Chapter 9. Groundwater & SurfaceWater Pollution

Principles of GroundwaterFlow

9.1GROUNDWATER AND AQUIFERS Definition of Groundwater Aquifers Physical Properties of Soils and

Liquids Physical Properties of Soils Physical Properties of Water

Physical Properties of Vadose Zones andAquifers

Physical Properties of VadoseZones

Physical Properties of Aquifers

9.2FUNDAMENTAL EQUATIONS OF GROUNDWATERFLOW Intrinsic Permeability Validity of Darcy’s Law Generalization of Darcy’s Law Equation of Continuity Fundamental Equations

9.3CONFINED AQUIFERS One-Dimensional Horizontal Flow Semiconfined Flow Radial Flow Radial Flow in a Semiconfined

Aquifer Basic Equations

9.4UNCONFINED AQUIFERS Discharge Potential and Continuity

Equation Basic Differential Equation One-Dimensional Flow Radial Flow Unconfined Flow with Infiltration One-Dimensional Flow with Infiltra-

tion Radial Flow with Infiltration Radial Flow from Pumping with Infil-

tration

9.5COMBINED CONFINED AND UNCONFINEDFLOW One-Dimensional Flow Radial Flow

Hydraulics of Wells

9.6TWO-DIMENSIONAL PROBLEMS Superposition

A Two-Well System A Multiple-Well System

Method of Images Well Near a Straight River Well Near a Straight Impervious

Boundary Well in a Quarter Plane

Potential and Flow Functions

9Groundwater and SurfaceWater Pollution

©1999 CRC Press LLC

GROUNDWATER POLLUTION CONTROL

Yong S. Chae | Ahmed Hamidi

9.7NONSTEADY (TRANSIENT) FLOW Transient Confined Flow (Elastic

Storage) Transient Unconfined Flow (Phreatic

Storage) Transient Radial Flow (Theis Solu-

tion)

9.8DETERMINING AQUIFER CHARACTER-ISTICS Confined Aquifers

Steady-State Transient-State

Semiconfined (Leaky) Aquifers Steady-State Transient-State

Unconfined Aquifers Steady-State Transient-State

Slug Tests

9.9DESIGN CONSIDERATIONS Well Losses Specific Capacity Partially Penetrating Wells (Imperfect

Wells) Confined Aquifers Unconfined Aquifers

9.10INTERFACE FLOW Confined Interface Flow Unconfined Interface Flow Upconing of Saline Water Protection Against Intrusion

Principles of GroundwaterContamination

9.11CAUSES AND SOURCES OF CONTAMI-NATION Waste Disposal

Liquid Waste Solid Waste

Storage and Transport of CommercialMaterials

Storage Tanks Spills

Mining Operations Mines

Oil and Gas Agricultural Operations

Fertilizers Pesticides

Other Activities Interaquifer Exchange Saltwater Intrusion

9.12FATE OF CONTAMINANTS IN GROUND-WATER Organic Contaminants

Hydrolysis Oxidation–Reduction Biodegradation Adsorption Volatilization

Inorganic Contaminants Nutrients Acids and Bases Halides Metals

9.13TRANSPORT OF CONTAMINANTS INGROUNDWATER Transport Process

Advection Dispersion Retardation

Contaminant Plume Behavior Contaminant Density Contaminant Solubility Groundwater Flow Regime Geology

Groundwater Investigation andMonitoring

9.14INITIAL SITE ASSESSMENT Interpretation of Existing Information

Site-Specific Information Regional Information

Initial Field Screening Surface Geophysical Surveys Downhole Geophysical Surveys Onsite Chemical Surveys

9.15SUBSURFACE SITE INVESTIGATION Subsurface Drilling

Drilling Methods Soil Sampling



Monitoring Well Installation Well Location and Number Casings and Screens Filter Packs and Annular Seals Well Development

Groundwater Sampling Purging Collection and Pretreatment Quality Assurance and Quality

Control

Groundwater Cleanup andRemediation

9.16SOIL TREATMENT TECHNOLOGIES Excavation and Removal Physical Treatment

Soil–Vapor Extraction Soil Washing Soil Flushing

Biological Treatment Slurry Biodegradation Ex Situ Bioremediation and Land-

farming In Situ Biological Treatment

Thermal Treatment Incineration Thermal Desorption

Stabilization and Solidification Treat-ment

Stabilization Vitrification

9.17PUMP-AND-TREAT TECHNOLOGIES Withdrawal and Containment

Systems Well Systems Subsurface Drains

Treatment Systems Density Separation Filtration Carbon Adsorption Air Stripping Oxidation and Reduction

Limitations of Pump-and-Treat Technol-ogies

9.18IN SITU TREATMENT TECHNOLOGIES Bioremediation

Design Considerations Advantages and Limitations

Air Sparging Design Considerations Advantages and Limitations

Other Innovative Technologies Neutralization and Detoxification Permeable Treatment Beds Pneumatic Fracturing Thermally Enhanced Recovery

9.19INTEGRATED STORM WATER PROGRAM Integrated Management Approach

Federal Programs State Programs Municipal Programs

9.20NONPOINT SOURCE POLLUTION Major Types of Pollutants Nonpoint Sources

Atmospheric Deposition Erosion Accumulation/Washoff

Direct Input from Pollutant Source

9.21BEST MANAGEMENT PRACTICES Planning

Land Use Planning Natural Drainage Features Erosion Controls

Maintenance and OperationalPractices

Urban Pollutant Control Collection System Maintenance Inflow and Infiltration Drainage Channel Maintenance

9.22FIELD MONITORING PROGRAMS Selection of Water Quality Parameters Acquisition of Representative Samples

Sampling Sites and Location Sampling Methods Flow Measurement

Sampling Equipment Manual Sampling Automatic Sampling Flowmetering Devices

QA/QC Measures Sample Storage Sample Preservation

Analysis of Pollution Data Storm Loads Annual Loads Simulation Model Calibration

Statistical Analysis

9.23DISCHARGE TREATMENT Biological Processes Physical-Chemical Processes Physical Processes

Swirl-Flow Regulator-Concentrator Sand Filters Enhanced Filters Compost Filters


STORM WATER POLLUTANT MANAGEMENT

David H.F. Liu | Kent K. Mao

This section defines groundwater and aquifers and dis-cusses the physical properties of soils, liquids, vadosezones, and aquifers.

Definition of GroundwaterWater exists in various forms in various places. Water canexist in vapor, liquid, or solid forms and exists in the at-mosphere (atmospheric water), above the ground surface(surface water), and below the ground surface (subsurfacewater). Both surface and subsurface waters originate fromprecipitation, which includes all forms of moisture fromclouds, including rain and snow. A portion of the precip-itated liquid water runs off over the land (surface runoff),infiltrates and flows through the subsurface (subsurfaceflow), and eventually finds its way back to the atmospherethrough evaporation from lakes, rivers, and the ocean;transpiration from trees and plants; or evapotranspirationfrom vegetation. This chain process is known as the hy-drologic cycle. Figure 9.1.1 shows a schematic diagram ofthe hydrologic cycle.

Not all subsurface (underground) water is groundwa-ter. Groundwater is that portion of subsurface water whichoccupies the part of the ground that is fully saturated andflows into a hole under pressure greater than atmosphericpressure. If water does not flow into a hole, where thepressure is that of the atmosphere, then the pressure in wa-ter is less than atmospheric pressure. Depths of ground-water vary greatly. Places exist where groundwater has notbeen reached at all (Bouwer 1978).

The zone between the ground surface and the top ofgroundwater is called the vadose zone or zone of aeration.This zone contains water which is held to the soil parti-cles by capillary force and forces of cohesion and adhe-sion. The pressure of water in the vadose zone is negativedue to the surface tension of the water, which produces anegative pressure head. Subsurface water can therefore beclassified according to Table 9.1.1.

Groundwater accounts for a small portion of theworld’s total water, but it accounts for a major portion ofthe world’s freshwater resources as shown in Table 9.1.2.

Table 9.1.2 illustrates that groundwater representsabout 0.6% of the world’s total water. However, exceptfor glaciers and ice caps, it represents the largest source of

freshwater supply in the world’s hydrologic cycle. Sincemuch of the groundwater below a depth of 0.8 km is salineor costs too much to develop, the total volume of readilyusable groundwater is about 4.2 million cubic km (Bouwer1978).

Groundwater has been a major source of water supplythroughout the ages. Today, in the United States, ground-water supplies water for about half the population andsupplies about one-third of all irrigation water. Some three-fourths of the public water supply system uses ground-water, and groundwater is essentially the only water sourcefor the roughly 35 million people with private systems(Bouwer 1978).

AquifersGroundwater is contained in geological formations, calledaquifers, which are sufficiently permeable to transmit andyield water. Sands and gravels, which are found in allu-vial deposits, dunes, coastal plains, and glacial deposits,are the most common aquifer materials. The more porousthe material, the higher yielding it is as an aquifer mater-ial. Sandstone, limestone with solution channels, and otherKarst formations are also good aquifer materials. In gen-eral, igneous and metamorphic rocks do not make goodaquifers unless they are sufficiently fractured and porous.

Figure 9.1.2 schematically shows the types of aquifers.The two main types are confined aquifers and unconfinedaquifers. A confined aquifer is a layer of water-bearing ma-terial overlayed by a relatively impervious material. If theconfining layer is essentially impermeable, it is called anaquiclude. If it is permeable enough to transmit water ver-tically from or to the confined aquifer, but not in a hori-zontal direction, it is called an aquitard. An aquifer boundby one or two aquitards is called a leaky or semiconfinedaquifer.

Confined aquifers are completely filled with ground-water under greater-than-atmospheric pressure and there-fore do not have a free water table. The pressure condi-tion in a confined aquifer is characterized by a piezometricsurface, which is the surface obtained by connecting equi-librium water levels in tubes or piezometers penetratingthe confined layer.


Principles of Groundwater Flow

9.1GROUNDWATER AND AQUIFERS

An unconfined aquifer is a layer of water-bearing ma-terial without a confining layer at the top of the ground-water, called the groundwater table, where the pressure isequal to atmospheric pressure. The groundwater table,sometimes called the free or phreatic surface, is free to riseor fall. The groundwater table height corresponds to theequilibrium water level in a well penetrating the aquifer.Above the water table is the vadoze zone, where water

pressures are less than atmospheric pressure. The soil inthe vadoze zone is partially saturated, and the air is usu-ally continuous down to the unconfined aquifer.

Physical Properties of Soils andLiquidsThe following discussion describes the physical propertiesof soils and liquids. It also defines the terms used to de-scribe these properties.

PHYSICAL PROPERTIES OF SOILS

Natural soils consist of solid particles, water, and air.Water and air fill the pore space between the solid grains.Soil can be classified according to the size of the particlesas shown in Table 9.1.3.

Soil classification divides soils into groups and sub-groups based on common engineering properties such astexture, grain size distribution, and Atterberg limits. Themost widely accepted classification system is the unifiedclassification system which uses group symbols for identi-fication, e.g., SW for well-graded sand and CH for inor-ganic clay of high plasticity. For details, refer to any stan-dard textbook on soil mechanics.

Figure 9.1.3 shows an element of soil, separated in threephases. The following terms describe some of the engi-neering and physical properties of soils used in ground-water analysis and design:


Return Flowfrom Irrigation

Groundwater Flow(Saturated Flow)

Groundwater Table

Flow fromSeptic Tanks

Freshwater-Salt WaterInterface

Tr

Return

SRLake

ESR

ET

Spring

E

ET (fromVegetation)

ET = EvapotranspirationE = EvaporationTr = TranspirationSR = Surface RunoffIn = Infiltration

SR

SR

E

In

Snow and Ice

Sublimation Precipitation(on Land)

Clouds

Clouds

Tr

Precipitation(on the Coast)

Movement of MoistAir Masses

Ocean

Sea Water

Leakage

Unsaturated Flow

River

FIG. 9.1.1 Schematic diagram of the hydrologic cycle.

TABLE 9.1.1 CLASSIFICATION OF SUBSURFACEWATER

Vadoze Soil Water

Subsurface Zone Intermediate Vadoze Water

Water Capillary Water

Zone of Groundwater

Saturation (Phreatic Water)

Internal Water

POROSITY (n)—A measure of the amount of pores in thematerial expressed as the ratio of the volume of voids(Vv) to the total volume (V), n = Vv/V. For sandy soilsn = 0.3 to 0.5; for clay n > 0.5.

VOID RATIO (e)—The ratio between Vv and the volumeof solids VS, e = Vv/VS; where e is related to n as e =n/(1 – n).

WATER CONTENT (v)—The ratio of the amount of waterin weight (WW) to the weight of solids (WS), v =WW/WS.

DEGREE OF SATURATION (S)—The ratio of the volume ofwater in the void space (VW) to Vv, S = VW/Vv. S variesbetween 0 for dry soil and 1 (100%) for saturated soil.

COEFFICIENT OF COMPRESSIBILITY (a)—The ratio of thechange in soil sample height (h) or volume (V) to thechange in applied pressure (sv)

a = 2}1

h} }

d

d

s

h

v

} = 2}V

1} }

dds

V

v

} 9.1(1)

The a can be expressed as

a = }(1 +

E

m

(1

)(

2

1 2

m)

2m)} = 9.1(2)

where:

E 5 Young’s modulusm 5 Poisson’s ratioB 5 bulk modulusG 5 shear modulus

Clay exists in either a dispersed or flocculated structuredepending on the arrangement of the clay particles with

1}

B + }4

3} G


TABLE 9.1.2 ESTIMATED DISTRIBUTION OF WORLD’S WATER

Volume Percentage of1000 km3 Total Water

Atmospheric water 13.25 000.001Surface water

Salt water in oceans 1,320,000.25 097.2Salt water in lakes and inland seas 104.25 000.008Fresh water in lakes 125.25 000.009Fresh water in stream channels (average) 1.25 000.0001Fresh water in glaciers and icecaps 29,000.25 002.15Water in the biomass 50.25 000.004

Subsurface waterVadose water 67.25 000.005Groundwater within depth of 0.8 km 4200.25 000.31Groundwater between 0.8 and 4 km depth 4200 0.31

Total (rounded) 1,360,000.25 100

Source: H. Bouwer, 1978, Groundwater hydrology (McGraw-Hill, Inc.).

Aquifer C

Aquifer B

Aquifer A

Interface

LeakageInterface

SeaWater

Sea

PerchedWater

WaterTable

FlowingWell

GroundSurface

RechargeArea

Leakage

Piezometric surface (B)

Piezometric surface (C)

ConfinedPhreatic Leaky Artesian Confined Leaky

Aqui er B

Impervious Stratum

Semipervious Stratum

FIG. 9.1.2 Types of aquifers.

the type of cations that are adsorbed to the clay. If thelayer of adsorbed cation (such as Ca

11) is thin and the clayparticles can be close together, making the attractive vander Waals forces dominant between the particles, then theclay is flocculated. If the clay particles are kept some dis-tance apart by adsorbed cations (such as N1

a), the repul-sive electrostatic forces are dominant, and the clay is dis-persed. Since clay particles are negatively charged, whichcan adsorb cations from the soil solution, clay can be con-verted from a dispersed state to a flocculant conditionthrough the process of cation exchange (e.g. N1

a ® Ca11)

which changes the adsorbed ions. The reverse, changingfrom a flocculated to a dispersed clay, can also occur. Claystructure change is used to handle some groundwater prob-lems in clay because the hydraulic properties of soil aredependent upon the clay structure.

PHYSICAL PROPERTIES OF WATER

The density of a material is defined as the mass per unitvolume. The density (r) of water varies with temperature,pressure, and the concentration of dissolved materials andis about 1000 kg/m3. Multiplying r by the acceleration ofgravity (g) gives the specific weight (g) as g < rg. For wa-ter, g < 9.8 kN/m3.

Some of the physical properties of water are defined asfollows:

DYNAMIC VISCOSITY (m)—The ratio of shear stress (tyx) inx direction, acting on an x–y plane to velocity gradient(dvx/dy); tyx 5 m dvx/dy. For water, m 5 1023 kg/m z s.

KINEMATIC VISCOSITY (y)—Related to m by y 5 m/r. Itsvalue is about 1026 m2/s for water.

COMPRESSIBILITY (b)—The ratio of change in densitycaused by change in pressure to the original density

b 5 }1r

} }ddpr} 5 2}

V1

} }ddVp}

b < 0.5 3 1029 m2/N 9.1(3)

The variation of density and viscosity of water withtemperature can be obtained from Table 9.1.4.

Physical Properties of Vadose Zonesand AquifersA description of the physical properties of vadose zonesand aquifers follows.

PHYSICAL PROPERTIES OF VADOSEZONES

As discussed earlier, the pressure of water in the vadosezone is negative, and the negative pressure head or capil-lary pressure is proportional to the vertical distance abovethe water table. Figure 9.1.4 shows a characteristic curve


TABLE 9.1.3 USUAL SIZE RANGE FOR GENERAL SOILCLASSIFICATION TERMINOLOGY

Material Upper, mm Lower, mm Comments

Boulders, cobbles 10001 752

Gravel, pebbles 75 2–5 No. 4 or larger sieveSand 2–5 0.074 No. 4 to No. 200 sieveSilt 0.074–0.05 0.006 InertRock flour 0.006 InertClay 0.002 0.001 Particle attraction, water

absorptionColloids 0.001

Source: J.E. Bowles, 1988, Foundation analysis and design, 4th ed. (McGraw-Hill).

FIG. 9.1.3 Three-phase relationship in soils.

(c)

Air Wg > 0

Ww

Ws

1 1

e

1 5

Ws /g

wG

se

Ww

/gw

V

Vs

Vy

Vw

Va

(b)

Vy

Vs1.0

e

1 2

n

1.00

n

V 5

1 1

e

Air

Water

Soil

(a)

V

Vs

Vy

Vw

Va

Ws

Ww

W

of the relationship between volumetric water content andthe negative pressure head (height above the water tableor capillary pressure).

For materials with relatively uniform particle size andlarge pores, the water content decreases abruptly once theair-entry value is reached. These materials have a well-de-fined capillary fringe. For well-graded materials and ma-terials with fine pores, the water content decreases moregradually and has a less well-defined capillary fringe.

At a large capillary pressure, the volumetric water con-tent tends towards a constant value because the forces ofadhesion and cohesion approach zero. The volumetric wa-ter content at this state is equal to the specific retention.The specific retention is then the amount of water retainedagainst the force of gravity compared to the total volumeof the soil when the water from the pore spaces of an un-confined aquifer is drained and the groundwater table islowered.

PHYSICAL PROPERTIES OF AQUIFERS

As stated before, an aquifer serves as an underground stor-age reservoir for water. It also acts as a conduit throughwhich water is transmitted and flows from a higher levelto a lower level of energy. An aquifer is characterized bythe three physical properties: hydraulic conductivity, trans-missivity, and storativity.

Hydraulic Conductivity

Hydraulic conductivity, analogous to electric or thermalconductivity, is a physical measure of how readily anaquifer material (soil) transmits water through it. Mathe-matically, it is the proportionality between the rate of flowand the energy gradient causing that flow as expressed inthe following equation. Therefore, it depends on the prop-erties of the aquifer material (porous medium) and the fluidflowing through it.

K 5 k }m

g} 9.1(4)

where:

K 5 hydraulic conductivity (called the coefficient of per-meability in soil mechanics)

k 5 intrinsic permeabilityg 5 specific weight of fluidm 5 dynamic viscosity of fluid

For a given fluid under a constant temperature and pres-sure, the hydraulic conductivity is a function of the prop-erties of the aquifer material, that is, how permeable thesoil is. The subject of hydraulic conductivity is discussedin more detail in Section 9.2.

Transmissivity

Transmissivity is the physical measure of the ability of anaquifer of a known dimension to transmit water throughit. In an aquifer of uniform thickness d, the transmissivityT is expressed as

T 5_Kd 9.1(5)

where _K represents an average hydraulic conductivity.

When the hydraulic conductivity is a continuous functionof depth

_K 5 }

1d

} Ed

oKz dz 9.1(6)

When a medium is stratified, either in horizontal (x) orvertical (y) direction with respect to hydraulic conductiv-ity as shown in Figure 9.1.5, the average value

_K can be

obtained by_Kx 5 ^

n

m51}Km

d

dm} 9.1(7)


TABLE 9.1.4 VARIATION OF DENSITY ANDVISCOSITY OF WATER WITHTEMPERATURE

Temperature Density Dynamic Viscosity(°C) (kg/m3) (kg/m s)

0 999.868 1.79 3 1023

5 999.992 1.52 3 1023

10 999.727 1.31 3 1023

15 999.126 1.14 3 1023

20 998.230 1.01 3 1023

Source: A. Verrjuitt, 1982, Theory of groundwater flow, 2d ed. (MacmillanPublishing Co.).

FIG. 9.1.4 Schematic equilibrium water-content distributionabove a water table (left) for a coarse uniform sand (A), a fineuniform sand (B), a well-graded fine sand (C), and a clay soil(D). The right plot shows the corresponding equilibrium water-content distribution in a soil profile consisting of layers of ma-terials A, B, and D.

0.1 0.2 0.3 0.4 0.5

100

200

300

0.1 0.2 0.3 0.4 0.5

100

200

300

0 0

D

B

AA

DC

B

VOLUMETRIC WATER CONTENT

DIS

TA

NC

E A

BO

VE

WA

TE

R T

AB

LE IN

CM

_Ky 5 9.1(8)

Storativity

Storativity, also known as the coefficient of storage or spe-cific yield, is the volume of water yielded or released per

d}

^n

m51}Kdm

m

}

unit horizontal area per unit drop of the water table in anunconfined aquifer or per unit drop of the piezometric sur-face in a confined aquifer. Storativity S is expressed as

S 5 }A1

} }ddQf} 9.1(9)

where:

dQ 5 volume of water released or restoreddf 5 change of water table or piezometric surface

Thus, if an unconfined aquifer releases 2 m3 water asa result of dropping the water table by 2m over a hori-zontal area of 10 m2, the storativity is 0.1 or 10%.

—Y.S. Chae

ReferenceBouwer, H. 1978. Groundwater hydrology. McGraw-Hill, Inc.


FIG. 9.1.5 Permeability of layered soils.

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Directionofflow

Direction of flow

x

y

d

k1

k2

kn dn

d1

d2

��

��

�

9.2FUNDAMENTAL EQUATIONS OF GROUNDWATERFLOW

The flow of water through a body of soil is a complexphenomenon. A body of soil constitutes, as described inSection 9.1, a solid matrix and pores. For simplicity, as-sume that all pores are interconnected and the soil bodyhas a uniform distribution of phases throughout. To findthe law governing groundwater flow, the phenomenon isdescribed in terms of average velocities, average flow paths,average flow discharge, and pressure distribution across agiven area of soil.

The theory of groundwater flow originates with HenryDarcy who published the results of his experimental workin 1856. He performed a series of experiments of the typeshown in Figure 9.2.1. He found that the total dischargeQ was proportional to cross-sectional area A, inverselyproportional to the length Ds, and proportional to thehead difference f1 2 f2 as expressed mathematically inthe form

Q 5 KA }f1

D

2

s

f2} 9.2(1)

where K is the proportionality constant representing hy-draulic conductivity. This equation is known as Darcy’sequation. The quantity Q/A is called specific discharge q.If f1 2 f2 5 Df and Ds ® 0, Equation 9.2(1) becomes

q 5 2K }ddf

s} 9.2(2)

This equation states that the specific discharge is directlyproportional to the derivative of the head in the directionof flow (hydraulic gradient). The specific discharge is alsoknown as Darcy’s velocity. Note that q is not the actualflow velocity (seepage velocity) because the flow is limitedto pore space only. The seepage velocity v is then

Reference level

p1lpgp2lpg

f2

f1

z1

z2

Area A

Flow

Ds

FIG. 9.2.1 Darcy’s experiment.

v 5 }nQz A} 5 }

qn

} 9.2(3)

where n is the porosity of the soil. Note that v is alwayslarger than q.

Intrinsic PermeabilityThe hydraulic conductivity K is a material constant, andit depends not only on the type of soil but also on the typeof fluid (dynamic viscosity m) percolating through it. Thehydraulic conductivity K is expressed as

K 5 k }m

g} 9.2(4)

where k is called the intrinsic permeability and is now aproperty of the soil only. Many attempts have been madeto express k by such parameters as average pore diame-ter, porosity, and effective soil grain size. The most famil-iar equation is that of Kozeny-Carmen

k 5 Cd2 }(1 2

n3

n)2} 9.2(5)

where:

n 5 porosityd 5 the effective pore diameterC 5 a constant to account for irregularities in the geom-

etry of pore space

Another equation by Hazen states

k 5 CD2 5 C1D210 9.2(6)

where:

D 5 the average grain diameterD10 5 the effective diameter of the grains retained

Values of hydraulic conductivity can be obtained from em-pirical formulas, laboratory experiments, or field tests.Table 9.2.1 gives the typical values for various aquifer ma-terials.

Validity of Darcy’s LawDarcy’s law is restricted to a specific discharge less than acertain critical value and is valid only within a laminar

flow condition, which is expressed by Reynolds numberRe defined as

Re 5 }qD

m

r} 5 }

qn

D} 9.2(7)

Experiments have shown the range of validity of Darcy’slaw to be

Re # 1 , 10 9.2(8)

In practice, the specific discharge is always small enoughfor Darcy’s law to be applicable. Only cases of flowthrough coarse materials, such as gravel, deviate fromDarcy’s law. Darcy’s law is not valid for flow through ex-tremely fine-grained soils, such as colloidal clays.

Generalization of Darcy’s LawIn practice, flow is seldom one dimensional, and the mag-nitude of the hydraulic gradient is usually unknown. Thesimple form, Equation 9.2(2), of Darcy’s law is not suit-able for solving problems. A generalized form must beused, assuming the hydraulic conductivity K to be the samein all directions, as

qx 5 2K }¶¶f

x}

qy 5 2K }¶¶f

y}

qz 5 2K }¶¶f

z} 9.2(9)

For an anisotropic material, these equations can be writ-ten as

qx 5 2Kxx }¶¶f

x} 2 Kxy }

¶¶f

y} 2 Kxz }

¶¶f

z}

qy 5 2Kyx }¶¶f

x} 2 Kyy }

¶¶f

y} 2 Kyz }

¶¶f

z}

qz 5 2Kzx }¶¶f

x} 2 Kzy }

¶¶f

y} 2 Kzz }

¶¶f

z} 9.2(10)

In the special case that Kxy 5 Kxz 5 Kyx 5 Kyz 5 Kzx 5Kzy 5 0, the x, y, and z directions are the principal direc-tions of permeability, and Equations 9.2(10) reduce to

qx 5 2Kxx }¶¶f

x} 5 2Kx }

¶¶f

x}

qy 5 2Kyy }¶¶f

y} 5 2Ky }

¶¶f

y}

qz 5 2Kzz }¶¶f

z} 5 2Kz }

¶¶f

z} 9.2(11)

This chapter considers isotropic soils since problems foranisotropic soils can be easily transformed into problemsfor isotropic soils.


TABLE 9.2.1 THE ORDER OF MAGNITUDE OF THEPERMEABILITY OF NATURAL SOILS

k (m2) K (m/s)

Clay 10217 to 10215 10210 to 1028

Silt 10215 to 10213 10282 to 1026

Sand 10212 to 10210 10252 to 1023

Gravel 10292 to 1028 10222 to 1021

Source: A. Verrjuit, 1982, Theory of groundwater flow, 2d ed. (MacmillanPublishing Co.).

Equation of ContinuityDarcy’s law furnishes three equations of motion for fourunknowns (qx, qy, qz, and f). A fourth equation notes thatthe flow phenomenon must satisfy the fundamental phys-ical principle of conservation of mass. When an elemen-tary block of soil is filled with water, as shown in Figure9.2.2, no mass can be gained or lost regardless of the pat-tern of flow.

The conservation principal requires that the sum of thethree quantities (the mass flow) is zero, hence when di-vided by Dx z Dy z Dz

}¶(

¶r

xqx)} 1 }

¶(¶r

yqy)} 1 }

¶(¶r

zqz)} 5 0 9.2(12)

When the density is a constant, then Equation 9.2(12) isreduced to

}¶¶qxx

} 1 }¶¶qyy

} 1 }¶¶qzz

} 5 0 9.2(13)

This equation is called the equation of continuity.

Fundamental EquationsDarcy’s law and the continuity equation provide four equa-tions for the four unknowns. Substituting Darcy’s lawEquation 9.2(9) into the equation of continuity Equation9.2(13) yields

}¶¶

2

xf2} 1 }

¶¶

2

yf2} 1 }

¶¶

2

zf2} 5 0 9.2(14)

or

=2f 5 0 9.2(15)

which is Laplace’s equation in three dimensions.Solving groundwater flow problems amounts to solv-

ing Laplace’s equation with the appropriate boundary con-ditions. It is essentially a mathematical problem.Sometimes a problem must be simplified before it can besolved, and these simplifications involve considering thephysical condition of groundwater flow.

—Y.S. Chae


Dy

(pqz)2

(pqy)2

(pqz)1

(pqx)1

(pqy)1

(pqx)2

Dx

Dz

FIG. 9.2.2 Conservation of mass.

9.3CONFINED AQUIFERS

This section discusses groundwater flow in confinedaquifers including one-dimensional horizontal flow, semi-confined flow, and radial flow. It also discusses radial flowin a semiconfined aquifer.

One-Dimensional Horizontal FlowOne-dimensional horizontal confined flow means thatwater is flowing through a confined aquifer in one di-rection only. Figure 9.3.1 shows an example of such aflow. Since qy 5 qz 5 0, the governing Equation 9.2(14)reduces to

}dd

2

xf2} 5 0 9.3(1)

and the general solution of this equation is f 5 Ax 1 B.Using the boundary conditions from Figure 9.3.1 of

x 5 0 f 5 f1

x 5 L f 5 f2

gives

f 5 f1 2 }f1 2

Lf2

} x 9.3(2)

Equation 9.3(2) indicates that the piezometric head f de-

L

H

x

D

D

z

z

f2f11

FIG. 9.3.1 One-dimensional flow in a confined aquifer.

creases linearly with distance. The specific discharge qx isthen found using Darcy’s law

qx 5 2K }¶¶f

x} 5 K }

f1 2

Lf2

} 9.3(3)

which follows that the specific discharge does not varywith position. The discharge flowing through the aquiferQx per unit length of the river bank is then

Qx 5 qx z H 5 KH }f1 2

Lf2

} 9.3(4)

Semiconfined FlowIf an aquifer is bound by one or two aquitards which al-low water to be transmitted vertically from or to the con-fined aquifer as shown in Figure 9.3.2, then a semicon-fined or leaky aquifer exists, and the flow through thisaquifer is called semiconfined flow. Small amounts of wa-ter can enter (or leave) the aquifer through the aquitardsof low permeability, which cannot be ignored. Yet in theaquifer proper, the horizontal flow dominates (qz 5 o isassumed).

The fundamental equation of semiconfined flow is de-rived from the principle of continuity and Darcy’s law asfollows:

Consider an element of the aquifer shown in Figure9.3.2. The net outward flux due to the flow in x and y di-rections is

2K 1}¶¶

2

xf2} 1 }

¶¶

2

yf2}2 Dx z Dy z H 9.3(5)

The amount of water percolating through the layers perunit time is

K1 }f 2

d1

f1} Dx z Dy

K2 }f 2

d2

f2} Dx z Dy 9.3(6)

Continuity now requires that the sum of these quantitiesbe zero, hence

KH 1}¶¶

2

xf2} 1 }

¶¶

2

yf2}2 2 }

f 2

c1

f1} 2 }

f 2

c2

f2} 5 0 9.3(7)

where c1 5 d1/K1 and c2 5 d2/K2, which are calledhydraulic resistances of the confining layers. The terms(f 2 f1)/c1 and (f 2 f2)/c2 represent the vertical leakagethrough the confining layers.

Defining leakage factor l 5 ÖTc where T 5 KH, thetransmissivity of the aquifer, Equation 9.3(7), can be writ-ten as

}¶¶

2

xf2} 1 }

¶¶

2

yf2} 2 }

f

l

221

f1} 2 }

f

l

222

f2} 5 0 9.3(8)

This equation is the fundamental equation of semiconfinedflow. When the confining layers are completely imperme-able (K1 5 K2 5 0), Equation 9.3(8) reduces to Equation

9.2(14).

Radial FlowRadial flow in a confined aquifer occurs when the flow issymmetrical about a vertical axis. An example of radialflow is that of water pumped through a well in an openfield or a well located at the center of an island as shownin Figure 9.3.3. The distance R, called the radius of influ-ence zone, is the distance to the source of water where thepiezometric head f0 does not vary regardless of the amountof pumping. The radius R is well defined in the case ofpumping in a circular island. In an open field, however,the distance R is theoretically infinite, and a steady-statesolution cannot be obtained. In practice, this case does notoccur, and R can be obtained by empirical formula or mea-surements.

The differential equation governing radial flow is ob-tained when the cartesian coordinates used for rectilinearflow are transformed into polar coordinates as

}¶¶

2

xf2} 1 }

¶¶

2

yf2} 5 }

¶¶

2

rf2} 1 }

1r} }

¶¶f

r} 1 }

r12} }

¶¶

2

rf2} 1 }

r12} }

¶¶

2

u

f2} 9.3(9)


��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

K1

K

K2

Flow

d1

d2

H

DyDx

x

qx 1qx Dx

x

1

2f 5 f2

f 5 f1

qy 1 qy Dy 2y

qx

qy

z

y

��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��@@@��ÀÀÀ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

2R

fo

Qo

H

sw

2rw

(b) Well in circular island

Qo

H

(a) Well in open field

f (r)fo

��

��

��

��

��

��

��

FIG. 9.3.2 Semiconfined flow.

FIG. 9.3.3 Radial flow in a confined aquifer.

Since f is independent of angle u, the last term of thisequation can be dropped. The fundamental equation ofradial flow is then

}¶¶

2

rf2} 1 }

1r} }

¶¶f

r} 5 0 9.3(10)

or

}1r} }

ddr} 1r }

ddf

r}2 5 0 9.3(11)

The solution of this differential equation with boundaryconditions (Gupta 1989) yields

f 5 }2p

QKH} ln }

Rr} 1 fo 9.3(12)

This equation is known as the Thiem equation.To calculate the head at the well fw using Equation

9.3(12), substitute the radius of the well rw for r, whichgives

fw 5 }2p

QKH} ln 1}

rRw}2 1 fo 9.3(13)

Since the flow is confined, the head at the well must beabove the upper impervious boundary (f must be greaterthan H). Otherwise, the flow in that situation becomes un-confined flow, and Equation 9.3(13) is not applicable.

If the radius of influence zone is known or can be de-termined, the discharge rate is obtained by

Qo 5 2pKH 9.3(14)

and the drawdown s at any point is given by

s 5 fo 2 f 5 }2p

QKH} ln 1}

Rr}2 9.3(15)

Radial Flow in a SemiconfinedAquiferRadial flow in a semiconfined aquifer occurs when theflow is towards a well in an aquifer such as the one shownin Figure 9.3.4.

When leakage through the confining layer is considered,Equation 9.3(4) becomes

}¶¶

2

rf2} 1 }

1r} }

¶¶f

r} 2 }

f

l

221

f1} 5 0 9.3(16)

The general solution of this equation is

f 5 fo 1 AIo 1}l

r}2 1 BKo 1}

l

r}2 9.3(17)

where A and B are arbitrary constants, and Io and Ko aremodified Bessel functions of zero order and of the first andsecond kind, respectively. Table 9.3.1 is a short table ofthe four types of Bessel functions. The two constants are

fo 2 fw}

ln 1}rR

w

}2


TABLE 9.3.1 BESSEL FUNCTIONS

x I0(x) I1(x) K0(x) K1(x)

0.0 01.0000 0.0000 ` `0.1 01.0025 0.0501 2.4271 9.85380.2 01.0100 0.1005 1.7527 4.77600.3 01.0226 0.1517 1.3725 3.05600.4 01.0404 0.2040 1.1145 2.18440.5 01.0635 0.2579 0.9244 1.65640.6 01.0920 0.3137 0.7775 1.30280.7 01.1263 0.3719 0.6605 1.05030.8 01.1665 0.4329 0.5653 0.86180.9 01.2130 0.4971 0.4867 0.71651.0 01.2661 0.5652 0.4210 0.60191.1 01.3262 0.6375 0.3656 0.50981.2 01.3937 0.7147 0.3185 0.43461.3 01.4693 0.7973 0.2782 0.37261.4 01.5534 0.8861 0.2436 0.32081.5 01.6467 0.9817 0.2138 0.27741.6 01.7500 1.0848 0.1880 0.24061.7 01.8640 1.1963 0.1655 0.20941.8 01.9896 1.3172 0.1459 0.18261.9 02.1277 1.4482 0.1288 0.15972.0 02.2796 1.5906 0.1139 0.13992.1 02.4463 1.7455 0.1008 0.12282.2 02.6291 1.8280 0.0893 0.10792.3 02.8296 2.0978 0.0791 0.09502.4 03.0493 2.2981 0.0702 0.08372.5 03.2898 2.5167 0.0624 0.07392.6 03.5533 2.7554 0.0554 0.06532.7 03.8416 3.0161 0.0493 0.05772.8 04.1573 3.3011 0.0438 0.05112.9 04.5028 3.6126 0.0390 0.04533.0 04.8808 3.9534 0.0347 0.04023.1 05.2945 4.3262 0.0310 0.03563.2 05.7472 4.7342 0.0276 0.03163.3 06.2426 5.1810 0.0246 0.02813.4 06.7848 5.6701 0.0220 0.02503.5 07.3782 6.2058 0.0196 0.02223.6 08.0277 6.7927 0.0175 0.01983.7 08.7386 7.4358 0.0156 0.01763.8 09.5169 8.1404 0.0140 0.01573.9 10.3690 8.9128 0.0125 0.01404.0 11.3019 9.7595 0.0112 0.0125

FIG. 9.3.4 Radial flow in an infinite semiconfined aquifer.(Reprinted from A. Verrjuit, 1982, Theory of groundwater flow,2d ed., Macmillan Pub. Co.)

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

H

2rw

Qo

fo

determined with the two boundary conditions as r ® `,f 5 fo and r 2 rw, Qo 5 22prHqr. The solution of thisequation is then

f 5 fo 2 }2Qp

o

T} Ko 1}

l

r}2 9.3(18)

When r approaches 4l, Ko (4) approaches zero whichmeans that at r . 4l, drawdown is practically negligible.Note that when r/l ,, 1, Ko(r/l) < 2ln(r/1.123l), f be-comes

f 5 fo 1 }2Qp

o

T} ln 1}1.1

r23l}2 9.3(19)

This equation is similar to the governing equation for aconfined aquifer, Equation 9.3(13), with the equivalent ra-dius Req equal to 1.123l. Therefore, the equation can berewritten as

f 5 fo 1 }2Qp

o

T} ln 1}

Rr

eq

}2 9.3(20)

Equation 9.3(20) indicates that the drawdown near thewell sw can be expressed as

sw 5 fo 2 fw 5 2}2Qp

o

T} ln 1}1.1

r23l}2 9.3(21)

Basic EquationsThe fundamental equations of groundwater flow can bederived in terms of the discharge vector Qi rather than thespecific discharge qi. For two-dimensional flow, the dis-charge vector has two components Qx and Qy and is de-fined as

Qx 5 Hqx

Qy 5 Hqy 9.3(22)

With the use of Darcy’s law

Qx 5 Hqx 5 H 12K }¶¶f

x}2

Qy 5 Hqy 5 H 12K }¶¶f

y}2 9.3(23)

These equations can be rewritten as

Qx 5 2}¶(K

¶Hx

f)}

Qy 5 2}¶(K

¶Hy

f)} 9.3(24)

With the substitution of a new variable F, defined as

F 5 KHf 1 Cc 9.3(25)

where Cc is an arbitrary constant, Equations 9.3(24) canbe simplified since the derivatives of Cc with respect x andy are zero as

Qx 5 2}¶¶F

x}

Qx 5 2}¶¶F

y} 9.3(26)

The function F is referred to as the discharge potential forhorizontal flow or simply as the potential.

Now the governing equation for horizontal confinedflow, Equation 9.2(13), expressed in terms of the head fis

}¶¶

2

xf2} 1 }

¶¶

2

yf2} 5 0 9.3(27)

and can be written in terms of the potential F as

}¶¶

2

xF2} 1 }

¶¶

2

yf2} 5 0 9.3(28)

or

=2F 5 0 9.3(29)

Solutions to horizontal confined flow can be obtainedwhen F is determined from this Laplace’s equation withproper boundary conditions satisfied.

The following equations give solutions for horizontalconfined flow in terms of F.

(1) One-dimensional flow

F 5 KHf 5 F1 2 }F1 2

LF2

} x 9.3(30)

(2) Radial flow

F 5 KHf 5 }2Qp} ln }

Rr} 1 Fo 9.3(31)

Two-dimensional flow problems expressed by the differ-ential Equation 9.3(29) are discussed in more detail inSection 9.6.

—Y.S. Chae

ReferenceGupta, R.S. 1989. Hydrology and hydraulic systems. Prentice-Hall, Inc.


As defined in Section 9.1, an unconfined aquifer is a wa-ter-bearing layer whose upper boundary is exposed to theopen air (atmospheric pressure), as shown in Figure 9.4.1,known as the phreatic surface. Problems with such aboundary condition are difficult to solve, and the verticalcomponent of flow is often neglected. The Dupuit-Forchheimer assumption to neglect the variation of thepiezometric head with depth (¶f/¶z 5 0) means that thehead along any vertical line is constant (f 5 h). Physically,this assumption is not true, of course, but the slope of thephreatic surface is usually small so that the variation ofthe head horizontally (¶f/¶x, ¶f/¶y) is much greater thanthe vertical value of ¶f/¶z. The basic differential equationfor the flow of groundwater in an unconfined aquifer canbe derived from Darcy’s law and the continuity equation.

Discharge Potential and ContinuityEquationThe discharge vector, as defined in Section 9.3, is the prod-uct of the specific discharge q and the thickness of theaquifer H. For an unconfined aquifer, the aquifer thick-ness h varies, and thus

Qx 5 qxh 5 2Kh }¶¶f

x}

Qy 5 qyh 5 2Kh }¶¶f

y} 9.4(1)

Since h 5 f and K is a constant, Equation 9.4(1) becomes

Qx 5 2}¶

¶x} 1}

12

} Kf22Qy 5 2}

¶¶y} 1}

12

} Kf22 9.4(2)

the discharge potential for unconfined flow introducing as

F 5 }12

} Kf2 1 Cu 9.4(3)

where Cu is an arbitrary constant. Now Equations 9.4(2)can be rewritten as

Qx 5 2}¶¶F

x}

Qy 5 2}¶¶F

y} 9.4(4)

These equations are the same as those derived for confinedflow, Equation 9.3(26).

The continuity equation for unconfined flow, withoutregard for inflow or outflow along the upper boundarydue to precipitation or evaporation, is the same as that forconfined flow as

}¶

¶Qx

x} 1 }

¶¶Qy

y} 5 0 9.4(5)

Basic Differential EquationThe governing equation for unconfined flow is obtainedwhen Equation 9.4(4) is substituted into Equation 9.4(5)as

}¶¶

2

xF2} 1 }

¶¶

2

yF2} 5 0 9.4(6)

The governing equation for both confined and uncon-fined flows is the same, in terms of the discharge poten-tial, and problems can be solved in the same manner math-ematically. The only difference between confined andunconfined flows lies in the expression for F as

F 5 KHf 1 Cc for confined flow 9.4(7)

and

F 5 }12

} Kf2 1 Cu for unconfined flow 9.4(8)

One-Dimensional FlowThe simplest example of unconfined flow is that of an un-confined aquifer between two long parallel bodies of wa-ter, such as rivers or canals, as shown in Figure 9.4.2. Inthis case, f is a function of x only, and the differentialEquation 9.4(6) reduces to


9.4UNCONFINED AQUIFERS

q

f = h

Phreatic or free surface

FIG. 9.4.1 Unconfined aquifer.

}dd

2

xF2} 5 0 9.4(9)

with the general solution

F 5 Ax 1 B 9.4(10)

Constants A and B can be found from the boundary con-ditions

x 5 0, F 5 F1 B 5 F1

x 5 L, F 5 F2 A 5 }F2 2

LF1

}

Substitution of A and B into Equation 9.4(10) yields

F 5 }F2 2

LF1

} x 1 F1 9.4(11)

An expression for the head f can be found by

}12

} Kf2 5 x 1 }12

} Kf21 Cu 5 o

f2 5 }f2

2 2

Lf2

1} x 1 f2

1 9.4(12)

This equation shows that the phreatic surface varies par-abolically with distance (Dupuit’s parabola).

The discharge Qx is now

Qx 5 2}¶¶F

x} 5 }

F1 2

LF2

} 9.4(13)

or

Qx 5 }K(f2

1

22

Lf2

2)} 9.4(14)

Radial FlowIn the case of radial flow in an unconfined aquifer as shownin Figure 9.4.3, the results obtained for confined flow canbe directly applied to unconfined flow because the gov-erning equations are the same in terms of the dischargepotential. From Equation 9.3(31), the governing equationfor radial unconfined flow is

F 5 }2Qp} ln 1}

Rr}2 1 Fo 9.4(15)

}12

} K(f22 2 f2

1)}}

L

The governing equation in terms of the head f is

}12

} Kf2 5 }2Qp} ln 1}

Rr}2 1 }

12

} Kf2o

f2 5 }p

QK} ln 1}

Rr}2 1 f2

o 9.4(16)

or

f 5 !}p

Q¤K}¤ l¤n¤ 1¤}

Rr}¤2¤1¤ f¤2

o¤ 9.4(17)

Note that the expression for the head f for radial uncon-fined flow is different from that for radial confined floweven though the discharge potential for both types of flowis the same. Also, the principle of superposition applies toF but not to f. Superposition of two solutions in Equation9.4(15), therefore, is allowed, but not in Equation 9.4(17).

The introduction of the drawdown s as s 5 fo 2 fmeans f2 5 (fo 2 s)2 5 f2

o 2 2fos 1 s2 5 f2o 2 2fos

(1 2 s/2fo). Hence, Equation 9.4(16) can be written as

s 11 2 }2f

s

o

}2 5 2}2p

QKfo

} ln 1}Rr}2 9.4(18)

If drawdown s is small compared to fo, then s/2fo < 0,and Equation 9.4(18) can be written as

s 5 }2p

QKfo

} ln 1}Rr}2 s ! fo 9.4(19)

This equation is identical to the drawdown equation forconfined flow, Equation 9.3(15). This fact is true only ifthe drawdown is small compared to the head fo. However,Equation 9.4(19) can be accurate enough as a first ap-proximation.

Unconfined Flow with InfiltrationWater can infiltrate into an unconfined aquifer throughthe soil above the phreatic surface as the result of rainfallor artificial infiltration. As shown in Figure 9.4.4, waterpercolates downward into the acquifer at a constant infil-tration rate of N per unit area and per unit time.

The continuity equation for unconfined flow, Equation9.4(5), can be modified to read


��

��

��

��

��

��

z

L

x

f1f

f2

��

��

��

FIG. 9.4.2 One-dimensional flow in an unconfined aquifer.

��

��

��

��2R

fo fo

��

��

Qo

2rw

FIG. 9.4.3 Radial flow in an unconfined aquifer.

}¶

¶Qx

x} 1 }

¶¶Qy

y} 2 N 5 0 9.4(20)

Hence, the differential equation for the potential becomes

}¶¶

2

xF2} 1 }

¶¶

2

yF2} 1 N 5 0 9.4(21)

In terms of f, this equation reads

}¶¶

2

xf2} 1 }

¶¶

2

yf2} 1 }

2KN} 5 0 9.4(22)

One-Dimensional Flow withInfiltrationFor one-dimensional flow shown in Figure 9.4.5, Equation9.4(21) becomes

}dd

2

xF2} 1 N 5 0 9.4(23)

The general solution of this equation is

F 5 2}N2

} x2 1 Ax 1 B 9.4(24)

Use of the boundary condition that

x 5 0 F 5 F1 5 }12

} Kf21

x 5 L F 5 F2 5 }12

} Kf22

gives

F 5 2}N2

} (x2 2 Lx) 2 }F1 2

LF2

} x 1 F1 9.4(25)

and

Qx 5 2}ddF

x} 5 Nx 2 }

N2L} 1 }

F1 2

LF2

} 9.4(26)

The location of the divide xd, where f is maximum, is ob-tained from

}ddF

x} 5 0 5 Nxd 2 }

N2L} 1 }

F1 2

LF2

} 5 0

[xd 5 }F1

N2

LF2

} 1 }L2

} (0 # xd # L) 9.4(27)

Note that xd could be larger than L or could be negative.In those cases, the divide does not exist, and the flow oc-curs in one direction throughout the aquifer.

Radial Flow with InfiltrationFigure 9.4.6 shows radial flow in an unconfined aquiferwith infiltration. If a cylinder has a radius r, the amountof water infiltrating into the cylinder is equal to Qin 5Npr2, and the amount of water flowing out of the cylin-der is equal to 2pr z hqr 5 2prQr. The continuity of flowrequires that 2prQr 5 Npr2, giving

Qr 5 }N2

} r 9.4(28)

which can be written as

Qr 5 2}¶¶F

r} 5 ¾

N

2¾ r 9.4(29)

yielding

F 5 2}N4

} r2 1 C 9.4(30)


��

N

f

x

FIG. 9.4.4 Unconfined flow with rainfall.

��

�� X

f2

f1

N

fxd

FIG. 9.4.5 One-dimensional unconfined flow with rainfall.(Reprinted from A. Verrjuit, 1982, Theory of groundwater flow,Macmillan Pub. Co.)

R

z

r

f0f0

N

r

FIG. 9.4.6 Radial unconfined flow with infiltration. (Reprinted from O.D.L.Strack, 1989, Groundwater mechanics, Vol. 3, Pt. 3, Prentice-Hall, Inc.)

The constant C in this equation can be determined fromthe boundary condition that r 5 R, F 5 Fo. The expres-sion for F then becomes

F 5 2}N4

} (r2 2 R2) 1 Fo 9.4(31)

The location of the divide is obviously at the center of theisland where dF/dr 5 0 and rd 5 0.

Radial Flow from Pumping InfiltrationFigure 9.4.7 shows radial flow in an unconfined aquiferwith infiltration in which water is pumped out of a welllocated at the center of a circular island.

The principle of superposition can be used to solve thisproblem. In the first case, the radial flow is from pump-

ing alone; in the second, the flow is from infiltration. Sincethe differential equations for both cases are linear(Laplace’s equation and Poisson’s equation), the solutionfor each can be superimposed to obtain a solution for thewhole with the sum of both solutions meeting the bound-ary conditions.

The addition of the two solutions, Equations 9.4(15)and 9.4(31), with a new constant C gives

F 5 2}N4

} (r2 2 R2) 1 }2Qp} ln 1}

Rr}2 1 C 9.4(32)

The constant C can be obtained from the boundary con-dition r 5 R, F 5 Fo. Hence,

F 5 2}N4

} (r2 2 R2) 1 }2Qp} ln 1}

Rr}2 1 Fo 9.4(33)

The discharge Qr is now obtained as

Qr 5 2}¶¶F

r} 5 }

N2

} r 2 }2Qpr} 9.4(34)

The divide rd is a circle and occurs when Qr 5 ¶F/¶r 5 0as

}N2

} rd 2 }2p

Qrd

} 5 0

[ rd 5 !}p

Q¤N}¤ (rd # R) 9.4(35)

—Y.S. Chae


FIG. 9.4.7 Radial flow from pumping with infiltration.(Reprinted from A. Verrjuit, 1982, Theory of groundwater flow,2d ed., Macmillan Pub. Co.)

��

��

2R

2rw

NQo

fo

9.5COMBINED CONFINED AND UNCONFINED FLOW

As water flows through a confined aquifer, the flowchanges from confined to unconfined when the piezomet-ric head f becomes less than the aquifer thickness H. Thiscase is shown in Figure 9.5.1. At the interzonal boundary,the head f becomes equal to the thickness H. The conti-nuity of flow requires no change in discharge at the inter-zonal boundary. Hence, the following equation governingthe discharge potential is the same throughout the flow re-gion:

}¶¶

2

xF2} 1 }

¶¶

2

yF2} 5 0 9.5(1)

where

F 5 KHf 1 Cc for f $ H

F 5 }12

} Kf2 1 Cu for f , H

At the interzonal boundary, F yields the same value, giving

KH2 1 Cc 5 }12

} KH2 1 Cu

Cc 5 Cu 2 }12

} KH2 9.5(2)

FIG. 9.5.1 Combined confined and unconfined flow.

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

H fConfinedFlow

UnconfinedFlow

If one of the two constants Cu is set to zero, then

Cc 5 2}12

} KH2, Cu 5 0 9.5(3)

The potential F can be expressed as

F 5 KHf 2 }12

} KH2 (f $ H)

F 5 }12

} Kf2 (f , H) 9.5(4)

One-Dimensional FlowFigure 9.5.2 shows combined confined and unconfinedflow in an aquifer of thickness H and length L. The aquiferis confined at x 5 0 and unconfined at x 5 L.

The expression for the potential F is the same through-out the flow region as

F 5 2(F1 2 F2) }Lx

} 1 F1 9.5(5)

However, the expression for F in terms of f is differentfor each zone as given in Equation 9.5(4). The expressionfor the discharge Q is

Qx 5 }F1 2

LF2

} 5 9.5(6)

KHf1 2 }12

} KH2 2 }12

} Kf22

}}}L

The location of the interzonal boundary xb is obtainedfrom Equation 9.5(5) when Equation 9.5(4) is substitutedfor F1 and F2, and F 5 1/2 KH2 as

xb 5 z L 9.5(7)

Note that xb is independent of the hydraulic conductivityK. Also note that when f1 5 H, xb 5 0 (entirely uncon-fined flow) and when f2 5 H, xb 5 1 (entirely confinedflow).

Radial FlowIf the drawdown near the well caused by pumping dipsbelow the aquifer thickness H, then unconfined flow oc-curs in that region as shown in Figure 9.5.3. The expres-sion for the potential F is the same for the entire flow re-gion as

F 5 }2Qp} ln 1}

Rr}2 1 Fo 9.5(8)

In this equation, Fo is the potential at r 5 R when theflow is confined. Hence

Fo 5 KHfo 2 }12

} KH2 (fo . H) 9.5(9)

The potential at well Fw for unconfined flow is

Fw 5 }12

} Kf2w (fw , H) 9.5(10)

Equation 9.5(8) can now be rewritten as

}12

} Kf2w 5 }

2Qp} ln 1}

rRw}2 1 KHfo 2 }

12

} KH2 9.5(11)

and solving for Q gives

Q 5 9.5(12)

The distance rb to the interzonal boundary, which is a cir-cle, can be obtained from Equation 9.5(8) with F 5 1/2KH2 as

}12

} KH2 5 }2Qp} ln 1}

rRb}2 1 Fo

[ rb 5 R z e2pKH(H2fo)/Q 9.5(13)

—Y.S. Chae

2p 1}12

} Kf2w 2 KHfo 1 }

12

} KH22}}}}

ln 1}rRw}2

Hf1 2 H2

}}}Hf1 2 }

12

} H2 2 }12

} f22


��

��

��

��

��

��

��

��

L

Hf1

f2

Z

X

��

@@@@@

��

ÀÀÀÀÀ

��

@@@@@

��

ÀÀÀÀÀ

��

@@@@@

��

ÀÀÀÀÀ

��

��@@��ÀÀ��@@��ÀÀ��@@��ÀÀ��

H ffw

Qo

z

r r

FIG. 9.5.3 Radial combined flow. (Reprinted from O.D.L.Strack, 1989, Groundwater mechanics, Vol. 3, Pt. 3, PrenticeHall, Inc.)

FIG. 9.5.2 One-dimensional combined flow.

This section describes methods for handling two-dimen-sional groundwater flow problems including superposi-tion, the method of images, and the potential and flowfunction.

SuperpositionThe differential equation for two-dimensional steady flowin a homogeneous aquifer is

}¶¶

2

xF2} 1 }

¶¶

2

yF2} 5 0 9.6(1)

Because this equation is a linear and homogeneous dif-ferential equation, the principle of superposition applies.The principle states that if two different functions F1 andF2 are solutions of Laplace’s equation, then the function

F(x,y) 5 c1F1(x,y) 1 c2F(x,y) 9.6(2)

is also a solution.Superposition of solutions is valuable in several ground-

water problems. For example, the case of groundwaterflow due to simultaneous pumping from several wells canbe solved by the superposition of the elementary solutionfor a single well.

A TWO-WELL SYSTEM

Consider the case of two wells in an infinite aquifer asshown in Figure 9.6.1, in which water is discharged (pos-itive Q) from well 1 and is recharged (negative Q) intowell 2. This case is referred to as a sink-and-source prob-lem.

The potential F at a point which is located at a dis-tance r1 from well 1 and r2 from well 2 can be expressedwhen the potential F1 is superimposed with respect to well1 and F2 is superimposed with respect to well 2 as

F 5 F1 1 F2 5 }2Q

p1

} ,n r1 2 }2Q

p2

} ,n r2 1 C 9.6(3)

The constant C 5 Fo @ r1 5 r2 5 RIf Q1 5 Q2 5 Q in a special case, then

F 5 }2Qp} ,n 1}

rr1

2

}2 1 Fo 9.6(4)

or

f 5 }2QpT} ,n 1}

rr1

2

}2 1 fo for a confined aquifer 9.6(5)

f2 5 }p

QK} ,n 1}

rr1

2

}2 1 f2o for an unconfined aquifer 9.6(6)

Figure 9.6.2 shows the flow net for a two-well sink-and-source system. Equation 9.6(4) shows that along they axis where r1 5 r2 5 ro, F 5 constant. This statementmeans that the y axis is an equipotential line along whichno flow occurs, and the drawdown is zero (f 5 fo). Thisresult occurs because the system is in symmetry about they axis and the problem is linear. Note that the distance Rdoes not appear in Equation 9.6(4). This omission is be-cause the discharge from the sink is equal to the rechargeinto the source, indicating that the system is in hydraulicequilibrium requiring no external supply of water.

Another example of using the principle of superposi-tion is the case of two sinks of equal discharge Q. Equation9.6(3) now reads

F 5 F1 1 F2 5 }2Qp} ,n (r1r2) 1 C 9.6(7)


Hydraulics of Wells

9.6TWO-DIMENSIONAL PROBLEMS

y

P (x,y)

Q1x

12

2Q2

r2

r1

FIG. 9.6.1 Discharge and recharge wells.

Use of the boundary condition r 5 R, F 5 Fo yields

F 5 }2Qp} ,n1}

rR1r

22

}2 1 F0 9.6(8)

Figure 9.6.3 shows the flow net for a two-well sink-and-sink system. The y axis plays the role of an imperviousboundary along which no water flows across. This resultoccurs because the flow at points on the y axis is directedalong the axis due to the equal pull of flow from the twowells located equidistance from the points.

A MULTIPLE-WELL SYSTEM

The principle of superposition previously discussed for twowells can be applied to a system of multiple wells, n wellsin number from i 5 1 to n. The solution for such a sys-tem can be written with the use of superposition as

F 5 }21p} 3^

n

i5iQi ,n 1}

Rri}24 1 Fo 9.6(9)

or

f 5 }2p

1T

} 3^n

i51Qi ,n 1}

Rri}24 1 fo for a confined aquifer

9.6(10)

and

f2 5 }p

1K} 3^

n

i51Qi ,n 1}

Rri}24 1 f2

o for an unconfined aquifer

9.6(11)

The drawdown at the jth well is then

fw 5 }2p

1T

} 3Qj ,n 1}rRw}2 1 ^

n21

i51Qi ,n 1}

rRi,j}24 1 fo

for a confined aquifer 9.6(12)

and

f2w 5 }

p

1K} 3Qj ,n 1}

rRw}2 1 ^

n21

i51Qi ,n 1}

rRi,j}24 1 f2

o

for an unconfined aquifer 9.6(13)

where ri,j is the distance between the jth well and ith wells.The quantities inside the brackets [ ] in these equationsare called the drawdown factors, Fp at a point and Fw ata well, respectively. These equations can be rewritten as

F 5 Fo 1 }21p} Fp at a point 9.6(14)


FIG. 9.6.2 Source and sink in unconfined flow. (Reprinted from R.S. Gupta, 1989,Hydrology and hydraulic systems, Prentice-Hall, Inc.)

Dischargingreal wellE

quip

oten

tial li

ne

Flow

line

Rechargingimage well

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

Imaginaryrecharge well

Stream boundaryzero drawdown

PumpedwellQ

Q

Due toimage well

Cone of impression

Due toreal well

Static water table

< <

[0Resu

ltant cone

Cone of depression

Fw 5 Fo 1 }21p} Fw at a well 9.6(15)

where

Fp 5 ^n

i51Qi ,n 1}

rRw}2 9.6(16)

Fw 5 Qj ,n 1}rRw}2 1 ^

n21

i51Qi ,n 1}

rRi,j}2 9.6(17)

The following examples give the drawdown factors ofwells in special arrays:

a. Circular array, n wells in equal spacing (Figure 9.6.4a)

Fp 5 nQ ,n }Rr

} 9.6(18)

Fw 5 Q ,n }nrw

Rrn

n21

} 9.6(19)

b. Rectangular array (Figure 9.6.4b)• Approximate method:

Equivalent radius re 5 4 awb/pThen use Equation 9.6(11)

• Exact method:Use Equation 9.6(9), 9.6(12) or 9.6(13)

c. Two parallel lines of equally spaced wells (Figure9.6.4c)

Fc 5 4Q ^i5n/4

i51,n 9.6(20)

Fw 5 2Q ^i5n/2

i51,n 9.6(21)

Method of ImagesA special application of superposition is the method of im-ages. This method can be used to solve problems involv-ing the flow in aquifers of relatively simple geometricalform such as an infinite strip, a half plane, or a quarterplane. The following problems are specific examples.

WELL NEAR A STRAIGHT RIVER

To solve the problem of a well near a long body of water(river, canal, or lake) shown in Figure 9.6.5, replace thehalf-plane aquifer by an imaginary infinite aquifer with animaginary well placed at the mirror image position fromthe real well. This case now represents the sink and sourceproblem discussed previously, and Equation 9.6(4) satis-

R}}}As z S2w(2wiw2w 3w)w2 1 B2

R}}}As z S2w(2wiw2w 1w)2w 1w Bw2w


FIG. 9.6.3 Sink and sink in unconfined flow. (Reprinted from R.S. Gupta, 1989,Hydrology and hydraulic systems, Prentice-Hall, Inc.)

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

QDue to

image well

< <

PumpedwellQ

[0A

xis

of im

perm

eabl

e bo

unda

ry

Dischargingreal well

Dischargingimage well

Potential line

Streamline

Due toreal well

Zeroflow line

Static water table

Image well cone

Real well coneResultant cone

fies all conditions associated with the case. Accordingly,the solution is given by

F 5 }2Qp} ln 1}

rr1

2

}2 1 Fo 9.6(22)

If n number of wells are on the half plane, use Equation9.6(7) for solution as follows:

F 5 Fo 1 }21p} F9p at a point 9.6(23)

Fw 5 Fo 1 }21p} F9w at a well 9.6(24)

where

F9p 5 ^n

i51Qi ,n 1}

rr9i

i

}2 9.6(25)

F9w 5 Qi ,n 1}rrw

9j}2 1 ^

n21

i51Qi ,n 1}

rr9i

i

,

,

j

j

}2 9.6(26)

and

r9i 5 distance between point and imaginary ith well.r9i,j 5 distance between jth well and imaginary ith well.

WELL NEAR A STRAIGHT IMPERVIOUSBOUNDARY

The problem of a well near a long straight imperviousboundary (e.g. a mountain ridge or fault) is solved in asimilar manner as that of a well near a straight river. Inthis case, the type of image well is a sink rather than asource as shown in Figure 9.6.6.


FIG. 9.6.4 Wells in special arrays. (Reprinted from G.A. Leonards, ed., 1962, Foundationengineering, McGraw-Hill, Inc.)

RzR

b S

a

S

i 5 1 2 3 4 5 6

S

B

cL

cL

(c)(c)(b)(a)Circular array Rectangular array Two parallel lines

of equal spacing

R

FIG. 9.6.5 Well near a straight river. (Reprinted from G.A.Leonards, ed., 1962, Foundation engineering, McGraw-Hill,Inc.)

��@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

2QQ

rw

f

fofw

ø ø

Line source

Image wellReal well

`

`

ø ø

r' r

P

sf

��@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

QQ

rw

ffo

fw

ø ø

Imperviousboundary

Image wellReal well

`

`

ø ø

r' r

P

sf

FIG. 9.6.6 Well near a straight impervious boundary.

The solution for this case, therefore, is the same as the caseof a sink-and-sink problem given by Equation 9.6(8),which is

F 5 }2Qp} ,n 1}

r1

Rr2}2 1 Fo 9.6(27)

WELL IN A QUARTER PLANE

Figure 9.6.7 shows the case of a well operating in anaquifer bounded by a straight river and an imperviousboundary. To solve this problem, place a series of imagi-nary wells (wells numbered 2, 3, and 4), and use super-position. Figure 9.6.7 indicates that wells 2 and 3 aresources, and well 4 is a sink. Hence,

F 5 }2Qp} 3,n 1}

rR1}2 2 ,n 1}

rR2}2 2 ,n 1}

rR3}2 1 ,n 1}

rR4}24 1 Fo

9.6(28)

Potential and Flow FunctionsIn the Basic Equations section, the fundamental equationof groundwater flow expressed in terms of discharge po-tential F is:

}¶¶

2

xF

2

} 1 }¶¶

2

yF2} 5 0 9.6(29)

The potential F(x,y) is a single-value function everywherein the x, y plane. Therefore, lines of constant F1, F2, . . .,called equipotential lines, can be drawn in the x, y planeas shown in Figure 9.6.8. When the lines are drawn witha constant interval between the values of the two succes-sive lines (DF 5 F1 2 F2 5 F2 2 F3 5 . . .), then anequal and constant amount of potential drop is betweenany two of the equipotential lines.

At any arbitrary point on the equipotential line, flowoccurs only in the direction perpendicular to the line (n di-rection), and no flow occurs in the tangential direction (mdirection) as

qn 5 }¶¶F_n} 5 2q; qm 5 }

¶¶F_m} 5 0 9.6(30)

Accordingly, lines can be drawn perpendicular to theequipotential lines as shown in Figure 9.6.8. These linesare called flow or stream lines.

At this point a second function, called flow or streamfunction C, is introduced. Since the specific discharge vec-tor must satisfy the equation of continuity, the function Cis defined by

qx 5 2}¶¶C

y} , qy 5 }

¶¶c

x} 9.6(31)

It now follows that

}¶¶

2

xC2} 1 }

¶¶

2

yC2} 5 0 9.6(32)

or

=2c 5 0 9.6(33)

which shows that C is, like the potential F, a harmonicfunction and should satisfy Laplace’s equation.

The directional function C with respect to m and n cannow be easily written as follows because no flow compo-nent is in m direction:

}¶¶_mc} 5 2q, }

¶¶

c_n} 5 0 9.6(34)

The lines of constant F and C form a set of orthogo-nal curves called a flow net. Also,

q 5 2}¶¶F_n} 5 2}

¶¶_mc} 9.6(35)


P(x,y)

2

3 4

1

r2

r3

r4

<2

<2

r1

`

`

<1<1

FIG. 9.6.7 Well in a quarter plane.

FIG. 9.6.8 Potential and flow lines.

y

m n

x

C3

C2

C1

a

F1

F2

F3

meaning that if lines are drawn with constant F and C atintervals DF and Dc, then

}D

D

F_n} 5 }

D

Dc_m} 9.6(36)

where Dn is the distance between two potential lines, andDm is the distance between two flow lines. Thus, theequipotential lines and flow lines are not only orthogonal,but they form elementary curvelinear squares. This prop-erty is the basis of using a flow net as an approximategraphic method to solve groundwater problems. With a

flow net drawn, for example, the rate of flow (Q) can beobtained by

Q 5 Kfo }nn

f

f} 9.6(37)

where:

nf 5 number of flow zonesnf 5 number of equipotential zonesfo 5 total head loss in flow system

—Y.S. Chae


9.7NONSTEADY (TRANSIENT) FLOW

Nonsteady or transient flow in aquifers occurs when thepressure and head in the aquifer change gradually untilsteady-state conditions are reached. During the course oftransient flow, water can be either stored in or releasedfrom the soil. Storage has two possibilities. First, water cansimply fill the pore space in soil without changing the soilvolume. This storage is called phreatic storage, and usu-ally occurs in unconfined aquifers as the groundwater tablemoves up or down. In the other storage, water is storedin the pore space increased by deformation of the soil andinvolves a volume change. This storage is called elastic stor-age and occurs in all types of aquifers. However, in con-fined aquifers, it is the only form of storage.

Transient Confined Flow (ElasticStorage)In a completely saturated confined aquifer, water can bestored or released if the change in aquifer pressure resultsin volumetric deformation of the soil. The problem is com-plex because the constitutive equations for soil are highlynonlinear even for dry soil, and coupling them withgroundwater flow increases the complexity.

The basic equation for the phenomenon is the storageequation (Strack 1989), as

=2F 5 H 3}¶

¶to

} 1 nb }¶¶pt}4 9.7(1)

where o 5 volume strain, and b 5 compressibility of wa-ter. From soil mechanics

}¶

¶to

} 5 mv }¶¶pt} 9.7(2)

where mv 5 modulus of volume change. Equation 9.7(1)can then be written as

=2F 5 H(mv 1 nb) }¶¶pt} 9.7(3)

When the variation of K, H, and r with time are neglected,then

}¶¶F

t} 5 KH }

¶¶t} 3}

r

pg} 1 z4 5 }

Kr

Hg} }

¶¶pt} 9.7(4)

so that Equation 9.7(3) can be written as

=2F 5 }SK

s} }

¶¶F

t} 9.7(5)

or

}¶¶F

t} 5 }

SK

s

} =2F 9.7(6)

where Ss [(1/m)] is the coefficient of specific storage

Ss 5 rg(mv 1 nb) 9.7(7)

If the compressibility of water b is ignored, then Ss 5 rgmv.Some typical values of mv are given in Table 9.7.1.

Equation 9.7(5) can also be written in terms of f as

=2f 5 }ST

e} }

¶¶f

t} 9.7(8)

where Se 5 coefficient of elastic storage 5 Ss z H.

Transient Unconfined Flow (PhreaticStorage)The vertical movement of a phreatic surface results in wa-ter being stored in soil pores without causing the soil to

deform. Phreatic storage is, therefore, several orders ofmagnitude greater than elastic storage, which can be ig-nored.

The basic differential equation for the transient uncon-fined flow (Strack 1989), such as shown in Figure 9.7.1,can be given as

=2F 5 Sp }¶¶f

t} 9.7(9)

where Sp 5 coefficient of phreatic storage.Equation 9.7(9) can be linearized in terms of the po-

tential F as

=2F 5 }SK

s} }

¶¶F

t} 9.7(10)

or

}¶¶F

t} 5 }

SK

s

} =2F 9.7(11)

This equation is the same as that for transient confinedflow. However, Ss is related to Sp as Ss 5 Sp/

_f, where

_f is

the average piezometric head in the aquifer.

Transient Radial Flow (TheisSolution)The governing equation for the transient radial flow (flowtoward a well in an aquifer of infinite extent) is obtainedwhen Equation 9.7(10) is rewritten in terms of radial co-ordinate r as

}¶¶

2

rF2} 1 }

1r} }

¶¶F

r} 5 }

SK

s} }

¶¶F

t} 9.7(12)

The solution to this equation is commonly given as

F 5 2}4Qp} Ei(u) 1 Fo 9.7(13)

known as the Theis solution. Ei is the exponential inte-gral, and u is a dimensionless variable defined by

u 5 }4Ss

Kr2

t} 9.7(14)

or

u 5 }S4

e

Tr2

t} for confined flow (T 5 KH) 9.7(15)

or

u 5 }S4

p

Tr2

t} for unconfined flow (T 5 K

_f) 9.7(16)

The exponential integral Ei(u) is referred to as the wellfunction W(u). Ei(u) can be approximated by

Ei(u) 5 320.577216 2 ,n u 1 u 2 }2u.2

2

!} 1 }

3u.3

3

!} 2 ••4 9.7(17)

Using the well function W(u), the Theis solution can bewritten as

F 5 2}4Qp} W(u) 1 Fo 9.7(18)

or in terms of the head f as

f 5 2}4QpT} W(u) 1 fo 9.7(19)

The drawdown s is obtained by

s 5 }4QpT} W(u) 9.7(20)

Values of W(u) for different values of u are shown in Table9.7.2. The drawdown s at a given distance r from the wellat given time t can be calculated from Equation 9.7(20)and Table 9.7.2.

Figure 9.7.2, accompanied by Table 9.7.3, shows anexample of drawdown versus a time curve for a transientradial flow in a confined aquifer with T 5 1000 m2/d andS 5 0.0001 for a pumping rate of Q 5 1000 m3/d. Thefigure shows that even in a transient flow, the rate of draw-down (Ds) achieves a steady state after a short period ofpumping, two days in this example.

If u is small (e.g., less than 0.01), only the first twoterms of the brackets in Equation 9.7(17) are significant.Equation 9.7(19) can be simplified to


TABLE 9.7.1 TYPICAL VALUES OFCOMPRESSIBILITY (mv)

Compressibility,(m2/N or Pa21)

Clay 1026–1028

Sand 1027–1029

Gravel 1028–10210

Jointed rock 1028–10210

Sound rock 1029–10211

Water (b) 4.4 3 10210

Source: R.A. Freeze and J.A. Cherry, 1979, Groundwater (Prentice-Hall,Inc.).

��

t + Dt

t

f

t. Dt–

FIG. 9.7.1 Storage change due to unconfined flow.

TABLE 9.7.2 VALUES OF W(U) FOR DIFFERENT VALUES OF U

1 34.0 31.7 29.4 27.1 24.8 22.4 20.1 17.8 15.5 13.2 10.9 8.63 6.33 4.04 1.82 0.2191.2 33.8 31.5 29.2 26.9 24.6 22.3 20.0 17.7 15.4 13.1 10.8 8.45 6.15 3.86 1.66 0.1581.5 33.6 31.3 29.0 26.6 24.3 22.0 19.7 17.4 15.1 12.8 10.5 8.23 5.93 3.64 1.46 0.1002 33.3 31.0 28.7 26.4 24.1 21.8 19.5 17.2 14.8 12.5 10.2 7.94 5.64 3.35 1.22 0.04892.2 33.2 30.9 28.6 26.3 24.0 21.7 19.4 17.1 14.8 12.4 10.1 7.84 5.54 3.26 1.15 0.03722.5 33.0 30.7 28.4 26.1 23.8 21.5 19.2 16.9 14.6 12.3 10.0 7.72 5.42 3.14 1.04 0.02493 32.9 30.6 28.3 26.0 23.7 21.3 19.0 16.7 14.4 12.1 09.84 7.53 5.23 2.96 0.906 0.01303.2 32.8 30.5 28.2 25.9 23.6 21.3 19.0 16.7 14.4 12.1 09.77 7.47 5.17 2.90 0.858 0.01013.5 32.7 30.4 28.1 25.8 23.5 21.2 18.9 16.6 14.3 12.0 09.68 7.38 5.08 2.81 0.794 0.006974 32.6 30.3 28.0 25.7 23.4 21.1 18.8 16.5 14.2 11.9 09.55 7.25 4.95 2.68 0.702 0.003784.2 32.5 30.2 27.9 25.6 23.3 21.0 18.7 16.4 14.1 11.8 09.50 7.20 4.90 2.63 0.670 0.003004.5 32.5 30.2 27.9 25.5 23.2 20.9 18.6 16.3 14.0 11.7 09.43 7.13 4.83 2.57 0.625 0.002075 32.4 30.0 27.7 25.4 23.1 20.8 18.5 16.2 13.9 11.6 09.33 7.02 4.73 2.47 0.560 0.001155.2 32.3 30.0 27.7 25.4 23.1 20.8 18.5 16.2 13.9 11.6 09.29 6.98 4.69 2.43 0.536 0.0009095.5 32.3 30.0 27.7 25.3 23.0 20.7 18.4 16.1 13.8 11.5 09.23 6.93 4.63 2.38 0.503 0.0006416 32.2 29.9 27.6 25.3 23.0 20.7 18.4 16.1 13.7 11.4 09.14 6.84 4.54 2.30 0.454 0.0003606.2 32.1 29.8 27.5 25.2 22.9 20.6 18.3 16.0 13.7 11.4 09.11 6.81 4.51 2.26 0.437 0.0002866.5 32.1 29.8 27.5 25.2 22.9 20.6 18.3 16.0 13.7 11.4 09.06 6.76 4.47 2.22 0.411 0.0002037 32.0 29.7 27.4 25.1 22.8 20.5 18.2 15.9 13.6 11.3 08.99 6.69 4.39 2.15 0.374 0.0001157.2 32.0 29.7 27.4 25.1 22.8 20.5 18.2 15.9 13.6 11.3 08.96 6.66 4.36 2.12 0.360 0.00009227.5 32.0 29.6 27.3 25.0 22.7 20.4 18.1 15.8 13.5 11.2 08.92 6.62 4.32 2.09 0.340 0.00006588 31.9 29.6 27.3 25.0 22.7 20.4 18.1 15.8 13.5 11.2 08.86 6.55 4.26 2.03 0.311 0.00003778.2 31.9 29.6 27.3 24.9 22.6 20.3 18.0 15.7 13.4 11.1 08.83 6.53 4.23 2.00 0.300 0.00003018.5 31.8 29.5 27.2 24.9 22.6 20.3 18.0 15.7 13.4 11.1 08.80 6.49 4.20 1.97 0.284 0.00002169 31.8 29.5 27.2 24.9 22.6 20.3 17.9 15.6 13.3 11.0 08.74 6.44 4.14 1.92 0.260 0.00001249.2 31.7 29.4 27.1 24.8 22.5 20.2 17.9 15.6 13.3 11.0 08.72 6.41 4.12 1.90 0.251 0.000009999.5 31.7 29.4 27.1 24.8 22.5 20.2 17.9 15.6 13.3 11.0 08.68 6.38 4.09 1.87 0.239 0.00000718

10 31.7 29.4 27.1 24.8 22.4 20.1 17.8 15.5 13.2 10.9 08.63 6.33 4.04 1.82 0.219


N

u

N

f 5 2}4QpT} ,n 1}2.2

Sr52

Tt}2 1 fo 9.7(21)

then

s 5 }4QpT} ,n 1}2.2

Sr52

Tt}2 9.7(22)

Equation 9.7(21) can be rewritten as

f 5 }2QpT} ,n 3 4 1 fo 9.7(23)

f 5 }2QpT} ,n 1}R

r

eq.

}2 1 fo 9.7(24)

where

r}}

12.25 }TSt

}21/2

Req. 5 12.25 }TSt

}21/2

Note that Equation 9.7(24) is similar in expression to thesteady-state flow. Equation 9.7(22) allows direct calcula-tion of drawdown in terms of distance r and time t forgiven aquifer characteristics T and S at a known pumpingrate Q.

The exact solution of Equation 9.7(13) is difficult forunconfined aquifers because

_T 5 K

_f is not constant but

varies with distance r and time t. The average head _f can

be estimated and used in the Theis solution for small draw-downs. For large drawdowns, however, the use of

_f for

the Theis solution is not valid.For large drawdowns, Boulton (1954) presents a solu-

tion which is valid if the water depth in the well exceeds0.5 fo. Boulton’s equation is:

s 5 }2p

QKfo

} (1 1 Ck)V(t9, r9) 9.7(25)

where V is Boulton’s well function, and Ck is a correctionfactor. The t9 and r9 are defined as

r9 5 }f

1

o

} z r 9.7(26)

t9 5 }Sp

Kfo

} t 9.7(27)

The values of V(r9,t9) and Ck are given in Table 9.7.4 andTable 9.7.5 respectively.

The head at the well fw can be calculated from theequation (Bouwer 1988) as

Q2w 5 f2

o 2 }p

Qk} ,n 11.5 !}

SK¤pr¤t

w

}¤2 9.7(28)

which is valid if t9 5 (Kt/foS) . 5. If t9 is smaller than 5,fw is calculated as

fw 5 fo 2 }2p

QKfo

} 1m 1 ,n }f

rw

o}2 9.7(29)


7 8 9 106543210

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

Time t in Days

r = 200m

r = 100m

TABLE 9.7.3 CALCULATION OF S IN RELATION TO T

r 5 100 m r 5 200 m

t, Days u W(u) s, m u W(u) s, m

0.001 0.25 01.044 0.083 1 0.219 0.0170.005 0.05 02.468 0.196 0.2 1.223 0.0970.01 0.025 03.136 0.249 0.1 1.823 0.1450.05 0.005 04.726 0.376 0.02 3.355 0.2670.1 0.002 5 05.417 0.431 0.01 4.038 0.3220.5 0.000 5 07.024 0.559 0.002 5.639 0.4491 0.000 25 07.717 0.614 0.001 6.331 0.5045 0.000 05 09.326 0.742 0.000 2 7.940 0.632

10 0.000 025 10.019 0.797 0.000 1 8.633 0.687


FIG. 9.7.2 Drawdown versus time due to pumping from a well.

TABLE 9.7.4 VALUES OF THE FUNCTION V(T9,R9) FOR DIFFERENT VALUES OF T9 AND R9

t9 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

0.01 2.99 2.30 1.90 1.64 1.42 1.28 1.15 1.04 0.95 0.875 0.474 0.322 0.240 0.192 0.158 0.135 0.118 0.1040.02 3.68 2.97 2.58 2.30 2.09 1.92 1.76 1.64 1.52 1.42 0.860 0.610 0.468 0.378 0.316 0.270 0.236 0.2100.03 4.08 3.40 3.00 2.70 2.46 2.28 2.13 2.00 1.88 1.79 1.18 0.860 0.675 0.555 0.465 0.400 0.350 0.3100.04 4.35 3.68 3.26 2.98 2.75 2.58 2.42 2.29 2.17 2.06 1.42 1.07 0.850 0.710 0.600 0.525 0.460 0.4100.05 4.58 3.90 3.49 3.20 2.96 2.79 2.64 2.50 2.38 2.28 1.60 1.24 1.010 0.850 0.725 0.630 0.560 0.5000.06 4.76 4.06 3.65 3.36 3.15 2.96 2.80 2.68 2.56 2.45 1.78 1.40 1.15 0.970 0.840 0.735 0.650 0.5850.07 4.92 4.20 3.80 3.51 3.30 3.12 2.96 2.82 2.70 2.60 1.91 1.54 1.28 1.09 0.950 0.835 0.740 0.6700.08 5.08 4.34 3.94 3.65 3.42 3.24 3.09 2.95 2.84 2.72 2.04 1.65 1.39 1.20 1.04 0.925 0.825 0.7500.09 5.18 4.47 4.05 3.75 3.54 3.35 3.20 3.05 2.95 2.84 2.14 1.75 1.50 1.29 1.14 1.02 0.910 0.8250.1 5.24 4.54 4.14 3.85 3.63 3.45 3.30 3.15 3.04 2.94 2.25 1.85 1.58 1.38 1.22 1.09 0.985 0.8900.2 5.85 5.15 4.78 4.50 4.28 4.10 3.93 3.80 3.66 3.56 2.87 2.46 2.20 1.98 1.80 1.65 1.52 1.420.3 6.24 5.50 5.12 4.85 4.61 4.43 4.28 4.14 4.01 3.90 3.24 2.84 2.54 2.32 2.14 1.98 1.85 1.740.4 6.45 5.75 5.35 5.08 4.85 4.67 4.50 4.38 4.26 4.15 3.46 3.05 2.76 2.54 2.36 2.20 2.07 1.960.5 6.65 6.00 5.58 5.25 5.00 4.85 4.70 4.55 4.45 4.30 3.65 3.24 2.95 2.72 2.52 2.38 2.24 2.140.6 6.75 6.10 5.65 5.40 5.15 4.98 4.82 4.68 4.56 4.45 3.76 3.37 3.09 2.85 2.67 2.50 2.38 2.260.7 6.88 6.20 5.80 5.50 5.25 5.08 4.92 4.80 4.68 4.55 3.90 3.50 3.20 2.99 2.80 2.64 2.50 2.380.8 7.00 6.25 5.85 5.60 5.35 5.20 5.00 4.90 4.80 4.65 3.96 3.55 3.26 3.05 2.86 2.71 2.58 2.460.9 7.10 6.35 6.00 5.70 5.50 5.30 5.12 5.00 4.90 4.75 4.05 3.65 3.36 3.15 2.96 2.80 2.66 2.551 7.14 6.45 6.05 5.75 5.55 5.35 5.20 5.05 4.95 4.83 4.10 3.74 3.45 3.22 3.04 2.90 2.75 2.642 7.60 6.88 6.45 6.15 5.92 5.75 5.60 5.50 5.35 5.25 4.59 4.18 3.90 3.68 3.50 3.34 3.20 3.093 7.85 7.15 6.70 6.45 6.20 6.00 5.85 5.75 5.60 5.50 4.82 4.42 4.12 3.90 3.72 3.57 3.45 3.314 8.00 7.28 6.85 6.58 6.35 6.15 6.00 5.90 5.75 5.70 4.95 4.55 4.26 4.04 3.86 3.70 3.59 3.465 8.15 7.35 7.00 6.65 6.50 6.25 6.10 6.00 5.85 5.80 5.05 4.68 4.40 4.19 4.00 3.85 3.71 3.606 8.20 7.50 7.10 6.75 6.55 6.35 6.20 6.10 5.95 5.85 5.20 4.78 4.50 4.26 4.09 3.92 3.80 3.697 8.25 7.55 7.15 6.85 6.62 6.40 6.30 6.20 6.05 5.95 5.25 4.85 4.58 4.35 4.18 4.00 3.90 3.788 8.30 7.60 7.20 6.90 6.70 6.50 6.35 6.25 6.10 6.05 5.30 4.92 4.65 4.40 4.25 4.10 3.95 3.829 8.32 7.65 7.25 7.00 6.75 6.55 6.40 6.30 6.15 6.10 5.35 5.00 4.70 4.49 4.30 4.15 4.00 3.90

10 8.35 7.75 7.35 7.05 6.80 6.60 6.45 6.35 6.20 6.14 5.40 5.02 4.80 4.52 4.35 4.19 4.05 3.92

Continued on next page

©1999 C

RC

Press LL

C

©1999 C

RC

Press LL

C

TABLE 9.7.4 Continued

t9 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2 3 4 5

0.01 0.093 0.0430 0.0264 0.0180 0.0132 0.0100 0.0078 0.0062 0.0049 0.0040 0.00057 0.000150.02 0.187 0.0865 0.0530 0.0365 0.0268 0.0205 0.0160 0.0125 0.0100 0.0081 0.00118 0.000200.03 0.278 0.130 0.0800 0.0550 0.0405 0.0310 0.0240 0.0190 0.0150 0.0122 0.00184 0.000320.04 0.368 0.174 0.107 0.0735 0.0540 0.0415 0.0322 0.0255 0.0202 0.0165 0.00244 0.000430.05 0.450 0.215 0.133 0.0920 0.0675 0.0520 0.0400 0.0320 0.0255 0.0206 0.00305 0.000550.06 0.530 0.257 0.160 0.110 0.0810 0.0610 0.0478 0.0380 0.0305 0.0250 0.00365 0.000650.07 0.610 0.298 0.186 0.130 0.0950 0.0725 0.0565 0.0450 0.0360 0.0292 0.00430 0.000780.08 0.680 0.340 0.214 0.148 0.108 0.0825 0.0645 0.0510 0.0412 0.0336 0.00500 0.000900.09 0.750 0.378 0.236 0.164 0.122 0.0930 0.0730 0.0585 0.0470 0.0380 0.00570 0.001050.1 0.815 0.415 0.260 0.180 0.134 0.103 0.0805 0.0640 0.0515 0.0420 0.00635 0.001180.2 1.32 0.750 0.500 0.359 0.268 0.208 0.165 0.132 0.107 0.0880 0.0145 0.002780.3 1.64 1.02 0.700 0.515 0.392 0.308 0.246 0.200 0.164 0.135 0.0238 0.004900.4 1.86 1.22 0.870 0.650 0.510 0.405 0.328 0.268 0.220 0.182 0.0350 0.00750 0.00160 0.000380.5 2.03 1.37 1.00 0.770 0.610 0.490 0.400 0.330 0.275 0.230 0.0450 0.0104 0.00240 0.000560.6 2.16 1.49 1.12 0.875 0.700 0.570 0.468 0.390 0.325 0.276 0.0580 0.0138 0.00320 0.000800.7 2.28 1.60 1.22 0.965 0.775 0.640 0.525 0.445 0.375 0.320 0.0715 0.0175 0.00425 0.001080.8 2.36 1.69 1.30 1.04 0.850 0.715 0.600 0.500 0.425 0.364 0.0840 0.0212 0.00525 0.001400.9 2.45 1.75 1.38 1.11 0.920 0.775 0.650 0.550 0.475 0.404 0.0980 0.0260 0.00630 0.001651 2.54 1.85 1.45 1.18 0.975 0.825 0.700 0.595 0.510 0.444 0.113 0.0310 0.00840 0.002352 2.97 2.29 1.88 1.60 1.38 1.22 1.07 0.950 0.840 0.750 0.259 0.0950 0.0330 0.01153 3.20 2.50 2.10 1.82 1.60 1.42 1.28 1.15 1.05 0.960 0.388 0.165 0.0700 0.02754 3.36 2.66 2.25 1.97 1.75 1.58 1.42 1.30 1.20 1.10 0.495 0.235 0.112 0.05355 3.49 2.78 2.38 2.09 1.87 1.69 1.54 1.42 1.30 1.21 0.580 0.300 0.150 0.07156 3.59 2.90 2.47 2.18 1.95 1.78 1.65 1.52 1.40 1.30 0.660 0.360 0.195 0.09907 3.66 2.96 2.55 2.25 2.04 1.85 1.70 1.58 1.48 1.38 0.730 0.415 0.230 0.1258 3.74 3.00 2.60 2.32 2.11 1.94 1.79 1.66 1.55 1.44 0.790 0.465 0.272 0.1559 3.80 3.09 2.67 2.39 2.17 2.00 1.85 1.72 1.60 1.50 0.850 0.515 0.307 0.182

10 3.84 3.12 2.74 2.45 2.24 2.05 1.90 1.77 1.65 1.55 0.890 0.550 0.340 0.210

Note: For t9 . 5, V(t9,r9) is about equal to 0.5W[(r9)2/4t], which is the well function in Table 9.7.2.Source: From N.S. Boulton, 1954, The drawdown of water table under non-steady conditions near a pumped well in an unconfined formation, Proc. Inst. Civ. Eng. (London) 3, Pt. 2:564–579.

r9

where m is a function of t9 and can be obtained from acurve plotted through the following points:

t9 0.05 0.2 1 5

m 20.043 0.087 0.512 1.288

—Y.S. Chae

ReferencesBoulton, N.S. 1954. The drawdown of water table under non-steady con-

ditions near a pumped well in an unconfined formation. Proc. Inst.Civ. Eng. (London) 3, pt 2:564–579.

Bouwer, H. 1978. Groundwater hydrology. McGraw-Hill, Inc.Strack, O.D.L. 1989. Groundwater mechanics. Vol. 3, pt. 3:564–579.

Prentice-Hall, Inc.


TABLE 9.7.5 CORRECTION FACTOR Ck

r9 0.03 0.04 0.06 0.08 0.10 0.20 0.40 0.60 0.80 1 2 4Ck 20.27 20.24 20.19 20.16 20.13 20.05 0.02 0.05 0.05 0.05 0.03 0

9.8DETERMINING AQUIFER CHARACTERISTICS

Hydraulic conductivity K, transmissivity T, and storativ-ity S are the hydraulic properties which characterize anaquifer. Before the quantities required to solve ground-water engineering problems, such as drawdown and rateof flow, can be calculated, the hydraulic properties of theaquifer K, S, and T must be determined.

Determining the hydraulic properties of an aquifer gen-erally involves applying field data obtained from a pump-ing test. Other techniques such as auger-hole and piezome-ter methods can be used to determine K where thegroundwater table or aquifers are shallow.

Pumping test technology is prominent in the evaluationof hydraulic properties. It involves observing the draw-down of the piezometric surface or water table in obser-vation wells which are located some distance from thepumping well and have water pumped through them at aconstant rate. Pumping test analysis applies the field datato some form of the Theis equation in general, such as

s 5 }4QpT} W(u, a, b, . . .) 9.8(1)

where u 5 Sr2/4Tt and a,b 5 dimensionless factors defin-ing particular aquifer system conditions. In general, match-ing the field data curve (usually a plot of s versus r2/t) withthe standard curve (known as the type curve) drawn be-tween W and u for various control values of a, b, . . ., cal-culates the values of S and T. This process is explained inthe next section. Techniques requiring no matching havesince been developed.

Various site conditions are associated with a pumpingtest in a well–aquifer system. The following list summa-rizes different site conditions (Gupta 1989):I. Type of pumping

A. DrawdownB. RecoveryC. Interference

II. State of flowA. Steady-stateB. Nonsteady (transient) state

III. Area extent of aquiferA. Aquifer of infinite extentB. Aquifer bound by an impermeable boundaryC. Aquifer bound by a recharge boundary

IV. Depth of wellA. Fully penetrating wellB. Partially penetrating well

V. Confined aquiferA. Nonleaky aquiferB. Leaky confining bed releasing water from storageC. Leaky confining bed not yielding water from stor-

age but transmitting water from overlying layerD. Leaky aquifer in which the head in the overlying

aquifer changesVI. Unconfined aquifer

A. Aquifer in which significant dewatering occursB. Aquifer in which vertical flow occurs near the wellC. Aquifer with delayed yield

Selecting a proper type curve is essential for the dataanalysis. During the last decades, several contributors havedeveloped type curves for various site conditions or com-binations of categories. Starting with Theis, who made theoriginal type curve concept, other contributors to this fieldinclude Cooper and Jacob (1946) and Chow (1952) forconfined aquifers, and Hantush and Jacob (1955), Neu-man and Witherspoon (1969), Walton (1962), Boulton(1963) and Neuman (1972) for unconfined aquifers.

Confined AquifersThis section discusses the methods used in determiningaquifer characteristics for confined aquifers.

STEADY-STATE

The Thiem equation, Equation 9.3(12), gives the draw-down between two points (s1 and s2) measured at distancesr1 and r2, respectively, as

s1 2 s2 5 s 5 }2QpT} ,n 1}

rr2

1

}2 9.8(2)

Hence, T can be calculated by

T 5 }2p(s1

Q2 s2)} ,n 1}

rr2

1

}2 9.8(3)

or from Figure 9.8.1, T can be obtained by

T 5 }22,3p

Q} }

D

D

losg r} 9.8(4)

Figure 9.7.2 shows that the drawdown between twopoints s1 2 s2 reaches a constant value after a day or two.Therefore, Equation 9.8(3) can be used to determine T be-fore the flow achieves a steady state.

Once T has been calculated, S can be determined withthe transient-flow equations, Equations 9.7(14) and9.7(20), as

W(u) 5 }4p

QTs} ® T 5 }

4Qps} W(u) 9.8(5)

u 5 }4r2

TSt

} ® S 5 }4T

r2

tu} 9.8(6)

Since T, Q, and s are known for a given r and t, W(u) canbe obtained. With the use of Table 9.7.2, the correspond-ing value of u can be found. S can be calculated fromEquation 9.8(6).

TRANSIENT-STATE

Three methods of analysis are the type-curve method(Theis), the Cooper–Jacob method, and the Chow method.These methods are briefly described.

Type Curve Method (Theis)

The Theis equation, Equations 9.7(20) and 9.7(14), canbe written respectively as

log s 5 log }4QpT} 1 log W(u) 9.8(7)

log }rt

2

} 5 log }4ST} 1 log u 9.8(8)

If these two equations are plotted on the same log–log pa-per, the resulting curves are the same shape but horizon-tally and vertically offset by the constants Q/4 pT and4T/S. If each curve is plotted on a separate sheet, the curvescan be made to match when the sheets are overlapped asshown in Figure 9.8.2. An arbitrary point on the match-ing curve is selected, and the coordinates of this matchingpoint are read horizontally and vertically on both graphs.


s1

s2

s

r1 r2

log r

Dlog r

Ds

FIG. 9.8.1 Plot of drawdown s versus distance r.

10-310-4 10-2 10-1 1 1u

102

10

1

10-1

W(u

)

104103 105 10610-3

108107

10-2

10-1

1MatchingPoint

s IN

ME

TE

RS

r2/t IN m2/DAY

FIG. 9.8.2 Relations s versus r2/t and W(u) versus u.

These values, s, r2/t, u, and W(u) can then be used to cal-culate T and S from Equations 9.7(20) and 9.7(14).

The following example illustrates the Theis solution (H.Bouwer 1978). With the use of the drawdown data inTable 9.7.2, the data curve and type curve are overlappedto make the two curves match as shown in Figure 9.8.2.Four coordinates of the matching point are:

s 5 0.167m r2/t 5 3 3 106 m2/d

W(u) 5 2.1 u 5 8 3 1022 9.8(9)

Therefore,

T 5 }4Qps} W(u) 5 }

4p

1(00.01067)

} (2.1) 5 1001 m2/d 9.8(10)

S 5 }4rT2/

ut

} 5}4(100

31)

3

(813

06

1022)}5 0.0001 9.8(11)

Cooper-Jacob Method

Cooper and Jacob (1946) showed that when u becomessmall (u ,, 1), the drawdown equation can be repre-sented by Equation 9.7(22) as

s 5 }24.p

3QT} log 1}2.2

Sr52

Tt}2 9.8(12)

On semilog paper, this equation represents a straight linewith a slope of 2.3Q/4 pT. This equation can be plottedin three different ways: (1) s versus log t, (2) s versus logr, or (3) s versus log t/r2 or log r2/t.

DRAWDOWN–TIME ANALYSIS (s VERSUS log t)

The drawdown measurements s at a constant distance rare plotted against time as shown in Figure 9.8.3. Theslope of the line is 2.3Q/4pT and is equal to

5 }24.p

3QT} 9.8(13)Ds

}log }

tt2

1

}

If a change in drawdown Ds is considered for one log cy-cle, then log (t2/t1) 5 1, and this equation reduces to

Ds 5 }24.p

3QT} 9.8(14)

or

T 5 }42p

.3(DQs)

} 9.8(15)

When the straight line intersects the x axis, s 5 0 and thetime is to. Substituting these values in Equation 9.8(12)gives

0 5 }24.p

3QT} log }

2.2r52STto} 9.8(16)

so

1 5 }2.2

r52STto} 9.8(17)

and

S 5 }2.2

r52

Tto} 9.8(18)

Example: Figure 9.8.3 shows that to 5 1.6 3 1023 days andslope Ds 5 0.181. These values yield T 5 1011 m2/d andS 5 0.00009, which agree with the values for T and S ob-tained by the Theis solution.

DRAWDOWN–DISTANCE ANALYSIS (s VERSUS log r)

The drawdown measurements s are plotted against dis-tance r at a given time t as shown in Figure 9.8.4. Fromsimilar considerations as in drawdown–time analysis

T 5 }22p

.3(DQs)

} 9.8(19)

S 5 }2.2

r52o

Tt} 9.8(20)


0.18

Tangent at s = 0.2Extension of Straight-Line Portion of Curve

A

r = 200 m

s in

Met

ers

0.7

0.6

0.5

0.4

0.3

0.2

0.1

010-3 10-2 10-1 1 10

to t in Days

FIG. 9.8.3 Drawdown versus time plot.

20

15

10

5

0

101 102 103 104 105

One cycle

Distance from pumped well, r (log scale) (ft or m)

Dra

wdo

wn,

s (

ft or

m)

Ds

r0

FIG. 9.8.4 Drawdown versus distance plot. (Reprinted fromR.S. Gupta, 1989, Hydrology and hydraulic systems, Prentice-Hall, Inc.)

DRAWDOWN–COMBINED-TIME–DISTANCE ANALYSIS (s VERSUS log r2/t)

The drawdown measurements in many wells at varioustimes are plotted as shown in Figure 9.8.5. Similarly as be-fore

T 5 }42p

.3(DQs)

} 9.8(21)

S 5 2.25T 1}rt2}2

0

9.8(22)

Chow Method

Chow’s procedure (1952) combines the approach of Theisand Cooper–Jacob and introduces the function

F(u) 5 }W

2(u.3)eu

} 5 }Ds/log

s(t2/t1)} 9.8(23)

where s is the drawdown at a point. The relation betweenF(u), W(u), and u is shown in Figure 9.8.6. For one logcycle on a time scale

log(t2/t1) 5 1 9.8(24)

and

F(u) 5 }D

ss} 9.8(25)

From the drawdown–time curve, obtain s at an arbitrarypoint and Ds over one log cycle. The ratio s/Ds is equal toF(u) in the Equation 9.8(25). F(u), W(u), and u can be ob-tained from Figure 9.8.6. With W(u), u, s, and t known,T and S can be calculated with Equations 9.7(20) and9.7(14).

Example: Point A in Figure 9.8.3 gives s 5 0.2m and Ds 50.18m at r 5 200m and F(u) 5 0.2/0.18 5 1.11. From Figure9.8.6, W(u) 5 2.2 and u 5 0.065. Substituting into Equations9.7(20) and 9.7(14) yields T 5 875 m2/d and S 5 0.00011,which reasonably agree with the values obtained by the twomethods just described.

Recovery Test

Figure 9.8.7 schematically shows a recovery test in whichthe water level in the observation wells rises when pump-ing stops after the pumping test is complete. Since the prin-ciple of superposition applies, the drawdown s9 after thepumping test is complete can be expressed as

s9 5 }4QpT} ,n 1}2.2

r2

5STt

}2 2 }Q

42

pTQ9

} ,n 1}2.2r2

5STt9}2 9.8(26)


20

15

10

5

0

1 102 103

One cycleDra

wdo

wn,

s (

ft or

m)

Ds

(t/r02)0

10

(t/r 2 (min/ft2 or min/m2) (log scale)

FIG. 9.8.5 Plot of drawdown versus combined time–distance. (Reprintedfrom R.S. Gupta, 1989, Hydrology and hydraulic systems, Prentice-Hall, Inc.)

10221021

10212 5 2 5 1 2 5 10

2

5

1

2

5

10

W(u)

F(u

)

3.0

2.5

2.0

1.5

1.0

0.80.7

0.6

0.40.3

0.20.1

0.050.01

0.0010.0001

FIG. 9.8.6 Relations between F(u), W(u), and u. (Reprintedfrom V.T. Chow, 1952, On the determination of transmissivityand storage coefficients from pumping test data, Trans. Am.Geoph. Union 33:397–404.) FIG. 9.8.7 Recovery test.

to

s9

t

s

Q

t9

where Q9 is the rate of flow, and t9 is the time after thepumping stops, respectively. Since Q9 5 0, s9 becomes

s9 5 }4QpT} ,n }

tt9} 5 }

24.p

3QT} log }

tt9} 9.8(27)

Thus, T can be calculated as

T 5 }42p

.3D

Qs9

} 9.8(28)

However, S cannot be determined from the recovery test.

Semiconfined (Leaky) AquifersThis section discusses the methods used in determiningaquifer characteristics for semiconfined (leaky) aquifers(Bouwer 1978).

STEADY-STATE

The DeGlee–Hantush–Jacob method (DeGlee 1930, 1951;Hantush and Jacob 1955) and the Hantush method (1956,1964) are used to determine the aquifer characteristics insemiconfined aquifers under steady-state conditions.

De Glee–Hantush–Jacob Method

The drawdown in a semiconfined aquifer is given byEquation 9.3(18) as

s 5 }2QpT} Ko 1}

l

r}2 9.8(29)

where Ko(r/l) 5 modified Bessel function of zero orderand second kind, and l 5 Twcw as defined before. The val-ues of Ko(r/l) versus r/l are shown in Table 9.8.1. Thevalue T can be determined as in a confined aquifer withthe use of the matching procedure. The data curve is ob-tained from a plot of s versus r on log–log paper, and thetype curve is obtained from a plot of Ko(r/l) versus r/l.Overlapping these two plots matches the two curves, andfour coordinates of an arbitrary selected print on thematching curve are noted. The value T is then calculatedfrom Equation 9.8(29) as

T 5 }2Qps} Ko 1}

l

r}2 9.8(30)

The resistance c can be determined from c 5 l2/T whenT and the values of r and r/l of the matching point aresubstituted into this equation.

Hantush Method

Equation 9.3(19) shows that when r/l ,, 1, the draw-down can be approximated by

s 5 }22.p

3QT} log }

1.1r23l} 9.8(31)

A plot of s versus log r forms a straight line, the slope ofwhich is 2.3Q/2pT. If Ds is taken over one log cycle, thenT can be calculated as

T 5 }22p

.(3D

Qs)

} 9.8(32)

Extending the straight line into the abscissa yields the in-tercept ro where s 5 0. Then, from Equation 9.8(31)

0 5 log }1.1

r2

o

3l} 9.8(33)

so

l 5 ro/1.123 9.8(34)

and

c 5 }l

T

2

} 5 }1.2

r2o

5T} 9.8(35)

Note that this method does not require the matching pro-cedure.

TRANSIENT-STATE

Hantush and Jacob (1955) showed that the drawdown ina semiconfined aquifer is described by


TABLE 9.8.1 VALUES OF THE FUNCTIONS K0(X)AND EXP (X)K0(X)

x K0(x) exp (x)K0(x) x K0(x) exp (x)K0(x)

0.01 4.72 4.77 0.35 1.23 1.750.015 4.32 4.38 0.40 1.11 1.60.02 4.03 4.11 0.45 1.01 1.590.025 3.81 3.91 0.50 0.92 1.520.03 3.62 3.73 0.55 0.85 1.470.035 3.47 3.59 0.60 0.78 1.420.04 3.34 3.47 0.65 0.72 1.370.045 3.22 3.37 0.70 0.66 1.330.05 3.11 3.27 0.75 0.61 1.290.055 3.02 3.19 0.80 0.57 1.260.06 2.93 3.11 0.85 0.52 1.230.065 2.85 3.05 0.90 0.49 1.200.07 2.78 2.98 0.95 0.45 1.170.075 2.71 2.92 1.0 0.42 1.140.08 2.65 2.87 1.5 0.21 0.960.085 2.59 2.82 2.0 0.11 0.840.09 2.53 2.77 2.5 0.062 0.7600.095 2.48 2.72 3.0 0.035 0.6980.10 2.43 2.68 3.5 0.020 0.6490.15 2.03 2.36 4.0 0.011 0.6090.20 1.75 2.14 4.5 0.006 0.5760.25 1.54 1.98 5.0 0.004 0.5480.30 1.37 1.85

Source: Adapted from M.S. Hantush, 1956, Analysis of data from pumpingtests in leaky aquifers, Transactions American Geophysical Union 37:702–14and C.W. Fetter, 1988, Applied hydrogeology, 2d ed., Macmillan.

s 5 }4QpT} W 1u, }

l

r}2 9.8(36)

where

u 5 }4r2

TSt

} 9.8(37)

Equation 9.8(36) is similar to Equation 9.7(20) for aconfined aquifer except that the well function contains theadditional term r/l. The values of W(u, r/l) are given inTable 9.8.2.

Walton Method

Walton’s solution (1962) of Equation 9.8(36) is similar tothe Theis method for a confined aquifer. Plotting s versust/r2 gives the data curve. Plotting W(u, r/l) versus u forvarious values of r/l gives several type curves. Figure 9.8.8shows the type curves. The data curve is superimposed onthe type curves to get the best fitting curve. Again, fourcoordinates of a match point are read on both graphs. Theresulting values of W(u, r/l) and s are substituted intoEquation 9.8(36) to calculate T. The value of S is obtainedfrom Equation 9.8(37) when u, t/r2, and T are substituted.The value c is calculated from c 5 l2/T where l is ob-tained from the r/l value of the best fitting curve.

Hantush’s Inflection Point Method

Hantush’s procedure (1956) for calculating T, S, and cfrom pumping test data utilizes the halfway point or in-flection point on a curve relating s to log t. The inflectionpoint is the point where the drawdown s is one-half thefinal or equilibrium drawdown as

s 5 }4QpT} Ko 1}

l

r}2 9.8(38)

The value u at the inflection point is

}2rl} 5 u 5 }

4rT

2Sti

} 9.8(39)

where ti is t at the inflection point. The ratio between thedrawdown and the slope of the curve at the inflection pointDs expressed as the drawdown per unit log cycle of t isderived as

2.3 }D

ss} 5 er/l z Ko 1}

l

r}2 9.8(40)

The values of function er/l z Ko(r/l) versus r/l are in Table9.8.1.

To determine T, S, and c from pumping test data, fol-low the following procedure:

1. Plot drawdown–time on semilog paper (s–log t).2. Locate the inflection point P where s 5 1/2 3 final

drawdown.3. Draw a line tangent to the curve at point P, and de-

termine the corresponding value of ti and the slope Ds.

4. Substitute s and Ds values into Equation 9.8(40) to ob-tain er/l z Ko(r/l), and determine the correspondingvalue of r/l and Ko r/l from Table 9.8.1.

5. Determine T from Equation 9.8(38).6. Determine S from Equation 9.8(39).7. Determine c from c 5 l2/T.

Unconfined AquifersThis section discusses the methods used in determiningaquifer characteristics for unconfined aquifers.

STEADY-STATE

As previously explained, the equation of groundwater flowfor unconfined aquifers reduces to the same form as thatfor confined aquifers except that the thickness of theaquifer is not constant but varies as the aquifer is dewa-tered. Therefore, the flow must be expressed through anaverage thickness of the aquifer fav. The Thiem equationis then

Q 5 5

5 9.8(41)

where f2 5 fo 2 s2 and f1 5 fo 2 s1.From Equation 9.8(41),

Tav 5 9.8(42)

which is the same form as that for a confined aquifer. Thetransmissibility of the aquifer T is then

T 5 }2fo 2

2f

so

1 2 s2

} z Tav 9.8(43)

Once T has been determined, S can be obtained in thesame manner as a confined aquifer. Note that when thesteady-state method is applied, pumping does not have tocontinue until true steady-state conditions are reachedsince Ds 5 s1 2 s2 reaches an essentially constant value af-ter a few days of pumping.

TRANSIENT-STATE

As explained previously, the transient flow of groundwa-ter in an unconfined aquifer occurs from two types of stor-age: phreatic and elastic. As water is pumped out of theaquifer, the decline in pressure in the aquifer yields waterdue to the elastic storage of the aquifer storativity Se, andthe declining water table also yields water as it drains un-der gravity. Unlike the confined aquifer, the release of wa-

Q ,n 1}rr1

2}2

}}

2pTav(f2 2 f1)}}

,n 1}rr2

1

}2

pK2fav(f2 2 f1)}}

,n 1}rr2

1

}2pK(f2

2 2 f21)

}}

,n 1}rr2

1

}2


TABLE 9.8.2 VALUES OF W(U,R/L) FOR DIFFERENT VALUES OF U AND R/L

u 0.002 0.004 0.006 0.008 0.01 0.02 0.04 0.06 0.08 0.1 0.2 0.4 0.6 0.8 1 2 4 6 8

0 12.7 11.3 10.5 9.89 9.44 8.06 6.67 5.87 5.29 4.85 3.51 2.23 1.55 1.13 0.842 0.228 0.0223 0.0025 0.00030.000002 12.1 11.2 10.5 9.89 9.44

4 11.6 11.1 10.4 9.88 9.446 11.3 10.9 10.4 9.87 9.448 11.0 10.7 10.3 9.84 9.43

0.00001 10.8 10.6 10.2 9.80 9.42 8.062 10.2 10.1 09.84 9.58 9.30 8.064 09.52 09.45 09.34 9.19 9.01 8.03 6.676 09.13 09.08 09.00 8.89 8.77 7.98 6.678 08.84 08.81 08.75 8.67 8.57 7.91 6.67

0.0001 08.62 08.59 08.55 8.48 8.40 7.84 6.67 5.87 5.292 07.94 07.92 07.90 7.86 7.82 7.50 6.62 5.86 5.294 07.24 07.24 07.22 7.21 7.19 7.01 6.45 5.83 5.29 4.856 06.84 06.84 06.83 6.82 6.80 6.68 6.27 5.77 5.27 4.858 06.55 06.55 06.54 6.53 6.52 6.43 6.11 5.69 5.25 4.84

0.001 06.33 06.33 06.32 6.32 6.31 6.23 5.97 5.61 5.21 4.83 3.512 05.64 05.64 05.63 5.63 5.63 5.59 5.45 5.24 4.98 4.71 3.504 04.95 04.95 04.95 4.94 4.94 4.92 4.85 4.74 4.59 4.42 3.48 2.236 04.54 4.54 4.53 4.48 4.41 4.30 4.18 3.43 2.238 04.26 4.26 4.25 4.21 4.15 4.08 3.98 3.36 2.23

0.01 04.04 4.04 4.03 4.00 3.95 3.89 3.81 3.29 2.23 1.55 1.132 03.35 3.35 3.35 3.34 3.31 3.28 3.24 2.95 2.18 1.55 1.134 02.68 2.68 2.68 2.67 2.66 2.65 2.63 2.48 2.02 1.52 1.13 0.8426 02.30 2.30 2.29 2.29 2.28 2.27 2.26 2.17 1.85 1.46 1.11 0.8398 02.03 2.03 2.02 2.02 2.01 2.00 1.94 1.69 1.39 1.08 0.832

0.1 01.82 1.82 1.82 1.81 1.80 1.75 1.56 1.31 1.05 0.819 0.2282 01.22 1.22 1.22 1.22 1.22 1.19 1.11 0.996 0.857 0.715 0.2274 00.702 0.702 0.702 0.701 0.700 0.693 0.665 0.621 0.565 0.502 0.2106 00.454 0.454 0.454 0.454 0.453 0.450 0.436 0.415 0.387 0.354 0.177 0.02228 00.311 0.311 0.310 0.310 0.310 0.308 0.301 0.289 0.273 0.254 0.144 0.0218

1 00.219 0.219 0.218 0.213 0.206 0.197 0.185 0.114 0.0207 0.00252 00.049 0.049 0.048 0.047 0.046 0.044 0.034 0.011 0.0021 0.00034 00.0038 0.0038 0.0037 0.0037 0.0036 0.0031 0.0016 0.0006 0.00026 00.0004 0.0004 0.0003 0.0002 0.0001 08 00 0

Source: From M.S. Hantush, 1956, Analysis of data from pumping tests in leaky aquifers, Transactions American Geophysical Union 37:702–14. Reference to the original article is made for more extensive tables and expres-sion of W(u,r/l) in more significant figures (See also M.S. Hantush, 1964, Hydraulics of wells, In Vol. 1 of Advances in hydroscience, edited by V.T. Chow [New York and London: Academic Press]:281–432) and H. Bouwer,1978, Groundwater hydrology, McGraw-Hill, Inc.

©1999 C

RC

Press LL

C

ter from storage is not immediate in response to the dropof the water table. The yield is delayed depending on theelastic and phreatic storativity of the aquifer. Accordingly,the delayed yield produces a sigmoid drawdown curve asshown in Figure 9.8.9.

Essentially, three distinct phases of drawdown–time(s–t) relations occur as shown in Figure 9.8.9: initial phase,intermediate phase, and final phase.

Initial Phase

As the pumping begins, a small amount of water is re-leased from the aquifer under the pressure drop due to the

compression of the aquifer. During this stage, the aquiferbehaves as a confined aquifer, and the flow is essentiallyhorizontal. The drawdown–time data follow a Theis-typecurve (type A) for elastic storativity Se, which is small.

Intermediate Phase

Following the initial phase, as the water table begins todecline, water is drawn primarily from the gravity drainageof the aquifer. The flow at this stage has both horizontaland vertical components, and the s–t relationship is a func-tion of the ratio of the horizontal to vertical hydraulic con-


FIG. 9.8.8 Type curves for a leaky aquifer. (Reprinted from C.W. Fetter, 1988, Applied hydrogeology, 2d ed., MacmillanPub. Co.)

10-1 1.0 10 102 103 104 105 106 107

102

10

1.0

0.1

0.01

1/u

W(u

,r/l

)

0.050.03

0.010.015

0.005

0.0750.10.150.2

0.30.4

0.50.6

0.70.81.0

1.5

2.0

r/ l = 2.5

Nonequilibriumtype curve

10-4 10-2 10-1 100 101 102 103

101

100

10-1

1/uB

W(u

A,u

B,G

)

10-3 104

10-2

1/uA

10-1 100 101 102 103 104

7.06.0

5.04.0

3.02.5

2.01.5

1.00.8

0.60.4

0.20.1

0.06

0.004

0.030.01

0.001

G = r 2Kv / f2oKh

Type B curves

Type

Acu

rves

Thei

scu

rve

Theis curve

FIG. 9.8.9 Type curves and curves for a delayed yield. (Reprinted from C.W. Fetter, 1988, Applied hydrogeol-ogy, 2d ed., Macmillan Pub. Co.)

ductivity of the aquifer, the distance to the pumping well,and the aquifer thickness.

Final Phase

As time elapses, the rate of drawdown decreases, and theflow is essentially horizontal. The s–t data now follow aTheis-type curve (type B) corresponding to the phreaticstorativity Sp, which is large.

Several type-curve solutions have been developed(Walton 1962), such as the one shown in Figure 9.8.9.The flow equation for unconfined aquifers is given by

s 5 }4QpT} W(uA,uB,G) 9.8(44)

where W(uA,uB,G) is the well function, and

uA 5 }S4

e

Tr2

t} (for early drawdown data) 9.8(45)

uB 5 }S

4p

T

r2

t} (for later drawdown data) 9.8(46)

and

G 5 }f

r2

2o

KK

v

h

} 9.8(47)

The values of W(uA,G) and W(uB,G) are given in Tables9.8.3 and 9.8.4. The type curves are used to evaluate thefield data for drawdown and time with the use of the fol-lowing procedure (Fetter 1988):

1. Superpose the late drawdown–time data on the type-B curves for the best fit. At any match point, determinethe values of W(uB,G), uB, t, and s. Obtain the value G fromthe type curve. Calculate T and Sp from

T 5 }4Qps} W(uB,G) 9.8(48)

Sp 5 }4T

rt2

uB} 9.8(49)

2. Superpose the early drawdown data on the type-Acurve for the value G of the previously matched type-Bcurve. Determine a new set of match points, and calculateT and Se from

T 5 }4Qps} W(uA,G) 9.8(50)

Se 5 }4T

rt2

uA} 9.8(51)

The calculated value of T should be approximatelyequal to that computed from the type-B curve.

3. Determine Kh and Kv from

Kh 5 }f

T

o

} 9.8(52)

Kv 5 }Gf

r

2o2

Kh} 9.8(53)

Slug TestsA slug test is a simple and inexpensive way of determin-ing local values of aquifer properties. Instead of the wellbeing pumped for a period of time, a volume of water issuddenly removed or added to the well casing, and re-covery or drawdown are observed over time. Throughcareful evaluation of the drawdown curve and knowledgeof the well screen geometry, the hydraulic conductivity ofan aquifer can be derived (Bedient 1994).


TABLE 9.8.3 VALUES OF THE FUNCTION W(UA,G) FOR WATER TABLE AQUIFERS

1/uA G 5 0.001 G 5 0.01 G 5 0.06 G 5 0.2 G 5 0.6 G 5 1.0 G 5 2.0 G 5 4.0 G 5 6.0

4.0 3 1021 2.48 3 1022 2.41 3 1022 2.30 3 1022 2.14 3 1022 1.88 3 1022 1.70 3 1022 1.38 3 1022 9.33 3 1023 6.39 3 1023

8.0 3 1021 1.45 3 1021 1.40 3 1021 1.31 3 1021 1.19 3 1021 9.88 3 1022 8.49 3 1022 6.03 3 1022 3.17 3 1022 1.74 3 1022

1.4 3 100 3.58 3 1021 3.45 3 1021 3.18 3 1021 2.79 3 1021 2.17 3 1021 1.75 3 1021 1.07 3 1021 4.45 3 1022 2.10 3 1022

2.4 3 100 6.62 3 1021 6.33 3 1021 5.70 3 1021 4.83 3 1021 3.43 3 1021 2.56 3 1021 1.33 3 1021 4.76 3 1022 2.14 3 1022

4.0 3 100 1.02 3 100 9.63 3 1021 8.49 3 1021 6.88 3 1021 4.38 3 1021 3.00 3 1021 1.40 3 1021 4.78 3 1022 2.15 3 1022

8.0 3 100 1.57 3 100 1.46 3 100 1.23 3 100 9.18 3 1021 4.97 3 1021 3.17 3 1021 1.41 3 1021

1.4 3 101 2.05 3 100 1.88 3 100 1.51 3 100 1.03 3 100 5.07 3 1021

2.4 3 101 2.52 3 100 2.27 3 100 1.73 3 100 1.07 3 100

4.0 3 101 2.97 3 100 2.61 3 100 1.85 3 100 1.08 3 100

8.0 3 101 3.56 3 100 3.00 3 100 1.92 3 100

1.4 3 102 4.01 3 100 3.23 3 100 1.93 3 100

2.4 3 102 4.42 3 100 3.37 3 100 1.94 3 100

4.0 3 102 4.77 3 100 3.43 3 100

8.0 3 102 5.16 3 100 3.45 3 100

1.4 3 103 5.40 3 100 3.46 3 100

2.4 3 103 5.54 3 100

4.0 3 103 5.59 3 100

8.0 3 103 5.62 3 100

1.4 3 104 5.62 3 100 3.46 3 100 1.94 3 100 1.08 3 100 5.07 3 1021 3.17 3 1021 1.41 3 1021 4.78 3 1022 2.15 3 1022

Source: Adapted from S.P. Neuman, 1975, Water Resources Research 11:329–42 and C.W. Fetter, 1988, Applied hydrogeology, 2d ed., Macmillan.

Hvorslev (1951) developed the simplest slug testmethod in a piezometer, which relates the flow rate Q(t)at the piezometer at any time to the hydraulic conductiv-ity and unrecovered head distance Ho 2 h in Figure 9.8.10by

Q(t) 5 pr2 }ddht} 5 FK(Ho 2 h) 9.8(54)

where F is a factor that depends on the shape and the di-mensions of the piezometer intake. If Q 5 Qo at t 5 0,then Q(t) decreases toward zero as time increases. Hvorslevdefined the basic time lag To 5 pr2/FK and solved Equation9.8(54) with initial conditions h 5 Ho at t 5 0. Thus

}HH

2

2

Hh

o

} 5 e2t /To 9.8(55)

When recovery H 2 h/H 2 Ho versus time is plotted onsemilog paper, To is noted at t where recovery equals 37%of the initial change. For the piezometer intake length di-vided by radius, L/R greater than 8, Hvorslev has evalu-ated the shape factor F and obtained an equation for K as

K 5 }r2,

2nL(TL/

o

R)} 9.8(56)

Several other slug test methods have been developed forconfined aquifers by Cooper et al. (1967) and Papadop-oulos et al. (1973). These methods are similar to Theis’sin which a curve-matching procedure is used to obtain Sand T for a given aquifer. Figure 9.8.11 shows the slugtest curves developed by Papadopoulos for various valuesof variable a, defined as

a 5 }rr2

2

c

s} S 9.8(57)

The obtained data are plotted and matched to the plottedtype curves for a best match, from which a is selected fora particular curve. The vertical time axis t which overlaysthe vertical axis for Tt/r2

c 5 1.0 is selected, and a value ofT can then be found from T 5 1.0r2

s/t1. Then, the valueof S can be found from the definition of a. The method isrepresentative of the formation only in the immediate vicin-ity of the test hole and should be used with caution (Bedient1994).

The most commonly used method for determining hy-draulic conductivity in groundwater investigation is theBouwer and Rice (1976) slug test shown in Figure 9.8.12.Although it was originally designed for unconfined


TABLE 9.8.4 VALUES OF THE FUNCTION W(UB,G) FOR WATER TABLE AQUIFERS

1/uB G 5 0.001 G 5 0.01 G 5 0.06 G 5 0.2 G 5 0.6 G 5 1.0 G 5 2.0 G 5 4.0 G 5 6.0

4.0 3 1024 5.62 3 100 3.46 3 100 1.94 3 100 1.09 3 100 5.08 3 1021 3.18 3 1021 1.42 3 1021 4.79 3 1022 2.15 3 1022

8.0 3 1024 4.80 3 1022 2.16 3 1022

1.4 3 1023 4.81 3 1022 2.17 3 1022

2.4 3 1023 4.84 3 1022 2.19 3 1022

4.0 3 1023 5.08 3 1021 3.18 3 1021 1.42 3 1021 4.88 3 1022 2.21 3 1022

8.0 3 1023 5.09 3 1021 3.19 3 1021 1.43 3 1021 4.96 3 1022 2.28 3 1022

1.4 3 1022 5.10 3 1021 3.21 3 1021 1.45 3 1021 5.09 3 1022 2.39 3 1022

2.4 3 1022 5.12 3 1021 3.23 3 1021 1.47 3 1021 5.32 3 1022 2.57 3 1022

4.0 3 1022 5.16 3 1021 3.27 3 1021 1.52 3 1021 5.68 3 1022 2.86 3 1022

8.0 3 1022 1.09 3 100 5.24 3 1021 3.37 3 1021 1.62 3 1021 6.61 3 1022 3.62 3 1022

1.4 3 1021 1.94 3 100 1.10 3 100 5.37 3 1021 3.50 3 1021 1.78 3 1021 8.06 3 1022 4.86 3 1022

2.4 3 1021 1.95 3 100 1.11 3 100 5.57 3 1021 3.74 3 1021 2.05 3 1021 1.06 3 1021 7.14 3 1022

4.0 3 1021 1.96 3 100 1.13 3 100 5.89 3 1021 4.12 3 1021 2.48 3 1021 1.49 3 1021 1.13 3 1021

8.0 3 1021 5.62 3 100 3.46 3 100 1.98 3 100 1.18 3 100 6.67 3 1021 5.06 3 1021 3.57 3 1021 2.66 3 1021 2.31 3 1021

1.4 3 100 5.63 3 100 3.47 3 100 2.01 3 100 1.24 3 100 7.80 3 1021 6.42 3 1021 5.17 3 1021 4.45 3 1021 4.19 3 1021

2.4 3 100 5.63 3 100 3.49 3 100 2.06 3 100 1.35 3 100 9.54 3 1021 8.50 3 1021 7.63 3 1021 7.18 3 1021 7.03 3 1021

4.0 3 100 5.63 3 100 3.51 3 100 2.13 3 100 1.50 3 100 1.20 3 100 1.13 3 100 1.08 3 100 1.06 3 100 1.05 3 100

8.0 3 100 5.64 3 100 3.56 3 100 2.31 3 100 1.85 3 100 1.68 3 100 1.65 3 100 1.63 3 100 1.63 3 100 1.63 3 100

1.4 3 101 5.65 3 100 3.63 3 100 2.55 3 100 2.23 3 100 2.15 3 100 2.14 3 100 2.14 3 100 2.14 3 100 2.14 3 100

2.4 3 101 5.67 3 100 3.74 3 100 2.86 3 100 2.68 3 100 2.65 3 100 2.65 3 100 2.64 3 100 2.64 3 100 2.64 3 100

4.0 3 101 5.70 3 100 3.90 3 100 3.24 3 100 3.15 3 100 3.14 3 100 3.14 3 100 3.14 3 100 3.14 3 100 3.14 3 100

8.0 3 101 5.76 3 100 4.22 3 100 3.85 3 100 3.82 3 100 3.82 3 100 3.82 3 100 3.82 3 100 3.82 3 100 3.82 3 100

1.4 3 102 5.85 3 100 4.58 3 100 4.38 3 100 4.37 3 100 4.37 3 100 4.37 3 100 4.37 3 100 4.37 3 100 4.37 3 100

2.4 3 102 5.99 3 100 5.00 3 100 4.91 3 100 4.91 3 100 4.91 3 100 4.91 3 100 4.91 3 100 4.91 3 100 4.91 3 100

4.0 3 102 6.16 3 100 5.46 3 100 5.42 3 100 5.42 3 100 5.42 3 100 5.42 3 100 5.42 3 100 5.42 3 100 5.42 3 100

8.0 3 102 6.47 3 100 6.11 3 100 6.11 3 100 6.11 3 100 6.11 3 100 6.11 3 100 6.11 3 100 6.11 3 100 6.11 3 100

1.4 3 103 6.67 3 100 6.67 3 100 6.67 3 100 6.67 3 100 6.67 3 100 6.67 3 100 6.67 3 100 6.67 3 100 6.67 3 100

2.4 3 103 7.21 3 100 7.21 3 100 7.21 3 100 7.21 3 100 7.21 3 100 7.21 3 100 7.21 3 100 7.21 3 100 7.21 3 100

4.0 3 103 7.72 3 100 7.72 3 100 7.72 3 100 7.72 3 100 7.72 3 100 7.72 3 100 7.72 3 100 7.72 3 100 7.72 3 100

8.0 3 103 8.41 3 100 8.41 3 100 8.41 3 100 8.41 3 100 8.41 3 100 8.41 3 100 8.41 3 100 8.41 3 100 8.41 3 100

1.4 3 104 8.97 3 100 8.97 3 100 8.97 3 100 8.97 3 100 8.97 3 100 8.97 3 100 8.97 3 100 8.97 3 100 8.97 3 100

2.4 3 104 9.51 3 100 9.51 3 100 9.51 3 100 9.51 3 100 9.51 3 100 9.51 3 100 9.51 3 100 9.51 3 100 9.51 3 100

4.0 3 104 1.94 3 101 1.94 3 101 1.94 3 101 1.94 3 101 1.94 3 101 1.94 3 101 1.94 3 101 1.94 3 101 1.94 3 101

Source: Adapted from S.P. Neuman, 1975, Water Resources Research 11:329–42 and C.W. Fetter, 1988, Applied hydrogeology, 2d ed., Macmillan.

aquifers, it can be used for confined aquifers if the top ofthe screen is some distance below the upper confining layer.The method is based on the following equation:

K 5 }r2

c ,n2(LR

e

e/rw)} }

1t} ,n }

yy

o

t

} 9.8(58)

where:

Re 5 effective radial distance over which the head differ-ence is dissipated

rw 5 radial distance between the well center and theundisturbed aquifer (including gravel pack)

Le 5 height of the perforated, screened, uncased, or oth-erwise open section of the well through whichgroundwater enters

yo 5 y at time zeroyt 5 y at time tt 5 time sine yo

In Equation 9.8(58), y and t are the only variables. Thus,if a number of y and t measurements are taken, they canbe plotted on semilog paper to give a straight line. Theslope of the best-fitting straight line provides a value for,n(yo/yt)/t. All other parameters in Equation 9.8(58) areknown from well geometry, and K can be calculated.

—Y.S. Chae

ReferencesBedient, P.B. et al. 1994. Groundwater contamination, PTR Prentice-

Hall, Inc.


FIG. 9.8.10 Hvorslev piezometer test.

FIG. 9.8.11 Slug test type curves. (Reprinted from I.S.Papadopoulos, J.D. Bredehoeft, and H.H. Cooper, Jr., 1973, Onthe analysis of slug test data, Water Resources Res. 9, no. 4:1087–1089.)

(b)

1.0

0.5

0.37

0.2

0.1420 6 8 10

H -

hH

- H

o

t (hr)

To

(a)

t = 0

q

R

Date

tt + dt

t ¥ 0 (and r < 0)

Rec

over

y

H h Ho

dh

r

L

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.010-3 10-2 10-1 1021 10

10-2 10-1 1 10 100 1000

(log10 a = -1)

(-10)(-9)(-8)(-7)(-6)(-5)(-4)(-3)(-2)

=rs

2

rc2

Sa

Line source approximation

=rc

2

Tt1.0

Tt /rc2

Time (s)

H/H

0

��@@@@@@��ÀÀÀÀÀÀ��@@@@@@��ÀÀÀÀÀÀ��@@@@@@��ÀÀÀÀÀÀ�� @@@@@ �� ÀÀÀÀÀ �� @@@@@ �� ÀÀÀÀÀ �� @@@@@ �� ÀÀÀÀÀ ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

2rc

WATER TABLEy

Lw

2rw Le

IMPERMEABLE

H

FIG. 9.8.12 Slug test setup. (Reprinted from H. Bouwer, 1978,Groundwater hydrology, McGraw-Hill, Inc.)

Boulton, N.S. 1963. Analysis of data from non-equilibrium pumping testsallowing for delayed yield from storage. Proc. Inst. Civ. Eng.16:469–482.

Bouwer, H. 1978. Groundwater hydrology. McGraw-Hill, Inc.Bouwer, H., and R.C. Rice. 1976. A slug test for determining hydraulic

conductivity of unconfined aquifers with completely or partially pen-etrating wells. Water Resour. Res. 12:423–428.

Chow, V.T. 1952. On the determination of transmissivity and storagecoefficients from pumping test data. Trans. Am. Geoph. Union 33:397–404.

Cooper, H.H., Jr., J.D. Bredchoeft, and I.S. Papadopoulos. 1967.Response of a finite-diameter well to an instantaneous charge of wa-ter. Water Resour. Res. 3:263–269.

Cooper, H.H., Jr., and C.E. Jacob. 1946. A generalized graphical methodfor evaluating formation constants and summarizing well-field his-tory, Trans. Am, Geoph. Union 27:526–534.

DeGlee, G.J. 1930. Over grondwaterstromingen bij wateronttrekkingdoor middel van putten. Doctoral dissertation, Techn. Univ., Delft.The Netherlands. Printed by J. Waltman.

DeGlee, G.J. 1951. Berekeningsmethoden voor de winning van grond-water. In Drinkwaterroorzlening 3e Vacantie cursus, 38–80. Moor-man’s periodieke pers. The Hague, Netherlands.

Fetter, C.W. 1988. Applied hydrogeology. 2d ed. Macmillan Pub. Co.Gupta, R.S. 1989. Hydrology and hydraulic systems. Prentice-Hall, Inc.Hantush, M.S. 1956. Analysis of data from pumping tests in leaky

aquifers. Trans. Am. Geophys. Un. 37:702–714.Hantush, M.S. 1964. Hydraulics of wells. In Advances in Hydroscience.

Vol. 1: edited by V.T. Chow, 281–432, New York and London:Academic Press.

Hantush, M.S., and C.E. Jacob. 1955. Non-steady radial flow in an in-finite leaky aquifer. Am. Geophys. Un. Trans. 36:95–100.

Hvorslev, M.J. 1951. Time lag and soil permeability in groundwater ob-servations. U.S. Army Waterways Experiment Station Bull. 36.

Neuman, S.P. 1972. Theory of flow in unconfined aquifers consideringdelayed response of the water table. Water Resources Res. 8:1031–1045.

Neuman, S.P., and P.A. Witherspoon. 1969. Theory of flow in a con-fine two-aquifer system. Water Resource Research 5:803–816.

Papadopoulos, I.S., J.D. Bredehoeft, and H.H. Cooper, Jr. 1973. On theanalysis of slug test data. Water Resources Res. 9, no. 4:1087–1089.

Walton, W.C. 1962. Selected analytical methods for well and aquiferevaluation. Illinois State Water Surrey Bull. 49.


9.9DESIGN CONSIDERATIONS

In well design, well losses, specific capacity, and partiallypenetrating wells must be considered.

Well LossesIn the previous sections, the drawdown in a pumping wellwas assumed to be due only to head losses in the aquifer.In reality, however, additional drawdown is caused byhead losses in the well system itself (screen or perforatedcasing) as water flows through it. The former is known asthe drawdown due to the formation loss (sw) and the lat-ter as the drawdown due to the well loss (sf) as shown inFigure 9.9.1. The total drawdown st is then st 5 sw 1 sf.

Since the flow in the aquifer is laminar, sw varies lin-early with Q(sw 5 CwQ). However, the flow through thewell system (screen and perforated casing) is turbulent, andthus sf can vary with some power of Q(sf 5 CfQ

n). Thetotal drawdown can be expressed as

st 5 sw 1 sf 5 CwQ 1 CfQn 9.9(1)

where

Cw 5 }2p

1T

} ,n 1}rR

w

}2 for steady-state 9.9(2)

Cw 5 }4p

1T

} W 1}4r2w

TSt

}2 for unsteady-state 9.9(3)

n 52 to 3.5

The best way to determine the values of Cw, Cf, and n isby experiment utilizing a step-drawdown test (Bear 1979;

Bouwer 1978). Well efficiency is related to the well lossand is defined as

Ew 5 }ssw

t

} 5 1 2 }ssf

t

} 9.9(4)

Specific CapacityThe specific capacity of a well is defined as the well flowper unit drop of water level in the well.

Specific capacity 5 }Qst

} 5 }Cw 1

1CfQ

n21} 9.9(5)

Specific capacity decreases with the pumping rate andtime as shown in Figure 9.9.2. It is a useful concept be-cause it describes the productivity of both the aquifer andthe well in a single parameter. A reduction of up to 40%in the specific capacity has been observed in one year inwells deriving water entirely from storage (Gupta 1989).The specific capacities of wells in an aquifer system can beused to calculate the transmissivity distribution of theaquifer based on pumping a well of a known diameter fora given period of time.

Partially Penetrating Wells (ImperfectWells)In practice, the underlying impermeable soil layer is oftenabsent or is encountered at a great depth. Wells which do

not penetrate through the entire thickness of the aquiferare called partially penetrating wells or imperfect wells.Imperfect wells are encountered more often in practice thanperfect, fully penetrating, wells. Imperfect wells can alsohave open or closed ends. Figure 9.9.3 shows various con-figurations of wells in an unconfined aquifer (ordinary

wells), and Figure 9.9.4 shows configurations for wells ina confined aquifer (artesian wells).

Compared to a fully penetrating well, a partially pene-trating well has an additional head loss, as shown in Figure9.9.5, due to the convergence of flow lines and their ex-tended length. Hence, for a given pumping rate Q, thedrawdown in an imperfect well is larger than in a perfectwell.

Starting with Forchheimer in 1898, numerous analyti-cal and empirical equations have been developed to solveimperfect wells: Kozney (1933), Muskat (1937), Li et al,Hantush (1962), and Kirkham (1959) among others.

CONFINED AQUIFERS

As previously stated, an imperfect well requires more draw-down for a given Q than does a perfect well. The addi-tional drawdown can be represented by

swp 5 }4QpT} 1,n }

2.2r2

5STt

} 1 2sp2 9.9(6)

where swp is the drawdown at the well. The values of sp

as a function of H/rw for various values of Le/H can beobtained from Figure 9.9.6, which was developed bySternberg (1967).

The performance of an imperfect well related to a per-fect well is expressed as an efficiency, defined as the ratioof Qp to Q at a given drawdown as

}Q

Qp

} 5 9.9(7)

or

}QQ

p} 5

1 19.9(8)

Equation 9.9(8) applies to wells with their perforatedor open section at the top (Figure 9.9.5-c) or at the bot-

sp},n 1}

rr

w

}2

1}}1 1 }

s

s

w

p} }

2QpT}


��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Q

st

sw

sf

FIG. 9.9.1 Formation and well losses in a pumping well.(Reprinted from R.S. Gupta, 1989, Hydrology and hydraulic sys-tems, Prentice-Hall, Inc.)

100010010118

27

36

45

54

Spe

cific

wel

l dis

char

ge (

g/m

in/ft

)

Time (days)

Qw 5 450 g/min

Qw 5 900 g/min

Qw 5 1350 g/min

FIG. 9.9.2 Examples of variation of specific discharge withpumping rate and time. (Reprinted from J. Bear, 1979,Hydraulics of groundwater, McGraw-Hill, Inc.)

��

GS

G.W.T.

2ro

QQQQQQ

2ro

Per

mea

ble

fo

fw

Le

S

Le

S

fw

Le

S

fwLe

Perfect well

fw

S

654321

FIG. 9.9.3 Types of ordinary imperfect wells. (Reprinted from A.R. Jumikis, 1964,Mechanics of soils, Van Nostrand.)

1

tom. If the open section of the well is in the center of theaquifer (Figure 9.9.5-a), vertical flow components occur atboth the top and bottom of the section. For this case, Qp/Qcan be obtained for the half-section when the section issplit symmetrically along the midway.

UNCONFINED AQUIFERS

The sp values in Figure 9.9.6 also give reasonable estimatesof Qp/Q for wells in unconfined aquifers, particularly if thedrawdown is small compared to the aquifer thickness andthe well has been pumped for some time (Bouwer 1978).

Forchheimer (1898) observed that the discharge Qp of awell with a perforated casing and closed end becomes larger,at equal drawdowns, as the immersed depth Le of the per-forated well increases. He assumed that the increase withdepth varies with the geometric mean between the parabolicand elliptic ordinates and showed an efficiency to be

}Q

Qp

} 5 !}Lz¤e}¤ 4!}

2¤z¤2

z¤ L¤e}¤ for closed end 9.9(9)

and

}Q

Qp

} 5 !}L¤e¤1¤z

0¤.5¤rw}¤ z 4!}

2¤z¤2

z¤ L¤e}¤ for open end 9.9(10)

where z 5 distance from the water level in the well to theimpervious stratum in meters.

—Y.S. Chae

ReferencesBear, J. 1979. Hydraulics of groundwater. McGraw-Hill, Inc.Bouwer, H. 1978. Groundwater hydrology. McGraw-Hill, Inc.Forchheimer, Ph., 1898. Grundwasserspiegel bei Brunnenanlagen.

Zeitschrift des Osterreichischen Ingenieur-und Architekten-Vereins50, no. 45:645.

Gupta, R.S. 1989. Hydrology and hydraulic systems. Prentice-Hall, Inc.Hantush, M.S. 1962. Aquifer tests on partially penetrating wells. Trans.

Am. Soc. Civ. Eng. Vol. 127, pt. 1:284–308.Kirkham, D. 1959. Exact theory of flow into a partially penetrating well.

J. Geophys. Research 64:1317–1327.Kozeny, J. 1933. Theorie und Berechnung der Brunnen. Wasserkraft und

Wasserwirtschaft 28:104.Li, W.H. et al. A new formula for flow into partially penetrating wells

in aquifers. Trans. Am. Geophys. Union 35:806–811.Muskat, M. 1937. The flow of homogeneous fluids through porous me-

dia. McGraw-Hill.Sternberg, Y.M. 1967. Efficiency of partially penetrating wells. Ground

Water 11, no. 3:5–8.


��

GS

QQQQQ

Le

Perfect well54321

Imperfect wells

Impermeable layer

Contact surface

Piezometric level

H

Impe

rmea

ble

laye

r

Confinedaquifer

��

fo

fw

FIG. 9.9.4 Types of artesian imperfect wells. (Reprinted fromA.R. Jumikis, 1964, Mechanics of soils, Van Nostrand.)

��

��

b2

bs

b1

BP

z

r

Qw

2rw

Piezometricsurfaces

(a)

��

Qw

2rw

Bottom

Average

(c)

��

Qw

2rw

(b)

��2rw

(d)

R

Top

FIG. 9.9.5 Partially penetrating wells. (a) In a confined aquifer.(b) In a phreatic aquifer. (c) Drawdown curves along streamlines.(d) Zero penetration in a thick aquifer.

FIG. 9.9.6 Graph of sp versus H/rw. (Reprinted from Y.M.Sternberg, 1967, Efficiency of partially penetrating wells, GroundWater II, no. 3:5–8.)

60

50

40

30

20

10

01 10 100 1000 10000

H/rw

CURVE PARAMETERLe

H

0.1

0.2

0.4

0.6

0.8

Sp

The flow of groundwater in coastal aquifers, as shown inFigure 9.10.1, can be treated as an interface flow problemin which two fluids of different densities, fresh and saltwater, have a clear interface rather than a transition zone.This flow problem assumes that the fresh water flows overthe salt water which is at rest. These flows are denoted asthe Ghyben–Duipint approximations.

The pressure distribution in the salt water rs is

ps 5 rsg(fs 2 z) 9.10(1)

and the pressure distribution in the fresh water pf is

pf 5 rfg(f 2 z) 9.10(2)

where fs and f are the head in the salt and fresh waterrespectively, and z is the distance from the reference planeto the interface. The pressure at any point of the interfacemust be a single value, that is pf 5 ps. Therefore, withEquations 9.10(1) and 9.10(2) and with z 5 Hs 2 hs, then

rsg(fs 2 Hs 1 hs) 5 rfg(f 2 Hs 1 hs) 9.10(3)

If fs 5 Hs and f 5 Hs 1 hf, Equation 9.10(3) yields

hs 5 }rs 2

rf

rf

} hf 5 ahf 9.10(4)

This equation is known as the Ghyben–Herzberg equa-tion. This equation is also valid for confined aquifers, inwhich the upper boundary of the aquifer is a horizontalimpermeable boundary rather than a phreatic surface andhf represents the piezometric head with respect to sea level.The ratio between the densities of salt and fresh water isof the order of 1.025. Then, Equation 9.10(4) shows that

hs is about 40 times h. Therefore, in coastal aquifers, stor-age of 40 m of fresh water exists below sea level for everymeter of fresh water above sea level.

Confined Interface FlowFigure 9.10.2 shows the shallow confined interface flowwhen the aquifer is bounded above by a horizontal im-pervious boundary and below by an interface. Since h 5hs 2 (Hs 2 H) and the head f 5 hf 1 Hs, use of theGhyben–Herzberg equation gives the thickness of theaquifer h as

h 5 }rs 2

rf

rf

} f 2 }rs 2

rs

rf

} Hs 1 H 9.10(5)

The elevation z of the interface above the reference levelequals z 5 Hs 2 hs. Use of the Ghyben–Herzberg equa-tion then yields


9.10INTERFACE FLOW

sea

salt water at rest

interface

Hs hs

P

hf

groundwater table

z

FIG. 9.10.1 Interface flow in coastal aquifers. (Reprinted fromO.D.L. Strack, 1989, Groundwater mechanics, Vol. 3, Pt. 3,Prentice-Hall, Inc.)

N

Hshs h

hf

z

H

interface

reference level

FIG. 9.10.2 Shallow confined interface flow. (Reprinted from O.D.L. Strack,Groundwater mechanics, Vol. 3, Pt. 3, Prentice-Hall, Inc.)

z 5 }rs 2

rs

rf

} Hs 2 }rs 2

rf

rf

} f 9.10(6)

Unconfined Interface FlowA shallow unconfined aquifer is shown in Figure 9.10.2.The aquifer thickness h can now be expressed with theGhyben–Herzberg equation as

h 5 }rs 2

rs

pf

} f 2 }rs 2

rs

rf

} Hs 9.10(7)

and the elevation of the interface z is obtained in the sameway as the confined interface flow as

z 5 }rs 2

rs

pf

} Hs 2 }rs 2

rf

rf

} f 9.10(8)

Upconing of Saline WaterFigure 9.10.3 depicts a situation in which water is pumpedfrom a freshwater zone underlaid by a saline water layer.The interface between fresh water and saline water risestoward the well in a cone shape as shown in the figure.This phenomenon is known as upconing.

The height of upconing under the steady-state condi-tion (Gupta 1989) is given by

z 5 }2p

Q

Kd} }

rs 2

rf

rf

} 9.10(9)

where d 5 depth to the initial interface below the bottomof the well. Salt water reaches the well, contaminating thesupply, when the rise becomes critical at z 5 0.3 to 0.5 d.Thus, the maximum discharge that keeps the rise belowthe critical limit is obtained when z 5 0.5 d is substitutedin Equation 9.10(9) as

Qmax 5 pKd2 }rs 2

rf

rf} 9.10(10)

In reality, brackish water occurs between fresh and saltwater. Even with a low rate of pumping, some saline wa-ter inevitably reaches the pump. Increasing the distance dand decreasing the rate of pumping Q minimizes the up-coning effect.

Protection Against IntrusionControlling the intrusion of saline water before it contam-inates an aquifer system is desirable because removing itonce it has developed is difficult. Years may be requiredto restore normal conditions. Table 9.10.1 summarizesmany control methods suggested for various categories ofproblems.

—Y.S Chae

ReferenceGupta, R.S. 1989. Hydrology and hydraulic systems. Prentice-Hall, Inc.


��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

Interfacereaching the

well

Fresh water

Criticaldepth

Saline water

Initial interfaceof fresh andsaline water

Z

d

Q

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Piezometricsurface

FIG. 9.10.3 Upconing of salt water under a pumping well.(Reprinted from R.S. Gupta, 1989, Hydrology and hydraulic sys-tems, Prentice-Hall, Inc.)

TABLE 9.10.1 CONTROLLING SALTWATERINTRUSION OF VARIOUSCATEGORIES

Source or Cause of Intrusion Control Methods

Seawater in coastal aquifer Modification of pumping patternArtificial rechargeExtraction barrierInjection barrierSubsurface barrier

Upconing Modification of pumping patternSaline scavenger wells

Oil field brine Elimination of surface disposalInjection wellsPlugging of abandoned wells

Defective well casings Plugging of faulty wellsSurface infiltration Elimination of sourceSaline water zones in Relocation and redesign of wells

freshwater aquifers

Source: D.K. Todd, 1980, Groundwater hydrology (John Wiley and Sons).

A groundwater contaminant is defined by most regulatoryagencies as any physical, chemical, biological, or radio-logical substance or matter in groundwater. The contam-inants can be introduced in the groundwater by naturallyoccurring activities, such as natural leaching of the soil andmixing with other groundwater sources having differentchemistry. They are also introduced by planned human ac-tivities, such as waste disposal, mining, and agriculturaloperations. Because the contamination from naturally oc-curring activities is usually small, it is not the focus of thischapter. However, human activities are the leading causeof groundwater contamination and the focus of most reg-ulatory agencies.

The most prevalent human activities that cause ground-water contamination are (1) waste disposal, (2) storageand transportation of commercial materials, (3) mining op-erations, (4) agricultural operations, and (5) other activi-ties as shown in Figure 9.11.1.

This section discusses the principal sources and causesof groundwater contamination from these activities withregard to their occurrence and effects on groundwaterquality.

Waste DisposalWaste disposal includes the disposal of liquid waste andsolid waste.

LIQUID WASTE

Underground or aboveground disposal practices of do-mestic, municipal, or industrial liquid waste can causegroundwater contamination. Among all disposal practicesof domestic liquid waste, septic tanks and cesspools con-tribute the most wastewater to the ground and are themost frequently reported sources of groundwater contam-ination (U.S. EPA 1977). Septic tanks and cesspools con-tribute filtered sewage effluent directly to the ground whichcan introduce high concentrations of BOD, COD, nitrate,organic chemicals, and possibly bacteria and viruses into

groundwater (Mallmann and Mack 1961; Miller 1980).Also, chlorination of the wastewater effluent and the useof chemicals to clean septic systems can produce additionalpotential pollutants (Council on Environmental Quality1980).

With regard to municipal liquid waste, land applicationof sewage effluent and sludge is perhaps the largest con-tributor to groundwater contamination. Treated waste-water and sludge have been applied to land for many yearsto recharge groundwater and provide nutrients that fertil-ize the land and stimulate plant growth (Bauer 1974; U.S.EPA 1983). However, land application of sewage effluentcan introduce bacteria, viruses, and organic and inorganicchemicals into groundwater (U.S. EPA 1974).

Another major municipal source of groundwater con-tamination is urban runoff from roadway deicing. In manyurban areas, large quantities of salts and deicing additivesare applied to roads during the winter months. These saltsand additives facilitate the melting of ice and snow; how-ever, they can percolate with the water into the groundand cause groundwater contamination of shallow aquifers(Field et al. 1973). In addition, the high solubility of thesesalts in water and the relatively high mobility of the re-sulting contaminants such as chloride ions in groundwa-ter can cause the zone of contamination to expand (Terry1974).

With regard to industrial liquid waste, surface im-poundments and injection wells are probably the largestcontributors to groundwater contamination. As legislationto protect surface water resources has become more strin-gent, the use of surface impoundments and injection wellshas become an attractive wastewater effluent disposal op-tion for many industries. However, leakage of contami-nants through the bottom of a surface impoundment ormigration of fluids from an injection well into a hydro-logically connected usable aquifer can cause groundwatercontamination (Council on Environmental Quality 1981).The extent and severity of groundwater contaminationfrom these sources is further complicated by the fact that,in addition to being hazardous, many of the organic and


Principles of GroundwaterContamination

9.11CAUSES AND SOURCES OF CONTAMINATION

©1999 C

RC

Press LL

C

FIG. 9.11.1 Sources of groundwater contamination. (Reprinted from National Geographic, 1993, Water, National Geographic Special Editions [November], Washington, D.C.:National Geographic Society.)

inorganic chemicals in industrial wastewater effluent andsludge are persistent.

SOLID WASTE

The land disposal of municipal and industrial solid wasteis another potential cause of groundwater contamination.Buried waste is subject to leaching by percolating rain wa-ter and surface water or by groundwater contact with thefill. The generated leachate can contain high levels of BOD,COD, nitrate, chloride, alkalinity, trace elements, and eventoxic constituents (in industrial waste landfill) that can de-grade the quality of groundwater (Hughes et al. 1971;Zanoni 1972). In addition, the biochemical decompositionof the organic matter in waste generates gases such asmethane, carbon dioxide, ammonia, and hydrogen sulfidethat can migrate through the unsaturated zone into adja-cent terrains and cause potential hazards such as methaneexplosions (Flower 1976; Mohsen 1975).

Stockpiles of materials and waste tailings can also be asource of groundwater contamination. Precipitation fallingon uncovered or unlined stockpiles or waste tailings causesleachate generation and seepage into the ground. Theleachate can transport heavy metals, salts, and other in-organic and organic constituents as pollutants to ground-water.

Storage and Transport of CommercialMaterialsGroundwater contamination from the storage and trans-port of commercial materials results from leaking storagetanks and spills.

STORAGE TANKS

Underground and aboveground storage tanks and trans-mission pipelines are another cause of groundwater pol-lution. Among all underground storage tanks andpipelines, gasoline and home oil fuel tanks probably con-tribute the most to groundwater contamination. Thesetanks and pipelines are subject to corrosion and structuralfailures with subsequent leaks that introduce a variety ofcontaminants into groundwater. Leakage is particularlyfrequent from bare steel tanks that are not protectedagainst corrosion. Even if a leakage is small, it can pose asignificant threat to groundwater quality.

Gasoline and petroleum products contain hydrocarboncomponents such as benzene, toluene, and xylene that arehighly soluble and mobile in groundwater and can be haz-ardous to humans if consumed. One gallon of gasoline isenough to render one million gallons of groundwater un-usable based on U.S. Environmental Protection Agency(EPA) drinking water standards (Noonan and Curtis1990). In addition, vapors and immiscible compounds

trapped in the pore spaces of the unsaturated zone con-tinue to feed groundwater with contaminants as precipi-tation moves into and through the subsurface or as thegroundwater table fluctuates (Dietz 1971; Van Dam1967).

SPILLS

Spills and discharges on the ground of chemical productscan migrate downward and contaminate groundwater.Spills and discharges vary from casual activities at indus-trial sites, such as leaks from pipes and valves, to accidentsinvolving aboveground storage tanks, railroad cars, andtrucks. The discharged chemicals are usually entrained bystormwater runoff and transported to the subsurfacewhere they reach the groundwater and degrade its quality(Scheville 1967).

Mining OperationsGroundwater can be contaminated by the drainage frommines and by oil and gas mining operations.

MINES

Drainage of both active and abandoned surface and un-derground mines can produce a variety of groundwaterpollution problems (Emrich 1969). Rainwater, particularlyacid rain, overexposed surface mines, and mine tailingsproduce highly mineralized runoff frequently referred toas acid mine drainage. This runoff can percolate into theground and degrade the quality of groundwater. In addi-tion, water seepage through underground mines can leachtoxic metals from exposed ores and raw materials and in-troduce them to groundwater (Barnes and Clarke 1964).Oxidation and leaching connected with coal mining pro-duce high iron and sulfate concentrations and low pH ingroundwater (Miller 1980).

OIL AND GAS

Oil and gas mining operations can also cause groundwa-ter contamination. These operations generate a substantialamount of wastewater, often referred to as brine. The brineis usually disposed of in surface impoundments or injectedin deep wells. Therefore, it can reach groundwater, and itsconstituents, such as ammonia, boron, calcium, dissolvedsolids, sodium, sulfate, and trace metals, can subsequentlydegrade the quality of groundwater (Fryberger 1975;Warner 1965).

Agricultural OperationsThe use of fertilizers and pesticides in agricultural opera-tions can contaminate groundwater.


FERTILIZERS

Fertilizers are the primary cause of groundwater contam-ination beneath agricultural lands. Both inorganic (chem-ically manufactured) and organic (from animal or humanwaste) fertilizers applied to agricultural lands provide nu-trients such as nitrogen, phosphorous, and potassium thatfertilize the land and stimulate plant growth. A portion ofthese nutrients usually leaches through the soil and reachesthe groundwater table. Phosphate and potassium fertiliz-ers are readily adsorbed on soil particles and seldom con-stitute a pollution problem. However, only a portion ofnitrogen is adsorbed by soil or used by plants, and the restis dissolved in water to form nitrates in a process callednitrification. Nitrates are mobile in groundwater and havepotential to harm infant human beings and livestock if con-sumed on a regular basis (Hassan 1974).

PESTICIDES

Pesticides, herbicides, and fungicides used for destroyingunwanted animal pests, plants, and fungal growth can alsocause groundwater contamination. When applied to landor disposed of in landfills, these chemicals degrade in theenvironment by a variety of mechanisms. However, theirparent compounds and their byproducts persist longenough to adversely impact the soil and groundwater(California Department of Water Resources 1968).

Other ActivitiesInteraquifer exchange and saltwater intrusion are twoother human activities that cause groundwater contami-nation.

INTERAQUIFER EXCHANGE

In interaquifer exchange, two aquifers are hydraulicallyconnected. Contamination occurs when contaminants aretransferred from a contaminated aquifer to a clean aquifer.Interaquifer exchange is common when a deep well pene-trates more than one aquifer to provide increased yield orwhen an improperly cased or abandoned well serves as adirect connection between two aquifers of different po-tential heads and different water quality. The hydraulicconnection (well or fractures) can allow contaminantsfrom aquifers with the greatest hydraulic head to move toaquifers of less hydraulic head (Deutsch 1961).

SALTWATER INTRUSION

Saltwater intrusion, in which saline water displaces ormixes with fresh groundwater, is another source ofgroundwater contamination. Saltwater intrusion is usually

caused when the hydrodynamic balance between the freshwater and the saline water is disturbed, such as when freshgroundwater is overpumped in coastal aquifers (TaskCommittee on Salt Water Intrusion 1969). Saltwater in-trusion can also occur when the natural barriers that sep-arate fresh and saline water are destroyed, such as in theconstruction of coastal drainage canals that enable tidalwater to advance inland and percolate into a freshwateraquifer (Todd 1974).

—Ahmed Hamidi

ReferencesBarnes, I., and F.E. Clarke. 1964. Geochemistry of groundwater in mine

drainage problems. U.S. Geotechnical Survey Prof. Paper 473-A.Bauer, W.J. 1974. Land treatment designs, present and future.

Proceedings of the International Conference on Land for WasteManagement, edited by J. Thomlinson. 343–346. Ottawa, Canada:National Research Council.

California Department Water Resources. 1968. The fate of pesticides ap-plied to irrigated agricultural land Bv11.174-1. Sacramento, Calif.

Council on Environmental Quality. 1980. The eleventh annual report ofthe Council on Environmental Quality. December.

———. 1981. Contamination of groundwater by toxic organic chemi-cals. (January). Washington, D.C.: U.S. Government Printing Office.

Deutsch, M. 1961. Incidents of chromium contamination of groundwa-ter in Michigan. Proceedings of Symposium on GroundwaterContamination, April. Cincinnati, Ohio: U.S. Dept. of Health,Education and Welfare.

Dietz, D.N. 1971. Pollution of permeable strata by oil components. InWater pollution by oil, edited by Peter Hepple, 128–142. Elsevier,Amsterdam.

Emrich, G.H., and G.L. Merritt. 1969. Effects of mine drainage ongroundwater. Groundwater 7, no. 3:27–32.

Field, R. et al. 1973. Water pollution and associated effects from streetsalting. EPA-R2-73-257. Cincinnati, Ohio: U.S. EPA.

Flower, F.B. 1976. Case history of landfill gas movement through soils,edited by E.J. Genetilli and J. Cirello, 177–184. Cincinnati, Ohio:U.S. EPA.

Fryberger, J.S. 1975. Investigation and rehabilitation of a brine-contam-inated aquifer. Groundwater 13, no. 2:155–160.

Hassan, A.A. 1974. Water quality cycle—reflection of activities of na-ture and man. Groundwater 12, no. 1:16–21.

Hughes, G. et al. 1971. Pollution of groundwater due to municipaldumps. Tech. Bull. no. 42. Ottawa, Ont.: Canada Dept. of Energy,Mines and Resources, Inland Waters Branch.

Mallmann, W.L., and W.N. Mack. 1961. Biological contamination ofgroundwater. Proceedings of Symposium on GroundwaterContamination. April. U.S. Department of Health, Education andWelfare.

Miller, D.W. 1980. Waste disposal effects on groundwater. Berkeley,Calif.: Premier Press.

Mohsen, M.F.N. 1975. Gas migration from sanitary landfills and asso-ciated problems. Ph.D. thesis, University of Waterloo.

Noonan, D.C., and J.T. Curtis. 1990. Groundwater remediation and pe-troleum: A guide for underground storage tanks. Chelsea, Mich.:Lewis Publishers.

Scheville, F. 1967. Petroleum contamination of the subsoil, a hydrolog-ical problem. In The joint problems of the oil and water industries,edited by Peter Hepple. 23–53. Elsevier, Amsterdam.

Task Committee on Salt Water Intrusion. 1969. Saltwater intrusion inthe United States. Journal of Hydraulics Division, ASCE 95, no.Hy5:1651–1669.


Terry, R.C., Jr. 1974. Road salt, drinking water, and safety. Cambridge,Mass.: Ballinger.

Todd, D.K. 1974. Salt water intrusion and its controls. Journal of AWWA66:180–187.

U.S. Environmental Protection Agency. 1974. Land application of sewageeffluents and sludge, selected abstracts. Washington, D.C.:Government Printing Office.

———. 1977. Waste disposal practices and their effects on groundwa-ter. Report to Congress, 81–107.

———. 1983. Process design manual: Land application of municipal

sludge. EPA-625/1-83-016. Cincinnati, Ohio: U.S. EPA, MunicipalEnvironmental Lab.

Van Dam, J. 1967. The migration of hydrocarbons in water bearing stra-tum. In The joint problems of the oil and water industries, edited byPeter Hepple. London: Institute of Petroleum.

Warner, D.L. 1965. Deep-well injection of liquid waste. Publ. no. 999-WP-21. U.S. Public Health Service.

Zanoni, A.E. 1972. Groundwater pollution and sanitary landfills—a crit-ical review. Groundwater 10, no. 1:3–16.


9.12FATE OF CONTAMINANTS IN GROUNDWATER

When a contaminant is introduced in the subsurface en-vironment, its fate and concentration are controlled by avariety of physical, chemical, and biochemical processesthat occur between the contaminant and the constituentsof the subsurface environment. A complete discussion andassessment of all these processes for all contaminants arebeyond the scope of this chapter. However, this section il-lustrates some of the most important processes for severalgroups of contaminants and the impact of these processeson the concentration and mobility of contaminants.

Organic ContaminantsThe physicochemical reactions that can alter the concen-tration of an organic contaminant in groundwater can begrouped into five categories as suggested by Arthur D.Little (1976) and Rao and Jessup (1982). These categoriesinclude (1) hydrolysis of the contaminant in water, (2) ox-idation–reduction, (3) biodegradation of the contaminantby microorganisms, (4) adsorption of the contaminant bythe soil, and (5) volatilization of the contaminant to theair present in the unsaturated zone. The relative impor-tance of each of these reactions depends on the physicaland chemical characteristics of the contaminant and onthe specific conditions of the subsurface environment.

HYDROLYSIS

Hydrolysis is a chemical reaction in which an organicchemical (RX) reacts with water or a hydroxide ion (OH)as follows:

R 2 X 1 H2O ® R 2 OH 1 H1 1 X2 9.12(1)

R 2 X 1 OH2 ® R 2 OH 1 X2 9.12(2)

During these reactions, a leaving group (X) is replaced bya hydroxyl ion (OH), and a new carbon–oxygen bond is

formed. The R represents the carbonium ion and the Xthe leaving group. Common leaving groups include halides(Cl2, Br2), alcohols (R—O2), and amines (R1R2N

2). Theacquisition of a new polar functional group increases thewater solubility of the organic chemical.

Examples of hydrolysis include the following (Valentine1986):

RCl 1 H2O ® ROH 1 H1 1 Cl2 9.12(3)an alkyl halide an alcohol

R1COOR2 1 H2O ®R2OH 1 R1COOH 9.12(4)an ester an alcohol 1 a carboxylic acid

RC(ON)R1R2 1 H2O ®RCOOH 1 R1R2NH 9.12(5)an amide a carboxylic acid1 an amine

RCH2CN 1 H2O ®RCH2COOH 1 NH3 9.12(6)a nitrile a carboxylic acid1 ammonia

The hydrolysis of organic chemicals in water is generallyconsidered first-order with respect to the organic chemi-cal’s concentration; thus, the rate of hydrolysis can be cal-culated with the following equation (Dragun 1988b):

k z C 5 2}d

d

C

t} 9.12(7)

or

k 5 }2.3

t

03} log 1}

C

C0

}2 9.12(8)

where:

k 5 rate constant, 1/timet 5 timeC0 5 initial concentration, ppmC 5 concentration at time t, ppm

The time needed for half of the concentration to react,half-life, can be calculated if k is known with use of thefollowing equation:

t1/2 5 }0.6

k

93} 9.12(9)

where t1/2 is equal to the half-life.Table 9.12.1 lists the hydrolysis half-lives for several or-

ganic chemicals. Half-lives vary from seconds to tens ofthousands of years. Certain compounds such as alkylhalides, chlorinated amides, amines, carbamates, esters,epoxides, phosphonic acid esters, phosphoric acid esters,and sulfones are potentially susceptible to hydrolysis(Dragun 1988b). Other compounds such as aldehydes,alkanes, alkenes, alkynes, aliphatic amides, aromatic hy-drocarbons and amines, carboxylic acids, and nitro frag-ments are generally resistant to hydrolysis (Dragun 1988b;Harris 1982).

For organic chemicals undergoing an acid- and base-catalyzed hydrolysis (in the case of acid or alkaline solu-tions), the total hydrolysis rate constant kT can be ex-pressed (Harris 1982; Mabey and Mill 1978) as

kT 5 kH [H1] 1 kN 1 kOH [OH2] 9.12(10)

where:

kT 5 total hydrolysis rate constantkH 5 rate constant for acid-catalyzed hydrolysis[H1] 5 hydrogen ion concentration[OH2] 5 hydroxyl ion concentrationkN 5 rate constant for neutral hydrolysiskOH 5 rate constant for base-catalyzed hydrolysis

Several other parameters can affect the rate of hydrol-ysis including temperature, the pH of the soil particle sur-faces, the presence of metals in soils, the adsorption of theorganic chemical, and the soil water content (Burkhardand Guth 1981; Konrad and Chesters 1969).

After the hydrolysis rate constant k is estimated, the be-havior of a compound can be modeled with a form of theadvection–dispersion equation. Equation 9.13(1), that in-cludes a first-order degradation term.

OXIDATION–REDUCTION

In organic chemistry, oxidation–reduction (redox) refersto the transfer of atoms rather than direct electrons as isthe case of inorganic chemistry. Oxidation of an organiccompound frequently involves a gain in oxygen and a lossin hydrogen atoms, and the reduction involves a gain inhydrogen and a loss in oxygen content. Oxidation–reduc-tion reactions greatly affect contaminant transport and areusually closely related to the microbial activity and the typeof substrates available to the organisms. Organic contam-inants provide the reducing equivalents for the microbes.After the oxygen in the subsurface environment is depleted,the most easily reduced materials begin to react and, alongwith the reduced product, dictate the dominant potential.

The occurrence of oxidation in the subsurface is a func-tion of the electrical potential in the reacting system(Dragun and Helling 1985). For oxidation to occur, the

potential of the soil system must be greater than that ofthe organic chemical. Soil reduction potentials can be gen-erally classified as follows:

Aerated soils: 10.8 to 10.4 volts

Moderately reduced soils: 10.4 to 10.1 volts

Reduced soils: 10.1 to 20.1 volts

Highly reduced soils: 20.1 to 20.5 volts

A number of organic chemicals can hydrolyze, oxidize,and reduce quickly and sometimes violently upon contactwith groundwater. Table 9.12.2 lists several classes of or-ganic chemicals that react rapidly and violently withgroundwater.

The hydrolysis, oxidation, or reduction of one organicchemical usually results in the synthesis of one or morenew organic chemicals. Organic chemistry textbooks iden-tify the basic reaction products. Soil minerals can signifi-cantly influence the chemical structure of reaction prod-ucts. In addition, certain organic chemicals can formsignificant amounts of residues that bind to the soil.Examples of such chemicals include anilines, phenols, tri-azines, urea herbicides, carbamates, organophosphates,and cyclodiene insecticides (Sax 1984).

BIODEGRADATION

Biodegradation of an organic chemical in soil is the mod-ification or decomposition of the chemical by soil mi-croorganisms to produce ultimately microbial cells, car-bon dioxide, oxygen, and water.

Soil serves as the home for numerous microorganismscapable of degrading organic chemicals. The most pre-dominant microorganisms in soil include bacteria, actino-mycetes, and fungi. One gram of surface soil can containfrom 0.1 to 1 billion cells of bacteria, 10 to 100 millioncells of actinomycetes, and 0.1 to 1 million cells of fungi(Dragun 1988b). The microorganism population in soilsis generally greatest in the surface horizons where tem-perature, moisture, and energy supply is favorable for theirgrowth. As the depth increases, the number of aerobic mi-croorganisms decreases; however, anaerobic microorgan-isms can exist depending on availability of nutrients andorganic material.

The biodegradation of an organic chemical by a mi-croorganism is catalyzed by enzymes which are producedas part of the metabolic activity of the living organism.The biotransformation occurs either inside the microor-ganism via intracellular enzymes or outside the microor-ganism by the action of extracellular enzymes. After an or-ganic chemical and an enzyme collide, an enzyme–chemicalcomplex forms. Then, depending on the alignment be-tween the functional groups of the chemical and the en-zyme, a reaction product (modified or decomposed organicchemical) is formed by the removal of one or more func-tional groups by oxidation or reduction reactions (Dragun


1988a). Some typical biochemical reactions are as follows(Valentine and Schnoor 1986):

Decarboxylation:R—OCH3 ® ROH 1 CO2 9.12(11)

Oxidation of an amino group:R—NH2 ® RNO2 9.12(12)

Reductive dehalogenation:R—CCl2—R ® RCHClR 1 Cl2 9.12(13)

Hydrolysis:R—CH2CN ® RCHONH3 9.12(14)

An organic chemical has two levels of the biodegrada-tion. Primary degradation refers to any biologically in-


TABLE 9.12.1 HYDROLYSIS HALF-LIVES FOR VARIOUS ORGANIC CHEMICALS

Chemical Half-life Chemical Half-life

acetamide 3,950 y ethion 9.9 datrazine 2.5 h N-ethylacetamide 70,000 yazirdine 154 d ethyl acetate 136 dbenzoyl chloride 16 s ethyl butanoate 5.8 ybenzyl bromide 1.32 h ethyl trans-buteonate 17 ybenzyl chloride 15 h ethyl difluoroethanoate 23 mbenzylidene chloride 0.1 h ethyl dimethylethanoate 9.6 ybromoacetamide 21,200 y ethyl methylthioethanoate 87 dbromochloromethane 44 y ethyl phenylmethanoate 7.3 ybromodichloromethane 137 y ethyl propanoate 2.5 ybromoethane 30 d ethyl propenoate 3.5 y1-bromohexane 40 d ethyl propynoate 17 d3-bromohexane 12 d ethyl pyridylmethanoate 0.41 ybromomethane 20 d fluoromethane 30 ybromomethylepoxyethane 16 d 2-fluor-2-methylpropane 50 d1-bromo-3-phenylpropane 290 d hydroxymethylpropane 28 d1-bromopropane 26 d iodoethane 49 d3-bromopropene 12 h iodomethane 110 dchloroacetamide 1.46 y 2-iodopropane 2.9 dchlorodibromoethane 274 y 3-iodopropene 2.0 dchloroethane 38 d isobutyramide 7,700 ychlorofluoriodomethane 1.0 y isopropyl bromide 2.0 dchloromethane 339 d isopropyl ethanoate 8.4 ychloromethylepoxyethane 8.2 d malathion 8.1 d2-chloro-2methylpropane 23 s methoxyacetanide 500 y2-chloropropene 2.9 d N-methylacetamide 38,000 y3-chloropropene 69 d methyl chloroethanoate 14 hcyclopentanecarboxamide 5,500 y methyl dichloroethanoate 38 mdibromoethane 183 y methylepoxyethane 14.6 d1,3-dibromopropane 48 d methyl parathion 10.9 ddichloroacetamide 0.73 y methyl trichloroethanoate ,3.6 mdichloroiodomethane 275 y monomethyl phosphate 1.0 ddichloromethane 704 y parathion 17 ddichloromethyl ether 25 s phenyl dichloroethanoate 3.7 mdiethyl methylphosphonate 990 y phenyl ethanoate 38 ddimethoxysulfone 1.2 m phosphonitrilic hexamide 46 d1,2-dimethylepoxyethane 15.7 d propadienyl ethanoate 110 d1,1-dimethylepoxyethane 4.4 d ronnel 1.6 ddiphenyl phosphate 20.6 d tetrachloromethane 7,000 y (1 ppm)epoxyethane 12 d tribromomethane 686 y3,4 epoxycyclohexene 6 m trichloroacetamide 0.23 y3,4 epoxycyclooctane 52 m trichloromethane 3,500 y1,3-epoxy-1-oxopropane 3.5 trichlorethylbenzene 19 s

triethyphosphate 5.5 ytri(ethylthio)phosphate 8.5 y

Source: J. Dragun, 1988, The soil chemistry of hazardous materials (Silver Springs, MD: Hazardous Materials Control Research Institute).d 5 days, h 5 hours, s 5 seconds, m 5 minutes, y 5 years.

duced structural alteration in the organic chemical. Ulti-mate biodegradation refers to the degradation of the or-ganic chemical into carbon dioxide, oxygen, water, andother inorganic products.

Primary biodegradation of an organic chemical can gen-erate a variety of degradation products that can contami-nate groundwater. For example, the degradation oftrichloroethylene (TCE) can lead to dichloroethylenes(DCEs), dichloroethanes (DCAs), vinyl chloride, andchloroethane (Dragun 1988b; Alexander 1981; Goringand Hamaker 1972). The degradation of cyclic hydrocar-bons can lead to aliphatic hydrocarbons, and aliphatic hy-drocarbons can be converted in successive reactions intoalcohols, aldehydes, and then aliphatic acids (Tabak et al.1981).

The biodegradation of many organic chemicals is gen-erally first-order with respect to the organic chemical’s con-centration (Scow 1982). As a result, the biodegradationrate constant can be calculated with use of the followingfirst-order equation as in hydrolysis:

k z C 5 2}d

d

C

t} 9.12(15)

or

k 5 }2.3

t

03} log 1}

C

C0

}2 9.12(16)

where:

k 5 rate constant, 1/timet 5 timeC0 5 initial concentration, ppmC 5 concentration at time t, ppm

The time needed for half of the concentration to react,half-life, can be calculated if k is known with use of thefollowing equation:

t1/2 5 }0.6

k

93} 9.12(17)

where t1/2 is equal to half-life.Table 9.12.3 lists the biodegradation rates for many

pesticides; the biodegradation rates for other organic chem-icals are in Dragun (1988b). However, note that the esti-mate of biodegradation rates of organic chemicals may notbe accurate. Biodegradation rates can be affected by manyfactors such as pH, temperature, water content, carboncontent, clay content, oxygen, nutrients, microbial popu-lation, acclimation, and concentration. Most of these fac-tors are interrelated. For example, the pH can affect boththe availability of a substrate as well as the compositionof the microbial community.

After the degradation rate constant k is estimated, thebehavior of a specific compound can be modeled with useof a form of the advection–dispersion equation, Equation9.13(1), that includes a first-order degradation term.

ADSORPTION

Adsorption is the bonding of an organic chemical to the soilmineral surfaces (clay) or to the organic matter surfaces.The bonding is usually temporary and is accomplished byionic, ligand, dipole, hydrogen, or Van der Waal’s bonds.Adsorption is important in the movement of organic chem-icals in groundwater because it decreases the mobility andretards the migration of an organic chemical in groundwa-ter. Furthermore, the adsorbed portion of an organic chem-ical may not be available in solution for other chemical re-actions such as hydrolysis and biodegradation.

The degree and extent of adsorption of an organicchemical to soil is determined by the chemical’s structureand the soil’s physical and chemical characteristics.Organic chemicals with large molecular structures, such asPCBs, PAHs, toluene, and dichlorodiphenyl trichloro-ethane (DDT), tend to be extensively adsorbed onto soil(Landrum et al. 1984). Organic chemicals with positivecharges, such as the herbicides paraquat and diquat, arereadily adsorbed onto the cation exchange sites (clay min-


TABLE 9.12.2 ORGANIC CHEMICALS THAT MAYRAPIDLY REACT WITH SOIL WATERAND GROUNDWATER

acetic anhydride methacrylic acidacetyl bromide 2-methylaziridneacetyl chloride methyl isocyanateacrolein methyl isocyanoacetateacrylonitrile oxopropanedinitrile3-aminopropiononitrile perfluorosilanesbis(difluoroboryl)methane peroxyacetic acidbutyldichloroborane peroxyformic acidcalcium cyanamide peroxyfuroic acid2-chloroethylamine peroxpivalic acidchlorosulfonyl isocyanate peroxytrifluoriacetic acidchlorotrimethylsilane phenylphosphonyl dichloridecyanamide phosphorus tricyanide2-cyanoethanol pivaloyloxydiethylboranecyanoformyl chloride potassium bis(propynyl)palladatecyanogen chloride potassium bis(propynyl)platinatedichlorodimethylsilane potassium diethynylplatinatedichlorophenylborane potassium hexaethynylcobaltatedicyanoacetylene potassium methanediazoatediethylmagnesium potassium tert-butoxidediethylzinc potassium tetracyanotitanatediketene potassium tetraethynylnickelatedimethyaluminum chloride propenoic aciddimethylmagnesium sulfur trixoide-dimethylformamidedimethylzinc sulfinylcyanamidediphenylmagnesiuim 2,4,6-trichloro-1,3,5-triazine2,3-epoxypropionaldehyde trichlorovinylsilane

oxime triethoxydialuminum tribromideN-ethyl-N-propylcarbamolyl vinyl acetate

chlorideglyoxalisopropylisocyanide dichloridemaleic anhydride

Source: J. Dragun, 1988, The soil chemistry of hazardous materials (SilverSprings, MD: Hazardous Materials Control Research Institute).

eral surfaces). In addition, the adsorption of organic chem-icals depends on the organic matter content of the soil.The relationship between the organic content of soil andthe adsorption coefficient of organic chemicals is generallylinear for soils with an organic carbon content greater than0.1 (Hamaker and Thompson 1972).

The adsorption process is usually reversible. At equi-librium, the adsorption coefficient, which is the rate atwhich the dissolved organic chemical in water transfersinto the soil, can be described with the linear Freundlichisotherm equation as

Kd 5 }C

C

w

s} 9.12(18)

where:

Kd 5 distribution coefficientCs 5 concentration adsorbed on soil surfaces, ug/gCw 5 concentration in water, ug/ml

Other nonlinear isotherm equations are also used(Lyman 1982), such as:

Kd 5 }C

C

w1

s

/n} 9.12(19)

where n is a constant usually between 0.7 and 1.1.

As in Equations 9.12(18) and 9.12(19), the distributionor adsorption coefficient Kd is directly proportional to theorganic carbon content of the soil; thus, Kd can be writ-ten as

Kd 5 }K

fo

o

c

c} 9.12(20)

where:

Koc 5 normalized adsorption coefficientfoc 5 soil organic carbon content

The normalized adsorption coefficient can be estimatedfrom the organic chemical’s water solubility or octanol wa-ter partition coefficient with use of regression equations(Dragun 1988b), such as

log(Koc) 5 a z log(S) 1 b 9.12(21)

log(Koc) 5 c z log(Kow) 1 d 9.12(22)

where:

S 5 water solubilityKow 5 octanol water partition coefficienta,b,c,d 5 coefficients that depend on the organic chemi-

cal

Table 9.12.4 lists the adsorption coefficient Koc for sev-eral organic chemicals. The regression coefficients a, b, c,and d for several chemicals are in Brown and Flagg (1981),Briggs (1973), and Keneya and Goring (1980). Therefore,after the distribution coefficient Kd is estimated, the effectof adsorption on the mobility of a specific compound canbe calculated with use of a form of the advection–disper-sion equation, Equations 9.13(1) and 9.13(3), that includesthe retardation factor R.

VOLATILIZATION

Volatilization is the loss of chemicals in vapor form fromthe soil water (liquid phase) or the soil surfaces (solidphase) to the soil air (gas phase) of the unsaturated zone.Only the first type of volatilization, from the liquid phaseto the gas phase, is discussed in this section. The volatiliza-tion from the solid phase to the gas phase is relatively smalland usually neglected. However, information on this typeof volatilization is presented by Mayer, Letey, and Farmer(1974); Baker and Mackay (1985); and Jury, Farmer, andSpencer (1984).

The extent of volatilization of an organic chemical fromwater to the soil air can be determined by Henry’s lawwhich states that when a solution becomes dilute, the va-por pressure of a chemical is proportional to its concen-tration (Thomas 1982) as

Ca 5 H z Cw 9.12(23)


TABLE 9.12.3 BIODEGRADATION RATECONSTANTS FOR ORGANICCOMPOUNDS IN SOIL

Compound k (Day21)

Aldrin, Dieldrin 0.013Atrazine 0.019Bromacil 0.0077Carbaryl 0.037Carbofuran 0.047DDT 0.00013Diazinon 0.023Dicamba 0.022Diphenamid 0.123b

Fonofos 0.012Glyphosate 0.10Heptachlor 0.011Lindane 0.0026Linuron 0.0096Malathion 1.4Methyl parathion 0.16Paraquat 0.0016Phorate 0.0084Picloram 0.0073Simazine 0.014TCA 0.059Terbacil 0.015Trifluralin 0.0082,4-D 0.0662,4,5-T 0.035

Source: W. Mabey and T. Mill, 1978, Critical review of hydrolysis of or-ganic compounds in water under environmental conditions, Jour. Phys. Chem.Ref. Data 17, no. 2:383–415.

where:

Ca 5 concentration of the chemical in airH 5 Henry’s law constantCw 5 concentration of the chemical in water

Henry’s law constant for a chemical can be calculatedwith the following equation:

H 5 }P

7v

6

z

0

M

z Sw

} 9.12(24)

where:

Pv 5 vapor pressure of the chemical in mmHg

Mw 5 molecular weight of the chemicalS 5 solubility in mg/l

Published texts report Henry’s law constant in variousunits such as atm-m3/mole, atm-cm3/g, or dimensionlessdepending on the units used for Ca and Cw. Table 9.12.5lists Henry’s constants for several organic chemicals.According to Lyman and others (1982), if H is less than1027 atm-m3/mole, the substance has a low volatility. IfH is less than 1025 but greater than 1027 atm-m3/mole,the substance volatilizes slowly. However, the volatiliza-tion becomes an important transfer mechanism when H isgreater than 1025 atm-m3/mole.

Several soil characteristics affect the volatilization of or-ganic chemicals in groundwater. Volatilization decreasesas the soil porosity decreases or as the soil water contentincreases. Soils with high clay content tend to have a highwater content and hence low volatilization (Jury 1986).

Inorganic ContaminantsComprehensive information on the behavior of most in-organic chemicals in groundwater is limited. Agriculturallyimportant compounds have been studied for many years;however, inorganic compounds such as metals have onlyrecently begun to attract widespread interest as ground-water and soil contamination become a concern. This sec-tion illustrates some of the most important processes forseveral groups of inorganic contaminants and the impactof these processes on the concentration and mobility ofcontaminants.

Inorganic constituents in the subsurface environmentcan be classified into the following four categories: nutri-ents, acids and bases, halides, and metals. The origin andsources of these inorganics are discussed in Section 9.11.

NUTRIENTS

Nutrients such as nitrogen, phosphorous, and sulfur areessential for plant and microorganism growth. They areeither applied to the land surface to increase its fertility ordiscarded with waste streams that contain appreciableamounts of these nutrients. These nutrients, however, canhave appreciable concentrations that can leach into theground and adversely affect the quality of groundwater.

Nitrogen (N) is found in waste, soil, and the atmos-phere in various forms such as ammonia, ammonium, ni-trite, nitrate, and molecular nitrogen. Nitrogen is convertedto ammonium (NH1

4) by a process called ammonification.Because of its positive charge, ammonium can be held inthe soil on cation exchange sites. Ammonium can also beconverted temporarily to nitrite (NO2

2) and then to nitrate(NO2

3) by aerobic nitrifying organisms through a processcalled nitrification. Ammonification and nitrification nor-mally occur in the unsaturated zone where microorgan-


TABLE 9.12.4 MEASURED KOC VALUES FORVARIOUS ORGANIC CHEMICALS

Chemical Koc Chemical Koc

acetophenone 35 ipazine 1,660alachlor 190 isocil 130aldrin 410 isopropalin 75,250ametryn 392 leptophos 9,3006-aminochrysene 162,900 linuron 813anthracene 26,000 malathion 1,778asulam 300 methazole 2,620atrazine 148 methomyl 160benefin 10,700 methoxychlor 80,000alpha-BHC 1995 methylparathion 5,129beta-BHC 1995 metobromuron 602,29-biquinoline 10,471 metribuzin 95bromacil 72 monolinuron 200butraline 8,200 monuron 100carbaryl 229 napthalene 1,300carbofuran 105 napropamide 680carbophenothion 45,400 neburon 2,300chloramben 21 nitralin 960chlorobromuron 460 nitrapyrin 458chloroneb 1159 norfluorazon 1,914chloroxuron 3200 oxadiazon 3,241chloropropham 589 parathion 4,786chlopyrifos 13,490 pebulate 630crotoxyphos 170 phenathrene 23,000cyanazine 200 phenol 27cycloate 345 phorate 3,2002,4-D 57 picloram 17DBCP 129 profluralin 8,600p,p9-DDT 129 prometon 350diallate 1,900 prometryn 48diamidaphos 32 pronamide 200dicamba 0.4 propachlor 265dichlobenil 235 propazine 158dinitramine 4,000 propham 51dinoseb 124 pyrazon 120dipropetryn 1,170 pyrene 62,700disulfoton 1,780 pyroxychlor 3,000diruon 398 silvex 2,600DMSA 770 simazine 135EPTC 240 2,4,5-T 53ethion 15,400 tebuthiuron 620fenuron 27 terbacil 51

Source: J. Dragun, 1988, The soil chemistry of hazardous materials (SilverSprings, MD: Hazardous Materials Control Research Institute).

isms and oxygen are abundant, but nitrate can be readilyleached from the soil into groundwater where it may pre-sent a health hazard; nitrate is highly mobile in ground-water because of its negative charge. Denitrification is aprocess whereby NO2

3 is reduced to nitrous oxide (N2O)and elemental nitrogen (N2) by facultative anaerobic bac-teria (Downing, Painter, and Knowles 1964; Freeze andCherry 1979; Bemner and Shaw 1958).

Phosphorous (P) is found in organic waste, rock phos-phate quarries, fertilizers, and pesticides in concentrationshigh enough to potentially leach into groundwater. Thedecomposition of organic waste and dissolution of inor-ganic fertilizers provide soluble phosphorous, soluble or-thophosphate, and a variety of condensed phosphates,tripolyphosphates, adsorbed phosphates, and crystallizedphosphates (U.S. EPA 1983). The hydrolysis and mineral-ization of these products provide soluble phosphate whichcan be used by plants and microorganisms, adsorbed tosoil particles, or leached to groundwater. Although phos-phorous is not a harmful constituent in drinking water, itspresence in groundwater is environmentally significant ifthe groundwater discharges to a surface water body wherephosphorous can produce algae growth and cause eu-

trophication of the aquatic system (Freeze and Cherry1979).

Sulfur (S) is found in appreciable amounts in wastestreams from kraft mills, sugar refining, petroleum refin-ing, and copper and iron extraction facilities (Overcashand Pal 1979). Aerobic bacteria can oxidize the reducedforms of sulfur to form sulfate which can be highly ad-sorbed to soil when the cation adsorbed on the clay is alu-minum; moderately adsorbed when the cation is calcium;and weakly adsorbed when the cation is potassium (Tisdaleand Nelson 1975). Leaching losses of sulfur to ground-water can be large because of the anionic structure of sul-fur and the solubility of most of its salt. Leaching is great-est when monovalent cations such as potassium andsodium predominate; moderate when calcium and man-ganese predominate; and minimal when the soil is acidicand appreciable levels of exchangeable iron and aluminumare present (Tisdale and Nelson 1975).

ACIDS AND BASES

Industrial liquid wastes are comprised of large volumes ofinorganic acids and bases that can alter the soil’s proper-


TABLE 9.12.5 VALUES OF HENRY’S LAW CONSTANT FOR SELECTED CHEMICALS

Low Volatility H H9 High Volatility H H9(H , 3 3 1027) attm-m3 (non-dim.) (H , 1023) attm-m3 (non-dim.)

3-Bromo-1-propanol 1.1 3 1027 4.6 3 1026 Ethylene dichloride 1.1 3 1023 2.4 3 1022

Diedrin 1.2 3 1027 8.9 3 1026 Naphthalene 1.15 3 1023 4.9 3 1022

Biphenyl 1.5 3 1023 6.8 3 1022

Middle Range Aroctor 1254 2.7 3 1023 1.6 3 1021

(3 3 1027 , H , 1023) Methylene chloride 2.3 3 1023 1.3 3 1021

Aroctor 1248 3.5 3 1023 1.6 3 1021

Lindane 4.8 3 1027 2.2 3 1025 Chlorobenzene 3.7 3 1023 1.65 3 101

m-Bromonitrobenzene 1.6 3 1026 7.4 3 1025 Chloroform 4.7 3 1023 2.0 3 1021

Pentachlorophenol 3.4 3 1026 1.5 3 1024 o-Xylene 5.1 3 1023 2.2 3 1021

4-tert-Butylphenol 9.1 3 1026 3.8 3 1024 Benzene 5.5 3 1023 2.4 3 1021

Triethylamine 1.3 3 1025 5.4 3 1024 Toluene 6.6 3 1023 2.8 3 1021

Aldrin 1.4 3 1025 6.1 3 1024 Aroclor 1260 7.1 3 1023 3.0 3 1021

Nitrobenzene 2.2 3 1025 9.3 3 1024 Perchloroethylene 8.3 3 1023 3.4 3 1021

Epichlorohydrin 3.2 3 1025 1.3 3 1023 Ethyl benzene 8.7 3 1023 3.7 3 1021

DDT 3.8 3 1025 1.7 3 1023 Trichloroethylene 2.1 3 1022 4.2 3 1021

Phenanthrene 3.9 3 1025 1.7 3 1023 Mercury 1.1 3 1022 4.8 3 1021

Acenaphthene 1.5 3 1024 6.2 3 1023 Methyl bromide 1.3 3 1022 5.6 3 1021

Acetylene tetrabromide 2.1 3 1024 8.9 3 1023 Cumene (isopropyl) 1.5 3 1022 6.2 3 1021

Aroclor 1242 5.6 3 1024 2.4 3 1022 1,1,1-Trichloroethane 1.8 3 1022 7.7 3 1021

Ethylene dibromide 6.6 3 1024 2.8 3 1022 Carbon tetrachloride 2.3 3 1022 9.7 3 1021

Methyl chloride 2.4 3 1022 9.7 3 1021

Ethyl bromide 7.3 3 1022 3.1Vinyl chloride 2.4 992,2,4-Trimethyl pentane 3.1 129n-Octane 3.2 136Fluorotrichloromethane 5.0 —Ethylene .8.6 ,360

Source: R.G. Thomas, 1982, Volatilization from water, In Handbook of chemical property estimation methods (New York: McGraw-Hill, Inc.).

ties. Acids can increase the amount of aluminum (Al), iron(Fe), and other cations in the water phase of the soil sys-tem as the hydrogen ion (H1) cation competes for cationexchange sites. If significant amounts of H1 are present,they can dissolve the more acid-solid minerals, releasingcations which are previously fixed to the mineral structureinto the water phase (Dragun 1988b). In addition, acidscan cause the dissolution of some of the clay minerals andgenerally increase soil permeability. Bases can increase theamount of cations in the water phase by dissolving themore base-soluble soil minerals. Bases can also cause thedissolution of some of the soil’s predominant clay miner-als and generally decrease soil permeability.

HALIDES

Halides are the stable anions of the highly reactive halo-gens: fluoride (F), chloride (Cl), bromine (Br), and iodine(I). Halides occur naturally in soils and are also present inmany industrial waste streams.

Fluoride is present in phosphatic fertilizers, hydrogenfluoride, fluorinated hydrocarbons, and certain petroleumrefinery waste. The leaching losses and mobility of fluo-ride can be large because of the anionic structure of fluo-ride and the solubility of some of its salt (Bemner and Shaw1958). Sodium salts of fluoride (NaF) are soluble and re-sult in high soluble fluoride levels in soils low in calcium.Calcium salts of fluoride (CaF2), however, are relativelyinsoluble and limit the amount of fluoride leached togroundwater. Fluoride solubility depends on the kind andrelative quantity of cations present in soil that has formedsalts with the fluoride ion (F2). Fluorosis disease can oc-cur in animals who consume water containing 15 ppm offluoride (Lee 1975).

Chloride (Cl) is present in chlorinated hydrocarbon pro-duction and chlorine gas production wastes as well as otherwastes. Chloride is soluble and mobile in groundwater be-cause of its anionic structure.

Bromide (Br) is present in synthetic organic dyes, mixedpetrochemical wastes, photographic supplies, and phar-maceutical and inorganic wastes. Other forms of bromidesuch as bromate and bromic acid occur naturally in soilsat smaller concentrations. Most bromide salts (CaBr,MgBr, NaBr, and Kbr) are soluble and readily leachableinto water percolating through the soil and down togroundwater (U.S. EPA 1983).

Iodine (I) is present in pharmaceutical and chemical in-dustrial wastes. Iodine is only slightly water soluble andtends to be retained in soils by forming complexes withorganic matter and being fixed to phosphates and sulfates.

METALS

Metals are found in industrial wastes in a variety of forms.When these metals are introduced into the subsurface en-vironment, they can react with water and soil in several

physicochemical processes to produce appreciable con-centrations that affect the quality of groundwater. Themost important processes that affect the concentration andmobility of metals in groundwater include filtration, pre-cipitation, complexation, and ion exchange.

Filtration occurs when dissolved and solid matter aretrapped in the pore spaces clogging the pore spaces anddecreasing the permeability of the soil system (Dragun1988b).

Precipitation occurs when metal ions react with waterto form reaction products which precipitate in soil as ox-ide and oxyhydroxide minerals or form oxyde and oxy-hydroxide coatings on soil minerals. Precipitation of met-als as hydroxides, sulfides, and carbonates is common(Dragun 1988b).

Complexation involves the formation of soluble,charged or neutral complexes between metal ions and in-organic or organic anions called ligands. The complexesformed influence the mobility and concentration of themetal in groundwater. For example, the mobility of zincin groundwater is affected by the formation of complexspecies between the zinc ion and inorganic anions presentin the water, such as HCO2

3, CO322, SO4

22, Cl2, F2, andNO2

3 (Freeze and Cherry 1979). The complexation ofcobalt-60 ions by synthetic organic compounds enhancesits mobility in groundwater (Killey et al. 1984). Othermetal species are reported to be highly mobile in ground-water after soluble complexes are formed with humic sub-stances or organic solvents (Bradbent and Ott 1957;Griffin and Chou 1980).

The predominant complex species in an aqueous solu-tion are influenced by the redox and pH of the soil. Therelationship between the redox, pH, and the complexspecies is commonly expressed in Eh–pH diagrams for eachmetal; Eh is the electronic potential. Figure 9.12.1 showsan example of an Eh–pH diagram for mercury. Methodsfor calculating Eh–pH diagrams are discussed by severalauthors (Brookings 1980; Garrells and Christ 1965;Verink 1979).

Using Eh–pH diagrams, environmental engineers canqualitatively determine the most important complexesformed by the metal in water and estimate the concentra-tion and mobility of the metal in groundwater. The con-centration of cations reported in chemical analyses ofgroundwater normally represents the total concentrationof each element in water. However, most cations exist inmore than one molecular or ionic form. These forms canhave different valences and, therefore, different mobilitiesdue to different affinities to sorption and different solu-bility controls.

Adsorption is another process affecting the concentra-tion and mobility of metals in groundwater. Positive ad-sorption involves the attraction of metal cations in waterby negatively charged soil particles. Therefore, adsorptioncan decrease the concentration of dissolved metals in wa-ter and retard their movement. The cation exchange ca-


pacity (CEC) of a soil, defined as the amount of cationsadsorbed by the soil’s negative charges, is usually expressedas milliequivalents (meq) per 100 grams of soil. In general,clay soils and humus have a higher CEC than other soils.

Some cations are more attracted to a soil surface thanothers based on the size and charge of their molecule. Forexample the Cu21 cation in water can displace and replacea Ca21 cation present at the soil surface through a processknown as ion exchange. Also, trivalent cations are prefer-entially adsorbed over divalent cations which are prefer-entially adsorbed over monovalent cations. The release of

ions by exchange processes can aggravate a contaminationproblem. For example, increases in water hardness result-ing from the displacement of calcium and magnesium ionsfrom geological materials by sodium or potassium in land-fill leachate have been documented (Hughes, Candon, andFarvolden 1971).

The cation exchange is reversible, and its extent can bedescribed by the adsorption or distribution coefficient(Dragun 1988b) as

Kd 5 }C

C

w

s} 9.12(25)

where:

Kd 5 adsorption or distribution coefficientCs 5 concentration adsorbed on soil surfaces (ug/g of

soil)Cw 5 concentration in water (ug/ml)

Table 9.12.6 lists the adsorption coefficients for sev-eral metals. The greater the coefficient Kd, the greaterthe extent of adsorption. Furthermore, changes in metalconcentration, as well as pH, can have a significant ef-fect on the extent of adsorption as shown in Figure9.12.2.

A negative adsorption occurs when anions (negativelycharged metal ions) are repulsed by negative soil particlecharges. This repulsion causes high mobility and migra-tion of anions in water. This process is also known as an-ion exclusion.

—Ahmed Hamidi


FIG. 9.12.1 Stability fields of solid phases and aqueous species of mercury asa function of pH and Eh at 1 bar total pressure. (Reprinted from J.D. Hem, 1967,Equilibrium chemistry of iron in groundwater, In Principles and applications ofwater chemistry, edited by S.D. Faust and J.V. Hunter, New York: John Wileyand Sons.)

0 2 4 6 8 10 12 14215

210

25

0

5

10

15

20

pE

pH0 2 4 6 8 10 12 14

215

210

25

0

5

10

15

20

pE

pH

HgS(s)Hg˚(I)

HgO(s)

Hg2Cl2(s)

HgCl2(s)

Normal pH rangeof groundwater

HgCl2˚(aq)

1023 Cl

1021 Cl

Hg221

(aq)

Hg(OH)2˚(aq)

Hg(HS)2˚(aq)Hg̊ (aq)

Hg˚(aq)

HgS222(aq)

100

80

60

40

20

00 2 4 6 8

CdCuPb

Fe(III)

pH

% A

DS

OR

BE

D

FIG. 9.12.2 Adsorption of metal ions on amorphous silica asa function of pH. (Reprinted from U.S. Environmental ProtectionAgency, 1989, Transport and fate of contaminants in the sub-surface, Seminar Publication EPA/625/4-89/019, Cincinnati: U.S.EPA.)

ReferencesAlexander, M. 1981. Biodegradation of chemicals of environmental con-

cern. Science 211:132–138.Arthur D. Little, Inc. 1976. Physical, chemical, and biological treatment

techniques for industrial wastes. Report to U.S. EPA, Office of SolidWaste Management Programs, PB-275-054/56A Vol. 1 and PB-275-278/1GA Vol. 2.

Baker, L.W., and K.P. Mackay. 1985. Screening models for estimatingtoxic air pollution near a hazardous waste landfill. Journal of the AirPollution Control Association 35, no. 11:1190–1195.

Bemner, J.M., and K. Shaw. 1958. Denitrification in soils-factors affect-ing denitrification. Journal of Agricultural Science 51, no. 1:22–52.

Bradbent, F.E., and J.B. Ott. 1957. Soil organic matter: Metal com-plexes—factors affecting various cations. Soil Science 83:419–427.

Briggs, G.G. 1973. A simple relationship between soil adsorption of or-ganic chemicals and their octanaol/water partition coefficients.Proceedings of the 7th British Insecticide and Fungicide Conference.Vol. 1. Nottingham, Great Britain: The Boots Company, Ltd.

Brookings, D.G. 1980. Eh–pH diagrams for elements of interest at theOklo Natural Reactor at 25°C, 1 bar pressure and 200°C, 1 bar pres-sure. Report to Los Alamos National Laboratory, CNC-11.

Brown, D.S., and E.W. Flagg. 1981. Empirical prediction of organic pol-lutant adsorption in natural sediments. Journal of EnvironmentalQuality 10:382–386.

Burkhard, N., and J.A. Guth. 1981. Chemical hydrolysis of 2-chloro-4,6-bis (alkylamino)-1,3,5-triazine herbicides and their breakdown insoils under the influence of adsorption. Pestic. Sci 12:45–52.

Downing, A.L., H.A. Painter, and C. Knowles. 1964. Nitrification in theactivated sludge process. Journal and Proceedings of the Institute ofSewage Purification, Part 2.

Dragun, J. 1988a. Microbial degradation of petroleum products in soil.Proceedings of a Conference on Environmental and Public Health,Effects of Soils Contaminated with Petroleum Products, October,New York: John Wiley and Sons.

———. 1988b. The soil chemistry of hazardous materials. Silver Springs,Md.: Hazardous Materials Control Research Institute.

Dragun, J., and C.S. Helling. 1985. Physicochemical and structural re-lationships of organic chemicals undergoing soil and clay catalyzedfree-radical oxidation. Soil Science 139:100–111.

Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Englewood Cliffs,N.J.: Prentice-Hall.

Garrells, R.M., and C.L. Christ. 1965. Minerals, solutions, and equilib-ria. New York: Harper and Row.

Goring, C.A.J., and J.W. Hamaker. 1972. Organic chemicals in the soilenvironment. Vols. 1 and 2. New York: Marcel Dekker.

Griffin, R.A., and S.F.J. Chou. 1980. Attenuation of polybrominatedbiphenyls and hexachlorobenzene by earth materials. EnvironmentalGeology Notes 87, Illinois State Geological Survey. Urbana, Ill.

Hamaker, J.W., and J.M. Thompson. 1972. Adsorption. In Organicchemicals in the soil environment, edited by C.A.I. Goring and J.W.Hamaker. New York: Marcel Dekker.

Harris, J.C. 1982. Rate of hydrolysis. In Handbook of chemical prop-erty estimation methods. New York: McGraw-Hill.

Hughes, G.M., R.A. Candon, and R.N. Farvolden. 1971. Hydrogeologyof solid waste disposal sites in northern Illinois. Solid WasteManagement Series, report SW-124. U.S. EPA.

Jury, W.A. 1986. Volatilization from soil. In Vadoze modeling of or-ganic contaminants, edited by Stephen Hern and Susan Melancon,159–176. Chelsea, Mich.: Lewis Publishers.

Jury, W.A., W.J. Farmer, and W.F. Spencer. 1984. Behavior assessmentmodel for trace organics in soils. Journal of Environmental Quality13, no. 4.

Keneya, E.E., and C.A.I. Goring. 1980. Relationship between water sol-ubility, soil-sorption, octanol-water partitioning, and bioconcentra-tion of chemicals in biota. In Aquatic toxicology, ASTM STP 707.Philadelphia, Pa.: ASTM.

Killey, R.W. et al. 1984. Subsurface cobalt-60 migration from a low-levelwaste disposal site. Environmental Science Technology 18, no.3:148–156.

Konrad, J.G., and G. Chesters. 1969. Degradation in soils of ciodrin andorganophosphate insecticide. J. Agr. Food Chem. 17:226–230.

Landrum, P.F. et al. 1984. Reverse-phase separation method for deter-mining pollutant binding to aldrich humic acid and dissolved organiccarbon of natural waters. Environmental Science and Technology18:187–192.

Lee, H.L. 1975. Trace elements in animal production. In Trace elementsin soil-plant-animal systems, edited by D. Nicholas and R. Egan. NewYork: Academic Press.

Lyman, W. 1982. Adsorption coefficient for soils and sediments. InHandbook of chemical property estimation. New York: McGraw-Hill.

Lyman, W.J., W.F. Reehl, and D.H. Rosenblatt. 1982. Handbook ofchemical property estimation methods: Environmental behavior oforganic compounds. New York: McGraw-Hill.

Mayer, R., J. Letey, and W.J. Farmer. 1974. Models for predictingvolatilization of soil-incorporated pesticides. Soil Science Society ofAmerica Proceedings. Vol. 38:563–568.


TABLE 9.12.6 RANGES FOR KD FOR VARIOUSELEMENTS IN SOILS AND CLAYS

Observed StandardElement Range (ml/g) Meana Deviationb

Ag 10–1,000 04.7 1.3Am 1.0–47,230 06.7 3.0As(III) 1.0–8.3 01.2 0.6As(V) 1.9–18 01.9 0.5Ca 1.2–9.8 01.4 0.8Cd 1.3–27 01.9 0.9Ce 58–6,000 07.0 1.3Cm 93–51,900 08.1 1.9Co 0.2–3,800 04.0 2.3Cr(III) 470–150,000 07.7 1.2Cr(VI) 1.2–1,800 03.6 2.2Cs 10–52,000 07.0 1.9Cu 1.4–333 03.1 1.1Fe 1.4–1,000 04.0 1.7K 2.0–9.0 01.7 0.5Mg 1.6–13.5 01.7 0.5Mn 0.2–10,000 05.0 2.7Mo 0.4–400 03.0 2.1Np 0.2–929 02.4 2.3Pb 4.5–7,640 04.6 1.7Po 196–1,063 06.3 0.7Pu 11–300,000 07.5 2.3Ru 48–1,000 06.4 1.0Se(IV) 1.2–8.6 01.0 0.7Sr 0.2–3,300 03.3 2.0Te 0.003–0.28 03.4 1.1Th 2,000–510,000 11.0 1.5U 11–4,400 03.8 1.3Zn 0.1–8,000 02.8 1.9

Source: C.F. Baes III and R.D. Sharp, 1983, A proposal for estimationof soil leaching and leaching constants for use in assessment models,Journal of Environmental Quality 12, no. 1:17–28.

aMean of the logarithms of the observed values.bStandard deviation of the logarithms of the observed values.

Mabey, W., and T. Mill. 1978. Critical review of hydrolysis of organiccompounds in water under environmental conditions. Jour. Phys.Chem. Ref. Data 17, no. 2:383–415.

Overcash, M.R., and D. Pal. 1979. Design of land treatment systems forindustrial wastes; Theory and practice. Ann Arbor, Mich.: Ann ArborScience.

Rao, P.S.C., and R.E. Jessup. 1982. Development and verification of sim-ulation models for describing pesticide dynamics in soils. Ecol.Modeling 16:67–75.

Sax, N.I. 1984. Dangerous properties of industrial materials. 6th ed. NewYork: Van Nostrand Reinhold.

Scow, K.M. 1982. Rate of biodegradation. In Handbook of chemicalproperty estimation methods. New York: McGraw-Hill.

Tabak, H.N. et al. 1981. Biodegradability studies with organic prioritypollutants compounds. Journal Water Pollution Control Federation53:1503–1518.

Thomas, R.G. 1982. Volatilization from water. In Handbook of chem-ical property estimation methods. New York: McGraw-Hill.

Tisdale, S.L., and W.L. Nelson. 1975. Soil fertility and fertilizers. 3d ed.New York: Macmillan.

U.S. Environmental Protection Agency 1983. Hazardous waste land treat-ment. SW-874. Washington, D.C.: U.S. EPA, Office of Solid Wasteand Emergency Response.

Valentine, R.L. 1986. Vadoze zone modeling of organic pollutants, editedby Stephen Hern and Susan Melancon, 233–243. Chelsea, Mich.:Lewis Publishers.

Valentine, R.L., and J.L. Schnoor. 1986. Biotransformation. In Vadozezone modeling of organic pollutants, edited by Stephen Hern andSusan Melancon. Chelsea, Mich.: Lewis Publishers.

Verink, E.D. 1979. Simplified procedure for constructing pourbaix dia-grams. Journal of Education Modules Math. Sci. Eng. 1:535–560.


9.13TRANSPORT OF CONTAMINANTS INGROUNDWATER

This section discusses the transport of contaminants ingroundwater and describes the transport process and thebehavior of the contaminant plume.

Transport ProcessWhen a contaminant is introduced in groundwater, itspreads and moves with the groundwater as a result of (1)advection which is caused by the flow of groundwater, (2)dispersion which is caused by mechanical mixing and mo-lecular diffusion, and (3) retardation which is caused byadsorption. The mathematical relationship between theseprocesses can be written as follows (Javandel, Doughtly,and Tsang 1984):

}¶

¶xi

} 3Dij }¶¶x

C

j

}4 2 }¶

¶xi

} (Cvi) 2 }C9

n

W9} 5 R }

¶¶C

t} 9.13(1)

vi 5 ¾2

n

Kij¾ }¶¶x

h

j

} 9.13(2)

R 5 31 1 }rb

n

Kd}4 9.13(3)

where:

C 5 contaminant concentrationvi 5 seepage or average pore water velocity in the di-

rection xi

Dij 5 dispersion coefficientKij 5 hydraulic conductivityC9 5 solute concentration in the source or sink fluidW95 volume flow rate per unit volume of the source or

sink

n 5 effective porosityh 5 hydraulic headR 5 retardation factorxi 5 cartesian coordinate

The following discussion uses a simplified two-dimen-sional representation of Equation 9.13(1) to describe thetransport of contaminants in groundwater. In a homoge-neous, isotropic medium having a unidirectional steady-state flow with seepage velocity V, Equation 9.13(1) canbe rewritten as

DL }¶¶

2

x

C2} 1 DT }

¶¶

2

y

C2} 2 V }

¶¶C

x} 5 R }

¶¶C

t} 9.13(4)

where:

C 5 contaminant concentrationV 5 seepage or average pore water velocityDL 5 longitudinal dispersion coefficientDT 5 transversal dispersion coefficientR 5 retardation factor

ADVECTION

A contaminant moves with the flow of groundwater ac-cording to Darcy’s law. Darcy’s law states that the flowrate of water through soil from point 1 to point 2 is pro-portional to the head loss and inversely proportional tothe length of the flow path as

Q 5 2K z A }h2 2

L

h1} 9.13(5)

where:

Q 5 groundwater flow rateA 5 cross-sectional area of flowh2 2 h1 5 head loss between point 1 and point 2L 5 distance between point 1 and point 2K 5 hydraulic conductivity

The actual seepage or average pore water velocity canbe calculated as

V 5 }n

Q

z A} 5 2}

K

n} }

h2 2

L

h1} 9.13(6)

where n is the effective porosity or percent of intercon-nected pore spaces that actually contributes to the flow.

The average pore water velocity calculated in Equation9.13(6) is a conservative estimate of the migration veloc-ity of the contaminant in groundwater. Therefore, whenonly advection is considered, a contaminant moves withthe groundwater flow at the same rate as water, and nodiminution of concentration is observed. In reality, how-ever, the movement of the contaminant is also influencedby dispersion and retardation.

DISPERSION

Dispersion is the result of two processes, molecular diffu-sion and mechanical mixing.

Molecular diffusion is the process whereby ionic or mo-lecular constituents move under the influence of their ki-netic activity in the direction of their concentration gradi-ents. Under this process, constituents move from regionsof higher concentration to regions of lower concentration;the greater the difference, the greater the diffusion rate.Molecular diffusion can be expressed by Fick’s law as

F 5 2Df }d

d

C

x} 9.13(7)

where:

F 5 mass flux per unit area per unit timeDf 5 diffusion coefficientC 5 contaminant concentrationdC/dx 5 concentration gradient

Fick’s law was derived for chemicals in unobstructedwater solutions. When this law is applied to porous me-dia, the diffusion coefficient should be smaller because theions follow longer paths between solid particles and be-cause of adsorption. This application yields an apparentdiffusion coefficient D* represented by

D* 5 w z Df 9.13(8)

where w is an empirical coefficient less than 1. Perkins andJohnston (1963) suggest an approximate value of 0.707for w. Bear (1979) suggests that w is equivalent to the tor-tuosity of the porous medium with a value close to 0.67.Values of D* for major ions can be obtained fromRobinson and Stokes (1965).

Mechanical mixing is the result of velocity variationswithin the porous medium. The velocity is greater in thecenter of the pore space between particles than at the edges.As a result, the contaminant spreads gradually to occupyan ever-increasing portion of the flow field. Mechanicalmixing dispersion can occur both in the longitudinal di-rection of the flow as well as in the transverse direction.According to Bachmat and Bear (1964), the mechanicalmixing component of dispersion can be assumed propor-tional to the seepage velocity as

D11 5 aL z V 9.13(9)

D22 5 aT z V 9.13(10)

where:

D11 5 longitudinal mechanical mixing component ofdispersion

D22 5 transversal mechanical mixing component of dis-persion

aL 5 longitudinal dispersivityaT 5 transversal dispersivityV 5 average linear pore water velocity

Finally the hydrodynamic dispersion coefficients can bewritten as

DL 5 D11 1 Df 5 aL z V 1 D* 9.13(11)

DT 5 D22 1 Df 5 aT z V 1 D* 9.13(12)

The dispersivity coefficients aL and aT are characteris-tic of the porous medium. Representative values of dis-persivity coefficients can be determined from breakthroughcolumn tests in the laboratory or tracer tests in the field(Anderson 1979).

Figure 9.13.1 shows how dispersion can cause some ofthe contaminant to move faster than the average ground-water velocity and some of the contaminant to moveslower than the average groundwater velocity. The frontof the contaminant plume is no longer sharp but rathersmeared. Therefore, when dispersion is also considered,


FIG. 9.13.1 Effect of dispersion–advection on concentrationdistribution. (Reprinted from U.S. Environmental ProtectionAgency, 1989, Transport and fate of contaminants in the sub-surface, Seminar Publication EPA/625/4-89/019 (Cincinnati: U.S.EPA.)

1

0.8

0.6

0.4

0.2

0

RE

LAT

IVE

CO

NC

EN

TR

AT

ION

DISTANCE

ADVECTION

PLUGFLOW

PLUGFLOW

DISPERSION

PLUGFLOW

X1 X2 X3

x1 = vt1

x2 = vt2

x3 = vt3

CONCENTRATION DISTRIBUTION

the contaminant actually moves ahead of what would havebeen predicted by advection only.

RETARDATION

Retardation in the migration of contaminants in ground-water is due to the adsorption mechanism, which was de-scribed in Section 9.12 for both organic and inorganic con-stituents. The retardation coefficient can be calculatedbased on the distribution or adsorption coefficients of thecontaminant and the characteristics of the porous mediumas

R 5 31 1 Kd }r

nd}4 9.13(13)

where Kd is the distribution or adsorption coefficient de-scribed previously. The values pd and n are the bulk den-sity and porosity of the soil. The velocity of the contami-nant in groundwater can be calculated as follows:

Vc 5 }V

R} 9.13(14)

where Vc is the velocity of the contaminant movement ingroundwater, V is the groundwater velocity, and R is theretardation factor. A high retardation factor, i.e., high ad-sorption coefficient, significantly retards the movement ofthe contaminant in groundwater. Figure 9.13.2 illustratesthe effect of advection, dispersion, and retardation on themobility of a contaminant in groundwater.

Contaminant Plume BehaviorThe behavior and movement of contaminants in ground-water depend on the solubility and density of the contam-inant, groundwater flow regime, and the local geology.This section qualitatively discusses the effect of each ofthese factors on the contaminant plume.

CONTAMINANT DENSITY

Immiscible fluids such as oils do not readily mix with wa-ter; therefore, they either float on top of the water tableor sink into the groundwater depending on their density.Immiscible fluids with densities less than water, also calledlight nonaqueous phase liquids (LNAPLs) or floaters, forma separate phase that can float on the groundwater table.For example, if a light-bulk hydrocarbon is released froma surface spill as shown in Figure 9.13.3, it migrates down-ward in the unsaturated zone due to gravity and capillaryforces. If the volume of the released hydrocarbon is large,the hydrocarbon reaches the groundwater and forms apancake on top of the water table. The pancake tends tospread laterally and in the downgradient direction until itreaches residual saturation. A portion of the pancake dis-


FIG. 9.13.2 Effect of advection, dispersion, and retardation onthe mobility of a contaminant in groundwater. (Reprinted fromM. Barcelona, 1990, Contamination of groundwater: Prevention,assessment, restoration, Pollution Technology Review No. 184,Park Ridge, N.J.: Nayes Data Corporation.)

Distance from Slug-Release Contaminant Source

Distance from Continuous Contaminant Source

Rel

ativ

e C

once

ntra

tion

Rel

ativ

e C

once

ntra

tion

A + D + S

A + D + S + B A + D

A

A + DA + D + S

A

A + D + S + B

A AdvectionD DispersionS SorptionB Biotransformation

FIG. 9.13.3 Movement of LNAPLs into the subsurface.(Reprinted from U.S. Environmental Protection Agency, 1989,Transport and fate of contaminants in the subsurface, SeminarPublication EPA/625/4-89/019 (Cincinnati: U.S. EPA.)

GROUNDWATER

FLOW

GROUNDWATER

FLOW

GROUNDWATER

FLOW

GROUNDWATER

FLOW

GROUNDWATER

FLOW

GROUNDWATER

FLOW

PRODUCT

PRODUCTSOURCEINACTIVE

PRODUCTAT RESIDUALSATURATION

TOP OFCAPILLARY

FRINGE

WATERTABLE

PRODUCTAT RESIDUALSATURATION

WATERTABLE

TOP OFCAPILLARY

FRINGE

PRODUCTENTERING

SUBSURFACE

PRODUCT SOURCE

PRODUCT SOURCE

PRODUCTENTERING

SUBSURFACETOP OF

CAPILLARYFRINGE

WATERTABLE

solves in groundwater and eventually migrates with thewater. The maximum spread of the pancake over thegroundwater table can be estimated (CONCAWESecretariat 1974) by

S 5 }10

F

00} 3V 2 }

A

K

z D}4 9.13(15)

where:

S 5 maximum spread of the pancake, m2

F 5 thickness of the pancake, mmV 5 volume of infiltrating bulk hydrocarbon, m3

A 5 area of infiltration, m2

d 5 depth to groundwater, mK 5 constant dependent on the soil’s retention capacity

for oil

Table 9.13.1 lists K values for different types of hy-drocarbons and soil textures.

Immiscible fluids with densities greater than water, alsocalled dense nonaqueous phase liquids (DNAPL) orsinkers, sink through the saturated zone and show a con-centration gradient through the aquifer, becoming moreconcentrated near the aquifer base as shown in Figure9.13.4. Fingering of the dense fluid into the water can alsooccur depending on the characteristics of the aquifer andthe viscosity of the fluid (Dragun 1988). The downwardmigration of the sinker can continue until a zone of lowerpermeability, such as a clay confining layer or a bedrocksurface, is encountered. Halogenated hydrocarbons andcoal tars are the principal solvents possessing densitiesgreater than that of water. Examples of DNAPLs includemethylene chloride, chloroform, trichloroethylene (TCE),tetrachloroethylene or perchloroethylene (PCE), and vari-ous Freons.

Another important factor of both LNAPL and DNAPLplume behavior is residual contamination. As the plumemigrates downward through the unsaturated or saturatedzone, a small amount of fluid remains attached to soil par-ticles and within the soil pore spaces via capillarity forces.This residual contamination can reside in the soil for many

years and serve as a continuous source of contamination.For more information on density flow, see Schwille (1988)and Fenestra and Cherry (1988).

CONTAMINANT SOLUBILITY

The solubility of a substance in water is defined as the sat-urated concentration of the substance in water at a giventemperature and pressure. This parameter is important inthe prediction of a contaminant plume in groundwater andin planning for its recovery. Substances with high watersolubility have a tendency to remain dissolved in the wa-ter column, not adsorbed onto soil particles, and are moresusceptible to biodegradation. Conversely, substances withlow water solubility tend to adsorb onto soil particles andvolatilize more readily from water. The water solubility ofseveral substances is listed in Montgomery (1989). Severalcompounds, such as bulk hydrocarbons, are comprised ofnumerous individual chemicals and substances with dif-ferent solubilities in water and different adsorption coeffi-


TABLE 9.13.1 TYPICAL VALUES FOR K FORVARIOUS SOIL TEXTURES

K

LightSoil Texture Gasoline Kerosene Fuel Oil

Stone & Coarse Gravel 400 200 100Gravel & Coarse Sand 250 125 62Coarse & Medium Sand 130 66 33Medium & Fine Sand 80 40 20Fine Sand & Silt 50 25 12

Sources: CONCAWE Secretariat, 1974, Inland oil spill clean-up manual,Report no. 4/74 (The Hague, The Netherlands: CONCAWE). D.N. Dietz, 1970,Pollution of permeable strata by oil components, In Water pollution by oil,edited by P. Hepple (New York: Elsevier Publishing and Institute of Petroleum).

FIG. 9.13.4 Movement of DNAPLs into the subsurface.(Reprinted from S. Fenestra and J.A. Cherry, Dense organic sol-vents in groundwater: An introduction, In Dense chlorinated sol-vents in groundwater, Progress Report 0863985 (Ontario, Can-ada: Institute of Groundwater Research, University of Waterloo.)

��

DNAPL SOURCE

DNAPL SOURCE

DENSE VAPORSRESIDUAL DNAPL

TOP OFCAPILLARYFRINGE

WATERTABLE

GROUND-WATER FLOW

LOWERPERMEABILITYSTRATA

DISSOLVEDCHEMICALPLUME

A

GROUND-WATER FLOW

GROUND-WATER FLOW

GROUND-WATER FLOW

TOP OFCAPILLARYFRINGE

RESIDUAL DNAPL DENSE VAPORS

B

CDNAPL SOURCE

RESIDUAL DNAPL

GROUND-WATER FLOW


��

��

WATERTABLE


DNAPLLAYER DISSOLVED

CHEMICALPLUME

��DISSOLVEDCHEMICALPLUME

DENSE VAPORSTOP OFCAPILLARYFRINGE

WATERTABLE

GROUND-WATER FLOW

DNAPLLAYER

DNAPL LAYER

cients in soil. When these compounds are introduced ingroundwater, they generate contaminant plumes with dif-ferent shapes and rates of migration.

GROUNDWATER FLOW REGIME

The length and width of the plume are affected by thegroundwater velocity and the aquifer’s hydraulic conduc-tivity. The plume is more elongated in groundwater withhigh velocity than in groundwater with low velocity. Theplume also tends to move slower in formations with lowhydraulic conductivity than in formations with high hy-draulic conductivity. A higher hydraulic conductivity canresult in more rapid movement and a longer and narrowerplume (Palmer 1992). The contaminant plume usuallymoves in the same direction as groundwater; however, thismovement may not occur with a DNAPL that can sink tothe bottom of the aquifer and flow by gravity in the op-posite direction to groundwater flow.

Perched water is another important consideration in theeffect of a groundwater flow regime on a contaminantplume. Perched water does not usually follow the regionalgroundwater flow direction but rather flows along an in-terface of hydraulic conductivity contrast. Therefore, acontaminant plume present in perched water can be mov-ing in a different direction than the regional groundwatergradient. Groundwater fluctuations can move trapped con-taminants from the vadose zone to the saturated zone.

GEOLOGY

The behavior of a contaminant plume depends largely onthe type of geological profile through which it is moving.Geological structures such as dipping beds, faults, cross-bedding, and facies can affect the rate and direction of amigrating plume. Dipping beds can change the directionof a migrating plume. Faults can act as a barrier or a con-duit to the contaminant plume depending on the materialin the fault. Interbedded clay lenses in a permeable sand

formation can split or retard a sinking contaminant plumeand change its shape and course. Fractures and cracks infractured bedrock formations can act as a conduit to thecontaminant plume depending on their size and intercon-nections. Interaquifer exchange can move a plume of con-tamination from formations with the greatest hydraulichead to formations of a lesser hydraulic head (Deutsche1961).

—Ahmed Hamidi

ReferencesAnderson, M.P. 1979. Using models to simulate the movement of con-

taminants through groundwater systems. CRC Crit. Rev. Env. Con-trol 9, no. 2:97–156.

Bachmat, Y., and J. Bear. 1964. The general equations of hydrodynamicdispersion in the homogeneous isotropic porous mediums. J.Geophys. Res. 69, no. 12:2561–2567.

Bear, J. 1979. Hydraulics of groundwater. New York: McGraw-Hill.CONCAWE Secretariat. 1974. Inland oil spill clean-up manual. Report

no. 4/74. The Hague, Netherlands: CONCAWE.Deutsche, M. 1961. Incidents of chromium contamination of ground-

water in Michigan. Proceedings of Symposium on GroundwaterContamination, April. Cincinnati, Ohio: U.S. Dept. of Health,Education and Welfare.

Dragun, J. 1988. The soil chemistry of hazardous materials. Silver Spring,Md.: Hazardous Materials Control Research Institute.

Fenestra, S., and J.A. Cherry. 1988. Subsurface contamination by dense-non aqueous phase liquid (DNAPL) chemicals. International Ground-water Symposium, May. Halifax, Nova Scotia: InternationalAssociation of Hydrogeologists.

Javandel, I., C. Doughtly, and C.F. Tsang. 1984. Groundwater trans-port: Handbook of mathematical models. Water Resources Mono-graph 10. Washington, D.C.: American Geophysical Union.

Montgomery, J.H., and L.M. Wekom. 1989. Groundwater chemicalsdesk reference. Chelsea, Mich.: Lewis Publishers.

Palmer, C.M. 1992. Principles of contaminant hydrogeology. Chelsea,Mich.: Lewis Publishers.

Perkins, T.K., and O.C. Johnston. 1963. A review of diffusion and dis-persion in porous media. Soc. Pet. Eng. J. 3:70–84.

Robinson, R.A., and R.H. Stokes. 1965. Electrolytes solutions. 2d ed.London: Butterworth.

Schwille, F. 1988. Dense chlorinated solvents in porous and fracturedmedia. Chelsea, Mich.: Lewis Publishers.


The purpose of a groundwater remedial investigation is todetermine the nature and extent of contamination, iden-tify current or potential problems caused by the contami-nation, and assist in the evaluation and selection of the re-medial action. The remedial investigation generally hastwo phases. The first phase, called initial assessment, in-volves the use of existing site information and initial fieldscreening techniques to identify potential sources of con-tamination; develop a conceptual understanding of the siteand contamination process; and optimize subsequent,more intrusive, field investigation. The second phase in-volves a detailed subsurface investigation to assess the mag-nitude and extent of contamination and evaluate remedialactions.

Interpretation of Existing InformationBecause the potential costs involved in groundwater re-medial investigations are large, the best use of existing dataand information must be made. Existing information anddata can be site-specific, such as records of operations andrecords of previous investigations, or regional includingsurveys of geology, hydrology, surface soils, and meteo-rology. Existing data, however, can vary in quality; there-fore, a thorough review and interpretation of these dataprior to the investigation is necessary.

SITE-SPECIFIC INFORMATION

Existing data on site history can provide useful informa-tion on potential causes and sources of groundwater con-tamination. Data that should be collected include old mapsand aerial photographs, interviews with present and for-mer employees at the plant site, records of operations,records of product losses and spills, waste disposal prac-tices, and the list of contaminants generated over the op-erating history of the site. The inventory must also includea history of the raw materials used and wastes disposedof over the years as industrial processes changed. Particularattention should be paid to potential sources of ground-water contamination such as locations of abandoned and

active landfills and wastewater impoundments, buriedproduct pipelines, old sewers, tanks, cesspools, dry wells,product storage areas, product loading areas, storm watercollection areas, and previous spill areas.

In addition, foundation borings or construction detailsof supply wells can provide firsthand information on thetypes and characteristics of subsurface soils and ground-water at the site. Chemical data may be available from theresults of previous monitoring activities at the site or atadjacent properties. These data should be analyzed andplotted on base maps and used to estimate backgroundgroundwater and soil quality.

REGIONAL INFORMATION

Regional information can be used to identify potential off-site sources of contamination and to provide backgroundinformation on regional geology, hydrology, surface soils,and meteorology. This information can provide insightinto the complexities of the groundwater contaminationand help guide future site investigations.

A regional inventory of potential offsite sources of con-tamination can be developed through aerial photographs,land-use maps, and field inspections. Old aerial pho-tographs are especially useful because they may be the onlymeans of identifying abandoned facilities such as old land-fills, lagoons, and industrial facilities. Land-use maps canidentify unsewered residential areas that can be a poten-tial source of contamination, especially where organicchemical septic tank cleaners have been used. Topographicmaps can identify surface drainage patterns that can carrycontaminants to the plant site and recharge the underly-ing groundwater system.

Regional geologic reports, maps, and cross sections canprovide details on the regional subsurface geology includ-ing areal extent, thickness, composition, and structure ofthe geological units present in the region. Regional hy-drogeologic reports can provide information on the re-gional groundwater flow direction and quality as well asthe groundwater usage in the region. A survey of state filescan reveal long-term groundwater quality problems in thegeneral area of the plant site.


Groundwater Investigation andMonitoring

9.14INITIAL SITE ASSESSMENT

Soil maps can be used to evaluate the migration po-tential of contaminants through the unsaturated zone.Climatological data can be used to determine precipitationrates and patterns as well as surface runoff and ground-water recharge rates. In addition, climatological data canbe used to determine evapotranspiration rates from shal-low groundwater tables and their effect on the gradientand direction of groundwater.

An inventory of regional information is available fromstate and federal agencies such as the U.S. GeologicalSurvey (USGS), the U.S. EPA, the U.S. Department ofAgriculture (USDA), and the Soil Conservation Service(SCS). Other sources of information include computerizeddatabases on environmental regulations and technical in-formation on a variety of chemical compounds (Lynne etal. 1991). Examples of these databases include theComputer-Aided Environmental Legislative Data System(CELDS), which provides a collection of abstracted fed-eral and state environmental regulations and standards;HAZARDLINE, which provides information on over 500hazardous workplace substances as defined by theOccupational Safety and Health Administration (OSHA);and the Chemical Information System (CIS), which pro-vides a variety of subjects related to chemistry.

Initial Field ScreeningData collection in a groundwater remedial investigationcan begin with minimally intrusive techniques, called ini-tial field screening techniques. These techniques are lessexpensive than the more intrusive techniques such as soilborings, test pits, and well monitoring. In addition, fieldscreening techniques provide information which stream-lines data collection and optimizes the use of intrusive tech-niques. The principal categories of initial field screeningtechniques include surface and downhole geophysicalsurveys and onsite chemical screening, such as a soil-gassurvey.

SURFACE GEOPHYSICAL SURVEYS

Surface geophysical surveys are applied at the surface toprovide a rapid reconnaissance of the hydrogeologic con-ditions at the site, such as depth to bedrock, degree ofweathering, and the presence of clay lenses, fracture zones,or buried waste. In addition, surface geophysical surveyscan be used to detect and map inorganic contaminantplumes, obtain the flow direction, and estimate the con-centration gradients (Benson et al. 1985).

Surface geophysical surveys include electromagneticconductivity, electrical resistivity, seismic refraction, andground-penetrating radar as described in Pitchford,Mazzella, and Scarbrough (1988); Benson, Glaccum, andNoel (1984); and the U.S. EPA desk reference guide on

subsurface characterization and monitoring techniques(1993a). A description of the most commonly used sur-face geophysical surveys follows.

Electromagnetic (EM) Methods

The EM methods use a transmitter coil to generate an elec-tromagnetic field that induces eddy currents in the groundbelow the instrument. A receiver coil measures secondaryelectromagnetic fields created by the eddy currents andproduces an output voltage that can be related to varia-tions in subsurface conductivity as shown in Figure 9.14.1.Variations in subsurface conductivity may be caused bychanges in the basic soil or rock types, thickness of the soiland rock layers, moisture content, fluid conductivity, anddepth to the water table.

Environmental engineers can use EM surveys to obtaindata by profiling or sounding. In profiling, the engineermakes measurements at a number of stations along a sur-vey line to map lateral changes in the subsurface electricalconductivity to a given depth. In sounding mode, the en-gineer places the instrument at one location and takes mea-surements at increasing depths, by changing coil orienta-tion or coil spacing, to map vertical changes in electricalconductivity and, therefore, the soil and rock type at thatlocation.

An advantage of the EM methods is that the surveyscan be done quickly because direct contact of the instru-ment with the ground is not required. The disadvantage,however, is that the EM surveys are susceptible to thepresence of metals and powerlines on the surface of theground.

Electrical Resistivity (ER) Methods

In ER methods, environmental engineers measure the re-sistivity of subsurface materials by injecting an electricalcurrent into the ground through a pair of surface electrodes(current electrodes) and measuring the resulting potentialfield (voltage) from a second pair of electrodes (potentialelectrodes) as shown in Figure 9.14.2. Several types of elec-trode geometries can be used for resistivity measurementsincluding the Wenner, Schlumberger, dipole, and others.The Wenner array is the simplest in terms of geometry andconsists of four electrodes spaced equally in a line.

The ER measurements are a function of the soil or rocktypes, thickness of the soil and rock layers, moisture con-tent, fluid conductivity, and depth to the water table. TheER of a geological formation is calculated based on theelectrode separation, the geometry of the electrode array,the applied current, and the measured voltage.

As with the EM surveys, environmental engineers canuse the ER surveys to obtain data by profiling or sound-ing. In profiling, engineers take measurements at a num-ber of stations along a survey line to map lateral changes


in the subsurface electrical properties to a given depth.Then, they can use the data to delineate hydrogeologicalanomalies or map inorganic plumes. Sounding measure-ments, on the other hand, are made at increasing depthsso that engineers can map vertical changes in electricalproperties. Engineers use data from sounding measure-ments to determine the depth, thickness, and type of soilor rock layer at the site. The data from ER surveys can beinterpreted with the use of computer models or mastercurves to create geoelectric sections (Orellana and Mooney1966). These sections illustrate changes in the vertical andlateral resistivity conditions at the site.

The ER surveys are useful for identifying shallow con-taminated groundwater bodies where (1) a significant con-trast exists in water quality; (2) the water table is less than40 feet deep; (3) the geology of the water table aquifer is

relatively homogeneous; and (4) local interferences, suchas buried pipelines, power lines, or metal fences, are notpresent.

The advantages of the ER methods are that they arewell established and their equipment is inexpensive, mo-bile, and easy to operate and provides relatively rapidareal coverage. In addition, the ER methods are supe-rior to the EM methods for detecting thin resistive lay-ers. The disadvantage, however, is that continuous pro-filing is not possible, and the requirement for groundcontact can cause problems in resistive material and gen-erally makes the ER surveys slower to use than the EMsurveys. Furthermore, use of the ER methods is limitedin wet weather and on paved areas, and the methods areless sensitive to conductive pollutants than the EM meth-ods.


PrimaryField

Coil Transmitter

InducedCurrentLoops

Receiver Coil

Ground Surface

Secondary FieldsFrom Current Loops

Sensed byReceiver Coil

PhaseSensingCircuits

andAmplifiers Chart and

Mag TapeRecorders

10 Meters

15 Meters

6 Meters

Surface

3

21

RECEIVERCOIL

tow line

CHANNEL BOTTOM

AIR

WATER

PRIMARYELECTROMAGNETICFIELD

TRANSMITTERCOIL

FIG. 9.14.1 Electromagnetic survey. (Reprinted from U.S. Environmental Protection Agency, 1993, Subsurface characterization andmonitoring techniques, a desk reference guide, U.S. EPA/625/R-93/003a [May] U.S. EPA.)

Seismic Refraction (SR) Methods

Environmental engineers often use the SR methods to de-termine the top of bedrock or depth of the water table, lo-cate fractures or faults, and characterize the type of rockor degree of weathering. The SR methods are based on thefact that elastic waves travel through different earth ma-terials at different velocities; the denser the material, thehigher the wave velocity.

The elastic waves are initiated by an energy source(hammer or controlled explosive charge) at the ground sur-face. A set of receivers, called geophones, is set up in a lineradiating outward from the energy source as shown inFigure 9.14.3. Waves initiated at the surface and refractedat the critical angle by a high-velocity layer at a depth reachthe more distant geophones quicker than the waves thattravel directly through the low-velocity surface layer. Thetime between the shock and the arrival of the elastic waveat a geophone is recorded on a seismograph. Using a set

of seismograph records, engineers can derive a graph ofarrival time versus distance from the shot point to the geo-phone. They can then analyze the line segments, slope, andbreak points in the graph to identify the number of layersand the depth of each layer. In addition, they can use typ-ical seismic velocity ranges to determine the type of soil ofeach layer (U.S. EPA 1993a).

The advantages of SR methods are that the equipmentis readily available, portable, and relatively inexpensive. Inaddition, the methods are accurate and provide rapid arealcoverage with depths of penetration up to 30 meters. Thedisadvantage, however, is that the resolution might be ob-scured by layer sequences where the velocity of the layersdecreases with depth, and thin layers, called blind zones,might not be detected. Furthermore, the methods are sus-ceptible to noise from adjacent areas (such as constructionactivities) and do not detect contaminants in groundwa-ter.


Current MeterCurrentSource

Volt Meter

SurfaceC1

Current FlowThrough Earth

Current

Voltage

aaa

A M N B

A MN B

WENNER ELECTRODE ARRAY

SCHLUMBERGER ELECTRODE ARRAY

r

A Q

AXIAL OR POLAR

1.5

m

electrodes

1.5

m

1.5

m

1 V

B M O N

AB/2 AB/2

N

P1 P2 C2

FIG. 9.14.2 Electrical resistivity sur-vey. (Reprinted from U.S. Environ-mental Protection Agency, 1993,Subsurface characterization and moni-toring techniques, a desk referenceguide, U.S. EPA/625/R-93/003a [May]U.S. EPA.)

Ground Penetrating Radar (GPR) Methods

Environmental engineers often use the GPR methods tolocate buried objects, map the depth to shallow water ta-bles, and delineate soil horizons. The principles involvedin GPR technology are similar to those in seismic refrac-tion, except that in GPR, electromagnetic energy is usedinstead of acoustic energy, and the resulting image is rel-atively easy to interpret.

In a GPR survey, a transmitting and a receiving antennaare dragged along the ground surface as shown in Figure9.14.4. The small transmitting antenna radiates short

pulses of high-frequency radio waves into the ground, andthe receiving antenna records variations in the reflected re-turn signal. The attenuation loss of the signal in the groundincreases with ground conductivity and with frequency fora given material. Changes in ground electric conductivityare associated with natural hydrogeological conditionssuch as bedding cementation, moisture, clay content, voids,and fractures. Therefore, an interface between two soil orrock layers with sufficient contrast in electric conductivityshows up in the radar profile (Benson and Glaccum 1979).

The advantages of the GPR methods include rapid arealcoverage, where site conditions are favorable, and great


FIG. 9.14.3 Seismic refraction survey. (Reprinted from U.S. Environmental Protection Agency, 1993,Subsurface characterization and monitoring techniques, a desk reference guide, U.S. EPA/625/R-93/003a[May] U.S. EPA.)

��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQ¢¢��QQ¢¢�Q¢��QQ¢¢�Q¢�Q¢�Q¢��QQ¢¢

�Q¢

��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQ¢¢��QQ¢¢�Q¢�Q¢�Q¢

�Q¢�Q¢��QQ¢¢

�Q¢

��QQQQQ¢¢¢¢¢��QQQQQ¢¢¢¢¢��QQQQQ¢¢¢¢¢

��QQ¢¢��QQ¢¢�Q¢��QQ¢¢�Q¢�Q¢�Q¢��QQ¢¢

�Q¢

��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQ¢¢��QQ¢¢�Q¢�Q¢�Q¢�Q¢�Q¢��QQ¢¢

�Q¢

��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQ¢¢��QQ¢¢�Q¢�Q¢�Q¢

�Q¢�Q¢��QQ¢¢

�Q¢

��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQQQ¢¢¢¢

��QQ¢¢

��QQ¢¢��QQ¢¢��QQ¢¢

��QQ¢¢�Q¢

�Q¢�Q¢�Q¢�Q¢�Q¢��QQ¢¢

�Q¢

�Q¢

��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQQQ¢¢¢¢��QQ¢¢��QQ¢¢�Q¢��QQ¢¢�Q¢�Q¢�Q¢��QQ¢¢

�Q¢

��QQQ¢¢¢��QQQ¢¢¢��QQQ¢¢¢��QQQ¢¢¢��QQQ¢¢¢��QQQ¢¢¢

�Q¢��QQ¢¢

��QQ¢¢

��QQ¢¢��QQ¢¢�Q¢

��QQ¢¢�Q¢

�Q¢

��QQ¢¢

Single Channel

VelocityTable

GeologicData

T D Plots

Manual Pick ofArrival Times

Computer

Geophones

Magnetic Tape

InterpretedGeologic Section

Seismograph

HammerSources

Geophone Array Trigger Cable

Soil

Bedrock

Direct WavesArrowhead

Multi Channel

Refracted Waves

resolution and penetration in dry, sandy, or rocky areas.The use of GPR, however, is limited in moist and clayeysoils and soils with high electrical conductivity.

DOWNHOLE GEOPHYSICAL SURVEYS

Downhole geophysical surveys provide localized details onsoil, rock, or fluid along the length of an existing moni-toring well or a borehole. The surveys can also identifypermeable zones, such as sand lenses in glacial tills, weath-ered zones, and fractures or solution cavities in rocks.

Several downhole logging techniques are available in-cluding nuclear, electromagnetic, and acoustic or seismicas described in Keys and MacCary (1976) and the U.S.EPA desk reference guide on subsurface characterizationand monitoring techniques (1993a). Some of these tech-niques provide measurements from inside plastic or steelcasing, and some allow measurements in the unsaturatedzone as well as the saturated zone. A description of themost commonly used logs follows.

Nuclear Logging Methods

Nuclear logging includes methods that detect the presenceof unstable isotopes or create such isotopes in the vicinityof a borehole. Several nuclear logging techniques are avail-able including natural gamma logs, gamma–gamma logs,and neutron–neutron logs. Natural gamma logs are prob-ably the most common nuclear methods used in ground-water studies.

Environmental engineers use natural gamma logging, ingeneral, to identify lithology and stratigraphic correlationand, in particular, to evaluate the presence, variability, andintegrity of clays and shales.

The natural gamma log records the amount of naturalgamma radiation emitted by rocks and unconsolidated ma-terials from a borehole. Different formations can be dis-tinguished from different levels of natural radioactivity asshown in Figure 9.14.5. The gamma-emitting radioiso-topes normally found in all rocks and unconsolidated ma-terials are potassium-40 and daughter products of the ura-nium and thorium decay series (Benson 1991). Clays andshales concentrate these heavy radioactive elementsthrough the process of ion exchange and adsorption; there-fore, their natural gamma activity is much higher than thatof other materials.

The natural gamma log instrumentation is relativelysimple and inexpensive and involves radiation detectiononly. However, only qualitative analysis is possible withthis method, and the sensitivity of the probe is reduced bylarge diameter holes, drilling fluid, and casing (U.S. EPA1993a).

Electromagnetic Logging Methods

As with the EM method, the electromagnetic loggingmethod measures the electrical conductivity of soil or rockin open or polyvinyl-chloride- (PVC) cased boreholesabove or below the water table. Environmental engineersuse this method to perform lithological characterization,locate the zones of saturation, and perform chemical char-


FIG. 9.14.4 Ground-penetrating radar survey. (Reprinted from U.S.Environmental Protection Agency, 1993, Subsurface characterization and mon-itoring techniques, a desk reference guide, U.S. EPA/625/R-93/003a [May] U.S.EPA.)

GRAPHICRECORDER

RECEIVER

TAPERECORDER

POWERSUPPLY

TRANSMITTER

ANTENNAGROUND SURFACE

LAYERED MATERIAL

ISOLATED REFLECTOR

REFLECTED IMPULSESTRANSMITTED IMPULSES

TRANSMITTER---RECEIVER SWITCH

acterization of groundwater. Several electromagnetic log-ging techniques are available including induction logs, mi-crowave-sensing logs, nuclear magnetic resonance logs,and surface-borehole logs. Induction logs are probably themost common electromagnetic methods used in ground-water studies.

The probe in an induction log contains a transmittercoil on the upper part, which induces eddy current in the

formation around the borehole, and a receiver on the lowerpart. Engineers measure conductivity using the same prin-ciples as the EM methods. Because the response of the logis a function of the specific conductance of the pore flu-ids, it is an excellent indicator of the presence of inorganiccontaminants (Benson 1991). Variations in conductivitywith depth also indicate changes in clay content, perme-ability of a formation, or fractures.


FIG. 9.14.5 Well log suites in sedimentary and fractured rocks. (Reprinted from U.S.Environmental Protection Agency, 1993, Subsurface characterization and monitoring tech-niques, a desk reference guide, U.S. EPA/625/R-93/003a [May] U.S. EPA.)

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

FineAlluylumCoarse

Quartzsandstone

SandstoneCoal

Shale

Limestone

Gypsum

Freshwater

Salinewater

AnhydriteSandstone

Limestone

Arkose

Arkosicconglomerate

Granite

Fra

ctur

ed

6-in

chop

en h

ole

8-in

chop

en h

ole

10-in

chca

sing

GAMMA NEUTRON ACOUSTIC VELOCITY CALIPER LITHOLOGYSPONTANEOUS

POTENTIALLONG-NORMAL

RESISTIVITY

SINGLE-POINTRESISTANCE

TEMPERATURE

GAMMA NEUTRONACOUSTICVELOCITYCALIPER LITHOLOGY RESISTIVITY

Quartzmonzonite

Granite

Weatheredor altered

Low-angleopen fracture

High-angleopen fracture

Gneiss

Fracturedand altered

Biotite schist

Granodiorite

Pegmatite

Gabbro

Fracturedand

altered

�

Other Logging Methods

Several other types of logging techniques are useful forcharacterizing lithology and hydrogeology inside a well ora borehole. Examples of these logs include caliper logs,temperature logs, fluid-flow logs, and borehole televisionlogs.

A caliper log provides information on the diameter,lithology, fractures, and construction details of an openborehole. Many types of caliper logs are available includ-ing mechanical, electric, and acoustic. The mechanicalcaliper is the most commonly used. The probe in a me-chanical caliper consists of spring-loaded arms which ex-tend from the logging tool so that they follow the sides ofthe borehole. Mechanical caliper tools have from one tosix arms and can measure variations as small as 1/4 inch inborehole diameter.

A temperature log can provide a continuous record ofthe temperature of the fluid inside the borehole or well.Environmental engineers can use changes in temperatureto identify leaks in the casing where damage or corrosionhas occurred.

Fluid-flow logs measure the fluid flow within a bore-hole or well (Keys and MacCary 1976). Examples of suchlogs include thermal and electromagnetic boreholeflowmeters that sense water movement either vertically orhorizontally (or both) at low velocities. Fluid-flow mea-surements can locate zones of high permeability (fractures)and areas of leakage in artisan wells.

Borehole television camera logs allow visual inspectionof a borehole or well for fractures or casing defects(Morahan and Dorrier 1984).

ONSITE CHEMICAL SURVEYS

Environmental engineers are increasingly using onsitechemical surveys as field screening techniques to pinpointsource areas or approximately delineate the extent of ex-isting contaminant plumes. The use of onsite chemical sur-veys optimizes the number of samples taken by more ex-pensive intrusive techniques and sent to the laboratory forconfirmatory chemical analysis. Several techniques areavailable for volatile and nonvolatile organics as well asfor inorganic compounds.

Onsite chemical screening techniques vary from quali-tative chemical analyses using indicators such as organicvapor analyzers (OVAs) or HNU meters to more quanti-tative soil-gas surveys using gas chromatography and massspectrometry (GC/MS).

Qualitative Onsite Chemical Surveys

Generally, environmental engineers use these field screen-ing techniques to collect preliminary site information andguide future and more intrusive field investigations.Engineers can measure the pH of the soil, waste, or ground-

water in the field with a pH meter and use the results ofthese measurements to characterize the subsurface envi-ronment or classify the corrosivity of waste materials. Theycan also electrometrically measure the Eh of groundwaterin the field using a platinum electrode and a reference elec-trode (Holm, George, and Barcelona 1986; Ritchey 1986).Then, they can use the results of the measurements to char-acterize oxidation-reduction conditions in the subsurfaceand evaluate the potential for mobility of heavy metals ingroundwater.

OVAs, photoionization detectors (PID/HNU meter),flame ionization detectors (FIDs/OVAs), argon ionizationdetectors (AIDs), and combustible gas indicators (EDs) areall total organic vapor survey instruments that locatesource areas of volatile compounds within the vadose zoneor track these compounds within groundwater (U.S. EPA1993b).

Test kits are commercially available for preliminaryfield screening of many inorganic compounds (Hatch kits)and some organic compounds (Handy kits). These kits arebased on the principles of colorimetry. Colorimetry in-volves mixing the reagents of known concentrations witha test solution in specified amounts. This mixing results inchemical reactions in which the color of the solution is afunction of the concentration of the analyte of interest(Davis et al. 1985; Fishman and Friedman 1989).

Soil-Gas Surveys

Environmental engineers use soil-gas surveys to locatesource areas of volatile compounds within the vadose zone,track plumes of volatile compounds in groundwater, iden-tify migration patterns of landfill gases, and optimize thenumber and location of more expensive and intrusive mon-itoring points such as soil borings and groundwater mon-itoring wells.

Soil-gas surveys are based on several in situ soil sam-pling techniques such as headspace analysis, surface fluxchambers, downhole flux chambers, surface accumulators,and suction ground probes. The most commonly used tech-niques, however, are the surface accumulators and the suc-tion probes.

Surface accumulators involve the passive sampling ofsoil gas by trapping volatile organic compounds (VOCs)onto an adsorbent contained within an inverted glass tube(Zdeb 1987). The inverted glass tube is buried in the soilfor a few days to weeks. The adsorbent consists of a fer-romagnetic wire coated with activated charcoal and is con-tained in an inverted test tube. The adsorbent passivelycollects diffusing VOCs which adsorb onto the activatedcharcoal. After a few days or weeks, the glass tube is sealedand taken to the laboratory for VOC analysis.

Ground probe sampling techniques for soil gas involveinserting a tube into the ground and pumping the soil gaswith a vacuum pump. Engineers then analyze the extractedgas in the field for VOCs using portable analytical instru-


ments. The probes can be manually or pneumatically dri-ven or installed in boreholes. Grab samples can be takenat the same depth (or at different depths) at several loca-tions for areal (or vertical) characterization of soil-gas con-centrations.

The vertical and horizontal spacing of the probes canbe affected by many factors such as soil moisture and or-ganic matter content, presence of perched water, depth togroundwater, permeability of the subsurface materials, andthe Henry’s Law constant of the VOC in question (Silka1986). The upward diffusion of vapors is usually blockedby soil strata containing a finer grained soil with a highermoisture content or higher organic carbon content.

—Ahmed Hamidi

ReferencesBenson, R.C. 1991. Remote sensing and geophysical methods for eval-

uation of subsurface conditions. In Practical handbook of ground-water monitoring, edited by David Nielson. Chelsea, Mich.: LewisPublishers.

Benson, R.C., and R.A. Glaccum. 1979. Radar surveys for geotechnicalsite assessment. Proceedings of the Geophysical Methods in Geo-technical Engineering, Specialty Session, 161–178. Atlanta, Ga.:American Society of Civil Engineers.

Benson, R.C., R.A. Glaccum, and M.R. Noel. 1984. Geophysical tech-niques for sensing buried wastes and waste migration. Worthington,Ohio: National Water Well Association.

Benson, R.C. et al. 1985. Correlation between geophysical measurementsand laboratory water sample analysis. Proceedings of the NationalWater Well Association, Environment Protection Agency Conferenceon Surface and Borehole Geophysical Methods in GroundwaterInvestigation. National Water Well Association.

Davis, S.N. et al. 1985. Introduction to groundwater tracers. EPA/600/2-85/022, NTIS PB86-100591.

Fishman, M.J., and L.C. Friedman, eds. 1989. Methods for determina-tion of inorganic substances in water and fluvial sediments. 3d ed.U.S. Geological Survey Techniques of Water Resources Investigations,TWRI 5-A1.

Holm, T.R., G.K. George, and M.J. Barcelona. 1986. Dissolved oxygenand oxidation-reduction potentials in groundwater. EPA/600/2-86/042, NTIS PB86-179678.

Keys, W.S., and L.M. MacCary. 1976. Application of borehole geo-physics to water resources investigations. Techniques of WaterResources Investigations of the United States Geophysical Survey.

Lynne, M. et al. 1991. The overall philosophy and purpose of site in-vestigation. In Practical handbook of groundwater monitoring, editedby David Nielsen. Chelsea, Mich.: Lewis Publishers.

Morahan, T., and R.C. Dorrier. 1984. The application of television bore-hole logging to groundwater monitoring programs. GroundwaterMonitoring Review 4, no. 4:172–175.

Orellana, E., and H.M. Mooney. 1966. Master tables and curves for ver-tical electrical sounding over layered structures. Madrid, Spain:Interciencia.

Pitchford, A.M., A.T. Mazzella, and K.R. Scarbrough. 1988. Soil andgeophysical techniques for detection of subsurface organic contami-nation. USEPA/600/4-88-019, NTIS. U.S. EPA.

Ritchey, J.D. 1986. Electronic sensing devices used for in situ ground-water monitoring. Groundwater Monitoring Review 6, no. 2:108–113.

Silka, L.R. 1986. Simulation of the movement of volatile organic vaporthrough the unsaturated zone as it pertains to soil–gas surveys.Proceedings of the NWWA/API Conference on Petroleum Hydro-carbons and Organic Chemicals in Groundwater: Prevention, Detec-tion, and Restoration. Dublin, Ohio: National Water WellAssociation.

U.S. Environmental Protection Agency. 1993a. Subsurface characteriza-tion and monitoring techniques, a desk reference guide. USEPA/625/R-93/003a (May). U.S. EPA.

———. 1993b. Subsurface characterization and monitoring techniques,a desk reference guide, Vol. 2. USEPA/625/R-93/003b (May). U.S.EPA.

Zdeb, T.F. 1987. Multi-depth soil–gas analysis using passive and dy-namic sampling techniques. Proceedings of Petroleum Hydrocarbonsand Organic Chemicals in Groundwater: Prevention, Detection, andRestoration. Dublin, Ohio: National Water Well Association.


9.15SUBSURFACE SITE INVESTIGATION

The purpose of a subsurface investigation is to collect sam-ples and obtain actual quantitative measurements of chem-ical concentrations, hydraulic parameters, and lithologicaldata within a particular hydrogeologic strata or group ofstrata. Environmental engineers can use these samples andmeasurements to assess the magnitude and extent ofgroundwater or soil contamination and support the selec-tion and design of engineering options for remediation.

Engineers can conduct subsurface investigations usingtemporary groundwater and soil sampling techniques suchas HydroPunch, soil probes, and cone penetrometers (Edge

and Cordy 1989) or more permanent techniques such asthe installation of monitoring wells and soil borings.Temporary techniques are less expensive but less reliable;therefore, they are usually used for screening purposes andthe optimization of the location and number of permanentsystems. Permanent techniques, on the other hand, aremore expensive and more reliable; therefore, their use isusually limited to confirm actual concentrations and sub-surface conditions.

Subsurface investigations involve several field activitiessuch as drilling, installation, development, and sampling

of monitoring wells. These activities are intrusive to thesubsurface environment; therefore, engineers should con-duct them with care to prevent cross-contamination andobtain representative groundwater and soil samples thatretain both the physical and chemical properties of the sub-surface environment. A description of these field activitiesfollows.

Subsurface DrillingSubsurface drilling for groundwater remedial investiga-tions uses much of the same technology as conventionalgeotechnical exploration but with some significant differ-ences. Geotechnical exploration requires the collection ofan intact physical specimen which can be tested for geo-technical properties. In comparison, groundwater remedialinvestigations require that the specimen also be represen-tative of existing conditions and valid for chemical analy-sis. Therefore, the selection of drilling methods and sam-pling protocols in a groundwater remedial investigation ismore restrictive and should be based upon site-specific con-ditions and the type of testing to be done.

The criteria used in the selection of a drilling methodinclude the type of geological formation, depth of drilling,depth of screen setting, types of pollutants expected, ac-cessibility to the site, and availability of drilling equipment.The following section briefly describes the drilling meth-ods used in groundwater remedial investigations.

DRILLING METHODS

Several drilling methods are used in groundwater remedialinvestigations including air rotary, direct mud-rotary, re-verse mud-rotary, hollow-stem augers, solid-stem augers,and cable tools among others (Davis, Jehn, and Smith1991). The following discussion focuses on the two meth-ods most commonly used for monitoring well installations:hollow-stem auger and direct mud-rotary.

Hollow-Stem Auger

The hollow-stem auger is a form of continuous flight augerusually used for drilling monitoring wells in unconsoli-dated materials. The auger consists of a tubular steel cen-ter shaft or axle around which is welded a continuous steelstrip in the form of a helix, also known as flight, as shownin Figure 9.15.1. As the auger column rotates and axiallyadvances in the ground, the dug material is simultaneouslyconveyed to the surface by the helix.

The main advantage of hollow-stem auger drilling isthat no drilling fluids or lubricants are used; therefore, nocontaminants are introduced into the aquifer. In addition,the hollow stem of the auger allows sampling of soil ma-terial as the borehole is advanced and installation of cas-ings and screens for monitoring wells when the requireddepth has been reached. The drill head, or cutting bit, lo-

cated at the bottom of the auger can be removed (tripped)through the center of the auger to the surface. This fea-ture allows the auger to stay in place providing an open,cased hole into which samplers, downhole drive hammers,casings, screens, and other instruments can be inserted.

The hollow-stem auger cannot be used, however, inconsolidated, rock, or well-cemented formations. In addi-tion, depths are usually limited to no more than 150 feet,and vertical leakage of water through the borehole duringdrilling is likely to occur.

Direct Mud-Rotary

Direct mud-rotary drilling is a drilling method in which afluid is forced down the drill stem, out through the bit,and back up the borehole to remove the cuttings as shownin Figure 9.15.2. The cuttings are removed by settling ina sedimentation tank or pond, and the fluid is circulatedback down the drill stem. The drilling fluid can be a liq-uid, such water or mud (water with special additives, e.g.,bentonite and polymers), or it can be gas, such as air orfoam (air with additives, e.g., detergents) (Davis, Jehn, andSmith 1991).

Mud-rotary drilling is a flexible and rapid drillingmethod in all types of geologic materials and depth ranges.The circulating fluid serves to cool and lubricate the bit,


AugerconnectorAuger headReplaceablecarbide insertauger tooth

Center plug

Pilot Bit

Pilot assemblycomponents

Center rod

Hollow stemauger section

Auger connector

Rod to capadapter

Drive cap

FIG. 9.15.1 Hollow-stem auger system. (Reprinted from U.S.Environmental Protection Agency, 1993, Subsurface characteri-zation and monitoring techniques, a desk reference guide, Vol.1, USEPA/625/R-93/003a [May] U.S. EPA.)

stabilize the borehole, remove the cuttings, and prevent theinflow of formation fluids, thus minimizing cross-contam-ination of aquifers. In addition, samples can be obtaineddirectly from the circulated fluid when a sample-collectingdevice is placed in the discharge pipe before the settlingtank.

Mud-rotary drilling, however, requires the introductionof some foreign liquids into the aquifer, which can com-promise the validity of subsequent monitoring well sam-ples. In addition, contaminants might be circulated withthe fluid, and the collection of representative samples isdifficult due to the mixing of drill cuttings. Other limita-tions of mud-rotary drilling include the inability to pro-vide information on the position of the water table andthe loss of drilling fluids in fractured materials.

SOIL SAMPLING

Soil samples are usually taken at regular intervals duringdrilling to be analyzed for chemical composition and testedfor physical properties such as particle size distribution,textural classification, and hydraulic conductivity. Thesamples are generally taken from the bottom of the bore-hole and at the necessary depth when the sampling deviceis driven with the aid of a 140-pound hammer. The ham-mer is connected to the sampling device by drill rods. The

number of hammer blows, usually counted for each 6-inchincrement of the total drive, indicates the compaction anddensity of the formation being penetrated.

The most commonly used soil sampling devices are thesplit-spoon sampler and the Shelby tube. The split-spoonsampler is a 12- or 18-inch long hollow cylinder consist-ing of two equal semicylindrical halves held together ateach end with threaded couplings. The sampler is loweredto the bottom of the borehole and driven with a hammerto the necessary depth. When the sampler is brought tothe surface, it is disassembled, or split, to remove the soilsample. Split-spoon sampling provides representative soilsamples for physical or chemical testing. The samples,however, are disturbed; therefore, the results of the analy-sis should be used with caution. When the same sampleris used to collect different samples, the engineer should de-contaminate it after each sampling event to prevent cross-contamination.

The Shelby tube sampler is a thin-walled tube made ofsteel, aluminum, brass, or stainless steel. The cutting edgeof the tube is sharpened, and the upper end is attached toa coupling head by cap screws. The sampler must meetcertain criteria, such as a clearance ratio of 0.5 to 1.5 andend area ratio of 10, to ensure the least disturbance to thesample (U.S. EPA 1993). The sample collection procedureis similar to split-spoon sampling except that the tube ispushed into the soil by the weight of the drill rig ratherthan driven. When the sampler is brought to the surface,the sample is sealed and preserved for laboratory analysis.Shelby tube sampling is used for soil analyses that requireundisturbed soil samples, such as hydraulic conductivitytesting.

Monitoring Well InstallationThe installation of groundwater monitoring wells involvesselecting the location and number of wells, drilling theboreholes, installing the casings and screens, placing thefilter pack materials and annular seals, and finally devel-oping the wells. Each of these steps should be designed tomeet the objectives of the monitoring program and suitthe conditions of the site. A typical monitoring well designis shown in Figure 9.15.3.

WELL LOCATION AND NUMBER

The selection of the proper number and locations of mon-itoring wells is obviously one of the most important deci-sions in any groundwater monitoring program or ground-water remedial investigation. For monitoring purposes,most guidances suggest a minimum of four monitoringwells per potential source of groundwater contamination(U.S. EPA 1986). Three of these wells are placed down-gradient of the potential source, and one is placed upgra-dient. The purpose of these wells is to provide informa-tion on the background quality of the groundwater


��

@@@@@@

��

ÀÀÀÀÀÀ

��

��

@@@@ @ @@@@

��

ÀÀÀÀ À ÀÀÀÀ

��

��

@@@@ @ @@@@

��


��

��

@@@@ @ @@@@

��


��

��@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��

��@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

Pumpsuction

Screen

Settling Pit

Casing

Cuttings circulated to surfacethrough annular space

Tricone bit

Borehole wall

FIG. 9.15.2 Direct mud-rotary circulation system. (Reprintedfrom U.S. Environmental Protection Agency, 1993, Subsurfacecharacterization and monitoring techniques, a desk referenceguide, Vol. 1, USEPA/625/R-93/003a [May] U.S. EPA.)

1080

��

��

��

��

��

6 in (152 mm) Clearance For Sampler

Top of Riser 3 ft (1.0 m)Above GradeSlope Bentonite/Soil

Mixture or 4 in (101 mm) ThickConcrete Pad Away FromCasing

Slope Grout Away From Casing or Riser ToPrevent Infliltration, But Do Not Create a Mushroom For Grout Which Will Be SubjectTo Frost Heave

Dry Bentonite Pellets

Minimum 2 in (50 mm) ID Riser With FlushThreaded Connections WrappedWith Pipe Tape or With O-rings(Varies With Riser Material)

Centralizers As Necessary

Borehle

Centralizer(s) As Necessary

PlugSediment Sump (As Appropriate)

Well Screen Length Varies

1 ft - 2 ft (303 mm to 608 mm), First SecondaryFilter Pack Where Conditions WarrantExtend Primary Filter Pack 20% of ScreenLength or 2 ft (608 mm) Above Slotted WellScreen, Unless Conditions Warrant Less

6 in - 1 ft (152 mm to 304 mm) Final Secondary Filter Pack

Grout Length Varies

1/4 in (6.3 mm) Weep Hole at 6 in Above Ground Level

Washed Pea Gravel or Coarse Sand Mixture

Protective Casing

Vented Cap

Well Identification LabeledInside and Outside The Cap

Proctective CoverWith Locking Cap

3 ft - 5 ft (1.0 m to 1.5 m) Bentonite Seal

3 ft - 5ft (1.0 to 1.5 m) Extend Protective CasingDepth To Below Frost Line

��

��

��

�

��

�

��

6 in (152 mm) Clearance For Sampler

Top of Riser 3 ft (1.0 m)Above GradeSlope Bentonite/Soil

Mixture or 4 in (101 mm) Thick Concrete Pad Away FromCasing

Slope Concrete Grout Away From Casingor Riser To Prevent Infiltration, But Do NotCreate a Mushroom Which Will Be SubjectTo Frost Heave

Dry Bentonite Pellets

Minimum 2 in (50 mm) ID Riser With FlushThreaded Connections Wrapped With Pipe Tape or With O-rings(varies With Riser Material)

Centralizers as Necessary

Borehole

Centralizers as Necessary

PlugSediment Sump (As Appropriate)

Well Screen Length Varies

1 ft - 2 ft (303 mm to 608 mm) SecondaryFilter Pack Where Conditons Warrant Extend Primary Filter Pack 20% of ScreenLength or 2 ft (608 mm) Above Slotted WellScreen, Unless ConditionsWarrant Less

6 in - 1ft (152 to 304 mm) Final Secondary Filter Pack

Grout Length Varies

1/4 in (6.3 mm) Weep Hole At 6 in (152 m) Above Ground Level

Washed Pea Gravel or Coarse Sand Mixture

Protective Casing

Vented Cap

Well Identification LabeledInside and Outside The Cap

Protective CoverWith Locking Cap

3 ft - 5 ft (1.0 m to 1.5 m) Bentonite Seal

3 ft - 5ft (1.0 to 1.5 m) Extend Protective Casing Depth To Below Frost Line��

Steel Casing Installed and Stabilized Ata Minimum 2 ft (60 mm) Into Confining Layer

FIG. 9.15.3 Typical monitoring well design and construction detail. (Reprinted from U.S. Environmental Protection Agency, 1993,Subsurface characterization and monitoring techniques, a desk reference guide, Vol. 1, USEPA/625/R-93/003a [May] U.S. EPA.)

upgradient of the source and detect or monitor any con-taminant plumes emanating from the source.

In a remedial investigation, however, the preliminaryselection of the location and number of wells needed todelineate and monitor the plume is usually based on theresults of initial field screening techniques such as gas sur-veys, HydroPunch, geophysical surveys, and borings.Environmental engineers use the data from these investi-gations to estimate the extent of the contaminant plumeand establish the basic hydrogeologic parameters of thesite. Once the hydrogeology of the site is understood andthe migration path of the suspected contaminant plume isestablished, the location and number of wells can be fi-nalized. In general, the more complicated the hydrogeol-ogy, the more complex the migration path of the con-taminant plume and the greater the number of requiredmonitoring wells (Barcelona et al. 1990).

CASINGS AND SCREENS

The purpose of the casing and screen in a groundwatermonitoring well is to provide access from the surface tothe groundwater in order to collect groundwater samplesor measure groundwater elevations. The casing preventsgeologic materials from collapsing into the borehole, whilethe screen allows groundwater to enter the monitoringwell. The screen is generally attached to the subsurface endof the casing.

Several types of casing and screen materials are avail-able including stainless steel, galvanized steel, carbon steel,PVC, Teflon, and aluminum. Selecting the monitoring wellcasing and screen material depends on the physical strengthand chemical reactivity of the material under subsurfaceconditions. With regard to physical strength, the casingand screen should be able to withstand the forces exertedon them by the surrounding geologic materials. Theseforces can be significant for deep monitoring wells (greaterthan 30 meters). Nielson and Schalla (1991) provide dataon the physical strength of different types of casing andscreen materials.

With regard to chemical reactivity, the material of thecasing and screen should neither adsorb nor leach chemi-cal constituents which would bias the representativenessof the samples collected. In addition, the material must bedurable enough to endure chemical attacks (corrosion orchemical degradation) from the natural chemical con-stituents or the contaminants in the groundwater. Teflonis probably the most chemically resistant material used inmonitoring well installation, but the cost of Teflon is high(Barcelona et al. 1990). Stainless steel offers good strengthand chemical resistance in most environments (except inhighly acidic conditions), but it too is expensive. Galvan-ized steel is less expensive; however, it can impart iron,manganese, zinc, and cadmium to many waters. PVC hasgood chemical resistance except to low molecular weight

ketones, aldehydes, and chlorinated solvents (Miller 1982).Two types of screens are commonly used in monitor-

ing wells: machine-slotted pipes and continuous-clot wire-wound screens. Machine-slotted pipes are readily availableand inexpensive, but the low amount of open area in thesescreens makes development of the well difficult.Continuous-slot screens, in contrast, are more efficient, buttheir cost is relatively high. The design of the slot size ofthe screen must be based on the characteristics of the fil-ter pack material and the grain size of the stratum. Theoptimum slot size should provide maximum open area forwater to flow through and minimum entry of fine parti-cles into the well during piping (Nielson and Schalla 1991;Aller et al. 1991).

The depth of placement of the screen as well as its lengthare usually determined based on the depth and thicknessof the water-bearing zone to be monitored. When the ob-jective of the well is to monitor a potable water supplyaquifer, then a longer screen, perhaps over the entire thick-ness of the aquifer, might be selected. On the other hand,when the objective of the well is to vertically delineate aplume, such as with cluster wells, then shorter screens atspecific intervals might be selected.

The screen should be fully submerged to prevent con-tact between the contaminated groundwater and the at-mosphere, particularly for volatile compounds. The screenis, however, extended above the water table for wells con-structed to monitor floating products. In this case, thescreen length and position must accommodate variationsin water table elevation.

The casings are produced in various diameters (2, 4, 6,and 8 inches) and various lengths (5, 10, and 20 feet) thatare joined by various coupling methods during installa-tion. The casing diameter depends on the future use of thewell, the type of pumping equipment, and the method ofdrilling. Small diameters (2 and 4 inches) are used for mon-itoring wells, while large diameters (6 and 8 inches) areused for recovery wells.

The casing must extend from the top of the screen tothe ground surface level. The casing is protected at the sur-face by a metal protective casing or a manhole. Multiplecasings are installed for wells penetrating more than onewater-bearing formation. The purpose of multiple casingsis to prevent a hydraulic connection and potential crosscontamination between the water-bearing formationsalong the annular space produced by the installation ofwell casings. Figure 9.15.3 shows an example of a dou-ble-cased well installation where the outer casing is an-chored into the confining layer before the borehole is ad-vanced and the well is installed through the inner casing.

FILTER PACKS AND ANNULAR SEALS

Filter packs placed around the well screen allow ground-water to flow freely into the well while keeping fine par-


ticles from entering the well. Two types of filter packs areused in monitoring wells: naturally developed filter packsand artificially introduced filter packs. Naturally developedfilter packs are produced in situ when the fine-grained ma-terials around the screen are removed during the well de-velopment process. Environmental engineers construct ar-tificial filter packs by backfilling the annular spacesurrounding the screen with a granular, relatively inert ma-terial such as clean silica sand.

In an artificially filter-packed well, the filter materialcan be selected for optimum efficiency of well operation,but the procedure of introducing the filter pack is timeconsuming and expensive. Furthermore, bridging can pre-vent complete filling around the well screen, and the filterpack material can introduce contaminants into the aquifer;a leaching test can determine whether this contaminationis a problem. Naturally developed filter packs are, on theother hand, simpler, less expensive, and do not introducenew contaminants into the aquifer. However, well devel-opment for these filter packs is more difficult, and successis less assured.

Engineers can use a tremie pipe or a reverse circulationmethod to place the artificial filter pack. The tremie pipemethod allows funneling of the material directly into theinterval around the well screen. In a reverse circulationmethod, a mixture of sand and water is fed into the an-nulus around the screen, and the water entering the screenis pumped up to the surface through the casing. The en-gineer progressively pulls back the temporary casing (forhollow-stem augers) to expose the screen as the filter packmaterial builds up around the well screen.

Artificially introduced filter packs usually extend fromthe bottom of the screen to at least 3 to 5 feet above thebottom of the screen. This extension accounts for settle-ment of the filter pack material and allows a sufficientbuffer zone between the well screen and the annular sealabove.

After the filter pack is placed around the well screen,the engineer seals the annular space between the well cas-ing and the formation to prevent upward or downwardmovement of water and contaminants along this pathway.In addition, the engineer places a surface seal of concretearound the protective casing to prevent surface drainageinto the borehole. The annular seal is usually composedof bentonite or neat cement (Williams and Evans 1987).Bentonite is readily available and inexpensive but can causeconstituent interference due to ion exchange. Neat cementis also readily available and inexpensive, but channelingbetween the casing and seal can develop due to tempera-ture changes during the curing process (U.S. EPA 1993).

The engineer places the sealing mixture in the annularspace using a side-discharge tremie pipe through which thegrout is pumped from the surface. Complete sealing of theannular space is necessary to avoid potential bridging ofthe grout with formation material (Campbell and Lehr1975).

WELL DEVELOPMENT

The purpose of well development is to remove the residuesof drilling fluids and fine particles of filter packs so thatsubsequent sampling is representative of the groundwater.The development should be performed as soon as possi-ble after the well is installed and the annular seal is cured.

Development methods include bailing, overpumping,air surging, and high-velocity jetting. In bailing, a bailer isdropped and retrieved in and out of the well causing anoutward surge of water through the well screen and filterpack. Such surging forces the loosely bound fine particlesthrough the screen and into the well where they can be re-moved by the bailer. Bailing has the advantage of being asimple technique which does not introduce new fluids intothe aquifer. However, bailing is time consuming and inef-fective in unproductive wells.

In overpumping, a submersible pump is lowered intothe well and alternatively turned on and off, usually at aslightly higher rate than what the formation can deliver.This action, along with the repeated raising and loweringof the pump into the well, causes the water to move backand forth through the well screen, moving fine particlesand drilling fluids into the well where they can be removed.Overpumping is convenient for small wells or pooraquifers; however, excessive pumping rates can cause wellcollapse, especially in deep wells.

Air surging consists of injecting compressed air in thewell, causing the water column to lift almost to the sur-face, and shutting off the air supply to allow the columnto fall back into the well. Repeated use of this techniquecauses an outward surging action in the well intake whichforces the loosely bound fine particles through the screenand into the well where they can be removed. Environ-mental engineers must filter the injected air so that con-taminants, such as lubricants of the compressor, are notintroduced into the well.

High-velocity jetting uses nozzle devices to force a hor-izontal stream of water against the well screen opening.Engineers can remove the material that enters the screenin the backwash of the jet stream by pumping or bailing.High-velocity jetting is effective in removing the mud cakeand breaking the bridges in the filter pack. However, thistechnique can introduce potential air and water contami-nants to the aquifer.

Groundwater SamplingThe objective of any groundwater sampling program is tocollect and analyze samples that are representative of ex-isting groundwater conditions at the site. This goal isachieved with a sampling plan that incorporates samplingprocedures designed to minimize sources of error or mis-representation in each stage of the sampling process. Thekey stages of sampling involve well purging, sample col-lection and pretreatment, sample handling and preserva-


tion, and analysis and reporting of analytical data. A briefdescription of these sampling stages follows. Figure 9.15.4presents a generalized flow diagram of groundwater sam-pling protocol.

PURGING

Before environmental engineers can sample a monitoringwell, they must remove the water standing in the well toallow fresh water from the aquifer to enter the well.Purging is necessary because the stagnant water in the wellis subject to chemical reactions from contact with well con-struction materials and the atmosphere for extended peri-ods of time (Seanor and Brannaka 1983; Wilson andDworkin 1984). The volume of water which should be re-moved from the well is based on the hydraulic character-istics of individual wells and geological settings (Gibb,Schuller, and Griffin 1981). A general rule is to remove

three to five well volumes or to remove water until thewater quality indicators, such as pH, conductance, andtemperature are stable.

When purging a well, engineers should not allow thewater level to drop below the level of the well screen toavoid aeration and loss of volatile or redox-sensitive com-pounds. In addition, the pumping rate should not exceedlevels that might cause turbulent flow in the well and sub-sequent pressure changes and loss of dissolved gases (Mer-idith and Brice 1992). Overpumping can also dilute thesample or increase its turbidity because of the fine parti-cles that may be drawn into the well.

Engineers should use the same equipment for purgingand sampling to minimize the number of items that enterthe well and therefore, the possibility of cross contamina-tion. Furthermore, placing the purging device at the topof the well screen or at the top of the column of water en-sures that all stagnant water is removed (Unwin and Huis1983).


Well Inspection Hydrologic Measurements Water-LevelMeasurements

Measure the water level to 60.3cm (60.01 ft).

Well Purging Removal or Isolation of Stagnant Water

Sample CollectionFiltration*FieldDeterminations**

PreservationField BlanksStandards

StorageTransport

Determination of Well-Purging Parameters(pH, Eh, T, V-1)**

Unfiltered

Volatile Organics, TOX

Dissolved Gases, TOC

Large Volume Sam-ples for Organic

Compound Determi-nations

Assorted SensitiveInorganic Species NO2

–, NH4+, Fe(II)

(as needed for goodQA/QC)

Trace Metals forMobile Substance

Load+++

Field Filtered*

Alkalinity/Acidity**

Trace Metal Samplesfor Specific Geochemical

Imformation+++

S a, SensitiveInorganics

Major Cations andAnions

Representative WaterAccess

Verification ofRepresentative Water

Sample Access

Sample Collection byAppropriate Mechanism

Minimal Sample Handling

Head-SpaceFree Samples

Minimal Aeration orDepressurization

Minimal Air Contact,Field Determination

Adequate Rinsing againstContamination

Minimal Air Contact,Preservation

Minimal Loss of SampleIntegrity Prior to Analysis

Pump water until well purgingparameters (e.g., pH, T V-1, Eh)stabilize to 610% over at leasttwo successive well volumespumped.

Pumping rates should be limitedto ,100 mL/min for volatileorganics and gas-sensitiveparameters.

Filter: Trace metals, inorganicanions/cations, alkalinity.Do not filter: TOC, TOX, volatileorganic compound samples. Filterother organic compoundsamples only when required.

Samples for determinations ofgases, alkalinity, and pH shouldbe analyzed in the field if at allpossible.

At least one blank and onestandard for each sensitiveparameter should be made up inthe field on each day ofsampling. Spiked samples are also recommended for good QA/QC.

Observe maximum sampleholding or storage periodsrecommended by the agency.Documentation of actual holdingperiods should be carefullyperformed.

** Denotes analytical determinations that should be made in the field.+++ See Puls and Barcelona (1989).

Step Procedure Essential Elements Recommendations

* Denotes samples that should be filtered to determine dissolved constituents. Filtration should be accomplished preferably with inline filters and pump pressure or by N2 pressure methods. Samples for dissolved gases or volatile organics should not be filtered. In instances where well development procedures do not allow for turbidity-free samples and may bias analytical results, split samples should be spiked with standards before filtration. Both spiked samples and regular samples should be analyzed to determine recoveries from both types of handling.

FIG. 9.15.4 Generalized flow diagram of groundwater sampling protocol. (Reprinted from U.S.Environmental Protection Agency, 1993, Subsurface characterization and monitoring techniques, a desk ref-erence guide, Vol. 1, USEPA/625/R-93/003a [May] U.S. EPA.)

COLLECTION AND PRETREATMENT

Groundwater samples can be collected with portable ordedicated in situ sampling equipment. Portable equipmentincludes bailers, syringes, suction-lift pumps, submersiblepumps, and gas-driven devices. In situ sampling equipmentincludes cone penetrometer samplers (e.g., Hydropunch,BAT, CPT, or DMLS), chemical-sensitive probes, ion-se-lective electrodes, fiber-optic chemical sensors, multilevelcapsule samplers, and multiport casings. A description ofthese types of equipment as well as their advantages anddisadvantages is in the U.S. EPA desk reference guide onsubsurface characterization and monitoring techniques(1993).

Selecting sampling equipment should be based on thepurpose of the sampling as well as the construction mate-rials of the sampling equipment and the method of sam-ple delivery. The construction materials of the samplingdevice could affect the integrity of the sample because con-stituents can leach from the materials into the water sam-ples or contaminants from the water sample can adsorbonto the sampler materials (Barcelona, Gibb, and Miller1983). Therefore, inert materials should be specified whennecessary. The method of sample delivery is important be-cause devices that cause aeration, degassing, or pressurechanges of the sample may not preserve the chemical qual-ity of the sample. For example, devices that introduce dis-solved oxygen into the sample could cause oxidation offerrous iron to ferric iron, which affects the speciation andconcentration of many chemical constituents in the sam-ple (Hrzog, Pennimo, and Nielson 1991). Turbulence anddepressurization can affect the sample’s original contentof dissolved oxygen, carbon dioxide, and volatile organiccompounds (Barcelona, Gibb, and Miller 1983).

Another decision environmental engineers should makebefore sampling is whether to filter the sample in the field.This decision should be based on the characteristics of theconstituents and the purpose of the sampling program. Forexample, samples requiring analysis for dissolved metals,alkalinity, and anionic species should be filtered. In con-trast, samples for dissolved gases or volatile organicsshould not be filtered since the handling required by fil-tration could lose these chemicals. Furthermore, filtrationshould be performed when the sampling program is con-cerned only with those constituents that are dissolved ingroundwater, excluding all constituents which can be ad-sorbed onto particulate matter in suspension, such as PCBsor polynuclear aromatic hydrocarbons. However, when adrinking water source is studied, samples should not befiltered before analysis because water taken from privatewells is generally not filtered before use. In some instances,engineers must run parallel sets of filtered and unfilteredsamples to determine the dissolved and adsorbed portionsof the constituent of interest.

Filtration is accomplished by vacuum, pressure, or in-line filtration devices. Stolzenburg and Nichols (1986) de-

scribe a variety of filtration equipment and their effects onsampling. The preferred device is the inline filter becauseit reduces the aeration and degassing of the sample as wellas the potential of sample cross contamination caused byimproper equipment decontamination.

To prevent cross contamination, engineers should de-contaminate the equipment used for sample collection orpretreatment prior to and after each use. The decontami-nation should involve a minimum of scraping or brushingto remove any soil or residue from the device, washingwith potable or deionized water, washing with detergentsor cleaning fluids such as acetone, and pressure cleaningwith a high-pressure steam cleaner.

QUALITY ASSURANCE AND QUALITYCONTROL

Groundwater sampling requires a quality assurance andquality control (QA/QC) plan which is designed to mini-mize sources of error in each stage of the sampling process,from sample collection to analysis and reporting. TheQA/QC plan should include procedures and requirementsfor chain-of-custody, sample storage and holding time, useof quality control samples, instrument calibration, sampleanalysis, laboratory validation, documentation, and recordkeeping.

A chain-of-custody must be filed and maintained fromthe moment the sample bottles are released from the lab-oratory until the samples are received by the laboratory.The samples must be stored in conditions that preservetheir integrity. Some samples require acidification to a spec-ified pH or cooling to a specified temperature. In addition,the recommended maximum holding time for the analyteof interest should not be exceeded. Required holding timescan range from hours to days as shown in Table 9.15.1.

The purpose of quality control samples is to detect ad-ditional sources of contamination in the field or labora-tory that might potentially influence the analytical valuesreported in the samples. Examples of quality control sam-ples include trip blanks and field blanks.

Trip blanks consist of a set of sample bottles filled atthe laboratory with laboratory demonstrated analyte-freewater. Trip blanks travel to the site with the empty sam-ple bottles, at a rate of one per shipment, and back fromthe site with the collected samples to simulate sample han-dling conditions. Contaminated trip blanks indicate inad-equate bottle cleaning or blank water of questionable qual-ity.

Field blanks serve the same purpose as trip blanks butare also used to indicate potential contamination from am-bient air or sampling instruments. At the field location, an-alyte-free water is passed through clean sample equipmentand placed in an empty sample container for analysis.Therefore, by being opened in the field and transferredover a cleaned sampling device, the field blank can indi-



TABLE 9.15.1 REQUIRED HOLDING TIMES FOR SEVERAL ANALYTES

Volume MaximumParameters Required (mL) Containers Preservation Holding(Type) 1 Samplea (Material) Method Period

Well purgingpH (grab) 50 T,S,P,G None; field det. ,1 hrb

V4 (grab) 100 T,S,P,G None; field det. ,1 hrb

T (grab) 1000 T,S,P,G None; field det. NoneEh (grab) 1000 T,S,P,G None; field det. None

ContaminationIndicators

pH, V4 (grab) As above As above As above As aboveTOC 40 G,T Dark, 4°C 24 hrTOX 500 G,T Dark, 4°C 5 days

Water qualityDissolved gases 10 mL minimum G,S Dark, 4°C ,24 hr

(O2CH2CO2)Alkalinity/acidity 100 T,G,P 4°C/None ,6 hrb

,24 hrFiltered under

pressure withappropriatemedia

(Fe, Mn, Na1, All filtered T,P Field acidified 6 monthsc

K1, Ca11, 1000 mLf to pH ,2 withMg11) HNO2

(PO24, Cl23 @50 (T,P,G 4°C 24 hrf

Silicate) glass only) 7 daysc,7 days

NO23 100 T,P,G 4°C 24 hrd

SO24 50 T,P,G 4°C 7 daysc

OH14 400 T,P,G 4°C/H2SO4 to 24 hrf

pH ,2 7 daysPhenols 500 T,G 4°C/H2PO4 to 24 hours

pH ,4Drinking Water Same as Same as Same as above 6 months

suitability above for aboveAs, Ba, Cd, Cr, waterPb, Hg, Se, Ag quality

cations(Fe, Mn,etc.)f

F Same as Same as Same as above 7 dayschloride aboveabove

Remaining organic As for TOX/ 24 hoursTOC, exceptwhere analyti-cal parametersmethod callsfor acidifi-cation of sample

Source: U.S. Environmental Protection Agency, 1993, Subsurface characterization and monitoring techniques, a desk reference guide, Vol. 1, USEPA 625/R-93/003a,May (U.S. EPA).

T 5 Teflon; S 5 stainless steel; P 5 PVC, polypropylene, polyethylene; G 5 borosilicate glass.aIt is assumed that at each site, for each sampling date, replicates, a field blank, and standards must be taken at equal volume to those of the samples.bTemperature correction must be made for reliable reporting. Variations greater than 610% can result from a longer holding period.cIn the event that NHO2 cannot be used because of shipping restrictions, the sample should be refrigerated to 4°C, shipped immediately, and acidified on receipt at

the laboratory. Container should be rinsed with 1:1 HNO3 and included with sample.d28-day holding time if samples are preserved (acidified).eLonger holding times in U.S. EPA (1986b).fFiltration is not recommended for samples intended to indicate the mobile substance lead. See Puis and Barcelona (1989a) for more specific recommendations for

filtration procedures involving samples for dissolved species.

cate ambient and equipment conditions that can poten-tially affect the quality of the associated samples.

—Ahmed Hamidi

ReferencesAller, L. et al. 1991. Handbook of suggested practices for the design and

installation of groundwater monitoring wells. EPA/600/4-89/034.Barcelona, M. et al. 1990. Contamination of groundwater: Prevention,

assessment, restoration. Pollution Technology Review 184. ParkRidge, N.J.: Noyes Data Corporation.

Barcelona, M.J., J.P. Gibb, and R.A. Miller. 1983. A guide to the selec-tion of materials of monitoring well construction and groundwatersampling. Illinois State Water Survey Report 327.

Campbell, M.D., and J.H. Lehr. 1975. Well cementing. Water WellJournal 29, no. 7:39–42.

Davis, H.E., J. Jehn, and S. Smith. 1991. Monitoring well drilling, soilsampling, rock coring, and borehole logging. In Practical handbookof groundwater monitoring, edited by David Nielsen. Chelsea, Mich.:Lewis Publishers.

Edge, R.W., and K. Cordy. 1989. The HydroPunch: An in situ samplingtool for collecting groundwater from unconsolidated sediments.Groundwater Monitoring Review (summer):177–183.

Gibb, J.P., R.M. Schuller, and R.A. Griffin. 1981. Procedures for the col-lection of representative water quality data from monitoring wells.Cooperative Groundwater Report 7, Illinois State Water andGeological Surveys.

Hrzog, B., J. Pennimo, and G. Nielson. 1991. Groundwater sampling.In Practical handbook of groundwater monitoring, edited by DavidNielsen. Chelsea, Mich.: Lewis Publishers.

Meridith, D.V., and D.A. Brice. 1992. Limitations on the collection ofrepresentative samples from small diameter monitoring wells.Groundwater Management II (6th NOAC): 429–439.

Miller, G.D. 1982. Uptake and release of lead, chromium, and trace levelvolatile organics exposed to synthetic well castings. Proceedings ofSecond National Symposium on Aquifer Restoration andGroundwater Monitoring, 26–28 May, Columbus, Ohio: NWWA.

Nielson, D.M., and R. Schalla. 1991. Design and installation of ground-water monitoring wells. In Practical handbook of groundwater mon-itoring, edited by David Nielsen. Chelsea, Mich.: Lewis Publishers.

Seanor, A.M., and L.K. Brannaka. 1983. Efficient sampling techniques.Groundwater Age 17, no. 8:41–46.

Stolzenburg, T.R., and D.G. Nichols. 1986. Effects of filtration methodsand sampling on inorganic chemistry of sampled well water.Proceedings of the Sixth National Symposium and Exposition onAquifer Restoration and Groundwater Monitoring, 216–234. Dublin,Ohio: National Water Well Association.

U.S. Environmental Protection Agency. 1986. RCRA groundwater mon-itoring technical enforcement guidance document. OSWER-9950.1.Washington, D.C.: U.S. Government Printing Office.

———. 1993. Subsurface characterization and monitoring techniques, adesk reference guide, Vol. 1. USEPA/625/R-93/003a (May). U.S. EPA.

Unwin, J.P., and D. Huis. 1983. A laboratory investigation of the purg-ing behavior of small diameter monitor well. Proceedings of the ThirdAnnual Symposium on Groundwater Monitoring and AquiferRestoration, 257–262. Dublin, Ohio: National Water WellAssociation.

Williams, C., and L.G. Evans. 1987. Guide to the selection of cement,bentonite and other additives for use in monitor well construction.Proceedings of First National Outdoor Action Conference, 325–343.Dublin, Ohio: National Water Well Association.

Wilson, L.G., and J.M. Dworkin. 1984. Development of a primer onwell water sampling for volatile organic substances. Bk. 1, Chap.D-2 of U.S. Geological Survey techniques of water resources investi-gations.


Groundwater Cleanup andRemediation

Restoration or cleanup of a contaminated aquifer usuallyinvolves also addressing the contaminated soils at the va-dose zone. The residual contaminants in the vadose needto be treated or removed to prevent the continuous feedof contaminants to groundwater due to leaching, rainfallpercolation, or groundwater table fluctuations. Severaltechniques are available to treat contaminants in the va-dose zone. These techniques include excavation and re-moval, physical treatments, biological treatments, thermaltreatments, and stabilization treatments. Selecting the ap-

propriate method depends on the volume of soils to behandled, the type of soils and contaminants, the regula-tory requirements, and costs.

Excavation and RemovalExcavation and soil removal is one of the most commonactivities in groundwater remediation and cleanup.Excavation involves removing contaminated soil from theunsaturated zone to prevent further groundwater contam-

9.16SOIL TREATMENT TECHNOLOGIES

ination by the residuals present in that zone. Excavationis often used at sites where site conditions preclude onsitetreatment, stabilization, or capping of the contaminatedunsaturated zone. Factors affecting excavation include thevolume of soils to be handled, the location of the area tobe excavated, the type of soils and contaminants, and theregulatory requirements. The excavated material is oftendisposed of at a permitted landfill or treated and reused.

Physical TreatmentThe physical treatments for treating contaminants in thevadose zone include soil–vapor extraction, soil washing,and soil flushing.

SOIL–VAPOR EXTRACTION

This treatment technology removes volatile compoundsfrom the vadose zone. Airflow is injected through extrac-tion wells creating a vacuum and a pressure gradient thatinduces volatiles to diffuse through the extraction wells.The volatiles are collected as gases and treated above-ground. The technology is effective for halogenatedvolatiles and fuel hydrocarbons (U.S. EPA 1991a). Thetechnology is also cost effective when large volumes of soilare involved; since treatment takes place onsite, the risksand costs associated with transporting large volumes ofcontaminated soils are eliminated.

The technology, however, is less effective in soils withlow air permeability, low temperatures, or high carboncontent. In addition, although this technology reduces thevolume of the contaminants, the toxicity of the contam-inants is not reduced.

SOIL WASHING

Soil washing removes adsorbed contaminants from soilparticles. The process involves excavating the contami-nated soil and washing it with a leaching agent, a surfac-tant, or chelating agency or adjusting the pH (U.S. EPA1990c). Sometimes extraction agents are added to enhancethe process. The process reduces the volume of contami-nant; however, residual suspended solids and sludges fromthe process may need further treatment since they containa higher concentration of contaminant than the original.The technology is effective for halogenated semivolatiles,fuel hydrocarbons, and inorganics (U.S. EPA 1993a).

The technology, however, is less effective when the soilcontains a high percentage of silt and clay particles or highorganic content. In addition, this technology reduces thevolume of the contaminants, but the toxicity of the con-taminants is unchanged.

SOIL FLUSHING

Soil flushing is an in situ process whereby environmentalengineers apply a water-based solution to the soil to en-

hance the solubility of the contaminant (U.S. EPA 1991b).The water-based solution is applied through injection wellsor shallow infiltration galleries. The contaminants are mo-bilized by solubilization or through chemical reactions withthe added fluid. The generated leachate must be interceptedby extraction wells or subsurface drains and pumped tothe surface for aboveground treatment. The technology iseffective for nonhalogenated volatile organics and for soilswith high permeability.

The technology, however, is less effective for soils withlow permeability or with particles that strongly adsorbcontaminants such as clays. In addition, special precau-tions are necessary to prevent groundwater contamination.

Biological TreatmentSlurry biodegradation, ex situ bioremediation and landfarming, and in situ biological treatment are biologicaltreatments for treating contaminants in the vadose zone.These treatment are discussed next.

SLURRY BIODEGRADATION

The slurry biodegradation process involves excavating thecontaminated soil and mixing it in an aerobic reactor withwater and nutrients. This process maximizes the contactbetween the contaminants and the microorganisms capa-ble of degrading those contaminants. The temperature inthe reactor is usually maintained at an appropriate level,and neutralizing agents are often added to adjust the pHto an acceptable range (U.S. EPA 1990b). After the treat-ment is complete, the slurry is dewatered, and the soil canbe redeposited on site. This technology is effective for soilscontaminated with fuel hydrocarbons (U.S. EPA 1993a).In addition, the contaminants can be completely destroyedand the soil reused.

The technology, however, is less effective for contami-nants with low biodegradability. In addition, the presenceof chlorides or heavy metals as well as some pesticides andherbicides in the soil can reduce the effectiveness of theprocess by inhibiting the microbial action.

EX SITU BIOREMEDIATION ANDLANDFARMING

This process involves excavating the contaminated soil, pil-ing it in biotreatment cells, and periodically turning it overto aerate the water (U.S. EPA 1993a). The moisture, heat,nutrients, oxygen, and pH are usually controlled in theprocess. In addition, volatile emissions as well as leachatefrom the biotreatment cells should be controlled. The tech-nology is effective for soils contaminated with fuel hydro-carbons. Also, the contaminants can be completely de-stroyed and the soil reused.


IN SITU BIOLOGICAL TREATMENT

This process enhances the naturally occurring biologicalactivities in the contaminated subsurface soil. Circulatingeither a nutrient and oxygen-enriched water-based solu-tion or a forced air movement which provides oxygen inthe soil enhances the naturally occuring microbes (U.S.EPA 1991a). In the latter process, also called bioventing,the air flow rate is lower than in vapor extraction sincethe objective is to deliver oxygen while minimizingvolatilization and the release of contaminants to the at-mosphere. The technology is effective for nonhalogenatedvolatiles and fuel hydrocarbons. In addition, the contam-inant toxicity is reduced or even eliminated. The technol-ogy, however, is less effective for nonbiodegradable com-pounds and for soils with low permeability.

Thermal TreatmentThermal treatment is used to treat contaminants in the va-dose zone and includes incineration and thermal desorp-tion. A brief description of these processes follows.

INCINERATION

Incineration is a process whereby organic compounds incontaminated soil are destroyed in the presence of oxygenat high temperatures (U.S. EPA 1990a). Three commontypes of incinerators are rotary kilns, circulating fluidizedbeds, and infrared incinerators. The excavated contami-nated soil is fed into the incinerator and incinerated at tem-peratures ranging from 1600 to 2200°F. Because the resid-ual ash may contain residual metals, it must be disposedof in accordance with appropriate regulations. In addition,the generated flue gases must be handled with appropri-ate air pollution control equipment.

Incineration is potentially effective for halogenated andnonhalogenated volatiles as well as fuel hydrocarbons andpesticides. Most organic contaminants are destroyed bythis technology; however, metals are not destroyed andend up in the flue gases or the ash. In addition, certaintypes of soils such as clay soils or soil containing rocksmay need screening prior to incineration.

THERMAL DESORPTION

Thermal desorption is a physical separation process inwhich the excavated contaminated soil is heated to a tem-perature at which the water and organic contaminants arevolatilized (U.S. EPA 1991d). The volatilized contaminantsare then sent to a gas treatment system. Low-temperaturethermal desorption is potentially effective for halogenatedsemivolatiles, nonhalogenated volatiles, and pesticides(U.S. EPA 1993a). High-temperature thermal desorption

is effective for halogenated volatiles and semivolatiles aswell as fuel hydrocarbons.

The contaminants, however, are not destroyed by thistechnology and require further gas treatment. In addition,the technology is less effective for tightly aggregated soilsor those containing large rock fragments.

Stabilization and SolidificationTreatmentStabilization and vitrification treatments are also used totreat contaminants in the vadose zone. These treatmentsare described next.

STABILIZATION

The soil stabilization process can be used in either in situor ex situ treatment. The process involves mixing the con-taminated soil with binding materials such as cement, lime,or thermoplastic binders to bind the contaminants to thesoil and reduce their mobility (U.S. EPA 1993b).Depending on the process and binding material, the finalproduct ranges from a loose, soil-like material to concrete-like molded solids. Pretreatment is usually required for soilswith high contents of oil and grease, surfactants, or chelat-ing agents. The process is effective for soils, sludges, orslurries contaminated with inorganics.

The technology, however, is not effective for soils con-taminated with organics or soils with high water or claycontent. Organics, sulfates, or chlorides can interfere withthe curing of the solidified product. Clay can interfere withthe mixing process, adsorbing the key reactants and in-terrupting the polymerization chemistry of the solidifica-tion agents. Furthermore, the stabilization process in-creases the volume of treated soil since reagents are added.

VITRIFICATION

Soil vitrification is used in both in situ and ex situ treat-ment. The process involves inserting large graphite elec-trodes into the soil and applying a high current of elec-tricity to the electrodes (U.S. EPA 1992). The electrodesare typically arranged in 30-foot squares and connectedby graphite on the soil surface. The heat causes a melt thatgradually works downward through the soil incorporat-ing inorganic contaminants into the melt and paralyzingorganic components. After the process is complete and theground has cooled, the fused waste material is dispersedin a chemically inert, stable, glass-like product with lowleaching characteristics.

The technology is potentially effective for halogenatedand nonhalogenated volatiles and semivolatiles as well asfuel hydrocarbons, pesticides, and inorganics. The processreduces the mobility of the contaminants, and the vitrifiedmass resists leaching for geological time periods. The tech-


nology, however, is energy-intensive, and the off-gasesmust be collected and treated before release.

—Ahmed Hamidi

ReferencesU.S. Environmental Protection Agency. 1990a. Mobile/transportable in-

cineration treatment. EPA/540/2-90/014. Washington, D.C.: U.S.EPA.

———. 1990b. Slurry biodegradation. EPA/540/2-90/016 (September).Washington, D.C.: U.S. EPA.

———. 1990c. Soil washing treatment. Engineering Bulletin EPA/540/2-90/017 (September). Washington, D.C.: U.S. EPA.

———. 1991a. Bioremediation in the field. EPA/540/2-91/018.Washington, D.C.: U.S. EPA.

———. 1991b. In situ soil flushing. EPA/540/2-91/021. Washington,D.C.: U.S. EPA.

———. 1991c. In-situ soil vapor extraction treatment. EPA/540/2-91/006 (May). Washington, D.C.: U.S. EPA.

———. 1991d. Thermal desorption treatment. EPA/540/2-91/008(May). Washington, D.C.: U.S. EPA.

———. 1992. Vitrification technologies for treatment of hazardous andradioactive waste. EPA/625/R-92/002 (May). Washington, D.C.: U.S.EPA.

———. 1993a. Remediation technologies screening matrix and referenceguide. EPA/542/B-93/005 (July). Washington, D.C.: U.S. EPA.

———. 1993b. Solidification/stabilization of organics and inorganics.EPA/540/S-92/015 (May). Washington, D.C.: U.S. EPA.


9.17PUMP-AND-TREAT TECHNOLOGIES

Pump-and-treat systems consist of a groundwater with-drawal system and an aboveground treatment system. Thegroundwater withdrawal system, also called the contain-ment system, includes pumping wells or subsurface drainsdesigned to remove the contaminants from the ground-water system and control the plume from further migra-tion. In some cases, injection wells are used to inject treatedwater back into the aquifer. Aboveground treatment sys-tems include chemical, physical, and biological treatmenttechnologies.

Withdrawal and Containment SystemsAs previously stated, the withdrawal and containment sys-tems include well systems and subsurface drains. A de-scription of these systems follows.

WELL SYSTEMS

Well systems remove contaminants from groundwater andstop the plume from further migration by manipulatingthe subsurface hydraulic gradients. Three general classesof well systems are well points, deep wells, and injectionwells. Well points use suction lifting as the standard tech-nique for pumping water; therefore, they can be used onlyfor shallow aquifers where the suction lifting is less than25 feet. Figure 9.17.1 shows several closely spaced pointwells connected to a centrally located suction lift pumpthrough a single main header pipe. Deep-well systems areused for greater depths and are usually pumped individu-ally by submersible pumps. Dual pumps are used for float-ing product recovery as shown in Figure 9.17.2. In injec-tion wells, the injection of clean or treated water into the

aquifer flashes the aquifer or forms a barrier to ground-water flow.

Design Considerations

The design of a well system involves determining the num-ber of wells needed, placing and spacing the wells, and de-termining the pumping cycles and rates of the wells. Thenumber and spacing of the wells should completely cap-ture the plume of contamination and produce as little un-contaminated water as possible to reduce treatment costs.In addition, the well’s capture zones should intersect eachother to prevent dead spots where contaminants stay stag-nant or routes where the contaminant can escape the zoneof capture. Environmental engineers determine the zone ofcapture by plotting the drawdowns within the radius ofinfluence of each well on the potentiometric surface mapof the site and calculating the cumulative drawdowns. Theradius of influence of each well is determined by pumpingtest analysis as discussed in Section 9.8 or estimated fromthe following formulas when pumping test data are lack-ing (Kuffs et al. 1983):

Equilibrium:

R0 5 3(H 2 hw)(0.47K)1/2 9.17(1)

Nonequilibrium:

R0 5 rw 1 (Tt/4790 S)1/2 9.17(2)

where:

R0 5 radius of influence, ftK 5 permeability, gpd/ft2

H 5 total head, fthw 5 head in well, ft

Q 5 pumping rate, gpmrw 5 well radius, ftT 5 transmissivity, gpd/ftt 5 time, minS 5 storage coefficient, dimensionless

Methods of Construction

The construction of a well system involves setting up thedrilling equipment, drilling the well hole, installing casingsand liners, grouting and sealing annular spaces, installingwell screens and fittings, packing gravel and placing ma-terial, and developing the well. Detailed discussions onthese aspects can be found in Johnson Division, UOP, Inc.(1975). In addition, Figure 9.17.3 shows typical well con-struction detail. In recent years, several innovative well in-stallation techniques have been developed including in-

stalling horizontal wells which act as subsurface drains butrequire less soil excavation and disturbances (Oakley et al.1994).

Operation and Maintenance

Equilibrium pumping is often used for plume management;however, nonequilibrium pumping has advantages in casesof floating and sinking plumes and can be used to flushsorbed contaminants associated with the residual phase.Pulsed pumping of recovery wells can be used to washoutresiduals from unsaturated zones, allow contaminants todiffuse out of low permeability zones, and flush and bringstagnant zones into active flow paths. Pulsed pumping,however, incurs additional costs and concerns that mustbe evaluated for site specific conditions (U.S. EPA 1989).

Cost

The costs of well systems vary from site to site dependingon the geology, the depth of the aquifer, the extent andtype of contamination, the periods and durations of pump-ing, and the electrical power requirements. A cost analy-sis study for a variety of well systems can be found inCambel and Lehr (1977) and in Powers (1981).

Advantages and Limitations

Well system technology is an efficient and effective meansof assuring groundwater pollution control. Wells can be


FIG. 9.17.1 Suction lift and a series of point wells. (Reprintedfrom S. Sommer and J.F. Kichens, 1980, Engineering and devel-opment support of general decon technology for the DARCOMinstallation and restoration program, Task 1: Literature reviewon groundwater containment and diversion barriers, Draft re-port by Atlantic Research Corp. to U.S. Army HazardousMaterials Agency, Contract No. DAK 11-80-C-0026, [October],Aberdeen Proving Ground.)

FIG. 9.17.2 Dual pumping wells. (Reprinted from E.K. Nyer,1992, Groundwater treatment technology, 2d ed., New York:Van Nostrand Reinhold.)

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

WellPoint

Saturated SandSubsoil

Water Levelwhile pumping

Wellpoint System

StaticWaterLevel

��

��

��

��

��

��

��

��

��

��

��

��

��

WATERPROBE

WATERTABLE

DEPRESSIONPUMP

AQUIFER

OIL LAYER

WATER TABLE

PRODUCTPROBE

PROBESCAVENGER

PUMP

CLEAN WATER

WATER TABLEDEPRESSION

PUMP CONTROLS

OIL RECOVERY TANK

WATER TABLEDEPRESSION

OIL/WATER SEPARATION

OUTPUT

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

LOCKING STEEL CAPelevation from GS ____

VENTED CAPelevation from GS ____

CEMENT COLLARdepth from GS ____

CASING SEALgrouting material ____depth from GS ____

depth from GS ____

CASING SEALgrouting material ____

WEATHERED BEDROCKdepth from GS _______

COMPONENT BEDROCKdepth from GS _______

depth from GS _______

GROUND SURFACE (GS)

OUTER CASING material ______ diameter ______ schedule ______ total length _____ depth from GS ___

OUTER BOREHOLE diameter ______ depth from GS ___

INNER CASING material _______ diameter _______ schedule _______ total length ______ depth from GS ____

INNER BOREHOLE diameter _______ depth from GS ______

OPEN HOLE diameter _______ total length ______

total depth from GS __________

��

��

��

��

FIG. 9.17.3 Typical well installation.

readily installed, or previously installed monitoring wellscan sometimes be used as part of a well system. In addi-tion, the technology provides high-design flexibility, andthe construction costs can be lower than artificial barriers.However, wells require continued maintenance and mon-itoring after installation; therefore, operation and mainte-nance costs can be high. In addition, the application of thistechnology to fine soils is limited due to the low yield andsmall radius of influence in these soils.

SUBSURFACE DRAINS

Subsurface drains involve excavating a trench and placinga perforated pipe and coarse material such as gravel in thetrench. The drain usually drains by gravity to a sump wherethe water is pumped to the surface for treatment.Subsurface drains essentially function like an infinite lineof extraction wells, creating a continuous zone of depres-sion in which groundwater flows towards the drain. Twotypes of subsurface systems are relief drains and intercep-tor drains. The major difference between these drains isthat the drawdown created by an interceptor drain is pro-portional to the hydraulic gradient, whereas the drawdowncreated by a relief drain is a function of the hydraulic con-ductivity and depth to the impermeable layer below thedrain.

Environmental engineers use relief drains primarily tolower the water table and prevent its contact with wastematerial or to contain a plume in place and prevent con-tamination from reaching a deeper aquifer. Relief drainscan be installed in parallel on either side of a waste site orcompletely around the perimeter of the waste site as shownin Figure 9.17.4. The areas of influence of relief drainsshould overlap to prevent the contaminated groundwaterfrom escaping between the drain lines.

Engineers use interceptor drains to intercept a plumehydraulically downgradient from its source and preventthe contamination from reaching wells and surface waterlocated downgradient from the site. Interceptor drains areinstalled perpendicular to groundwater flow and down-gradient of the plume of contamination as shown in Figure9.17.5. In some cases, engineers use interceptor drains inconjunction with a barrier wall to prevent infiltration ofclean water from downgradient of the drain thereby re-ducing treatment costs (see Figure 9.17.6). A series of in-terceptor drains or collector pipes (laterals) can be con-nected to a main pipe (header) as shown in Figure 9.17.7.


The primary design components of a subsurface drain sys-tem are (1) the location of the drains, (2) the spacing of


��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Waste Disposal Site

Subsurface Drain

Collected GroundwaterPumped to Receiving

Stream

Map View

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

Lowered Groundwater TableUnder Disposal Site

Original Water Table

SubsurfaceDrain

Waste Disposal Site

Cross Section

FIG. 9.17.4 Relief drains around the perimeter of a waste site. (Reprinted from U.S.Environmental Protection Agency, 1985, Leachate plume management, EPA/540/2-85/004,Washington, D.C.: U.S. EPA.)

the drains, (3) the pipe diameter and slope, and (4) the en-velope and filter materials around the pipe.

An interceptor drain should be installed perpendicularto the groundwater flow direction and downgradient fromthe plume of contamination. The drain should be installedon top of a layer of low hydraulic permeability to preventunderflow beneath the drain. The location of the drainshould be selected so that the upgradient and downgradi-ent influences of the drain completely capture the con-tamination plume. The upgradient and downgradient in-fluences of an interceptor drain can be calculated using thefollowing equations described by Van Hoorn andVandemolen (1974) and Kuffs (1983):

Du 5 1.33msI 9.17(3)

Dd 5 }K

Q

I} z (de 2 hd 2 D2) 9.17(4)

where:

Du 5 effective distance of drawdown upgradient, ftms 5 saturated thickness of aquifer not affected by

drainage, ftI 5 hydraulic gradientDd 5 downgradient influence, ftK 5 hydraulic conductivity, ft/dayQ 5 drainage coefficient, ft/day


FIG. 9.17.6 Interceptor drains in conjunction with a barrier wall. (Reprinted fromU.S. Environmental Protection Agency, 1985, Leachate plume management, EPA/540/2-85/004, Washington, D.C.: U.S. EPA.)

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Collected groundwateris pumped to treatmentsystem.

Subsurface Drainage

Waste DisposalSite

Groundwater FlowDirection

Map View

ContaminatedGroundwater

Plume

Waste DisposalSite

Original Water Table

ContaminatedGroundwater

Plume

Cross Section

FIG. 9.17.5 Interceptor drains downgradient of the plume ofcontamination. (Reprinted from U.S. Environmental ProtectionAgency, 1985, Leachate plume management, EPA/540/2-85/004,Washington, D.C.: U.S. EPA.)

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@

@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@

@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀ

ÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ






ÀÀÀÀÀÀÀÀÀ

ÀÀÀÀ

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@

@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@

@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀ

ÀÀÀÀÀÀÀÀÀ












ÀÀÀÀÀÀÀÀÀ

ÀÀÀÀ

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@

@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@

@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀ

ÀÀÀÀÀÀÀÀÀ












ÀÀÀÀÀÀÀÀÀ

ÀÀÀÀ

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@

@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@

@@@@@@@@@

@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀ

ÀÀÀÀÀÀÀÀÀ












ÀÀÀÀÀÀÀÀÀ

ÀÀÀÀ

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@@@

@@@@@@@@@@

@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@

@@@@@@@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ




ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ

ÀÀÀÀÀÀÀÀÀÀÀÀÀ


��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@@@

@@@@@@@@@@

@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@

@@@@@@@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀÀÀ














��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@@@

@@@@@@@@@@

@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@

@@@@@@@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀÀÀ














��

��

��

��

��

��

��

��

��

��

��

��

��

��

@@@@@

@@@@@@@@@@

@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@

@@@@@@@@@@@@@

@@@@@@@@@@

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ÀÀÀÀÀ














��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

ConventionalSubsurface Drain

OriginalWater Table

CleanWater

Rechargingfrom Stream

ContaminatedPlume

Low Permeability

(a) The conventional subsurface drain receives recharges from the stream as well as the leachate plume resulting in larger collection and treatment requirements.

Subsurface Drain withClay or Plastic Barrier

Original WaterTable

(b) One-sided drainage reduces flow to drain.

Low Permeability

de 5 depth of drain, fthd 5 depth of drawdown, ftD2 5 distance from ground surface to water table prior

to drainage at the distance Dd downgradient fromthe drain, ft

The spacing between two parallel relief drains shouldbe selected so that their combined drawdown is adequateto lower the water table beneath the waste. The minimumspacing, however, is often imposed by the boundaries ofthe waste material. The drain spacing depends on the hy-draulic conductivity of the aquifer, the depth of the im-permeable layer beneath the drain, the cross-sectional areaof the drain, the water level in the drain, and precipitationand other sources of recharge. The spacing between twoparallel drains resting on an impermeable barrier can becalculated with the use of the Wasseling (1973) equationas

L 5 1}8KDH

Q

1 4KH2

}20.5

9.17(5)

where:

L 5 drain spacing, ftK 5 hydraulic conductivity, ft/secD 5 distance between the water level in the drain and

the impermeable layer, ftH 5 height of the water table above the water level in

the drain midway between the two drains, ftQ 5 drain drainage rate per unit surface area, ft/sec

For a two-layered soil, Hooghoudt, as described byWasseling (1973), developed a modified equation of

L 5 1}8K1DH

Q

1 4K2H2

}20.5

9.17(6)

where K1 and K2 are the hydraulic conductivities aboveand below the drain, and d is the equivalent depth of theaquifer below the drain as illustrated in Figure 9.17.8.Using this equation involves either a trial-and-error pro-cedure or the use of monographs which have been devel-oped specifically for equivalent depth and drain spacing(U.S. EPA 1985b; Repa et al. 1982). Other equations fordifferent subsurface configurations are available in Cohenand Miller (1983).

The diameter of the pipe can be calculated with the useof Manning’s equation, assuming that the carrying capac-ity of the pipe is equal to the design seepage. The result-ing equation (Luthin 1957) is

d 5 0.892(q z A)0.375 z A20.1875 9.17(7)

where:

d 5 inside diameter of pipe, inA 5 drainage area, acresq 5 seepage coefficient, in/dayI 5 hydraulic gradient


FIG. 9.17.7 Interceptor drains connected to a header. (Reprinted fromU.S. Environmental Protection Agency, 1985, Leachate plume management,EPA/540/2-85/004, Washington, D.C.: U.S. EPA.)

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

CollectionSump

��

��

��

��

��

��

��

��

��

�

��

DischargeLateral

Main

Collector

Dd

Low PermeabilityLayer

RadialFlow

Water TableH

rd

K2

K1

L

HorizontalFlow

FIG. 9.17.8 Subsurface drain formulation. (Reprinted fromU.S. Environmental Protection Agency, 1985, Leachate plumemanagement, EPA/540/2-85/004, Washington, D.C.: U.S. EPA.)

The slope of the pipe should be selected so that the flowvelocity in the drain is greater than the critical velocity ofsiltation of the soil that enters the drain (Soil ConservationService 1973). When the velocity is less than 1.4 ft/sec, fil-ter fabrics and silt traps or cleanouts should be installedaround the pipe and along the subsurface drain.


The construction of a subsurface drain involves trench ex-cavation, dewatering, wall stabilization, pipe installation,and backfilling. Trench excavation is the most significantstep in the construction of a subsurface drain. A varietyof excavation equipment can excavate the trench; the op-timum is determined by the depth, width, length of thetrench, and the type of material being excavated.Dewatering can be performed by open pumping,predrainage using well points, or groundwater cutoff. Wallstabilization methods include shoring for deep excavationsor open cuts for shallow excavations. Continuous trench-ing machines can accomplish all excavation and pipe in-stallation operations simultaneously (Oakley et al. 1994);however, this machinery is limited to small diameter andrelatively shallow subsurface drains.

Another important aspect in subsurface drain installa-tion is the placement of filter and envelope materialsaround the pipe to prevent soil particles from entering andclogging the pipe. Geotextiles and well-graded sand andgravel can be used as filter materials. The general re-quirement for envelopes is that their hydraulic conductiv-ity is higher than that of the base material. Design proce-dures for filters and envelopes are in the Soil ConservationService (1973).


Subsurface drains require frequent inspection and mainte-nance during the first year or two of operation. Typicalproblems that can develop in drainage systems and requiremaintenance include clogging of the drain or manhole bysediment buildup or buildup of chemical compounds suchas iron and manganese. Clogged pipes can be cleaned byhydraulic jetting, mechanical scrapping, or chemical treat-ment in cases of chemical buildup.

Cost

Costs of subsurface drain systems vary from site to site de-pending on the geology, the depth of the aquifer, the ex-tent and type of contamination, the periods and durationsof pumping, and the electrical power requirements. Themajor costs of a subsurface drain system occur during sys-tem installation. These costs include excavation, dewater-ing, pipe bedding, filter and envelop materials, pipes, man-holes, and pumps. Typical costs for subsurface draininstallation are in Means (1994).


For shallow contaminations, subsurface drains are morecost-effective than wells particularly in aquifers with lowor variable permeability. Construction methods for sub-surface drains are simple, and operation costs are relativelylow since flow to the underdrain is by gravity. In addition,subsurface drains provide considerable design flexibilitysince adjusting the depth or modifying the envelope ma-terial can alter the spacing to some extent. However, sub-


GROUNDWATERWELLS

COAGULANT POLYMER POLYMER

STRIPPER

FLASH MIXER

CONTAMINATED GROUNDWATER FROM SLUDGE DEWATERING

GRANULAR ACTIVATED CARBONADSORBERS

RIVERDISCHARGE

SLUDGEPUMP

SLUDGE TOSANITARYLANDFILL

HYDROGENPEROXIDE

ANDCHLORINEADDITIONS

AIRSTRIPPER

ANDFLASHMIXER

CHLORINEDIOXIDE

ADDITION

PULSATINGCLARIFIER

TWOSAND

FILTERS

STORAGERESERVOIR

FIRSTSTAGE

SECONDSTAGE

GACADSORBERS

SLUDGESTORAGE

TANK

SLUDGEDEWATERING

CHAMBER

CIO2

CI2H2O2

GR

OU

ND

WA

TE

RP

UM

PIN

GW

ELL

S

BLOWER

FIG. 9.17.9 Typical treatment trains. (Reprinted from North Atlantic Treaty Organization, 1993, Demonstrationof remedial action technologies for contaminated land and groundwater, Vol. 2, pt. 2, no. 190, EPA/600/R-93/012,NATO Committee on the Challenges of Modern Society.)

surface drains are not suited to poorly permeable soils, todeep contaminant plumes, or beneath existing sites. In ad-dition, subsurface drains require continuous and carefulmonitoring to assure adequate leachate collection and pre-vent pipe clogging.

Treatment SystemsThe most commonly used treatment in pump-and-treattechnologies is physical treatment. Physical treatment in-cludes density separation, filtration, adsorption, air strip-ping, and reverse osmosis. Each of these processes can beused individually or in conjunction with others (e.g., treat-ment trains) as shown in Figure 9.17.9. In addition, mosttreatment systems include equalization and spill control toprotect the treatment works from shock pollutants and hy-draulic loadings. The selection of the process is usuallybased on the type of contaminant, influent concentration,effluent requirements, and cost.

DENSITY SEPARATION

Density separation is a process whereby the water and con-taminant are separated based on their individual densities.This treatment is often used for the pretreatment of sus-pended solids or floating immiscible products that couldbe present in pumped groundwater. For suspended solids,the most commonly used equipment is clarifiers, settlingchambers, and sedimentation basins as discussed inChapter 8. For immiscible products, such as oil and grease,the most commonly used equipment is oil–water separa-tors. Both suspended solids and oil and grease must gen-erally be removed from contaminated groundwater priorto further treatment because these materials can foul in-struments and interfere with other processes. Furthermore,oil and grease and suspended solids can damage the envi-ronment and cause a significant pollution problem to thereceiving body of water.

The most common oil–water separators are theAmerican Petroleum Institute (API) gravity separators andthe parallel-plate separators. The design of an oil–waterseparator is based on the amount of oil present in the wa-ter, the oil droplet size distribution, the presence of sur-factants, the specific gravity of the oil, and the water tem-perature. A step-by-step procedure for the design of anoil–water separator is in Corbitt (1990). Once the oil orfloating product is at the surface, it can be removed fromthe water by slotted pipes, dip tubes, or belt or rope skim-mers.

FILTRATION

Filtration is a process whereby suspended solids are re-moved from the influent by forcing the water through afilter of porous medium such as sand or sand with an-thracite or coal. The purpose of filtration is to reduce the

concentration of suspended solids, such as carbon col-umns, prior to certain treatment processes. The most com-mon filter is a dual-media system with a layer of anthraciteover a layer of sand. This filter provides better suspendedsolids removal with longer filter runs at higher flow ratesthan the more conventional single-medium filter (Corbitt1990). The design of filters is based on the flow rate, flowscheme, and the type of medium used in the filter as dis-cussed in Chapter 8. Up to 75% of suspended solids canbe removed by dual-media filters operating at flow ratesranging from 2 to 8 gpm/ft (Oakley et al. 1994), bed depthsof 24 to 48 inches, sand to anthracite ratios of 1:1 to 4:1,and a filter run of 8 to 148 hours (Corbitt 1990).

Filtration is a reliable and effective means of removinglow levels of solids provided that the solid content doesnot vary greatly. Also, periodic filter backwashing is nec-essary to remove collected materials from the media. Typ-ical backwash flow rates are 15 to 25 gpm/ft (Oakley etal. 1994) for eight to ten minutes (Corbitt 1990). The spentbackwash water can be routed to the plant’s headworksor to an intermediate process which provides settling.

CARBON ADSORPTION

In adsorption, the molecules of a dissolved contaminantbecome attached to the surface of a solid adsorbent. Themost widely used adsorbent is granular activated carbon(GAC) because its porous structure provides a relativelylarge surface area per unit volume (1000–2000 m2/g).Collection of the molecules on the surface of the adsor-bent is due to chemical or physical forces. Chemical ad-sorption is due to actual chemical bonding at the solid’ssurface. Physical adsorption is due to van der Waals’forces, which are weak bonds compared to chemical ad-sorption. However, because of the weak nature of thesebonds, adsorbed molecules can be easily removed with achange in the solute concentration or the addition ofenough energy (regeneration) to overcome the bonds. Thiscapacity to remove certain molecules adsorbed on carbonand, thus, the possibility for repeated carbon reuse is whatallows activated carbon adsorption to be a cost-effectivetechnology.

Environmental engineers commonly use carbon ad-sorption to remove organic contaminants from water orair; however, they also use it to remove a limited numberof inorganic contaminants as shown in Table 9.17.1. Theeffectiveness of GAC depends on the molecular weight,structure, and solubility of the contaminant as well as theproperties of the carbon, the water temperature, and thepresence of impurities such as iron and manganese. Theinfluence of each of these parameters on the absorbabilityof organic contaminants is shown in Table 9.17.2. Asshown in this table, carbon adsorption is suitable for highmolecular weight and low solubility and polarity com-pounds (U.S. EPA 1988), such as chlorinated hydrocar-bons, organic phosphorous, carbonates, PCBs, phenols,


and benzenes. GAC can also be used in conjunction withother treatment technologies. For example, GAC can beused to treat the effluent water or offgas from an air strip-per (Crittenden 1988).


The most important variables in designing carbon treat-ment systems are the contact time and the carbon usage

rate, both of which depend on the flow rate, type of con-taminant, and influent and effluent concentrations. Thecontact time is the time allowed for the pollutant to reactwith the carbon exterior and enter and react with the sur-face of the interior pores. The contact time is the result ofdividing the volume of carbon by the flow rate. The car-bon usage rate is the result of dividing the volume of car-bon online by the volume of water treated when the re-quired effluent concentration is exceeded (i.e., breakpoint).


TABLE 9.17.1 POTENTIAL FOR REMOVAL OF INORGANICMATERIAL BY ACTIVATED CARBON

Potential forConstituents Removal by Carbon

Metals of high sorption potentialAntimony Highly sorbable in some solutionsArsenic Good in higher oxidation statesBismuth Very goodChromium Good, easily reducedTin Proven very high

Metals of good sorption potentialSilver Reduced on carbon surfaceMercury CH3HgCl sorbs easily Metal filtered out

CobaltTrace quantities readily sorbed, possibly

as complex ionsZirconium Good at Low pH

Elements of fair-to-good sorptionpotential

Lead GoodNickel FairTitanium GoodVanadium VariableIron FE31 good, FE21 poor, but may

oxidizeElements of low or unknownsorption potential

Cooper Slight, possibly good if complexedCadmium SlightZinc SlightBeryllium UnknownBarium Very lowSelenium SlightMolybdenum Slight at pH 6–8, good as complex ionManganese Not likely, except as MnO4

Tungsten SlightFree halogens

F2 fluorine Will not exist in waterCl2 chlorine Sorbed well and reducedBr2 bromine Sorbed strongly and reducedI2 iodine Sorbed very strongly, stable

HalidesF, flouride May sorb under special conditionsCl2, Br2, I2 Not appreciably sorbed

Source: U.S. Environmental Protection Agency, 1985, Handbook, remedial action at waste dis-posal sites, EPA/625/6-85/006 (Washington, D.C.: U.S. EPA).

The goal of the design is to find the optimum contact timewhich provides the lowest carbon usage rate. Typical de-sign parameters for carbon adsorption are shown in Table9.17.3.

The contact time and carbon usage rate for a compoundare usually determined through laboratory testing. A com-mon test method is the bed depth service time (BDST)analysis, also called the dynamic column test study (Adamsand Eckenfelder 1974). In this test method, three to fourcolumns are connected in series as shown in Figure9.17.10. Each column is filled with an amount of carbonwhich provides superficial contact times of fifteen to sixtyminutes per column. Effluent from each column is ana-

lyzed for the chemicals of concern, and the effluent-to-in-fluent concentration ratio is plotted against the volume ofwater treated by each column. Figure 9.17.11 shows anexample of a dynamic test where four columns are usedand each column represents fifteen minutes of contact timeTc. The curves obtained are called breakthrough curvessince they represent the amount of contaminated waterthat has passed through the carbon bed before the maxi-mum allowable concentration appears in the effluent.

Once the breakthrough curves are determined, the car-bon usage rates can be calculated as:

qc 5 }V

V

w

c} 9.17(8)


TABLE 9.17.2 SUMMARY OF INFLUENCE OF CONTAMINANTPROPERTIES ON THE ABSORBABILITY OFORGANICS

Parameter Influence on Absorbability

Molecular weight High molecular-weight compounds adsorbbetter than low molecular-weight compounds.

Solubility Low-solubility compounds are adsorbed betterthan high-solubility compounds.

Structure Nonpolar compounds adsorb better than polarcompounds.

Branched chains are usually more adsorbablethan straight chains.

Large molecules are more adsorbable thansmall molecules.

Substituent group Hydroxyl generally reduces absorbability.Amino generally reduces absorbability.Carbonyl effect varies according to host

molecule.Double bonds effect varies.Halogens effect varies.Sulfonic usually decreases absorbability.Nitro often increases absorbability.

Temperature Adsorptive capacity decreases when the watertemperature increases.

Properties of carbon Adsorption is directly proportional to thesurface area of the carbon used.

Virgin carbon has more adsorptive capacitythan regenerated carbon.

Other Iron and manganese (if present at significantlevels in the water) can precipitate onto thecarbon, clog its pores, and cause rapid headloss.

Biological growth on the surface of the carboncan enhance the removal efficiency andincrease the carbon service life. If the growthis excessive, however, it can clog the carbonbed.

Excessive amounts of suspended solids (above50 ppm) or oil and grease (above 10 ppm) canaffect the efficiency of the carbon.

Source: U.S. Environmental Protection Agency, 1985, Handbook, remedial action at waste dis-posal sites, EPA/625/6-85/006 (Washington, D.C.: U.S. EPA).

where:

qc 5 carbon usage rate, lb/galVc 5 volume of carbon, lbVw 5 volume of water treated when the required efflu-

ent concentration is exceeded, gal

Carbon usage rates are then plotted for each contacttime as shown in Figure 9.17.12. The optimum contacttime tcopt is determined as the time which provides the low-est carbon usage rate. The optimum volume of carbon bedneeded is calculated as

Vcopt 5 Tcopt z Q 9.17(9)

where:

Vcopt 5 optimum volume of carbon bed, lbQ 5 flow rate, gal

The optimal tradeoff point between a lower carbon us-age rate and a smaller carbon bed size can be foundthrough analysis. A typical minimum contact time forgasoline contaminants is fifteen minutes. This contact timecorresponds to a liquid loading rate of 2 gpm/ft2 in a stan-dard 20,000-lb and 10-ft-diameter carbon vessel (Noonanand Curtis 1990). Table 9.17.4 lists the contact times aswell as carbon usage rates for several organics.


GAC is available from a number of suppliers in vessels ofdifferent sizes. The vessels are typically open-top, cylin-drical steel tanks for gravity systems and closed-top, cylin-drical steel tanks for pressure systems. Gravity systems areoperated like sand filters and are generally used for highflows, such as at municipal wastewater treatment plants.Pressure systems are generally used for smaller flows andallow higher surface loading rates (5–7 gpm/ft2 comparedto 2–4 gpm/ft2 for gravity systems) and pressure dischargeto the distribution system, saving pumping costs (Noonanand Curtis 1990).

Activated carbon is commonly made from coal; othermaterials such as coconut shells, lignite, wood, tires, andpulp residues can also be used. In the formation of GAC,the material is subjected first to a high temperature to re-move water and other vapors from it. Then, a superheatedsteam is released into the material (activation) to enlargethe pores and remove ashes from it (Noonan and Curtis1990).


TABLE 9.17.3 TYPICAL DESIGN PARAMETERS FORCARBON ADSORPTION

Parameters Requirements

Contact time Generally 10–50 min; may beas high as 2 hours for someindustrial wastes

Hydraulic load 2–15 gpm/ft2 depending ontype of contact system; seeTable 9.17.1

Backwash rate Rates of 20–30 gpm/ft2 usuallyproduce 25–50% bedexpansion

Carbon lossduring 4–9%regeneration 2–10%

Weight of COD 0.2–0.8removed per weightof carbon

Carbon requirementsPCT plant 500–1800 lb/106 galTertiary plant 200–500 lb/106 gal

Bed depth 10–30 ft

Source: U.S. Environmental Protection Agency, 1985, Handbook, remedialaction at waste disposal sites, EPA/625/6-85/006 (Washington, D.C.: U.S. EPA).

FIG. 9.17.10 Dynamic column test. (Reprinted from E.K. Nyer, 1992, Groundwater treatment technol-ogy, 2d ed., New York: Van Nostrand Reinhold.)

Sampleout

Sampleout

Sampleout

Sampleout

Backwash(to drain)

4-in heightshot gravel

Filter meshStrainer

Output

Flow-controlvalve

Samplevalve

To drain

Bleedoff valve

Sample in

Pump

Backwash inletFlow-control

valve

Carbon

Carbon Carbon Carbon


Activated carbon systems can be operated as upflow, ex-panded-bed columns or downflow, fixed-bed columns.Upflow expanded beds can tolerate higher suspended-solids loading than downflow beds. They also make effi-cient use of the carbon since fully exhausted carbon canbe removed from the bottom of the bed while fresh car-bon is added to the top. In addition, carbon beds can beoperated in parallel (single-stage) or in series (multiple-stage) as shown in Figure 9.17.13. When operated in se-ries, the leading contactor removes the majority of the con-tamination, while the second contactor removes anyresidual organics from the water. Furthermore, multiple-stage use allows a contactor to be completely exhaustedbefore regeneration, while effluent quality remains pro-tected by the subsequent contactor. When operated in par-allel, contactors should stagger startup to permit bed-by-bed regeneration without reducing effluent quality.

When the adsorption capacity of the carbon is ex-hausted, the spent carbon can either be disposed of at adisposal site, regenerated, or reactivated for reuse. Offsitedisposal at a landfill or an incinerator is the preferredmethod when the amount of carbon is small. For disposalat a landfill, testing and classifying the spent carbon arenecessary to ensure that all regulations for disposal are be-ing met. Spent carbon may be considered hazardous wasteand may need to be disposed of at a hazardous waste land-fill or burned at an incinerator where both the carbon andthe hazardous waste are destroyed.

If the amount of spent carbon is large or the user hasaccess to an offsite, multiuser facility, regeneration or re-activation for reuse may be the preferred solution.Regeneration exposes the spent carbon to steam to desorbthe contaminants. Reactivation is conducted in electricalor multiple-heart furnaces where the temperature is highenough (up to 1800°F) to thermally destroy the contami-nants and reactivate the carbon. Regeneration and reacti-vation can incur a 10 to 20% material loss and can changethe adsorptive properties of the virgin grade material.

Cost

The capital costs of a GAC system include the costs of car-bon, carbon vessels, pumps and piping, electrical equip-ment and controls, housing, design, and contingencies. Thecost depends on the flow rates, type of contaminant, con-centrations, and discharge requirements. Costs can varyfrom $0.10–1.50/1000 gal treated for flow rates of 100mgd to $1.20–6.30/1000 gal treated for flow rates of 0.1mgd (O’Brien 1983).

Operation and maintenance costs include labor, energy,carbon replacement, and sampling and monitoring. Themajor cost, however, is carbon replacement which is afunction of the carbon usage rate. Typical carbon costs


FIG. 9.17.11 Dynamic column test results breakthrough curves.(Reprinted from E.K. Nyer, 1992, Groundwater treatment technology, 2ded., New York: Van Nostrand Reinhold.)

Col. 115 Min.

Col. 230 Min.

Col. 345 Min.

Col. 460 Min.

VOLUME TREATED

% A

DS

OR

BA

TE

BR

EA

KT

HR

OU

GH

100

75

50

25

FIG. 9.17.12 Optimum carbon contact time. (Reprinted fromE.K. Nyer, 1992, Groundwater treatment technology, 2d ed.,New York: Van Nostrand Reinhold.)

15 30 45 60CONTACT TIME (MIN.)

CA

RB

ON

US

AG

E R

AT

E

range from $0.60 per pound for regenerated carbon to$0.75 per pound for virgin, high-quality carbon (Noonanand Curtis 1990).


Carbon adsorption is an effective and simple treatmenttechnology for volatile organic compounds. In addition,GAC can be used in conjunction with other treatmenttechnologies.

However, GAC is not recommended for low-molecu-lar-weight and high-polarity compounds. In addition, high-suspended solids, oil and grease, and a high concentrationof iron and manganese can foul the carbon and requirefrequent backwashing. GAC showed poor adsorption ca-pacity for wastewaters with high fatty acids (i.e., leachatefrom young landfills) or wastewaters with high BOD/CODand COD/TOC ratios (U.S. EPA 1977). Furthermore, theamount of carbon required, the frequency of regenerationand reactivation, and the potential need to handle the dis-


TABLE 9.17.4 CARBON ADSORPTION WITH PPM INFLUENT LEVELS

Typical Typical Total CarbonInfluent Effluent Surface Contact Usage

System Conc. Conc. Loading Time Rate OperatingNo. Contaminants (mg/liter) (mg/liter) (gpm/ft2) (min) (lb/1000 gal) Mode

1 Phenol 63.45 ,1 1.0 201 5.8 Three fixed bedsOrthorchlorophenol 100.45 ,1 in series

2 Chloroform 3.45 ,1 0.5 262 11.6 Two fixed bedsCarbon tetrachloride 135.45 ,1 in seriesTetrachloroethylene 3.45 ,1Tetrachloroethylene 70.45 ,1

3 Chloroform 0.85 ,1 2.3 58 2.8 Two fixed beds inCarbon tetrachloride 10.05 ,1 seriesTetrachloroethylene 15.05 ,1

4 Benzene 0.45 ,1 1.21 112 1.9 Two fixed beds inTetrachloroethylene 4.55 ,1 series

5 Chloroform 1.45 ,1 1.6 41 1.15 Two fixed beds inCarbon tetrachloride 1.05 ,1 series

6 Trichloroethylene 3.85 ,1 2.4 36 1.54 Two fixed beds inXylene 0.2–0.5 ,1 seriesIsopropyl Alcohol 0.25 ,10Acetone 0.15 ,10

7 Di-isoproply methyl 1.25 ,50 2.2 30 0.7 Single fixed bedphosphonate

Dichloropentadiene 0.45 ,101 1,1,1-Trichloroethane 143.55 ,1 4.5 15 0.4 Single fixed bed

Trichloroethylene 8.45 ,1 in seriesTetrachloothylene 26.55 ,1

2 Methyl T-butyl ether 30.55 ,5 5.7 12 0.62 Two single fixedDi-isopropyl ether 35.55 ,1 beds

3 Chloroform 400.55 ,100 2.5 26 1.19 Four single fixedTrichloroethylene 10.55 ,1 beds

4 Trichloroethylene 35.55 ,1 3.3 21 0.21 Three single fixedTetrachloroethylene 170.55 ,1 series

5 1,1,1-Trichlorethane 70.55 ,1 4.5 30 0.45 Two fixed beds in1,1-Dichloroethylene 10.55 ,1 series

6 Trichlorethylene 25.55 ,1 2.0 35 0.32 Single fixed bedCis-1,2-dicloroethylene 15.55 ,1

7 Trichlorethylene 50.55 ,1 1.6 42 0.38 Two single fixedbeds

8 Cis-1,2-dichloroethylene 5.55 ,1 1.91 70 0.25 Two fixed beds inTrichloroethylene 5.55 ,1 seriesTetrachloroethylene 10.55 ,1

Source: R.P. O’Brien, 1983, There is an answer to groundwater contamination, Water/Engineering and Management (May).

carded carbon as a hazardous waste make GAC a rela-tively expensive technology.

AIR STRIPPING

Air stripping is a mass-transfer process whereby volatilecontaminants are stripped out of the aqueous solution andinto the air. The process exposes the contaminated waterto a fresh air supply which results in a net mass transferof contaminants from the liquid phase to the gaseousphase. Contaminants are not destroyed by air strippingbut rather are transferred into the air stream where theymay need further treatment. Air stripping applies tovolatile and semivolatile organic compounds. It does notapply to low volatility compounds, metals, or inorganiccontaminants.

Several types of air stripping technologies are availableincluding tray aeration, spray aeration, and packed tow-ers. Among these technologies, packed tower aeration

(PTA) is the most commonly applied to remove volatileorganics from groundwater. In a packed tower, the con-taminated water comes in contact with a countercurrentflow of air. The packing material in the tower breaks thewater into small droplets and thin films causing a largecontact area where the mass transfer can take place. Figure9.17.14 shows a typical treatment process using air strip-ping.

Design Consideration

The design of an air stripper is based on the flow rate, typeof contaminant, concentration, temperature, and effluentrequirements. The major design variables are the type ofpacking, gas pressure drop, and air-to-water ratio. Giventhose design variables, environmental engineers can deter-mine the gas and liquid loading rates, tower diameter, andpacking height by using the following mass-balance equa-tion (Noonan and Curtis 1990):


FIG. 9.17.13 Single-stage and multiple-stage contactors. (Reprinted from U.S. Environmental ProtectionAgency, 1985, Handbook, remedial action at waste disposal sites, EPA/625/6-85/006, Washington, D.C.: U.S.EPA.)

��

��

��

��

��

��

��

��

��

out

inout

in

out

in

Upflow Expanded in SeriesDownflow in ParallelDownflow in Series

��

in

out

Moving Bed

FIG. 9.17.14 Typical treatment process using air stripping. (Reprinted from U.S. EnvironmentalProtection Agency, 1991, Air stripping of aqueous solutions, Engineering Bulletin, EPA/540/2-91/022,Washington, D.C.: U.S. EPA.)

Gas

Liquid

EFFLUENTTREATMENT

(4)

OFFGAS TREATMENT(5)

HEAT EXCHANGERS

(optional)(2)

PRE-TREATMENT

STORAGETANKS

(1)

AIR STRIPPER(3)

Pump AirBlower

Packed Bed

Mist Eliminator

Stack

Treated LiquidRecycle (optional)

FeedContaminatedGroundwater

orSurface Water

��Water

Air

STRIPPEROFFGAS

Water Spray

Zt 5 }K

L

1a} 3}R

R

2 1}4 z ln 9.17(10)

D 5 3}4

p

Q

L}4

0.59.17(11)

where:

Zt 5 depth of packing, mD 5 diameter of the tower, mL 5 liquid loading rate, m3/m2/secK1a 5 overall liquid mass-transfer coefficient, sec21

R 5 stripping factor, dimensionlessCi 5 influent concentration, mg/lCe 5 effluent concentration, mg/lQ 5 flow rate m3/sec

The key variables to define in the preceding equationsare the overall mass-transfer coefficient K1a and the strip-ping factor R. The mass-transfer coefficient is a functionof the type of packing, the liquid and gas flow rates, andthe viscosity and density of the water. Therefore, the mass-transfer coefficient is usually determined from a pilot teston actual field data. When pilot testing is not feasible, the-oretical correlations, such as those developed by Onda,Takeuchi, and Okumoto (1968), can be used.

The stripping factor R is related to the air–water ratioas follows (Noonan and Curtis 1990):

R 5 }(

(

G

G

/

/

L

L

)

)

m

ac

in

t} 9.17(12)

}C

C

e

i} (R 2 1) 1 1

}}R

(G/L)min 5 }H

1} }

Ci 2

Ci

Ce} 9.17(13)

where:

(G/L)min 5 minimum air–water ratio, dimensionless(G/L)actual 5 actual air–water ratio, dimensionlessG 5 gas (air) loading rate, m3/m2/secL 5 liquid loading rate, m3/m2/secH 5 Henry’s constant, dimensionless

The actual air–water ratio, however, is related to thegas pressure drop through the column as shown in Figure9.17.15 (brand-specific pressure drop curves are availablefrom packing vendors). Therefore, engineers should ex-amine several combinations of air–water ratio and pres-sure drop to determine the most cost-effective design. Ahigh pressure drop reduces the size of the tower and cap-ital costs; however, it increases the size of the blower andoperation costs. Studies have shown that the most cost-ef-fective stripping factor R usually falls between 3 and 5(Hand et al. 1986).

After a stripping factor is selected, the actual air–waterratio can be calculated with Equation 9.17(12), and thegas (air) loading rate can be obtained from Figure 9.17.15for a given pressure drop. Then the tower height and di-ameter can be calculated with Equations 9.17(10) and9.17(11), respectively. This procedure should be repeatedfor several combinations of stripping factor and pressuredrop until the most cost-effective design is obtained.Several computer cost models can be used in this process(Nirmalakhandan, Lee, and Speece 1987; Cummins andWestrick 1983; Clark, Eilers, and Goodrick 1984).


Approximateflooding

0.4

0.2

0.10

0.08

0.06

0.04

0.02

0.01

0.008

0.006

0.004

0.002

0.001

0.01 0.02 0.04 0.1 0.2 0.4 1.0 2 4 10

50

1200800

100

400

200

Gas pressure drop

N/m2

m

N/m2

m= 6.37 ´ 10-3

= 1.224 ´ 10-3 N/m2

minH2O

ft

lbf /ft2

ftDpZ

rg

rL - rG

L¢G¢

1/221

r G 1r

L -

r G2 g

c

G¢2 C

fmL0.

1 J

FIG. 9.17.15 Generalized pressure drop curves. (Reprinted from R.E.Treybal, 1980, Mass transfer operations, 3d ed., New York: McGraw-Hill.)


The components of a stripping tower include the towershell, tower internals, packing, and air delivery systems.The tower shell can be made of aluminum, fiberglass, stain-less steel, or coated carbon steel. Selecting the shell con-struction material is usually based on cost, structuralstrength, resistance to corrosion, and esthetics. Table9.17.5 shows the advantages and disadvantages of severalmaterials of construction.

Tower internals include the water distributor system in-side the tower, the mist eliminator system, and the air ex-haust ports. The environmental engineer should select thetype of components that ensure optimal mass-transfer con-ditions at the most economical cost.

The packing material is an important component of theair stripping tower. Several types of packing materials arecommercially available including plastic, metal, or ceramic

(Perry and Green 1984). The selection is based on mate-rials exhibiting a high mass-transfer rate and a low gaspressure drop. Plastic packings are often used because oftheir low price, corrosion resistance, and light weight.Table 9.17.6 shows the physical characteristics of com-mon packing materials.

Other components of an air stripping system includethe blower, noise control devices, and air filters. Theblower is designed based on the air–water ratio and canbe mounted on top of the tower or at the base. Soundmufflers control noise, and air filters prevent contact be-tween the water and the air outside the tower.


In a packed air stripping tower, the water flows counter-current to the air stream which is introduced at the bot-


TABLE 9.17.5 CONSTRUCTION MATERIALS FOR TOWER SHELLS WITH A PACKED AIR STRIPPER

Material Advantages Disadvantages

Aluminum Lightweight Poor resistance to water with pHLow cost less than 4.5 and greater than 8.6Corrosion resistant Pitting corrosion occurs in theExcellent structural properties presence of heavy metalsLong life (. fifteen years) Not well suited to high chlorideNo special coating required water

Carbon Steel Mid-range capital cost Requires coating inside andGood structural properties outside to prevent corrosion,Long life if properly painted leading to increased maintenance

and maintained Heavier than aluminum or FRPFiberglass Low cost Poorly defined structural

High chemical resistance to propertiesacidic and basic conditions, Short life (, ten years) unless morechlorides, and metals expensive resins used

Poor resistance to ultraviolet (UV) light(can be overcome with specialcoatings that must be maintained)

Requires guy wires in most situationsSusceptible to extremes of temperature

differential disturbing tower shapeand interfering with distribution

Stainless Steel Highly corrosion resistant Most expensive material forExcellent structural properties prefabricated towersLong life (. twenty years) Susceptible to stress fracture corrosion inNo special coating required the presence of high chloride levels

Concrete Aesthetics Difficult to cast in one place leading toLess prone to vandalism potential difficulties with cracks and leaks

More expensive than self-supportingprefab towers

Metal lined block Aesthetics More expensive than self-supportingand brick Less prone to vandalism prefab towers

Prefab air stripper insert eliminates problemsassociated with cast in place towers

Source: K.E. Nyer, 1992, Groundwater treatment technology, 2d ed., New York: Van Nostrand Reinhold.

tom of the tower. In some configurations (e.g., induceddraft systems), the air is drawn through the tower by theblower instead of being forced. The water supply pumpsusually control the blowers to coordinate the air and wa-ter flows. The offgas from an air stripper may need to betreated, depending on air emission requirements, with theuse of granular activated carbon, catalytic oxidation, orincineration (U.S. EPA 1985a). The liquid effluent fromthe air stripper may contain trace amounts of contami-nants which can be treated by GAC.

Maintenance of air stripping systems is minimal andusually involves the blower. However, periodic inspectionof the packing is required if the water contains high lev-els of iron, suspended solids, or microbial population.

During the aeration process, dissolved iron and manganesecan be oxidized and deposited on the packing material.This deposit can build up and clog the packed bed and,therefore, reduce system efficiency. Pretreating the influ-ent can control iron deposition. A high microbial popula-tion can lead to a biological build up within the packedbed and reduce system performance. This problem can alsobe prevented through pretreatment of the influent.

Cost

The capital costs of an air stripper include the costs of thetower shell, packing, tower internals, air delivery system,electrical equipment and controls, housing, design, andcontingencies. The addition of an air treatment systemroughly doubles the cost of an air stripping system (Lenzoand Sullivan 1989; U.S. EPA 1986a). The cost depends onthe flow rate, volatility of the contaminant, concentration,and removal efficiency. Costs vary from $0.07–0.70/1000gal for Henry’s law coefficients of 0.01–1.0 to $7.00/1000gal for Henry’s law coefficients lower than 0.005 (Adamsand Clark 1991).


Air stripping is a proven technology for treating water con-taminated with volatile and semivolatile organic com-pounds. Removal efficiencies of greater than 98% forvolatile organics and greater than or equal to 80% forsemivolatile compounds have been achieved. Recent de-velopments in this technology include high temperature airstripping and air rotary stripping to increase removal effi-ciencies (Bass and Sylvia 1992). The use of diffused air orbubble aeration air strippers for flows less than 50 gpmhave also increased during the last five years.

The air stripping technology, however, is not effectivein treating low volatility compounds, metals, or inorgan-ics. Air emissions of volatile organics from the air strippermay need a separate treatment. In addition, the removalefficiency of air strippers is reduced for aqueous solutionswith high levels of suspended solids, iron, manganese, ormicrobial population. Periodic cleaning of the packing ma-terial removes the deposits of these products.

OXIDATION AND REDUCTION

In chemical oxidation, the oxidation state of a contami-nant is increased by the loss of electrons, while the oxi-dation state of the reactant is lowered. Conversely, in re-duction, the oxidation state of a contaminant is decreasedby the addition of electrons. Oxidizing or reducing agentscan be added to contaminated water to destroy, detoxify,or convert the contaminants to less hazardous compounds.Many hazardous substances including various organics,sulfites, soluble cyanide- and arsenic-containing com-pounds, hydroxylamine, and chromates can be oxidized


TABLE 9.17.6 PHYSICAL CHARACTERISTICS OFCOMMON PACKING MATERIALS

Surface Area Void Space PackingType Size (ft3/ft3) (%) Factora

Dumped PackingsGlitsch 0A 106 89 60Mini-rings 1A 60.3 92 30(Plastic) 1 44 94 28

2A 41 94 282 29.5 95 153A 24 95.5 12

Tellerettes 10(#1) 55 87 40(Plastic) 20(2-R) 38 93 18

30(3-R) 30 92 1630(2-K) 28 95 12

Intalox 10 63 91 33Saddles 20 33 93 21(Plastic) 30 27 94 16Pall rings Gk0 104 87 97(Plastic) 10 63 90 52

1As0 39 91 402As0 31 92 253As0 26 92 16

Raschig rings As0 111 63 580(Ceramic) Df 0 80 63 255

10 58 73 1551As0 38 71 9520 28 74 6530 19 78 37

Jaegar 10 85 90 28Tri-Packs 20 48 93 16(Plastic) 3As0 38 95 12

Stacked PackingDelta — 90 98 —(PVC)Flexipac Type 1 170 91 33(Plastic) Type 2 75 93 22

Type 3 41 96 16Type 4 21 98 9

Source: R.E. Treybal, 1980, Mass transfer operations, 3d ed., New York:McGraw-Hill.

or reduced to forms which are more readily removed fromgroundwater (Huibregts and Kastman 1979).

Chemical Oxidation

Chemical oxidation involves adding oxidizing agents tothe contaminated water and maintaining the pH at aproper level. The choice of an oxidizing agent depends onthe substance or substances to be detoxified. Numerousoxidizing agents are available to detoxify a variety of com-pounds. The most commonly used agents are hydrogenperoxide, ozone, hypochlorite, chlorine, and chlorine diox-ide because they tend not to form toxic compounds orresiduals and are relatively inexpensive. Ozone and hy-drogen peroxide have an advantage over oxidants con-taining chlorine because potentially hazardous chlorinatedcompounds are not formed (U.S. EPA 1986b).

Hydrogen peroxide is a stable and readily available sub-stance that can oxidize many compounds. Industrial treat-ment plants have used hydrogen peroxide to detoxifycyanide and organic pollutants including formaldehyde,phenol, acetic acid, lignin sugars, surfactants, amines andglycol ethers, aldehydes, dialkyl sulfides, dithionate, andcertain nitrogen and sulfur compounds (EnvirosphereCompany 1983).

Ozone is a strong oxidizing agent (gas) that is unstableand extremely reactive. Therefore, ozone cannot beshipped or stored but must be generated onsite immedi-ately prior to application (U.S. EPA 1985b). Ozone rapidlydecomposes to oxygen in solutions containing impurities.Ozone’s half-life in distilled water at 68°F is twenty-five

minutes, while in groundwater it drops to eighteen min-utes (Envirosphere Company 1983).

Hypochlorite is used in drinking water and municipalwastewater systems for the treatment and control of algaeand biofouling organisms (U.S. EPA 1985b). In industrialwaste treatments, hypochlorite is used for the oxidation ofcyanide, ammonium sulfide, and ammonium sulfite(Huibregts and Kastman 1979). Sodium hypochlorite so-lutions at concentrations of 2500 mg/l are also used forthe detoxification (by oxidation) of cyanide contaminationfrom indiscriminate dumping (Farb 1978). However, be-cause the principal products from chlorination of organiccontaminants are chlorinated organics which can be asmuch of a problem as the original compound, hypochlo-rite treatment is limited.

Advanced Oxidation

Advanced oxidation uses UV radiation combined withozone or hydrogen peroxide to enhance the oxidation rateof the compounds; reaction times can be 100 to 1000 timesfaster in the presence of UV light (U.S. EPA 1986b). UVlight reacts with hydrogen peroxide molecules to form anhydroxyl radical, a powerful chemical oxidant.Specifically, hydrogen peroxide and UV light are used asshown in Figure 9.17.16 for the treatment of volatile or-ganic compounds and other organic contaminants in con-taminated groundwater (U.S. EPA 1993). In addition, hy-drogen peroxide, ozone, and UV radiation are used asshown in Figure 9.17.17 for the oxidation of dissolved or-ganic contaminants including chlorinated hydrocarbons


FIG. 9.17.16 Perox-pure chemical oxidation technology. (Reprinted from U.S. EnvironmentalProtection Agency, 1993, Perox-pure chemical oxidation technology—Perioxidation Systems,Inc., EPA/540/AR-93/501 Superfund Innovative Technology Evaluation, Washington, D.C.: U.S.EPA.)

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

GROUNDWATER FROMSITE 300

EXTRACTION WELL

CARTRIDGEFILTERS

SPIKINGSOLUTION

STATICMIXER

BLADDERTANK

STATICMIXER

OXIDATION UNIT

REACTOR

UV LAMP

TO DISPOSAL

HYDROGENPEROXIDE

SULFURICACID

HYDROGENPEROXIDESPLITTER

SODIUMHYDROXIDE

and aromatic compounds in groundwater (U.S. EPA1990).

Chemical Reduction

Environmental engineers have proposed chemical reduc-tion to detoxify wastes and contaminated waters, but itsapplication does not appear to have the potential thatchemical oxidation has. For example, they have proposedsodium sulfites to treat groundwater contaminated bysodium hypochlorite (Huibregts and Kastman 1979) andferrous sulfate in conjunction with hydroxides to detoxifyand insolubilize hexavalent chromium (Tolman et al.1978; Metcalf and Eddy, Inc. 1972). Little work has beendone in the use of chemical reduction for organic wastes.

Cost

Costs for oxidation systems include the costs for storageand handling equipment, chemicals, feed systems and con-trols, and electricity to operate the ozone generator or theUV lamps. Costs for enhanced oxidation range from $0.15to $70/1000 gal treated depending on the type of con-taminants, their concentration, and the cleanup level (U.S.EPA 1993, 1990).


The principal advantage of chemical oxidation technologyis the ability of oxidizing agents to degrade carbonaceouscompounds, theoretically to carbon dioxide and water(Roy 1990b). Adequate oxidant and operating conditions(i.e., temperature, pH, and contact time), however, mustbe present to facilitate a complete reaction. Incomplete re-actions can generate partially oxidized products which mayrequire further treatment. Oil and grease in the water canminimize the efficiency of the oxidation process. In addi-tion, UV lamps do not perform well in turbid waters be-cause of the reduced light transmission (Roy 1990a).

Limitations of Pump-and-TreatTechnologiesPump-and-treat is the most commonly used technology forgroundwater remediation and plume containment.However, recently pump-and-treat technology has beensubject to increasing scrutiny and controversy. One sig-nificant problem with the technology is its inability toachieve cleanup goals within reasonable time frames(Galya 1994). At many sites where this technology is used,contaminant removal rates follow a relatively consistentpattern. After a period of initially steady reductions,


FIG. 9.17.17 Ultrox, UV/oxidation technology. (Reprinted from U.S. Environmental Protection Agency, 1990,Ultraviolet radiation/oxidation technology—Ultrox International, EPA/540/AS-89/012 Superfund InnovativeTechnology Evaluation, Washington, D.C.: U.S. EPA.)

Hydrogen PeroxideFeed Tank Contaminated Water

Feed Tank

Ozone Diffuser(typical)

From ShallowGroundwaterMonitoring Wells

EffluentSample Tap

TreatedEffluentStorageTank

StainlessSteelReactor

UV Lamp(typical)

Ozonefrom OzoneGenerator

Needle Valve(typical)

Rotameter(typical)

HeadspaceOverflow Weir

(typical)

Ozone Manifold

Sight Glass

CatalyticOzone

Decomposer

groundwater contaminant concentrations tend to level offand remain fairly constant, with random fluctuationsaround an assumptotic limit (Tucker et al. 1989) as shownin Figure 9.17.18. The assumptotic concentration levelmay be higher than the specified cleanup target, andachieving cleanup goals within reasonable time frames maynot be possible.

Therefore, pump-and-treat technology is not an effec-tive approach by itself for the ultimate remediation ofaquifers to health-based cleanup concentrations.

—Ahmed Hamidi

ReferencesAdams, C.E., and W.W. Eckenfelder. 1974. Process design techniques

for industrial waste treatment. Nashville, Tenn.: Enviro Press.Adams, J.Q., and R.M. Clark. 1991. Evaluating the cost of packed-tower

aeration and GAC for controlling selected organics. Journal AWWA1:49–57.

Bass, D.H., and T.E. Sylvia. 1992. Heated air stripping for the removalof MTBE from recovered groundwater. Proceedings for the 1992Petroleum Hydrocarbons and Organic Chemicals in Groundwater,4–6 November, Houston, Tex.

Campbell, M.D., and J.H. Lehr. 1977. Well cost analysis. In Water welltechnology. 4th ed. McGraw-Hill.

Clark, R.M., R.G. Eilers, and J.A. Goodrick. 1984. VOCs in drinkingwater: Cost of removal. Journal of Environmental Engineering 110,no. 6:1146–1162.

Cohen, R.M., and W.J. Miller. 1983. Use of analytical models for eval-uating corrective actions at hazardous waste disposal facilities.Proceedings of the Third National Symposium on Aquifer Restorationand Groundwater Monitoring. Worthington, Ohio: National WaterWell Association.

Corbitt, R.A. 1990. Wastewater disposal. In Standard handbook of en-vironmental engineering, edited by R.A. Corbitt. New York:McGraw-Hill.

Crittenden, J.C. et al. 1988. Using GAC to remove VOCs from air strip-per off-gas. Journal AWWA 80, no. 5(May):73–84.

Cummins, M.D., and J.J. Westrick. 1983. Trichloroethylene removal bypacked column air stripping: Field verified design procedure. InProceedings ASCE Environmental Engineering Conference, 442–449.Boulder, Colo.: ASCE.

Envirosphere Company. 1983. Evaluation of systems to accelerate sta-bilization of waste piles or deposits. Cincinnati, Ohio: U.S. EPA.

Farb, D. 1978. Upgrading hazardous waste disposal sites: Remedial ap-proaches. EPA-SW-677. Cincinnati, Ohio: U.S. EPA.

Galya, D. 1994. Evaluation of the effectiveness of pump and treat ground-water remediation system. In WEF Specialty Conference SeriesProceedings on Innovative Solutions for Contaminated SiteManagement, 6–9 March, 323–342. WEF.

Hand, D.W. et al. 1986. Design and evaluation of an air stripping towerfor removing VOCs from groundwater. Journal of AWWA 78, no.9:87–97.

Huibregts, K.R., and K.H. Kastman. 1979. Development of a system toprotect groundwater threatened by hazardous spills on land. Oil andHazardous Material Spills Brands, Industrial Environmental ResearchLaboratory. Edison, N.J.: U.S. EPA.

Johnson Division, UOP, Inc. 1975. Groundwater and wells. Saint Paul,Minn.: Edward F. Johnson, Inc.

Kuffs, C. et al. 1983. Procedures and techniques for controlling the mi-gration of leachate plumes. Ninth Annual Research Symposium onLand Disposal, Incineration and Treatment of Hazardous Waste.

Lenzo, F., and K. Sullivan. 1989. Groundwater treatment techniques, anoverview of state-of-the-art in America. First US/USSR Conferenceon Hydrogeology, July. Moscow.

Luthin, J.N. 1957. Drainage of agricultural lands. Madison, Wis.:American Society of Agronomy.

Means. 1994. Means site work and landscape cost data. Means SouthamConstruction Information Network.

Metcalf and Eddy, Inc. 1972. Wastewater engineering: Collection, treat-ment, and disposal. New York: McGraw-Hill.

Nirmalakhandan, N., Y.H. Lee, and R.E. Speece. 1987. Designing a costeffective air stripping process. Journal of AWWA 79, no. 1:56–63.

Noonan, D.C., and J.T. Curtis. 1990. Groundwater remediation and pe-troleum: A guide for underground storage tanks. Chelsea, Mich.:Lewis Publishers.

Oakley, D. et al. 1994. The use of horizontal wells in remediating andcontaining a jet fuel plume—preliminary findings. WEF SpecialtyConference Series Proceedings on Innovative Solutions for Contam-inated Site Management, March, 331–342. Water EnvironmentFederation.

O’Brien, R.P. 1983. There is an answer to groundwater contamination.Water/Engineering and Management (May).

Onda, K.H., Takeuchi, and Y. Okumoto. 1968. Mass transfer coeffi-cients between gas and liquid phases in packed columns. Journal ofChemical Engineering, Japan 72, no. 12:684.

Perry, R.H., and D. Green. 1984. Perry’s chemical engineer’s handbook.6th ed. New York: McGraw-Hill.

Powers, J.P. 1981. Construction dewatering: A guide to theory and prac-tice. New York: John Wiley and Sons.

Repa, E. et al. 1982. The establishment of guidelines for modeling ground-water contamination from hazardous waste facilities. JRB Assoc. re-port prepared for the Office of Solid Waste, U.S. EPA.

Roy, K. 1990a. Researchers use UV light for VOC destruction. HazmatWorld (May):82–93.

———. 1990b. UV-oxidation technology, shining star or flash in thepan? Hazmat World (June):35–50.

Soil Conservation Service. 1973. Drainage of agricultural land. Syosset,N.Y.: Water Information Center.

Tolman, A. et al. 1978. Guidance manual for minimizing pollution fromwaste disposal sites. EPA/600/2-78/142. Cincinnati, Ohio: U.S. EPA.

Tucker, W.A. et al. 1989. Technological limits of groundwater remedi-ation: A statistical evaluation method. Proceedings of the PetroleumHydrocarbons and Organic Chemicals in Groundwater, 15–17 Nov.,Houston, Tex.

U.S. Environmental Protection Agency. 1977. Wastewater treatment fa-cilities for sewered small communities. Washington, D.C.: U.S. EPA,Technology Transfer Division.


Concentration

What isClean

Time

FIG. 9.17.18 Assumptotic behavior of pump-and-treatcleanup technologies. (Reprinted from K.E. Nyer, 1992,Groundwater treatment technology, 2d ed., New York: VanNostrand Reinhold.)

———. 1985a. Handbook, remedial action at waste disposal sites.EPA/625/6-85/006. Washington, D.C.: U.S. EPA.

———. 1985b. Leachate plume management. EPA/540/2-85/004.Washington, D.C.: U.S. EPA.

———. 1986a. Mobile treatment technologies for superfund wastes.EPA/540/2-86/003(f). Washington, D.C.: U.S. EPA.

———. 1986b. Systems to accelerate in situ stabilization of waste de-posits. EPA/540/2-86/002. Cincinnati, Ohio: U.S. EPA.

———. 1988. A compendium of technologies used in the treatment ofhazardous waste. EPA/540/2-88/1004. Washington, D.C.: U.S. EPA.

———. 1989. Seminar publication on transport and fate of contami-nants in the subsurface. EPA/625/4-89/019. Cincinnati, Ohio: U.S.EPA.

———. 1990. Ultraviolet radiation/oxidation technology—Ultrox

International. EPA/540/A5-89/012, Superfund Innovative Technol-ogy Evaluation. Washington, D.C.: U.S. EPA.

———. 1993. Perox-pure chemical oxidation technology—PerioxidationSystems, Inc. EPA/540/AR-93/501 Superfund Innovative TechnologyEvaluation. Washington, D.C.: U.S. EPA.

Van Hoorn, J.W., and W.H. Vandemolen. 1974. Drainage of sloppingof lands, drainage principles and applications. Vol. 4 of Design andmanagement of drainage systems. Publ. 16. 329–339. Wageningen,The Netherlands: International Institute of Land ReclamationImprovements.

Wasseling, J. 1973. Theories of field drainage and watershed runoff:Subsurface flow into drains. In Drainage, principles and applications.Wageningen, The Netherlands: International Institute for LandReclamation and Improvement.


9.18IN SITU TREATMENT TECHNOLOGIES

In situ treatment is an alternative to pump-and-treat tech-nology and involves the underground destruction and neu-tralization of contaminants. The technology has the ad-vantage of requiring minimal surface facilities and reducingpublic exposure to the contaminant. Theoretically, thetechnology could be applied to both organic and inorganiccontaminants. However, in situ treatment is still relativelynew and for the most part has been limited to organiccompounds. The most commonly used in situ treatmentmethods include bioremediation, air sparging, and chem-ical detoxification.

BioremediationBioremediation is a relatively new technology that has re-cently gained considerable attention. Bioremediation usesnaturally occurring microorganisms to degrade and breakdown organic contaminants into harmless products con-sisting mainly of carbon dioxide and water. In situ biore-mediation has two basic approaches. The first approachrelies on the natural biological activities of indigenous mi-croorganisms in the subsurface. The second approach iscalled enhanced bioremediation and involves stimulatingthe existing microorganisms by adding oxygen and nutri-ents. Most organic compounds are biodegradable, somefaster than others. The rate of biodegradation, however,depends on the chemical structure of the compound as dis-cussed in Section 9.12 and shown in Table 9.12.3. Figure9.18.1 shows a simplified representation of a groundwa-ter bioremediation system.

DESIGN CONSIDERATIONS

The design variables of bioremediation include the amountof bacteria, oxygen, and nutrients needed for the

biodegradability of the contaminant as well as the char-acteristics of the subsurface environment. Given those vari-ables, environmental engineers can determine an appro-priate hydraulic design of the bioremediation system.Computer models such as BIOPLUME II (1986) can as-sist in the design of bioremediation systems.

The number of bacteria must be sufficient to consumeall of the organic contaminants in a timely manner. Mostsites have significant populations of indigenous microor-ganisms that can degrade a variety of organic contami-nants. One gram of surface soil can contain from 0.1 to1 billion cells of bacteria, 10 to 100 million cells of actin-omycetes, and 0.1 to 1 million cells of fungi (Dockins 1980;Whitelaw and Edwards 1980). The microorganism popu-lation in soils is generally greatest in the surface horizonswhere the temperature, moisture, and energy supply is fa-vorable for their growth. As the depth increases, the num-ber of aerobic microorganisms decreases; however, anaer-obic microorganisms can exist depending on theavailability of nutrients and organic material. The type ofmicroorganisms present on site and their optimal livingconditions can be determined in the laboratory. If indige-nous microorganisms are not present on site or if theirnumber is not sufficient to consume all organic contami-nants, appropriate exogenous microorganisms can be im-ported, or existing microorganisms can be stimulated withthe addition of oxygen and nutrients.

In addition, aerobic bacteria require oxygen for theirgrowth. Because the concentrations of dissolved oxygen ingroundwater are generally low, adding oxygen supportsthe aerobic biodegradation of organic compounds ingroundwater. The theoretical quantities of oxygen requiredto degrade an organic compound can be determined fromstoichiometric analysis. For example, degradation of a sim-ple organic acid, such as acetic acid, theoretically requires

1.1 mg of oxygen. Oxygen can be added in several ways,including aeration, oxygenation, and the use of hydrogenperoxide and other oxygen-containing compounds.Obviously, the use of these compounds requires carefulcontrol of the geochemistry and hydrology of the site.

Inorganic nutrients including nitrogen, phosphorous,and potassium are needed for proper bacterial growth andcan limit cell growth if they are not present at sufficientlevels. The groundwater may already contain levels ofphosphorous and nitrogen, but these levels are probablyinsufficient for bacterial growth (Bouwer 1978; Doetschand Cook 1973). The addition of nutrients, however, cancontaminate the aquifer. Therefore, only the amountneeded to sustain biological activity should be added.

Other factors limit the growth rate of bacteria and,therefore, the biodegradation of organic contaminants in

groundwater. These factors include the pH, temperature,and toxicity of the contaminant. The appropriate rangefor these parameters should be determined in a treatabil-ity study.

ADVANTAGES AND LIMITATIONS

Bioremediation has several advantages over other cleanuptechnologies including cost, minimal surface facilities, andminimum public exposure to the contaminant. However,bioremediation suffers from several drawbacks (Lee et al.1988). The technology is limited to aquifers with high per-meability. Bacterial growth can be inhibited by one or morecompounds at sites with mixed wastes. In addition, in-


��@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@��ÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀÀ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Injection Well

Direction of Groundwater Flow

Extraction Well

Subsurface Aeration Wells

LeachatePlume

Boundary

Nutrients

In-lineOxygen Source In-situ

Aeration

��

Soil Flushing

ContaminatedSoil

InjectionWell

Aeration Zone

AerationWell Bank

Extraction Well

Direction of Groundwater Flow

Leachate Plume

Groundwater Table

��

��

��

��

��

FIG. 9.18.1 Simplified representation of a groundwater bioremediation system. (Reprinted fromU.S. Environmental Protection Agency, 1985, Handbook, remedial action at waste disposal sites,EPA/625/6-85/006, Washington, D.C.: U.S. EPA.)

complete degradation of some substances can lead to othertypes of contamination.

Air SpargingAir sparging, also called in situ stripping, is an innovativetechnology that injects air into the saturated zone to re-move contaminants from the water. The air injected in thesaturated area creates bubbles that rise and carry trappedand dissolved contaminants into the unsaturated zoneabove the water table (Camp Dresser & McKee, Inc.1992). This technology is typically used in conjunctionwith soil vapor extraction (SVE) to enhance the removalrate of contaminants from the saturated and unsaturatedzones (Bohler et al. 1990). As volatile organic compoundsreach the unsaturated zone, they are captured by the SVEvapor wells that are screened in the unsaturated zone, asillustrated in Figure 9.18.2. Air sparging also provides anoxygen source which may stimulate bioremediation ofsome contaminants. Air sparging is applicable for conta-minants which have a high Henry’s constant or high va-por pressure in soils with high permeability.

DESIGN CONSIDERATIONS

The design variables for an air sparging system include thevolatility and concentration of the compound, the poros-ity and permeability of the soil, and the temperature of thewater. Given those variables, environmental engineers candetermine the radius of influence of the air sparge wells,the air flow rate, and the vacuum pressure needed.Although the technology has been used at several sites(Loden and Fan 1992), references to the design of an airsparging system are limited (Sellets and Schreiber 1992;Marley, Li, and Magee 1992), and in most cases the de-sign is based on empirical formulas or the results of pilotstudies.

ADVANTAGES AND LIMITATIONS

Air sparging is a promising technology which has severaladvantages. Air sparging can extend the effectiveness ofSVE systems to include volatile contaminants from the sat-urated zone, and the contaminants can be treated onsitewithout removal or potential public exposure to them. Inaddition, air sparging can expedite groundwater cleanup.The technology, however, is limited to aquifers with highpermeability and contaminants with high volatility. In ad-dition, the technology is relatively new, and the numberof case studies where the technology has been successfullyapplied is limited.

Other Innovative TechnologiesOver the last few years, several innovative technologieshave been proposed for the in situ treatment of ground-water. Although these technologies have not yet been de-veloped to the extent of previously discussed technologies,some of them have demonstrated success in actual site re-mediations (Wagner et al. 1986). Laboratory and pilot test-ing, however, are necessary to evaluate the applicability ofa particular technology to a site. Examples of innovativegroundwater technologies are described next.

NEUTRALIZATION ANDDETOXIFICATION

In situ neutralization and detoxification involves injectinga substance into groundwater that neutralizes or destroysa contaminant. The technology is limited to contaminantsthat can be neutralized or degraded to nontoxic byprod-ucts. Neutralization and detoxification is applicable toboth organic and inorganic compounds. Selecting a treat-ment agent depends on the type of contaminant and thecharacteristics of the subsurface environment such as tem-


��

MonitoringWell

Air InjectionWell

AirCompressor

VOC Gases

Vacuum Pump

Vapor Extraction System

WaterTable

UnsaturatedZone

SaturatedZone

Soil GasMonitor

TreatmentSystem�

FIG. 9.18.2 Simplified representation of an air sparging system.

perature, permeability, pH, salinity, and conductivity.Examples of in situ treatment agents include hydrogen per-oxide that can be injected directly into groundwaterthrough existing monitoring wells or subsurface drains(Vigneri 1994). Hydrogen peroxide produces the hydro-gen free radical OH, an extremely powerful oxidizer whichprogressively reacts with organic contaminants to producecarbon dioxide and water.

Other in situ neutralization and detoxification tech-nologies include precipitation and polymerization.Precipitation involves injecting substances into the ground-water plume which form insoluble products with the con-taminants, thereby reducing the potential for migration ingroundwater (U.S. EPA 1985). This technique is mainlyapplicable to dissolved metals, such as lead, cadmium, zinc,and iron. Some forms of arsenic, chromium, and mercuryand some organic fatty acids can also be treated by pre-cipitation (Huibregts and Kastman 1979). The most com-mon precipitation reagents include hydroxides, oxides, sul-fides, and sulfates. As with other in situ techniques,precipitation is only applicable to sites with aquifers hav-ing high hydraulic conductivities. The major disadvantagesof precipitation are that it can only be applied to a nar-row, specific group of chemicals (mainly metals); that apotential groundwater pollutant may be injected; that toxicgases (as in sulfide treatment) may form; and that the pre-cipate may resolubilize (U.S. EPA 1985).

In situ polymerization involves injecting a polymeriza-tion catalyst into the nonaqueous organic phase of a con-taminant plume to cause polymerization (U.S. EPA 1985).The resulting polymer is gel-like and nonmobile in thegroundwater flow regime. Polymerization is a specific tech-nique that is applicable to organic monomers such asstyrene, vinyl chloride isoprene, methyl methacrylate, andacrylonitrile (Huibregts and Kastman 1979). In a haz-ardous waste site where groundwater pollution has oc-curred over time, any organic monomers originally pre-sent would most likely have polymerized upon contactwith the soil (U.S. EPA 1985). Therefore, in situ poly-merization is a technique most suited for groundwatercleanup following land spills or underground leaks of apure monomer. The major disadvantages of polymeriza-tion include its limited application and the difficulty of ini-tiating sufficient contact of the catalyst with the dispersedmonomer (Huibregts and Kastman 1979).

PERMEABLE TREATMENT BEDS

Permeable treatment beds are also in situ treatment tech-niques used at sites with relatively shallow groundwatertables. The concept of a permeable treatment bed involvesexcavating a trench, filling the trench with a permeabletreatment material, and allowing the plume to flowthrough the bed thus physically removing or chemicallyaltering the contaminants. The function of a permeabletreatment bed is to reduce the quantities of contaminants

in the plume to acceptable levels. Potential problems withusing a permeable treatment bed include saturation of thebed material, plugging of the bed with precipitates, andthe short life of the treatment material (U.S. EPA 1985).

The selection of the appropriate bed material to treatthe contaminants and the design of the bed are two ele-ments that determine the effectiveness of a permeable treat-ment bed. The types of available treatment bed fill mate-rial include limestone, crushed shell, activated carbon,glauconitic greensands, and synthetically produced ion ex-change resins. Ensuring proper physical design of thetreatment bed requires a knowledge of the hydrogeologyof the site (e.g., groundwater flow rate and direction, hy-draulic conductivities) and the chemical characteristics ofthe plume (U.S. EPA 1985).

PNEUMATIC FRACTURING

Environmental engineers use pneumatic fracturing extrac-tion and hot gas injection to treat in situ contaminationlocated within low permeable formations (AccutechRemedial Systems, Inc. 1994). The process has beendemonstrated at numerous sites and significantly increasessubsurface permeability and contaminant mass removal(U.S. EPA 1993b). The process applies controlled burstsof high pressure air into a well through a proprietary in-jection and monitoring system. When the down-hole pres-sure exceeds the pressure of the formation, channels orfractures are created propagating from the fracture well.Once the permeability of the formation is increased, engi-neers inject hot gas air (250 to 300°F for pilot-scale and300 to 600°F for full-scale design) under pressure to ele-vate the temperature of the fracture surface and volatilizecontaminants located within the formation matrix. The ex-tracted vapors are then treated by activated carbon dur-ing low-concentration process streams or by catalytic tech-nology during high-concentration process streams.

The technology can be applied at depths to 50 feet andhas a radius of influence of as much as 40 feet from theinjection point (well). Subsurface air flow has been in-creased 150 times compared with the site’s natural per-meability. The technology, however, is not applicable fortreating inorganic or nonvolatile organic compounds. Inaddition, applying the pneumatic fracturing process maybe unnecessary at a site with a high natural permeability.

THERMALLY ENHANCED RECOVERY

The in situ steam enhanced extraction process, called ther-mally enhanced recovery (Praxis Environmental ServicesInc. 1994), removes volatile and semivolatile organic com-pounds from an area of contaminated soil or groundwa-ter without excavation. The process operates through theuse of wells constructed in the contaminated soil. High-quality steam is added to the soil through some wells, calledinjection wells. Other wells, known as extraction wells,


operate under vacuum to remove liquid and vapor con-taminants and water from the soil. Injecting steam into theground raises the temperature of the soil and causes themost volatile compounds to vaporize. In addition, pres-sure gradient is formed between the injection and extrac-tion wells which drives the flow of steam and vaporizedcontaminants towards the extraction wells (U.S. EPA1993a). Raising the temperature of the soil matrix also as-sists in removing less volatile compounds by increasingtheir in situ vapor pressure. After the entire soil mass be-ing treated has reached the steam temperature, as deter-mined by soil–temperature monitors, and steam break-through occurs at the extraction wells, the flow of steamcontinues only intermittently with a constant vacuum ap-plied to the extraction wells. The vacuum extraction re-moves much of the remaining contamination. As the soilin the high permeability region cools, the steam remainingin the low permeability region evaporates the contami-nants.

The technology is cost-effective for large and deep ar-eas of contamination where technologies requiring exca-vation are difficult or impossible. The process can be ap-plied in sections to treat an area of any size and depth. Ifthe site, however, contains a high concentration (.200ppm) of heavier-than-water organics, a possibility existsthat these compounds might be mobilized downward intogroundwater. In addition, treatment of shallow (,10 feet)contaminated areas is less cost-effective than deeper areascompared to other technologies.

—Ahmed Hamidi

ReferencesAccutech Remedial Systems, Inc. 1994. Pneumatic fracturing. Keyport,

N.J.BIOPLUME II. 1986. Computer model of two dimensional contaminant

transport under the influence of oxygen limited biodegradation ingroundwater. Houston, Tex.: National Center for GroundwaterResearch, Rice University.

Bohler, U. et al. 1990. Air injection and soil air extraction as a combinedmethod for cleaning contaminated sites: Observations from test sitesin sediments and solid rocks. In Contaminated Soil ’90, edited by F.Arench et al., 1039–1044. The Netherlands: Kluwser Academic Publ.

Bouwer, H. 1978. Groundwater hydrology. New York: McGraw-Hill.Camp Dresser & McKee, Inc. 1992. A technology assessment of soil va-

por extraction and air sparging. Risk Reduction EngineeringLaboratory, Office of Research and Development. Cincinnati, Ohio:U.S. EPA.

Dockins, W.S. et al. 1980. Dissimilatory bacterial sulfate reduction inMontana groundwaters. Geomicrobiology Journal 2, no. 1:83–98.

Doetsch, and T.M. Cook. 1973. Introduction to bacteria and their eco-biology. Baltimore, Md.: University Park Press.

Huibregts, K.R., and K.H. Kastman. 1979. Development of a system toprotect groundwater threatened by hazardous spills on land. Oil andHazardous Material Spills Brands, Industrial Environmental ResearchLaboratory. Edison, N.J.: U.S. EPA.

Lee, M.D. et al. 1988. Biorestoration of aquifers contaminated with or-ganic compounds. CRC Crit. Rev. Environ. Control 18:29–89.

Loden, M.E., and C.Y. Fan. 1992. Air sparging technology evaluation.Proceedings of 2nd National Research and Development Conferenceon the Control of Hazardous Materials, 328–334. San Francisco,Calif.

Marley, M.C., F. Li, and S. Magee. 1992. The application of a 3-D modelin the design of air sparging systems. Proceedings of the PetroleumHydrocarbons and Organic Chemicals in Groundwater 4–6 Nov.,377–392. Houston, Tex.: NGWA.

Praxis Environmental Services, Inc. 1994. Thermally enhanced recoveryin situ. San Francisco, Calif.

Sellets, K.L., and R.P. Schreiber. 1992. Air sparging model for predict-ing groundwater cleanup rate. Proceedings of the PetroleumHydrocarbons and Organic Chemicals in Groundwater, 4–6 Nov.,365–376. Houston, Tex.: NGWA.

U.S. Environmental Protection Agency. 1985. Leachate plume manage-ment. EPA/540/2-85/004. Washington, D.C.: U.S. EPA.

———. 1993a. In-situ steam enhanced extraction process. In SuperfundInnovative Technology Evaluation Program, Technology Profiles. 6thed. EPA/540/R-93/526. Washington, D.C.: U.S. EPA.

———. 1993b. Pneumatic fracturing extraction. In Superfund InnovativeTechnology Evaluation Program, Technology Profiles. 6th ed.EPA/540/R-93/526. Washington, D.C.: U.S. EPA.

Vigneri, R. 1994. Groundwater remediation primer. Wilmington, N.C.:Cleanox Environmental Services, Inc.

Wagner, K. et al. 1986. Remedial action technology for waste disposalsites. 2d ed. Park Ridge, N.J.: Noyes Data Corporation.

Whitelaw, K. and R.A. Edwards. 1980. Carbohydrates in the unsatu-rated zone of the chalk. England Chemical Geology 29, no.314:281–291.


Storm water is defined as storm water runoff, snowmeltrunoff, and surface runoff and drainage. Storm water man-agement is important in urban water systems, includingwater supply systems and wastewater systems. With in-creasing residential, commercial, and industrial develop-ment, stormwater has become an important issue.

Growing urbanization has a significant impact on thesurrounding environment, creating problems such as non-point sources of water pollution. Because of changes inland-use patterns, pollutants in developed areas build upduring dry periods and are washed off as runoff passesover land surfaces. Nonpoint sources account for about45%, 76% and 65% of the degradation of estuaries, lakes,and rivers respectively (EPA 1989). In comparison, mu-nicipal and industrial point source discharges underNational Pollution Discharge Elimination System(NPDES) control account for about 9–30% of the degra-dation of these water sources.

In contrast to our complex urban environment, the hy-drological cycle shown in many hydrology textbooks israther simplistic. Modification of natural drainage paths,damming of waterways, impoundment of water, reuse ofstormwater, and implementation of new stormwater man-agement processes result in highly intricate hydrologicalprocesses. The development of storm water runoff and itspossible superimposition on dry weather flow in combinedsewer systems are summarized in Figure 9.19.1. A detailedurban drainage subsystem is shown in Figure 9.19.2.

Integrated Management ApproachStorm water system components and functions interactwith, and may also interfere with, each other. Integratedsystem management coordinates actions to achieve waterquantity and quality control, focusing on issues such asfloodplain management, erosion and sediment control,nonpoint source pollution, and preservation of wetlandsand wildlife habitat. System management also facilitatescooperation among all levels of government, and helps toimplement laws and regulations to control storm waterpollution.

FEDERAL PROGRAMS

In the 1987 amendments to the Clean Water Act, Congressmandated development of a permit system for certainsources of storm water discharge, thus the EPA has es-tablished permit application requirements for industrialstorm water discharges and municipal storm sewer systemdischarges. Pollutants entering storm water and surfacewater systems are now regulated as point sources underSection 402(p) and subject to the NPDES permit process.

The EPA also provides assistance and guidance to mu-nicipalities developing storm water management pro-grams. Although there are several agencies with possibleauthority in this field, no federal agency has assumed gen-eral responsibility or control. Most actions taken to datehave been local initiatives. Only the Soil ConservationService has long-standing programs of storm water man-agement.

However, many federal agencies are directly involvedin flood hazard mitigation, flood control, and floodplainmanagement. Although there is no federal agency directlymandated to plan and implement stormwater managementprograms, there are several agencies engaged in related ac-tivities.

The federal government exerts a broad influence via itsmany agencies. For example, in the Corps of Engineers’major structural flood control program, the federal agencyconsults with local agencies, but maintains field offices andstaff for planning, construction, operation, and mainte-nance. In another approach, the Soil Conservation Service(SCS) has a nationwide network of conservation districts.The districts perform some functions autonomously, whileother functions are carried out by the federal staff. In flood-plain management, the Federal Emergency ManagementAgency (FEMA) has established fairly complete federalcontrol, although actions affecting individuals are legallymandated by state laws and local ordinances. In this case,the financial incentives of the flood insurance program arethe prime motivation for obtaining required state legisla-tion and local ordinances.


Storm Water PollutantManagement

9.19INTEGRATED STORM WATER PROGRAM

STATE PROGRAMS

State governments enable legislation providing for in-volvement in storm water management. For example, theWashington State Environmental Policy Act (SEPA) en-sures that environmental values are considered by stateand local government officials when making decisions. TheDepartment of Ecology (State of Washington) recentlycompleted a storm water rule, a highway storm water rule,and a model storm water ordinance for local governments.These rulings require the development of storm water man-agement programs for cities and counties.

MUNICIPAL PROGRAMS

County-level involvement plays an important role in im-plementing comprehensive storm water managementplans. The principal authority for storm water manage-ment is the government with jurisdiction, usually a mu-nicipality. Municipalities usually have legal control of:

• Erosion and sedimentation ordinances• Floodplain ordinances• Storm water drainage ordinances• Zoning ordinances• Building codes• Grading ordinances

Storm water management is closely tied to future landuse development and management. Existing and future


FIG. 9.19.2 The urban storm drainage subsystem.

Precipitation

Outflow

Majordrainage

Intialdrainage

Steamflow

Div

ersi

ons

and

tran

sfer

Sanitarywater

Com

bine

dse

wer

Stormsewer

Evaporationand loses

Aquiferrecharge

Streamflow

Subsurface flow

InflowCollectionconveyance

Intermediatestorage

Thunder- stormsTorrential rainsHeavy rainfall and snowmelt

Land surfaceSwalesGuttersStreetsStorm sewers

DepressionsStreetsSumps

LakesReservoirsFlood- plains

GulchesCreeksStreamsFlood- plains

Light rainfallNormal snowmeltIrrigation runoff

GulchesStorm sewers

CreeksRivers

Storage

Potable water treatment

Distribution

Surfacerunoff

Lawnirrigation

Potablewater

Other urbanSubsystems

Building and DevelopmentPublic Health and SafetyTranportationSolid WasteRecreation/ Open SpaceSanitary WasteIrrigationWater Supply

Waste water treatment

Naturalstorage

Atmosphere

Sheet (Overland) Flow

Detention

Sew

er N

etw

ork

Run

off

Sur

face

Run

off

EvaporationEvapotranspiration

Wetting Surfaces and Filling Depressions

Excess (Effective) Rainfall

Flow in Gutters and Ditches

Percolation

SewerInfiltration

Flow in CollectingSewer (Laterals)

Domestic, Commercialand Industrial

Sewage

Rainfall

Infiltration

Interceptor Flow

Confluence and FlowDiversion Within The Network

Flow Diversion atOverflows

Sewage Treatment Plant Receiving Waters

dry weather

flow

stor

m-

wat

er

Sub

surf

ace

and

Gro

undw

ater

Flo

w

com

-bi

ned

runo

ff

FIG. 9.19.1 Development of Stormwater Runoff and Flow inCombined Sewer Systems. (Reprinted, with permission, fromW.F. Geiger, 1984, Combined sewage quantity and quality—acontribution to urban drainage planning, [Muenchen TechnicalUniversitat, Muenchen].)

land use development are incorporated into an integratedstorm water management program as presented in Figure9.19.3.

Many municipalities now require developers to considerfuture development of watersheds when designing stormwater drainage systems for new development. Detention

facilities are frequently required in subdivision laws, zon-ing ordinances, building codes, and water pollution regu-lations.

—Kent K. Mao


Fiscal Scrutiny offlood controlexpenditures

Regulatefloodplain development

Develop acomprehensivedrainage plan

Ensure efficientstorm water

drainage design

Preserve naturaldrainage system

Protecthuman life

OB

JEC

TIV

ES

AC

TIO

NP

AS

SIV

ER

ES

OLU

TIO

NS

OR

DIN

AN

CE

S

Minimizepropertydamage

Provide forsediment anderosion control

Comprehensivestorm water

drainagemaster plan

Storm watermanagement and

design manual

Floodplainmanagement

board

Comprehensiveland use plan

Ensure new developmentcreates no demand for

public flood controlinvestment

Preservation ofecosystems andflood plain areas

is desirable

Intergovernmentalcooperationis essential

STORM WATERUTILITY ORDINANCE

FLOODPLAINS ANDWETLANDS

MANAGEMENTORDINANCE

STORM WATERDRAINAGE

ORDINANCE

SEDIMENT ANDEROSION CONROL

ORDINANCE

FIG. 9.19.3 Developing a comprehensive storm management program.

9.20NONPOINT SOURCE POLLUTION

Urban storm water pollution and most pollution in com-bined sewer overflows originates from nonpoint or diffusesources. The processes controlling storm water quality arerather complex, as shown in Figure 9.20.1. In contrast topoint source pollution, such as industrial and municipaltreatment plant outfalls, these sources of pollution are nu-merous and their contributions are difficult to quantify.Diffuse pollution is a hydrologic process that closely fol-lows the statistic character of rainfall, and must be evalu-ated similarly.

Urban nonpoint sources have been identified as a ma-jor cause of pollution of surface water bodies by the U.S.EPA (EPA 1984). In the 1988 Report to Congress (EPA1990), the EPA stated that urban storm water runoff is

the fourth most extensive cause of impaired water qualityin the nation’s rivers, and the third most extensive causeof impaired water quality in lakes. Combined sewer out-flows (CSOs) are tenth on the list of significant sources ofimpairment for both surface-water bodies.

Major Types of PollutantsUrban storm water runoff may transport many undesir-able pollutants. The pollutants present, and their concen-trations, are a function of the degree of urbanization, thetype of land use, the densities of automobile traffic andanimal population, and the degree of air pollution beforerainfall. Major pollutant types are classified as follows:

• Suspended sediments• Oxygen-demanding substances• Heavy metals• Toxic organics—pesticides, PCBs• Nutrients—nitrogen, phosphorous• Bacteria and viruses• Petroleum-based substances or hydrocarbons• Acids and• Humic substances—precursors for trihalomethane

Annual pollutant loadings for storm water and com-bined sewer overflows are given in Table 9.20.1 (UNESCO1987).

Nonpoint SourcesThree basic processes generate pollutants during runoff:

ATMOSPHERIC DEPOSITION

Atmospheric deposition is generally divided into wet anddry deposition. Wet deposition is closely related to the lev-els of atmospheric pollution by traffic, industrial and do-mestic heating, and other sources. Urban rainfall is gener-ally acidic, with below 5 pH units. The elevated acidity ofurban precipitation damages pavements, sewers, and other


RemoteSources

Sources

TransportProcesses

Sinks

LocalSources

MotorVehicles

PlantMaterial Erosion Fertili-

zationSolid

WastesWasteDumps

Sewer Cross-Connections

SpillsAbrasionof Solid

Surfaces

Anti-Skidand De-

icingMaterials

AnimalExcrementAir Pollution

Dust and Dirt Accumulation on The Catchment Surface

Air Transport(Including Rainfall)

Sewer Pipe Transport

Solid Waste Disposal

Stormwater Ponds

Water Course

Overland Flow

Removal By StreetSweeping

FIG. 9.20.1 Sources of pollutants in stormwater and pollutant pathways. (Reprinted from UnitedNations Educational, Scientific and Cultural Organization (UNESCO) 1987, Manual on drainagein urbanized areas, vol. 1, Planning and design of drainage systems, UNESCO Press.)

TABLE 9.20.1 ANNUAL UNIT POLLUTANT LOADINGS FOR STORMWATER AND COMBINED SEWEROVERFLOWS

Annual Pollutant Loadings (kg/ha/yr)*

TotalSuspended Total Total

Source Solids BOD COD N P

Runoff in storm sewers 100–6300 5–170 20–1000 2–12 0,2–2,2Residential area runoff 600–2300 5–100 20–800 2–12 0,2–2,2Commercial area runoff 100–800 40–90 100–1000 5–12 1,2–2,2Industrial area runoff 400–1700 10–90 200–1000 5–10 1,0–2,1Highway runoff 120–6300 90–170 180–3900 — —Combined sewer overflows 1200–5000 500–1300 500–3300 15–40 4–8

*1 kg/ha/yr 5 0.89 lb/acre/yrSource: Reprinted from United Nations Educational, Scientific and Cultural Organization (UNESCO), 1987, Manual on drainage in urbanized areas, vol. 1,

Planning and design of drainage systems, (UNESCO Press).

building materials. Particles are then washed off the sur-face by stormwater.

Dry atmospheric deposits are fine particles originatingfrom a distance (fugitive dust) or locally (traffic on un-paved roads, construction and industrial) sources. Dustfallrates vary from region to region. Rural dustfalls dependon soil condition; urban dustfalls are more related to lo-cal air pollution.

EROSION

Erosion of construction areas represents the largest sourceof sediments in urban runoff. Reported unit loads of sed-iment from urban construction sites ranged from 12 to500 tons/half-yr (Novotny and Chesters 1981). Further-more, building activities generate other pollutants such aschemicals from fertilizers and pesticides, petroleum prod-ucts, construction chemicals (cleaning solvents, paints,acids and salts), and various solids. Grading exposes sub-soil, increasing surface erosion due to stormwater runoff.

Erosion of urban lawns and park surfaces is usually low.Exceptions are open, unused lands, and construction sites.

Soil is a source of suspended solids, organics, and pes-ticide pollution. Despite the SCS’s active promotion of ero-sion control, the U.S. Department of Agricultural estimates57–76 million acres (21–31 Mha), about 15–25% of thenation’s agricultural land, is in need of sediment controlmeasures.

ACCUMULATION/WASHOFF

Most urban watersheds are dominated by accumulationand wash-off processes, depending on impervious areas.The accumulation of solids on impervious urban surfaceareas is described by Sartor and Boyd (1972), as shown inFigure 9.20.2.

The sources of urban diffused pollution are:

• Litter, including large-sized materials (greater than3.2 mm) containing items such as cans, brokenglass, vegetable residues and pet waste. Pet fecaldeposits can reach alarming proportions in urbancenters where large numbers of people reside inhighly impervious zones.

• Medium size deposits (street dirt) represent thebulk of street surface-accumulated pollution. Thesources are numerous and very difficult to identifyand control. They may include traffic, road dete-rioration, vegetation resides, pets and other animalwaste and residues, and decomposed litter.

• Traffic emissions are responsible for potentiallytoxic pollutants found in urban runoff, includinglead, chromium, asbestos, copper, hydrocarbons,phosphorous, and zinc. Pollution also comes fromparticles of rubber abraded from tires.

• Road deicing salts applied in winter cause highlyincreased concentrations of salts in urban runoff.

Road salts are applied at rates of 100–300 kg/kmof highway and contain sodium and calcium chlo-ride.

• Pesticides and fertilizers applied onto grassed ur-ban lands.

In fully developed urban areas, where most land sur-faces are impervious because of paving and rooftops,washoff of deposited particles and their transport to thewatercourse become the important mechanism. The rela-tionship of imperviousness to the quantity of some pollu-tants are shown in Table 9.20.2.

Table 9.20.3 shows values and ranges of accumulationof street and surface pollutants estimated by Ellis (1986).A list of specific nonpoint sources is presented in Table9.20.4. The list is not exhaustive. The importance of thesources varies with local conditions.

Direct Input from Pollutant SourceNonpoint pollutants can also reach receiving waters by di-rect input from a pollutant source. Drainage systems in-clude depressions, ditches, culverts, catch basins, wetlands,and creeks that collect water and transport pollutants toreceiving waters. Pollutants may be directly introduced atspecific sites in the system, independent of storm condi-tions. For example, substances may be poured into a catchbasin, traveling directly into a creek or other receiving wa-ter.

In addition to cross-connections of sewage and indus-trial wastes from sanitary sewers, solid waste dumps, andfailing septic tanks, solids accumulations and growth insewers can also enter into storm sewers. Excess water fromlawn watering and car washing is another example of di-rect input. Pollutant loadings from direct inputs are diffi-cult to document and quantify.

—Kent K. Mao


1400

1200

1000

800

600

400

200

00 1 2 3 4 5 6 7 8 9 10 11 12 DAYS

Industrial

All Land-UseCategories Combined

Residential

Commercial

Elapsed Time Since Last Sweeping or Rain

Acc

umul

ated

Sol

ids

Load

ing

(lb/c

urb

mi)

FIG. 9.20.2 Pollutant Accumulation for Different Urban LandUses. (Reprinted, with permission from J.D. Sartor and S. Boyd,1972, Water pollution aspects of street surface contaminants,U.S. Environmental Protection Agency (EPA), EPA ReportR2–72–087, Washington, D.C.)


TABLE 9.20.2 ANNUAL STORM POLLUTANT EXPORT FOR VARIOUS LAND USE TYPES BY PERCENTIMPERVIOUS COVER (POUNDS/ACRE/YEAR)

General Percent Total Total BODLand Use Imperviousness Phosphorous Nitrogen 5-Day Zinc Lead

Rural to 0 0.11 0.80 2.10 0.02 0.01residential 5 0.20 1.60 4.00 0.03 0.01

10 0.30 2.30 5.80 0.04 0.02Large lot, 10 0.30 2.30 5.80 0.04 0.02

single family 15 0.39 3.00 7.70 0.06 0.0320 0.49 3.80 9.60 0.07 0.04

Medium density 20 0.49 3.80 9.60 0.07 0.04single family 25 0.58 4.50 11.40 0.08 0.05

30 0.68 5.20 13.30 0.10 0.0535 0.77 6.00 15.20 0.11 0.06

Townhouse 35 0.77 6.00 15.20 0.11 0.0640 0.86 6.70 17.10 0.12 0.0745 0.97 7.40 18.90 0.14 0.0750 1.06 8.20 20.80 0.15 0.08

Garden apartment 50 1.06 8.20 20.80 0.15 0.08buildings 55 1.16 8.40 22.70 0.16 0.09

60 1.25 9.60 24.60 0.18 0.09High rise to light 60 1.25 9.60 24.60 0.18 0.09

commercial/industrial 65 1.35 10.40 26.40 0.19 0.1070 1.44 11.10 28.30 0.21 0.1075 1.54 11.80 30.20 0.22 0.1180 1.63 12.60 32.00 0.23 0.11

Heavy commercial to 80 1.63 12.60 32.00 0.23 0.11shopping center 85 1.73 13.30 33.90 0.25 0.12

90 1.82 14.00 35.80 0.26 0.1395 1.92 14.80 37.70 0.27 0.13

100 2.00 15.40 39.20 0.28 0.14

NOTES: Assumed rainfall of 40 in/yrRural residential 5 0.25–.5 dwelling units/acreLarge lot, single family 5 1–1.5 dwelling units/acreMedium density, single family 5 2–10 dwelling units/acreTownhouse and garden apartment 5 10–20 dwelling units/acrePollutant loadings are for new developments only.

TABLE 9.20.3 SOLIDS ACCUMULATION AND ASSOCIATED POLLUTANT CONCENTRATIONS IN URBAN AREAS

Residential LightLand Use Low Density High Density Commercial Industrial Highways

Solids accumulation 10–182 30–210 13–180 80–288 13–1100(g/curb m)

Pollutant BOD5 5260 3370 7190 2920 2300–10,000concentration COD 39,300–40,000 40,000–42,000 39,000–61,730 25,100 53,650–80,000(mg/g) Tot.N 460–480 530–610 410–420 430 223–1600

Pb 1570 1980 2330 1390 450–2346Cd 3.2 2.7 2.9 3.6 2.1–10.2

Fecal Coliforms 60,570–82,500 25,621–31,800 36,900 30,700 18,768–38,000(MPN/g)

Source: Reprinted, with permission, from J.B. Ellis Pollutional aspects of urban runoff, Urban runoff pollution, ed. H.C. Torno, J. Marsalek, and M. Desbordes,1–38. (New York, N.Y.: Springer, Verlag, Berlin).

ReferencesBrowne, F.X. and J.T. Grizzard. 1979. Nonpoint sources, J. Water Pol.

Cont. Fed. 51, p. 1428.Ellis, J.B. 1986. Pollutional aspects of urban runoff, in Urban runoff pol-

lution, eds. H.C. Torno, J. Marsalek, and M. Desbordes, 1–38. NewYork, N.Y.: Springer Verlag, Berlin.

Novotny, V. and G. Chesters. 1981. Handbook of nonpoint pollution:source and management. New York, N.Y.: Van Nostrand Reinhold.

Rechow, K.H., M.N. Beaulac, and J.T. Simpon. 1980. Modeling phos-phorous loading and lake response under uncertainty: A manual and

compilation of export coefficients. U.S. Environmental ProtectionAgency (EPA) EPA 440–5–80–011. Washington, D.C.

Sartor, J.D. and G. Boyd. 1972. Water pollution aspects of street surfacecontaminants. U.S. Environmental Protection Agency (EPA) EPAR2–72–081. Washington, D.C.

United Nations Educational, Scientific and Cultural Organization (UN-ESCO). 1987. Manual on drainage in urbanized areas. vol. I. Planningand design of drainage systems. UNESCO Press.

U.S. Environmental Protection Agency (EPA). 1990. National water qual-ity inventory—1988 report to Congress. U.S. EPA Office of Water.EPA Report 440–4–90–003. Washington, D.C.


TABLE 9.20.4 NONPOINT SOURCE POLLUTANTS

Source N O/G T S O M P H

AgriculturalNurseries X X X XCrop farms X X X XLivestock/hobby farms X X X X XFeed/seed/fertilizer supply X X X

Commercial/RetailRestaurants X X X XDry cleaners X XGarden centers X X X X XPrinting shops X

Urban StormwaterRoof washoff X X X X X X XLawn/landscape washoff X X X X XYard debris X X X XSeptic systems X X X XHousehold X X X

MiscellaneousIllicit dumps X X X X X X XCemeteries X X XWarehouses X X X X XFuel storage facilities X X XStreambank erosion X X XDitch cleaning/defoliating X X X XFilling/diverting streams X X XLoss of buffer zones X X XBoating and marinas X X X X X

ConstructionClearing/grading X X X X XBuilding X X X

TransportationRoadways/parking lots X X X X X X XService/repair stations X X X X X XCar/truck washes X X X X X X XOil change shops X X X X X X

N 5 nutrients; O/G 5 oils and greases; T 5 toxic chemicals; S 5 sediments; O 5 organics; M 5 metals; P 5 pathogens, bacteria; H 5 heat

Much emphasis is currently placed on controlling stormwater pollution by attacking the problem at the source, in-stead of using more costly downstream treatment facili-ties. These source controls, termed “Best ManagementPractices” (BMPs), are judged most effective in reducingnonpoint source pollution to a level compatible with wa-ter quality goals.

Best Management Practices are classified into twogroups:

• Planning, with efforts directed at future develop-ment and redevelopment of existing areas

• Maintenance and operational practices to reducethe impact of nonpoint source contaminationfrom existing developed areas

Successful storm water pollution control depends on theeffective implementation of proposed planning efforts

and/or control practices. Legislation or ordinances en-couraging or requiring conformance with intended BMPshas proven to be effective. Table 9.21.1 lists activities in-cluded in a typical source control program.

PlanningThe first goal of planning is to develop a macroscopic man-agement concept, preventing problems from short-sighteddevelopment of individual areas. The planner is interestedin controlling storm water volume, rate, and pollutionalcharacteristics of storm water runoff. Since the size ofstorm sewer networks and treatment plants relates directlyto flow quantity, particularly the peak flowrate, reducingtotal volume or smoothing out peaks will result in lowerconstruction costs.


9.21BEST MANAGEMENT PRACTICES

TABLE 9.21.1 TYPICAL SOURCE CONTROLS

Activity or Area Overview

S1.10—Fueling stations (both commercial Covered, concrete-paved pump island with separateand private) drainage

S1.20—Vehicle/equipment wash and Wash building or designated paved area with separatesteam cleaning drainage containing oil-water separator

S1.30—Loading and unloading liquid materials Conduct activities inside building or at dock withoverhang or skirts to prevent drainage to storm drains;rail and tanker truck transfers require drip pans or pavedareas, and operations and spill cleanup plans

S1.40—Above-ground tanks for liquid storage Diked secondary containment area with stormwaterdrainage passing through oil-water separator

S1.50—Container storage of liquids, food Containers kept indoors or under designated coveredwastes, or dangerous wastes area with separate drainage and secondary containment

for liquid wastesS1.60—Outside storage of raw materials, Place materials under covered area, place temporary

by-products, or finished product (i.e., sand plastic sheeting over material, or pave the area and installand gravel, lumber, concrete and metal) treatment drainage system

S1.70—Outside manufacturing Alter, enclose, cover, or segregate the activity; dischargerunoff to sewer or process wastewater system; or usestormwater BMPs

S1.80—Emergency spill cleanup plans Required for storing, processing, or refining oil productsand producers of dangerous wastes

S1.90—Vegetation management Specific BMPs for seeding and planting, and pest/integrated pest management management, including use of pesticides

S2.00—Maintenance of storm drainage facilities Specific BMPs for maintenance (inspection, repair, andcleaning), disposal of contaminated water, and disposalof contaminated sediments

Source: Stormwater Management Manual for the Puget Sound Basin (Ecology 1992).

LAND USE PLANNING

Computer simulations are used to examine interacting pol-lutant sources in the watershed. By modeling the runoffprocess, a planner can predict the effects of proposed plans,and the ability of controls to solve potential problems.Several models are described in Section 10.9. Water qual-ity criteria standards can be recommended after investi-gating pollution sources and the ability of receiving waterto absorb loadings.

When watershed goals are set, the planning agency hastwo choices for achieving water quality standards.Individual sites can be forced to comply with the practicesand performance standards set forth in the master plan,or the basin system must be designed and maintained asa public utility. Isolated development tracts can be con-trolled by requiring developers to follow specific sourcecontrol practices, or a simple set of performance standardscan be applied and the choice of practices can be left upto the developer. For example, the agency can require thatrunoff from developed sites must not exceed predevelop-ment intensity. The developer will have to minimize runoff-producing areas and provide detention facilities at the site.

Planners must also consider the effects of their actionson areas outside the watershed. For example, a systemwhere storm flow is detained in a downstream watershedwhile it remains unregulated upstream can cause higherflood levels in a river than a completely unregulated sys-tem.

NATURAL DRAINAGE FEATURES

The key to preserving a natural drainage system for ur-banizing areas is understanding the predevelopment waterbalance and designing to minimize interference with thatsystem. The soil and hydrology of the site must be stud-ied so that high-density, highly impervious locations, such

as shopping centers and industrial complexes, are locatedin areas with low infiltration potential. Recharge areasshould be preserved as open, undisturbed space in parksand woodlands. Runoff from developed areas should bedirected to recharge areas and detained to use the full in-filtration potential. Broad, grassy swales will slow runoffand maximize infiltration. The drainage plan can includevariable depth detention ponds that rise during a runoffevent and return to a base level during dry weather.

Realizing that the design goal is maximizing infiltration-recharge and minimizing runoff, the planner should in-corporate the following techniques into a site plan:

• Roof leaders should discharge to pervious areasor seepage pits. Dry (French) wells, consisting ofborings filled with gravel, can be used for infil-tration of rooftop runoff.

• As much area as possible should be left in a nat-ural, undisturbed state. Earthwork and construc-tion traffic will compact soil and decrease infil-tration.

• Steep slopes should be avoided. They contributeto erosion and lessen recharge.

• Large impervious areas should be avoided.Parking lots can be built in small units and drainedto pervious areas.

• No development should be permitted in floodplains.

Porous pavement is an alternative to conventional pave-ment. (Thelen and Howe 1978; Dinitz 1980). It providesstorage, enhancing soil infiltration to reduce surface andvolume from an otherwise impervious area.

For parking lots and access roads, planners can usemodular pavement systems. Pavers are placed on a pre-pared sand and gravel base, which overlays the subsoil.The voids of the pavers are filled with either sand, gravel,or sod. Frost problems are minimal.


Reversebench

Ditch

Haul road

Interceptor dike

Vegetativebuffer area

Concretechute

FIG. 9.21.1 Interception and diversion measures.


TABLE 9.21.2 SELECTING BMPs BY POLLUTANTS

Methods of Control Structural Vegetative Management

Sediment (TSS, cobble embeddedness, turbidity)Control erosion on land Terraces; diversions; Cover crops and Contour farming; riparian

and streambank grade stabilization rotations; conservation area protection; properstructures; streambank tillage; critical area grazing use and rangeprotection and planting managementstabilization

Route runoff through Sediment basins Filter strip; grassedBMPs that capture waterway; stripcropping;sediment field borders

Dispose of sediment Beneficial use ofproperly sediment—wetland

enhancement

Nutrients: N, P (nuisance algae, low dissolved oxygen, odor)Minimize sources Animal waste system Range management; crop Range and pasture

(lagoon, storage area); rotations management; properfences (livestock stocking rate; wasteexclusion); diversions; composting; nutrientterraces management

Uptake all that is Terrace; tailwater pit; Cover crop; strip Recycle/reuse irrigationapplied to the land or runoff retention pond; cropping; riparian buffer return flow and runoffcontain and recycle/ wetland development zone; change crop or water; nutrientreuse (dissolved form grass species to one that management; irrigationcontrol—commercial is more nutrient water managementnutrients) demanding

Contain animal Diversion; pit/pond/lagoon; See 2(a) Lagoon pump out; properwaste, process and compost facility irrigation managementland apply, or exportto a differentwatershed (dissolvedform control—animalwaste)

Minimize soil erosion Terrace; diversion; Conservation tillage; Nutrient managementand sediment delivery stream-bank protection filter strip; riparian(adsorbed form control) and stabilization; buffer zone; cover crop

sediment pond; criticalarea treatment

Intercept, treat runoff See 1–3; water treatment Riparian buffer zone See four precedingbefore it reaches the (filtration or flocculation) items, this columnwater (suspended for high-value cropsform control)

Pathogens (bacteria, viruses, etc.)Minimize source Fences Animal waste

management, especiallyproper application rateand timing

Minimize movement so Animal waste storage; Filter strips; riparian Proper site selection forbacteria dies detention pond buffer zones animal feeding facility;

proper application rateof waste

Treat water Waste treatment lagoon; Artificial wetland/rock Recycle and reusefiltration reed microbial filter

MetalsControl soil sources Crop/plant selection Avoid adding materials

containing trace metals

Continued



Methods of Control Structural Vegetative Management

Control added sources Tailwater pit; reuse/recycle Crop selection Irrigation watersystem management; integrated

pest controlTreat water Filtration Artificial wetland/rock

reed microbial filtersystem

Salts/salinityLimit availability Drip irrigationControl loss Evaporation basins; Crop selection; saline Irrigation water

tailwater recovery pits; wetland buffer; land-use managementditch lining; replace conversionditches with pipe

Pesticides and other toxinsMinimize sources Plant variety/crop IPM; change planting

selection dates; proper containerdisposal

Minimize movement and Terrace; sediment control Buffer zone; Irrigation waterdischarge basin; retention pond with conservation tillage; management; IPM

water reuse/recycle system filter strips (adsorbedcontrol only); wetlandenhancement

Treat discharge water Carbon filter system (high- Rock-reed microbialvalue crops) filter system/artificial

wetland

Physical habitat alterationMinimize disturbance Road and turnrow Buffer strips; riparian Proper grazing

within 100 feet of water realignment; fencing/ buffer zones management, includinglivestock water crossing limiting livestock accessfacility

Control erosion on land See sediment BMPsMaintain or restore natural Streambank stabilization; Wetland enhancement Proper grazing use and

riparian area vegetation channel integrity repair range management;and hydrology limit livestock access

Sources: U.S. EPA (1993); Brach (1990); Alexander (1993a); USDA, Soil Conservation Service (1988).

EROSION CONTROLS

Erosion control for construction and developing sites willhave a major impact on the total pollution loads in re-ceiving waters. Current estimates show that approximately1500 sq mi of the United States is urbanized annually. Allof this land area is exposed to accelerated erosion.

Following are basic guidelines and principles of erosioncontrol. Reduce the area and duration of soil exposure.For example, various mining operation stages should bescheduled so that clearing, grubbing, scalping, grading andrevegetation occur concurrently with extraction, so that aminimum area is exposed at one time.

Protect the soil with mulch and vegetable cover. For ex-ample, covering the soil surface with wood chips reducesconstruction site soil loss by 92%. Vegetation also has amarked effect on water quality. Temporary fast-growinggrass can reduce erosion by an order of magnitude; sod-

ding can reduce erosion by two orders of magnitude. Strawmulch application can be combined with grass seeding forpermanent surface protection.

Reduce the rate and volume of runoff by increasing theinfiltration rate. A properly roughened and loosened soilsurface will benefit plant growth, enhance water infiltra-tion, and slow surface runoff.

Diminish runoff velocity with planned engineeringworks. A key concept in controlling soil erosion is to in-tercept runoff before it reaches a critical area and divert itto a safe disposal area. Interception and diversion are ac-complished through various structures, including earthdikes, ditches, and combined ditch and dike structures(Figure 9.21.1).

Protect and modify drainage ways to withstand con-centrated runoff from paved areas. To reduce the rate offlow and the resulting detachment and transport of soilparticles in natural and manmade drainageways, grade can

be controlled by the construction of flumes or other flowbarriers across the channel. Bends in the channel, eithernatural or manmade, also impede flow.

Trap as much sediment as possible in temporary or per-manent sedimentation basins.

Maintain completed works and assure frequent inspec-tion for maintenance needs.

Principal cropland erosion control practices and BMPsfor pollutants are summarized in Table 9.21.2.

Maintenance and OperationalPracticesProper maintenance and cleanliness of an urban area canhave a significant impact on the quantity of pollutantswashed from an area by storm water. Cleanliness of anurban area starts with control of litter, debris, deicingagents, and agricultural chemicals such as pesticides andfertilizers. Regular street repair and sweeping can furtherminimize pollutants in stormwater runoff. Proper drainagecollection system use and maintenance can maximize con-trol of pollutants by directing them to treatment or dis-posal.

URBAN POLLUTANT CONTROL

Litter Control

Used food containers, cigarettes, newspapers, sidewalksweepings, lawn trimmings, and other materials carelesslydiscarded become street litter. Unless this material is pre-vented from reaching the street or is removed by streetcleaning, it is often found in stormwater discharges.Enforcement of antilitter laws, convenient location of dis-posal containers, and public education programs aresource control measures.

Chemical Use Control

Reducing the indiscriminate use and disposal of fertilizers,pesticides, oil and gasoline, and detergents is a frequentlyoverlooked measure for reducing stormwater runoff pol-lution. Tree spraying, weed control, municipal fertilizationof parks and parkways, and homeowner use of pesticidesand fertilizers can be controlled by increasing public aware-ness of the potential hazards to receiving waters. Directdumping of chemicals and debris into catch basins, inlets,and sewers is a significant problem that can only be ad-dressed through educational programs, ordinances, andenforcement.

Street Sweeping

Street sweeping is used by most cities to remove accumu-lated dust, dirt and litter from street surfaces, but clean-

ing is usually done for aesthetic reasons. Street cleaningpractices effectively attack the source of stormwater-re-lated problems.

The type of cleaning equipment has an effect on theoverall effectiveness of debris removal. Public awarenessof street cleaning practice is essential for more efficient op-erations. Vehicles parked on the street during sweeping op-erations hamper efficiency and prevent cleaning of de-posits.

Street Maintenance

Pavement conditions have an effect on the amount of streetpollutants. Vehicles travelling over rough streets shake offmore particulate matter. A large portion of solids alsocomes from cracks in the pavement.

Highway Deicing Management

Effective management of highway deicing practices canlessen receiving water impacts associated with chlorides,sodium, and suspended solids. Recommended alternativesfor modifying deicing practices include: (1) judicious ap-plication of salt and abrasives; (2) reducing applicationrates using sodium and calcium salt premixers; (3) usingbetter spreading and metering, and calibrating applicationrates; (4) prohibiting use of chemical additives; (5) pro-viding improved salt storage areas; and (6) educating thepublic and operators about the effect of deicing technol-ogy and best management practices.

COLLECTION SYSTEM MAINTENANCE

The major objective of maintaining storm or combinedsewer systems is to provide maximum transmission offlows to treatment and disposal, while minimizing over-flows, bypasses, and local flooding conditions. This ob-jective can be achieved by maintaining system facilities atpeak capacity.

The significance of collection system maintenance as abest management practice is that when properly applied,extraneous solids and debris are removed in a controlledmanner, not accumulated as pollutant sources to be flushedinto receiving waters under storm conditions.

The basic part of a maintenance program is regular sys-tem inspection. Specific tasks include: (1) catchbasin main-tenance; (2) cleaning (both deposits and root infestation)and flushing of pipes; (3) removal of excess shrubbery anddebris from flood control channels and ditches; and (4)control of inflow and infiltration sources.

Sewer cleaning involves routine inspection of the sewersystem. All plugged or restricted lines should be cleaned.Major problems in large-diameter sewers are siltation andaccumulation of large debris like shopping carts and treebranches. In small-diameter sewers, siltation and penetra-tion of tree roots are major problems. Benefits of sewer


cleaning include reducing local flooding, emergency re-pairs, and pollutant loading. Increased carrying capacitiesand reduced blockages in interceptor/regulator works maydirectly reduce overflows.

Many types of sewer cleaning equipment are used, in-cluding hydraulic, mechanical, manual, and combinationdevices. The cleaning tool is pushed or pulled through thesewer to remove obstructions or cause them to be sus-pended in the flow and carried out of the system. However,large sewer and interceptor cleaning involves unique prob-lems because several feet of sludge blanket can accumu-late.

Regular flushing of sewers can ensure that sewer later-als and interceptors continue to carry their design capac-ity, as well alleviate solids buildup and reduce solid over-flow.

Sewer flushing can be particularly beneficial in sewerswith very flat slopes. If a modestly large quantity of wa-ter is periodically discharged through these flat sewers,small accumulations of solids can be washed from the sys-tem. This cleaning technique is effective only on freshlydeposited solids.

Internally automatic flushing devices have been devel-oped for sewer systems. An inflatable bag is used to stopflow in upstream reaches until a volume capable of gen-erating a flush wave is accumulated. When the correct vol-ume is reached, the bag is deflated with the assistance ofa vacuum, releasing impounded water and cleaning thesewer segment.

INFLOW AND INFILTRATION

Extraneous flow entering a sewer is generally categorized

as inflow or infiltration. Inflow generally occurs from sur-face runoff via roof leaders, yard and area away drains,and flooding of manhole covers. Infiltration usually occursby water seeping into pipes or manholes from leaky joints,crushed or collapsed pipe segments, leaky lateral connec-tions or other pipe failures. Extraneous flows may resultin unnecessary pollution, as these reduce effective collec-tion system and treatment plant facilities.

Details of principal methods of reducing both infiltra-tion and inflow through rehabilitation are found in EPA1977.

DRAINAGE CHANNEL MAINTENANCE

Maintenance of flood control channels covers a wide rangeof cleaning tasks. Debris to be removed ranges from trash,garbage, and yard trimmings to used tires and shoppingcarts.

—Kent K. Mao

ReferencesDinitz, E.V. 1980. Porous pavement. Phase I. Design and operational

criteria. U.S. Environmental Protection Agency (EPA). EPA Report600–2–80–135.

Stewart, B.A., D.A. Woodhiser, W.H. Wischmeier, J.H. Caro, and M.H.Frere. 1975. Control of water pollution from cropland. Vol. I. U.S.Environmental Protection Agency (EPA). EPA Report600–2–75–026a. Washington, D.C.

Thelen, E. and L.F. Howe. 1978. Porous pavement. Philadelphia, Pa.:The Franklin Institute.

U.S. Environmental Protection Agency (EPA). 1977. Sewer system eval-uation, rehabilitation, and new construction: A manual of practice.EPA Report 600–2–77–017d.


9.22FIELD MONITORING PROGRAMS

The objectives of field monitoring water quality in drainagestudies include:

• Analyzing the impact on receiving waters of (1)storm sewer discharges, (2) combined sewer over-flows, (3) atmospheric fallout and urban activi-ties, and (4) new facilities or treatment plants de-signed to reduce environment impacts.

• Identifying the contributions of various land usesto total pollution discharge, to optimize urban de-velopment and derive some regulations such assource control.

• Increasing existing treatment efficiency during wetweather in combined sewer systems.

• Analyzing of scour and deposit problems in sew-ers to define optimal cleaning sequences or to de-sign facilities for better hydraulic conditions.

To fulfill these objectives, storm water discharges needto be sampled during dry-weather and wet-weather con-ditions. Water quality data gathered during dry weatherprovide a baseline and indicate point source impacts.

To trace contaminants and identify pollutant sources,a phased monitoring approach requires repeated investi-

gation of land use activities in a basin. The program is ex-pected to be an iterative process, as several rounds of sam-pling are generally required. Precise data are essential forcalibrating and verifying nonpoint source models.

Experience proves that water quality data collectionprograms can be costly. Collection procedures have highmanpower requirements, as frequent site visits are re-quired. The cost of analyzing collected samples may in-crease rapidly with the number and types of pollutantsstudied. It is important that the parameters to be studiedare carefully selected and limited to the essentials.

This section presents an outline of water quality param-eters important in studies on urban stormwater discharges,and reveals the main difficulties in obtaining representa-tive samples. Also included is a brief discussion on dataanalysis.

Selection of Water QualityParametersWater quality parameters included in urban hydrologicalstudies may be divided into seven groups. Those parame-ters, relating to a specific drainage problem, are listed inTable 9.22.1, along with their detection limits, precisionlevel of analysis, and study objectives. In most cases, onlybiochemical oxygen demand (BOD), chemical oxygen de-mand (COD), and total suspended solids (TSS) are initiallystudied, but if these parameters show high values, someother parameters can be taken into account (i.e. Kjeldahlnitrogen, total phosphorous, and volatile suspended solids[VSS]). As the program continues, some special investiga-tion should be made on trace elements and other specialparameters mentioned in Table 9.22.1.

Solids are good indicators of urban water quality, asthey may contain pollutant materials. Suspended solids areclosely related to other pollutant concentrations. In fact,sample uniformity is not easily achieved. Suspended solidconcentrations are affected by the flow level, which is nottaken into account by manual or automatic sampling. Thesampler itself may also introduce effects that can modifythe gradient profile of suspended solids. Conditions at thesampler intake cannot be adapted to the flow variationsencountered in storm sewers or combined sewers duringhigh flows.

If the sampler cannot be precisely measured in the col-lected samples, sample uniformity can be questionable. Inmost cases, suspended solids are regarded as a rough in-dicator of water quality, so this should be among the pa-rameters selected.

Acquisition of RepresentativeSamplesThe number of sampling sites, the frequency of measure-ments, and the quality parameters to be measured should

be chosen. This requires knowledge of the sewer network,significant building activities, street cleaning practices, at-mospheric pollution sources affecting the experimentalsites, erosion patterns in surrounding natural areas, in-dustrial activities, seasonal or climatic changes, etc., in or-der to avoid erroneous judgements in understanding thephenomena studied.

The experimental design must be in agreement with thestudy objectives (Geiger 1981, Gideometeozdat 1984a,Wong and Marsalek 1981). However, trial and error pro-cedures should be used at the beginning of the study fora few basic parameters (for example, BOD, COD, TSS) ata few sampling sites. This information should be usedwhen determining the experimental design.

SAMPLING SITES AND LOCATION

Sampling sites must be chosen according to study objec-tives, but hydraulic conditions and constraints necessaryto the adopted procedures should be given attention. Thesampling site must be located at a section downstream ofthe study site, i.e. corresponding to well-known sewer sys-tems, land use types, special activities, etc. It is recom-mended that highly turbulent sections with well mixedflow be sampled. However, for the study of sediment trans-port deposit, these conditions may not be suitable, as sus-pended solids in the highly turbulent section may be scat-tered.

For monitoring in-stream impacts, the area of interestshould be bracketed by upstream and downstream sta-tions. A control station on a hydrologically similar butundisturbed watershed can be used to determine baselineconditions.

Two types of monitoring stations are employed for non-point source surveys:

1. Small catchment stations ranging from 12 to 125acres (5 to 50 ha) in size, are used to gather data on spe-cific land uses or special areas. They are usually found onstorm sewers, drainage ditches or small tributaries.

2. Another type of station is built to monitor largerbasins of greater than 125 acres (50 ha), and measure non-point source pollution loads impacting a receiving body,such as stream channels or rivers.

There are cases where the final choice must be madefrom a group of catchments. In such cases, the techniqueof weighted suitability ratings, as developed for land use,is recommended (Alley 1977). Assignment of suitabilityvalues is perhaps the most subjective part of the schedule.

SAMPLING METHODS

Due to the transient nature of storm runoff phenomena,random collection of grab samples does not allow a truerepresentation of pollutant transport. Even if grab samplesare modified to concentrate on storm events, the error po-


tential remains quite high because of variations in pollu-tant concentrations during runoff events.

Two basic methods can provide estimates of pollutantloading during a storm event:

1. To determine total pollution loading during a stormevent, a flow-weighted composite method is adequate. Inthese methods, either aliquot volume or time between

aliquots is varied to construct a truly flow-weighted com-posite from many samples. Analyzing the composite sam-ple and using synoptic flow data allow computation of anaccurate estimate of runoff pollution loads, if the intervalsbetween samples are short.

2. When, in addition to total pollution loading, it isnecessary to investigate load variations during a stormevent, the sequential discrete procedure must be used. A


TABLE 9.22.1 WATER QUALITY PARAMETERS SAMPLED IN URBAN DRAINAGE STUDIES

Detection Precision Level Study ObjectivesParameters Limits (absolute or relative) or Observations

Common Constituents and IndicatorsChlorides 2.5–5%* Impact of salts used for deicingWater temperature 0.1°C–0.5°C Cross-connections; parasites in

watersConductivity (at 20°C) 5mS/cm 5% Changes during runoff, monitoring

and controlpH 0.05–0.1 unit* Rainfall quality analysisTurbidity Sediment transport

NutrientsKjeldahl nitrogen 0.1 mg/l Impact on receiving watersTotal phosphorus 5–15%* Eutrophication processAmmonia 0.001 mg/l 1–10%* Impacts on detention basins with

recreational purposesNitrites and nitrates 0.05 mg/l 4.5–18% Cross-connections

Organic IndicatorsBOD5 (5 day BOD) 2–25 mg/l Impact on receiving waters by

oxygen depletionCOD (with ,1.5 g/l 1–5% (if COD Cross-connections

chlorides) .50 mg/l)

Trace ElementsLead Impact on receiving waters; toxicsZinc and other 0.05–3 mg/l 2–10%† accumulation in sediments

heavy metals

SolidsTSS (at 105°C) 0.5 mg/l 2–5% Turbidity, oxygen reduction,

transport of toxics; increase ofhydraulic roughness

VSS (at 550°C) 1 mg/l Organic part, oxygen depletionSettleable solids Maintenance problems in sewers

and detention basins inrecreational areas

Bacterial IndicatorsTotal coliforms Impact on receiving waters with

recreational useFecal coliforms Detection of cross-connections

Special ParametersPersistent toxic substances 0.00005 mg/l 0.005–0.05 g/l† Impact on receiving waters

(PTS) such as organochloride Pollution of receiving waterspesticides sediments

Polyaromatic hydrocarbons 1–5 mg/l* Bioaccumulation in food chainsChlorinated benzenes 0.002–0.02 mg/l†

*Depending on the instrument and/or analysis method.†Depending on the substance analyzed.

series of samples is retrieved during a monitored runoffevent. Following laboratory analysis of each sample andanalysis of synoptic flow data, the runoff hydrograph anda curve of pollutant concentration or loading as a func-tion of time may be plotted as shown in Figure 9.22.1. Bydetermining the area under the curve, an accurate estimateof the total pollutant load for an event may be determined.

The interval between sample collection for the aboveprocedures depends on the response time and duration ofthe storm. In general, at least four samples on the risinglimbs and six samples on the recessing limbs should be col-lected for proper resolution of nonpoint source pollutionloads in urban areas.

Samples may be collected either manually or by auto-matic samplers. Table 9.22.2 shows a matrix of advan-tages and disadvantages related to each sampling tech-nique. A summary of methods used in urban stormwatersampling and comments on each was prepared by Shelley(Shelley and Kirkpatrick 1975).

Experimental results show sediment distribution in astream cross section flowing at 5 ft/sec. An analysis of wa-ter quality constituents in the stream cross-section shouldbe made to determine the distribution across the width andfrom top to the bottom of the stream. Samples should betested for a suspended parameter (such as TSS) and a sol-uble parameter (such as orthophosphate). The testingshould be carried out at a small runoff event and a mod-erate-to-high flow event if possible. Vertical samplingshould be done using depth samplers (such as Kenmeyerbottles) or closeable bottles if the stream is more than 4to 5 ft (1.2 to 1.5 m) deep. This factor should be consid-ered in designing manual and automatic sampling proce-dures.

FLOW MEASUREMENT

Flow measurement is perhaps one of the most importantaspects of designing an urban collections program. No datacollecting task will be capable of achieving its goals if theprecision and accuracy of the flow data required for loadcalculations are not considered.

The flow measurement devices and methods can be clas-sified according to the physical principles upon which theirprimary elements are based.

Channel Friction Coefficient Method

This indirect method, also referred to as the slope-areamethod, consists of measuring flow depth at a suitablecross-section and substituting the measured depth into anequation for uniform flow (such as the Manning equation)or critical flow. To complete the calculation, one must es-timate the friction coefficient of the channel where the flowis to be measured, and know the channel slope and geom-etry.

The inference of flow rates from measured depths offlow is a rather inaccurate procedure. The main sourcesof error arise from the lack of uniformity and steadinessof flow, and the lack of certainty in estimating the frictioncoefficient.

Improved accuracy can be achieved by performing cal-ibration in place, and developing an empirical rating curvefor each measuring cross-section. In this case, the channeldischarge (Q) is measured, generally by current meters, forvarious depths of flow, and the cross-section rating curve(Q vs depth of flow) is developed. This curve is then usedto convert the observed stage to discharge.

Weirs

Measuring weirs are overflow structures built across a flowchannel to measure discharge. For a given set of weir andchannel geometry conditions, a single head value on thedevice may exist for each discharge under a free-flow,steady state regimen. The existence of such a relationshipmakes constructing a rating curve of head versus dischargea simple task. Such rating curves are available in the lit-erature for most common configurations (such as rectan-gular weirs, V-notch weirs, vertical slot weirs, and trape-zoidal weirs without the bottom part) (U.S. Departmentof Interior 1975).

One advantage of weirs is their large relative measure-ment range. However, weir installation in sewers reducespipe capacity, may lead to solids accumulation (particu-larly in combined sewers), may distort flow hydrographs,and may limit operating range because of surcharging orsubmerging. These constraints will eliminate weirs fromconsideration for certain locations, but many of the abovedifficulties can be avoided in open-channel installations atoutfalls. For these reasons, weirs should be used only un-der carefully controlled conditions, such as at detentionbasin outlets, where suspended solid concentrations arelikely to be low.

Flumes

A measuring flume creates a constriction in the channelcross-section, causing a velocity change and, consequently,


0 8 16 24 32 40 48 56

1.2

1.0

0.8

0.6

0.4

0.2

0.0

PH

OS

PH

OR

US

LO

AD

ING

, lb/

hr

RUNOFF DURATION, min

FIG. 9.22.1 Plot of total phosphorus loading at irongate catch-ment.

a depth change. In critical flow flumes, the surface profilein the constriction passes through the critical depth. Theflume discharge can then be directly related to the depthimmediately upstream of the throat.

Flumes are sometimes classified according to throat

shape. Common types include rectangular, trapezoidal,semicircular, and composite throat flumes. Flumes with abottom contraction (a hump) are suitable for installationin sewers. The Parshall flume, the cut-throat flume, andthe Palmer-Bowlus flume are also popular.


TABLE 9.22.2 COMPARISON OF MANUAL AND AUTOMATIC SAMPLINGTECHNIQUES

Advantages Disadvantages

Manual GrabsAppropriate for all pollutants Labor-intensiveMinimum equipment required Environment possibly dangerous

to field personnelMay be difficult to get personnel

and equipment to the storm wateroutfall within the 30 min requirement

Possible human error

Manual Flow-Weighted Composites (multiple grabs)Appropriate for all pollutants Labor-intensiveMinimum equipment required Environment possibly dangerous

to field personnelHuman error may have significant

impact on sample representativenessRequires flow measurements taken

during sampling

Automatic GrabsMinimizes labor requirements Samples collected for O&G mayLow risk of human error not be representativeReduced personnel exposure to Automatic samplers cannot

unsafe conditions properly collect samples for VOCSampling may be triggered analysis

remotely or initiated according Costly, numerous sampling sitesto present conditions require the purchase of equipment

Requires equipment installationand maintenance

Requires operator trainingMay not be appropriate for pH

and temperatureMay not be appropriate for

parameters with short holdingtimes (e.g., fecal streptococcus,fecal coliform, chlorine)

Cross-contamination of aliquot iftubing/bottles not washed

Automatic Flow-Weighted CompositesMinimizes labor requirements Not acceptable for VOC samplingLow risk of human error Costly if numerous sampling sitesReduced personnel exposure to require the purchase of equipment

unsafe conditions Requires equipment installationMay eliminate the need for and maintenance, may malfunction

manual compositing of aliquots Requires initial operator trainingSampling may be triggered Requires accurate flow

remotely or initiated according measurement equipment tied toto on-site conditions sampler

Cross-contamination of aliquot iftubing/bottles not washed

Rating curves for critical flume geometry may be con-structed from solution of the Bernouilli Equation at pointsupstream of and in the flume throat. While they generallyexhibit excellent characteristics of self-cleaning, flumes donot share the brand flow measurement characteristics ofweirs.

There are a large number of other flume designs thatcan be used in drainage studies. For example, the SoilConservation Service has HS, H, and HL flumes designedto measure small, moderate, and large runoff flows, re-spectively. These devices combine the best features of bothflumes and weirs, with wide ranges of measurement, goodself-cleaning characteristics, small head loss, and relativeinsensitivity to submergence.

Differential Pressure Methods

Traditionally, differential pressure flowmeters have beenused to measure flows in full closed conduit. Two excep-tions to this rule are the U.S. Geological Survey andUniversity of Illinois sewer meters. Although these func-tion as differential meters in the pressure flow region, theyare also fully functional in the open-channel flow region,where they act as Venturi flumes. This dual mode of op-eration represents the main advantage of these flowmeters.

The U.S. Geological Survey (USGS) flowmeter is simi-lar to flumes with a U-shaped throat. The flume does notobstruct the part of the pipe immediately below the crown,thus transition from open-channel flow to pressure flowis fairly smooth and head losses are reduced. Rating curvesfor the USGS flowmeter are available (Smoot 1975).

Dilution Method

In this method, a tracer is continuously injected at a con-stant rate into the flow, and tracer dilution by the meteredflow is monitored at a downstream point. If a tracer ab-sent in the meter flow is used, the following relationshipapplies

QD 5 qTCT/CD 9.22(1)

where:

QD5 flow upstreamqT 5 tracer input flowCT 5 tracer input concentrationCD 5 tracer concentration downstream

The dilution method has some definite advantages, be-cause it is independent of flow characteristics, does not in-terfere with the flow and, consequently, does not causeany head loss. Using fluorescent dyes and ensuring com-plete tracer mixing, the method has a good range of mea-surement (1000 : 1), and can be fairly accurate (5%) (Alley1977). Disadvantages are the discrete nature of measure-ment, as opposed to the preferred continuous measure-ments; the problem with automating the method; and theneed for well-trained personnel. Consequently, the dilu-

tion method is mostly used for in-situ calibration of con-ventional flowmeters.

Basic characteristics of flow measurement methods dis-cussed are summarized in Table 9.22.3.

Sampling EquipmentMANUAL SAMPLING

Certain manual techniques cannot be avoided in studiesof urban runoff quality. Manual sampling is useful whensetting up automatic equipment, selecting the sampling sec-tion, and the inlet location.

Manual sampling requires good logistic preparation.Field crews must be dispatched to sampling sites beforethe start of a runoff event, so that sampling can start atthe beginning of runoff. This is particularly important incombined sewers which exhibit the first flush phenome-non with high pollutant loads occurring early duringrunoff events. Therefore, field crews may have to be sta-tioned at sampling sites. Extensive field training is essen-tial to ensure collection of adequate samples.

AUTOMATIC SAMPLING

To obtain necessary flow measurements along with stormwater samples, two devices are required: one for flow me-tering and one for flow metering with an interconnectionto insure synoptic collection of sample and flow data.Common characteristics of adequate devices are summa-rized below:

• Sample transport velocity of 3.0 fps or more toprevent sedimentation

• Minimum of 24 discrete sample bottles or abilityto composite samples in one container

• 12 v dc supply option• Constant sample size over different sampling lines

for rising and falling streams• Air purging of sampling intake line before and af-

ter sample collection• Minimum 3/8 in (or 1 cm) sample line• No solids deposition in sample train• Chemically inert surfaces in contact with sample

In general, the intake should point upstream, extendedslightly upstream from any obstacles in the flow, andshould not excessively obstruct flow to avoid clogging ordamage. Locations are recommended along the pipe pe-riphery at about one third of the average water depth abovethe bottom. The intake should be placed at a cross-sectionwhere the flow is highly turbulent and well mixed. At suchlocations, a single intake, instead of multiple intakes, maybe acceptable.

Sample withdrawal is accomplished by a pump con-trolled by timers or flow meters. The best devices for ur-ban pollution studies fall into the following categories ofpumping methods: positive displacement, peristaltic, and


vacuum. Suction lift devices are the best means of samplewithdrawal. Such devices have to operate near the flowsampled because the lift is limited to about 15 ft (5 m).Submersible positive displacement pumps are commonlyused where equipment installation is restricted to locationstoo high above the water surface to operate in a suctionlift mode. Although such pumps allow sampling at greaterdepth, they are susceptible to malfunction and clogging.

FLOWMETERING DEVICES

Selection of secondary devices for the continuous mea-surements necessary to convert from stage to discharge isan important facet of developing an automated monitor-ing program. Important criteria for these secondary de-vices include:

• Wide measurement range• Accuracy and precision over the entire range• Minimal calibration loss with time• Insensitivity to suspended solids in flow• Capacity to internally convert stage to discharge• Capacity to trigger an associated sampler• Unattended operations

Secondary devices are divided into four categories: float-operated devices; ultrasonic devices; bubbler devices(manometers and transducers); and combination bubbler-magnetic devices.

Bubbler-Operated Devices

In the simplest of designs, a float is connected to a stripchart or digital recorder via flexible steel tape. In most ap-plications, float-type devices require a stilling well to dampout surges and rapid fluctuations in water surface eleva-tion. In addition, most float-operated devices do not pro-vide an internal stage-to-discharge conversion.

Ultrasonic Devices

These secondary devices rely upon the travel time of anultrasonic signal from a transponder to the water surfaceand back. This type of meter functions in a noncontactmode, and is therefore free from clogging and freezing.However, ultrasonics are sometimes subject to spurioussignals from floating matter and foam. Some devices haveinternally programmable read-only memories (PROMs)and microprocessor circuitry to provide stage-to-dischargeconversion using the unique relationships of the primarydevice.

Bubbler Devices

In bubbler devices, gas is forced through a fixed orifice,oriented to assure that only static head is measured. Thestatic pressure required to maintain a given bubble rate isproportional to the height of the water column above the


TABLE 9.22.3 CHARACTERISTICS OF SELECTED FLOW MEASUREMENT METHODS

CharacteristicsSuitable For Applicable At

EstimatedOpen Channel Pressure In Sewer Accuracy* RelativeFlow Flow Outfall Manhole Pipes (%) Costs

Depth and channelfriction coefficient X X X X 15–20 low

Depth and stage-dischargerelationship X X X 10–15 low

WeirsRectangular X X 5† low toV-notch X X 5† mediumModified trapezoidal X X X X 5†Vertical slot X X X X 5†

FlumesCut-throat X X 5†Palmer-Bowlus X X X X 5† mediumParshall X X 5†USDA (H, HL and HS) X X 5†

Differential pressure flowmetersU. of Illinois X X X 5 mediumUSGS X X X 5 to high

Tracer dilution X X X X 5 medium

*Under favorable conditions.†These relatively high accuracies correspond to well-designed, installed, and operated installations. Under less favorable circumstances, the accuracies would be

somewhat lower, between 5 and 10%.

TABLE 9.22.4 SUMMARY OF DATA ANALYSIS METHODS

Level of Analysis and Methods Examples References

Design of ExperimentsFactor analysis Choosing experimental catchments or measuring Cochran & Cox, 1957; Kendall

sites for a given experimental program: & Stuart, 1973; Snedecor &land uses Cochran, 1957catchments parameterswater quality sampling

Choosing number of experiments using physicalmodels

Raw Data CriticismDouble mass analysis Testing for systematic errors in time data series

such as cumulative rainfall or runoff depths atvarious points in the same climatic areas

Parametric tests Testing the random aspect of a data series such Bennet & Franklin, 1967;(Anderson test) as rainfall and runoff Dagnelie, 1970; Haan, 1977;

Pearson & Hartley, 1969Nonparametric tests Testing of the hypothesis on equal variance of two Dagnelie, 1970; Kendall &

Variance ratio test, populations: rainfall runoff, runoff quality data Stuart, 1973; Kite, 1976;Bartlett’s test, et al. Pettitt, 1979

Wilcoxon, Mann-Whitney, Testing of the hypothesis on equal means andKruskal-Wallis, Wilks tests identical location of population: rainfall or

runoff, runoff quality data from several catchments

Statistical parameters Comparison of samples and homogeneity testing All books on statisticalArithmetic mean (or Parameters can be time and/or flow weighted for methods

geometric mean for data runoff quality datalognormally distributed)

Variance or standarddeviation

Ranges Preliminary statistical analysisPearson’s and Fisher’s

coefficients

Point-Frequency AnalysisEmpirical frequency plotting Analysis of a separate variable considered as a Adamowski, 1981; Bennet &

random variable: Franklin, 1967; Cunnane, 1973;Probability papers rainfall depths for various time intervals Dagnelie, 1970; Haan, 1977;Plotting formulae (I.D.F. curves) Kendall & Stuart, 1977a; Kite,

peak runoff 1976; Snedecor & Cochran, 1957;risk analysis Yevjevich, 1972b

Choice of a theoretical probability distributionTheoretical probability Almost all hydrological variables (rainfall, runoff, Chow, 1964; Dagnelie, 1970;

(distributions discrete and quantity, quality) considered as a random Gumbel, 1960; Haan, 1977;continuous) variable Kendall & Stuart, 1977a; Kite,

1976; Linsley et al., 1975;Method of moments Snedecor & Cochran, 1957;Method of maximum Viessman et al., 1977;

likelihood Yevjevich, 1972bHypothesis testing and Testing the adequacy of a given probability Chow, 1964; Dagnelie, 1970;

confidence intervals distribution to a given sample Haan, 1977; Kendall & Stuart,1977a; Kendall & Stuart, 1973;

Tests of means and variances Kite, 1976; Snedecor & Cochran,Goodness-of-fit tests 1957; Yevjevich, 1972b

Multivariate AnalysisSimple Regression Analysis Applied to a pair of hydrological variables Chatfield & Collins, 1980;

best fit procedure choice rainfall and runoff volumes Haan, 1977; Morrison, 1976;tests of fit runoff coefficients and imperviousness Draper & Smith, 1966; Haan,spurious correlations rainfall depths at two sites 1977; Viessman et al., 1977

overland flow detention storage and dischargepollutants loads and peak runoff etc.

Continued




Level of Analysis and Methods Examples References

Multivariate probability Applied to several independent variables considered Adamowski, 1981; Dagnelie,distributions to be purely random variables 1970; Kendall & Stuart, 1977a;

risk analysis in urban water management Kite, 1976; Yevjevich, 1972a;spatial rainfall depths distribution Yevjevich, 1972bhydrological stochastic processes (discrete andcontinuous)

Multiple regressions analysis Applied to one explained variable Y and to several Chatfield & Collins, 1980;explanatory variables Xi: Draper & Smith, 1966; Haan,

Simple matrix procedure interpolation between a set of raingauges 1977; Pearson & Hartley, 1969;of best fit generation of data for incomplete data series Robitaille and Bobbée, 1975;

Stepwise regression rainfall-runoff modeling at a given location versus Stone, 1974; Yevjevich, 1972aprocedure rainfall and/or runoff at other locations

Orthogonal regression runoff coefficients versus urban catchmentprocedure parameters and rainfall characteristics

Ridge regression procedure lag times and times of concentration versus Cross validation procedure catchments and rainfall parameters

Better results when Xi Urban runoff pollutant loads versus rainfall and variables are correlated runoff parameters, catchment characteristics

such as land uses, imperviousness, slopes, etc.Interdependence analysis Mostly for qualitative analysis. Not frequently Chatfield & Collins, 1980;

applied in urban hydrology Haan, 1977; Morrison, 1976Correlation analysis Just two variablesPrincipal components More than two variables: reduction of dimensionality,

analysis (P.C.A.) preliminary analysis for regression procedures.Sometimes quantitative

spatial distribution of rainfallurban runoff pollutants loads

Factor analysis Similar aims as P.C.A. but with assumption of a proper statistical model. Covariance analysis

Cluster analysis Grouping tests of individualsDiscriminant analysis Separation of individuals in two populations.

Preliminary analysis for regression proceduresTime Series Analysis Testing the random aspect of a given variable for Bartlett, 1966; Box & Jenkins,

preliminary statistical analysis 1970; Cox & Miller, 1968;Jenkins & Watts, 1968;

Stochastic modelling of hydrological processes (not Kendall & Stuart, 1977b;very frequent in urban hydrology due to time and Yevjevich, 1972aspace intervals to be considered)

Trend analysis Testing gradual natural or man-induced changes inTests of randomness data seriesLeast squares procedures Changes in urban hydrological data due to Moving average methods continuous urbanization

Periodic analysis Testing the existence of cycles:Seasonal aspects of rainfall, runoff, quality,

quantity dataShort cycles due to some industrial or domestic

water usesSpectral analysis on Testing the random aspect of a given process

the time domain(Autocorrelation function)

Spectral analysis on the Identifying Instantaneous Unit Hydrographs (IHU)frequency domain for small urbanized catchments(Spectral density function)

orifice. The static pressure may be measured either by theinclined manometer or pressure tranducer devices. Somedevices are available with internal PROMs for flow datareduction. In fast flowing water, the dip tube may be pro-tected by a simple still well: a concentrically placed perfo-rated tube. Shortcomings include contaminant build-up onthe dip tube in the vicinity of the measuring tube, and rel-atively low accuracy in the total part of the total pressurerange.

Combination Bubbler-Magnetic Devices

These instruments rely on velocity-area measurement tocompute instaneous flow rates. Stage measurements aremade using conventional transducer bubblers. These dataare converted to area measurements using a PROM thatdescribes conduit geometry. Flow velocity measurementsare made at the same time with an electromagnetic devicelocated in a band attached to the conduit wall. Using theindependent values of area and velocity, the device com-putes discharge.

Other Monitors

When planning any atmospheric precipitation measure-ment, contact the National Meteorological Institute orequivalent organization for expert assistance.

Various types of open containers and man-made nat-ural surfaces are used to collect impurities deposited by at-mospheric forces. Open containers are generally polyeth-ylene, polypropylene, or glass funnels or cylinders. Variousmodified gauge types are designed for special purposes.

Precipitation intensity, volume, and duration datashould be collected during qualified sample events, and formonitoring programs. Nonrecording gauges are used formeasurement by most government hydrological and me-teorological services. The ordinary rain gauge used fordaily readings is a collector above a funnel leading to a re-ceiver. Continuous registration has also been incorporatedinto rain gauges. Precipitation recorders in general use arethe weighing, tipping-bucket, or float type. If the standardrain gauge is sited in the wind direction, it should be sur-rounded by an 0.4 m-high screen. A wind shield consist-ing of a frame of hanging strips is placed within 1 m ofthe recorder.

QA/QC MeasuresA quality assurance/quality control program should be de-veloped and implemented as part of a long-term monitor-ing program to provide assessment of techniques used dur-ing sample collection, storage, and analysis (EPA 1979b,1980). EPA programs require QA/QC plans to be ap-proved by the EPA prior to sample collection and analy-sis. The QA/QC plan should specify sample collection andpreservation methods, maximum sample holding time,chain-of-custodian procedure, analytical techniques, accu-

racy and precision checks, detection limits, and datarecording and documentation procedures.

SAMPLE STORAGE

The preceding steps will not guarantee a representativesample unless container selection and sample preservationmethods meet the required standards. The choice of con-tainer and cap materials is very important due to the pos-sibility of interference with constituents to be analyzed.Containers and all elements involved in sampling or com-positing operations must be properly cleaned. More de-tailed information on container types and cleaning is foundin EPA 1980b. Recommended operations are as follows:

Container Selection

Containers can introduce positive or negative errors intrace metal and inorganic measurements by contributingcontaminants through leaching or surface desorption or,depleting concentrations through absorption. Samples tobe analyzed for toxic metals can be stored in 1-l polyeth-ylene or glass bottles with polypropylene caps. Teflon lidliners should be purchased or cut from sheet teflon andinserted in caps to prevent possible contamination fromcaps supplied with bottles.

Container Cleaning

Due to the sensitivity of tests examining waterborne tracemetals, sample containers must be thoroughly cleaned. Thefollowing schedule must be followed for the preparationof all sample bottles and accessories, whether glass, poly-ethylene, polypropylene or Teflon:

• Wash with detergent and tap water• Rinse with 1:1 nitric acid• Rinse with tap water• Rinse with 1:1 hydrochloric acid• Rinse with tap water• Triple rinse with distilled (or deionized) water

SAMPLE PRESERVATION

Water samples are susceptible to rapid physical or bio-logical reactions that may take place between samplingand analysis. This time period can exceed 24 hr due tolaboratory capacity needed to handle unpredictably vary-ing amounts of samples resulting from aleatory rainfalls(Geiger 1981).

Preservation techniques are recommended to avoidsample changes resulting in large errors. Refrigeration ofsamples at 4°C is commonly used in fieldwork and helpsto stabilize samples by reducing biological and chemicalactivity. All samples except metals must be refrigerated.

In addition to refrigeration, specific techniques are re-quired for certain parameters. They consist of the addition


of chemical compounds, biocides, etc. More detailed in-formation can be found in EPA 1979a, 1980b.

The decision to eliminate a portion of the drainage sys-tem from further sampling must include a review of dataQA/QC procedures. Review of contaminant data fordrainage systems must be performed to ensure that ana-lytical results are properly interpreted, and that detectionof potential sources is not missed because of field or lab-oratory constraints.

Analysis of Pollution DataWhen considering the significance of runoff pollutant con-tributions, both concentrations and total loadings must beexamined. Receiving water concentrations are usually ofprime concern. In principle, if pollutant concentrations donot exceed certain allowable maxima, detrimental effectswill not occur. However, for many pollutants, allowableconcentrations in water are not known. Because of sedi-mentation, accumulation of benthal deposits such as phos-phorous, hydrocarbons, and heavy metals may be moresignificant than concentrations in the water. Receiving wa-ter column concentration and benthal accumulation de-pends more on mass loads of pollutants than of pollutantconcentrations in the runoff.

Judging from the above, an estimate of mass loadingsof the principal pollutants are of great importance for plan-ning and management purposes. Priority should be placedon obtaining a reliable estimate of mass loadings enteringa body of water because of urban runoff during a specifictime, such as a year, and data collection plans should beso designed.

STORM LOADS

Data from nonpoint source monitoring studies are usuallyreduced to a pollutant load per storm basis. An event ex-pected concentration (EMC) is multiplied by the value ofrunoff. EMCs are calculated by integrating the polluto-graph (instantaneous load with time) with the hydrograph.After sampling several storms, these load per storm dataare used to estimate annual load from the basin. It is as-sumed that the monitored storms are representative sam-ples of storms usually occurring in the basin during theyear (Whipple 1983).

ANNUAL LOADS

Regressions of total load versus total runoff from a fam-ily of storms give a slope in concentration units, which canbe used to predict pollutant load for a specific quantity ofrunoff. To better represent the actual runoff process, baseflows were abstracted from storm runoff or low-flow loadsfrom storm load. Good correlations have been found us-ing log-load versus log runoff volume, reflecting log-nor-mal distribution of the concentration data. Nonpoint

source data are usually log-normal distributed, as are hy-drologic events.

After mass loadings for a given land-use type are accu-mulated over a considerable period of time, results can beexpressed in terms used to estimate loadings from that typeof land use for the rest of the watershed(s) of interest. Theapproaches below are commonly used:

1. Annual loading/area of given land use, lbs/acre/yr2. Annual loading/curb mi of given land use, lb/mi/yr3. Annual loading/traffic volume, lbs/vehicle/yr4. Annual loading/air pollution index, lbs/avg in/yr5. Annual loading/runoff volume, lbs/million gal6. Annual loading/precipitation amount, lbs/in (for spe-

cific area)

Number 1 assumes that pollution varies according toland use. This is the most commonly used method of pre-dicting loading under future conditions. Number 2 as-sumes that pollution loading varies with the number ofcurb mi in various stages of development. Numbers 3 and4 make similar assumptions regarding automobile trafficand air pollution. Numbers 5 and 6 are designed to con-vert loading data from specific storm events to annual av-erage loadings, which are then converted to relationshipswith land use for predictive purposes.

SIMULATION MODEL CALIBRATION

Monitoring data can be used to calibrate sophisticatednonpoint source computer models. These models attemptto interpret the mechanisms involved in nonpoint sourcegeneration. Alternatives can then be evaluated using com-puter-simulated processes. Models exist for different typesof basins, levels of complexity, and nonpoint source prob-lems. They need good calibration data.

Statistical AnalysisUrban hydrological phenomena, especially those involvingstorms, give historical data that can be observed only once,and then will not occur again. Such collected data forman ever-growing sample of measurements. Even if somephenomena can be described by means of physical or ra-tional theory, the input, rainfall, is commonly stochasticin nature and whole phenomena can be amenable to sta-tistical interpretation and probability analysis.

Table 9.22.4 summarizes data analysis methods alongwith examples and references. Although it is impossible tosummarize the many references in general or applied sta-tistics, good basic knowledge is provided in books such asKendall and Stuart’s theory of statistics, or Haan’s (1977)statistical methods in hydrology.

—Kent K. MaoDavid H.F. Liu


ReferencesAlley, W.M. 1977. Guide for collection, analysis, and use of urban storm-

water data, conference report. American Society of Civil Engineers(ASCE) New York, N.Y.: American Society of Civil Engineers(ASCE).

Geiger, W.F. 1981. Continuous quality monitoring of storm runoff.Water Science Technology. Vol. 13, pp. 117–123. Munich:IAWRR/Pergamon Press Ltd.

Gideometeozdat. 1984. Complex assessments of surface water quality.(In Russian). p. 140. Leningrad.

Haan, C.T. 1977. Statistical methods in hydrology. Ames, Iowa: TheIowa State University Press.

Shelley, P.E., and G.A. Kirkpatrick. 1975. An assessment of automaticflow samplers—1975. U.S. Environmental Protection Agency (EPA).U.S. EPA 600–2–75–065. Washington, D.C.

Smoot, G.F. 1975. A rain-runoff quantity—quality collection system.Proceedings of a research conference on Urban Runoff Quantity and

Quality. American Society of Civil Engineers (ASCE). New York,N.Y.

U.S. Department of Interior, Bureau of Reclamation. 1975. WaterMeasurement Manual.

U.S. Environmental Protection Agency. 1979a. Methods for chemicalanalysis of water and waste. Washington, D.C.: EPA.

U.S. Environmental Protection Agency. 1979b. Monitoring requirements,methods, and costs for the nationwide urban runoff program. EPAReport 600–9–76–014. Washington, D.C.

U.S. Environmental Protection Agency. 1980. Monitoring toxic pollu-tants in urban runoff, a guidance manual. U.S. EnvironmentalProtection Agency (EPA), Office of Water Regulation and Standards.

Wong, J. and J. Marsalek. 1981. Persistent toxic substances in urbanrunoff. Proceedings of Storm Water Management Model Users GroupMeeting, Niagara Falls, Ontario, Canada: U.S. EnvironmentalProtection Agency (EPA). EPA Report, pp. 455–468.

Whipple, W., et al. 1983. Stormwater management in urbanizing areas.Englewood Cliffs, N.J.: Prentice-Hall.


9.23DISCHARGE TREATMENT

Three types of treatment are used for wastewater dis-charges: biological, physical-chemical, and physicalprocesses. Some systems use two or all types to achievebest water quality. The efficiency of various storm waterand combined sewer overflow (CSO) treatment processesis given in Table 9.23.1 (Lager et al. 1977).

Biological ProcessesThe biological processes used for point sources are diffi-cult to implement for stormwater discharges, which havelow biochemical oxygen demand (BOD) nonpoint con-centrations. These processes perform poorly or not at allwhen treating flows with irregular quantity or quality. Itis very difficult to keep the biota alive between stormevents. Wet weather reduces low organic concentrations,and the biomass is sensitive to toxic substances often pres-ent in urban stormwater runoff.

Physical-Chemical ProcessesPhysical-chemical processes show promise in overcomingshock loadings. Chemical coagulants enhance the separa-tion of particles from liquid (see Chapter 7). Chemical ad-dition is also effective in removing phosphorous, metals,and some organic colloids.

Physical ProcessesSuccessfully demonstrated physical processes include fine-mesh screening, fine-mesh screening/high-rate filtration,

sedimentation, sand and peat-sand filtration, fine-meshscreening/dissolved-air flotation, and swirl separation.

SWIRL-FLOW REGULATOR-CONCENTRATOR

The dual purpose swirl-flow regulator-solids-concentratorhas shown a potential for simultaneous quality control(Field 1990). These devices have been applied to CSO;however, they can also be used for storm water runoff pol-lution control.

The swirl concentrator uses a swirl action to separateparticles from liquids (Figure 9.23.1). Flow from combinedsewers enters a diversion chamber and bar screen, remov-ing the debris. The swirl facility is automatically activatedwhen storm flows enter the lower portion of the circularchamber. Rotary motion causes liquids to follow a longspiral path, to be discharged from the chamber top througha downshaft. This overflow water can be disinfected anddischarged or stored for later treatment. Because a flowdeflector prevents chamber flow from completing its firstrevolution and merging with continuing inlet flow, thereis a gently swirling rotational movement.

The settleable solids entering the chamber are spreadover the full cross-section of the channel and settle quickly.Solids are entrained along the bottom around the cham-ber and are concentrated at the foul sewer outlet, wherethey are transported to the treatment plant.

The scum acts as a baffle, keeping floatables outside theoverflow weir and preventing these from overflowing into

the clean effluent. Floatables are directed by a floatable de-flector to a floatable trap. The floatable trap is connectedto a floatable storage area under the clear overflow weirplate. Floating material is drawn beneath the weir plate bythe vortex and dispersed around the downshaft. Floatingsolids are retained here until after a storm event, when thewater level recedes in the swirl chamber. As this occurs,trapped floatables are dropped and enter the foul seweroutlet, where they are transported to a sewage treatmentplant.

A partial list of U.S. installations with experience inswirl-flow regulator-concentrator use was presented byPisano (1989).

SAND FILTERS

Sand systems are usually off-line. A typical sand filtrationsystem is comprised of an inlet structure with a presettingbasin, a flow disperser, filtration media, an underdrain sys-tem, and a basin liner (Figure 9.23.2). For piped storm wa-ter systems, the inlet structure might be a manhole usinga weir to divert low flows into the filtration system. Foropen-channel conveyance systems, the inlet might be a weirconstructed within the flow path to divert low flows tothe filtration system, while allowing higher flows to by-pass the filtration system. Without a presettling basin, thefilter medium may quickly become plugged with large sed-iments. This basin may not be necessary, if the sand fil-tration basin is used in place of an API oil/water separa-tor, and if the contributing drainage area is small andcompletely impervious.

The primary pollutant removal mechanisms are filtra-tion and sedimentation. Particulate matter such as sedi-ments, oils and greases, and trace metals are removed byfiltration as stormwater percolates through the sand filter.Sedimentation removes large particles, and filtration re-moves silt and clay-size particles.

Over time, sediment eventually penetrates the filter me-dia surface, requiring replacement of the filter media.


TABLE 9.23.1 EFFICIENCY OF VARIOUS STORM-WATER AND CSO TREATMENT PROCESSES

Efficiency (%)

Process Suspended Solids BOD5 COD Total P TKNa

Physical—ChemicalSedimentation 20–60 50 34 20

without chemicals 38with chemicals 68 68 45

Vortex separation 40–60 25–60 50–60Screening

microstrainers 50–95 10–50 35 20 30rotary screens 20–35 1–30 15 12 10

Sand–peat filtersb 90 90 NA 70 50

Biologicalc

Contact stabilization 75–95 70–90 50 50Biodiscs 40–80 40–80 33Oxidation ponds 20–57 10–17 22–40 57Aerated lagoons 92 91Facultative lagoons 50 50–90

Source: Reprinted from J.A. Lager, W.G. Smith, W.G. Lynard, R.M. Finn, and E.J. Finnemore, 1977, Urban stormwater management and technology: updates andusers’ guide (U.S. Environmental Protection Agency (EPA), EPA Report 600–8–77–014, Municipal Environmental Research Laboratory, Office of Research andDevelopment, Washington, D.C.).

aTotal Kjeldahl nitrogen.bAfter Galli (1990), peat-sand filters are similar to biological anaerobic-aerobic slow filters. They are applicable for treatment of urban runoff.cBiological treatment is feasible only for CSOs.

FIG. 9.23.1 An isometric view of a swirl regulator-concentra-tor.

Inflow

Overflow

Foul Sewer

GA

B

EF

HI

J

D

C

K

A Inlet rampB Flow deflectorC Scum ringD Overflow weir and weir plateE SpoilersF Floatables trapG Foul sewer outletH Floor gutterI DownshaftJ Secondary overflow weirK Secondary gutter

Maintenance requirements can be intensive, dependingupon sediment concentrations in surface runoff. Fifty acresis recommended as the maximum contributing drainagearea for a sand filtration system (Schueler, Kumble andHeraty 1992).

ENHANCED FILTERS

Enhanced (peat-sand) filters use layers of peat, lime, and/ortopsoil, and may also use a grass cover (Figure 9.23.3) toremove particulate pollutants. To minimize clogging, bothsand and enhanced filters should be preceded by a solid-removing unit, such as a pond or a filter strip.

Peat-sand filters provide high phosphorous, BAD, ni-trogen and silt removal. Peat has a high removal affinityfor adsorbing and removing toxic compounds (Novotny1994), hence peat-containing filters are effective for re-moving priority pollutants.

COMPOST FILTERS

W&H Pacific conceived the idea of utilizing yard debriscompost as a treatment and filtration medium for storm-water runoff. This medium removes organic and inorganicpollutants through adsorption, filtration, and biologicalprocesses (ion exchange and bioremediation). TheCompost Storm Water Treatment System (CSFy) has beenconstructed at eight different sites throughout Oregon. Sixof the eight systems are enclosed facilities, located belowgrade, while the remaining two are open channel systemsretrofitted into existing swales. The technology is beingtested and field modified.

The filtering capacity of the medium removes sedimentsfrom the runoff. Ion exchange and adsorption removes oilsand greases, heavy metals, and non-dissolved nutrients.Following adsorption, organic material is further brokendown into carbon dioxide and water by microbial actionwithin the compost. Treated stormwater then passesthrough a 6 in to 8 in gravel layer underneath the filter-ing media, and is conveyed to a surface water body or toa storm drainage system by an underdrain system.

Prototype test results for nine events show good solidsremoval: 67% removal of COD, 40% removal of totalphosphorous, 67% removal of copper, and better than


FIG. 9.23.2 Conceptual sand filtration basin system. (Reprinted from the City of Austin, 1988.)

��

PresettlingBasin

Energy Dissipators

6 Mo-24 HrStorm

Stormwater Channel

Drop Inlet

To StormwaterDetention Basin

Channel Sloped toFacilitate SedimentTransport intoPresettling Basin

Perforated Riserwith Trash Rack

Underdrain Piping System

Sand Bed

Weir To AchieveUniform Discharge

Elevation A - A

StoneRiprap

Filtered Outflow

AA

��

Filtration Basin

Plan View

��

��

��

��

Grass Layer

Peat

Peat/Sand Mix

Fine-Medium Sand

Gravel Underdrain

30-4

5 cm

10 c

m50

-60

cm15

cm

105-

130

cm

FIG. 9.23.3 Peat sand filter for storm water treatment.(Reprinted from J. Galli, 1990, Peat sand filters: A proposedstorm water management practice for urban areas [Departmentof Environmental Programs, Metropolitan Washington Councilof Governments, Washington, D.C.].)

87% removal of zinc, aluminum and iron. The leaf com-post has very good cation exchange capacity. However,like sand and enhanced filters, operating life depends onthe frequency of preventive maintenance.

—Kent K. Mao

ReferencesField, R. 1990. Combined sewer overflows: Control and treatment. In

Control and treatment of combined-sewer overflows. Moffa, P.E. ed.New York, N.Y.: Van Nostrand Reinhold.

Lager, J.A., W.G. Smith, W.G. Lynard, R.M. Finn, and E.J. Finnemore.1977. Urban stormwater management and technology: Updates andusers’ guide. U.S. Environmental Protection Agency (EPA). EPAReport 600–8–77–014. Municipal Environmental ResearchLaboratory, Office of Research and Development.

Novotny, V., and H. Olem. 1994. Water quality: Prevention, identifica-tion, and management of diffuse pollution. New York, N.Y.: VanNostrand Reinhold.

Pisano, W.C. 1989. Recent United States experience with designs andnew German technology. In Design of urban runoff quality controls.L.A. Roesner, B. Urbonas, and M.B. Sonnen, eds. American Societyof Civil Engineers (ASCE). New York, N.Y.

Schueler, T.R., P.A. Kumble, and M.A. Heraty. 1992. A current assess-ment of urban best management practices. Techniques for reducingnon-point source pollution in the coastal zone. Technical guidancemanual. Metropolitan Washington Council of Government. Officeof Wetlands, Oceans, and Watersheds. Washington, D.C.

BibliographyAravin, V.L., and S.N. Numerov. 1965. Theory of fluid flow in unde-

formable porous media. New York: Daniel Davey.Bear, J. 1972. Dynamics of fluids in porous media. Elsevier, Amsterdam.Beasley, D.B. 1976. Simulation of the environmental impact of land use

on water quality. In Best management practices for non-point sourcepollution control. U.S. Environmental Protection Agency (EPA). EPAReport 905-9-76-005. Washington, D.C.

Bennett, G.D. 1976. Introduction to groundwater hydraulics, Techniquesof Water Resources Investigations, Chap. B2, Book 3, U.S. GeologicalSurvey, Washington, D.C.

Bowen, R. 1980. Groundwater. Barking, Essex, England: Applied SciencePublishers Ltd.

Cooper, H.H. 1966. The equation of groundwater flow in fixed and de-forming coordinates. J. Geophys. Res. 71, no. 20:4785–4790.

Council on Environmental Quality. 1980. Environmental quality—1978:The ninth annual report of the council of environmental quality.Washington, D.C.: U.S. Government Printing Office.

Crawford, N.H., and R.K. Linsley. 1966. Digital simulation in hydrol-ogy: The Stanford Model IV. Technical Report No. 39. StanfordUniversity, Department of Civil Engineering. Palo Alto, Calif.

Davis, S.N., and R.J.M. DeWiest. 1966. Hydrogeology. New York: JohnWiley & Sons, Inc.

DeVries, J.J. 1975. Groundwater hydraulics. Aqua-Vu, Ser. A, no. 6,Communications of the Institute of Earth Sciences. Amsterdam: FreeReformed University.

DeWiest, R.J.M. 1965. Geohydrology. New York: John Wiley & Sons,Inc.

DeWiest, R.J.M., ed. 1969. Flow through porous media. New York:Academic Press, Inc.

Hantush, M.S. 1964. Hydraulics of wells. In Advances in hydroscience.Vol. 1, edited by V.T. Chow. New York: Academic Press, Inc.

Harr, M.E. 1962. Groundwater and seepage. New York: McGraw-HillBook Company.

Heaney, J.P., and W.C. Huber. 1972. Storm water management model;refinements, testing and decision making. University of Florida,Department of Environmental Science. Gainesville, Fla.

Heath, R.C. 1983. Basic groundwater hydrology. Water Supply Paper2220, U.S. Geological Survey. Washington, D.C.

Hubbert, M.K. 1940. The theory of groundwater motion. J. Geol.48:785–944.

Jacob, C.E. Flow of groundwater. In Engineering hydraulics, edited byH. Rouse. John Wiley.

Javandel, I., C. Doughty, and C.F. Tsang. 1984. Groundwater transport:Handbook of mathematical models. Washington, D.C.: AmericanGeophysical Union.

Lohman, S.W. 1972. Groundwater hydraulics. Professional paper 708,U.S. Geological Survey. Washington, D.C.

Marino, M.A., and J.N. Luthin. 1982. Seepage and groundwater.Developments in Water Science Series no. 13. New York: ElsevierScience Publishing Co., Inc.

McWhorter, D.B., and D.K. Sunada. 1977. Groundwater hydrology andhydraulics. Fort Collins, Colo.: Water Resources Publications.

Meinzer, O.E. [1923] 1960. Outline of groundwater hydrology. WaterSupply Paper 494, U.S. Geological Survey. Washington, D.C.

Novotny, V., and G. Chesters. 1981. Handbook of nonpoint pollution:Sources and management. New York, N.Y.: Van Nostrand Reinhold.

Novotny, V., R. Imhoff, M. Olthoff, and P.A. Krenkel. 1989. KarlImhoff’s handbook of urban drainage and wastewater disposal. JohnWiley.

Novotny, V., and H. Olem. 1994. Water quality: Prevention, identifica-tion, and management of diffuse pollution. New York, N.Y.: VanNostrand Reinhold.

Petersen, D.F. 1957. Hydraulics of wells. Trans. Am. Soc. Cir. Eng.122:502–517.

Pinneker, E.V., ed. 1983. General hydrogeology. Cambridge: CambridgeUniversity Press.

Polubarinova-Kochina, P.Y. 1962. Theory of groundwater movement.Translated from Russian by R.J.M. DeWiest. Princeton, N.J.:Princeton University Press.

Todd, D.K. 1964. Groundwater. In Handbook of applied hydrology,edited by V.T. Chow. New York: McGraw-Hill Book Company.

United Nations Educational, Scientific and Cultural Organization (UN-ESCO). 1987. Manual on drainage in urbanized areas. Vol. 1,Planning and design of drainage systems. Paris, France: UNESCOPress.

U.S. Bureau of Reclamation. 1960. Studies of groundwater movement.Technical Memorandum 657. Denver, Colo.: U.S. Dept. of Interior.

U.S. Department of Interior, Water and Water Resources Service. [1977]1981. Groundwater manual. Washington, D.C.: U.S. GovernmentPrinting Office.

U.S. Soil Conservation Service (SCS). 1975. Procedure for computingstreet and rill erosion on project areas. SCS technical release no. 51.Washington, D.C.

U.S. Soil Conservation Service (SCS). 1975. Urban hydrology for smallwatersheds. SCS technical release no. 55. Washington, D.C.

Whipple, W., Jr. et al. 1983. Storm water management in urbanizing ar-eas. Englewood Cliffs, N.J.: Prentice Hall.


Documents

Chapter 9. Groundwater & SurfaceWater Pollution