this heterogeneity adequately is afundamental problem ingroundwater hydrology andwill continue, even with im-proved models, to place lim-its on the reliability of modelpredictions. The links be-tween spatial heterogeneityand model uncertainty alsodepend on the type of ques-tions being asked. For exam-ple, reasonable estimation ofhead distributions in an aqui-fer may require only limitedunderstanding of spatial het-erogeneity. On the otherhand, confidence in predic-tions of chemical concentra-tions at a specific locationcan be very sensitive to minoruncertainty in the spatial dis-tribution of hydraulic proper-ties, even for relatively ho-mogeneous porous media.

Tracer techniques havebeen widely applied for es-timating the residence timeof subsurface waters, aswell as the amounts and tim-ing of recharge and dis-charge (23). Most tracertechniques require knowl-edge (or assumption) of thetime history of tracer appli-cation at the land surface orthe water table (Fig. 3). Thistemporal pattern is then cor-related to a concentration-depth pattern in the subsurface at a point intime. Other approaches [e.g., the 3H/3Hetechnique (24 )] use information on decayproducts to determine age. Tracers can benaturally occurring (the stable isotopes 2Hand 18O, Cl, heat), can occur in the atmo-sphere as a result of anthropogenic activi-ties [tritium, 36Cl, chlorofluorocarbons(CFCs)], or can be applied intentionally onthe land surface (N and P fertilizers, organ-ic pesticides). Isotopes of elements dis-solved from host rocks (222Rn, 87Sr/86Sr)can also be used to estimate residence timesand interactions with surface water.

Over the past decade, advances in agedating and tracking young groundwater(,50 years old) using multiple tracers havebeen major breakthroughs in understandingthe dynamics of groundwater systems. Forexample, multiple tracers have been used todefine the amount and locations of riverwater recharging the Upper Floridan aqui-fer and causing deterioration of well-waterquality near Valdosta, Georgia (25), to con-strain groundwater flow models in the At-lantic Coastal Plain (26 ), and to help ex-plain unusual nutrient regimes fromgroundwater inputs to Florida Bay (27 ).

Determining the time that water has beenflowing within the groundwater system isparticularly useful in understanding the op-eration of highly heterogeneous aquifersystems. However, special care should betaken in interpretations of tracer concentra-tions in these settings, because the concen-trations may be affected greatly by hydro-dynamic dispersion and diffusion into therock matrix (28).

RechargeRecharge is an important factor in evaluatinggroundwater resources but is difficult to

quantify. The present discussion is limited torecharge to the water table (as opposed tointeraquifer recharge). Recharge can occur inresponse to individual precipitation events inregions having shallow water tables. In con-trast, unsaturated zone water in some desertregions is estimated to have infiltrated thesoil surface as long as 120,000 years ago(29). Perhaps nowhere is the importance anddifficulty of estimating recharge more appar-ent than in the assessment of the suitability ofYucca Mountain, Nevada, as a repository forhigh-level radioactive waste. More than 15years and tens of millions of dollars have

been spent to estimate re-charge rates and locationsthrough the thick, fracturedvolcanic tuffs at this site un-der past, current, and futureclimates (30).

Recharge can be diffuseor localized. Diffuse rechargerefers to the widespreadmovement of water from landsurface to the water table as aresult of precipitation overlarge areas infiltrating andpercolating through theunsaturated zone. Localizedrecharge refers to the move-ment of water from surface-water bodies to the ground-water system and is lessuniform in space than diffuserecharge. Most groundwatersystems receive both diffuseand localized recharge. Ingeneral, the importance ofdiffuse recharge decreases asthe aridity of a region in-creases (31). For example, insemiarid parts of Niger,localized recharge from natu-rally occurring runoff-collec-tion ponds accounts for virtu-ally all recharge (32).

Typically, most waterfrom precipitation that infil-trates does not become re-charge. Instead, it is stored inthe soil zone and is eventual-

ly returned to the atmosphere by evaporationand plant transpiration. The percentage ofprecipitation that becomes diffuse recharge ishighly variable, being influenced by factorssuch as weather patterns, properties of sur-face soils, vegetation, local topography,depth to the water table, and the time andspace scales over which calculations aremade. For example, over a 6-year period,recharge in the Great Bend area of centralKansas was estimated to be 10% of the an-nual precipitation of 585 mm; however, insome years, no recharge occurred (33).

Magnitudes of recharge fluxes are gen-

Fig. 3. Annually averaged atmospheric concentrations during the past 60 years ofsome environmental tracers used to determine groundwater ages. Environmentaltracers due to industrial production and release are CFCs (CFC-11, CFC-12, andCFC-113), SF6, and 85Kr. Environmental tracers produced by nuclear tests inaddition to natural production are 3H, 36Cl, and 14C. Units: TU, tritium units inprecipitation at Washington, DC; mBq m23, millibecquerels per cubic meter; pptv,parts per trillion by volume; pmc, percent modern carbon [(23); CFC-11 andCFC-113 data from L. N. Plummer and E. Busenberg, U.S. Geological Survey].



erally quite low and are difficult to measuredirectly. Measurement of fluxes can becomplicated by preferential flow (i.e.,macropore or unstable flow) in the unsat-urated zone, although preferential flowpaths are of greatest concern as potentialconduits for rapid contamination of aqui-fers. The above factors, in addition to tem-poral and spatial variability, greatly com-plicate estimation of basin-wide rechargerates. Estimation methods include use ofwater budgets, tracers, geophysics, andsimulation models (34 ). Recent develop-ments include improved age-dating tech-niques (23); geophysical monitoring, suchas time-domain reflectometry and ground-penetrating radar (35); land- and possiblysatellite-based gravity measurements to es-timate changes in subsurface water massover length scales of tens to thousands ofkm (36, 37 ); the linking of watershed andgroundwater flow models (38); and use ofpiezometers completed in thick clay layersto measure changing geostatic loadsassociated with mass changes in subsurfacewater (39). Because of inherent uncertain-ties in any method, it is recommended thatmultiple techniques be applied for anystudy.

Identifying human practices that influ-ence recharge is straightforward; quantify-ing effects of these practices is more diffi-cult. Clearing of native vegetation has led

to an order of magnitude increase in re-charge rates in areas such as the NigerBasin in Africa (32). Irrigation has resultedin increased recharge rates (Fig. 2) as wellas salinization of soils and aquifers, such asin the Nile River delta (40). Urbanizationhas modified natural recharge processes ap-preciably but does not usually lead to de-creased recharge (41), as is often assumed.Enhanced runoff from built-up and pavedareas may be channeled to a retention basinor infiltration gallery, resulting in reloca-tion of recharge areas and the transitionfrom slow, diffuse recharge to rapid, local-ized recharge. Canals, leaky water mains,and sewers are other artifacts of develop-ment that influence recharge processes. Forexample, it is estimated that 26% of thewater transmitted through water mains inGoteborg, Sweden, is lost to leakage (42).

The effects of climate change on re-charge also are difficult to assess. Areas ofhigh or low recharge in past climates canperhaps be mapped (43), and groundwatersthat infiltrated many thousands of years agohave been identified, but quantification ofrecharge rates during past climatic periodshas been attempted only in isolated aridregions such as the western United States,where, as estimated from tracer data fromthe unsaturated zone, recharge rates 15,000years ago were about 20 times the currentrates (44 ).

Interactions with Surface WaterThe interactions of surface-water bod-ies with groundwater are governed bythe positions of the water bodies rela-tive to the groundwater flow system,the characteristics of their beds andunderlying materials, and their climaticsetting (45). Whereas the geologicframework affects the flow pathsthrough which groundwater flows, thetype of sediments at the interface be-tween groundwater and surface watercan dictate the spatial variability of dis-charge to surface water and, in turn,affects the distribution of biota at theinterface. For example, silty streambeds with minimal groundwater ex-change may support a less diverse suiteof biota than do sandy or gravellystream beds with large groundwater ex-change (46). Discharge from springscan provide habitat for unique speciesthat are dependent on adequate ground-water flow (47). In some cases, biotarelated to groundwater discharge havebeen used to identify locations wherefocused discharge occurs into surfacewaters (48).

Exchange of water across the inter-face between surface water andgroundwater can result from down-stream movement of water in and out

of stream beds and banks (Fig. 4), tides, waveaction, filling or draining of reservoirs, ortranspiration of water by vegetation at theedges of wetlands and other surface waters(45). Water exchange across the surface wa-ter–groundwater interface has been exploredin some detail in the past decade, with moststudies focused on streams (49), and is in-creasingly studied with respect to effects onthe chemical composition of surface and sub-surface water and the distribution of biota(46, 50). Once thought to be of little conse-quence and thus ignored, the interactions ofgroundwater with lakes, wetlands, estuaries,and oceans now are recognized as importantprocesses. For example, discharge of salinesprings contributes to the salinity of LakeKinneret, Israel (51); peat wetlands can alter-nate between recharge and discharge statusbecause of flow reversals (52); coastalgroundwater discharge is equivalent to asmuch as 40% of riverine input in summeralong the coast of South Carolina, USA (53);and groundwater input from the Ganges-Brahmaputra delta is an important factor af-fecting the marine strontium isotope record(54).

Thermal effects also play a role in thedistribution of biota and biogeochemical pro-cesses. For example, thermal effects ofgroundwater discharge in inland waters havebeen directly related to fish habitat, both interms of spawning areas and refuge for adults

Fig. 4. Local geomorphic features such as stream bed topography, stream bed roughness, meandering, andheterogeneities in sediment hydraulic conductivities can give rise to localized flow systems within streambeds and banks. The near-stream subsurface environment with active exchange between surface waterand groundwater commonly is referred to as the hyporheic zone, although the transition betweengroundwater and surface water represents a hydrologic continuum, preventing a precise separation.



when ice forms in colder environments (55).Thermal differences between groundwaterand surface water also are used to provideinformation on location and amount of re-charge (56) and discharge (57), and thesedata enable indirect determination of geother-mal properties of groundwater flow systems,particularly from data gathered at springs(58).

When salt water and fresh water arepresent, a dynamic interface is present both inthe ground and at the discharge boundary offresh groundwater into salty surface water. Inrelatively homogeneous porous media, thedenser salt water tends to remain separatedfrom the overlying fresh water by a transitionzone, known as the zone of diffusion ordispersion. In coastal plain areas where theporous media is heterogeneous in nature, asystem of layered mixing zones can form.Advances in geophysical techniques, such asdirect-current resistivity and transient electro-magnetic induction (59), have enabled betterdefinition of the three-dimensional distribu-tion of salty water in the subsurface.

Large groundwater withdrawals can causesalt water to move into areas of use in coastal(60) and some inland (61) areas and decreasethe volume of fresh wateravailable. The importantrole of fresh groundwaterdischarge to coastal eco-systems is also increasing-ly being recognized (62).The time required for thesalt water–fresh water in-terface and freshwater dis-charge to respond to hu-man and natural changescan range from almost in-stantaneously to thou-sands of years. In somecoastal areas, such as NewJersey, USA, and Suri-name, South America(63), relatively freshgroundwaters located faroff the coast are hypothe-sized to be remainingfrom the last ice age,when sea levels weremuch lower.

Interactions with the Land SurfaceWhen groundwater is removed from storagein groundwater systems, hydraulic heads arelowered, and a portion of the mechanicalsupport for the overlying sediments and sub-surface water previously provided by pore-fluid pressure is transferred to the granularskeleton of the aquifer system. If enoughwater is withdrawn, the pore-fluid pressurecan be reduced enough so that the granularskeleton of the aquifer is irreversibly com-pressed, causing permanent compaction of

the more compressible fine-grained silt andclay layers (aquitards) interbedded within oradjacent to the aquifers. The resulting subsi-dence can severely damage structures andcreates problems in design and operation offacilities for drainage, flood protection, andwater conveyance. Examples of areas withlarge subsidence include the California Cen-tral Valley, Houston, and Mexico City (64).

The low permeability of thick aquitards(65) can cause vertical drainage to adjacentpumped aquifers to proceed slowly and to lagfar behind changing water levels in theseaquifers (Fig. 5). The drainage and compac-tion in response to a given stress in thickaquitards may require decades or centuries toapproach completion. Numerical modelinghas successfully simulated complex transienthistories of compaction observed in responseto measured water-level fluctuations at thesite scale (66), but considerable challengesremain at the regional scale to simulate com-paction histories of groundwater systemswith thick aquitards.

Technologies to measure the sometimessubtle and slow changes in land-surface ele-vations caused by groundwater withdrawalshave evolved considerably from borehole ex-

tensometry and terrestrial geodetic [spirit lev-eling and Global Positioning System (GPS)]surveys to remote-sensing using space-basedradar imaging (64). Interferometric syntheticaperture radar (InSAR) uses repeat radar sig-nals from satellites to measure deformation ofEarth’s crust at an unprecedented level ofspatial detail (changes in elevation on theorder of 10 mm or less) and a high degree ofmeasurement resolution (tens of meters). In-SAR results have provided detailed regionalmaps of land subsidence at seasonal andlonger time scales (67); have revealed the

timing, magnitudes, and patterns of seasonaldeformations that may occur in response toseasonal changes in pumping and large-scaleartificial recharge (68); have been used toestimate the elastic storage coefficient of anaquifer system at locations where contempo-raneous groundwater level observations wereavailable (69); have been used as an obser-vational constraint for inverse modeling ofregional groundwater flow and aquifer-system compaction (70); and have providednew insights into how subsidence is con-trolled by geologic structures and sedimentcomposition (71).

Future ChallengesFuture success in understanding the dynamicnature of groundwater systems will rely oncontinued and expanded data collection atvarious scales, improved methods for quanti-fying heterogeneity in subsurface hydraulicproperties, enhanced modeling tools and un-derstanding of model uncertainty, and greaterunderstanding of the role of climate and in-teractions with surface water.

Water-level measurements from wells re-main the principal source of information onthe effects of hydrologic stresses on ground-

water systems. Advancesin instrumentation nowenable the collection ofreal-time water-level data,allowing us to observe di-urnal and seasonal trendsfrom well networks acrosslarge areas. To understandthe true nature of changein a groundwater systemand to differentiate be-tween natural and human-induced changes, we re-quire records of water-level measurements oversubstantial periods (72).Despite their importance,groundwater-level datahave received little atten-tion in concerns expressedabout the continuity ofglobal water data, primar-ily because such concernshave focused on more vis-

ible surface-water monitoring networks (73).Because aquifers smooth out short-term

fluctuations of climate signals, analyses ofgroundwater systems typically have under-played the role of climate. Effects of decadal-scale fluctuations in wet and dry cycles, suchas those hypothesized from the Pacific Dec-adal Oscillation (74), may have large effectson groundwater systems, but these are rela-tively unexplored, as are the effects of possi-ble future climate change on the shallowaquifers that supply much of the water instreams, lakes, and wetlands. A greater un-

135

130

145

140

155

150DE

PTH

TO

WAT

ER

, IN

FE

ET

BE

LOW

LA

ND

SU

RFA

CE

Compaction progressessteadily despite

water-level recoveries

Further water-level declinesdo not cause

accelerated compaction

CO

MPA

CT

ION

, IN

FE

ET

19911990 1992 1993

Year

1994 1995 1996 1997

Seasonal high water levels

Aquifer-system compaction

Seasonal low water levels

0

0.20

0.10

0.40

0.30

Fig. 5. The hydrodynamic lag of fluid-pressure changes between aquifers and thick,slowly draining aquitards in Antelope Valley, California, causes continued compactionthat is relatively independent of the effects from seasonal and annual fluctuations inwater levels (66).



derstanding of feedbacks between water lev-els and atmospheric forcing at seasonal andinterannual scales is also needed (20). Be-cause conditions at the time of recharge in-fluence the geochemical composition of wa-ter percolating into the subsurface, aquifersmay prove to be invaluable archives of pastclimate and environmental change (75).

Surface-water depletion is viewed in-creasingly as the limiting factor to the long-term use of groundwater resources, yet thedistinctly different temporal and spatialscales at which groundwater and surface-water systems operate present major chal-lenges to their integrated analysis. The loca-tions, quantity, and timing of reductions insurface-water flow resulting from groundwa-ter development are fundamental questions atscales of years to decades, whereas ecologicalissues require attention to seasonal and evendiurnal changes in groundwater recharge anddischarge and more attention to fluvial plainand channel-scale flow processes (76).

Groundwater systems have value not onlyas perennial sources of water supply, but alsoas reservoirs for cyclical injection and with-drawal to modulate the variability inherent insurface-water supplies. Management approach-es increasingly involve the use of artificial re-charge of excess surface water or recycled wa-ter by direct well injection, surface spreading,or induced recharge from streams. As predic-tive links between hydrology and climate im-prove (e.g., prediction of El Nino conditions),opportunities exist to make better use of thestorage capacity of groundwater systems. Manyscientific challenges remain to understand morefully the long-term hydraulic response of aqui-fer systems, subsurface chemical and biologicalchanges of the injected water, and geochemicaleffects of mixing waters of different chemis-tries (77). With time and extensive use, muchof the local groundwater may be derived fromartificial recharge (78)—a further indicator ofthe dynamic nature of groundwater systems.

References and Notes1. R. Clarke, A. Lawrence, S. Foster, Groundwater: A

Threatened Resource (United Nations EnvironmentProgramme Environment Library No. 15, Nairobi,Kenya, 1996).

2. W. B. Solley, R. R. Pierce, H. A. Perlman, U.S. Geol.Surv. Circ. 1200 (1998).

3. S. Postel, Pillar of Sand: Can the Irrigation MiracleLast? (Norton, New York, 1999).

4. D. L. Sahagian, F. W. Schwartz, D. K. Jacobs, Nature367, 54 (1994).

5. Schwartz and Ibaraki (79) provide evidence of thematuring nature of groundwater hydrology as a dis-tinct field. This implies that the greatest future ad-vances are likely to be in integration with other fields.

6. H. W. Bentley et al., Water Resour. Res. 22, 1991(1986).

7. The “age” of water usually refers to the amount oftime since the water entered the top boundary of thesaturated system or water table. Depending on theage-dating technique, it may have other meanings,such as the amount of time since the water enteredat the land surface.

8. P. A. Domenico, F. W. Schwartz, Physical and Chem-ical Hydrogeology ( Wiley, New York, 1998).

9. Equation 1 is for a homogeneous, confined, constantdensity groundwater basin and is derived from thediffusion equation [(8), p. 72]. The system represen-tative of horizontal flow in a confined stream-aquifersystem, the lower aquifer of the Mississippi Riveralluvium in Iowa (80), has a characteristic length of1000 m and a hydraulic conductivity of 10 m/day.The system representative of vertical flow in a thickregional low-permeability unit, the Pierre Shale inNorth Dakota (81), has a characteristic length of200 m and a hydraulic conductivity of 1029 m/day. Aspecific storage of 1026 m21 was assumed to berepresentative of both systems.

10. The time of travel for the confined, stream-aquifersystem (80) is obtained by dividing the characteristiclength of 1000 m by the velocity of the fluid move-ment. The velocity (v) of the fluid movement iscalculated by dividing Darcy’s law by a porosity (n) of0.3, assuming a representative uniform head gradient(i) of 0.001 (v 5 Ki/n).

11. Rock Fractures and Fluid Flow (National Academy ofSciences, Washington, DC, 1996).

12. T. W. Rayne, K. R. Bradbury, M. A. Muldoon, Hydro-geol. J. 9, 432 (2001).

13. W. E. Sanford, S. Buapeng, Hydrogeol. J. 4, 26 (1996).14. G. Garven, Annu. Rev. Earth Planet. Sci. 23, 89

(1995).15. B. J. O. L. McPherson, J. D. Bredehoeft, Water Resour.

Res. 37, 857 (2001).16. B. W. Bruce, G. P. Oelsner, J. Am. Water Res. Assoc.

37, 1389 (2001).17. J. D. Bredehoeft, S. S. Papadopulos, H. H. Cooper Jr., in

Scientific Basis of Water Resources Management (Na-tional Academy Press, Washington, DC, 1982), pp.51–57.

18. W. M. Alley, T. E. Reilly, O. L. Franke, U.S. Geol. Surv.Circ. 1186 (1999).

19. D. A. Sverjensky, G. Garven, Nature 356, 481 (1992).20. J. P. York, M. Person, W. J. Gutowski, T. C. Winter, Adv.

Water Resour. 25, 221 (2002).21. P. Lichtner, C. Steefel, E. Oelkers, Eds., special issue on

Reactive Transport in Porous Media, Rev. Mineral. 34(1996).

22. E. P. Poeter, M. C. Hill, Ground Water 35, 250 (1997).23. P. G. Cook, J.-K. Bohlke, in Environmental Tracers in

Subsurface Hydrology, P. G. Cook, A. L. Herczeg, Eds.(Kluwer, Boston, 1999), pp. 1–30.

24. P. G. Cook, D. K. Solomon, J. Hydrol. 191, 245 (1997).25. L. N. Plummer et al., Appl. Geochem. 13, 995 (1998).26. T. E. Reilly, L. N. Plummer, P. J. Phillips, E. Busenberg,

Water Resour. Res. 30, 421 (1994).27. Z. Top, L. E. Brand, R. D. Corbett, W. Burnett, J. Chan-

ton, J. Coastal Res. 17, 859 (2001).28. A. M. Shapiro, Water Resour. Res. 37, 507 (2001).29. W. Edmunds, S. Tyler, Hydrogeol. J. 10, 216 (2002).30. A. L. Flint, L. E. Flint, E. M. Kwicklis, G. S. Bodvarsson,

J. M. Fabryka-Martin, Rev. Geophys. 39, 447 (2001).31. D. N. Lerner, A. S. Issar, I. Simmers, Groundwater

Recharge, vol. 8 of International Association of Hy-drogeologists, International Contributions to Hydro-geology (Heise, Hannover, West Germany, 1990).

32. G. Favreau et al., Ground Water 40, 144 (2002).33. M. Sophocleous, J. Hydrol. 137, 113 (1992).34. B. R. Scanlon, R. W. Healy, P. G. Cook, Hydrogeol. J.

10, 18 (2002).35. J. A. Huisman, C. Sperl, W. Bouten, J. M. Verstraten, J.

Hydrol. 245, 48 (2001).36. D. R. Pool, J. H. Eychaner, Ground Water 33, 425

(1995).37. J. Wahr, M. Molenaar, F. Bryan, J. Geophys. Res. 103,

30205 (1998).38. M. Sophocleous, S. P. Perkins, J. Hydrol. 236, 185

(2000).39. G. van der Kamp, R. Schmidt, Geophys. Res. Lett. 24,

719 (1997).40. T. H. S. Kotb, K. K. Tanji, T. Watanabe, Y. Ogino, Agric.

Water Manage. 43, 239 (2000).41. D. N. Lerner, Hydrogeol. J. 10, 143 (2002).42. M. Norin, A-M. Hulten, C. Svensson, in Groundwater

in the Urban Environment—Selected City Profiles,J. Chilton, Ed. (Balkema, Rotterdam, 1999), pp. 209–216.

43. U. Beyerle et al., Science 282, 731 (1998).

44. F. M. Phillips, Soil Sci. Soc. Am. J. 58, 15 (1994).45. T. C. Winter, J. W. Harvey, O. L. Franke, W. M. Alley,

U.S. Geol. Surv. Circ. 1139 (1998).46. P. Marmonier et al., in Subterranean Ecosystems, H.

Wilkens, D. C. Culver, W. F. Humphreys, Eds., vol. 30of Ecosystems of the World (Elsevier, Amsterdam,2000), pp. 513–531.

47. J. M. Sharp Jr., J. L. Banner, in The Earth Around Us:Maintaining a Livable Planet, J. L. Schneiderman, Ed.(Freeman, New York, 2000), pp. 154–165.

48. D. O. Rosenberry, R. G. Striegl, D. L. Hudson, GroundWater 38, 296 (2000).

49. J. A. Jones, P. J. Mulholland, Eds., Streams and Ground-water (Academic Press, San Diego, CA, 2000).

50. A. J. Boulton, Ecosyst. Health 6, 108 (2000).51. R. J. G. Mortimer, M. D. Krom, D. R. Boyle, A. Nishri,

Limnol. Oceanogr. 44, 1802 (1999).52. C. J. D. Fraser, N. T. Roulet, M. Laffleur, J. Hydrol. 246,

142 (2001).53. W. S. Moore, Mar. Chem. 65, 111 (1999).54. A. R. Basu, S. B. Jacobsen, R. J. Poreda, C. B. Dowling,

P. K. Aggarwal, Science 293, 1470 (2001).55. G. Power, R. S. Brown, J. G. Imhof, Hydrol. Proc. 13,

401 (1999).56. J. Constantz, Water Resour. Res. 34, 1609 (1998).57. F. Malard, A. Mangin, U. Uehlinger, J. V. Ward, Can. J.

Fish. Aquat. Sci. 58, 1319 (2001).58. M. Manga, Annu. Rev. Earth Planet. Sci. 29, 201

(2001).59. C.-H. Yang, L.-T. Tong, G.-F. Huang, Geophysics 64,

417 (1999).60. J. Bear, A. H.-D. Cheng, S. Soreck, D. Ouazar, I. Her-

rera, Eds., Seawater Intrusion in Coastal Aquifers—Concepts, Methods, and Practices (Kluwer, Dor-drecht, Netherlands, 1999).

61. T. D. Lahm, E. S. Bair, Water Resour. Res. 36, 3167(2000).

62. J. Laroche et al., Global Change Biol. 3, 397 (1997).63. H. Kooi, J. Groen, J. Hydrol. 246, 19 (2001).64. D. Galloway, D. R. Jones, S. E. Ingebritsen, Eds., U.S.

Geol. Surv. Circ. 1182 (1999).65. Equation 1 can provide an indication of the time

required for the excess pore pressure to dissipateafter an instantaneous change in head at the bound-aries of an aquitard by substituting the aquitardvalues for hydraulic conductivity, specific storage,and thickness (divided by 2 for doubly draining aqui-tard).

66. M. Sneed, D. L. Galloway, U.S. Geol. Surv. Water Res.Invest. Rep. 00-4015 (2000).

67. D. L. Galloway et al., Water Resour. Res. 34, 2573(1998).

68. G. W. Bawden, W. Thatcher, R. S. Stein, K. W. Hudson,G. Peltzer, Nature 412, 812 (2001).

69. J. Hoffmann, H. A. Zebker, D. L. Galloway, F. Amelung,Water Resour. Res. 37, 1551 (2001).

70. J. Hoffmann, D. L. Galloway, H. A. Zebker, Eos 82 no.(47), abstract H41E-0320 (2001).

71. F. Amelung, D. L. Galloway, J. W. Bell, H. A. Zebker,R. L. Laczniak, Geology 27, 483 (1999).

72. C. J. Taylor, W. M. Alley, U.S. Geol. Surv. Circ. 1217(2001).

73. C. J. Vorosmarty et al., Eos 82, 54 (2001).74. J. N. Mantua, S. R. Hare, Y. Zhang, J. M. Wallace, R. C.

Francis, Bull. Am. Meteorol. Soc. 78, 1069 (1997).75. H. H. Loosli et al., Eos 79, 576 (1998).76. W. W. Woessner, Ground Water 38, 423 (2000).77. Aquifer Storage and Recovery in the Comprehensive

Everglades Restoration Plan (National Academy ofSciences, Washington, DC, 2001).

78. J. L. Shelton et al., U.S. Geol. Surv. Water Resour.Invest. Rep. 01-4188 (2001).

79. F. W. Schwartz, M. Ibaraki, Ground Water 39, 492(2001).

80. R. A. Boyd, U.S. Geol. Surv. Water Resour. Invest. Rep.00-4274 (2001).

81. C. E. Neuzil, Water Resour. Res. 29, 2007 (1993).82. W. H. Schlesinger, in Biogeochemistry—An Analysis of

Global Change (Academic Press, San Diego, CA, ed. 2,1997), chapter 10.

83. R. H. Johnston, U.S. Geol. Surv. Prof. Pap. 1425(1997).

84. We thank A. J. Boulton, L. F. Konikow, A. M. Shapiro,E. P. Weeks, T. C. Winter, and an anonymous reviewerfor insightful comments and suggestions.



1985

Documents