Chapter 8

Groundwater—Multidimensional Flow and

Applications
SOLUTIONS TO QUESTIONS AND PRACTICE PROBLEMS

Section 8.2 Flow Net Solutions for Two-Dimensional Flow

8.1 One of the factors in Equation 8.27 is NF , yet when drawing a flow net we assume a
value for this parameter. How can this formula produce correct results when one of the
factors is assumed? (i.e., would assuming a higher NF produce a higher computed value
of Q?)

Solution
Equation 8.26 is expressed in terms of the ratio NF/ND, and it is this ratio that is a constant
for a given cross-section. If we had assumed a different value for NF and followed the
rules for drawing flow nets (constant b/a, etc.), the value of ND also would change
proportionately, thus maintaining the same NF/ND ratio and the same computed Q. The
advantage of using larger NF values is that they produce more precise flow nets, and thus
more precise NF/ND ratios. However, such flow nets also require more time to draw.

8.2 The flow net in Figure 8.38 is incorrect. Explain why.

Figure 8.38 Trial flow net for Problem 8.2. Note: This flow net is not drawn correctly
and should not be used as an example!
8-1

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-2 Groundwater—Multidimensional Flow and Applications Chap. 8

Solution
This flow net has many problems. The major ones are shown in the following figure.

8.3 Compute the total flow rate under the dam shown in Figure 8.2 under the following
conditions: Δh = 15 m, the soil beneath the dam has a hydraulic conductivity of 3×10-3
cm/s, and the length of the dam into the page is 80 m.

Figure 8.2 A sample flow net of seepage beneath a concrete dam

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-3

Solution
From Figure 8.2:

b/a=1

NF = 4

ND = 10

cm ⎛ m ⎞⎛ 86,400 s ⎞
k = 3 × 10 −3 ⎜ ⎟⎜ ⎟ = 2.59 m/day
s ⎝ 100 cm ⎠⎜⎝ day ⎟⎠

⎛ N ⎞⎛ b ⎞
Q = kLΔh⎜⎜ F ⎟⎟⎜ ⎟
⎝ N D ⎠⎝ a ⎠
⎛4⎞
= (2.59 m/day)(80 m )(15 m )⎜ ⎟(1) = 1244 m 3 /day
⎝ 10 ⎠

8.4 Redraw the cross-section in Figure 8.39 to a scale of 1:500 (1 cm = 5 m), then draw a
flow net that describes the seepage below this 150 m long concrete dam. Finally,
compute the flow rate through this soil, expressed in liters per second, and the pore water
pressure at Point A, which is at elevation 122.0 m.

Figure 8.39 Cross-section of dam for Problems 8.4 and 8.5. el. = elevation

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-4 Groundwater—Multidimensional Flow and Applications Chap. 8

Solution

Based on flow net:

NF=4

ND=10.4 note: the last equipotential drop on the right side is

approximated 0.4×a full drop

NF b
Q = klΔh
ND a
⎛ 1m ⎞ ⎛ 4 ⎞
= 8x10 −3 cm/s⎜ ⎟(150 m )(134.2 m − 126.1 m ) ⎜ ⎟ (1) = 0.0374 m /s
3

⎝ 100 cm ⎠ ⎝ 10.4 ⎠
3
⎛ 1x10 2 cm ⎞ ⎛ 1 liter ⎞
= 0.0374⎜⎜ ⎟⎟ ⎜ 3 ⎟
= 37 liter/s
⎝ m ⎠ ⎝ 1000 cm ⎠

Point A is 3.9 equipotential drops from the reservoir, and there are a total of 10.4
equipotential drops in the flow net. Therefore the total head at point A can be computed
as

⎛ 3.6 ⎞
h = 134.2 − (134.2 − 126.1) ⎜ ⎟ = 129.7 m
⎝ 10.4 ⎠

Scaling from the drawing, the elevation at A is 122 m. Therefore the pressure head and
pore water pressure at A are computed as

h p = h − hz = 129.7 − 122 = 7.7 m

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-5

( )
u = γ w h p = 9.81 kn/m 3 (7.7 m ) = 75 kPa

8.5 Using the flow net from Problem 8.4, develop a plot of seepage flow rate vs. the water
elevation in the reservoir. Consider reservoir elevations between 126.1 m and 135.0 m.

Solution

8.6 The earth dam shown in Figure 8.40 is to be built on a gravelly sand with silt and cobbles.
This dam will extend a distance of 850 ft perpendicular to the cross-section. To reduce
the flow rate through these soils, a concrete cutoff wall will be built as shown. Redraw
this cross-section to a scale of 1 in = 100 ft, draw a flow net, and compute Q. Then,
identify the area in the flow net that has the greatest hydraulic gradient.

Figure 8.40 Cross-section for Problem 8.6. el. = elevation

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-6 Groundwater—Multidimensional Flow and Applications Chap. 8

Solution
Completed flow net

Based on flow net:

NF=4

ND=19.8 note: the last equipotential drop on the right side is

approximated 0.8×a full drop

NF b
Q = klΔh
ND a
⎛ 1 ft ⎞ ⎛ 4 ⎞
= 6 x 10 −2 cm/s⎜ ⎟ (850 ft )(701.5 ft − 620.0 ft ) ⎜ ⎟ (1)
⎝ 30.5 cm ⎠ ⎝ 19.8 ⎠
= 28 ft 3 /s

The greatest hydraulic gradient occurs at the tip of the cutoff wall.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-7

8.7 A proposed twenty-story office building with three levels of underground parking will be
supported on a concrete mat foundation, as shown in Figure 8.41. The bottom of this mat
will be 40 ft below the street, and its plan dimensions will be 200 ft × 150 ft. The
groundwater table is currently at a depth of 25 ft below the ground surface, but could rise
to only 13 ft below the ground surface during the life of the building. Compute the total
hydrostatic uplift force to be used in the design.

Figure 8.41 Proposed underground parking area for Problem 8.7.

Solution
The most critical case would be when the groundwater table is 13 ft below the ground
surface, so the design should be based on this condition.

( )
u = γ w z w = 62.4 lb/ft 3 (40 − 13) ft = 1685 lb/ft 2

( ) ⎛ 1k ⎞
Puplift = uA = 1685 lb/ft 2 (200 x 150 )ft 2 ⎜ ⎟ = 50,600 k
⎝ 1000 lb ⎠

8.8 Compute the uplift forces acting on the dam in Figure 8.2 using the data in Problem 8.3.
Draw a diagram of the uplift pressure acting on the dam and compute the total uplift force.

Solution

Using the information from 8.3:

Δh = 15m

ND = 10

Δh 15
= = 1.5 m/equipotential drop
N D 10

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-8 Groundwater—Multidimensional Flow and Applications Chap. 8

Taking datum to be the lower left corner of the dam, point A shown below, and setting
Δh = 15m:
• The width of the dam is 32 m
• The height of the reservoir above point A is 20.5m
• The tributary area where the ith equipotential intersects the base of the dam is
Δxi × 80 m.

The pressures and forces along the base of the dam are then computed as:
Upliift
Equipotential u Δxi
Δh (m) Force
drop (kPa) (m)
(kN)
3 20.5 – 2(1.5) = 17.5 172 4.1 56,416
4 20.5 – 3(1.5) = 16.0 157 4.7 59,032
5 20.5 – 4(1.5) = 14.5 142 4.8 54,528
6 20.5 – 5(1.5) = 13 128 4.9 50,176
7 20.5 – 6(1.5) = 11.5 113 4.8 43,392
8 20.5 – 7(1.5) = 10.0 98 4.7 36,848
9 20.5 – 8(1.5) = 8.5 83 4.1 27,224
Total 327,616

0
50
100
150
200
Uplift Pressure (kPa)

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-9

8.9 The sheet pile in Example 8.5 was located near the upstream end of the spillway. Would
the total hydrostatic uplift force acting on the structure change if the sheet pile was
located near the downstream end? Explain. Which position would be best? Why?

Solution
Since the groundwater must pass through a narrow opening beneath the sheet pile, there
is extra head loss in this vicinity. Therefore, the uplift pressure immediately downstream
of the sheet pile is less than that immediately upstream, as shown in the plot that
accompanies the solution to Example 8.5. If the sheet pile was moved further
downstream, this jog in the uplift pressure diagram also would move downstream. Since
the total uplift force is the area created by this diagram, it would increase if the sheet pile
was moved further downstream.
If one of the objectives of the sheet pile is to reduce the uplift force, then a
position near the upstream end would be best because this position puts more of the dam
downstream of the jog in uplift pressure.

Section 8.3 Numerical Modeling of Two-Dimensional Flow

For problems 8.10 and 8.11, see Appendix D for guidance on finite difference solutions
to flow problems.

8.10 Create a finite difference model for the sheet pile system shown in Figure 8.4 using a
commercial spreadsheet program. Plot the equipotential lines using you program and
sketch in the flow lines. Compare your solution to that shown in Figure 8.4. If the
hydraulic conductivity of the soil in the aquifer is 2×10-4 cm/s, what is the total flow rate
per meter of wall length into the page?

Solution
An example finite difference solution is provided in the files
• Problem_08-10.xlsx
• Problem_08-10_solved.xlsx

8.11 Create a finite difference model for the spillway with cutoff wall shown in Figure 8.9
using a commercial spreadsheet program. Plot the equipotential lines within the aquifer
as well as the uplift forces on the base of the spillway. Compare your results to those
shown in Example 8.5.

Solution
An example finite difference solution is provided in the files
• Problem_08-11.xlsx
• Problem_08-11_solved.xlsx

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-10 Groundwater—Multidimensional Flow and Applications Chap. 8

Section 8.4 Two and Three Dimensional Flow to Wells

8.12 The proposed well shown in Figure 8.42 will be used to supply a municipal water system.
Compute its pumping capacity with the groundwater level as shown, and express your
answer in gallons per minute.

Figure 8.42 Proposed well for Problems 8.12 and 8.13.

Solution
We have no information about a recharge source so we must estimate r0 using Equation
8.34

( )
k = 5 x 10 −3 ft/s (30.5 cm/ft ) = 0.15 cm/s
r0 = 300(h0 − hw ) k = 300 (40 − 13) 0.15 cm/s = 3140 ft

The groundwater table is within the aquiclude, but will be drawn down into the aquifer.
Therefore, thus is mixed aquifer, and solved using Equation 8.38.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-11

πk (2 H a h0 − H a2 − hw2 )
Q=
⎛r ⎞
ln⎜⎜ 0 ⎟⎟
⎝ rw ⎠

Q=
( )[
π 5x10 −3 ft/s 2(15)(40) − (15) − (13)
2 2
]
⎛ 3140 ⎞
ln⎜ ⎟
⎝ 8/12 ⎠
= 1.50 ft 3 /s
= 673 gal/min

8.13 Using the well shown in Figure 8.42, if the drawdown is limited such that the level of
water in the well is maintained at the top of the gravelly sand layer, what pumping rate is
possible? Express your answer in gallons per minute.

Solution
We have no information about a recharge source so we must estimate r0 using Equation
8.34

( )
k = 5 x 10 −3 ft/s (30.5 cm/ft ) = 0.15 cm/s
r0 = 300(h0 − hw ) k = 300 (40 − 15) 0.15 cm/s = 2900 ft

If the water lever in the well is maintained at the top of the aquifer, pumping from the
aquifer will remain as a confined aquifer and Equation 8.31 will apply.

2πkH a (h0 − hw )
Q=
⎛r ⎞
ln⎜⎜ 0 ⎟⎟
⎝ rw ⎠

=
( )
2π 5 × 10 −3 (15)(40 − 15)
⎛ 2900 ⎞
ln⎜ ⎟
⎝ 8 / 12 ⎠
= 1.41 ft 3 /s
= 633 gal/min

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-12 Groundwater—Multidimensional Flow and Applications Chap. 8

8.14 After reaching steady-state conditions, the test well shown in Figure 8.43 is producing a
flow rate of 17 l/s. The aquifer is an alluvial soil with interbedded medium-to-coarse
sand and silty sand.

Figure 8.43 Cross-section for Problems 8.14 and 8.15

The water depths in the observation wells are as follows:

Water Depth From Ground Surface (m)
Well
Before Pumping During Pumping
Pumping 16.9 26.0
Observation A 16.9 23.5
Observation B 16.9 18.1
Observation C 16.9 16.9

Using the best available data, compute the hydraulic conductivity of the soil in the
aquifer. Is the computed k value reasonable? Explain why or why not.

Solution
The groundwater table is completely within the aquifer, so this is an unconfined aquifer
and the hydraulic conductivity may be computed using Equation 8.39.
Well C is beyond the radius of influence because the water level during pumping is the
same as that before pumping. Therefore, it cannot be used to compute k. The analysis
must be based on wells A and B.
Well A: r = (3)(100) = 300 cm, h = (19 + 11 − 23.5)(100) = 650 cm
Well B: r = (3 + 27)(100) = 3000 cm, h = (19 + 11 − 18.1)(100) = 1190 cm

( )
Q = (17 liters/s ) 1000 cm 3 /l = 17,000 cm 3 /s

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-13

k= =
( )
Q ln (r1 / r2 ) 17,000 cm 3 /s ln (3000/300 )
= 1.2 x 10 −2 cm/s
(
π h1 − h2
2 2
) (
π 1190 − 650
2 2
)
This is an alluvial soil which has horizontal stratifications, The flow to the well is
primarily horizontal, so the measured k will be very close to the horizontal hydraulic
conductivity, kh. According to the discussion of anisotropic soils on page 284-286, the
value of kh will be close to that of the medium-to-coarse sand layers. According to Table
7.1, clean coarse sand typically has k=10-2 to 1 cm/s and fine sand typically has k=10-3 to
10-1 cm/s, so medium-to-coarse sand typically has k≈10-2 to 10-1 cm/s. The measured
value is in the range. Therefore, the measured value seems reasonable.

8.15 For the well shown in Figure 8.43 assume the original depth from the ground surface to
the phreatic surface is 16.9 m and the hydraulic conductivity of the soil in the aquifer is
10-2 cm/s. What is the maximum pumping rate, Q, such that the distance from the ground
surface to the phreatic surface at observation well B is no greater than 18 m? Assume
rw = 0.06 m

Solution
Calculate the distance from the top of the Aquiclude:

h0 = 11 + 19 − 16.9 = 13.1 m
h B = 11 + 19 − 18 = 12 m

We have no information about a recharge source so we must estimate r0 using Equation

8.34. To do this we will have to estimate hw and then check our assumption later.
Assume hw = 5 m

r0 = 300(h0 − hw ) k = 300 (13.1 − 5) 0.02 cm/s = 344 m

Since this is an unconfined aquifer Equation 8.35 will apply.

πk (h12 − h22 )
Q=
ln (r1 / r2 )
π (2 × 10 −6 m/s )(13.12 − 12 2 ) m 2
=
ln (344 / 0.06)
= 2.00 × 10 −5 m 3 /s
= 1.2 l/min

This is a very low flow rate. If the restriction at Well B existed, it would not be possible
to use this well as a supply well.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-14 Groundwater—Multidimensional Flow and Applications Chap. 8

Section 8.5 Groundwater Control

8.16 A construction site needs to be predrained in order to allow for the excavation shown in
Figure 8.44. This will require dropping the groundwater table 5 m below its current
location. From observations of the drawdown from other wells in the area, the radius of
influence, r0, is estimated to be 825 m. Compute the pumping rate required to lower the
groundwater table using the well array shown.

Figure 8.44 Plan and elevation drawing for Problem 8.16.

Solution
Compute the equivalent radius of the well array using Equation 8.42

r'w =
ab
=
(50)(40) = 25.23 m
π π

Compute the flow rate using the Equation 8.35 for unconfined aquifers.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-15

πk (h02 − hw2 )
Q=
ln (r0 / rw )
π (0.003 m/s )(302 − 252 ) m 2
=
ln (825 / 25.23)
= 0.74 m3 /s
= 740 l/s

740 l/s
Q= = 53 l/s for each pump
14

Section 8.6 Contaminant Control and Remediation

8.17 A contaminate soil site is to be capped and provided with a slurry wall barrier similar to
the situation shown in Figure 8.29. A bioremediation process will be used to clean up the
contaminated soil. This bioremediation process is expected to take 15 years. The slurry
wall must be designed to prevent migration of the contaminants off of the site during the
15 year bioremediation process. If the regional hydraulic gradient is 0.0043 from left to
right in Figure 8.29 and the clay in the slurry wall has a hydraulic conductivity (k) of
2×10-5 cm/s with an effective porosity (ne) of 65%, how thick does the slurry have to be
to contain the contaminants for 15 years?

Solution
Use Equation7.13 to compute the seepage velocity

ki
vs =
ne

=
(2 × 10 )(0.0043)
−5

0.65
= 1.32 × 10 −7 cm/s
= 4.2 cm/yr

Thickness = (4.2 cm/yr)(15 yr) = 63 cm = 0.63 m

Section 8.7 Soil migration and Filtration

8.18 A proposed levee is to be built using the soil described in Figure 8.45. This levee will
include a toe drain similar to the one in Figure 8.34 to control the groundwater flow, and
thus maintain adequate stability. This toe drain must be coarse enough to adequately
collect and transmit the water, yet fine enough to provide sufficient filtration to prevent
migration of the main levee soils. There will be no separate filter layer; the drain must
act as the filter.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-16 Groundwater—Multidimensional Flow and Applications Chap. 8

To maintain sufficient hydraulic conductivity, the drain must have no more than 3%
passing the #200 sieve. In addition, to provide adequate filtration, it must meet the
criteria described in this chapter. Determine the acceptable range of grain-size
distribution for this material and plot it on a grain-size distribution curve.

Figure 8.45 Particle size distribution of proposed levee soils for Problems 8.18 and 8.19

Solution
Per gain size distribution curve, 61% of the levee soil passes the #200 sieve. Therefore, it
is in soil group 2, per Table 8.2, and D15 of the filter/drain should be ≤0.7 mm to prevent
migration of the levee material. To ensure filter has sufficient high hydraulic
conductivity D15 of the filter/drain should be > 4 × D15 of the levee material or
> 4 × 0.016 = 0.064. Also, the problem statement requires the filter/drain soil to have no
more than 3% passing the #200 sieve. Finally, the grain size distribution curve of the
filter/drain should be roughly parallel to that of the levee soils. Based on all of these
requirements, the grain size distribution of the filter/drain should be as shown below.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-17

8.19 An alternative design for the levee in Problem 8.18 uses a perforated pipe drain instead of
the toe drain. This perforated pipe drain would be surrounded with gravel and wrapped
with a filter fabric, similar to the one shown in Figure 8.37. The design flow rate into the
drain is 80 gal/min per square foot, and the maximum acceptable head loss is 0.10 ft.
Select an appropriate fabric from Table 8.2.

Solution
Per figure 8.40, (D85)soil = 0.2 mm

O95 < (2 or 3)(D85 )soil

< (2 or 3)(0.2 )
< 0.4 − 0.6 mm

Q ⎛ 80 gal/min ⎞⎛⎜ 0.1337 ft 3 ⎞⎟⎛ 1 min ⎞ 0.18 ft 3 /s

=⎜ ⎟⎜ =
A ⎝ ft 2 ⎠⎝ gal ⎟⎜⎝ 60 s ⎟⎠ ft 2
⎠

⎛Q⎞
⎜ ⎟
A 0.18 ft 3 /s/ft
Minimum required ψ = ⎝ ⎠ = = 1.8 s −1
Δh 0.10 ft

Per Table 8.2, can use TG600, TG550, or TG500 fabric.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-18 Groundwater—Multidimensional Flow and Applications Chap. 8

Comprehensive

8.20 Several years ago a state highway department built a highway across a shallow lake by
placing the clayey fill shown in Figure 8.46. The top of this fill is above the high water
level, thus protecting the highway from flooding.

It is now necessary to install a buried pipeline beneath the roadway. To install this pipe,
the contractor plans to make a temporary excavation using steel sheet piles as shown.
The contractor plans to use sump pumps at 50 ft intervals to maintain the water level at
the bottom of the excavation. Once the pipe has been installed, the excavation will be
backfilled and the pumps and sheet piles removed. You are to perform the following
tasks in connection with this project:
(a) Recognizing that the proposed cross-section is symmetrical, redraw half of it to a
scale of 1 in = 10 ft and construct a flow net.
(b) Determine where the largest hydraulic gradient occurs and mark this spot on the
cross-section.
(c) Using the flow net, compute the minimum required capacity for each pump,
expressed in gallons per minute.
(d) Describe two methods of reducing the flow rate into the excavation (and thus the
required pump size). Explain how each of them would reduce Q.

Solution
(a) The flow net for half of the cross-section is shown below.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-19

(b) The largest hydraulic gradient occurs where the equipotential lines are closest
together, which is at the tip of the sheet pile, as shown in the flow net.
(c) The flow rate is computed as

NF b
Q = 2kLΔh
ND a

= (2)(5 x 10 −2 cm/s)⎜
⎛ 1 ft ⎞ ⎛ 4 ⎞ ⎛ 449 gal/min ⎞
⎟(50 ft )(32.7 ft − 21.5 ft )⎜ ⎟(1)⎜ ⎟
⎝ 30.5 cm ⎠ ⎝ 16 ⎠ ⎝ ft 3 /s ⎠
= 206 gal/min

(d) The flow rate could be reduced by any of the following methods:
• Extend the clayey fill further into the lake (increases ND)
• Drive the sheet piles deeper (increases ND)
• Raise the invert elevation of the pipe, thus requiring a shallower excavation
(increases ND and decreases Δh)
• Inject chemical grout into the soil (decreases k)

8.21 Using the flow net from Problem 8.4, compute the uplift pressures acting on the bottom
of the dam in Figure 8.39 and develop a plot similar to the one shown in Figure 8.9.

Solution

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-20 Groundwater—Multidimensional Flow and Applications Chap. 8

Δh = 134.2 − 126.1 = 8.1 m

N D = 10.4

Compute the total head for each equipotential line based on the total head in the reservoir
and the number of equipotential drops from the reservoir to that equipontential line. Then,
use this total head to compute the pressure head and the pore water pressure along the
bottom of the dam.

Equipotential
Line Number h (m) hp (m) u (kPa)
2 134.2 - 2/10.4 (8.1) = 132.6 132.6 - 122.0 = 10.6 (9.8)(10.6) = 104
3 134.2 - 3/10.4 (8.1) = 131.8 131.8 - 122.0 = 9.8 (9.8)(9.8) = 96
4 134.2 - 4/10.4 (8.1) = 131.1 131.1 - 122.0 = 9.1 (9.8)(9.1) = 89
5 134.2 - 5/10.4 (8.1) = 130.3 130.3 - 122.0 = 8.3 (9.8)(8.3) = 81
6 134.2 - 6/10.4 (8.1) = 129.3 129.5 - 122.0 = 7.5 (9.8)(7.5) = 73
7 134.2 - 7/10.4 (8.1) = 128.7 128.7 - 122.0 = 6.7 (9.8)(6.7) = 66
8 134.2 - 8/10.4 (8.1) = 128.0 128.0 - 122.0 = 6.0 (9.8)(6.0) = 59
9 134.2 - 9/10.4 (8.1) = 127.2 127.2 - 122.0 = 5.2 (9.8)(5.2) = 51

8.22 A 30 m wide, 40 m long, 8 m deep construction excavation needs to be made in a silty

clay (CL). The groundwater table is at a depth of 2 m. The sides of the excavation will
be sloped at an angle of about 1 horizontal to 1 vertical, and no sensitive structures or
other improvements are nearby. Suggest an appropriate method of construction
dewatering for this site, and explain the reason for your choice. Include statements of
any assumptions, if any.

Solution
The Hydraulic conductivity of CL is very low, so the Q entering the excavation will be
small. Therefore, an open pumping system is probably the best choice. However, this is
dependent on the temporary slopes being stable with a relatively high groundwater table.
If slope stability is a problem, it will be necessary to either use flatter slopes, or some
other method of dewatering (such as wellpoints located outside the excavation) that will
draw the groundwater table down to a lower elevation.

8.23 After the analysis described in Example 8.6 was completed, the well was installed to the
depth indicated in Figure 8.18. However, when the pump was installed, it produced a
flow rate of only 102 gal/min. Is the difference between this value and the computed
flow rate within the normal range of uncertainty for these kinds of analyses? Explain.
What portion of the analysis usually introduces the greatest error?

Solution
The predicted flow rate was 276 gal/min, but the actual flow rate was only 102 gal/min.
This represents a difference of 63%. The most likely source of this difference is the
hydraulic conductivity, k. The in-situ value of k could easily be 63% different from the
design value. Thus, this difference is not unusual, and is within the expected range of

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-21

uncertainty. If this flow rate is not acceptable, it may be possible to increase it by drilling
the well deeper and installing the pump at a lower elevation.

8.24 Compute Q for the cross-section in Figure 8.39 using the following hydraulic
conductivities for the silty sand: kx = 5 × 10-2 cm/s, kz = 4 × 10-3 cm/s.

Solution

Figure 8.39 Cross-section of dam for Problems 8.4 and 8.5. el. = elevation.

The anisotropic soil condition requires the use of a transformed cross-section. The
vertical dimensions need to be expanded by:

kx 5 x 10 −2
= = 3.54
kz 4 x 10 −3

The equivalent keq used in the computations is computed as

k ea = k x k z
= (5 x 10 )(4 x 10 ) = 1.4 x 10
−2 −3 -2
cm/s

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-22 Groundwater—Multidimensional Flow and Applications Chap. 8

The flow net for the transformed section is shown below.

And the flow rate is compute as

NF b
Q = kLΔh
ND a
3
⎛ 4 ⎞ ⎛ 1x10 cm ⎞ ⎛ 1 liter ⎞
2
⎛ 1m ⎞
−2
= 1.4 x 10 cm/s⎜ ⎟ (150 m )(134.2 m − 126.1 m )⎜ ⎟ ⎜(1)⎜ ⎟⎟ ⎜ 3 ⎟
⎝ 100 cm ⎠ ⎝6⎠ ⎝ m ⎠ ⎝ 1000 cm ⎠
= 113 liter/s

8.25 By observing the groundwater drawdown in the vicinity of a proposed well, an engineer
had determined the radius of influence, r0. This engineer then used Equation 8.34 to
compute k for the aquifer. Write a 200–300 word memo to this engineer, explaining why
this is not a good method of computing k, then suggest a better method.

Solution
MEMORANDUM
I have reviewed your computations dated _____ for the _____ well project. These
computations used the Sichart and Kyrieleis equation to compute the hydraulic
conductivity for the aquifer at this site based on the observed drawdown data.
Unfortunately, this is not an appropriate use of this equation, so the computed k value is
not very reliable.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-23

The Sichart and Kyrieleis equation is an empirical relationship intended to estimate the
radius of influence based on k. It is only an approximate relationship, but is suitable for
use computing flow rates from proposed wells. However, using it in reverse to compute k
is not very precise because:
• The actual radius of influence is difficult to determine, even when observation
well data is available. Thus, precision of the computed k value is no better than
the r0 value.
• This approach determines k based on only two parameters: r0 and the depth of
drawdown at the pumped wellpoint. Many other factors also influence k, but they
have not been considered.
A better method would be to use observation well data and flow rate data and compute k
using Equations 8.32, 8.33, or 8.34 in Coduto (1999). This approach produces more
precise values of k because:
• These are derived equations, which have proven to be reliable.
• These equations are based on observation well and flow rate data, which are more
reliable than the radius of influence value.
Please reanalyze the field data using the approach and recomputed the in-situ hydraulic
conductivity.

8.26 A below-ground swimming pool was built twenty years ago and, until recently, has been
performing satisfactorily. About ten years ago it became necessary to temporarily drain
the pool to clean out some algae. It was then refilled without any problems. However,
the pool recently developed some cracks in its concrete shell, so it became necessary to
drain it once again. Soon after it was drained, the pool rose out of the ground a distance
of about 1 m. Provide a possible explanation for this behavior, and a possible
explanation for why the pool did not rise out of the ground when it was drained the first
time.

Solution
The recent damage to this pool appears to be the result of hydrostatic uplift pressures.
The groundwater table has risen to an elevation above the bottom of the pool, and thus
has imparted uplift pressures onto the shell. Until recently, these pressures were more
than offset by the weight of the water in the pool and the weight of the pool shell.
However, when the pool was emptied, most of this resisting weight was lost, and the
hydrostatic uplift force pushed the shell out of the ground.
It also appears that this mode of failure did not occur when the pool was emptied ten
years ago because the groundwater table was not imparting hydrostatic pressures onto the
shell. It is quite possible that the subsequent rise in groundwater was due to leaks in the
pool shell.

8.27 An excavation for a foundation is to be made in the upper clay soil as shown in Figure
8.47. In order to prevent the base of the excavation from being blow out by the uplift
pressure of the confined aquifer below the clay layer, the construction contractor is
planning to predrain the excavation with the well array shown. The excavation is half a
mile from a large lake that recharges the aquifer below the clay.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-24 Groundwater—Multidimensional Flow and Applications Chap. 8

(a) Determine the depth to which the ground water table must be lowered to provide a
factor of safety of 3 against uplift of the clay layer assuming the weight of the clay is
the only resistance to uplift (i.e. ignore any strength in the clay)
(b) Assuming only three of the four wells are in operation, compute the pumping rate
required to achieve the needed drawdown.

Solution
(a) The uplift pressure applied at the base of the clay is

u = γ w z w = γ w (6 − Δh )

Where

Δh = the required drop in head below the current phreatic surface

The downward vertical stress at the base of the clay layer is

3′(γ Clay )

And the factor of safety, F, is

3(γ clay )
F=
γ w (6 − Δh )

For a factor of safety of 3 we compute

3(123)
3=
62.4(6 − Δh )

Δh =
(62.4)(6) − (123)
62.4
= 4 ft

(b) Since the spacing of wells is not close, we will use superposition to compute the
drawdown. Since only three wells will be in operation, we will require the drawdown
at the point of the fourth (inoperative) well, to be equal to 4 feet. This will ensure the
drawdown is greater than 4 feet everywhere else in the excavation.

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 Chap. 8 Groundwater—Multidimensional Flow and Applications 8-25

A 70 ft B

60 ft 70 ft

D C

Excavation

Plan View

Assume Well D is the inoperative well. The radius from each well to Well D is

rA = 70 ft
rB = 70 ft

rA = 70 ft
rC = 70 ft
rB = 2 (70 ) = 99 ft

The aquifer is confined so Equation 8.32 applies and, using superposition, the
drawdown from all three wells will be the sum of the drawdown from each well.
Assuming each well pumps at the same rate, Q

⎛ ⎛ r0 ⎞ ⎛ ⎞ ⎛ ⎞⎞
Δh =
Q⎜ ln⎜ ⎟ + ln⎜ r0 ⎟ + ln⎜ r0 ⎟ ⎟
⎜ ⎜r ⎟
2πkH a ⎜r ⎟ ⎜ r ⎟⎟
⎝ ⎝ A⎠ ⎝ B⎠ ⎝ C ⎠⎠
2ΔhπkH a
Q=
⎛ ⎛ r0 ⎞ ⎛ ⎞ ⎛ ⎞⎞
⎜ ln⎜ ⎟ + ln⎜ r0 ⎟ + ln⎜ r0 ⎟ ⎟
⎜ ⎜r ⎟ ⎜r ⎟ ⎜ r ⎟⎟
⎝ ⎝ A⎠ ⎝ B⎠ ⎝ C ⎠⎠

We have no information about a recharge source so we must estimate r0 using

Equation 8.34. Assume the drawdown at the wells is equal to 6 feet

( )
k = 2 × 10 −3 ft/s (30.5 cm/ft ) = 0.061 cm/s
r0 = 300(h0 − hw ) k = 300 (6) 0.061 cm/s = 444 ft

Say r0 = 450 ft

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.
 8-26 Groundwater—Multidimensional Flow and Applications Chap. 8

2ΔhπkH a
Q=
⎛ ⎛ r0 ⎞ ⎛ ⎞ ⎛ ⎞⎞
⎜ ln⎜ ⎟ + ln⎜ r0 ⎟ + ln⎜ r0 ⎟ ⎟
⎜ ⎜r ⎟ ⎜r ⎟ ⎜ r ⎟⎟
⎝ ⎝ A⎠ ⎝ B⎠ ⎝ C ⎠⎠

=
(
2π (4 ) 2 × 10 −3 (20 ))
⎛ ⎛ 450 ⎞ ⎛ 450 ⎞ ⎛ 450 ⎞ ⎞
⎜⎜ ln⎜ ⎟ + ln⎜ ⎟ + ln⎜ ⎟ ⎟⎟
⎝ ⎝ 70 ⎠ ⎝ 99 ⎠ ⎝ 70 ⎠ ⎠
= 0.19 ft 3 /s
= 85 gal/min for each well

© 2011 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. This publication is protected by Copyright and written permission should be obtained
from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any form or by any means, electronic, mechanical, photocopying,
recording, or likewise. For information regarding permission(s), write to: Rights and Permissions Department, Pearson Education, Inc., Upper Saddle River, NJ 07458.