Groundwater hydraulics

Week 11

1
Characterizing aquifer systems – Pumping test

Level meter Discharge measuring

Logger

Groundwater surface

measuring:
Under water Pump
pH
El. Conductivity
Pressure
Redox potential
transducer
Temp.

Filter

2
Pumping test

3
http://hydrogeologistswithoutborders.org/wordpress/1979-english/chapter-8/
Phases of a pumping test
Start of pump Stop of pump
time (t) [s]
drawdown s [m]

transient quasi
steady
state

drawdown recovery

4
Pumping test & aquifer properties

5
Procedure of pumping tests
• Testing wells
• Max drawdown
• Well characteristic (drawdown-Pump rate)
• Reach of drawdown
• Location of observation wells
• Determine aquifer parameters
• Hydraul. Conductivity/Transmissivity
• Dispersivity
• Spec. Storage/Storativity
• Heterogeneity of the aquifer
• Hydrochemistry characteristics

6
 Testing wells (multi-phase testing)
time (t) [s]
drawdown s [m]

Identify wells max. capacity (m³/s)

Well characteristic
(drawdown vs. pump rate)
Pump rate Q [m³/s]

Longterm drawdown

time (t) [s]

7
 Well performance Specific Yield C 
Q 2 −1
C = m s 
(Spezifische Ergiebigkeit) s
1. Pump rate

2. Pump rate
Possible

3. Pump rate
Initial water level withdrawl (l/s)
Drawdown s [m]

4. Pump rate
Very high yield > 40
(Sehr ergiebig)
High yield 15 - 40
(ergiebig)
Low yield 5 - 15
Weniger oder wechselnd
ergiebig
discharge Q [m³/s] Hydrologischer Atlas von
Deutschland, BGR

8
Observation wells
• Use as many as possible
• Multiple-well tests are preferable
• Increase accuracy of storativity estimate
• Less sensitive to well construction, clogging, etc.
• Less sensitive to variations in pumping rate
• Analysis using distance-drawdown relationship possible
• Possible to detect aquifer anisotropy
• Preferably in logarithmic distance (5, 50, 500m instead 100, 200, 300)
• Also in aquifers above/below to detect possible leakage

9
Duration of pumping tests
• As long (at sustainable pumping rate) as possible
• Increase likelihood to identify boundaries, leakage etc.
• Observe delayed yield and other long-term effects
• Duration depends on aquifer properties
• Rule of thumb
• A couple hours: determine transmissivity & storativity
• Multiple days: detect leakage, boundaries, delayed yield
• Multiple weeks: effects on water quaility (e.g. salinity increase)

10
Optimal pumping rate
• As high as possible to stress aquifer
• Maintainable at a constant rate during full test
• Varyies preferably less than 5% during test
• Below maximum capacity of the pump
• Allow upward adjustments during tests
• Check type of pump (centrifugal, extended-shaft, submersible)
• well, application type, duration, expected drawdown, etc.

11
Additional concerns
• Check/Correct for nonpumping-induced water level changes, e.g. tide
• Determine optimal pumping rate at the beginning
• Short duration step-drawdown test (e.g. three steps of 1h each) with
increasing pumprates
• Start of pump is starting point for analysis
• Recovery data is very valuable!
• Less impacted by fluctuations of pumping rate
• Convey pumped water far away

12
 History of pumping test analysis methods
• 1856 DARCY
• 1863 – 1906 DUPUIT / THIEM Stationary Pumping
• 1935 – 1951 THEIS, COOPER & JACOB Instationary Pumping
• 1956 – 1979 HANTUSH, WALTON, BOULTON, Anisotrope, heterogeneous,
NEUMAN, STALLMAN, bounded aquifers
WARREN & ROOT
• 1979 GRINGARTEN et al. Drawdown in pumping well
• 1983 BOURDET et al. Well diagnosis
• Since Computer aided models and analysis

13
Dupuit 1863 – confined aquifer – Conditions:
• Aquifer:
• Homogeneous & isotrope
• Confined, no vertical intrusion or leakage
• Aquifer thickness M is constant
• Infinitely wide
• Aquifer top and bottom are horizontal
• Pumping well:
• Perfect
• Flow
• Stationary
• Horizontal pressure head prior to pumping
14
Dupuit 1863 – confined aquifer – Concept:

15
Source: P. Renard (2005)
Dupuit 1863 – confined aquifer – Formulas:
𝑄 𝜕ℎ
• Darcys law: = 𝑘𝑓 Assuming Q>0 besides symbolizing extraction!
𝐴 𝜕𝑥

• Assuming a cylinder around the well with radius r

𝐴 =2𝜋𝑟𝑀 𝑇 = 𝑀 𝑘𝑓
𝜕ℎ 𝜕ℎ 𝜕ℎ
𝑄 𝑟 =𝐴 𝑘𝑓 =2𝜋𝑟𝑀 𝑘𝑓 =2𝜋𝑟𝑇
𝜕𝑟 𝜕𝑟 𝜕𝑟

Q Discharge m³/s M Aquifer thickness m

A Area m² T Transmissivity m²/s
kf Hydraulic conductivity m/s h Hydraulic head m
r Radius of cylinder around well m x Spatial direction m 16
Dupuit 1863 – confined aquifer – Formulas:
• Switch from hydraulic head h(r) to drawdown s(r)
𝜕ℎ 𝜕𝑠
𝑄 𝑟 =2𝜋𝑟𝑇 𝑠 𝑟 = ℎ0 − ℎ(𝑟) 𝑄 𝑟 = −2 𝜋 𝑟 𝑇
𝜕𝑟 𝜕𝑟
• Solve for s(r) by integration w.r.t. r
−𝑄(𝑟) 𝜕𝑠 −𝑄
= 𝑠 𝑟 = ln 𝑟 + 𝐶
2𝜋𝑟𝑇 𝜕𝑟 2𝜋𝑇

• Determine the constant C through boundary conditions

17
Dupuit 1863 – confined aquifer – Formulas:
• Dupuits assumption: Radius R at which h is not affected by pumping
ℎ 𝑅 = ℎ0 s 𝑅 =0
−𝑄 𝑄
𝑠 𝑅 = ln 𝑅 + 𝐶 = 0 𝐶= ln 𝑅
2𝜋𝑇 2𝜋𝑇
𝑄
• Inser C into the solution 𝑠 𝑟 = (ln 𝑅 − ln(𝑟))
2𝜋𝑇

𝑄 𝑅 𝑄 𝑅
𝑠 𝑟 = ln 𝑇= ln
2𝜋𝑇 𝑟 2𝜋𝑠(𝑟) 𝑟
18
Approximation of depression cone radius
• Sichardt (1928) R = 3000  s  k f

• Kusakin (1935) R = 575  s  k f  M

• If M > 27m: Kusakin > Sichardt

• Purely empirical formulas
s Drawdown in the well m
M Aquifer thickness m
kf Hydraulic conductivity m/s
R Radius of cone m 19
Calculation exercise
Based on exploration results in the downstream flow of a former
chemical factory it was determined that no significant contamination
can be found in the aquifer below the former operating site. The
contamination has shifted together with the natural groundwater flow
in eastern direction. A further shift of the contamination from the
unsaturated zone into the groundwater can be excluded due to surface
sealing (=confined aquifer). With the help of an installed protection
well (6-inch) a decontamination of the contamination plume must be
accomplished. The groundwater level may be lowered by a maximum
of one meter. The kf-value is 8,5 · 10-4 m/s. Determine the maximum
abstraction rate. What is the diameter of cone of depression?
20
Calculation exercise ctd

21
Solution

22
Thiem 1906 – Confined aquifer – Conditions:
• Aquifer:
• Homogeneous & isotrope
• Confined, no vertical intrusion or leakage
• Aquifer thickness M is constant
• Infinitely wide
• Aquifer top and bottom are horizontal
• Pumping well:
• Perfect
• Flow
• Stationary
• Constant pressure head prior to pumping
23
Thiem 1906 – Confined aquifer – Formulas:
r2
r1
• Improve estimation of C using
two observation wells s2 s1

s 𝑟1 = 𝑠1 s 𝑟2 = 𝑠2
h2 h1 H
−𝑄 M
𝑠 𝑟1 = ln 𝑟1 + 𝐶
2𝜋𝑇
−𝑄
𝑠 𝑟2 = ln 𝑟2 + 𝐶
2𝜋𝑇
−𝑄 𝑄
𝑠 𝑟1 − 𝑠 𝑟2 = ln 𝑟1 + 𝐶 + ln 𝑟2 − 𝐶
2𝜋𝑇 2𝜋𝑇 24
Thiem 1906 – Confined aquifer – Formulas:
−𝑄 𝑄
𝑠 𝑟1 − 𝑠 𝑟2 = ln 𝑟1 + 𝐶 + ln 𝑟2 − 𝐶
2𝜋𝑇 2𝜋𝑇
𝑄
𝑠 𝑟1 − 𝑠 𝑟2 = (ln 𝑟2 − ln 𝑟1 )
2𝜋𝑇
𝑄 𝑟2
𝑠 𝑟1 − 𝑠 𝑟2 = ln
2𝜋𝑇 𝑟1
𝑄 𝑟2
𝑇= ln
2𝜋(𝑠 𝑟1 − 𝑠 𝑟2 ) 𝑟1
25
Exercise
A pumping test with constant pumping rate of 700 m³/d is conducted in a confined
aquifer of 17.82 m thickness. Once steady state in the drawdown was observed, the
drawdown in the closest observation well 1.20 m away from the pumping well was
1.44 m. In the second observation well, which is 6.00 m away, the drawdown was
4cm.

Determine the transmissivity of the aquifer with the method of Thiem.

Calculate the kf value

26
Solution (and exemplary points)
s2 = 0.04 m 1P
s1 = 1.44 m 1P
r2 = 6.00 m 1P
r1 = 1.20 m 1P
Q = 700 m³/day 1P
M = 17.82 m 1P
𝑄 𝑟2
𝑇= ln 1P
2𝜋(𝑠 𝑟1 − 𝑠 𝑟2 ) 𝑟1
T = 128 m²/day 1P
kf = T/M = 7.2 m/day 1P
27
Dupuit-Thiem – unconfined – Conditions:
• Aquifer:
• Homogeneous & isotrope
• Unconfined, no vertical intrusion or leakage For very small drawdowns
• Aquifer thickness M is constant (relative to aquifer thickness) the
equations for confined aquifers
• Infinitely wide can be used. Limit: 𝑠 ≈ 0.15 𝑀
• Aquifer bottom is horizontal
• Pumping well:
• Perfect
• Flow
• Stationary
• Constant pressure head prior to pumping
28
Dupuit-Thiem – unconfined – Formulas:
• Similar to confined aquifer:
𝐴 =2𝜋𝑟𝑀 𝑇 = 𝑀 𝑘𝑓
𝜕ℎ 𝜕ℎ 𝜕ℎ
𝑄 𝑟 = 𝐴 𝑘𝑓 = 2 𝜋 𝑟 𝑀 𝑘𝑓 =2𝜋𝑟𝑇
𝜕𝑟 𝜕𝑟 𝜕𝑟

• BUT NOW: 𝑀=ℎ

𝜕ℎ 𝑄 𝜕ℎ
𝑄 = 2 𝜋 𝑟 ℎ 𝑘𝑓 =ℎ
𝜕𝑟 2𝜋𝑟𝑘𝑓 𝜕𝑟

29
Dupuit-Thiem – unconfined – Formulas:
• Integrate w.r.t. r
𝑄 𝜕ℎ 𝑄 1 2
=ℎ ln 𝑟 + 𝐶 = ℎ
2𝜋𝑟𝑘𝑓 𝜕𝑟 2𝜋𝑘𝑓 2

𝑄 𝑅 𝑄 𝑅
• Dupuit: 2 2
ℎ0 − ℎ = ln 𝑘𝑓 = 2 2
ln
𝜋𝑘𝑓 𝑟 𝜋 (ℎ0 − ℎ ) 𝑟

𝑄 𝑟2 𝑄 𝑟2
• Thiem ℎ22 − ℎ12 = ln 𝑘𝑓 = 2 2 ln 𝑟
𝜋𝑘𝑓 𝑟1 𝜋(ℎ2 −ℎ1 ) 1 30
Calculation exercise
The production well has an effective well radius of 0.30 m and reaches
the low-permeable clay layer. During the pumping test 35 l/s were
taken from the production well. The lowered GW level was measured in
the production well itself (s = 3.00 m). A drop of 1.42 m was measured
in a 32 m distant GW observation well under stationary conditions. The
aquifer (GWL) is made up of quaternary sands with a thickness of 23 m.
The base is built of marls. The GW surface of the unstressed GWL is 1.5
m below ground level. Please determine the hydraulic permeability
coefficient kf.

31
Calculation exercise ctd.

32
Solution

Q = 0,035 m³/s
r1 = rw = 0,30 m
r2 = 32 m
Thiem – Dupuit: h1 = h = 23.00 – 1.50 – 3.00 = 18.50 m
h2 = 23.00 – 1.50 – 1.42 = 20.08 m
𝑄(ln 𝑟2 − ln 𝑟1)
𝑘𝑓 =
𝜋(ℎ22 − ℎ12 )

0.035 m³/s · ( ln 32 m− ln 0.3)

𝑘𝑓 = = 8.53 · 10-4 m/s
π · ((20.08 m) −(18.5 m ))
2 2

33
Pumping in a leaky aquifer

34
DeGlee – Semi-Confined – Conditions:
• Aquifer:
• Leaky
• Homogeneous & isotrope
• Unconfined, no vertical intrusion or leakage
• Aquifer thickness M is constant
• Infinitely wide
• Aquifer bottom is horizontal
• Well:
• Perfect
• Flow:
• Stationary
• Constant pressure head during pumping
• Vertical in the aquitard
• Drawdown in the unpumped aquifer is negligible
35
DeGlee – Semi-Confined – Formulas:
𝑄 𝑟 𝑚′
𝑠(𝑟) = 𝐾0 𝐿= 𝑇 ′
2𝜋𝑇 𝐿 𝑘𝑓

S Drawdown m
Q Discharge m³/s
T Transmissivity m²/s
r Radius m
L Leakage factor m
m‘ Water height in aquitard m
k‘f kf of aquitard m/s
K0 Modified Bessel fct.
36
Modified Bessel function
• Solutions to the general differential equation

• Mathematically rather complex and (usually) not calculated by hand

• Tabellized or graphically known (old times)
• Computed (nowadays)

37
 Tabular overview – stationary pumping
Aquifer type Flow equation Name Method Parameters Comment

Confined Dupuit Calculation T One well;

𝑄 𝑅
𝑇= ln Estimation of R
2𝜋𝑠(𝑟) 𝑟
Unconfined Dupuit Calculation T One well;
𝑄 𝑅
𝑘𝑓 = ln Estimation of R
𝜋 (ℎ02 − ℎ2 ) 𝑟
Confined Thiem Calculation T Two observation wells
𝑄 𝑟2
𝑇= ln
2𝜋(𝑠 𝑟1 − 𝑠 𝑟2 ) 𝑟1
Semi-confined DeGlee Typcurve T, r/L, kf ’ Several observation wells
𝑄 𝑟
𝑇= 𝐾0
2𝜋𝑠(𝑟) 𝐿
Unconfined Thiem Calculation T Two observation wells
𝑄 𝑟2
𝑘𝑓 = ln
𝜋(ℎ22 −ℎ12 ) 𝑟1
43
Corrections – incomplete well
• Th < h0 : QC < QB < QA

A Qvollk B Qunvollk C Qvollk

s
Th Th

h0
t1

44
Corrections – incomplete well
• Usually
• Applies to very thick aquifers
• Considering difference between complete and imcomplete well
• Correction terms to common models exist

Qvollk Qunvollk

s
h T
h0
t

45
Corrections – bounded aquifer
• „positive“ boundary = inflow
s
• Infiltrating rivers or lakes
• Waterfilled faults
• Artificial infiltration h

• „negative“ boundary = outflow/isolation

• Lithological changes s
• Joints & faults
• Water discharge h

46
Bounded Aquifers – Superposition method
• Superposition with imaginary well
Top view

Observation
Wellreal well Wellimaginary

sges = sr + si [m]

47
Imaginary wells – River infiltration

Real well
Imag. Injection well
Reflecting surface

48
Imaginary wells – Lithological boundary

Real well
Imaginary well
Imag. Injection well
Reflection surface

49
Corrections – Injection well – unconfined A.
• From Dupuit (pumping well):
𝑄 𝑅 R r
𝑘𝑓 = 2 2
ln
𝜋 (ℎ0 − ℎ ) 𝑟 z
• Drawdown here negative:
h0 h
ℎ = ℎ0 + 𝑧
𝑄 𝑅
𝑘𝑓 = 2 2
ln
𝜋 (ℎ0 − (ℎ0 + 𝑧) ) 𝑟
−𝑄 𝑅 Note that Q here is considered negative,
𝑘𝑓 = 2
ln
𝜋 (2ℎ0 𝑧 + 𝑧 ) 𝑟 because pumping was considered positive!
50
Corrections – Injection well – confined A.
• From Dupuit (pumping well):
𝑄 𝑅 r
R
𝑇= ln
2𝜋𝑠(𝑟) 𝑟 z
2    k f  M  (H - h 0 )
• Drawdown here negative: Q=
ln (R : r0 )

𝑠 = −𝑧 h0 h
M
−𝑄 𝑅
𝑇= ln
2𝜋𝑧 𝑟

51
Theis solution (1935) - assumptions
• Aquifer:
• Homogeneous & isotrope
• Confined, no vertical intrusion or leakage
• Aquifer thickness M is constant
• Infinitely wide
• Aquifer top and bottom are horizontal
• Pumping well:
• Perfect
• Negligible radius of well
• Flow
• Transient
• Constant pressure head prior to pumping
52
Theis solution – mathematical conditions
1 𝜕𝑠 𝜕 2 𝑠 𝑆 𝜕𝑠
• Formulate in terms of drawdown s + 2=
𝑟 𝜕𝑟 𝜕𝑟 𝑇 𝜕𝑡

• Initial condition 𝑠 𝑟, 𝑡 = 0 = 0

• Boundary conditions
• Zero drawdown in infinity lim 𝑠 = 0
𝑟→∞
𝜕𝑠
• Fixed flux equal to pumping rate at well lim 2𝜋𝑟𝑇 = −𝑄
𝑟→0 𝜕𝑟

53
Theis solution – The well function

𝑄 𝑟2𝑆
• An analytical solution can be obtained 𝑠 𝑟, 𝑡 = 𝑊
4𝜋𝑇 4𝑡𝑇
∞ −𝑦
𝑒 𝑢2
Using the „well function“ W(u) 𝑊 𝑢 = න
𝑦
𝑑𝑦 = −0.5772 − ln 𝑢 + 𝑢 −
2.2!
+⋯
𝑢

• The function W(u) is approximately calculated either numerically

(computer-based), taken from tables, or from curve matching

54
Theis solution – dimensionless drawdown
• We can express Theis solution using „dimensionless“ values
2𝜋𝑇 𝑟 𝑇𝑡
𝑠𝑑 = 𝑠 𝑟𝑑 = 𝑡𝑑 = 2
𝑄 𝑟𝑤 𝑟𝑤 𝑆
1 𝑟𝑑2
• Theis solution for infinite numbers 𝑠𝑑 = W −
2 4𝑡𝑑

𝜕𝑠𝑑 1 𝑟𝑑2
• Its logarithmic derivation = exp −
𝜕ln(𝑡𝑑 ) 2 4𝑡𝑑
55
Theis dimensionless drawdown
• Drawdown increases with time & decreases with distance
• Cone developes quickly in shape, then cone moves downward

𝑠𝑑

𝜕𝑠𝑑
𝜕ln(𝑡𝑑 )
t t 56
Theis curve fitting
• Similarity to well function
𝑄 𝑟2 4𝑇
log 𝑠 = log + log 𝑊(𝑢) log = log + log(𝑢)
4𝜋𝑇 𝑡 𝑆

constant constant

• Plot s vs. 𝑟 2 /𝑡 and W(u) vs. u on a log-log paper

• Curves will be of same shape
• Horizontal offset is equal to 𝑄/4𝜋T
• Vertical offset is equal to 4𝑇/𝑆
57
Theis curve fitting – more convenient
• Instead of W(u) vs. u – rather plot W(u) vs. 1/u & s vs. t/r²

10

1
W(u)

0.1

0.01
0.1 1 10 100 1000 10000
58
1/u
Curve fitting methodology
• Superimpose the data curves keeping the coordinate axes parallel,
• Adjust until most of the measurement points fall on the type curve
• give less weight to the early data (not optimal conditions yet)
• If no unique position is obtained (e.g. cause of flat curvature) use other
method
• Read coordinates of an arbitrary match point A
• it is not necessary for the match point to be located along the type curve.
• Substitute the values into the equations

59
Theis method – example
Match point A:

W(u) = 1
1/u = 10

t/r² = 1.5*10-3 min/m²

s = 0.16 m

60
Theis method – example
• With the additional information of Q=788 m³/day

• Use formulas from above

𝑄 788
𝑇= 𝑊 𝑢 = ∙ 1 = 392 𝑚²ൗ𝑑𝑎𝑦
4𝜋𝑠 4 ∙ 𝜋 ∙ 0.16

Unit conversion from minutes to days

4𝑇 𝑡ൗ 2 4 ∙ 392 ∙ 1.5 ∙ 10−3 /1440
𝑆= 𝑟 = = 1.6 ∙ 10−4
1/𝑢 10

61
Radius of compression
• Stimulated volume of aquifer?
• At small radial distances, the flux is equal to the pumping rate
• the aquifer transfers water to the well
• At infinity the aquifer is not affected
• Inbetween: Compression zone
• Water extracted from compressibility of aquifer and water
• Theis: drawdown never zero (besides at infinity)
• What is “negligible”?
• One approach (van Poolen, 1964): 𝑟𝑑𝑖 = 2 𝑡𝑑

62
The Jacob solution
• Approximation of Well function W(u) of Theis
• Small distance r and large times t
• Several relations become straight lines in semi-log plots
• s vs. t for r = constant
• s vs. r for t = constant
• s vs. t/r²
• Approximated formulas for S and T depending on the chosen
relationship

63
The Jacob solution – example
• s vs. t for r = constant
• Extend straight line until
it intercepts the time
axis
• Determine the slope of
the straight line
• Substitute the values
2.3𝑄
𝑇=
4𝜋∆𝑠

2.25𝑇𝑡0
𝑆=
𝑟2 64
The Jacob solution
• Conditions similar to Theis
• and u small enough
• If u < 0.01 -> 1% error
• If u < 0.05 -> 2% error
• If u < 0.1 -> 5% error
• For very small u it is difficult to visually identify the deviation from the
straight line (recommendation: u < 0.1

• The Thiem solution (steady state) is a special case of the Jacob

solution
65
Leaky aquifers – Hantush & Jacob (1956)
• Aquifer:
• Homogeneous & isotrope
• Leaky; drawdown in aquitard (& unpumped aquifer) is negligible
• Aquifer thickness M is constant
• Infinitely wide
• Aquifer top and bottom are horizontal
• Pumping well:
• Perfect
• Negligible radius of well
• Flow
• Transient
• Constant pressure head prior to pumping
66
Hantush & Jacob
𝑄 𝑟2𝑆 𝑟
• Add leakage factor in the well function 𝑠 𝑟, 𝑡 = 𝑊 ,
4𝜋𝑇 4𝑡𝑇 𝐿
𝑚′
• Leakage factor 𝐿= 𝑇 ′
𝑘𝑓

• Different values of r/L result in a „function family“

• Still apply curve fitting methodology similar to Theis

67
Hantush & Jakob – curve fitting
𝛽 = 𝑟/𝐿

68
Unconfined aquifers – Neumann (1972)
• Unconfined aquifer analysis analytically challenging
• Neumann: 2 2
• elastic early time storativity SA 𝑄 𝑟 𝑆𝐴 𝑟 𝑆𝑦
𝑠 𝑟, 𝑡 = 𝑊 , ,𝛽
• Specific late time yield Sy 4𝜋𝑇 4𝑡𝑇 4𝑡𝑇
• Sy/SA > 10
𝑟 2 𝐾𝑉
• 𝛽= (accounting for possible anisotropy)
𝑀2 𝐾𝐻

• Works with curve-matching method as before, just more complex

Kv Vertical hydraulic conductivity
Kh Horizontal hydraulic conductivity
M Aquifer thickness (m)
69
Neuman (unconfined) – curve fitting method

70
 Recovery curve following Theis (confined)
𝑄 𝑟2𝑆 𝑟2𝑆′
• Consider residual drawdown 𝑠′ 𝑟, 𝑡, 𝑡′ = 𝑊 −𝑊
4𝜋𝑇 4𝑡𝑇 4𝑡′𝑇
End of pumping
Start of pumping t´ = 0
t=0 time t

S Storativity during pumping

drawdown s

Instationär
(transient) S‘ Storativity during recovery
t Time since the start of the pumping
t' Time since the end of the pumping

71
Recovery (Theis) – confined aquifer
120

130

s [m ]
140

150
tp t'
160
-100 0 100 200 300 400 500 600 700 800 900
t [min]

120
Initial water level at t = 0
130
s [m ]

140

150

160
1 10 100 1000
(tp +t') / t' 72
Recovery (Theis) - confined
• For each s‘ calculate t‘/t
• Plot s‘ vs t‘/t on semi-log paper
2.3𝑄 𝑡
• Fit a straight line -> slope is 𝑠′ =
4𝜋𝑇
𝑙𝑜𝑔
𝑡′

• Assuming S = S‘

73
Well-bore storage
• Pumping well might contain significant amount of fluid
• Early times: Most water pumped from well itself
• Later times: Aquifer contribution becomes significant

• Well-bore storage coefficient 𝐶𝐷

• Well-bore storage is linear part in log-log plot s vs t

74
Skin effect
• Lower hydraulic conductivity in the immediate vicinity of the well
• poor well development,
• deposition of particles,
• development bacterial films.
• Increased hydraulic conductivity is possible
• Less impact on drawdown
• Expression: Additional drawdown through a cylindrical well
With skin factor σ (positive for clogged well)

𝑄 𝑇 − 𝑇𝑆 𝑟𝑠
𝑠𝐷 = 𝜎 𝜎= 𝑙𝑛
2𝜋𝑇 𝑇 𝑟𝑊

75
Skin effect

76
Pumping with steep hydraulic gradients

77
Diagnostikplots
Diagnostikplots
Diagnostikplots
Diagnostikplots
Diagnostikplots
Diagnostikplots
Diagnostikplots
Diagnostikplots
Diagnostikplots
Diagnostikplots
 Drawdown s (solid line)

ds/dt = Derivative of s (dashed line)

time
What kind of aquifer is this?

88
 What kind of aquifer is this?
ds/dt = Derivative of s (dashed line)
Drawdown s (solid line)

time
Confined Ideal Aquifer following Theis model 89
 Drawdown s (solid line)

ds/dt = Derivative of s (dashed line)

time
What kind of aquifer is this?

90
 Drawdown s (solid line)

ds/dt = Derivative of s (dashed line)

time
Leaky Aquifer
What kind of aquifer is this?

91
 Drawdown s (solid line)

ds/dt = Derivative of s (dashed line)

time
What kind of aquifer is this?

92
 Drawdown s (solid line)

ds/dt = Derivative of s (dashed line)

time
What kind of aquifer is this?

Double-Porosity Aquifer
93
 Drawdown s (solid line)

ds/dt = Derivative of s (dashed line)

time
What kind of aquifer is this?

94
 What kind of aquifer is this?
ds/dt = Derivative of s (dashed line)
Drawdown s (solid line)

time
95
Confined Aquifer with well-bore storage and skin effects
Calculation exercise
Draw a typical drawdown over time relationship following the model of
Neuman in an unconfined aquifer with a delayed reaction of the water
level and explain the curve.

96
Solution

97
Lessons learned
• Transition between drawdown s and hydraulic head h
• Dupuit-Thiem equations for stationary pumping
• Conceptual difference between Thiem and Dupuit
• Understanding type-curve fitting
• Working with bounded aquifers

98