Engineering Geology

ENGINEERING GEOLOGY
ENGINEERING GEOLOGY
LECTURE NOTES
by
Stephen J. Martel
Associate Professor
Department of Geology and Geophysics
School of Ocean and Earth Science and Technology
University of Hawai'i at Manoa
January, 2002
Copyright © 2001 by S. J. Martel
Lecture notes are available as PDF files. You may need to install Acrobat Reader to obtain the
PDF documents.
LECTURES
Class schedule 2002
1-Introduction
2-Properties of rocks
3-Characteristics and uses of rock
4-Structures in rock
5-Quaternary deposits
6-Soils
7-Earthquake case histories(1)
8-Earthquake case histories(2)
9-Surface faulting
10-Magnitude & Intensity
11-Mechanics: Stress(1)
12-Mechanics: Stress(2)
13-Characterizing earthquake sources
14-Seismic shaking
http://www.soest.hawaii.edu/martel/Courses/GG454/ (1 of 3) [9/2/03 6:46:56 AM]

ENGINEERING GEOLOGY
15-Characterizing ground response

16-Response of structures to shaking
17-Paleoseismology
18-Recurrence intervals & probability
19-Teton Dam
20-Landslide case histories(1)
24-Landslide classification
25-Landslide recognition
26-Mechanics: Effective stress
27-Limit Equilibrium Analysis
28-Landslide characterization
29-Case histories: coastal processes
30-Mechanics: The wave equation
31-Mechanics: Wave refraction
32-Coastal Landforms
33-Hurricanes
34-Coastal hazards and risk
35-Ground subsidence case histories 1
36-Ground subsidence case histories 2
37-Consolidation
38-The diffusion (heat flow) equation
39-Numerical solution for 1-D flow
40-The 1-D consolidation equation
41-Examples of 1-D consolidation
42-Subsidence in three dimensions
43-Review of ground subsidence
44-Nuclear waste disposal
REFERENCES
References
REPORT ORGANIZATION SUGGESTIONS AND EVALUATION CRITERIA

Report Organization Suggestions
Report Evaluation Criteria
HOMEWORK ASSIGNEMNTS
Homework #1
Homework #2

ENGINEERING GEOLOGY
GROUP PROJECTS
Project #3
MANOA VALLEY FIELD TRIP QUESTIONS

Manoa Valley questions

Geo 454 Class Schedule 2002
Day Date Topic Reading

Mon 1 / 1 4 / 0 2 Preliminaries, Course Content, General Approach, Topics
Wed 1 / 1 6 / 0 2 Rocks and their engineering characteristics I & Darcy's Law Rahn, Ch.3
Fri 1 / 1 8 / 0 2 Rocks and their engineering characteristics II Rahn, Ch.3
Mon 1 / 2 1 / 0 2 *Holiday
Wed 1 / 2 3 / 0 2 Rock structure & discontinuities in rock Rahn, Ch.4.9,4.10
Fri 1 / 2 5 / 0 2 Quaternary deposits and their engineering characteristics Rahn, Ch.3
Mon 1 / 2 8 / 0 2 Soils (from a geologist's view) and their engineering characteristics Rahn, Ch.3
Wed 1 / 3 0 / 0 2 Earthquakes: Case histories Rahn, Ch.11
Fri 2 / 1 / 0 2 Earthquakes: Case histories Rahn, Ch.11
Mon 2 / 4 / 0 2 Seismic hazard recognition & characterization: Surface faulting Rahn, Ch.11
Wed 2 / 6 / 0 2 Seismic hazard characterization: Magnitude & intensity Rahn, Ch.11
Fri 2 / 8 / 0 2 Stress transformations Rahn, Ch. 4.6
Mon 2 / 1 1 / 0 2 Stress transformations Rahn, Ch. 4.6
Wed 2 / 1 3 / 0 2 Characterizing seismic sources Rahn, Ch.11
Fri 2 / 1 5 / 0 2 Seismic hazard characterization: Shaking &Liquefaction Rahn, Ch.11
Mon 2 / 1 8 / 0 2 *Holiday
Wed 2 / 2 0 / 0 2 Seismic hazard characterization: Predicting ground response Rahn, Ch.11
Fri 2 / 2 2 / 0 2 Seismic hazard characterization: Response of structures Rahn, Ch.11
Mon 2 / 2 5 / 0 2 Paleoseismology: site selection Rahn, Ch.11
Wed 2 / 2 7 / 0 2 Paleoseismology: quantitative analysis Rahn, Ch.11
Fri 3 / 1 / 0 2 Engineering geology factors associated with dams
Mon 3 / 4 / 0 2 Landslides: Case histories Rahn, Ch. 6
Wed 3 / 6 / 0 2 Case histories of landslides: debris flows Rahn, Ch. 6
Fri 3 / 8 / 0 2 Landslide recognition & classification Rahn, Ch. 6
Mon 3 / 1 1 / 0 2 Landslide recognition Rahn, Ch. 6
Wed 3 / 1 3 / 0 2 Landslide mechanics (Effective stress) Rahn, Ch. 6
Fri 3 / 1 5 / 0 2 Landslide mechanics (Limit equilibrium analysis) Rahn, Ch. 6
Mon 3 / 1 8 / 0 2 Landslide characterization Rahn, Ch. 6
Wed 3 / 2 0 / 0 2 Coastal Processes: Case histories Rahn, Ch. 10
Fri 3 / 2 2 / 0 2 Wave mechanics: The wave equation Rahn, Ch. 10
Mon 3 / 2 5 / 0 2 Recess
Wed 3 / 2 7 / 0 2 Recess
Fri 3 / 2 9 / 0 2 Recess
Mon 4 / 1 / 0 2 Wave mechanics: Refraction Rahn, Ch. 10
Wed 4 / 3 / 0 2 Wave mechanics: Reflection and diffraction Rahn, Ch. 10
Fri 4 / 5 / 0 2 Effects of waves on coasts Rahn, Ch. 10
Mon 4 / 8 / 0 2 Effects of waves on coasts Rahn, Ch. 10
Wed 4 / 1 0 / 0 2 Dams: Case histories
Fri 4 / 1 2 / 0 2 Dams
Mon 4 / 1 5 / 0 2 Fluid withdrawl & land subsidence hazards: characterization Rahn, Ch. 7,9
Wed 4 / 1 7 / 0 2 Fluid withdrawl and land subsidence: Case histories Rahn, Ch. 7,9
Fri 4 / 1 9 / 0 2 Consolidation: Subsidence Mechanics Rahn, Ch. 7,9
Mon 4 / 2 2 / 0 2 Consolidation: Subsidence Mechanics Rahn, Ch. 7,9
Wed 4 / 2 4 / 0 2 Consolidation: Subsidence Mechanics Rahn, Ch. 7,9
Fri 4 / 2 6 / 0 2 Holiday
Mon 4 / 2 9 / 0 2 Consolidation: Subsidence Mechanics Rahn, Ch. 7,9
Wed 5 / 1 / 0 2 Consolidation: Subsidence Mechanics Rahn, Ch. 7,9
Fri 5 / 3 / 0 2 Consolidation: Subsidence Mechanics Rahn, Ch. 7,9
Mon 5 / 6 / 0 2 Open
Wed 5 / 8 / 0 2 Open
Fri 5/10/02
Mon 5/13/02
Wed 5/15/02
Fri 5 / 1 7 / 0 2 Projects due at 2:00
Page 1
Geo 454 Class Schedule 2002
Assignment Schedule
Day Date Topic

Mon 1/21/02 Holiday
Mon 1/28/02
Mon 2/4/02
Mon 2/11/02
Fri 2/15/02
Mon 2/18/02 Holiday
Mon 2/25/02 Earthquake mechanics homework due
Mon 3/4/02
Mon 3/11/02 Class project I due
Mon 3/18/02 Landslide homework due
Mon 3/25/02 Spring Recess
Mon 4/1/02
Mon 4/8/02
Mon 4/15/02 Class project II due
Mon 4/22/02
Mon 4/29/02 Homework 3 due
Mon 5/6/02
Mon 5/13/02 Class project III due?
Fri 5/17/02 Projects due at 2:00
Page 2
Geo 454 January 17, 2002 1
INTRODUCTION (01)
I Main Topics
A Engineers, Geologists, & Society
B Approach to Engineering Geology
C Importance of case histories
D Mechanics
I I Engineers, Geologists, & Society
A Engineers
a Solve problems
b Quantitative analysis emphasized
c Models often simplified/simplistic
B Geologists
a Study problems
b Qualitative analysis emphasized (traditionally)
c See earth as complex (heterogeneous & anisotropic)
C Society
a Pays the bills for problems and regulates response
b Confused by conflicting analyses
c Sees geologists and engineers as mysterious
I I Approach to Engineering Geology
A Hazard Recognition (Regional & site-specific)
1 Hazard = condition, process, or potential event that poses a threat
to personal or economic health, safety, or welfare
2 Province of geologist
B Hazard Characterization (Regional & site-specific)
1 Characterization: thorough description of system state & history
a What are the essential (and/or recurring) features/processes?
b Where are the features? (Geometry)
c What are their engineering and hydrologic properties?
d When did the geologic feature (structure/rock/deposit) form?
2 Province of geologist & engineer
C Risk Evaluation (Involves probabilities)
1 Risk = function (product) of probability of occurence and potential
loss. Example: Teton Dam. R = (1.5x10-4 /yr)($7x10 8 ) ≈ $105 /yr
2 Province of geologist & engineer
Geo 454 January 17, 2002 2
D Risk Assessment
1 Province of society at large
2 Is the level of risk tolerable?
III Importance of case histories
A Learn from the experience of others
B What has happened can happen
C Problems occur when all four of the above steps not executed
D Don't ignore heterogeneity, discontinuities, and anisotropy
E Demands vs. sufficiency of data often conflict
1 Too little time
2 Too little data (typical geologist's problem; exposures
incomplete)
3 Too much data
4 Incorrect or inadequate data
F Inadequate understanding of geologic processes ⇒ trouble
G Investigators with different backgrounds see the same thing
differently
I V Mechanics
A How do the processes operate?
B What factors are important?
C Increasingly emphasized as part of quantitative analyses
D What are the assumptions in the analyses?
GG 454 January 18, 2002 1
IMPORTANT ENGINEERING & HYDROGEOLOGIC PROPERTIES OF ROCKS (02)
I Main Topic: Variety of qualities and quantities relevant to the

mechanical and hydrologic behavior of rocks
I I Density (ρ )
A Density: mass/volume (e.g. kg/m3 ). ρ water = 103 kg/m3
B Specific gravity ( ρ / ρ water) and unit weight (γ = weight/volume)
C Factors affecting density: mineralogy and porosity
III Strength and stress
1 Strength: maximum stress that a rock can withstand
2 Compressive strength, tensile strength, & shear strength
3 Dimensions of stress (s): Force/area (e.g. N/m2 )
4 Conversion of units: 1 MPa = 106 N/m2 = 145 psi ≈ 10 atm
5 Pressure at bottom of a 10-m-deep pool ≈ 0.1 MPa ≈1 atmosphere
6 Stress as a tensor (quantities with 2 associated directions)
a σij acts on a plane ⊥ to the i direction, and in the j direction.
b Normal stress acts ⊥ to surfaces ; shear stress acts || to them
7 Factors affecting rock strength
a Rock type (crystalline vs. clastic) and mineralogy
b Weathering: generally decreases rock strength
c Discontinuities: these decrease rock strength
d Sample size: small samples stronger than large samples
IV Hardness
A Mohs hardness scale (1= talc, 3 = calcite, 7 = quartz, 10 = diamond)
B Factors affecting rock hardness
1 Mineralogy
2 Weathering: generally decreases rock hardness
V Strain (ε ) and Poisson's ratio (ν )
A Strain (ε ) = (L1- L o )/L o = ∆ L/L o . Strain measures deformation.
B ε is dimensionless
C Poisson's ratio ( ν ): ε 1/ ε 2 (strain in one direction/strain in another)
VI Hook's law of linear elasticity and Young's modulus (E)
A Hook's law of linear elasticity: σ x = E ε x (or ε x = σ x /E)
B E has dimensions of stress
C Significance of Young's modulus
1 For engineering projects: low E ⇒ large deformations from loads
Stephen Martel 2-1 University of Hawaii

GG 454 January 18, 2002 2
STRESS-STRAIN CURVE:
RESULT FROM A ROCK STRENGTH TEST
Strength
STRESS
SLOPE = E
Linear elastic behavior
STRAIN

GG 454 January 18, 2002 3
Components of stress at a point (the "on-in rule")

σij acts on the plane perpendicular to the xi axis
and in the xj direction
If i = j: normal stress: σ11, σ22, σ33
If i = j:shear stress: σ12, σ13, σ21, σ23, σ31, σ32
X2
σ22
σ21
σ23
σ12
σ11
σ32
σ31 σ13 X1
σ33
X3
σ11 σ12 σ13

σ = σ21 σ22 σ23
ij
σ31 σ32 σ33

GG 454 January 18, 2002 4
2 For seismology: Young's modulus affects seismic velocity

D Factors affecting Young's modulus
1 Rock type
2 Mineralogy
3 Weathering
4 Fractures
VII Creep behavior
A Creep: "slow" inelastic deformation at low stress
B Rocks that creep
1 Ice
2 Salt
VIII Porosity (n)
A Definition: volume of voids/total volume of rock
B Void ratio (engineering term) = volume of voids/volume of solids
C Fluids (e.g. air, water, petroleum, natural gas) occupy voids
D Factors affecting porosity
1 Rock type
2 Weathering
3 Sorting: excellent sorting (or poor grading) ⇒ high porosity
IX Hydraulic conductivity (K) See handout
A Hydraulic conductivity = how readily rock conducts fluid ≠ porosity!
B Q = -K A (∆h/ ∆L )
Discharge Hyd. cond. Area head gradient
m 3 / s e c = (m/sec) (m 2 ) (m/m)
C Factors affecting hydraulic conductivity
1 Dynamic viscosity and density of fluid
2 Character of rock ("intrinsic permeability")
a Interconnection and apertures of pores
b Interconnection and apertures of fractures
X Chemical stability and reactivity

GG 454 January 18, 2002 5
Conversions between the elastic moduli of rocks

(See Fung. p. 216-218; Turcotte and Schubert, p. 104-112)
Under relatively low loads of relatively short duration, nearly all solid
materials, including rocks, deform in a recoverable manner. In other
words, if the loads are relaxed, the material rebounds to its original
shape. This is called elastic deformation. Under sufficiently large loads
or over sufficiently large time frames, the materials will fail or flow,
and they will not recover their original shapes when the loads are
removed.
Elastic moduli describe how a material will deform elastically under a

given load or stress. Elastic moduli have dimensions of stress; they are
the ratio between an applied or induced stress and the associated strain.
Strain is a dimensionless term that describes the deformation of a
material.
There are a few common ways strain is described. Longitudinal strain (ε )

refers to the change in length of an element of material divided by its
original length: ε = ∆ L/L o . Volume strain (∆ ) refers to the change in area
of an element of material divided by its original area: ∆ = ∆ V/V o .
Consider the uniaxial loading arrangement below:
σ σ
∆L
E
Stress
Lo
Strain
Young's modulus (E) relates the normal stress σ to the longitudinal strain ε
as follows: E = σ /e.

GG 454 January 18, 2002 6
Now consider the three-dimensional loading arrangement below, where the

load on each side of the element is the pressure P
P
Vo V1
P P
P K
Pressure
P
∆V = V - V
o 1
∆
Volume strain = V/V ο =∆ Volume Strain
The bulk modulus (K) relates the pressure (P) to the volume strain ∆ as
follows: K = P/∆ . The reciprocal of the bulk modulus is the
compressibility ß. A highly compressible material has a low bulk
modulus.
As one might guess, K and E are related. For isotropic materials:

E = 3K (1-2ν ), where ν = Poisson's ratio, and K = E/{3(1-2ν )}

GG 454 January 18, 2002 7
The Mechanical Energy in Flowing Water
Water flows from high potential energy to low potential energy.
How is the mechanical energy partitioned in the water?
Consider the mechanical energy contained by a small mass of water m.

The water occupies a volume V, and the density of the water is ρ . We
consider steady state flow of an incompressible fluid (its density ρ is
constant) and neglect the loss of energy due to loss of heat. Fluid flow
driven by thermal and chemical gradients are also not treated.
Total energy = kinetic energy + elevation potential energy (1)

+ pressure potential energy
E total = 1/2 mv2 + mgh + Epressure . ( 2 )
Note that the dimension of pressure (Force/area) is the same as

energy/volume (i.e. Force x distance/volume). It turns out that pressure is
a measure of internal energy in a volume of fluid. Dividing (2) by the
volume of the water gives the energy density in the water:
E total /V = 1/2ρ v 2 + ρ gh + P. (3)
Dividing both sides of equation (3) through by the product ρ g, which in

many cases is a constant, yields:
E total /V ρ g = v2 / 2 g + h + P/gh. (4)

Total head Velocity head elevation head pressure head
The sum of all these factors is called the hydraulic head and has
dimensions of length. The hydraulic head can easily be measured in the
field: it is the standing elevation that water rises to in a well. Usually
the kinetic energy term is negligible for ground water flow, so the
hydraulic head is effectively the elevation head plus the pressure head.
Water flow from high head to low head. Note that this is very different
from the water flowing from high pressure to low pressure; if water did
that, it would flow from the bottom of a swimming pool to the top!

GG 454 January 18, 2002 8
DARCY'S LAW
h
1
Well 1
∆h
Well 2 h
2
∆L
Cross-sectional
area A
Q = -K A ( ∆ h/ ∆ L )
Discharge = (-hydraulic conductivity) (Area) (head gradient)
( m 3 /sec) = (m/sec) 2
(m ) (m/m)
The head gradient is the change in head divided by the length of the
flow tube. The minus sign indicates that flow is in the direction of
decreasing head (i.e. flow is from high potential energy to low potential
energy). In cases of unconfined aquifers where the flow direction is
nearly horizontal, the expression ∆ h/ ∆ L is effectively the slope of the
ground water table.
The hydraulic conductivity is really a function of both the porous
medium and the fluid that flows through it:
K = K i (ρ g / µ )
where K is the hydraulic conductivity, K i is the intrinsic permeability o f
the porous medium, ρ is the fluid density, g is gravitational acceleration,
and µ is the dynamic viscosity of the fluid. Low-density, viscous (i.e.
"thick") fluids will flow slower than high-density, "thin" fluids. Intrinsic
permeability is measured Darcies. 1 Darcy = 9.87 x 10-9 cm2 . For H2 0 at
15.6°C, an intrinsic permeability of 1 Darcy is equivalent to a hydraulic
conductivity of 8.61 x 10-4 cm/sec.

GG 454 January 18, 2002 1
ENGINEERING & HYDROGEOLOGIC CHARACTERISTICS OF ROCKS (03)
I Main Topics
A Engineering & hydrogeologic characteristics of rocks
B Engineering uses of rocks: Recognition of potential uses
I I Engineering & hydrogeologic characteristics of rocks: general trends
A "Hard" rocks generally have higher unconfined compressive strengths
and higher Young's moduli (E) than "soft" rocks.
B Compressive strength generally less than tensile strength
C Poisson's ratio (ν; ν ≈ 0.25)
1 Effect of Poisson's ratio for 1-D strain (for isotropic materials)
a εy = - ν εx; εz = - ν εx: εy = εz.
b ν > 0: shortening in x-direction ⇒ lengthening in y and z
c ν < 0: shortening in x-direction ⇒ shortening in y and z
2 Effect of Poisson's ratio for 3-D strain (for isotropic materials)
a e = ∆ V/V o = (V1 -V o )/V o ≈ ε x + ε y + ε z = ε x (1-2v)
b ν < 0.5: shortening in x-direction ⇒ volume decrease
c ν = 0.5: shortening in x-direction ⇒ no volume change
d ν > 0.5: shortening in x-direction ⇒ volume increase
D Chemical stability of rocks depends on chemical environment
E Hydraulic conductivity (K) of continuous porous media
1 K hard rocks > Ksoft rocks
2 K fractured rocks > Kunfractured rocks
F Shape of rock bodies relevant to isotropy and continuity of rock
G Wide range in properties requires site-specific characterization
III Engineering uses of rocks
A Building stone
1 Granite (Good if fractures are not closely spaced)
2 Slate (Very durable, high tensile strength, difficult to work)
3 Sandstone (Durability depends on porosity and cement)
4 Marble (e.g. Washington Monument)

GG 454 January 18, 2002 2
B Crushed stone (for road fill, dams, riprap; $3 B in 1973)

1 Limestone! (Needs to be pure; location & hardness keys)
2 Basalt
3 Granite
C Sand & gravel
1 Greatest extracted tonnage of any resource
2 Expensive to transport; needs to be available locally
D Concrete Aggregate
1 Material should be clean, poorly sorted, strong, and stable
1 Good rocks: granite, clean sandstone
2 Reactive (bad) rocks: chert, opal, rhyolite, andesite, phyllite,
shale, tuff, siliceous limestone, sulfide-bearing rocks
E Lime (CaO), Cement, and plaster
1 Slightly impure limestone
2 Coral
3 Gypsum
F Clay and sand (ceramics, glass, & abrasives)
1 Kaolinite clay for china, paper, paint, plastic, etc.
2 Bentonite (Na-rich montmorillonite) for drilling mud and sealer;
from devitrified volcanic ash
3 Clays for bricks (e.g. kaolinite, gibbsite)
4 Sand - needs to be 93%+ pure to avoid reactions and discoloration
a Foundry molds
b Glass
c Abrasives

GG 454
Crystalline Rocks ("Hard" rocks)

Mineral confined Streng Young's Poisson's Chemical Grain Discont. Nature of Hydraulic Porosity Shape
Composition mpressive/Tens Modulus ratio Stability size Permeability Conductivity
(MPa) (GPa) m/sec
Stephen Martel
Plutonic Rocks No frax <1E-13 - 2E-10
Fracture 6E-9 -- 2E-4
Granite Silicate 00-900/15-3 16-86 (0.28)-0.30 Stable mm-cm Fractures Fractures <0.01-0.05 Spheroidal
and pores or sheets
Gabbro Silicate 30-300/15-3 8 5 - 11 9 0.33 Stable mm-cm Fractures Fractures <0.01-0.05 Spheroidal
and pores or sheets
Diorite Silicate 00-800/15-3 55-107 0.22-0.30 Stable mm-cm Fractures Fractures <0.01-0.05 Spheroidal
and pores or sheets
Volcanic Rocks
Basalts Silicate 50-300/10-3 18-78 0.09-0.35 Stable mm Fractures Fractures 3E-9 -- 3E-5 ~0.10 Tongues/sheets
Beds and pores
Andesite Silicate ??? 65 0.18 Stable mm Fractures Fractures 3E-9 -- 3E-5 ~0.10 Tongues/sheets
Beds and pores
Welded Tuffs Ash ??? ??? ??? "Stable" µm-dm Fractures Fractures
(Silicate glass) Beds 3E-9 -- 3E-5 0.10-0.80 Tongues/sheets
Unwelded Tuffs Ash ??? ??? ??? "Stable" µm-dm Beds Fractures
(Silicate glass) Fractures and pores Sheets
Metamorphic rocks No frax <1E-13 - 2E-1 Similar to
3-3
Fracture 6E-9 -- 2E-4 plutonic rx
Slate Silicate 100-200/7-20 88 ? Stable µm Foliation Fractures Sheets
Fractures
Schists Silicate 25-162/? 14-98 ? Stable mm-cm Foliation Fractures Sheets
January 18, 2002
Fractures
Gneisses Silicate 50-230/5-20 37-103 0.06-0.29 Stable mm-cm Foliation Fractures Spheroidal
Fractures or sheets
Quartzite Silicate 50-629/10-3 1-171 0.10-0.20 Stable mm Fractures Fractures Sheets
Marble Carbonate 55-274/7-20 48-89 0.17-0.30 Soluble mm Fractures Fractures Sheets

and pores
Sources: Costa and Baker, 1981, p. 120, p.320; Carmichael, 1989, p. 689-707; Freeze and Cherry, 1979, p. 29.
3
University of Hawaii
GG 454
Sedimentary Rocks ("Soft" rocks)

Mineral confined Streng Young's Poisson's Chemical Grain Discont. Nature of Hydraulic Porosity Shape
Composition mpressive/Tens Modulus ratio Stability size Permeability Conductivity
(MPa) (MPa)
Stephen Martel
Shale* Clay 2-216/2-10 0.4-68 (0.43)-0.34 Stable µm Beds Fractures <1E-13 - 8E-10 Sheets
Fractures and pores
Siltstone* Silica 3-256/?? 0.1-26 0.05-0.35 Stable µm Beds Fractures Sheets
Fractures and pores
Sandstone* Varies 3-256/4-25 0.4-69 (0.51)-0.36 Depends mm Beds Fractures E-10 -- 3E-6 0.05-0.30 Tongues/sheets
on cement Fractures and pores
Conglom.* Varies epends on ceme ?? ?? Depends mm-cm Beds Fractures E-10? -- 3E-5 0.05-0.30 Tongues/sheets
on cement and pores
Breccia* Varies epends on ceme ?? ?? Depends mm-cm Beds Fractures Tongues/sheets
on cement and pores
Limestone†* CaCO3 6-245/5-25 2.9-97 (0.13)-0.64 Soluble µm-mm Beds Fractures 3E-10 -- E-2 0.01-0.10 Sheets
Fractures and pores
Dolomite CaMg (CO3)2 44-322 12.7-95 (0.09)-0.51 Stable µm-mm Beds Fractures 6E-10 -- 3E-6 Sheets
Fractures and pores
Evaporite†* NaCl, KCl, CaSO8-31 (salts flow 3.0-35 0.03-0.19 Soluble µm-cm Beds Fractures †† Sheets
and pores
† Sedimentary crystalline rock
3-4
†† Hydraulic conductivities may be off scale at both ends for evaporites
*Note: The strength and permeability depend heavily on the degree of cementationand the orientation of bedding or foliation
January 18, 2002
4
GG 454 January 25, 2002 1
STRUCTURES AND DISCONTINUITIES ROCKS (04)
I Main Topics
A Review of first three lectures
B Geologic, engineering, and hydrologic significance of structures
C Types of structures
I I Failure of geologic materials (e.g. earthquakes, landslides, avalanches)
and fluid flow very commonly is localized along pre-existing
structures (structures can be reactivated)
Recognition of potentially hazardous condition
III Types of structures (Characterization)
B Planar Discontinuities (see handout)
1 Fractures: structural discontinuities (all rock types).
A fracture is classified according to the relative motion across
it. Fractures commonly occur in parallel sets and thus impart
anisotropy (directional variability) to rocks.
a Joints: opening mode fractures
b Faults and fault zones: shearing mode fractures
c Dikes: opening mode fractures that contain igneous rock
d Dimensions of fractures
Dimension Joint Fault/ fault zone Dike
Thickness 0 + - 1 cm 0 + - 1 km 0 + - 10 km
Slip 0 0 + - 100's km 0
In-plane 0 - ~ 100 m 0 + - ~1000 km 0 + - ~1000 km
dimensions
2 Bedding planes (sedimentary rocks & volcanic rocks)
a Sedimentological discontinuities
b Some individual bedding planes extend for tens of km
b Bedding planes can slip and become faults or landslide slip
surfaces
3 Foliation (metamorphic rocks & igneous rocks)
a preferred alignment of minerals in planes.
b Concentrations of aligned weak minerals (e.g. mica, talc)
provide potential failure surfaces.

GG 454 January 25, 2002 2
B Shear zones
http://earth.leeds.ac.uk/shearzones/
1 Deformation is continuous across a shear zone
2 Rock within shear zones commonly is foliated
3 Shear zones common in plutonic & metamorphic rocks
C Folds
1 Require rocks to be layered or bound by parallel discontinuities;
folds occur in all rocks, including (rarely) some plutonic rocks!
2 Folding commonly causes bedding places to slip
3 Types
a Anticlines
i Oldest rocks in center of fold
i i Usually "A-shaped"
b Synclines
i Youngest rocks in center of fold
i i Usually "U-shaped"
Folds in Cross Section

Youngest Rock Layers
Axial surface
of anticline Axial surface
of syncline
Oldest Rock Layers

GG 454 January 25, 2002 3
FOUR PLANAR GEOLOGIC STRUCTURES
Dilatant
splay
cracks
JOINT FAULT
SHEAR
ZONE DIKE
For joints and dikes (opening mode fractures) the relative displacement of originally
neighboring points on opposing walls is perpendicular to the fracture
For shear zones and faults, the relative displacement of neighboring

points is parallel to the feature
Deformation (displacement) is discontinuous across a fault
Deformation (displacement) is continuous across a shear zone

GG 454 January 25, 2002 4
ORIENTATION OF LINES AND PLANES
PLANES
Strike direction
Horizontal Line Horizontal Line
Plane Dip angle
Dip direction Vertical plane
Horizontal Line
Horizontal Plane
Right hand rule for strike and dip directions: If thumb on right hand points in the direction
of strike the fingers on the right hand should point in the direction of dip
LINES
Horizontal Line Horizontal Line

Pitch
Plunge
Trend
Line 1
Line 2
Vertical plane
Need to define orientation of plane for the pitch (rake) to have meaning
The POLE to a plane is a line that is perpendicular to the plane.

The trend of the pole is opposite the direction a plane dips.
The plunge of a pole and the dip of a plane sum to 90°.

GG 454 January 25, 2002 5
Geologic Conventions for Measuring Orientations

Compass Bearings
By quadrant (relative to north or south). The angle does not exceed 90°
By 360° azimuth (0° - 360°)
Examples
N0°E N45°E N90°E S45°E S0°E S45°W S90°W N45°W
0° 45° 90° 135° 180° 225° 270° 315°
Lines
Trend: A compass bearing
Plunge: An inclination below horizontal
Examples: The lines below all plunge at 30°. Their trends vary
according to the table above
30°
Trend: N45°W
30° Trend: N45°E
30° Plunge: 30°
N Plunge: 30°
Trend: 270° Trend: 90°

Plunge: 30° Plunge: 30°
30° 30°
30°
Trend: S45°W 30° Trend: S45°E
Plunge: 30° 30° Plunge: 30°
Planes
Strike: A compass bearing along a horizontal line in a plane
Dip: An inclination below horizontal
Examples: The planes below all dip at 70°. Their strikes vary
according to the table above
Strike: 315° 70° 70°

Dip: 70° NE
N The "NE" is useful
70°
70°
Strike: N90°E
Dip: 70° S
70° 70° The "S" is useful
Strike: S45°W
Dip: 70° NW 70°
The "NW" is useful 70°

GG 454 January 25, 2002 6

GG 454 January 25, 2002 1
QUATERNARY DEPOSITS (05)
I Main Topics
A Distribution of Quaternary deposits
B Types of Quaternary deposits
C Dating methods for Quaternary deposits
I I Distribution of Quaternary deposits
http://www.wae.com/px/newcom/quaterna.htm
http://geology.er.usgs.gov/eastern/us-namap.html
http://geopubs.wr.usgs.gov/open-file/of00-444/
A Surficial deposits (mostly Quaternary) cover 95-97% of land area
B Quaternary deposits: poorly consolidated sediments
C Soils: these develop in-place
III Types and properties of Quaternary deposits; see attached table
A Depositional environment ⇒ composition, grain size, & sorting
B Tremendous variety of surficial deposits;
site-specific mapping and testing commonly is called for.
IV Dating methods for Quaternary deposits (not comprehensive!)
A Reasons for dating
1 To determine time intervals between periodic events (e.g. quakes)
2 To relate possible causes and effects (e.g. tsunamis and deposits)
3 To determine rates of geologic processes (e.g. erosion)
B Relative dating
1 Stratigraphic methods
a Classical methods
b Tephrochronology
c Magnetostratigrapy
2 Fossils (including pollen)
3 Relative amounts of weathering
a Degree of soil development
b Weathering of clasts
4 Relative amounts of erosion (e.g. by water and wind)
5 Relative amounts of deposition

GG 454 January 25, 2002 2
C "Absolute" (quantitative) dating

Technique Age range (years) Comments
14C 150-80,000 Most widely used
Uranium series Organic carbonates
e.g. 234 U- 230 T h 10,000 - 350,000 U-Th good for marine
3 9 Ar- 4 0 A r >10,000 Rapid improvements
Amino acid > ~50,000 Applicable to bone, shells
racemization
Desert varnish 15,000 - 500,000 Active debate
Dendrochronology <8200 Also indicates climate
Fission track > 100 Large uncertainties
Varve chronology <20,000 Geographically restricted
Obsidian rind hydration > 10,000 Large uncertainties
Tritium <30 V. good for groundwater

GG 454
Depositional Grain size Sorting Hydraulic Deposit Comments

Setting Conductivity Shape
Stephen Martel
Fluvial deposits
Meandering
Channels Gravel & sand Good Sinuous tongues Commonly superposed
Bars Gravel & sand Excellent Lens
Overbank Silt & clay Fair Sheet
Splays Sand Good Tongues
Oxbow lakes Sand, silt, clay Fair-good Horseshoe
Braided Gravel & sand Good Braided
Alluvial fans
Braided Gravel & sand Good Braided
Debris flow Gravel-clay Poor Tongue
Interfan Sand, silt, clay Fair-poor Irregular sheet
5-3
Lake deposits
Varves Clays Good Sheet
Channels Sands Fair-good Tongue
January 25, 2002
Marine clays Clays Good 6E-13 -- 2E-9 Sheet Can be "quick"
Swamps/marshes Peat Fair Sheet
Deltas
Topset Sand Fair-good
Foreset Sand & silt Fair
Bottomset Clay Fair
Distributaries Sand & silt Fair-good Fingers
3
Depositional Grain size Sorting Hydraulic Deposit Comments
Setting Conductivity Shape
GG 454
Beach deposits Gravel & sand Excellent Sheet

Tidal flats Sand, silt, clay Fair Sheet Beware of "quick sands"
Stephen Martel
Glacial deposits 8E-13 -- 2E-6
Unstratified Gravel-clay Poor
Ground Moraines Gravel-clay Poor Sheet
Lateral moraines Gravel-clay Poor Tongue
Terminal moraines Gravel-clay Poor Horseshoe
Stratified Good ″ E-5
Eskers Gravel & sand Good E-8 -- E-4 Sinuous tongues
Eolian Deposits
Sand dunes Sand Excellent 2E-5 -- 2E-4 Lens/sheet
Loess Silt Excellent E-9 -- 5E-6 Sheet Brittle: can collapse
5-4
"Gravity" deposits
Colluvium Gravel--clay Poor Sheet Blankets many slopes
January 25, 2002
Talus Gravel Excellent Cone Unstable
Reefs Coral NA Generally "high" Annulus or sheet These usually are hard; properties highly
depedent on degree of alteration
Gravel E-3 -- 1
Clean sand 3E-5 -- 8E-3
Med-coarse sand E-4 -- 3E-2
Fine sand E-7 -- E-4
Silty sand 9E-8 -- 8E-4
Silt E-9 -- 2E-5
Clay < E-7
4
GG 454 January 18, 2002 1
SOILS (06)
I Main Topics
A Pedologic classification schemes
B Engineering classification schemes
C Behavior of soils and influence of geologic history
I I Pedologic classification schemes
A Soils: the part of the regolith that can support rooted plants
1 Soils contain organic material
2 Factors influencing soil development (Hans Jenny)
a Climate
b Organic factors
c Topography
d Parent material
e Time (Soils are in many senses non-renewable resources)
B Master soil horizons
• O horizon (surface accumulation of organic material)
• A horizon (mixture of organic material and mineral soil)
a Zone of clay loss (zone of leaching of iron and aluminum)
b Moderately dark color
• E horizon
a Less organic material than A (so lighter color)
b Less iron, aluminum, and clay than B
• B horizon
a Zone of clay accumulation, ped development
b Clay can develop in place or be transported in
c Red color (iron and aluminum accumulation)
d Concentration of insoluble elements
• K horizon (Carbonate horizon; desert soils)
• C horizon (zone of weathered rock)
• R horizon bedrock)
III Engineering classification schemes
A Rock: requires blasting or heavy earth-moving equipment
B Soils: can excavate by hand or with light earth-moving equipment
C Soils as solid particles and fluid-filled voids (multiphase system)
IV Behavior of engineering soils (preview of consolidation theory)

GG 454 January 18, 2002 2
A Properties of sands, silts, and clays

Particle Grain size Comments
Gravel > 2 mm Hurts toes. Gritty
Sand 1/16 mm - 2 mm Visible to unaided eye.
Beware quick sands
Silt 1/256 mm - 1/16 mm Invisible to unaided eye. Gritty.
Washes off fingers easily.
Loess can fracture and collapse
Clay < 1/256 mm Invisible to unaided eye.
Gives soil cohesion. Sticks to fingers
when wet. Beware quick clays and
expansive clays (montmorillonite)
B Effective stress, Pore pressure, and total stress (see handout)

1 Effective stress: normal stress load born by the solid skeleton
2 Pore pressure: normal stress load born by the pore fluid
3 Total stress: effective stress + normal stress
C Consolidation (volume loss) of unconsolidated materials
During consolidation porosity (void ratio) and water content
decrease and strength increases. Usually soil strength increases
with depth. Time for consolidation primarily controlled by the time
it takes water to flow from material.
1 Soil memory and preconsolidation stress (see handout)
The mechanical behavior of an unconsolidated material hinges
varies depending on the past loads imposed on it.

GG 454 January 18, 2002 3
2 Influence of geologic history on soil behavior

a Normally consolidated soil: soil has been consolidated by a
load equivalent to that of the existing overburden
b Overconsolidated soil: soil has been consolidated by a load
greater than that of the existing overburden
i Erosion of overburden
i i Desiccation
c Underconsolidated soil: soil has been consolidated by a load
less than of the existing overburden
d Effect of moisture on soil color
i Wet conditions = reducing environment: blue or black color
i i Dry conditions = oxidizing environment: yellow or red color

GG 454 January 18, 2002 4
Hydromechanical Analog for Consolidation

(from Lambe and Whitman, 1969)
Initial
Porous
water surface A Piston Valve
piston
Spring
Soil Water
Physical Example Hydromechanical Analog

initial state
Valve
Valve open;
open; water
water flow
flows stops
B Valve C D Water
closed level
Water
Spring Water Spring Spring
pressure is not pressure compresses supports
rises compressed drops Water load
to pressure
support is
load hydrostaic
Hydromechanical Analog Hydromechanical Analog Hydromechanical Analog

Load applied, valve closed Valve open, water flows Equilibrium
Variation of Loads in Spring and Water
In spring
Load Applied
load
In water
tB tC tD
Time

GG 454 January 18, 2002 5
REPRESENTATIVE SOIL CONSOLIDATION CURVE
1.8 0.64
Virgin Compression Curve

1.6 0.62
Void Ratio = n/(1-n)
Porosity = n
1.4 Recompression 0.58
Preconsolidation
Stress
Decompression
1.2 0.55
1.0 0.50
0.8 0.44
10 100 1000
Vertical effective stress (kN/m2)
(Note logarithmic scale)

GG 454 January 18, 2002 1
EARTHQUAKE CASE HISTORIES (07)
I Main Topics for next two lectures

A Recognition of earthquake hazards from case histories
B Consistent lessons
C Key questions regarding characterization of earthquakes and faulting
I I Recognition of earthquake hazard: a broad perspective
A Magnitude of problem (see handout)
B Concentration of earthquakes worldwide near plate boundaries, with
a small but significant number of intra-plate quakes
IV Case Histories (The historic record)
http://wwwneic.cr.usgs.gov/neis/eqlists/USA/1872_03_26.html
A Owens Valley, 1872 (2:30 AM March 26, 1872; M="8.3+", M=7.8)
1 Epicenter at Lone Pine at base of Mt. Whitney. Second great
historical California earthquake. Rupture length of 100+ km.
2 Earthquake felt from Oregon to Mexico to Salt Lake City
3 Damage to Bidwell mansion in Chico, 300 miles from epicenter
4 Resilience of wood-frame construction and flaws of unreinforced
masonry highlighted. In Lone Pine (population 250-300), 23
killed, 50+ injured, most in the collapse of unreinforced masonry.
52 of 59 buildings wholly or partly destroyed, including every
adobe, brick, or stone building.
5 Deciding factor in G.K. Gilbert (1884) recognizing faulting hazard
in Basin and Range, including Salt Lake City
B San Francisco, 1906 (5:12 AM April 18, 1906; M=8.25, now M=7.7)
1 Epicenter: Pt. Reyes. Rupture length of ~270 miles (430 km).
2 Report of the State Earthquake Investigation Commission
a Establishment of elastic rebound theory as earthquake
mechanism (vol. 2, p. 27)
b Recognition that rupture was along a fault that could have been
identified in hindsight
c Recognition and documentation of association between shaking
intensity and and surficial geology; beginning of
characterization (vol. 2, p. 49-56)
d Resilience of wood-frame construction and flaws of
unreinforced masonry

GG 454 January 18, 2002 2
e Significance of Hayward fault

f Estimation of ~100 year recurrence interval (vol. 2, p. 18-19)
g Difference in attenuation characteristics vs. East coast quakes
(i.e. Charleston 1886)
h Calculation of energy release (vol. 2, p. 22)
i Estimate of regional principal stress field (vol. 2, p. 22-28)
j Documention of effect on ground water
3 700 people killed, $500 million damage
C San Fernando, 1971 (6 AM February 9, 1971; ML =6.4, M=6.6)
1 Rupture length of ~15 km
2 Quake occurred near the center of the largest concentration of
strong-motion instruments in U.S
3 First instrumental recording of vertical accelerations >1g
4 "Shattered ridge tops" recognized as potential evidence of
vertical accelerations >1g
5 Numerous houses damaged by ground rupture. Whether fault could
have been recognized in advance is controversial.
6 Initiated Alquist-Priolo Act
7 Near failure of Van Norman dam
8 Gave major boost to review of safety of dams in U.S.
9 Graphic evidence of potential for freeway disruption
10 Graphic evidence of inadequacy of "flexible first story" design.
Other buildings built to code generally did well
11 Resilience of wood-frame construction highlighted again.
Casualties in personal residences less than 1 in a million.
12 $500 million damage, or $71 per person in the L.A. area
13 58 deaths out of 7 million people in affected area. We were
lucky.

GG 454 January 18, 2002 3
Earthquakes since 1268 that have caused at least 50,000 deaths*

Place Year Estimated Deaths
Silicia, Turkey 1268 60,000
Chihli, China 1290 100,000
Shaanxi, China 1556 830,000
Shemska, "USSR" 1667 80,000
Naples, Italy 1693 93,000
Catania, Italy 1693 60,000
Beijing, China 1731 100,000
Calcutta, India 1737 300,000
Lisbon, Portugal 1755 60,000
Calabria, Italy 1783 50,000
Messina, Italy 1908 160,000
Gansu, China 1932 70,000
Tokyo/Yokahama, Japan 1923 143,000
Gansu, China 1932 70,000
Quetta, Pakistan 1935 60,000
Tangshan, China 1976 240,000
Iran 1990 52,000
Earthquake distributions for the entire Earth

Magnitude Number Maximum Modified Effects on
per year Mercalli Intensity populated areas
<3.4 800,000 I Recorded by seismographs only
3.5-4.2 30,000 II and III Felt by some people indoors
4.3-4.8 4,800 IV Felt by many, windows rattle
4.9-5.4 1,400 V Felt by all, dishes break
5.5-6.1 500 VI and VII Slight building damage,
plaster cracks, bricks fall
6.2-6.9 100 VIII and IX Houses move on foundations,
chimneys fall
7.0-7.3 15 X Bridges twisted, many masonry
buildings collapse
7.4-7.9 4 XI Most buildings collapse
>8.0 1/5 -/10 XII Total damage, Waves on ground,
objects thrown into air
(Modified from Skinner and Porter, 1992)

GG 454 January 18, 2002 1
EARTHQUAKE CASE HISTORIES II (08)
I Main Topics for next two lectures

A Recognition of earthquake hazards from case histories
B Consistent lessons
C Key questions regarding characterization of earthquakes and faulting
I I Case Histories (The historic record)
A Loma Prieta, 1989 (5:04 PM October 17, 1990; M=7.1)
1 The location and style of the ground response and ground water
response could have been predicted in many areas based on what
happened in 1906
2 Rupture was very similar to what had been mapped by Hall &
Sarna
3 Resilience of wood-frame construction and flaws of unreinforced
masonry demostrated again
4 Tremendous increase in ability/ speed to determine the location,
extent, and energy release (i.e. physical parameters) based on
seismologic and geodetic information
5 Possible precursor identified after the fact - long wavelength
radio waves (submarine communications)
6 No precursor recognized in advance
7 57 people killed (41 on Cypress structure). We were lucky.
8 ~$6 billion damage. Just for comparison, the 1906 quake damage
in 1987 dollars is estimated at ~$20 billion.
B Parkfield "nonearthquake" and earthquake prediction1992
1 Similar prior earthquakes of M≈6: 1881, 1901, 1922, 1934, 1966
2 Mean interevent time of 21.8 ± 5.2 years
3 Strain buildup comparable to 1966
4 Earthquake forecast issued in 1985 for a 1988 (± 5 years) quake
5 No earthquake yet

GG 454 January 18, 2002 2
C Kalapana, Hawaii (4:48 AM November 29, 1975; M=7.12)

1 Surface rupture along Kilina fault system
2 Magnitude 5.7 foreshock at 3:36
3 Maximum tsunami height 12.2-14.6 m
4 Maximum subsidence along coast: 3.5 m
5 Small eruption at Kilauea triggered 45 after quake
6 $4.1 million in damage
7 Earthquake caused by deep-seated failure of SE flank of Kilauea
8 Earthquake similar to April 2, 1868 earthquake
G Key Lessons (Common elements / recurring themes)
a Earthquake effects are a function of the strength of the source,
the path of the seismic waves, and the nature of the "receiver"
b Faults with prior Quaternary activity generate nearly all quakes
c Structural failures kill the most people. Securely tie houses to
foundations in earthquake country!
d Shaking depends on surficial materials but generally decreases
with distance from the epicenter
e Ground response can be consistent from quake to quake
f Ground rupture is devasting for structures but can be avoided
g Significant damage can occur for faults that don't rupture the
surface (Coalinga, almost Loma Prieta).
h We get surprised
i We can't predict earthquakes in the short term yet
H Some key characterization questions
a What controls how large an earthquake will be?
b How do surficial deposits respond to seismic waves?
c How do buildings respond to seismic waves?
d How can we predict when and where an earthquake is likely to
occur? What is the uncertainty associated with the predictions?

GG 454 January 18, 2002 1
SURFACE FAULTING (09)
I Main Topics
A Recognition of surface rupture hazard
B Characterization of surface rupture hazard
C Evaluation of surface rupture risk
D Assessment of surface rupture risk: is it tolerable?
I I Recognition of surface rupture hazard
A Nearly all documented surface ruptures have occurred near or along
along pre-existing faults that have broken Quaternary deposits or
have shown historic seismicity. Past is the key to the future.
B For many areas, the current tectonic regime is that of Quaternary
time. The record of prior faulting may be inappropriate.
C Surface rupture is generally confined to a zone no more than a few
hundred meters along the fault (20 m - 2 km range).
D In California, surface faulting ⇒ moment magnitudes > 5.5-6.
E Most structures are not designed to accommodate rupture through
their foundation. Exception: some earth-fill dams
F Geomorphic features along active faults
(See p. 29, 68, 85 of PP 1360)
1 Strike-slip faults
a Linear trace (linear valleys) narrow zone of deformation
b Fault scarp
c Sag ponds
d Deflected drainages
e Springs
f Shutter ridges
2 Dip slip faults
a Ragged arcuate trace, broad zone of deformation
b Fault scarp
c Springs
III Characterization of surface rupture
A Distribution of active faults and trace geometry
1 What is "active"? Slemmons (1982) lists 30 definitions!
a CA: Evidence of Holocene offset (<10,000 yrs)
b BuRec (dams): Slip in last 100,000 yrs

GG 454 January 18, 2002 2
c NRC: Multiple slip events in last 500,000 yrs

d PP 1360: evidence of late Quaternary activity (<750,000 yrs)
e Historical record and Holocene evidence of dormancy has not
proven adequate to judge fault activity.
2 Evidence for activity
a Historical surface faulting or creep
b Historical or contemporary seismicity
c Geodetic strain
d Geologic evidence (To be pursued later)
i Faulting or folding of Quaternary deposits
i i Geomorphic expression
B Character of surface rupture: seismic rupture vs. aseismic creep
C Likely sense and amount of rupture
Characteristic earthquake model: local displacements repeat
D Likely size and "timetable" of quakes
E Slip rate
F Fault geometry and segmentation
IV Risk evaluation (to be discussed later; relies upon Quaternary record)
V Risk assessment: It is generally not tolerable for new structures and
critical facilities in California to be sited on faults considered active.
http://www.trinet.org/hector/photos.html
http://www.scecdc.scec.org/lanplaya.html

GG 454 January 18, 2002 1
MAGNITUDE AND INTENSITY (10)
I Main Topics
A Types of seismic waves
B Magnitude
C Intensity
I I Seismic waves: Nearly-pure elastic waves
A Seismic waves a) cause damage and b) provide quantitative
information on source strength and effect of "path" on wave energy
B Speed of seismic waves: c = L/T (L = wavelength; T= period)
C Body waves: Seismic waves that travel inside the earth.
1 P - w a v e s : P rimary (compressional) waves. First arrival.
2 S - w a v e s : S econdary (shear) waves. Second arrival
3 Speeds of body waves
a V p = {(k + 4µ/3)/ ρ } 1/2 (ρ = density; µ = shear modulus;
b V s = {µ / ρ } 1 / 2 k = bulk modulus)
k = 1/compressibility
Material k (N/m2 ) µ (N/m2 ) ρ (kg/m 3 ) V p (m/sec) V s (m/sec)
Granite 48 x 109 32 x 109 2720 5800 3400
Water 22 x 108 0 1000 1500 0

c Key point: Vp > Vs (V p ≈ 1.7 Vs for rock)
D Surface waves: waves that travel at the earth's surface
1 These cause strong ground motion near the epicenter
2 Surface waves "feel" deeper as their wavelength increases
(Analog to water waves and wave base).
3 Surface wave velocity increases as wavelength increases.
4 Surface waves are polarized.
a Horiz. polarized waves: especially destructive; V ≈ Vs .
b Vertically polarized waves: V ≈ 0.92 Vs .
5 Amplitudes of surface waves tend to saturate (have a ceiling).
The amplitudes scale with stress drop ("strength") of fault.

GG 454 January 18, 2002 2
III Magnitude
http://www.scecdc.scec.org/measureeq2.html
A Magnitude ideally measures only the source strength
B Local (Richter) magnitude (ML ): ML = log A - log A0
1 A 0 = Base-ten log of the amplitude (in microns) of a Wood-
Anderson seismograph located 100 km from the epicenter.
2 Devised for southern California; reflects wave attenuation there
3 Does not discriminate between types of waves.
4 Sensitive to periods of 0.8 seconds.
5 Originally set up to give magnitude to nearest 1/4.
6 Because surface waves saturate, intensity tends to saturate too
7 Saturates at ML ≈ 6.5
C Surface wave magnitude (Ms )
1 Based on surface waves with 20-second periods.
2 Saturates at Ms ≈ 6.8-7.5.
3 Has been set to match Richter magnitude for ML < 7.5.
D Body wave magnitude (Mb )
1 Based on P-waves with periods of about 1 second
2 Primarily for deep focus events and for small shallow events
3 Saturates at Mb ≈ 6.8.
E Coda Magnitude: Based on decay of seismic wave amplitude.
F Moment magnitude (Mw ): based on seismic moment M o
1 M w = (2/3) log M o -10.7 (M o in dyne-cm)
2 Designed to dovetail with Ms for Mw < 7.5.
IV Intensity: A "semi-quantitative" measure of the local degree of shaking
A Function of source strength (and duration) and wave path
Intensity = f(magnitude, distance from source, local geology)
B An alternative to quantitative acceleration or velocity spectra
1 Modified Mercalli Scale (MMI = I ⇒ XII) Note Roman numerals
http://wwwneic.cr.usgs.gov/neis/general/handouts/mercalli.html
2 Rossi-Forel Scale (R-FI = I⇒ X )
http://www.seismo.nrcan.gc.ca/magnitudes/rossi_e.html
C Because magnitude saturates, intensity tends to also
D Evernden's equation for intensity decay

GG 454 January 18, 2002 3
Evernden's Rossi-Forel Intensity Formula
 1 1 
γ
  1011.8 +1.5M  γ  n −kγ 
(1) I = 3(0.5 + log  A 


  ∑ ( Ri + C )  )
 n i = 1  
 
The term n is the number of segments the fault will be broken into. For
simplicity, we will not subdivide the fault, so n = 1.
) (( R + C ) )
 1 
( −k γ
1 γ
(2) I = 3(0.5 + log  A 10 11.8 + 1.5 M γ
)
 
The term in the first set of parentheses is the energy (in ergs) as
calculated from the magnitude M. This will be a constant and defines the
contribution from the source. The energy term is multiplied by the
attenuation term in the second set of parentheses. This will will vary
with distance from the epicenter and defines part of the contribution from
the seismic wave path. This part of the expression simplifies for n = 1:
(10 ) (( R + C ) )
 +
1
γ −k 
(3) I = 3(0.5 + log A
11.8 1.5 M ) 
 
The log term can be broken down into simpler form:
{
(4) I = 3(0.5 + log A + γ log 10
1 (
11.8+1.5M
)
+ log ( R+ C ) )
−k
}
{
(5) I = 3(0.5 + log A + 1 γ (11.8 +1.5M ) − k log ( R+ C ) ) }
Now let's pick a spot on a fault (R=0) in coastal California and assume a
magnitude 6.5 earthquake occurs. This is the magnitude that the intensity
scale should saturate at, so the expression should yield a Rossi-Forel
intensity of X (10). Evernden gives the following values: A = 0.779; γ = 4,
k = 1.75, C (a pseudo-depth) of 25 km. So (5) reduces (7) in two steps:
{
(6) I = 3(0.5 + log(0.779) + 1 4 (11.8 +1.5[6.5]) − 1.75 log(25) ) }
(7) I = 3(0.5 + { −.108 + 5.388 − 2. 446} ) =10.002
This is the maximum intensity and can be corrected for the local geology
as shown on p. 161 of PP 1360.

GG 454 January 18, 2002 4
Evernden's Method for Calculating Seismic Intensity
1 Relates attenuation of "radiated seismic energy" (a function of

source strength), the distance from the source and the regional
attenuation characteristics of the basement rocks (the "path"), and
the near surface geology (part of the "path", part of the receiver").
2 Uses Rossi-Forel intensity instead of Modified Mercalli because

Rossi-Forel intensity apparently scales better with the RMS (root
mean square or "average") acceleration over the frequencies of
greatest interest. For example, accelerations increase by a factor
of 2 for each half unit of increase in the Rossi-Forel scale, but this
doesn't hold for the Modified Mercalli scale.
3 Focusing (e.g. due to surface topographic effects or the shape of the
bedrock surface at the bottom of an alluvial basin) are ignored.

GG 454 January 18, 2002 5
Why Use Intensity?
"All too often, peak acceleration is misguidedly used as a direct measure

of earthquake damage potential. It is convenient to discuss an earthquake
that has just occurred in terms of its location, magnitude, and peak
acceleration. But, as time goes on, magnitude and acceleration may no
longer be so relevant. There is a vast gap between peak instrumental
acceleration and the base shear coefficients used to design
buildings....They can be reconciled only if all the many factors related to
the earthquake are considered....it follows that the spikes observed on
records have little or no structural design significance....We need to
continue to record acceleration and to treat it as a valuable tool but to
recognize that it is not a reliable index of damage potential."
J. Blume (1981)
"Although intensity is subjective by nature and is by definition linked to

the loss extent, it is the only measure available which takes into account
such important damage factors as the spectral characteristics and the
duration of ground motion in addition to its severity. Therefore, Mercalli
intensity is still superior to all known instrumental quantities such as,
e.g., instrumental intensity (Housner intensity, Arias intensity) or
spectral acceleration, which have failed up to now in reflecting the whole
of important damage factors."
(Smolka and Berz, 1981)

GG454 Lecture 11 10/27/02 1
STRESS AT A POINT
I Main Topics
A Stress vector (traction) on a plane
B Stress at a point
C Principal stresses
D Initial 2-D stress transformation equations
→
I I Stress vector (traction) on a particular plane: τ
→ →
A τ = lim A→0 F / A . Dimensions of force per unit area
B Traction vectors can be added vectorially.
C A traction vector can be resolved into normal and shear components.
→
1 A normal traction τ n acts perpendicular to a plane
→
2 A shear traction τ s acts parallel to a plane
D The magnitudes of tractions depend on the orientation of the plane
III Stress at a point
A Stresses refer to balanced internal "forces (per unit area)". They
differ from force vectors, which, if unbalanced, cause accelerations
B "On -in convention": The stress component σ ij acts on the
plane normal to the i-direction and acts in the j-direction

1 Normal stresses: i=j
2 Shear stresses: i ≠ j
y
σyy y
σyx
σxx
σxy
σxx x σxy x
σyy σyx
Normal stresses Shear stresses

GG454 Lecture 11 10/27/02 2
 σ xx σ xy σ xz   σ11 σ12 σ13 

   
C σ ij =  σ yx σ yy σ yz  or σ ij =  σ 21 σ 22 σ 23 
σ   σ 31 σ 32 σ 33 
 zx σ zy σ zz 
D In 3-D, nine components are needed to define the state of stress at a
point. These nine components are the traction components that act
on three perpendicular planes (planes with normals in the x-, y-, and
z-directions)
E The state of stress can (and usually does) vary from point to point.
F For rotational equilibrium, σ xy = σ yx , σ xz = σ zx , σ yz = σ z y
IV Principal Stresses (these have magnitudes and orientations)

A Principal stresses act on planes which feel no shear stress
B The principal stresses are normal stresses.
C Principal stresses act on perpendicular planes
D The maximum, intermediate, and minimum principal stresses are
usually designated σ 1 , σ 2 , and σ 3 , respectively. Note that the
principal stresses have a single subscript.

E The principal stresses represent the stress state most simply.
 σ1 0 0
  σ1 0 
F σ ij = 0 σ 2 0  3-D or σ ij =   2-D
   0 σ2 
 0 0 σ 3 

GG454 Lecture 11 10/27/02 3
V Initial 2-D stress transformation equations

A Consider three planes A, A1, and A2 that form the sides of a
triangular prism; these have normals in the n-, 1-, and 2-directions,
respectively. Plane A1 is acted on by known normal stress σ . Plane
1
A2 is acted on by known normal stress σ . The n-direction is at
2
angle θ (=θ ) with respect to the 1-direction, and at angle θ with
1 2
respect to the 2-direction. The s-direction is at a counter-
clockwise 90° angle relative to the n-direction (like y and x).
s
σ2 θ2 = θ2n
A2 A1 n
τn
θ2
A θ1 = θ1n = θ
1-direction
τs σ1
θ1 + θ2 = 90°
cos θ1 = cos θ
cos θ2 = sin θ
To find the normal traction, τ n that acts on plane A, we determine

how much σ 1 and σ 2 each contribute individually to τ n and then sum
the contributions (Figures 16.1, 16.2). Similarly, to find the shear
traction, τ s that acts on plane A, we determine how much σ 1 and σ 2
each contribute to τ and then sum those contributions (Figures 16.3,
16.4). The results are:
1 τ n = σ1cos2θ + σ 2sin 2θ
2 τ s = (σ 2 - σ1) sinθ cosθ

GG454 Lecture 11 10/27/02 4
s 2-direction Fig. 1
Contribution of σ1 to τn
θ2 A1 n
τn
θ2
A θ1 = θ 1-direction
σ1
θ1
θ1 + θ2 = 90°
cos θ1 = cos θ
cos θ2 = sin θ
What does σ1 on face A1 of area A1 contribute to τn on face A of area A?
Start with the definition of stress: τn(1) = Fn(1)/ A.
The unknown quantities Fn and A must be found from the known quantities σ1 and θ.
To do this we first find the force F1 associated with σ1:

n
Force = (stress)(area) s (1)
Fn
F1 = σ1 A1 θ1
F1
The component of force F1 that acts along the n-direction is F1 cos θ1.
Fn(1) = F1 cos θ1
As can be seen from the diagram atop the page A1 = A cos θ1, so
A = A1/cos θ1
So the contribution to σn by σ1 is:

τn(1) = Fn(1) / A = F1 cos θ1 / (A1/cos θ1 ) = (F1 / A1) cos θ1 cos θ1 = σ1 cos θ1 cos
τn(1) = σ1 cos θ cos θ First contribution to τn

GG454 Lecture 11 10/27/02 5
s 2-direction Fig. 11.2

Contribution of σ2 to τn σ2
θ2 A2 n
τn
θ2
θ1 θ1 + θ2 = 90°
cos θ1 = cos θ
cos θ2 = sin θ
What does σ2 on face A2 of area A2 contribute to τn on face A of area A?
Start with the definition of stress: τn(2) = Fs(2)/ A.
The unknown quantities Fs and A must be found from the known quantities σ2 and θ.
To do this we first find the force F2 associated with σ2: F2
Force = (stress)(area)
F2 = σ2 A2
The component of force F2 that acts along the n-direction is F2 cos θ2.
Fn(2) = F2 cos θ2 n
s θ2 (2)
As can be seen from the diagram atop the page A2 = A cos θ2, so θ1 Fn
A = A2/cos θ2
So the contribution to σn by σ2 is:

τn(2) = Fn(2) / A = F2 cos θ2 / (A2/cos θ2 ) = (F2 / A2) cos θ2 cos θ2 = σ2 cos θ2 cos θ2
τn(2) = σ2 sin θ sin θ Second contribution to σn
τn = τn(1) + τn(2) = σ1 cos2 θ + σ2 sin2 θ Total value of σn

GG454 Lecture 11 10/27/02 6

Contribution of σ1 to τs
θ2 A1 n
θ1
θ2
τs σ1
θ1 + θ2 = 90°
cos θ1 = cos θ
cos θ2 = sin θ
What does σ1 on face A1 of area A1 contribute to τs on face A of area A?
Start with the definition of stress: τs(1) = Fs(1)/ A
s n
F1
Force = (stress)(area)
F1 = σ1 A1 (1) θ2
Fs
The component of force F1 that acts along the s-direction is - F1 cos θ2.
Fs(1) = - F1 cos θ2
As can be seen from the diagram atop the page A1 = A cos θ1, so
A = A1/cos θ1
So the contribution to τ by σ1 is:

τs(1) = Fs(1) / A = - F1 cos θ2 / (A1/cos θ1 ) = (- F2 / A2) cos θ2 cos θ1 = - σ1 cos θ2 cos θ1
τs(1) = - σ1 sin θ cos θ First contribution to τs

GG454 Lecture 11 10/27/02 7

Contribution of σ2 to τs σ2
θ2 A2 n
θ1
θ2
τs
θ1 + θ2 = 90°
cos θ1 = cos θ
cos θ2 = sin θ
What does σ2 on face A2 of area A2 contribute to τs on face A of area A?
Start with the definition of stress: τs(2) = Fs(2)/ A
s F2
(2)
Force = (stress)(area) Fs
F2 = σ2 A2
The component of force F2 that acts along the s-direction is F2 cos θ1.
θ1
Fs(2) = F2 cos θ1
θ2 n
As can be seen from the diagram atop the page A2 = A cos θ2, so θ1
A = A2/cos θ2
So the contribution to τ by σ2 is:

τs(2) = Fs(2) / A = F2 cos θ1 / (A2/cos θ2 ) = (F2 / A2) cos θ1 cos θ2 = σ2 cos θ2 cos θ1
τs(2) = σ2 sin θ cos θ Second contribution to τs
τs = τs(1) + τs(2) = (σ2 - σ1 )(sin θ cos θ ) Total value of τs

GG454 Lecture 12 10/27/02 1
MOHR CIRCLE FOR TRACTIONS
I Main Topics
A Stresses vs. tractions
B Mohr circle for tractions
I I Stresses vs. tractions
A Similarities between stresses and tractions
1 Same dimensions (force per unit area)
2 The normal stress acting on a plane matches the normal traction
B Differences between stresses and tractions
1 Stresses are tensor quantities and tractions are vectors.
2 The stress state is defined at a point using a fixed reference
frame, whereas a traction is defined on a plane with a reference
frame that floats with the plane.
3 Shear stress components on perpendicular planes have the same
sign, whereas shear tractions on perpendicular planes have
opposite signs.
σyy n τnn
s
σyx
τns
y s
σxy τns n
σxx τnn
x σxx τnn
σxx n τns
s
τns
σyx
s
σyy τnn n
Positive stresses at a point Positive tractions on perpendicular planes

GG454 Lecture 12 10/27/02 2
I I I Mohr circle for tractions
s
σ2 θ2 = θ2n
A2 A1 n
τn
θ2
A θ1 = θ1n = θ
1-direction
τs σ1
θ1 + θ2 = 90°
cos θ1 = cos θ
cos θ2 = sin θ
A1 τ n = σ1cos2θ + σ 2sin 2θ A2 τ s = (σ 2 - σ1) sinθ cosθ
1 + cos2θ 1 − cos 2θ sin 2θ

Now cos2 θ = , sin 2 θ = , and sin θ cosθ = , so
2 2 2
σ1 + σ 2 σ1 − σ 2 −(σ1 − σ 2 )
A3 τn = + cos2θ A4 τs = sin 2θ
2 2 2
σ1 + σ 2 σ − σ2
If = c and 1 = r , then
2 2
A5 τ n = c + r cos2θ = c + r cos(-2θ ) A6 τ s = −r sin 2θ = r sin(−2θ )
These are the equations of a Mohr circle for traction that relates the
tractions on planes of different orientation at a point. Its center
“c” is the mean normal stress (traction), and the absolute value of
“r” is the magnitude of the maximum shear traction (the circle
radius). The term σ is the most tensile stress.
1

GG454 Lecture 12 10/27/02 3
Suppose in "physical space" the angle from the normal to plane A to

the normal to plane B is +θAB (measured about the z-axis using a
right-hand rule). The tractions on these planes are represented by
points on a Mohr circle diagram. The circumferential angle from the
point for plane A to the point for plane B is the double angle -2θAB .
As θ goes counterclockwise from 0 to 180°, -2θ goes from 0 to -
360°; one complete clockwise circuit is made around the Mohr circle.
FORCE BALANCE DIAGRAM CORRESPONDING MOHR CIRCLE

("Physical space") ("Mohr circle space")
τs
s 2-direction
σ2 θ2 = θ2n
A2 A1 n σ2
τn σ
• • • 1
θ2 τn
−2θ
A θ1 = θ1n = θ
1-direction
τs σ1
(τn,τs) •
θ1 + θ2 = 90°
τn = ((σ1 + σ2)/2) + ((σ1 - σ2)/2) cos (-2θ)
cos θ1 = cos θ
τs = ((σ1 - σ2)/2) sin (-2θ)
cos θ2 = sin θ
Fig. 12.1
B Key points
1 θ is the angle between the normal to the plane σ acts on and the
1
normal to the plane of interest.
2 The maximum shear stress is at θ = ±45° from σ (−2θ = m90 o )
1
3 The normal stress maximum is at θ = 0° from σ (-2θ = 0°)
1
4 The normal stress minimum is at θ = 90° from σ (-2θ = -180°)
1

GG454 Lecture 12 10/27/02 4
C Conventions
1 The equations of lectures 16 and 17 are derived using a tensor
“on-in” convention for defining stresses, where tensile stresses
and tensile tractions are positive. Many geologists and civil
engineers, however, use a different convention in which
compressive normal stresses are positive and left-lateral shear
stresses are positive (see Figure 17.2). In this alternative
convention the leading minus sign in equation A4 is dropped. In
order to minimize confusion, we will stick with one convention,
the tensor convention, rather than deal with two.
2 Advantages of the alternative convention

a In the earth the normal stresses generally are compressive, so
having compressive stresses be positive is appealing.
b Angles on a Mohr circle can be turned in the same direction as
in “physical space”
c The convention for defining sense of shear (right-lateral or
left-lateral) is handy for describing one class of common
faults (strike-slip faults).
3 Advantages of the tensor convention
a Maintains contact with the essential underlying physics far
more readily than the alternative convention.
b It encourages consideration of the reference frames the
stresses are being transformed between.
c It is far more useful in dealing with 3-D problems.
d Tied to powerful mathematical methods that are otherwise
“hidden”.
Just remember to turn the angles on the Mohr diagram in the

opposite sense as in “physical space.”

GG454 Lecture 12 10/27/02 5
Fig. 12.2
SHEAR STRESS CONVENTIONS

Tension Positive
Tensor Convention Alternative Convention

y L(-)
σ1 σ1
R(+)
x L(-) R(+) R(+) L(-)
R(+)
L(-)
Positive shear Right-lateral shear stress positive;
stresses Left-lateral shear stress negative*
Compression Positive
Tensor Convention Alternative Convention
y R(-)
σ1 σ1
L(+)
x
R(-) L(+) L(+) R(-)
This convention is used
little if at all. It is included
for "completeness" only
L(+)
R(-)
Positive shear Left-lateral shear stress positive;
stresses Right-lateral shear stress negative*
*Positive theta direction is counterclockwise

GG454 Lecture 12 10/27/02 6
Example 1 using Mohr circle to find principal stresses

A) Draw a box in a reference frame and clearly label the stresses on its sides; this is
a critically important step.
Suppose σxx = +10 MPa (tension), σxy = +3 MPa (left lateral shear), σyy = +2 MPa (tension),
and σyx = +3 MPa (right lateral shear).
B) Determine the stresses and tractions on the faces of the box.
Here, we use the tensor "on-in" convention.
σyy = +2 MPa (tension) Stresses Tractions
n y
s σyx = +3 MPa (R.L) σxx +10 MPa τxn +10 MPa
σxy = +3 MPa (L.L) σxy +3 MPa τxs +3 MPa
σyx +3 MPa τys -3 MPa

σxx = +10 MPa (tension)
x σyy +2 MPa τyn +2 MPa
s
n Note that the magnitude of the
normal stresses and normal tractions
are equal. So τ1 = σ1 below.
C) Plot and label the points on a set of labelled τn, τs axes. Then draw the Mohr circle through the
points by finding the center (c) and radius (r) of the circle. Now label the principal magnitudes
τ1 and τ2 (τ1 > τ2); they come from the intersection of the circle with the normal stress (τn) axis.
Then assign reference axes to the principal directions; below I chose x' for the τ1-direction.
Label the negative double angle between the traction pair that act on a plane with a known normal
direction (here, x or y) and the traction pair that act on a plane with an unknown direction (e.g., x').
6 τs (MPa) 6 τs (MPa)
4 4
(τxn, τxs) (τxn, τxs)
r
2 2
τ2=τy'n c -2θxx' τ1=τx'n
τ (MPa) τ (MPa)
-2 2 4 6 8 10 12 n -2 2 4 6 8 10 12 n
-2 -2
(τyn, τys)
-4 (τyn, τys) -4
-6 -6
o o
Here, -2θxx' = -37 (clockwise),so θxx' = +18.5 (counterclockwise)
D) Draw and label a new reference frame and box showing the principal stresses.
y' y
τ2= τy'n = 1 MPa tension τ1= τx'n = 11 MPa tension

x'
θxx' = 18.5o
σ1 is the most tensile stress
x
σ2 is the least tensile stress

GG454 Lecture 12 10/27/02 7
Mohr Circle Exercises
These problems can be solved with either a Mohr circle or the stress tensor equations. They
are set up assuming a tensor convention for stresses (tensile stresses positive), so a
compressive stress of magnitude 8 MPa would correspond to a tensile stress of -8 Mpa.
x2' x2
30°
60°
x1'
120°
30°
x1
1 Suppose the most compressive stress is horizontal, trends east-west, and has a magnitude of
-4Mpa, where a negative normal stress is compressive. The least compressive stress is
horizontal, trends north-south (this has to be the case if the most compressive stress
trends east-west!) and has a magnitude of -2MPa. Find the normal stress and shear stress
that act on a vertical fault with a strike of N30°E.
2 Two locked vertical faults each support a left-lateral shear stress of 8MPa. One fault
strikes N30°E, and this fault has a normal stress of -14 Mpa, where a negative normal
stress is compressive. The second fault has a normal stress of -2 MPa. Assume that the
stress field around the faults is uniform. (a) Draw the Mohr circle. (b) Determine the
magnitude and the orientations of the principal stresses. (c) Determine the strike of the
second fault.
3 The pressure at the bottom of a swimming pool is ρgh, where ρ is the density of the water, g
is gravitational acceleration, and h is the depth of the water The water is still. (a) What is
the pressure (in Pa) at the bottom of a 10m-deep pool? (b) If we call the pressure
(normal stress) on the pool bottom σ 1 = σ 11, and the pressure (stress) that acts
horizontally σ 3 = σ 33 , then σ 11 = σ 33 . Plot the Mohr circle that describes the state of
stress at the bottom of our 10m-deep pool (this is an example of hydrostatic stress). (c)
What is the maximum shear stress at the bottom of the pool? (d) What must the level of the
intermediate principal stress σ 2 be at the bottom of the pool?
4 Two orthogonal planes have the following stresses acting on them: plane 1 (σ n = -9 MPa, τ =
-3 MPa); plane 2 (σ n = -1 MPa, τ = 3 MPa), where a negative normal stress is
compressive and a negative shear stress follows from the tensor convention. (a) Draw the
Mohr circle. (b) Calculate the greatest and least compressive stresses, and the maximum
shear stress. It should help to look at your Mohr circle while doing the calculations.
Remember that the mean normal stress is (σ 1 + σ 2 )/2, and the radius of the Mohr circle is
(σ 1 - σ 2 )/2. (c) Accurately and neatly draw a square showing σ 1 and σ 2 acting on the
sides of the square, and show the orientation of planes 1 and 2 within the square.
5 Find the τn and τs on a plane whose normal is 45° counterclockwise to the σ1 direction,
where σ 1 = -8 Mpa, and σ 2 = -3 MPa ; negative normal stresses are compressive.

GG454 Lecture 12 10/27/02 8
6 Find the magnitudes and orientations of σ1 and σ2 from known normal stress and shear
stress on two perpendicular planes P and Q. The known stresses acting on P and Q are:
Plane Normal stress Shear stress
P -2.2 kbar -1.1 kbar
Q -1.2 kbar 1.1 kabr
P -30kbar (a compression) -8 kbar
Q 0 kbar 8 kbar
Answer
σ1 σ2
Magnitude -34kbar (a compression) 2 kbar (a tension)
Orientation 14° clockwise from normal to 14° clockwise from normal to
plane P plane Q
P -28kbar (a compression) - 5 kbar
Q -2 kbar (a compression) 5 kabr
9 Consider the stresses at the surface of the earth in Kansas. Suppose the most compressive
horizontal stress trends N30°E and has a magnitude of 20 Pa (i.e., the most compressive
horizontal stress is -20 Pa). Ignoring the weight of the atmosphere, what is the vertical
normal stress at the earth's surface? Is this vertical normal stress a principal stress?
Why or why not? What would be the normal and shear stresses acting on a plane that
strikes N60W and dips 18.5° NE? What would be the stress on a plane that strikes 120°
and dips 26.5° SW? It might help to note that 18.5° ≈ 0.5*{tan-1 (3/4)} and 26.5° ≈
0.5*{tan-1(4/3)}
1 0 A rock is being prepared for a compression test, where one principal stress is -30 MPa
and the other is –10 MPa, where negative stresses are compressive. Suppose the shear
failure criterion for the rock is given by the following equation: |τ s| = |τ n|(tan 30°). Plot
the failure envelope and the Mohr circle corresponding to the state of the stress in the rock.
Can the rock sustain the indicated principal stresses without failing in shear? If the rock
will fail in shear, give the orientation of the shear failure plane(s) and the normal and
shear stress on the failure plane. If the rock will not fail in shear, give the normal and
shear stress on a plane whose normal makes a 60° angle with the σ1 axis.
1 1 A rock is being prepared for a compression test, with the principal stresses to be applied
being -40 MPa (the greatest compression) and -20 MPa (the least compression). Suppose
the shear failure criterion for the rock is given by the following equation: |τ s| = |σ n|(tan
30°). What must be the fluid pressure in the rock if it is to fail in shear, assuming that
the fluid pressure reduces the effective normal stresses by an amount equal to the fluid
pressure?

GG 454 February 8, 2002 1
CHARACTERIZING EARTHQUAKE SOURCES (13)
I Main Topics
A Elastic rebound theory
B Slip on a fault with a uniform stress drop
C Seismic moment
D Energy release during an earthquake
I I Elastic rebound theory (H.F. Reid, 1908, v. 2 of 1906 Earthquake report)
A Founded by comparing pre- and post-quake survey lines across SAF
B Seismic energy source: elastic potential energy of rock around fault
GEODETIC OBERVATIONS SUPPORTING REID'S ELASTIC REBOUND THEORY
SAN ANDREAS FAULT
st-quake
1906 po
1906
pre-q
1874- uake
1892
1851-1865
"1806"
1 6m slip on fault in 1906

2 Straight-line survey
in 1851-1865
3 ~3m far-field relative displacement
in ~50 years from 1851/1865 - 1906.
3m
This is half the 1906 slip.
4 ~2m far-field relative displacement
from 1874/1892 - 1906. 6 km
5 3m of slip probably
accumulated before
1851/1865.
6 Last quake 100 (= 50+50) years ago?

III Slip (∆ u) on a fault with a uniform shear stress drop (∆τ = τ1 − τ2 )

A Rock is elastic, homogenous, isotropic, isothermal material
B Shear stress on fault prior to slip = τ 1 ; post-slip shear stress = τ 2
C Slip profile is related to the shape and size of the rupture
D Slip distribution is particularly sensitive to the short dimension
E For a "2-D" rupture (one dimension >> other dimension = 2a)
FAULT GEOMETRY AND FAULT RUPTURE GEOMETRY
Long dimension
of rupture >> 2a
y
2a X
FAULT RUPTURE
FAULT

∆ u ave
∆u
-a 0 a
Distance (x) from rupture center
1 ∆u = 2(1-ν) ( ∆τ/µ ) (a2 -x2 ) 1/2 ν = Poisson's ratio

2 ∆ u max = ∆ u(x = 0) = 2(1-ν ) ( ∆τ/µ ) a ≈ 3 x 10-4 a
3 ∆ u min = ∆ u(x = ±a) = 0
4 ∆ u ave = (∫∆ u dx)/2a = (π /4)( ∆ u max )
IV Seismic moment M o
A M o = µ ∆uave A µ = shear modulus; A = rupture area
B M o has dimensions of energy; measures deformation
C A larger area of rupture or a larger slip ⇒ larger earthquake
D µ has been measured; geologists can estimate ∆uave and A
E Seismic moments (“earthquake size) M o can be predicted
V Radiated seismic kinetic energy Es (From Scholz, 1990)
Es + ∆E strain + ∆E friction + ∆E fracture surface + ∆E chemical ? = 0
A Kinetic energy (Es ) in seismic waves from a dynamics viewpoint
1 E s varies with amplitude and L wavelength
2 Waves of a frequency of zero ("ultra-long wavelength")
correspond to a static situation (“permanent” deformation).
3 Seismic moment, which describes the “permanent” deformation
after an earthquake, should be related to seismic energy release.
4 Empirical relationship of seismic energy (E s ) to magnitude (Mw ):
E s (joules)= 10(4.8 + 1.5 Mw )
B Strain energy (∆ E strain )
1 Energy in a linear spring
a Force in a spring (F):F=kx; k=spring constant, x=displacement
b Strain energy (∆ E strain ) equals area under a force-disp. curve:
[ ]
x x x
∫ ∫
∆Estrain = F dx = kx dx = kx 1 = 1 kx 2
2
0 0 2 0 2
∆ E strain = ("Average force on spring")(total displacement of spring)

FORCE-DISPLACEMENT PLOT FOR A LINEAR SPRING

kx
k = slope
Force
kx/2 dE
Elastic
Energy
0 dxDisplacement x
2 Energy of deformation (∆ E strain ) in an earthquake (Method of Reid)
a The change in strain energy = work of faulting
b Consider energy needed to restore rock to pre-quake conditions
c ∆ E strain = (1/2)(Peak shear stress)(Area of rupture)(slip)
(If the shear stress after slip occurred is zero, the relevant
average stress -τ is half the maximum stress. This is where
the factor of 1/2 comes from.)
d ∆ E strain = -τ A ∆ u ave
Note the similarity between this and M o (eq. IV.A)
e τ = (Shear strain)(Shear modulus of rock)
f Example: San Francisco, 1906
τ = (1/1500)(2 x 1010 J/m 2 ) = 1.33 x 107 J/m 2 )
∆ E strain = (107 N/m2 )(20x10 3 m)(435x10 3 m)(1/2 x 4 m)
≈ 1017 J
For comparison, ∆ E Bikini, 1946 = 1012 J
C Heat due to friction ( ∆ E friction )
1 ∆ E friction = ([τ friction ] [A] [∆ u ave ] )
2 Assuming τ friction = τ 2 , ∆ E friction = [τ 2 ] [∆ u ave ] [A]

D Energy in seismic waves (Es ) from a mechanics viewpoint

1 Es + ∆E strain + ∆E friction + ∆E fracture surface = 0
The strain energy in the earth decreases after a quake, so
∆ E strain <0. Energy appears in the form of heat, so ∆ E friction >0.
The energy to create fracture surfaces is assumed to be negligible.

2 Es ≈ −( ∆E strain ) − ( ∆ E friction )
3 Es ≈ (1/2[ τ1 + τ2 ] [∆u ave ] [A]) - (1/2)(2)([ τ 2 ] [∆ u ave ] [A])??
4 Es ≈ (1/2[ τ1 - τ2 ] [∆u ave ] [A]) = (1/2[ ∆τ ] [∆ u ave ] [A])??
5 So the seismic kinetic energy depends on the strength change on
the fault ∆ τ .
VI Formulas relating seismic energy release (Es ), moment (M o ), and
moment magnitude (Mw )
A) Mw ≈ 2/3 log M o - 6.067 M o in Nm
B) E s ≈ 10(4.8 + 1.5 Ms ) E s in joules
C) E s ≈ M o /20,000] E s in joules, M o in Nm
D) log Es ≈ log M o - 4.3 E s in joules, M o in Nm
E) log M o ≈ 1.5 Mw + 9.1 M o in Nm

Energy Release vs. Magnitude

The empirical relationship between energy content in radiated seismic
waves and magnitude is (Richter, 1958; Bolt, 1989):
E s (joules) = 10(4.8 + 1.5 Ms ) .
M s is the surface wave magnitude.
Consider two earthquakes, where Ms1 = 1 + Ms2 . Then

E s1 /E s 2 = 10(4.8 + 1.5 [Ms2 +1])/10 (4.8 + 1.5 Ms2 )
= {(104.8 )(10 1.5 [Ms2 ] )(10 1.5 )}/{(10 4.8 )(10 1.5 Ms2) }
= 101.5 ≈ 31.6
A unit increase in magnitude corresponds to (a) a factor of 10 increase in
amplitude of shaking, and (b) a factor of 31.6 increase in energy release.
One magnitude 8 quake releases the energy of 1000 magnitude 6 quakes.
__________________________________________________________
Relationship between moment magnitude (Mw ) and seismic moment (M o )
M w = 2/3 log M o - 6.067, where M o is measured in Nm p. 249 of Bolt
This empirical relation has been setup to dovetail with the surface wave
magnitude (i.e. Ms = Mw ). What is the relationship between M o and Es ?
Es = 10(4.8 + 1.5 Ms ) where E s is in joules p. 179 of Scholz
= 10(4.8 + 1.5 [2/3 log M o - 6.067])

= 10(4.8 + log M o - [1.5][6.067])
= 10(4.8 + log M o - 9.1)
= 10(log M o - 4.3)
= [10 log M o][10 -4.3 ]
Es = M o /20,000]
Alternatively, Es = [∆σ/2 ][∆u ave ][A] p. 165 of Scholz

Mo = [µ][∆ u ave ][A]
E s/M o = {[∆σ/2µ ][ ∆u ave ][A]} / {[µ ][∆u ave ][A]} = [∆σ/2µ ] p. 179 of Scholz
Typically ∆σ ≈ 3 MPa, and 2µ = (2)(3 x 104 MPa), so Es /M o = 1/20,000 √

SEISMIC SHAKING, TSUNAMIS, AND GROUND RESPONSE (14)
I Main Topics
A Recognition of direct effects of shaking as a seismic hazard
B Recognition of tsunamis and seiches as seismic hazards
C Recognition of types of shaking-induced hazards
http://earthquake.usgs.gov/faq//effects.html
I I Recognition of direct effects of shaking as a seismic hazard
A Shaking (and tsunamis) the most widespread seismic hazards
B Shaking generally causes the most damage and loss of life
C Unlike ground rupture, shaking is very difficult to avoid
D Shaking can be engineered against to a large extent
E Key factors controlling response of near-surface materials:
1 Thickness of Holocene/Quaternary deposits
2 Degree of consolidation (void ratio)
3 Degree of cementation
4 Depth to water table (some dispute here)
III Tsunamis and seiches
A Tsunamis: seismic sea waves generated by sharp vertical
displacement of the ocean floor
1 Can reflect dip-slip faulting or landsliding
2 Long-wavelength ("shallow-water") waves
3 Examples: Hilo, Hawaii, 1946; Lituya Bay, Alaska, 1958 (518m!)
http://www.tsunami.org/
http://www.geocities.com/CapeCanaveral/Lab/1029/Tsunami1958LituyaB.html
B Seiches: Shaking-induced oscillations of enclosed water bodies,
Most destructive when strong shaking is at resonant frequency.

IV Types of shaking-induced ground failures

A Settlement of cohesionless (sandy) alluvial materials
1 Vertical contraction resulting from loss of void space
2 Example: Homer, Alaska, 1964
B Liquefaction of cohesionless (sandy) alluvial materials
http://www.ce.washington.edu/~liquefaction/html/content.html
1 Loss of shear strength due to shaking; liquids have essentially no
shear strength
2 Sands and silty sands most susceptible to liquefaction are:
a Well-sorted
b Very young (<100 years)
c Have a high water table (< 3 meters below surface)
d River deposits susceptible
3 Examples: Niigata, Japan, 1964
C Ground oscillation
a Occurrence of large surface waves, without permanent lateral
translation, as a surficial layer that overlies a liquefied layer
fractures and blocks bob up and down. Sandblows form.
http://www.gi.alaska.edu/ScienceForum/ASF3/323.html
b San Francisco, California, 1906
D Lateral spreading (slopes <3°)
http://geohazards.cr.usgs.gov/wa/uwmeeting2.html
a Horizontal translation of surficial blocks in response to
liquefaction of underlying subhorizontal layers
b Example: Anchorage, Alaska, 1964
E Flow failure (slopes >3°)
a Translation of surficial blocks in response to liquefaction of
underlying inclined layers (Same mechanism as lateral spreading)
b Examples: Valdez, Alaska, 1964; Kansu Province, China, 1920
F Failure of quick clays
a Collapse of open clay structure and loss of shear strength
b Example: Turnagin Heights, Anchorage, Alaska, 1964
http://www.iris.washington.edu/EandO/slidescans/alaska/slideshow/
G Triggering of deep-seated slides
a Mechanism not understood
b Example: Santa Cruz Mtns, California, 1989

GG 454 March 11, 2002 1
RECOGNITION OF AREAS OF POTENTIAL SLOPE FAILURE (25)
I Main Topics
A Motivation
B Signs of slope instability
http://anaheim-landslide.com/
I I Motivation
A Most landslides (or potential slides) can be recognized or predicted
if the proper investigations are performed in advance.
B Except for small slides, slides typically cost far less to prevent
than to fix (An ounce of prevention is worth a pond of cure...).
C Massive slides may cost many times the cost of the desired facility;
they should be avoided.
III Signs of past slope instability (focus on topography and hydrology)
A Review of maps (Figs. 4.5, 3.9 in SR 176)
B Review of aerial photos & remote sensing imagery in SR 176
1 Hummocky topography
http://www.kgs.ukans.edu/Publications/pic13/pic13_2.html
a Oblique aerial photos: Figs. 2.19, 2,22, 2.23, 2.28, 2.33
b Stereophotos: Figs. 3.14, 3.26, 3.31
2 Arcuate fractures and old scarps or scars
http://anaheim-landslide.com/features.htm
a Oblique aerial photos: Figs. 2.19, 3.40,
b Stereophotos: Figs. 3.5, 3.7, 3.14, 3.19, 3.20, 3.24
3 Areas of anomalous vegetation (seasonal photos useful)
a Oblique aerial photos: Plate 3.3
a Stereophotos: Plates 3.1, 3.2
4 Areas of seepage: Vertical infrared imagery: Fig. 3.35
5 Sand flows and talus (Figs. 3.25, 3.30 )
C Features to examine in field reconnaissance
1 Ground cracks (Fig 4.2 in SR 176)
2 Closed hillside depressions
3 Phreatophytes and seeps
4 Patched or hummocky roads
5 Tilted walls, trees, and fences

GG 454 March 11, 2002 2
Characteristics of Real Landslides
Open cracks Topographic surface
Head scarp Failure Surface

(at depth)
Buckle at toe
Projection
of failure surface

GG 454 February 20, 2002 1
RESPONSE OF STRUCTURES (16)
I Main Topics
A Acceleration, velocity and displacement spectra
B Resonance and natural frequencies
C Response of structures
I I Displacement, velocity, and acceleration spectra
A Spectra represent parameters as a function of wave frequency (or
period), not time, and reveal the most energetic/forceful waves
B Examples of shaking vs. time and shaking vs. frequency plots
C Displacement, velocity, and acceleration spectra
y λ
x
A = amplitude
Displacement as a function of position along a wave
(1) y(x,t) = A sin {(2π / λ )(x + vt)}where v = λ /T = fλ , so λ = v/f
(2) y(x,t) = A sin {(2π f/v)(x + vt)}
Consider the displacement at a fixed point on the ground (e.g. x = 0)
( 3 ) y(t) = A sin {(2π f / v ) ( v t ) } or y = A sin (2π f t )
Displacement given as a function of the frequency and wave amplitude
Let ω = 2π f = angular frequency
( 4 ) y = [A] sin (ωt ) The term in square brackets gives ym a x
The shaking velocity of the ground (not the velocity of the wave) = dy/dt
( 5 ) y' = d(A sin (ω t))/dt = [ ω A] cos (ω t )
The acceleration of the ground = dy'/dt
( 6 ) y'' = d(ω A cos (ω t))/dt = -[ω 2 A] sin (ω t) = -ω 2 y
Note that (|y''/y|)1/2 = ω . For a given frequency (|y''/y|)1/2 = constant.

GG 454 February 20, 2002 2
Velocity and Acceleration vs. Frequency (Displacement is constant)

Frequency (f) 0.01* 0.1* 1* 10* 100
Hz
(ω ) 0.06 0.6 6 63 628
Hz
Maximum 10- 3 10- 3 10- 3 10- 3 10- 3
Displacement (y)
(m)
Maximum Velocity 6 x 1 0 - 5 6x10- 4 6x10- 3 6x10- 2 6x10- 1
y' (= ω y)
(m/sec)
Max. Acceleration 3 9 x 1 0 - 5 3 9 x 1 0 - 4 3 9 x 1 0 - 3 3 9 x 1 0 - 2 3 9 x 1 0 - 1
y" (= |-ω 2 y|)
(m/sec 2 )
*Significant frequency for design of large engineering structures
C Max. accelerations (forces) commonly at 2-10 Hz (T=0.1-0.5 sec),

max. velocities (kinetic energy) at 0.5-2 Hz (T=0.5-2 sec), and max.
displacements at 0.006-0.5 Hz (T=2-160 sec)
1 High frequency (small period) waves: high amplitudes of
acceleration, small amplitudes of displacement
2 Low frequency (long period) waves: low amplitudes of
acceleration, large amplitudes of displacement
D Avoid structure designs that are sensitive to wavelengths where
seismic forces/energy is concentrated (i.e. avoid structure designs
with natural periods that match the natural period of the underlying
materials [or the source]).

GG 454 February 20, 2002 3
III Resonance and natural frequencies

A Resonance: vibration of large amplitude due to arrival of energy at a
particular frequency
B Natural frequency: The frequency at which a structure will resonate
C Natural frequency of a pendulum
Simple Pendulum
L = length of pendulum arm
m = mass at end of arm

1 Natural period: T = 2π (L/g) 1/2 (Can use a pendulum to measure g!)
2 Natural frequency: f = 1/T = (g/L)1/2 /(2 π)
3 Natural angular frequency: ω = 2π f = ( L / g ) 1 / 2
D Natural frequency of a mass on a spring (simple harmonic oscillator)
Spring (spring constant = k)
m
Damper
1 Natural period: T = 2 π (m/k) 1 / 2

2 Natural frequency: f = 1/T = (k/m)1/2 /(2 π)
3 Natural angular frequency: ω = 2π f = ( k / m ) 1 / 2
IV Response of structures
Building More
m Simplest m1
Mechanical Complicated
Idealization Mechanical
m2
Idealization
A To first-order, a structure can be idealized as one mass on a spring

B Natural period: T = 2π(m/k)1/2 , where m = mass and k = stiffness
C k = f(building dimension, shape, elasticity)
D Rule of thumb: T (seconds) ≈ # of stories/10

GG 454 February 20, 2002 4
E Example: In Mexico City, the high rise buildings that were the most
damaged were 10-30 story (100'-300' tall) high-rise buildings.
These were calculated to have resonant period of 1-3 sec.
F Real buildings are sensitive to several different wave frequencies,
but will tend to be most sensitive to one frequency

GG 454 February 20, 2002 5
Response Spectra for a Hypothetical Earthquake

450
Displacement (cm)
Velocity (cm/sec)
400 Acceleration (cm/sec2)
350
300 ω = 2π /T, where T = period
250
200
150
factor of ω
100
50
factor of ω
0
0 0.5 1 1.5 2 2.5 3
Period (seconds)

GG 454 February 20, 2002 6
IDEALIZED UNDAMPED VELOCITY SPECTRUM CURVES
VELOCITY RESPONSE (M/SEC) 0.6
40 km from epicenter
of large quake
0.4

0.2
of large quake
of small (M=5.3) quake
0.0
0 1 2 3
PERIOD (SEC)
GROUND RESPONSE = f(SOURCE, PATH)

PATH = DISTANCE, GEOLOGY (ESPECIALLY NEAR-SURFACE GEOLOGY)
Source and distance effects highlighted here
A SMALL, NEARBY EARTHQUAKE MAY AFFECT SHORT-PERIOD STRUCTURES
MORE THAN A LARGE, DISTANT QUAKE
A LARGE, DISTANT EARTHQUAKE MAY AFFECT LONG-PERIOD STRUCTURES

MORE THAN A SMALL, NEARBY QUAKE
From Housner, 1970, in Wiegel, 1970)

PALEOSEISMOLOGY: SITES (17)
Schedule Updates and Reminders:

Reading for this lecture: Big Picture - Skim "Applications" in PP 1360
Reading for next lecture: Handouts from Active Tectonics, statistics book
Questions from last time?
I Main Topics
A Reasons for investigating the prehistoric record
B Characterization objectives of a paleoseismology investigation
C Reason for targeting paleoseismology investigations
D Criteria for siting paleoseismology investigations
I I Reasons for investigating the prehistoric record
A Nearly all ruptures occur along pre-existing faults
B Historical record is too short to rely on entirely (e.g. historic
inactivity is a poor way to gauge seismic hazard and risk)
C Past is key to future
III Characterization objectives of a paleoseismology investigation
A Location of fault(s)
B Geometry of fault (including segmentation)
1 Behavior of fault may hinge on its geometry
2 Most large faults do not rupture end-to-end during an earthquake.
3 Rupture in a given quake commonly confined to certain segments
of a fault. Geomorphology and geometry of a given segment may
be uniform, implying consistent behavior over large time frames.
4 Significance: Segment geometry controls size of earthquake
{Mo = (µ)(average slip)(Area of rupture)}
C Measurement of slip
1 Relative displacement (slip): a vector with magnitude & direction
2 Requirements to determine slip
a Orientation (and usually position) of fault
b Continuity of feature across fault (before and after offset)
c Piercing points (linear features, problems with streams)
d Piercing "curve" (e.g. cinder cone); not piercing planes

D Bracketing the timing of slip

1 Ages of offset units usually measured, not dates of earthquakes
2 Age of units above and below
3 Need to have time frame in mind to collect appropriate material
for dating and to apply the proper technique
4 Some radiometric techniques for Quaternary dating
14 C, K-Ar, 39 Ar- 40 Ar, U-Th, 36 Cl,
Amino acid racemization, Desert varnish
5 Relative dating techniques
a Fossils
b Geomorphic/weathering effects
6 Scarp degradation techniques
E Estimatation of slip rate(s)
Average slip rate at fault = Slip change/time interval
F Measurement of slip/event
G Estimation of recurrence intervals (Time between consecutive
events)
a Date (bracket) individual events
(Uncertainties with dates, missed events)
b Divide slip rate by slip/event
(Uncertainties with variation in slip rate and slip-per-event)
H Estimation of "size" of past quakes
a Intensity/Liquefaction/Ground response
b Amount of slip per event*
IV Reason for targeting paleoseismology investigations: One can't get the
desired information in most places along a fault!
V Criteria for siting paleoseismology investigations
(Focus on faults with surface rupture [Mw ≥ ~6])
A Fault must be located
B Simple fault expression
C Marker units (preferably several)
D Young materials (to allow a record of recent behavior)
E Datable materials (Dating techniques)
F Need to understand the geomorphology at the site
(Interaction of erosional, depositional, and faulting processes)

Useful References
Allen, C.R., 1968, The tectonic environments of seismically active and
inactive areas along the San Andreas fault system: Conference on
Geological Problems of the San Andreas fault system, Proceedings,
Stanford University Publications in the Geological Sciences, v. 11, p.
70-82,
Wallace, R.E., 1970, Earthquake recurrence intervals on the San Andreas
faul California: Geological Society of America Bulletin, v. 81, p. 2875-
2890.
Wallace, R.E., 1977, Profiles and ages of young fault scarps, north-central
Nevada: Geological Society of America Bulletin, v. 88, p. 1267-1281.
Sieh, K., 1978, Prehistoric earthquakes produced by slip on the San
Andreas fault at Pallet Creek: Journal of Geophysical Research, v. 83, p.
3907-3939.
Sieh, K., 1984, Lateral offsets and revised dates of large prehistoric
earthquakes at Pallet Creek: Journal of Geophysical Research, v. 89, p.
7641-7670.
Sieh, K., and Jahns, R.J., 1984, Holocene activity of the San Andreas fault
Wallace Creek, California: Geological Society of America Bulletin, v.
95, p. 883-896.
Schwartz, D., and Coppersmith, K., 1984, Fault behavior and characteristic
earthquakes: examples from the Wasatch and San Andreas faults:
Journal of Geophysical Research, v. 89, p. 5681-5698.
Lubetkin, L.K.C., and Clark, M.M., 1988, Holocene activity along the Lone
Pine fault, eastern California: Geological Society of America Bulletin,
v. 100, p. 755-766.
National Academy of Sciences, 1986, Active Tectonics, National Academy
Press, 266 p.
Earthquake Prediction, 1981, Earthquake prediction: an international
review: Simpson, D.W., and Richards, T.G., eds., Maurice Ewing Series 4,
American Geophysical Union, Washington, D.C.,680 p.
Working group on California earthquake probabilities, 1990, Probabilities
of large earthquakes in the San Francisco Bay region: U.S. Geological
Survey Circular 1053, 51 p.

HYPOTHETICAL TRENCH ACROSS THE WAIOWAIMI FAULT
Modern soil
Red clay A (Modern soil)

Blue clay
B (Red clay)
Gravelly sand
C (Blue clay
Buried A horizon
D (Gravelly sand)
Buried B horizon
Buried C horizon
E (Gravel)
Silt
F (Buried A horizon)
Silty Sand
G (Buried B horizon)
H (Buried C horizon)
I (Silt)
Bottom of Trench J (Silty sand)

Geologic History: The upper grey package is offset more than the
lower grey package, so there have been at least two faulting
events. One occurred after the formation of F and before depo-
sition of C; the other after deposition of B. The gravelly sand is
thicker on the right than on the left; either D was deposited over
an old fault scarp or some faulting occurred while D was deposited.

HYPOTHETICAL TRENCH ACROSS THE WAIOWAIMI FAULT
Modern soil
Red clay A (Modern soil)

Blue clay
B (Red clay)
Gravelly sand
C (Blue clay
Buried A horizon
D (Gravelly sand)
Buried B horizon
Buried C horizon
E (Gravel)
Silt
F (Buried A horizon)
Silty Sand
G (Buried B horizon)
H (Buried C horizon)
I (Silt)
Bottom of Trench J (Silty sand)

Geologic History: The upper grey package is offset more than the
lower grey package, so there have been at least two faulting
events. One occurred after the formation of F and before depo-
sition of C; the other after deposition of B. The gravelly sand is
thicker on the right than on the left; either D was deposited over
an old fault scarp or some faulting occurred while D was deposited.

GG 454 February 27, 2002 1
RECURRENCE INTERVALS AND PROBABILITY (18)
I Main Topics
A Recurrence intervals
B Simple empirical earthquake recurrence models
C Seismic gaps
D Probability distributions
http://www.seismo.berkeley.edu/seismo/hayward/probabilities_new.html
http://quake.usgs.gov/prepare/ncep/
E Exercise on probability of "The Big One" in So. Cal. in next 30 years
F Recognition, Characterization, Risk Evaluation, Risk Assessment
I I Recurrence interval:
A Used to evaluate when an earthquake is likely to occur
B Recurrence interval = time between consecutive earthquakes
(usually with reference to earthquakes of a given magnitude)
C Can be determined by geologic means
1 Dating of individual events (e.g. data from trench study)
2 Average recurrence int. = Average slip per event/average slip rate
III Simple empirical earthquake recurrence models
A Characteristic Earthquake Model
1 Same rupture length and slip distribution (and seismic moment)
2 Recurrence interval can vary through time
B Constant slip rate (time-predictable) Earthquake Model
1 Slip rate across fault is constant
2 Recurrence interval depends on slip during earthquake
C Random (Poisson) Model
1 Historical record too short to separate any patterns from "noise"
2 Earthquakes might best be considered as random events in time
D Problems with resolving dates of events
IV Seismic gaps
A Used to evaluate where an earthquake is likely to occur
B Along an active fault, the probability of an earthquake occurring is
~highest where the most time has elapsed since the last rupture

GG 454 February 27, 2002 2
Cummulative Slip on Fault Cummulative Slip on Fault Cummulative Slip on Fault COMPARISON OF THREE EARTHQUAKE MODELS
CHARACTERISTIC EARTHQUAKE
((Slip per event is constant;
Time between events can vary)
Earthquake
Time
CONSTANT SLIP RATE MODEL
(Long-term slip rate is constant;
Slip per event can vary)
Earthquake
Time
POISSON MODEL
(Slip per event and
time between events
is random)
Earthquake Time

GG 454 February 27, 2002 3
V Probability distributions
A Probability density functions [PDF = f(x)]: general comments
1 Probability (a<X<b) = probability of an outcome between a and b
b
= P(a<X<b) = area under f(x) from a to b = ∫ f ( x) dx
a
f is discontinuous f is continuous
f(x) f(x)
x
a b a b
a Example 1: Probability of Micheal Jordan scoring 25-35 points
b Example 2: Probability of quake (Mw = 7.5) in next 25-35 years
∞
2 (P(-∞<X<∞) = area under f(x) from -∞ to ∞ = ∫ f ( x) dx = 1 =100%
−∞
3 For continuous distributions, P(x=a) = area under f(x) from a to a
a
= ∫ f ( x) dx = 0
a
B The normal distribution ("The bell-shaped curve): one kind of PDF
1 Described by mean µ and standard deviation σ
n n n
∑ ( xi − µ )
2
∑ x i ∑ f i xi
i =1 i =1 i =1
µ = = σ =
n n n −1
∑ fi
i =1
2 P(µ−σ <x<µ+σ) ≈ 2/3; P(µ−2σ <x<µ+2σ) ≈95% P(µ−3σ<x<µ+3σ) ≈99%
VI Exercise on probability of "The Big One" in So. Cal. in next 30 years
VII Recognition, Characterization, Risk Evaluation, Risk Assessment
A Probabilistic assessment allows the likelihood of given effects (e.g.
intensities), and hence potential damages, to be estimated for a
given area for a given time frame. This is what is meant by
evaluating the level of risk.
B Steps 1 and 2 must be done in order to get to step 3 (and then 4)
C Outcome probabilities are sensitive to the model one chooses
D This approach can be (and has been) applied to many phenomena

GG 454 February 27, 2002 4
Probability distribution curves for three earthquake models

Characteristic Earthquake model
Probability density
Cummulative slip
Only quakes of
magnitude M x
occur P(M=M x)
Slip during event of

magnitude M x
0
Time 0 Elapsed time
Time-predictable model
P(M=2) P(M=4)
Probability density
Slip in one quake increases P(M=1) P(M=3) P(M=5)
Cummulative slip
with the time elapsed

since last quake
0
Time 0 Elapsed time
Random model
P = probability of an
Probability density
earthquake of any
Cummulative slip
magnitude; P does
not depend on
elapsed time.
Earthquakes of any size

can occur at any time.
Poisson Distribution
0
Time 0 Time interval

GG 454 February 27, 2002 5
Method for Predicting the Annual Likelihood of "The Big One" at Pallet Creek
Kerry Sieh, a professor at Cal Tech, has done more than any other single person to document the
hazard presented by recurring large earthquakes on the San Andreas fault in Southern
California. We will use some of Kerry's results to estimate the probability of a large
earthquake on the San Andreas fault in Southern California.
Here are Kerry's estimates (from his 1984 JGR paper) on the time of the last 12 large
earthquakes at Pallet Creek (the uncertainties associated with these events are dropped):
1857, 1720, 1550, 1350, 1080, 1015, 935, 845, 735, 590, 350, 260
1 Based on the time between the oldest event listed above and the 1857 quake, calculate the
average (mean) recurrence interval for large earthquakes at Pallet Creek.
Mean Recurrence Interval = (1857-260)years/11 intervals = 1597 yr/11 = 145 years
2 The earthquakes are not occurring at a perfectly regular pace. Calculate the recurrence
times between each successive pair of earthquakes.
137, 170, 200, 270, 65, 80, 90, 110, 145, 240, 90
3 Calculate the standard deviation of the 11 recurrence intervals associated with the 12
quakes. The equation to use is:
n
∑ (Ri - R*)2
σ= i=1
n-1
where σ is the standard deviation, Ri is the recurrence time between a given pair of events,
R* is the mean recurrence interval, and n is the number of recurrence intervals (not the
# of quakes!).
68 years

GG 454 February 27, 2002 6
4a Assuming the year is 1993, how many years have elapsed since the last large San Andreas
earthquake in southern California?
1993-1857 = 136 years
4 b How many years shy of the mean recurrence interval would we be?
145-136 = 9 years
4c How many standard deviations shy of the mean recurrence interval would we be?
9 years/68 years = 0.13 standard deviations
5 We will now suppose the distribution of recurrence intervals is normally distributed about
the mean recurrence interval. On the supplied paper, plot the equation
-(t - t* )2
f(t)= 1 exp
σ 2π 2σ2
where f(t) is the normal distribution, t is time, t* is the mean, and σ is the standard
deviation.
Plot this for 0 ≤ t ≤ 250 years.
6 What is the probability of an earthquake on the San Andreas fault at Pallet Creek in the next
30 years from 1993 given our model? This probability is the area under the curve from
1993 to 30 years hence divided by the area from 1993 to infinity.
Suppose the year is 1993 - 9 years (0.13 standard deviations) shy of the mean recurrence
interval. In 30 years we would be 21 years (or 21/68 = 0.31 standard deviations) past
the mean recurrence interval. The area under the probability density curve from the
mean to 0.13 standard deviations shy of the mean is 0.0517. The area under the
probability density curve from the mean to 0.31 standard deviations past the mean is
0.1217. The area under the probability density curve from 0.13 standard deviations shy
of the mean to ∞ is 0.5 + 0.0517. So:
P = (0.0517 + 0.1217) / (0.5 + 0.0517) = 0.1734/0.5517 = 31%
Even though Kerry doesn't think he missed a quake, suppose there were circumstantial evidence
(e.g. Indian legends) for one large quake in the year 490 and another in 1215.
7 What would the new mean recurrence interval and standard deviation be?

GG 454 February 27, 2002 7
New mean recurrence interval = 1597 years/13 intervals = 122.8 years =123 years
New standard deviation = 38 years (larger % change in standard deviation than in mean!)
8 Assuming the year is 1993, what would the recalculated probability be for a large quake at
Pallet Creek in the next 30 years?
The year 1993 would be 13 years (or 13/38 = 0.34 standard deviations) past the mean
recurrence interval. In 30 years we would be 43 years (or 43/38 = 1.13 standard
deviations) past the mean recurrence interval. The area under the probability density
curve from the mean to 0.34 standard deviations from the mean is 0.1331. The area under
the probability density curve from the mean to 1.13 standard deviations past the mean is
0.3708. The area under the probability density curve from 0.34 standard deviations past
the mean to ∞ is (0.5 - 0.1331). So:
P = (0.3708 - 0.1331) / (0.5 - 0.1331) = 0.2377/0.3669 = 65%
Now suppose we consider the earthquakes to be distributed randomly (i.e. they are characterized
by a Poisson distribution). Then the probability of an earthquake occurring does not depend on
how much time has elapsed since the last earthquake. The probability of “x” number of
earthquakes occurring in a given interval of time t is given by:
(vt) x e −vt
P( x) =
x!
where “v” is the average rate of occurrence. So if the average recurrence interval is 145
years, the probability of getting 1 event in 145 years is:
1 event
−( 145 yrs)
1 event
( 145 yrs)1 e 145 yrs
145 yrs
P(1) = = e-1 = 37%
1!
The probability of getting one event in 30 years is:
1 event
−( 30 yrs)
1 event
( 30 yrs)1 e 145 yrs
145 yrs
P (1) = = (30/145)(e - 3 0 / 1 4 5 )= 17%
1!
The probability of getting no event in 30 years is:
1 event
−( 30 yrs)
1 event 0 145 yrs
( 30 yrs) e
145 yrs
P (1) = = (e- 3 0 / 1 4 5 ) = 81%
0!

GG 454 February 27, 2002 8
-3
x 10
Probability of an Earthquake at Pallet Creek: Scenario A
6
MRI = 145.1818
σ = 68.0174
5
3 P = 0.3141
1993
0
1850 1900 1950 2000 2050 2100 2150 2200

GG 454 February 27, 2002 9
Probability of an Earthquake at Pallet Creek: Scenario B

0.012
0.01 MRI = 122.8462
σ = 38.1703
0.008
0.006
1993
P = 0.64643
0.004
0.002
0
1850 1900 1950 2000 2050 2100 2150 2200

GG 454 February 27, 2002 10
-3
x 10
Probability of an Earthquake at Pallet Creek: Scenario C
6
MRI = 145.1818
σ = 68.0174
5
3 P = 0.3403
2002
0
1850 1900 1950 2000 2050 2100 2150 2200

GG 454 February 27, 2002 11
Probability of an Earthquake at Pallet Creek: Scenario D

0.012
0.01 MRI = 122.8462
σ = 38.1703
0.008
0.006
2002
P = 0.69406
0.004
0.002
0
1850 1900 1950 2000 2050 2100 2150 2200

GG 454 March 8, 2002 1
DAM AND RESERVOIR FAILURES (19)
I Main Topics
A General comments
B Teton Dam
I I General comments
In the 1960s and 1970s, reservoirs impounded behind some of the
world’s tallest dams failed. The failures can be examined in terms of
(a) the physical conditions and processes that resulted in failure, and
(b) organizational failures - that is, a failure of the organizations that
constructed or oversaw the construction of the dams to adequately
recognize, characterize, evaluate, and asses the level of risk
associated with the conditions and processes at the dams. This lecture
addresses both issues.
III Teton Dam
http://www.lib.utah.edu/spc/photo/p211/p211.html
http://www.geol.ucsb.edu/~arthur/Teton%20Dam/narrative.html
http://www.rootsweb.com/~idfremon/flood.htm
On June 5, 1976, the Teton Dam, 44 miles northeast of Idaho Falls in

southeastern Idaho, failed. It was one of the world’s tallest earth-fill
dams (~89 m) at the time of the failure. Nearly 300,000 acre-feet of
water (or 80 billion gallons or ~300 x 106 m3 ) flooded farmland and
towns downstream. Damage estimates range around $1 billion and
fourteen people’s deaths were attributed to the dam failure. The
failure has been widely attributed to hydraulic fracturing of the dam
core by water that accessed the core via fractures. The failure can
also be attributed to a combination of (a) inadequate recognition,
characterization, and evaluation of the fractures in the foundation, and
(b) inadequate engineering of the foundation to seal the fractures.

Geo 454 March 3, 2002 1
MASS WASTING I ("LANDSLIDES") (20)
I Main Topics
A Emphases for Landslide Section of Course
B Case history 1: Nevados Huascaran, Peru, May 31, 1970
I I Emphases for Landslide Section of Course
A Hazard recognition (see also the tables)
1 Magnitude of mass wasting problem in U.S.
a Est. annual cost in U.S. as of 1978: $1 B (x 2 for 1991)
b Predicted losses in California (1970-2000):$10 B
c Portugese Bend (1956-1959): $10 M (x 5 for 1991)
d Utah (1983*): $250 million*
e S.F. Bay Area January1982 storm: 18,000 landslides!
2 Human activity as a contributing factor
a Contra Costa County, California: 80%
b Allegheny County, Pennsylvania: 90%
3 Improved geotechnical practices can greatly reduce slide costs:
a New York State (1969-1976): 90% reduction
b Los Angeles 1968-1969 storm: 97% reduction vs. pre-1952
c State of CA (Beach Leighton): 95-99% reduction
4 Key empirical factors relevant to mass wasting processes
a Geologic Factors
i Geomorphology: Topography; processes of erosion &
deposition and deformation that create topography
i i Composition: Rock types and weathering products
i i i Structure: Distribution of flaws (e.g., bedding and fractures)
and rock types; controls geometry of failure surface
i v Seismicity
v Geologic History (Time): Past is key to future
b Environmental Factors
i Climate and Hydrology: Rainfall, surface/ground water flow
i i Vegetation (Very sensitive to all factors above)
c Human Activity
i Deposition
i i Erosion
i i i Alteration of environmental factors
B Hazard characterization (and classification)
C Slide mechanics: how do empirical factors contribute mechanically?

Geo 454 March 3, 2002 2
I I I Case history: Nevados Huascaran, Peru, May 31, 1970
On May 31, 1970 the most catastrophic known avalanche in history
descended from Nevados Huascaran, the highest peak in the Peruvian
Andes. The avalanche cause ~18,00 casaulties, including 15,000 of
17,000 resident of the city of Yungay. The avalanche was triggered at
3:23 PM by a great earthquake (M=7.7) off the coast of Peru; the
epicenter was 130 km west of Nevados Huascaran. The avalanche
occurred at end of wet season, when snow cover was near a maximum
and the snow had begun to melt. The avalanche originated from a
partially overhanging cliff (average slope of 70-80°) at 5400-6500 m
elevation, where the fractured granitic rock of the peak was covered by
a 30-m-thick glacier. The avalanche had a volume of ~ 50-100 million
cubic meters and traveled 16 km to Rio Santa with a vertical drop of 4
km. The avalanche shot over a 230m-high ridge on its descent,
launching boulders as large as 60 metric tons as far as 4 km from the
launch point. The average speed of the 1970 avalanche was 280
km/sec– peak speeds probably reached ~1000km/hr. A previous
avalanche from the same peak in 1962, with a volume of 13 million
cubic meters, caused 4000 fatalities. That event prompted a prediction
that an avalanche even larger than the 1962 could fall from the face
and threaten Yungay. The hazard posed was thus recognized, partially
characterized, and partially evaluated. Mapping after the 1970 event
revealed deposits from an substantially larger pre-Columbian (pre
1492) avalanche that might have attained average speeds of 315-355
km/hr.

Geo 454 March 3, 2002 3
The 25 most catastrophic landslides of the 20th century
http://www.canadacollege.net/galloway/haz.9.html
Year Country Name and Type Triggering Impact

process
1911 Tadzhik Usoy rock slide Usoy 54 killed
Republic earthquake
M=7.4
1919 Indonesia Kalut lahra Kelut volcano 5,100 killed 104 villages
(Java) (volcanic mud (eruption) destroyed
flow)
1920 Ningxia Haiyuan landslide - ~100,000 killed
(China)
1921 Kazakh Alma-Ata debris Snow melt 500 killed
Republic flow
1933 Sichuan Deixi landslide Deixi 6,800 killed; 2,500 drowned
(China) earthquake when dam failed
M=7.5
1939 Hyogo Mount Rokko Rain 505 dead, 130,000 homes
(Japan) slide/mud flow (typhoon) destroyed
1949 Tadzhik Kahit rock slide Khait 12,000 to 20,000 killed; 33
Republic earthquake villages destroyed
M=7.5
1953 Wakayama Arita River Rain 460 dead, 4,722 homes
(Japan) slide/debris/mud (typhoon) destroyed
flow
1953 Kyoto Minamiyamashiro Rain 336 dead, 5,122 homes
(Japan) slides/debris/mud (typhoon) destroyed
flow
1958 Shizuoka Kanogawa Rain 1,094 dead; 19,754 homes
(Japan) slide/debris/mud (typhoon) destroyed
flow
1962 Ancash Nevados Huascaran - 4,000 to 5,000 killed most of
(Peru) debris avalanche Ranrahirca village destroyed
1963 Fruili- Vaiont rock slide Filling of ~2,000 killed; city of
Venezia- Vaiont Longarone damaged ~ $970
Griulia Reservoir million (1994 U.S.dollars)
(Italy)
1964 Alaska Alaska slides 1964 Estimated $860 million in
(United earthquake (1994 U.S. dollars)
States) M=9.4
1965 Yunnan Rock slide - 444 dead; 4 villages
(China)
1966 Rio de Rio de Janeiro Heavy rain ~1,000 dead
Janeiro slides/avalanches/
(Brazil) debris/mud flows
1967 Serra das Serra das Araras Heavy rain ~1,700 dead
Araras slides/avalanches/
(Brazil) debris/mud flows
1970 Ancash Nevados Huascaran Earthquake 18,000 dead; town of Yungay
(Peru) debris avalanche M=7.7 destroyed and Ranrahirca
partially destroyed

Geo 454 March 3, 2002 4
Year Country Name and Type Triggering Impact

process
1974 Huancavelia Mayunmarca Rainfall ~450 killed; Mayunmarca
(Peru) rock/slide/debris village destroyed failure of
/ 150-m-high landslide
avalanche dam caused major flooding
downstream
1980 Washington Mount St. Helens Eruption of World's largest historic
(United rockslide/ Mt. St. Helens landslide 23-km-long debris
States) debris avalanche avalanche with average
velocity of 125 km/hr;
surface remobilized into 95-
km long debris flow
1983 Utah Thistle debris Snow melt and Total losses ~$600 million in
(United slide heavy rain 1994 U.S. dollars
States)
1983 Gansu Saleshan landslide Loess landslide 237 dead; buried 4 villages
(China)
1985 Tolima Nevado del Ruiz Eruption of >20,000 dead; four towns
(Columbia) debris flows Nevado del destroyed; ~100-km long
Ruiz debris flow
1986 Papu New Bairaman rock Bairaman Village of Bairaman destroyed.
Guinea slide/avalanche earthquake Debris avalanche formed
M=7.1 210-m-high dam that
impounded 50-million cubic
meter lake; dam failed,
causing 100-m-deep debris
flow downstream
1987 Napo Reventador Reventodor ~1,000 killed $1.3 billion
(Ecuador) landslide earthquake (in 1994 U.S. dollars)
M=6.1 and
M=6.9
1994 Cauca Paez landslides Paez 271 dead; 1,700 missing;
(Columbia) earthquake 32,000 displaced; villages
M=6.4 destroyed
(Modified from Table 1, Schuster, R.L., 1996, in Landslide, Chacón, Irigaray and Fernádez, eds.
A.A. Balkema/Rotterdam

Geo 454 March 3, 2002 5
Reference Material
Special Report 176 (Chapters 1, 2, 3, 4, 5, 6, 7 in SR 176)
Reviews in Engineering Geology, volume III, 1977, Coates, D.R., ed.,
Geological Society of America, 278 p.
Landslide Dams: Processes, Risk, and Mitigation, 1986, Schuster, R.J., ed.,
Geotechnical Special Publication No. 3, American Society of Civil
Engineers, 164 p.
Fleming, R.W., and Johnson, A.M., 1990, Structures associated with strike-
slip faults that bound landslide elements: Jahns Memorial volume,
Journal of Engineering Geology, v. 27, p. 39-114.
Vonder Linden, K., 1990, The Portugese Bend Landslide: Jahns Memorial
volume, Journal of Engineering Geology, v. 27, p. 301-374.
Committee on Natural Disasters, 1984, The Utah Landslides, debris flows,
and floods of May and June 1983: National Academy Press, 96 p.
Rockslides and Avalanches, 1979, Voight, B., ed., Elsevier, N.Y., 2 volumes.
Crandell, D.R., Miller, C.D., Glicken, H.X., Christiansen, R.L., and Newhall,
C.G., 1984, Catastrophic debris avalanche from ancestral Mount Shasta
volcano, California: Geology, v. 12, p. 143-146.
Rahn, P., Engineering Geology.

GG 454 March 8, 2002 1
DAM AND RESERVOIR FAILURES: THE VAIONT LANDSLIDE (21)

http://www.sci.port.ac.uk/geology/staff/dpetley/imgs/enggeolprac/vaiont1.html
On October 9, 1963 a landslide with a volume of ~270 million m3 moved

into the reservoir behind the recently completed 265m-tall Vaiont dam in
Italy - the world’s tallest double thin arch dam. The mass moved
approximately 500 m at speeds as great as 30 m/sec. It displaced the
water in the reservoir (115 million m 3 of water prior to failure) and drove
a wave up the opposite side of the valley. The wave destroyed the village
of Casso, 260 m above lake level before over-topping the dam by as much
as 245 m. An estimated 30 million m3 of water then descended down the
valley as a wave initially more than 70 m tall, completely destroying the
villages of Longarone, Pirago, Villanova, Rivalta and Fae. About 2500
people were killed. The dam still stands today: the reservoir slopes
failed, but the dam did not. This disaster sparked renewed attention to
slope stability around the world. The walls of the valley of the reservoir
were particularly susceptible to large landslides: (a) the slopes were
steep (~28°), (b) the rock layers dip toward the valley floor, (c) the rock
consistent of highly fractured, clay-rich limestone with abundant
dissolution features, allowing a well-connected hydrologic system to
exist, and (d) the site was the scene of prior large slides in the geologic
past. The possibility of a landside into the reservoir was recognized in
advance but estimates of the maximum possible size of a slide
underestimated the size of the actual slide by two and one-half orders of
magnitude. Surveying measurements and water table (piezometric)
measurements made after the reservoir began to be filled showed that
changes in the reservoir level caused near simultaneous changes in the
water table in the valley walls, and that increases in the elevation of the
reservoir level corresponded to increased displacement rates of the slide
mass on the south valley wall. The failure of the dam can be attributed to
inadequate recognition, characterization, and evaluation of the conditions
of the slopes flanking the reservoir before and after the reservoir w a s
developed.

GG 454 March 8, 2002 1
CASE HISTORIES: TRANSLATIONAL LANDSLIDES (22)

I Main Topics
A Frank Landslide
B Love Creek Landslide
I I Frank Landslide
http://www.frankslide.com/boulder.html
http://www.frankslide.com/
http://cgrg.geog.uvic.ca/abstracts/ReadGeotechnicalThe.html
http://www.calgaryopera.com/NewFiles/TWrelease.html
The Frank rockslide occurred in the southern Canadian Rockies of

Alberta on 4:10 AM, April 29, 1903. The 30 million cubic meter rockslide
occurred on the east flank of Turtle Mountain. The side dimensions are
impressive: 150 m deep, 425 m long and 1 km wide. The slide destroyed
the southern end of the coal mine town of Frank and claimed 70 lives -
about one of every nine residents. The slide lasted 100 seconds, and
covered an area 3 km2 to an average depth of 14 m. The natural geologic
conditions at the Frank slide were highly conducive to sliding: slopes were
steep (~30°), precipitation in the years preceding the slide was relatively
high, and the bedding dipped parallel to the slope. The rock consists
mostly of highly fractured and faulted limestone. Bedding surfaces, and
pre-existing faults and fractures all were exploited during the slide
event. Human activity might also have played an important role in
triggering the slide: extraction of coal was at the base of the east slope
of Turtle Mountain weakened the slope to some degree, and explosives
used in the mining generated seismic waves. Miners in Frank apparently
did not recognize that the slope was prone to sliding or that their
activities might contribute to destabilization of the slope, although a rock
fall in 1853 had killed about 100 Blackfoot Indians during a battle. A
series of geologic investigations after the slide to characterize the
geologic structure at Turtle Mountain have yielded decidedly different
interpretations of the site geology. An accurate characterization of the
site certainly did not exist in 1903 (or for many years later). The lack of
recognition and characterization of the site precluded an adequate
evaIuation and assessment of the level of risk in 1903. Recent research
to characterize, and evaluate the level of risk at the site indicates that
the old coal mining activities will likely continue to destabilize the slope
as mine workings collapse due to rotting timbers, caving of open rooms,
and the burning of coal by the fire which cause the mine to close.

GG 454 March 8, 2002 2
I I I Love Creek Landslide

(Cotton, W.R., and Cochrane, D.A., 1982, Love Creek landslide disaster, Lanuary 5,
1982, California Geology, June issue, p. 153- 157).
In terms of lives lost (10), the most destructive naturally-occurring
landslide in the history of California occurred on January 5, 1982. On
January 3, the San Francisco Bay area was visited by a “500-year storm”
that dropped as much as 0.6 m of rain in the next 24-hours. The storm
triggered ~18,000 landslides in the San Francisco Bay area, caused 26
deaths and nearly 300 million dollars in damage. The most massive slide,
the Love Creek slide, happened at ~1:00 AM, shortly after the storm
cleared. The slide occurred on the west-facing slope of Love Creek, about
16 km (10 miles) north of the city of Santa Cruz. It was about 600m wide
(maximum), 250m long (maximum), and 10 m thick, with a volume of about
500,000 cubic meters. The slide and an accompanying debris flow dammed
Love Creek, forming a lake about 300 meters long that flooded several
homes. The landslide occurred on the northern flank of what has been
mapped as an ancient landslide. The rock involved consisted of fractured,
weathered sandstone and shale. The site is located on the west-dipping
limb of a synclinal fold, and the surface of sliding coincided with a
bedding plane, and locally with a thin clay layer. The January 5 slide
occurred in a region known for landslides, and at a site prone to sliding,
but the specific site was not recognized or characterized as having the
potential for a slide as large as the one that occurred. The slide was
judged to have removed enough support from the adjacent slope material
to justify the evacuation of nearly 30 homes adjacent to the slide.

Geo 454 March 11, 2002 1
DEBRIS FLOWS AND DEBRIS AVLANCHES (23)
I Main Topics
A Hazard recognition
B Debris flows on Mount Shasta, California
C Debris flows from Nevado del Ruiz, Colombia
D Debris flows & avalanches from Mount St. Helens, Washington
http://volcanoes.usgs.gov/Hazards/What/Lahars/RuizLahars.html
E Debris flows and debris avalanches in Hawaii
I I Hazard recognition
Debris flows are one of the most widespread types of mass wasting.
They are slurries that resemble flows of wet concrete. They are
particularly common where episodic pulses of water encounter
concentrations of loose rocky debris, such as stream channels in semi-
arid regions and on volcanoes. Alluvial fans are built largely by debris
flows. Debris avalanches commonly are associated with debris flows.
Iverson (1997) provides an outstanding description of the physics of

debris flows:
Iverson, R.M., 1997, The physics of debris flows: Reviews of Geophysics, v. 35, p. 245-
296.
I I I Debris flows on Mount Shasta, California

http://vulcan.wr.usgs.gov/Volcanoes/Shasta/description_shasta.html
I V Debris flows from Nevado del Ruiz, Colombia
http://volcanoes.usgs.gov/Hazards/What/Lahars/RuizLahars.html
A Casualty toll of ~23,000 people: largest toll from a mass wasting
event in the last 3/4 of a century
B Substantial damage at a run-out distance of100 km
C Occurred on a volcano
V Debris flows & avalanches from Mount St. Helens, Washington
http://vulcan.wr.usgs.gov/Volcanoes/MSH/framework.html
A Largest documented landslide in recorded history
B Debris avalanche generated devastating debris flows

Geo 454 March 11, 2002 2
V I Debris flows and debris avalanches in Hawaii
A Makaha Valley Towers 11/14/96
http://starbulletin.com/96/11/14/news/story1.html
http://the.honoluluadvertiser.com/2000/Nov/20/
B Submarine debris avalanches and debris flows
http://www.uhh.hawaii.edu/~kenhon/geol205/Landslides2/default.htm
Moore, J.G., Clague, D.A., Holcomb, R.T., Lipman, P.W., Normark, W.R., and Torresan,
M.E., 1989, Prodigious submarine landslides on the Hawaiian Ridge, Journal of
Geophysical Research, v. 94, p. 17,465-17,48.
1 Among the largest known mass wasting events in the solar
system
2 Nuuanu debris avalanche
a Volume: ~5,000 cubic kilometers
b Length: 235 km
c Run-up height at distal end: 300m+
d Minimum speed at base of Hawaiian Deep:
v = 2 gh = 2(10)( 300) m /sec = 80 m /sec = 280 km / hr

GG 454 March 11, 2002 1
CLASSIFICATION OF MASS WASTING PROCESSES (24)
I Main Topics: Varnes classification scheme (Fig. 2.1 of SR 176)

http://www.em.gov.bc.ca/Mining/Geolsurv/Surficial/landslid/ls2.htm
A Style of slope movements
B Types of geologic materials
C Landslide nomenclature
Varnes classification scheme

Process⇒ Sliding Toppling Fall Flow Avalanche
Material (complex)
Rock
“Coarse
soil”
(Debris)
“Fine
soil”
(Earth)
Ice analog Glaciers Glaciers
I I Style of slope movements

A Falls
1 Movement largely a free fall through air; accompanied by leaping,
bounding, bouncing, or rolling of material
2 Maximum rate of movement: ~102 m/sec
3 Example: rock falls from cliffs
B Topples
1 Movement is initially a rigid body tilting or overturning;
generally requires steep fractures parallel to free face
2 Maximum rate of movement: ~102 m/sec
3 Example: topples along banks of Mississippi River
C Slides
1 Material moves parallel to (and maintains contact with) one or
more surfaces or narrow zones of failure
2 Rate of movement: Highly variable (10-9 m/sec - 101 m/sec)
3 Two many types of slides
a Rotational slide (slump): slip surface curved in cross section
Pure rotational slides usually in uniform engineering material
b Translational slide: slip surface is roughly planar
4 Example: Manoa Valley slides; small failures along highway cuts
GG 454 March 11, 2002 2
D Lateral spreads
1 Material on a shallow slope is extended; there may or may not be
a well-defined failure surface
3 Commonly triggered by shock to young quick clays
4 Example: G street "slide" in Anchorage, 1964

GG 454 March 11, 2002 3
E Flows
1 Deformation is spread throughout the displaced material in a
relatively continuous fashion
3 Enormous variety of phenomena (e.g. bedrock flows, soil creep,
silt flows, dry sand flows, debris flows, debris avalanches)
4 Similarity between earthflows, lava flows, glaciers
F Complex
1 Movement is by a combination of the above main styles; this
implies gradations between styles
3 Examples: Elm, Switzerland event; Nevados Huascaran, Blackhawk
4 Includes many phenomena termed avalanches
III Types of geologic materials
1 Rock: material that requires blasting or heavy equipment to move
2 Debris: loose material that contains a significant portion of coarse
material (20%-80% > 2mm)
3 Earth: loose material that primarily consists of fine material
(80+% < 2mm)
IV Landslide nomenclature (see handout)

GG 454 March 11, 2002 4
TYPES OF SLOPE MOVEMENT PROCESSES
Cross-section views
Falls Topples
Slides Lateral spreads

Translational
Rotational
Weak layer (e.g. quick clay)
Flows Complex
Translational side
Fall
Flow

GG 454 March 19, 2002 1
EFFECTIVE STRESS AND MOHR-COULOMB FAILURE (26)
I Main Topics
A Driving and resisting stresses at the base of an inclined block
B Factor of safety
C Effective stress
D Mohr-Coulomb failure
I I Driving and resisting stresses at the based of a dry inclined block
F = F sin ß
s
F = F + F
n s
F = F cos ß
n F = mg
ß ß
A Driving stress: σ s = Fs / A
Slope-parallel component of block weight / block area
B Resisting stress
1 Shear strength = Shear strength of slip surface under no
normal stress + frictional resistance (which depends on Fn !)
2 τ = c + (Fn tan φ )/A
a c = cohesion
i Tape sticks to a vertical wall because of cohesion
i i Dry sand has virtually no cohesion
b tan φ = coefficient of friction = ∆τ / ∆σ n
c φ = angle of internal friction
τ φ
co
σ
n
Shear strength at the base of sliding block as a function of normal stress
(Compression taken as positive here!)

GG 454 March 19, 2002 2
III Factor of safety (F.S.)

Fresisting Fresisting / A σ resisting
1 F .S . = = =
Fdriving Fdriving / A σ driving
2 If resisting forces (stress) < driving forces (stresses) ⇒ f a i l u r e
3 If factor of safety < 1 ⇒ f a i l u r e
4 For dry cohesionless (sandy) soils
Factor of safety = {(F cos ß) (tan φ )}/ (F sin ß)
= tan φ / tan ß
Factor of safety < 1 if φ < ß (i.e. if angle of internal friction < slope)
IV Effective stress
A For soils: σ ' = σ total - P
1 σ ' = Effective stress = load supported by solid framework
2 σ total = total normal stress porous material is subjected to
3 P = pore pressure of pore fluid = load supported by fluid
B General case (For soil and rock): σ ' = σ total - α P
(Nur and Byerlee, 1971, JGR, v. 76, p. 6414-6419)
1 α = 1 - (Kagg /K s )
a K = Bulk modulus = 1/B = 1/compressibility =∆ pressure/( ∆ V/V)
b Low bulk modulus = high compressibility; low K = high B
c K agg = bulk modulus of dry aggregate;
d K s = bulk modulus of the solid component of the aggregate
2 α = 1 - (Bs /B agg )
3 For soils, Kagg<<Ks (or Bagg>>Bs) so α = 1
4 For some rocks α ≠ 1 (0 < α < 1)
C Pore pressure does not "lubricate" the failure surface; the pore
pressures acts to float the material overlying the failure surface
D Until the pore fluid has time to flow, increased loads on a saturated
"soil" are supported by an increase in the pore pressure.

GG 454 March 19, 2002 3
V Mohr-Coulomb failure
A τ failure = c + σ ' failure tanφ , where σ ' = effective normal stress
Failure
Envelope φ
(σ'f , fτ )
τ
(σ'f , fτ )
2θ 2θ
co
P1 P2 σn
( φ + 90°)/2 = θ If φ = 30°, then θ = 60°
B For soils: τ failure = c + (σ n -P) tanφ

GG 454 March 19, 2002 4
C Factor of safety for a tilted block accounting for pore pressure
Basal area = A H
Fresisting Fresisting / A σ resisting

F .S . = = =
Fdriving Fdriving / A σ driving
c + (σ n − P )tan φ
F .S . =
τ
σ n = (( m) g / A) cos β = (( ρHA) g / A) cos β = ρgH cos β (Here, compression is positive)
τ = (( m) g / A) sin β = (( ρHA) g / A) sin β = ρgH sin β
c + ( ρgH cos β − P )tan φ

F .S . =
ρgH sin β

GG 454 March 19, 2002 5
Factors That Increase Shear Driving Stress

Removal of Support
• Erosion of toe areas by streams, surf; lowering of reservoirs
• Holocene oversteepening of slopes by glacial erosion
• Weathering promotes erosion
• Mass wasting of supporting lower slope
• Quarries, canals, road cuts, tunnels
Surcharge
• Weight of precipitation
• Accumulation of talus
• Construction of fill
• Stockpiles of ore, waste
• Weight of buildings
• Weight of water from leaking pipes, lawn watering, pools, etc.
Transitory Earth Stresses
• Earthquakes
• Vibrations from machinery, traffic, blasting, etc.
Volcanic Processes
• Inflation of magma chambers (Mt. St. Helens)
• Harmonic tremors, earthquakes
Increase in Lateral Pressure
• Water in cracks
• Swelling (ice, clays)
Factors That Contribute To Low or Reduced Shear Strength
• Inherently weak geologic materials (clay, shale, organic material,
schists, soft tuffs, talc, serpentine
• Discontinuities (Bedding planes, faults, joints)
• Strata inclined towards free face
• Dip slopes
• Weathering of rocks
• Chemical exchange or leaching (e.g. quick clays)
• Hydration of clay minerals
• Dissolution of cement
• Shrink/swell effects
INCREASED PORE PRESSURE
• Rain, reservoir level rise, alteration of drainage

GG 454 March 19, 2002 6
Effective Stress Example

Suppose a block of impermeable, zero porosity quartz of density 2.67
g/cm 3 rests on a horizontal surface. What is the total normal stress and
effective stress at the base of a 10m tall block of pure quartz?
σn = (ρ quartz )(g)(h)
= (2.67 x 103 kg/m3 )(9.81 m/sec 2 )(10m)
= 2.62 x 105 kgm/sec2 /m 2 = 2.62 x 105 Pa
σ n ’ = σ n (no pore pressure at base)
Now suppose a dry block of quartz sandstone rests on a horizontal
surface. Further assume that the porosity of the sandstone is 20%. What
is the total normal stress at the base of a 10m tall block of this
sandstone ?
σn ’ = {ρ bulk }(g)(h)
= {(2.67 x 103 kg/m3 )(0.80) + (0 kg/m3 )(9.81 m/sec 2 )(10m)(0.20)
= (2.14 x 103 kg/m3 )(9.81 m/sec 2 )(10m)
= 2.10 x 105 kgm/sec 2 /m 2 = 2.10 x 105 Pa (i.e., 80% of first case)
Now suppose a saturated block of quartz sandstone rests on a
horizontal surface. Further assume that the porosity of the sandstone is
20%. What is the total normal stress and total effective stress at the
base of a 10m tall block of this sandstone?
σn= (ρsat )(g)(h)
= {(2.67 x 103 kg/m3 )(0.80) + (103 kg/m3 )(9.81 m/sec 2 )(10m)(0.20)
= (2.34 x 103 kg/m3 )(9.81 m/sec 2 )(10m)
= 2.30 x 105 kgm/sec2 /m 2 = 2.30 x 105 Pa
NOTE: ρsat ≠ ρ sat + ρ water !! (2.30 ≠ 2.13 + 1.00)
So the total stress is slightly greater for the saturated block because the
saturated block is denser than the dry block. But the pore pressure
increase at the base of the block more than compensates for the total
stress increase resulting from the higher bulk density:
σn ’= σn - P = (ρ bulk )(g)(h) - (ρ water )(g)(h)
= 2.30 x 105 Pa - (103 kg/m3 )(9.81 m/sec 2 )(10m)
= 2.30 x 105 Pa - 0.98 x 105 Pa = 1.32 x 105 Pa
So saturating the block increases the total stress but decreases the
effective stress.

GG 454 March 18, 2002 1
SLOPE STABILITY (LIMIT EQUILIBRIUM) ANALYSIS (27)
I Main Topics
A General procedure for slope stability (limit equilibrium) analyses
B Alternative definition of factor of safety
C Types of limit equilibrium analyses
D Comments on analyses for factors of safety
E Example(s)
II General procedure for slope stability analyses
Focus here is on the effect of various conditions that promote slope
failure and not on the numerical results.
Short version
A Postulate slip mechanism (failure criterion and failure geometry)
B Calculate shearing resistance by method of statics
C Compare calculated resistance with shear strength
D Find lowest factor of safety by iteration if failure surface is not
known
Long version
A Determine geometry of potential slide block, potential failure
surface(s), and identify other key factors (whatever they may be).
B Decide on appropriate type of stability analysis and failure criteria.
C Calculate the static vertical force due to the weight of the block.
D Determine the static components of force perpendicular (normal) to
the potential slip surface and the driving force parallel to the
potential slip surface. Calculate the driving moment if need be.
E If the slide material is saturated, calculate the pore pressure and
then calculate the effective normal stress acting on the potential
slip surface (divide the appropriate force by the area of the
potential slip surface). If the effect of flowing water in a slope is
accounted for, the pore pressure used should be for flowing water.
F Calculate the resisting stress using the effective normal stress.

GG 454 March 18, 2002 2
G Calculate the resisting force by multiplying the resisting stress by

the area of the potential rupture surface (or calculate the resisting
moment by multiplying the appropriate shear strength by the
appropriate lever arm).
H Calculate the factor of safety by dividing the resisting force
(moment) by the driving force (moment).
I Repeat with different conditions (failure surfaces, strengths, etc.)
to test sensitivity of result.
III Alternative definition of factor of safety (F.S.)
A Gauge for which the shear strength model parameters would be
reduced to bring slope into limiting equilibrium (verge of failure)
along a given slip surface
B Usually F.S.=Resisting moment or F.S. = Resisting shear strength
Driving moment Shear driving stress
C Moment = Force x lever arm = Stress x area x distance
D Key Points
1 The factor of safety is not a measure of stability at a point; it is
a number that represents averaging
2 The factor of safety cannot be measured in the field
3 The factor of safety is model-dependent
4 A factor of safety higher significantly greater than one is
desirable because uncertainty regarding the geologic conditions
and pore pressure variability.
Significance of factor of safety (from Sowers, 1979, p. 587)
Factor of Safety Significance
Less than 1.0 Unsafe
1.0 - 1.2 Questionable safety
1.3 - 1.4 Satisfactory for cuts, fills; questionable for dams
1.5 - 1.75 Safe for dams
IV Three main types of limit equilibrium analyses

A Planar slip surface (uses resisting/driving forces)
B Circular slip surface (uses resisting/driving moments): Rahn, p. 183
C Method of slices (uses resisting/driving moments): Rahn, p. 184-186.

GG 454 March 18, 2002 3
V Some comments on analyses for factors of safety

A Sowers (1979, p. 579) on computer analyses by method of slices:
"The author has found their indiscriminate use misleading and
sometimes unsafe... The computed minimum [factor of safety] is
often unrealistic or geometrically impossible, diverting attention
from the real answer."
B Morgenstern and Sangrey (SR 176, p. 165-166): "The analysis of a
rock slope in terms of a factor of safety is a subordinate activity to
achieving a clear understanding of the controlling geology and
water-pressure configuration."
C From p. 187 of Rahn (1996): “Most mathematical models and
computational abilities are more precise than the ability to
determine reliable values for the properties and parameters needed
for their application.”

GG 454 March 18, 2002 4
Method of Slices
r Moment arm r
1 associated with 2
resisting force:
Unsaturated
Moment arm R R Mresisting = W r
associated with Mean
driving force: Saturated water β
height
Mdriving = τ R h
Vector representing Basal Slip Surface

basal resisting shear force of area A
of slice
Vector representing
driving force (i.e. total weight) Total Weight = W
of slice (magnitude W) Component of
Total Weight
that is normal
to basal slip
surface:
Fn = W cos β
Mohr-Coulomb Failure Envelope

3
φ Total normal stress = σT = Fn / A
τ τ = σ’ tan φ + c
Pore pressure = P = ρΗ20 g h
NOTE: If φ = 0, then τ = c
c Effective normal stress = σ ’ = σT - P
P
Resisting shear stress = τ = σ’ tan φ + c
x x
σ’ σT σn Resisting shear force = τ A

GG 454 March 18, 2002 5
Effect of Flow Parallel to a Slope on Water Pressure
β
C Equipotential line
EC A
EA
Elevation
β
Equipotential lines are lines of equal total potential
so HA = HB; a well drilled to B will rise to the
EB elevation of A, less than the elevation of C.
B
Flow line
Htotal = Helevation + Hpressure
HA = EA +0
HB = EB + Hpressure(B)
Because HA = HB, Hpressure(B) = EA - EB

So the pressure head at B is less than EC - EB.
Flowing water reduces the pressure.
By examining the geometry of triangle ABC
EA - EB = {(EC - EB) cosβ} cosβ = Hpressure(B)
The pressure at B is the pressure head x (ρH 0)(g)

2
PB = [ (EC - EB) cosβ} cosβ ](ρH 0)(g)
2
This is the pressure that affects the effective stress at B.

GG 454 March 18, 2002 1
CHARACTERIZATION AND MONITORING OF SLOPE MOVEMENTS (28)
I Main Topics
A Evidence for different types of movement
B Slope stability reconnaissance procedure (Step I)
C Miscellaneous tips for aerial photography
D Ground surveys (Step 2)
E Comments on uses and limitations of borehole and piezometer data
I I Evidence for different types of movement (See p. 74-75 in SR 176)
A Type of rock or soil
B Topography/slope
C Fractures
D Water
III Slope stability reconnaissance procedure (Step I)
1 Obtain and review existing literature. Focus beyond immediate site
limits. {reports and maps (geologic, topographic, & soil)}
2 Obtain and review aerial photographs (or other imagery) of site.
Photos for different years or different times of year are helpful.
3 VISIT THE SITE and perform field reconnaissance
Note: The order of steps 1-3 should be rearranged as appropriate
4 Repeat steps 1-3 as necessary
5 Decide where more detailed field tests are required.
IV Miscellaneous tips for aerial photography
1 Aerial photography usually more effective and less costly than other
types of remote sensing imagery.
2 Low sun angle vs. high sun angle
3 Slope moisture and vegetation response depends of the time of year.
4 Fine-grained, low-permeability surficial materials commonly have
fine-grained drainage patterns.
5 Moist slopes often appear darker than light slopes
(probably partly a function of vegetation)
6 Vegetation sensitive to material, moisture, and stability

GG 454 March 18, 2002 2
V Engineering Geologic Mapping and Ground surveys (Step 2)

A Prepare engineering geologic map that defines major geomorphic
elements of slide, key geologic features (rock type, bedding dip,
faults, potential high- and low-permeability units, etc.),
particularly important vegetation features, sites of water
inflow/outflow, and slope angles.
B Construct geologic/topographic profile along slide.
C For complicated large slides a fence diagram may be useful.
D Establish survey points for monitoring slide; some must extend
beyond area of movement.
E Survey the diagonals between grid points.
F Displacement and strain in landslide (see handout)
G Determine best locations for exploratory boreholes, trenches, and
piezometers (if needed).
H Drill boreholes, update geologic maps and cross-sections, monitor
holes
VI Comments on uses and limitations of borehole and piezometer data
A Boreholes
1 Useful method for directly detecting features at depth.
2 Sampling by drilling alters sample properties and water content.
Lab tests on extracted samples may or may not be useful.
3 Boreholes sample a statistically insignificant part of a body.
Need to use judgment as to how far to extrapolate drilling data,
how many holes to drill, and where to drill them; this is a very
persistent problem in geotechnical practice.
B Piezometers
1 In many cases piezometers can be installed inexpensively and
give enormously useful information.
2 Coupling of displacement and piezometer information can be
invaluable. See GSA paper by Iverson and Major (1987).
3 See comment 3 above.
4 Movement of the slide can radically rearrange the hydrology of a
slide. Opening of cracks (or animal burrows) is a particular
problem. See GSA paper by Harp and others (1990) for details.
5 In some cases installing the piezometer can change the hydrology
of a slide.
6 Note any ground water withdrawal/recharge offsite.
GG 454 March 18, 2002 3
DISPLACEMENT AND STRAIN IN A LANDSLIDE
umax
Extension
Axial strain (e)
e = du/dx
0
Shortening
x
Axial Displacement (u)
u = ” e dx
umax

GG 454 March 30, 2002 1
COASTAL PROCESSES (29)
I Main Topics
A Recognition, characterization, risk evaluation, risk assessment
B Selected erosion rates
C Tsunamis in the northeastern Pacific Ocean
I I Recognition, characterization, risk evaluation, risk assessment
http://www.fema.gov/nwz00/erosion.htm (FEMA study, June 27, 2000)
http://www.csc.noaa.gov/crs/ALACE/techcd/htm/severity.htm
A ~ 25% of homes and other structures within 500 feet of the U.S.
coastline and the shorelines of the Great Lakes will fall victim to
the effects of erosion within the next 60 years
B Average costs to U.S. homeowners: $500,000,000+ / y r
C The Atlantic and Gulf coasts account for 45 percent of the U.S.
coastline and 63 percent of the structures within 500 feet of the
shoreline. The nation's highest average erosion rates - 6 feet+ /yr -
occur along the Gulf coast.
D The average erosion rate on the Atlantic coast is ~2-3’/yr.

However, actual erosion rates can vary widely from one location to
another and from one year to another. A hurricane or other major
storm can cause the coast to erode 100 feet or more in a single day.
E Development in several high-risk coastal areas studied has

increased by more than 60% over the last 20 years.
F Property owners within the first few hundred feet of the nation's
coasts face as large a risk of damage from erosion as they do from
flooding.
G Only about half of homeowners in high erosion areas on the Atlantic

and Gulf coasts currently hold flood insurance policies.
H To fully reflect erosion risk, insurance rates in the highest hazard

coastal areas would have to be double today's rates, on the average.
I ~ 10,000 structures exist on land expected to erode within 10 years.
J ~ 87,000 structures within the estimated 60-year erosion zone.

GG 454 March 30, 2002 2
III Selected erosion rates (recognition) (From Bascom, 1980)

A England
1 Sheringham
Sheringham Sheringham Harbor Today: No harbor
(1781) (1829)
50'
20'
Harbor now gone
2 Cromer: “Long-term” sea cliff retreat of 19'/yr (6m/yr)
3 Suffolk (1953) (A nearby lower cliff retreated 90’)

40'
40' One night!

GG 454 March 30, 2002 3
B Southern California coast
http://www.nationalgeographic.com/xpeditions/atlas/index.html
1 Encinitas (From Bascom, 1980)

a 1883-1976: six city blocks lost, maximum retreat of 800'
b Retreat concentrated in storms of 1884, 1886, 1889, and 1891
c Choice of action: acceptance of hazard & change in assessed
property value
2 Santa Barbara breakwater (1928)

http://ci.santa-barbara.ca.us/departments/waterfront/index2.html
a Project completed in 1928 to form a harbor
b Sand accretion immediately upon completion
c Federal dredging authorized in 1935 at $30,000 per year
d Federal dredging costs in 1992: ~$800,000 per year

GG 454 March 30, 2002 4
3 Santa Monica breakwater (1934)

http://www.smmirror.com/Volume1/issue41/city_plans_to.html
a Built in 1934 to protect boats along the Santa Monica Pier.
Built from scrapped cars.
b Design intended to allow sand to pass by breakwater
c Results: deposition behind breakwater, erosion “downstream”
d Status: Allowed to deteriorate. Being considered for
restoration to allow the long-unused harbor to be used again.
Normal beach Original coastline

Harbor
Spit Post-breakwater
coastline
2000'
Incoming waves
Incoming waves
Deposition at the Santa Barbara breakwater Deposition at the Santa Monica breakwater
Modified from Costa & Baker, 1981) Modified from Costa & Baker, 1981)
4 Channel Islands Harbor harbor and breakwater, Oxnard (1963)

http://www.spl.usace.army.mil/co/navigation/chnlisln/channel.html
a Dredging from mid-1950’s to 1963.
b Designed to lessen the effect of

downshore beach erosion, provide
dredged material for beach
replenishment for downcoast beaches,
and prevent sand loss to the submarine
canyon off of Port Hueneme.
c Currently, authorized by Congress for

dredging on a biennial basis; ~1.8
million cubic yards of material will be
dredged biennially.
D Hawaii http://www.soest.hawaii.edu/SEAGRANT/CEaBLiH.html
~25% (17 miles) of Oahu's beaches have been lost or significantly

narrowed over this century

GG 454 March 30, 2002 5
V Historic large tsunamis that have affected Hawaii

http://www.tsunami.org/
http://www.soest.hawaii.edu/tsunami/
http://www.usc.edu/dept/tsunamis/alaska/1946/webpages/
A Distant sources: 1946, 1960. Great damage in Hilo
B Local sources:1868, 1975. 14+ m run ups near Halape
http://pubs.usgs.gov/gip/hazards/tsunamis.html
http://www.soest.hawaii.edu/tsunami/

GG 454 March 31, 2002 1
THE WAVE EQUATION (30)
I Main Topics
A Assumptions and boundary conditions used in 2-D small wave theory
B The Laplace equation and fluid potential
C Solution of the wave equation
D Energy in a wavelength
E Shoaling of waves
I I Assumptions and boundary conditions used in 2-D small wave theory
Small amplitude surface wave
y L
H = 2A = wave height
η x
Still water u
level ε v
ζ
Depth above
Water depth = +d bottom = d+y Particle Particle
orbit velocity
y = -d
Modified from Sorensen, 1978
A No geometry changes parallel to wave crest (2-D assumption)

B Wave amplitude is small relative to wave length and water depth; it
will follow that particle velocities are small relative to wave speed
C Water is homogeneous, incompressible, and surface tension is nil.
D The bottom is not moving, is impermeable, and is horizontal
E Pressure along air-sea interface is constant
F The water surface has the form of a cosine wave
H x t
η = cos 2π −
2 L T

GG 454 March 31, 2002 2
III The Laplace equation and fluid potential
A Conservation of mass
Consider a box the shape of a cube, with fluid flowing in and out or it.
∆z v+∆v
(out)
u+∆u (out)
∆y
u
(in) ∆x
v
(in)
The mass flow rate in the left side of the box is:
∆m1 ∆ρV ∆V ∆x∆y∆z ∆x
1a = =ρ =ρ = ρ∆y∆z = ( ρ )( ∆y∆z )(u)
∆t ∆t ∆t ∆t ∆t
The mass flow rate out the right side is
∆m 2
1b = ( ρ )( ∆y∆z )(u + ∆u) .
∆t
For the bottom of the box, mass flow rate in is
∆m3
1c = ( ρ )( ∆x∆z )(v) ,
∆t
and the mass flow rate out the top is
∆m 4
1d = ( ρ )( ∆x∆z )(v + ∆v)
∆t
If the fluid is incompressible (so no fluid can be compressed and stored in
the box), then: a) the fluid mass flowing into the box in a given increment
of time must equal the fluid mass flowing out of the box in that same
increment of time, and b) the density of the fluid (ρ ) is a constant.
Remember that u and v are velocities in the x and y directions,
respectively. So, what goes in equals what comes out:
∆m1 ∆m3 ∆m 2 ∆m 4
2a + = + .
∆t ∆t ∆t ∆t
or
∆m1 ∆m 2 ∆m3 ∆m 4
2b − + − = 0.
∆t ∆t ∆t ∆t

GG 454 March 31, 2002 3
Substituting equations (1) into (2b)

3 ( ρ )( ∆y∆z )(u) − ( ρ )( ∆y∆z )(u + ∆u) + ( ρ )( ∆x∆z )(v) − ( ρ )( ∆x∆z )(v + ∆v) = 0.
Dividing out the terms ρ and ∆z
4 ( ∆y)(u) − ( ∆y)(u + ∆u) + ( ∆x )(v) − ( ∆x )(v + ∆v) = 0.
Now let ∆u = ∆x [∂u/∂x] and ∆v = ∆y [∂v/∂y]
5 ( ∆y)(u) − ( ∆y)(u + ∆x[∂u / ∂x ]) + ( ∆x )(v) − ( ∆x )(v + ∆y[∂v / ∂y]) = 0.
This simplifies to
∂u ∂v
6 −( ∆y)( ∆x ) − ( ∆x )( ∆y ) = 0.
∂x ∂y
After dividing out the ∆x and ∆y terms
∂u ∂v ∂u ∂v ∂u ∂v
7a − − = 0, or 7b + = 0, or 7c =− .
∂x ∂y ∂x ∂y ∂x ∂y
Equation 7c states that a change in flow in the x direction must be
compensated for by an opposite change in flow in the y-direction if mass
is to be conserved.

GG 454 March 31, 2002 4
B Conservation of angular momentum
[Conditions of irrotational flow (vorticity = 0)]
We don't want our box to rotate. Experiments with submerged floats

beneath waves show that the floats do not spin.
u + ∆u ∆u
v + ∆v
v
∆y ∆y
∆v
∆x ∆x
u
Flow velocities on box sides
If there is no rotation, then there can be no moment:

8 ( Force) y (lever arm) y = ( Force) x (lever arm) x
The shear force is related to the shear stress (τ ) as follows:
9 Fshear = (τ )(area)
Substituting equation (9) into equation (8) yields.
10 (τ xy [∆y∆z])(∆x) = (τ yx [∆x∆z])(∆y)
For linear fluids, the shear stress is proportional to the velocity gradient
du dv
11a τ xy = µ 11b τ yx = µ (µ = viscosity)
dy dx
Substituting equations (11) into (10) then gives:
 ∂u   ∂v 
12  µ [ ∆y∆z ] ( ∆x ) =  µ [ ∆x∆z ] ( ∆y)
 ∂y  ∂x
Hence
∂u ∂v ∂u ∂v
13a = or 13b − = 0 (vorticity = 0)
∂y ∂x ∂y ∂x

GG 454 March 31, 2002 5
C Irrotational potential flow and the Laplace equation
The Laplace equation is one of the most common equations in physics. It

describes how the second partial derivatives of a function ( φ ) are related:
∂ 2φ ∂ 2φ
14a + =0 or 14b ∇ 2 φ = 0 .
2 2
∂x ∂y
In the context of waves, φ will represent a fluid potential function. You
are already familiar with gravitational potential energy U:

15 U = mgy y = height
Partial derivatives of potential functions, taken with respect to position,

yield measurable physical quantities (e.g., gravitational potential U):
∂U
16 − = − mg = Fy F y= gravitational force (Note that ∇ 2U = 0).
∂y
Darcy's law for one-dimensional fluid flow can be written in terms of
fluid potential φ :
 ρ  ∂φ
17 q x = −k x  
 µ  ∂x
where qx = flux (m/s), kx = permeability in the x-direction (m2 ), ρ = density
(kg/m 3 ), µ = fluid viscosity (kg m-1 s-1 ), and x = position (m).
Suppose that a potential function φ exists that satisfies the Laplace
equation, and that the following conditions apply:

∂φ
18a = u, where u =horizontal component of particle velocity
∂x
∂φ
18b = v, where v =vertical component of particle velocity.
∂y
Now let us substitute the expressions of equations (18) into the Laplace
equation (equation 14a); this yields the continuity equation (7b).
∂φ  ∂φ 
∂  ∂ 
∂x ∂y ∂u ∂v
19 + = + =0
∂x ∂y ∂x ∂y
Does the vorticity condition hold? Inserting eqs. (18) into eq. (13b):

GG 454 March 31, 2002 6
 ∂φ   ∂φ 
∂  ∂ 
∂u ∂v  ∂x   ∂y  ∂ 2 φ ∂ 2 φ
20 − = − = − =0 So vorticity = 0.
∂y ∂x ∂y ∂x ∂x∂y ∂y∂x
So the Laplace equation can be used to study water waves.
Solutions to the Laplace equation also obey an averaging procedure, where

the value of the function at a point on a square grid is the average of the
values at the nearest four gridpoints.
y ø2
ø6
ø3 ø7 ø0 ø5 ø1
x
ø8
∆y
ø4
∆x
φ0 = 1/4 [ φ1 + φ2 + φ3 + φ4]

GG 454 March 31, 2002 7
IV Solutions for wave speed and particle velocities

(see appendix for derivation)
A General solutions
H g cosh([2 π / L][d + y] sin(2 πx / L − 2 πt / T )
φ =
2 (2 π / T ) cosh(2 πd / L)
1 Wave speed or wave celerity (C)
C = (gT /2 π ) tanh(2 π d/L) T = wave period = constant
Function of wave period and relative water depth
2 Wave length (L)
L = CT
3 Horizontal particle velocity amplitude (|u|)
|u| = (πH / T ) (cosh [2 π (d + y ) / L ])/(sinh [2 π d / L ] )
Function of wave height and relative water depth and relative
distance above bottom
4 Vertical particle velocity amplitude (|v|)
|v| = (πH / T ) (sinh [2 π (d + y ) / L ])/(sinh [2 π d / L ] )
5 Amplitude of horizontal particle displacement (|ζ|)
|ζ| = |u| (T/2π)
6 Amplitude of vertical particle displacement (|ε|)
| ε | = |v| (T/2π)
B Deep-water solutions (d/L > 0.5, or 2πd/L > π, so tanh(2πd/L) ≈ 1)
1 Wave speed (C)
a C = gT/2π (function of wave period; independent of depth)
2 Wave length (L)
a L = CT
b L = (gT 2 )/2π (function of wave period; independent of depth)
|u| = ( π H/T)(e 2 π y/ L ).
The horizontal velocities decrease exponentially with depth (y<0)
|v| = ( π H/T)(e 2 π y / L ).
The vertical velocities decrease exponentially with depth (y<0)
|ζ| = (H /2) (e 2 π y / L ) . Exponential decay with depth
6 Amplitude of vertical particle displacement (| ε | )
| ε| = (H /2) (e 2 π y / L ) . Exponential decay with depth

GG 454 March 31, 2002 8
C Shallow-water solutions (d/L < 0.05, or 2 π d/L < π /10)

Examples: tides, tsunamis,
1 Wave speed (C) [tanh(2 π d/L) ≈ 2 π d/L]
C = (gT)(d/L) = (gd)(T/L) = (gd)(1/C) ⇒ C = (gd) 1 / 2
i Function of water depth; independent of period
i i As water depth decreases, wave speed decreases
2 Wave length (L= CT)
L = (gd)1/2 T
i As water depth decreases, T stays constant, L decreases
i i Waves will bunch together as they enter shallower water
|u| = (πH/T)/(2 π d/L) = HL/2dT = (L/T)(H/2d) = C (H/2d)
|u| is independent of distance above bottom; |u| ≠0 at bottom
|v| = (πH/T) [(d + y)/d]
|v| = 0 at bottom and increases linearly to πH/T at surface (y=0)
|ζ| = (H/2)/(2 π d / L ) = (L/ π )(H/d) Independent of y
6 Amplitude of vertical particle displacement (|ε|)
| ε| = (H/2) [(d + y)/d] Decays linearly with y
7 Wave base: y = -L/2 (e- π = 0.04)
Shallow
water
L/2
(wave base)
Deep
water
Not to scale
Modified from Sorenson, 1978

GG 454 March 31, 2002 9
V Energy in a wavelength (per unit length along wave crest)

A Kinetic energy = (Ek )/z = ρ gH 2 L/16
B Potential energy = (Ep )/z = ρ gH 2 L/16 So Ek = Ep!
C Total energy = (ET )/z = (Ek +E p )/z = ρ gH 2 L/8
V I Shoaling of waves
Assuming no energy loss as a wave shoals*
Case1: N
Wave crest B1
is unstretched
B2
Case 2:
Wave crest
is shortened B1
B2
A E1 = E2
B (B 1 ρ gH 1 2 L 1 )/8 = (B2 ρ gH 2 2 L 2 )/8
C (H 2 /H 1 ) = (L1 /L 2 ) 1/2 (B1 /B 2 ) 1 / 2
1 As L decreases, H increases
2 As B decreases, H increases
D Wave steepness = H/L
E Waves get taller and steeper as they shoal because:
L decreases and H increases (conservation of energy)
F Waves break when (H/L) = 1/7 tanh (2πd/L)

GG 454 March 31, 2002 10
Hyperbolic Functions
eβ − e−β β3 β5
sinh(β ) = = β + + + ...
2 3! 5!
Shallow water: As ß → 0, sinh (ß) → ß (from series expansion)
Deep water: As ß → ∞, sinh (ß) → (eß )/2 (from definition)
sinh (π) = 11.549
eβ + e−β β2 β4
cosh(β ) = = 1 + + + ...
2 2! 4!
Shallow water: As ß → 0, cosh (ß) → 1 (from definition)
Deep water: As ß → ∞, cosh (ß) → (eß )/2 (from definition)
cosh (π) = 11.592
eβ − e−β β3 2β 5 π
tanh(β ) = = β − + + ... for β <
eβ + e−β 3 15 2
Shallow water: As ß → 0, tanh (ß) → ß (from series expansion)
Deep water: As ß → ∞, tanh (ß) → 1 (from definition)
tanh (π) = 0.9963
In the expressions below, k = 2π/L

ek(d+y) + e−k(d+y)
cosh k(d + y) 2 ekd eky + e−kd e−ky
= =
sinh kd e − e−kd
kd ekd − e−kd
2
Shallow water: As kd→ 0 ,cosh k(d+y)→ 1, sinh kd→ kd, so...
[cosh k(d+y)]/sinh kd → 1/kd
Deep water: As kd→π , [cosh k(d+y)]/sinh kd → eky
ek(d+y) − e−k(d+y)
sinh k(d + y) 2 ekd eky + e−kd e−ky
= =
sinh kd e −
kd e−kd ekd − e−kd
2
Shallow water: As kd→ 0 , sinh k(d+y)→ k(d+y), sinh kd→ kd, so...
[sinh k(d+y)]/sinh kd → (d + y)/d = height above bottom/depth
Deep water: As kd→π , [sinh k(d+y)]/sinh kd → eky

GG 454 March 31, 2002 11
Appendix
Derivation of the small amplitude wave equation
(from Sorenson, R.M., 1978, Basic coastal engineering: Wiley, New York,
227 p.)
Small amplitude surface wave
y L
H = 2A = wave height
η x
Still water u
level ε v
ζ
Depth above
Water depth = +d bottom = d+y Particle Particle
orbit velocity
y = -d
Modified from Sorensen, 1978
The original solution is attributed to Airy (Airy, C.B., 1845, On tides and
waves, in Encyclopedia Metropolitana, London, p. 241-396).
H x t
η = cos 2π − (30A.1)
2 L T
or
H
η= cos 2π ( kx − σt) (30A.2)
2
where
2π
k= ( wave number) (30A.3)
L
2π
σ= ( wave angular frequency ) (30A.4)
T
The flow normal to the sea bed is zero, so
∂φ
v= = 0 at y = −d (30A.5)
∂y
This is the first boundary condition.
The unsteady Bernoulli equation for irrotational flow is
1 2
u + v 2  + gy + p + ∂φ = 0 (30A.6)
2  ρ ∂t
where g = gravitational acceleration, p is the pressure, ρ is fluid density,
and the last term is a dynamic pressure term associated with

GG 454 March 31, 2002 12
accelerations. If the squares of the velocity terms are assumed to be

small relative to the other terms, and if the particle velocities are small
relative to the wave speed, then at the surface (i.e., at y = η ), where the
pressure is taken as zero, the unsteady Bernoulli equation yields
−1 ∂φ
y= at y = η (30A.7)
g ∂t
This yields the second boundary condition is at y=0,
−1 ∂φ
η= at y = 0 (30A.8)
g ∂t
The velocity potential should vary with depth, and should have the same
cycle as the wave. If the depth contribution (Y) can be separated from the
cyclic contribution (a common assumption in solving differential
equations), then the velocity potential φ would have the following form:
φ = Y sin( kx − σt) (30A.9)
where Y = Y(y). Upon insertion of this function into the Laplace equation
∂ 2φ ∂ 2φ
+ =0 (30A.10)
∂x 2 ∂y 2
one obtains
(
∂ 2 Y sin( kx − σt)) + ∂ 2 (Y sin(kx − σt)) = 0 (30A.11)
∂x 2 ∂y 2
This simplifies first to
∂ 2 (sin( kx − σt)) ∂ 2Y
Y + sin( kx − σt) =0 (30A.12)
∂x 2 ∂y 2
and then to
∂ 2Y ∂ 2Y
−k 2Y sin( kx − σt) + sin( kx − σt) = −k 2Y + =0 (30A.13)
∂y 2 ∂y 2
The solution to this differential equation is well known
Y = Ae ky + Be−ky (30A.14)
and can be verified by substitution into (30A.13). Substituting this into
(30A.9) yields a general solution of the Laplace equation .
φ =  Ae ky + Be−ky  sin( kx − σt) (30A.15)
The two constants A and B now need to be solved for using the two
boundary conditions (30A.5) and (30A.8). Inserting (30A.15) into (30A.5)

GG 454 March 31, 2002 13
∂ Ae ky + Be−ky  sin( kx − σt)

v=  = 0 at y = −d (30A.16)
∂y
or
k Ae−kd − Be kd  sin( kx − σt) 
v=  =0 (30A.17)
∂y
The only way this can hold for all values of x and t is if
Ae−kd − Be kd = 0 (30A.18)
or
e kd
A=B (30A.19)
e−kd
Inserting this back into (30A.15) yields φ with one unknown constant
 e kd   e ky e−ky 
φ = B e ky + Be−ky  sin( kx − σt) = Be kd  + sin( kx − σt) (30A.20)
−kd −kd e kd 
 e  e 
or
φ = Be kd e k ( y + d ) + e−k ( y + d )  sin( kx − σt) (30A.21)
The term in the large parentheses equals 2cosh[k(d+y)], so

φ = Be kd (2cosh k ( y + d )) sin( kx − σt) (30A.22)
Now the form of the water surface is
H x t
η = cos 2π − (30A.1)
2 L T
So at t=0, x=0
H
η= (30A.23)
2
Now the second boundary condition comes into play.
−1 ∂φ
η= at y = 0 (30A.8)
g ∂t
So at t=0, x=0, and y=0
H −1 ∂φ
= (30A.24)
2 g ∂t
The derivative on the right side of (30A.24) is found from (30A.22)
∂φ
= Be kd (2cosh( kd )) cos( kx − σt)(−σ ) (30A.25)
∂t
Substituting this back into (30A.24)
H −1 kd
= Be (2 cosh( kd )) cos( kx − σt)(−σ ) (30.A26)
2 g
This is solved for Bek d most readily where cos(kx-σ t) =1:

GG 454 March 31, 2002 14
gH
= Be kd (30.A27)
2σ (2 cosh( kd ))
This goes into (30A.22) to yield the expression for the velocity potential
gH
φ= (cosh k ( y + d)) sin(kx − σt) (30A.28)
2cosh[ kd ]
The wave speed (or celerity) is a key term we wish to find. We find it by
evaluating the vertical velocity at the surface in two ways. First, using
the expression for the water height above still water level
−1 ∂φ
η= at y = 0 (30A.8)
g ∂t
we obtain
∂η −1 ∂ 2φ
v= = at y = 0 (30A.29)
∂t g ∂t2
Also the vertical velocity is given by
∂φ
v= (30A.30)
∂y
So at the surface
−1 ∂ 2φ ∂φ
= at y = 0 (30A.31)
g ∂t2 ∂y
Inserting (30A.28) , gives
gH gH
∂2 (cosh k ( y + d)) sin(kx − σt) ∂ (cosh k ( y + d)) sin(kx − σt)
−1 2 cosh[ kd ] 2 cosh[ kd ]
= at y = 0
g ∂t2 ∂y
(30A.32)
Taking the derivatives yields
σ2 gH gH
(cosh k ( y + d)) sin(kx − σt) = k (sinh k ( y + d)) sin(kx − σt) at y = 0
g 2 cosh[ kd ] 2 cosh[ kd ]
(30A.33)
Now set y= 0, and solve for φ . Many terms can be dropped from both sides.
k sinh( kd )
σ 2 = −g = gk tanh( kd ) (30A.34)
cosh k ( d )
Now the wave speed C = L/T = σ /k, so
gk tanh( kd ) g gL 2πd
C= = tanh( kd ) = tanh (30A.34)
k k 2π L

GG454 April 1, 2002 1
REFLECTION, REFRACTION, AND DIFFRACTION (31)
I Main Topics
A Huygen's Principal
B Reflection
C Refraction
D Diffraction
I I Huygen's Principal
I I I Reflection: "to bend (bounce) back"
A Angle of incidence = angle of reflection
B Waves reflect off vertical walls, not gently sloped beaches
IV Refraction: bending of a wavefront due to changes in its speed
A Snell's Law
B Effect of water depth on wave speed
C Effects of refraction at headlands and in bays
1 Waves concentrate on headlands
2 Waves diverge in bays
D Ancient Polynesians exploited refraction in exploration!
V Diffraction: deflection of waves around obstacles
A A manifestation of Huygen’s principal
B Effects of diffraction at breakwaters

GG454 April 1, 2002 2
HUYGEN'S PRINCIPLE
New Wavefront
Source
Wavefronts

GG454 April 1, 2002 3
Reflection of Waves
A
n
A' θi θr
C' Huygen's wavefront
Arc B'
B
v = wavespeed
B'
γ
α β C
Consider wavefront AA' that advances to BB' and then to CC'

Ray A'B' | to wavefront B'B because wavefronts are normal to rays
Ray B'C' | to wavefront C'C because wavefronts are normal to rays
(Ray B'C' | to wavefront C'C because CC' is tangent to Arc B')
BC =B'C' = v∆t; B'C = CB'; and both ∆B'BC and ∆CC'B' are right triangles, so
∆BB'C ~ ∆C'CB'. So angle BB'C (i.e., α) = angle C'CB' (i.e., β).
Now A'B' ⊥ B'B, and nB' ⊥ B'C, so θi = α.
ο ο
Also, α+β+γ= 90 , and α+γ+θr= 90 , so θr = β.
Whereas α = β, θi = θr .
The angle of incidence = the angle of reflection.

GG454 April 1, 2002 4
Refraction of Waves & Snell's Law

A'
Huygen's wavefront
B'
A
θ1
B θ1 C' "Fast material": v = v 1
θ2
"Slow material": v = v2
C θ2
AB = A'B' = v ∆t BC = v2 ∆t; B'C' = v1∆ t; BC < B'C'

1
sin θ = B'C'/BC'; sin θ = BC/BC' sin θ /sin θ = B'C'/C'C' = v /v
1 2 1 2 1 2
If v1 > v , then
2
θ >θ .
2
If v > v , then θ >θ .
1 2 1 2 1
Fast
(Deep water)
Slow
(Shallow water)
Incoming wave energy converges
on headlands
Fast
(Deep water)
Slow
(Shallow water)
Incoming wave energy diverges
in bays

GG454 April 1, 2002 5
DIFFRACTION & REFLECTION
A Breakwater
T4
T3
Wavefronts
T2
T1
B Breakwater
Source
Transmitted wave
T7
e
av
w
ed
T6
ct
ra
iff
D
T5
C Breakwater
T
5
T
6
Reflected wave
T
7

GG 454 April 8, 2002 1
COASTAL LANDFORMS (32)
I Main topics
A Sediment sources and sinks
B Beaches and coastal landforms
I I Sediment budget
A Where sediment input > sediment loss: accretion
B Where sediment loss > sediment input: erosion
C Coastal landforms reflect long term geologic history and sediment

budget
I I I Beaches and coastal landforms

A Beach: Accumulation of rock fragments that can be moved by
ordinary wave action
1 Berm: The above water portion of a beach
2 Bar: The below water portion of a beach; bars typically extend to
depths of ~10m below low tide level
3 Effect of seasons on bars and berms
Berm
Mean sea level
Bar
B Wave-cut platform/bench (e.g.,, Hanauma Bay)

C Wave-cut terrace: result from sea-level fall or land rise
D Deltas: sedimentary bodies deposited at a river mouth
1 Sediment sources
2 Marine influences
a Currents and delta shapes
i Stream-dominated (“birdfoot”): Distributaries prominent
i i Wave-dominated: beaches prominent
b Tides and tide flats
E Spits: attached, shore-parallel ridges form where there is a supply
of sand, a transporting current, and a dumping ground (see D.2.a.ii)

GG 454 April 8, 2002 2
F Barrier islands: low, elongate, shore-parallel islands (see D.2.a.ii)

1 Emergence of offshore bars
2 Submergence of beaches and beach dunes
G Lagoons: protected elongate or circular bay
H Reefs
1 Indicate approximate level of sea level at the time of reef
formation
2 Elevated reefs indicate former sea level highstands or coastal
emergence
3 Submerged reefs indicate former sea level lowstands or coastal
subsidence
I Estuaries: inundated stream valleys
J Fjords: inundated glacial valleys

GG 454 April 8, 2002 3
LONGSHORE DRIFT OF BEACH SAND
Incident
Waves
Longshore Drift
Backwash
Swash
Beach (berm)
LONGSHORE CURRENT
Incident
Waves
Rip Currents Rip Currents
Breaker zone
Water height builds Longshore Current
Beach (berm)
Modified from Press and Siever, 1978

GG 454 April 9, 2002 1
HURRICANES (33)
I Main Topics
A Definition and characterization of hurricanes
B Conditions required for hurricanes
C Tornadoes vs. hurricanes
D Storm surges
E Case histories
I I Terminology and classification of hurricanes (cyclones)
http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqHED.html
A Terminology: hurricane = typhoon = strong tropical cyclone.
1 A tropical cyclone is the generic term for a non-frontal synoptic

scale low-pressure system over tropical or sub-tropical waters
with organized convection (i.e. thunderstorm activity) and
definite cyclonic surface wind circulation
2 Cyclone: a large-scale storm with heavy rain and winds that

rotates around a low pressure center (the eye). The rotation is
counterclockwise in the northern hemisphere and clockwise in the
southern hemisphere.
3 A large, concentrated atmospheric/oceanic heat engine with

sustained wind speeds greater than 74 mph.
B Classification: the Saffir/Simpson Hurricane Scale

http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqD.html#D7
Saffir-Simpson Maximum Maximum Maximum Minimum Storm Storm
Category sustained sustained sustained surface surge surge
wind speed wind speed wind speed pressure
mph m/s kts mb ft m
1
74-95 33-42 64-82 greater 3-5 1.0-1.7
than 980
2
96-110 43-49 83-95 979-965 6-8 1.8-2.6
3
111-130 50-58 96-113 964-945 9-12 2.7-3.8
4
131-155 59-69 114-135 944-920 13-18 3.9-5.6
5
156+ 70+ 136+ < 920 19+ 5.7+

GG 454 April 9, 2002 2
Damage common in hurricanes of different strength

Category/ Damage
Example
1 Damage primarily to shrubbery, trees, foliage, and unanchored homes. No real
damage to other structures. Some damage to poorly constructed signs. Low-
Hurricane lying coastal roads inundated, minor pier damage, some small craft in exposed
Earl anchorage torn from moorings.
(1998)
2 Considerable damage to shrubbery and tree foliage; some trees blown down.
Major damage to exposed mobile homes. Extensive damage to poorly
Hurricane constructed signs. Some damage to roofing materials of buildings; some window
Georges and door damage. No major damage to buildings. Coast roads and low-lying
(1998) escape routes inland cut by rising water 2 to 4 hours before arrival of
hurricane center. Considerable damage to piers. Marinas flooded. Small craft
in unprotected anchorages torn from moorings. Evacuation of some shoreline
residences and low-lying areas required.
3 Considerable damage to shrubbery and tree foliage; some trees blown down.
Major damage to exposed mobile homes. Extensive damage to poorly constructed
Hurricane signs. Some damage to roofing materials of buildings; some window and door
Fran damage. No major damage to buildings. Coast roads and low-lying escape routes
(1996) inland cut by rising water 2 to 4 hours before arrival of hurricane center.
Considerable damage to piers. Marinas flooded. Small craft in unprotected
anchorages torn from moorings. Evacuation of some shoreline residences and
low-lying areas required.
4 Shrubs and trees blown down; all signs down. Extensive damage to roofing
materials, windows and doors. Complete failures of roofs on many small
Hurricane residences. Complete destruction of mobile homes. Flat terrain 10 feet
Andrew of less above sea level flooded inland as far as 6 miles. Major damage to lower
(1992) floors of structures near shore due to flooding and battering by waves and
floating debris. Low-lying escape routes inland cut by rising water 3 to 5
hours before hurricane center arrives. Major erosion of beaches. Massive
evacuation of all residences within 500 yards of shore possibly required, and
of single- story residences within 2 miles of shore.
5 Shrubs and trees blown down; considerable damage to roofs of buildings; all
signs down. Very severe and extensive damage to windows and doors. Complete
Hurricane failure of roofs on many residences and industrial buildings. Extensive
Camille shattering of glass in windows and doors. Some complete building failures.
(1969) Small buildings overturned or blown away. Complete destruction of mobile
homes. Major damage to lower floors of all structures less than15 feet above
sea level within 500 yards of shore. Low-lying escape routes inland cut by
rising water 3 to 5 hours before hurricane center arrives. Massive evacuation
of residential areas on low ground within 5 to 10 miles of shore possibly
required.
http://hurricanes.noaa.gov/prepare/categories3.htm

GG 454 April 9, 2002 3
I I I Conditions required for hurricanes (necessary but not sufficient)

http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqHED.html
1 Warm ocean waters (at least 26.5°C [80°F]) throughout a

sufficient depth (at least on the order of 50 m [150 ft]). Warm
waters are necessary to the heat engine of the tropical cyclone.
2 An atmosphere which cools fast enough with height such that it

is potentially unstable to moist convection. Thunderstorm
activity allows the heat stored in the ocean waters to be
liberated for the tropical cyclone development.
3 Relatively moist layers near the mid-troposphere (5 km [3 mi]).

Dry mid levels are not conducive for allowing the continuing
development of widespread thunderstorm activity.
4 A minimum distance of at least 500 km [300 mi] from the equator

such that the Coriolis force is sufficient to allow the near-
surface wind conditions needed to maintain low pressures on the
water surface.
5 A pre-existing near-surface disturbance with sufficient vorticity

and convergence. Tropical cyclones cannot be generated
spontaneously. To develop, they require a weakly organized
system with sizable spin and low-level inflow.
6 Low values (less than about 10 m/s [20 kts 23 mph]) of vertical
wind shear between the surface and the upper troposphere.
Vertical wind shear is the magnitude of wind change with height.
Large values of vertical wind shear disrupt the incipient tropical
cyclone and can weaken or destroy a cyclone that has formed.

GG 454 April 9, 2002 4
Cutaway diagram of a hurricane
http://mmem.spschools.org/grade5science/weather/hurricanediagram.html
Key points http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqA.html#a11

* Hurricanes draw warm moist air from over the oceans up into the
atmosphere to an altitude of ~12 km where the water condenses. The
air rises around the eye and in spiral rainbands. Condensation of water
aloft liberates energy.
* Air sinks in the eye and along the rainbands, with the largest region of
air subsidence being in the eye. The air moves from high pressure at
high elevations to low pressure at the surface. Sinking air in the eye
usually descends to 1-3 km above the surface, not to the surface.
* High pressure aloft causes moisture raised from the warm ocean to
spread out in the spiral rain bands.
* The winds at the eyewall (edge of the large central spiral above) are
the highest and help draw moisture up in the hurricane.
* Eyes range in size from 8 km [5 mi] to over 200 km [120 mi] across, but
most are approximately 30-60 km [20-40 mi] in diameter

GG 454 April 9, 2002 5
7 Energy
http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqD.html#D7
1.the total amount of energy released by the condensation of water droplets or ...
2.the amount of kinetic energy generated to maintain the strong swirling winds of the
hurricane.
It turns out that the vast majority of the heat released in the condensation process is used to
cause rising motions in the thunderstorms and only a small portion drives the storm's
horizontal winds.
Total energy released through cloud/rain formation:
An average hurricane produces 1.5 cm/day (0.6 inches/day) of rain inside a circle of radius
665 km (360 n.mi) or a volume of rain of 2.1 x 101 6 cm3/day. A cubic cm of rain weighs 1 gm.
Using the latent heat of condensation, this amount of rain produced gives 5.2 x 1019 Joules/day
or 6.0 x 101 4 Watts. This is equivalent to 200 times the world-wide electrical generating
capacity - an incredible amount of energy
produced!
Total kinetic energy (wind energy) generated:
For a mature hurricane, the amount of kinetic energy generated is equal to that being dissipated
due to friction. The dissipation rate per unit area is air density times the drag coefficient
times the windspeed cubed. One could either integrate a typical wind profile over a range of
radii from the hurricane's center to the outer radius encompassing the storm, or assume an
average windspeed for the inner core of the hurricane. Doing the latter and using 40 m/s (90
mph) winds on a scale of radius 60 km (40 n.mi.), one gets a wind dissipation rate (wind
generation rate) of 1.5 x 1012 Watts. This is equivalent to about half the world-wide electrical
generating capacity - also an amazing amount of energy - but 400 times less than that generated
by the condensation of water.
.

GG 454 April 9, 2002 6
I V Hurricanes vs. tornadoes

http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqA.html#A7
Property Tornado Hurricane
Diameter 100s of meters 100s of *kilometers
Produced by … a single convective several to dozens of
storm convective storms
Vertical Wind shear “Substantial” “low” ( < 10 m/s [20
kt, 23 mph])
Horizontal large temperature near zero
temperature fields gradient
Primary Location Over land Over oceans
Duration Minutes Days
Note: tropical cyclones at landfall often provide the conditions necessary for tornado formation.
As the tropical cyclone makes landfall and begins decaying, the winds at the surface die off
quicker than the winds at, say, 850 mb. This sets up a fairly strong vertical wind shear that
allows for the development of tornadoes, especially on the tropical cyclone's right side (with
respect to the forward motion of the tropical cyclone). For the southern hemisphere, this
would be a concern on the tropical cyclone's left side - due to the reverse spin of southern
hemisphere storms.

GG 454 April 9, 2002 7
I I I Effects of hurricanes
A Wind
1 Drives waves
2 Affects buildings
B Storm surge
http://www.aoml.noaa.gov/general/lib/stormsurgey.html
http://hurricanes.noaa.gov/prepare/surge.htm
http://www.ncstormsurge.com
1 A dome of water as much as 50 miles wide, that sweeps across

the coastline near where the eye of the hurricane makes landfall.
2 Can increase the water level by as much as 33’ (10 m)
3 The stronger the hurricane, the higher the storm surge will be.
4 The most dangerous part of a hurricane. Nine out of ten hurricane

fatalities are caused by the storm surge.
Contributions to storm surges. http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqC.html#C1
5 Depend upon the coastal topography, angle of incidence of

landfall, speed of tropical cyclone motion, as well as the wind
strength.
C Flooding
D Erosion

GG 454 April 9, 2002 8
http://www.ncstormsurge.com
The low pressure in
the eye and high
pressure outside
the eye causes the
water surface to
become elevated a
few meet at most
The winds in a
hurricane are what
pile up most of the
water in a storm
surge, which can
exceed 25’ in
height.

GG 454 April 9, 2002 9

GG 454 April 9, 2002 10
VII Case histories
A Hurricane Andrew: Category 4 (8/16/92- 8/28/92)

http://www.nhc.noaa.gov/1992andrew.html
http://marine.usgs.gov/fact-sheets/hurricane/hurricane-txt.html
http://www.eqe.com
Fatalities: 38 confirmed.
Damage estimate: ~$25 billion.
People left homeless: 175,000 to 250,000 people;

85,000 homes and apartments are uninhabitable, many totally destroyed.
600,000 homes and businesses have lost electrical and phone services
Area of heavy damage: more than 40 miles in diameter.
Utilities temporarily lost

Power, drinking water, communications, and sewage facilities (no pumps)
Underground water and gas distribution mains damaged by uprooted trees.
Street and highway signs blown away and traffic lights lost, hampering
rescue and relief efforts.
Florida Power & Light reports 12,620 miles of local power distribution
line and more than 631 miles of high-power feeder lines knocked down.
~1.4 million customers affected. Estimate for time of final utility
repairs: six months.
Turkey Point Generating Station (two nuclear and two fossil fuel units),
sustained damage that may keep it off-line for several months. One of the
plant's 400-foot-tall concrete exhaust stacks sustained heavy structural
damage and had to be demolished by explosives. A tank of heavy bunker
fuel spilled an estimated 12,000 gallons of fuel, some of which spread to
the ground near the nuclear reactors. The two nuclear units were shut
down as a safety precaution before the hurricane hit and sustained
minimal damage, although the fire protection system's water tank
collapsed, rendering the system inoperable for nearly a week. Damage to
the site was estimated at about $90 million.
In Louisiana, 13 offshore oil and gas production platforms were reported

as destroyed--four of which disappeared leaving no trace. Approximately
40 additional offshore drilling rigs and production platforms were also
damaged. It has been reported that 5% of the nation's gas supply has been
temporarily lost.

GG 454 April 9, 2002 11
B Hurricane Iniki: Category 5 (9/11/92)

http://members.aol.com/Rosendalhe/hurrican2.htm
http://www.eqe.com/publications/iniki/iniki.htm
Fatalities: 6
Injuries: > 100
Damage estimate: ~$3 billion
14,350 damaged or destroyed homes on Kauai (Red Cross figures).

~90% of the island's wood-frame buildings damaged
Homes completely destroyed: 1,421.
Homes destroyed by wave action or storm surge on the south coast: 63
Homes with major damage: 5,152
Homes with minor damage: 7,178
Electric power and telephone service were lost throughout the island, and
only 20 percent of power restored four weeks after the storm. Forty
percent of the island's power is generated by sugar cane waste-burning
cogeneration plants, which were undamaged and operational. Utility
substations also appear to have incurred little damage. Even though power
generation is mostly functional, downed poles prevent power distribution.
Hawaii Telephone reports that about 6,000 poles (both power and
telephone) collapsed. Replacement estimated at one to two months,
leaving approximately 80% of Kauai without power.
Relief efforts initially hampered by sparse facilities for aircraft and

ships, in addition to damage to communications and roadway systems.
Crop damage extensive. Sugar cane stripped or severely set back. Tender
tropical plants, such as banana and papaya, were destroyed. Fruit and nut
trees were broken or uprooted.
Several houses on the shoreline at Poipu collapsed and were pushed

several hundred feet inland, away from their foundations. Residential
construction generally not engineered, and poorly designed and built. Dry
rot was prevalent, especially in older homes that were damaged.

GG 454 April 9, 2002 12
Wind damage varied, affecting mostly wood-frame and older light metal
buildings that had deteriorated due to rust or were weakened by
architectural modifications. According to the Uniform Building Code, the
design wind for the region is 80 mph, which is only slightly higher than
that for California. Local, state, and federal facilities were well-
designed and built of concrete and steel, and some unreinforced masonry.
None of these structures were damaged.
Hurricance Inkiki, 9/11/92. http://www.geocities.com/drgeorgepc/

GG 454 April 9, 2002 13
Hurricane Iniki storm surge damage, Poipu beach, Kauai. SJM 2/10/02.
Iniki storm surge effects. http://www.geocities.com/drgeorgepc/

GG 454 April 9, 2002 14
C Closing comments
Both hurricanes Andrew and Iniki graphically illustrate the potential of hazardous natural
phenomena (such as hurricanes, earthquakes, and conflagrations) to significantly harm the
lives and economy of a region. Nonetheless, the widespread extent of the damage caused by
Andrew is surprising because the sustained wind speed was within the design criteria of the
South Florida Building Code. This code is the most stringent code for wind loads in the United
States, so a greater proportion of buildings and structures should have withstood the
hurricane with little or no damage.
Coverage gaps in the code, such as mobile homes, and the frequency of non-engineered
structures put many people and their property unwittingly at risk. Most non-coastal damage
was caused by rain and wind infiltration of buildings. The code and its enforcement could be
improved to better safeguard the integrity of building envelopes, and thus significantly reduce
damage in future hurricanes.
Iniki, in striking Kauai, missed the state's major population center on Oahu. Observed damage
indicated that engineered structures withstood the wind forces well, and damage was confined to
those of poor design and construction. This was also true of Andrew, showing that the knowledge
exists to engineer economical structures that will withstand winds of the forces seen in Andrew
and Iniki.
Power and telephone services could have been restored quickly after both hurricanes had the
aboveground utility poles remained intact. Utility poles by their very nature are susceptible to
wind forces, but had these utilities been buried, there would have been minimal disruption of
services.
Hurricane risks, like those from fire and earthquake, are quantifiable and controllable.
Appropriate decisions with regard to siting, design, construction, and improving facilities can
provide good protection from such losses.

GG 454 April 9, 2002 15
Hurricane risk
http://www.colorado.edu/hazards/wp/wp94/wp94.html

GG 454 April 10, 2002 1
HAZARDS AND RISKS ALONG COASTS (34)
I Main topics
A Hazard recognition
B Characterization of hazards
C Risk evaluation
D Risk Assessment
I I Hazard recognition
A Hazards peculiar to coasts
1 Coastal erosion
2 Coastal deposition
3 Hurricanes
4 Tsunami
B Hazards accentuated at some coasts
1 Sea cliff retreat/sea cliff failure
2 Wind damage
3 Quick clays (where young sediments are uplifted)
4 Seismic shaking (because of proximity to subduction zones)
I I I Characterization of hazards and processes
Dynamic competition between waves and sediment supply
A Waves
1 Underwater nearshore topography
2 Continental shelf topography
3 Changes in relative sea level
4 Wave statistics
a Direction
b Period
c Height
d Pressure and Forces (6 tons/ft2 measured = 0.7 MPa)

GG 454 April 10, 2002 2
B Sediment
1 Sources
a Beach steepness
b Composition and grain size (90%+ are sand); coarse⇒ steep
c Human intervention
2 Sinks
a Submarine canyons
b Human intervention
3 Sand motion perpendicular to shore

a Flat waves (H/L < 0.025): sand moves onshore
b Steep waves (H/L < 0.025): sand moves offshore
4 Longshore drift rate (rate of sand movement in surf zone)

a Longshore currents set up by oblique approach of waves
b Swash + Backwash ⇒ longshore drift
c Example: 1,000,000 m3 /yr at Oxnard, So. Cal.)
5 Biology (e.g., kelp)

6 Geologic History
a What has happened can happen
b Helps to pin down the high-energy, low frequency events
c Paleo-hurricane studies now underway
B Hazards and processes accentuated at some coasts
1 Sea cliff failure: Landslides
2 Wind damage: Meterology
3 Seismic shaking (liquefaction): Distribution of faults and
Quaternary deposits
I V Risk evaluation
A Uncertainty regarding high-energy, low frequency events
B We commonly don't design for these (too expensive)
C These represent the tails of probability distributions

GG 454 April 10, 2002 3
V Risk Assessment: Is the level of risk acceptable

A Common answer: yes (in part because infrequent events are ignored)
B Coastal engineering structures for which risks can be assessed
1 Harbors
2 Buildings
3 Piers
4 Pipelines (e.g., oil, sewage) and cables
5 Floating structures (e.g., drill rigs)
6 Seawalls, jetties, breakwaters, groins, rip rap
C Engineering solutions
1 Location and orientation
a Don't locate structures where wave energy is focused
b Don't orient structures parallel to prevailing waves
c Jetty spacing
i Natural streams commonly near equilibrium in terms of
channel with and cross section dimensions
i i Tidal prism volume/cross section of entrance
i i i Tidal prism volume = average volume that flows in and out
in a tidal cycle. Volume between higher high water and
lower low water.
i v Spacing too high → sedimentation
v Spacing too low → erosion
2 Ability to reflect or absorb energy
a Shape
i Uniform shapes not good
i i Stepped shapes are good for breakwaters
b Armor
i Dense
i i Permeable
iii Not rectangular blocks
i v Problem: Commonly lead to beach loss
3 Dredging and replenishment of sand
4 Build with nature as much as possible
D Insurance

GG 454 April 12, 2002 1
SUBSIDENCE I (35)
I Main Topics
A Recognition of subsidence hazards (Case histories)
B Characterization of subsidence hazards
C Evaluation of subsidence hazards (Subsidence mechanics)
D Assessment and mitigation of subsidence hazards
II Recognition of subsidence hazards
http://water.wr.usgs.gov/sub/
http://water.usgs.gov/pubs/circ/circ1182/
http://www.env.go.jp/en/soe/ground.html
http://books.nap.edu/books/POD309/html/45.html
A Distribution of hazard
1 World-wide: see highly incomplete distribution below
2 U.S.
a >44,000 km2 (See Figure 1 from Panel on Land Subsidence)
b $125 million annually (as of 1991)
B Fluid withdrawl from porous media
http://geo0.eng.morgan.edu/center/gallery-hazarWD.html
1 Extraction of oil and gas (e.g., Long Beach, CA; Niigata, Japan)
2 Extraction of geothermal fluids
a Wairakei, NZ. 4.5m subsidence 1964-1975
b The Geysers, CA. 13 cm subsidence 1973-1977
3 Extraction of ground water (e.g., Tokyo; Mexico City; Venice; Santa
Clara and San Joaquin Valleys, CA; Houston, TX)
C Subsidence of organic soils

(e.g., Mississippi river Delta; Everglades, FL; Sacramento Delta, CA)
1 Magnitude: near 10 meters in the 20th century near Sacramento
2 Causes
a Void loss due to water withdrawl
b Compaction due to plowing
c Wind (and water) erosion
d Burning
e Biochemical oxidation

GG 454 April 12, 2002 1
SUBSIDENCE II (36)
I Main Topics
A Recognition of subsidence hazards (Case histories)
II Recognition of subsidence hazards
A Collapse into man-made and natural cavities
http://www.ene.ttu.ee/maeinstituut/poster/rez.html
1 Subsidence over coal mines (e.g., Pennsylvania)
http://www.netcentral.co.uk/steveb/geography/collapse.htm
http://gsa.confex.com/gsa/2001NC/finalprogram/abstract_5641.htm
2 Sinkholes in carbonate terrains (e.g., Florida)
http://landslides.usgs.gov/html_files/landslides/slides/slide10.htm
http://waterquality.ifas.ufl.edu/PRIMER/sinks.html
3 Extraction/dissolution of salt (e.g., New Mexico, Texas, Kansas)
http://www.sigmaxi.org/amsci/articles/98articles/martinez.html
B Hydrocompaction (Loess areas, San Joaquin Valley, CA)
Collapse of soil structure due to wetting
http://ncgmp.cr.usgs.gov/ncgmp/cencol/cencol.htm
C Natural compaction of sediments (e.g., Mississippi Delta)
http://abcnews.go.com/sections/science/DyeHard/dyehard000202.html
D Tectonic subsidence
1 Alaska, 1964
2 Hawaii, 1975
E Volcanic subsidence
1 Calderas (e.g., Kilauea).
Due to magma withdrawl
2 Pit craters (e.g., Devils Throat).
Due to stoping into subterranean fissures
http://www.pgd.hawaii.edu/~chriso/pubs/gsa97.html
http://volcanoes.usgs.gov/Products/Pglossary/PitCrater.html
I I I Characterization of subsidence hazards
A Conventional surveying
B GPS
C Interferometric Synthetic Aperture Radar (InSAR)
http://water.wr.usgs.gov/program/desert/insar/

GG 454 April 17, 2002 1
SUBSIDENCE MECHANICS: CONSOLIDATION (37)
I Main Topics
A Consolidation vs. compaction
B Mechanics of subsidence due to pore pressure changes
(Evaluation of subsidence)
I I Consolidation vs. compaction
A Consolidation: Defined by Lambe and Whitman (1969) as a decrease
in volume resulting from dissipation of excess pore pressure.
B Compaction: Defined by Lambe and Whitman (1969) as a densification
process involving mechanical equipment, usually a roller, and as
distinguished from pre-loading and dewatering.
I I I Mechanics of subsidence due to pore pressure changes
A Stages of consolidation
1 Initial consolidation: Void volume loss due to air loss
2 Primary consolidation: Void volume loss due to pore fluid loss
a Fluid "loss" requires fluid flow
b Flow reflects changes in pore pressure/effective stress
c Flow and pore pressure dissipation takes time, so primary
consolidation is time-dependent
3 Secondary consolidation: due to decrease in solid volume
B Pressure here is considered to be positive

GG 454 April 17, 2002 2
C Hydromechanical analog for consolidation

(from Lambe and Whitman, 1969)
Initial
Porous
water surface A Piston Valve
piston
Poro-elastic Spring
medium
Soil Water
Physical Example Hydromechanical Analog

initial state
Valve
Valve open;
open; water
water flow
flows stops
B Valve C D Water
closed level
Water
Spring Water Spring Spring
pressure is not pressure compresses supports
rises compressed drops Water load
to pressure
support is
load hydrostaic
Hydromechanical Analog Hydromechanical Analog Hydromechanical Analog

Load applied, valve closed Valve open, water flows Equilibrium
Variation of Loads in Spring and Water
In spring
Load Applied
load
In water
tB tC tD
Time

GG 454 April 17, 2002 3
D Consolidation as a reflection of poro-elastic strain
H1 − H 0 ∆H
∆H Hs∆e = =ε
Voids Hse0 H0 H0
Voids Hse1
H s∆e
ε=
H s (1 + e0 )
H0 H1
Solids Hs Solids Hs ∆H ∆e
=
H1 1 + e0
Initial state Final state
∆ H = final height - initial height = H 1 - Ho ∆H is negative here
Settlement = -∆H Settlement is positive here

∆H = (Ho ) (vertical strain) = Ho (∆H/Ho ) (37.1)
For 1-D consolidation

∆H = (Ho ) (vertical strain) = Ho (∆e/[1+eo ]) (37.2)
where e = void ratio; ∆e < 0

The change in void ratio scales approximately with the change in effective
stress, or the change in excess pore pressure: ∆e ∝ ∆σz ' or ∆e ∝ ∆ue
Two similar approaches relate ∆e to ∆σz '; both utilize the following
∆H = Ho (∆e/{1+eo }) [∆σz ' / ∆σz ']. (37.3)
∆H = [-Ho (- ∆ e/{1+e o }) /∆σz '] ∆ σz ' = -Ho (mv ) ∆σz ' (37.4)
where
m v = -(∆e/{1+e o }) / ∆σ z ' = -vertical strain: change in effective stress
m v = coefficient of volume change
∆H = (-Ho /{1+eo }) (- ∆ e/ ∆σz ') ∆σz ' = (-Ho /{1+eo }) (av ) ∆σz ' (37.5)
a v = -∆e/ ∆σ z ' = -void ratio change: change in effective stress
a v = coefficient of compressibility

GG 454 April 17, 2002 4
E The consolidation ratio U
U(t) = ∆H(t)/∆Hmax = ∆e/∆em a x (37.6)

Solving for ∆H using (37.4), but using ∆ue in place of -∆σ z ', yields
U = ∆H/∆Hmax = (Ho m v ∆ u e ) / (Ho m v ∆ u e max) (37.7)
U = ∆H/∆Hmax = ∆ue / ∆ u e max (37.8)
∆ue (negative)
Normal
u ∆ue (ultimate)
stress (or o ue
pressure)
at depth
h
utotal uhyd uequilibrium
Time
The excess pore pressure u e is what must be bled off for the water
pressure to return to an equilibrium (hydrostatic) case. The hydrostatic
pressure is ρ water gh; h is the depth below the water table. The term u o
is the initial excess pore pressure. The change in excess pore pressure
∆ u e is measured relative to uo and can be defined in three ways.
∆ u e = ue |t - ue |t=0 . (37.9a)
∆ u e = (u|t - uhyd ) - (u|t=0 - uhyd ) = (u|t - u|t=0 ) = ∆u (37.9b)
∆ u e = (u|t - u|t=0 ) = (σtotal | t - σ'| t ) - (σtotal | t=0 - σ'| t=0 ) = -∆σ' (37.9c)
As the pore pressure drops, ∆ue is negative and ∆σ ' (the change in
effective stress) is positive. Also, ∆ue is a maximum when ue is zero, so
the maximum change in ue is -uo : the negative of the initial excess pore
pressure.
∆ u e max = -uo (37.10)
Substituting (37.9a) and (37.10) into (37.8)
U = ∆ue / ∆ u e max = (ue - uo )/(-u o ) (37.11)

GG 454 April 17, 2002 5
This reduces to
U = 1- (ue /uo ) (37.12)
Equating (37.7) and (37.12),

U = ∆H/∆Hmax =1- (ue /uo ) (37.13)
Note: U does not need to be constant throughout a consolidating column of
material; U usually varies with position. Use the average consolidation
ratio in a column of material to find the height change for the column.

GG 454 April 24, 2002 1
SUBSIDENCE MECHANICS: HEAT FLOW ANALOG (38)
I Main Topics
A Motivation: Why investigate the flow of heat?
B Development of 1-D heat flow equation as analog for consolidation
C Solution of heat flow equation using finite-difference approximation
I I Motivation: Why investigate the flow of heat?
A The form of the heat flow equation (i.e., diffusion equation) is
identical to the 1-D consolidation equation but is easier to
understand and derive from a mechanical standpoint
B Diffusion of heat is analogous to diffusion of excess pore pressure
(i.e., pore pressure in excess of hydrostatic pressure)
C Hundreds of analytic solutions for heat flow problems exist; (see
Carslaw, H.S., and Jaeger,J.C., 1984, Conduction of heat in solids,
Clarendon Press, Oxford, 510 p.)
D The diffusion equation for the diffusion of solutes in water also has
the same form as the heat equation, so solutions to the heat
equation have many varied and important practical uses.
Conduction Analogies
Fluid Heat Electrical Chemical
Potential Total head h Temperature T Voltage V Concentration c
(cm) (°) (volts) (moles/cm 3 )
Storage Fluid volume W Thermal energy Charge Q moles
(cm 3 / c m 3 ) U (cal/cm 3 ) (Coulombs)
Conductivity Hydraulic k Thermal k Electrical Diffusion coeff.
(cm/sec) (cal/°/cm/sec) (volts/sec) (moles/sec)
Flow (discharge) Q (cm3/ s e c ) Q (cal/sec) Current I (amps) J (moles/ sec)
Flux (Spec. dis.)
q (cm/sec) 2
q (cal/[sec • cm ]) amps/cm 2 -1 2
moles sec cm
Gradient ∂h ∂T ∂V ∂c
i=− i=− i=− i=−
∂x ∂x ∂x ∂x
(cm/cm) (°/cm) (volts/cm) (moles/ cm 4 )
Conduction Darcy’s Law Fourier’s Law Ohm’s Law Fick’s Law
∂h ∂T ∂V V ∂c
Q = −k A Q = −k A I = −σ A= J = −D A
∂x ∂x ∂x R ∂x
Capacitance Coefficient of Volumetric heat Capacitance C ---------
volume change C (cal/°/ cm3 ) (farads)
∇2h = 0 ∇ 2T = 0 ∇ 2V = 0 ∇ 2c = 0
Steady state flow
1-D diffusion
∂h ∂2h ∂T k ∂ 2T ∂V σ ∂ 2V ∂c ∂ 2c
= cv = = =k
∂t ∂x 2 ∂t C ∂x 2 ∂t C ∂x 2 ∂t ∂x 2

GG 454 April 24, 2002 2
I I Development of 1-D heat flow equation as analog for consolidation

A Isotropic, uniform material
B Definition of terms
1 Position (x) Dimensions of meters
2 Time (t) Dimensions of seconds
3 Heat flux (q)
a Rate of heat energy transfer across a given unit area per unit
time. The flux can vary with time and position, so q = q(x,t).
b Dimensions: Joules/(meter 2 . sec)
c 1 heat flow unit (hfu) = 10-6 cal cm-2 sec-1 = 0.04184 W m- 2
4 Temperature (T) Dimensions of degrees (°K)
Temperature can vary with time and position, so T = T(x,t).
C Fourier's Law of Heat Conduction (1-D)
∂T
q = −k
∂x (38.1)
∂T
where ∂ x = temperature gradient
1 Heat flow is proportional to the temperature gradient
2 k = coefficient of thermal conductivity
Dimensions: Joules sec -1 m-1 °K- 1
k assumed to be constant [i.e., k ≠ k(x); k ≠ k(t);k ≠ k(T)]
∂T
3 Dimension check: q = −k
∂x
Joules  Joules   ° K 
=  
meter 2 sec  sec m ° K   m 
4 Partial derivative used because T is a function of x and t.
5 The minus sign: for heat to flow from x1 to x2 , where x1 <x 2 ,
T(x 1 ) > T(x2 ). So positive heat flow corresponds to a drop in
temperature.
∆T
6 Finite difference approximation: q = −k
∆x

GG 454 April 24, 2002 3
6 Analog: Darcy's Law

∂H
q = −k
∂x (38.2)
q = volumetric flow rate per unit area (specific discharge)
k = hydraulic conductivity
∂H
= head gradient
∂x
C The heat equation
1 Conservation of energy (see handout)
∂ T ∂ 2T
2 1-D form: C = a parabolic differential equation
∂t ∂x2
∂ T ∂ 2T ∂ 2T
3 2-D form: C = +
∂ t ∂ x 2 ∂ y2
∂ T ∂ 2T ∂ 2T ∂ 2T
4 3-D form: C = + +
∂ t ∂ x 2 ∂ y2 ∂ z 2
5 Laplace Equation (2-D or 3-D)
Applies to steady state distribution of temperature
a Temperature does not change as a function of time
∂T
b =0
∂t
∂ 2T ∂ 2T ∂ 2T ∂ 2T ∂ 2T
c 0= + or 0 = + 2 + 2 or 0 = ∇2T
∂x 2 ∂y 2 ∂x 2 ∂y ∂z
d In finite-difference form, the Laplace equation means that the
value of the function T (here T=temperature) at a given point is
the average of the values at the nearest neighboring points,
where the points are on a square grid (see notes on wave eqn).
6 Initial conditions and boundary conditions
The equations above indicate how temperature will change as a
function of time and position within a body. If we know the
temperature conditions at the boundaries of a body at specific
points in time and space, then we can solve these equations to find
the temperature distribution within the body

GG 454 April 24, 2002 4
DERIVATIVES AND PARTIAL DERIVATIVES

u
∆t
u2
u = u(t)
∆u
u*
u1
t/2 t/2
t t* t t
1 2
∆u= u -u du/dt | = du/dt | = du/dt |

2 1 t1 t* t2
∆t= t - t
2 1 However, du/dt | ≈ u - u , and
t* 2 1
∆ t/2= t* - t = t - t*
1 2
t 2- t1
du/dt | ≈ du/dt | t + du/dt | t

t*
1 2
2
THE PARTIAL DERIVATIVE
(u is a function of more than one variable)
6 u/ 6 t
u = u(z,t) t = constant along this curve
u t
6 u/ 6 z
z = constant
along this curve
z

GG 454 April 24, 2002 5
Heat Flow
1-D heat flow T = Temperature

T1 > T2
q = -k (6T/6x)
q
q = heat flux
T1 T2
∆t = increment of time
∆x
Area A
x 1 x2 x
Change in heat energy = heat in - heat out

(in the slab)
(∆T)(mass)(spec. heat) = A∆t [q(x1) - q(x2)] = A∆t [q(x1) - {q(x1)+∆q}]
(∆T)(ρ V)(c) = -A∆t [∆q]
(∆T)(ρ A∆x)(c) = -A∆t [( 6q / 6x) ∆x ], but q = -k 6T / 6x, so
(∆T)(ρ A∆x)(c) = A∆t k[ ( 62T / 6x2) ∆x ]

If the temperature profile isn't curved (i.e., if the underlined term
is zero), then there can be no change in heat energy in the slab.
(∆T/∆t)(ρ c/k) = ( 62T / 6x2) K = ρ c/k = diffusivity
(6 T/6 t)(ρ c/k) = ( 62T / 6x2) 1-D Heat Flow Eqn

To a good approximation, the rate of temperature change with time
is proportional to the curvature of the temperature profile.

GG 454 April 24, 2002 6
EFFECT OF TEMPERATURE GRADIENT CHANGES

ON HEAT FLOW
Heat in = Heat out
Temperature profile
Temperature
is not curved
-Temperature gradient = -Temperature gradient

going in going out
No heat is stored in the box
Heat in < Heat out
Temperature profile
Temperature
is concave-down
-Temperature gradient < -Temperature gradient

going in going out
Heat is lost from the box
Heat in > Heat out
Temperature profile
Temperature
is concave-up
-Temperature gradient > -Temperature gradient

going in going out
Heat is gained in the box

GG 454 April 24, 2002 7
Fluid Flow
1-D fluid flow H = head

H 1 > H2
q Q = volumetric flow rate

Q = -k ( 6H/6 x) A
H1 H2
∆t = time increment
∆x Area A
x1 x 2 x
Change in fluid volume = volume in - volume out

(in the slab)
(∆Vw) = ∆t [Q(x1)-Q(x2)] = ∆t [Q(x1)- {Q(x1)+∆Q}]
(∆Vw) = ∆t [-∆Q]
(∆Vw) = −∆t [(6 Q /6x) ∆x ], but Q = -kA 6H /6 x, so
(∆Vw) = kA ∆t [( 6 2H /6 x2) ∆x ]
If the head profile isn't curved (i.e., if the underlined term
is zero), then there can be no change in water volume in the slab.
(∆Vw/∆Vbox)/∆t) = k( 6 2H /6 x2)
To a good approximation, the rate of fluid volume change with time

is proportional to the curvature of the head profile.

GG 454 April 24, 2002 8
EFFECT OF HEAD GRADIENT CHANGES

ON FLUID FLOW
Fluid in = Fluid out
Head profile
is not curved
Head
-Head gradient = -Head gradient
going in going out
No fluid is stored in the box
Fluid in < Fluid out
Head profile
is concave-down
Head
-Head gradient < -Head gradient

going in going out
Fluid is lost from the box
Fluid in > Fluid out
Head profile
is concave-up
Head
|Head gradient > -Head gradient

going in going out
Fluid is gained in the box

GG 454 April 17, 2002 1
NUMERICAL SOLUTION OF THE 1-D DIFFUSION EQUATION (39)
I Main Topics
A Motivation: Why use a numerical technique?
B Non-dimensionalizing the diffusion (heat flow) equation
C Solution of heat flow equation using finite-difference approximation
See the figures on pages 5 and 8!
I I Motivation: Why use a numerical technique?
A It provides a useful alternative insight into the second order PDE
(partial differential equation) for transient flow
B It provides useful insight into the influence of different initial
conditions and boundary values on the solution to the equation
C It is useful for investigating a wide range of initial value/boundary
value combinations and geometries. In contrast, analytical solutions
are available only for a small range of initial value/boundary value
combinations and for simple geometries.
D Finite-difference technique is a good learning tool, in spite of its
limitations (e.g., need for fine mesh locally, numerical errors).
III Non-dimensionalizing the heat flow equation
∂T ∂2T
= κ
∂t ∂x2 (39.1)
where T = temperature, κ is the thermal diffusivity (= 1/C of lecture 38),
and x is position. T and x are variables, and we assume K is a constant.

"Nondimensionalizing" (or scaling) the equation eliminates the term K.
Let x* = x/xm a x , t* = Kt/xm a x 2 , and T* = T/Tm a x
or x = x* xmax , t = t* xmax 2 /K, and T = T* Tmax .
Note that all the starred terms are dimensionless numbers. The terms
x max and Tmax are the scaling terms and are constants. The chain rule
will be used to rewrite (39.1), so we take derivatives of x, t, and T:
dx/dx* = xmax = 1/(dx * /dx) = 1/(1/xmax ) (39.2)
dt/dt* = xmax 2 /k = 1/(dt * /dt ) = 1/( k/xmax 2 ) (39.3)
dT * /dT = 1/Tmax (39.4)

GG 454 April 17, 2002 2
We will re-write the left side of (39.1) first - it involves only a first-
order derivative. We will then attack the “harder” right side of (39.1).
Now the chain rule is brought to bear. In the equations below pay
attention to which terms are derivatives and which are constants.
∂T * ∂T dT * dt ∂T 1 ( x max )
2
= =
∂t * ∂t dT dt * ∂t Tmax κ (39.5)
∂T
We solve for to express the left side of (39.1) in dimensionless terms.
∂t
∂T ∂T * κ
= (Tmax )
∂t ∂t * ( xmax )2 (39.6)
We are done with the left side of (39.1) and move to the right side.
∂T * ∂T dT* dx ∂T 1
= = x
∂x * ∂x dT dx* ∂x Tmax max (First derivative) (39.7)
 ∂T *   ∂T *   ∂T 1 
∂  ∂  ∂ x max 
∂2 T *  ∂x *  =  ∂x *  dx  ∂x T max 
= =  xmax
∂x *2 ∂x* ∂x dx* ∂x (39.8)
∂2 T * ∂2 T 1
2
=
2 ( xmax )2
∂x * ∂x Tmax (39.9)
Substituting (39.6) and (39.9) into (39.1) yields:
∂T * κ ∂2T * Tmax
(Tmax ) =κ
∂t * ( x max )2 ∂x *2 ( x max )2 (39.10)
By eliminating the three terms common to both sides, this reduces to
∂T * ∂2T *
=
∂t * ∂x *2 (T*, x*, and t* are dimensionless) (39.11)

GG 454 April 17, 2002 3
III Finite-difference solution to the 1-D heat equation (diffusion equation)

The first step we take is to set up a dimensionless grid. The size of
our ∆ x* and ∆ t* steps are equal, so this is a square mesh (note: this could
mean that either ∆x or ∆t is awkwardly small):
x* T*(x*,t*) = T*i,j
4
3
i-va;ies
t*
0
0 1 2 3 4 5 6 7
j-values
There are two terms that need to be approximated: ∂T*/∂t*, and
∂ 2 T*/ ∂ x* 2 . We make use of the definition of a partial derivative to do
this (also see the next figure).
∂T * [T * (x*,t * +∆t*) − T * (x*,t*)]
≈
∂t * ∆t * (39.12)
In terms of the grid parameters i and j, this is
∂T * [T *i, j+1 −T *i, j ]
≈
∂t * ∆t * (39.13)
As to the second derivatives
 ∂T * 
2 ∂ 
∂ T*  ∂x * 
≈ (39.14)
∂x *2 ∂x *

GG 454 April 17, 2002 4
 ∂T *   ∂T * 
  − 
∂2 T *  ∂x *  i+1/2, j  ∂x *  i−1/2, j
≈
∂x *2 ∆x * (39.15)
The terms in the square brackets are partial derivatives, the first taken
at point (i+1/2, j) and the second at point (i-1/2, j).
T*i+1,j ∆t* The second partial derivative of T*

with respect to x* at the point (i,j) is
T*i+1/2,j approximately equal to the change in
∆x*
the first partial derivative with
T*i,j T*i,j+1
respect to x* between points (i+1/2,j)
T*i-1/2,j and (i-1/2,j), divided by ∆ x* [the
distance in space between the points

T*i-1,j (i+1/2,j) and (i-1/2,j)].
T *i +1, j −T *i, j T *i, j −T *i −1, j

−
∂2 T * ∆x ∆x *
≈ *
∂x *2 ∆x * (39.16)
∂2T * T *i+1, j −2T *i, j +T *i−1, j
≈
∂x *2 (
∆x * 2 ) (39.17)
So equation (39.11) can be written in finite difference form as
T *i, j +1 − T *i, j T *i+1, j −2T *i, j +T *i−1, j
≈
∆t * (∆x *)2 (39.18)
This equation can be solved for T * i , j + 1 :
    
 ∆t * 

T *i, j+1 ≈ 1− 
2∆t * 
 T *i, j + 
  ( ∆x *) 2  
[ T *i+1, j +T *i−1, j
 ( ∆x *) 2 
]
  (39.19)

GG 454 April 17, 2002 5
The terms in the braces can be considered as weighting terms. In that

context, equation (39.19) says (see the above figure!) that the value of
T * i,j+1 can be obtained from a weighted average of the nearest three
points at the previous time step. In practice, this only works if
∆ t*/[ ∆ x*] 2 ≤ 1/2 (i.e., ∆t* ≤ 1/2 [∆x*]2 ), so ∆ t* needs to be impractically
tiny. This is because the approximations used here for the partial
derivatives aren't sufficiently accurate unless ∆ t* is really small.
To do a better job, we use the Crank-Nicholson technique for
approximating the partial derivatives. This technique relies on using the
values at four surrounding nodes instead of three, and leads to an
amazingly simple equation that gives a considerable amount of insight
into the nature of our differential equation.
T*i+1,j ∆t* T*i+1,j+1
T*i+1/2,j ∆x* T*i+1/2,j+1
T*i,j T*i,j+1
T*i-1/2,j T*i,j+1/2 T*i-1/2,j

T*i-1,j T*i-1,j
The Crank-Nicholson technique works by finding the derivatives at point

(i, j+1/2), that is, at the midpoint on the line shared by the two boxes
show above, instead of at point (i, j). This is because equation (39.13)
∂T * [T *i, j+1 −T *i, j ]
≈
∂t * ∆t * (39.13)
is a better approximation of the derivative at point (i, j+1/2) than at point
i,j (see the handout on derivatives).

GG 454 April 17, 2002 6
To evaluate ∂ 2 T*/ ∂ x* 2 at point (i, j+1/2), we take the average of

∂ 2 T*/ ∂ x* 2 at points (i, j) and (i, j+1). The value of this derivative at
point (i,j) is:
∂2T * T *i+1, j −2T *i, j +T *i−1, j
≈
∂x *2 i, j (∆x * 2 )
(39.20)
The value of ∂2 T*/ ∂ x* 2 at point (i, j+1) is approximated by permuting the
indices in the previous expression:
∂2T * T *i+1, j +1 −2T *i, j +1 +T *i−1, j +1
≈
∂x *2 i, j + 1 (∆x *) 2
(39.21)
The value of ∂2 T*/ ∂ x* 2 at point (i, j+1/2) is thus approximated by:
∂2T * 1 ∂2T * 1 ∂2T *
≈ +
∂x *2 i, j +1 / 2
2 ∂x *2
i, j
2 ∂x *2
i, j +1 (39.22)
Because of the inherently greater accuracy of this technique, we can let
∆t* = ∆x* (we couldn't do this with our first technique; see the comment
on the previous page). We can pick values of ∆t and ∆x that give us

whatever values we wish for ∆t* and ∆x*. We now set ∆t* and ∆x* equal
to one, and the expressions for the derivatives simplify greatly.
∂T *
≈ T *i, j+1 − T *i, j
∂t * i, j +1 / 2
(39.23)
∂2T *
≈ T *i+1, j − 2T *i, j + T *i−1, j
∂x *2 i, j (39.24)
By permuting the indices of (24) we obtain
∂2T *
≈ T *i+1, j+1 − 2T *i, j+1 + T *i−1, j+1
∂x *2 i, j +1 (39.25)
Now use equation (39.22) to find ∂2 T*/ ∂ x* 2 at point (i, j+1/2)

GG 454 April 17, 2002 7
∂2T *
∂x *2
≈
1
2
{
T *i+1, j − 2T *i, j + T *i−1, j }
i, j +1 / 2
+
1
2
{T *i+1, j+1 − 2T *i, j+1 + T *i−1, j+1 } (39.26)
Now expressions (39.23) and (39.26) are set equal to give the finite
difference approximation of equation (39.11)
T *i, j+1 − T *i, j ≈
1
2
{
T *i +1, j − 2T *i, j + T *i −1, j }
1
{
+ T *i +1, j +1 − 2T *i, j +1 + T *i −1, j +1
2
} (39.27)
Multiplying both sides of (39.27) by 2
{
2{T *i, j+1 − T *i, j } ≈ T *i +1, j − 2T *i, j + T *i −1, j }
{
+ T *i +1, j +1 − 2T *i, j +1 + T *i −1, j +1 } (39.28)
Now like terms are collected and the terms - 2T*i , j on both sides of
(39.28) are dropped.

{T *i+1, j } { }
+ T *i−1, j + T *i+1, j+1 + T *i−1, j+1 − 4T *i, j+1 = 0
(39.29)
Now the term T * i , j+1 is solved for
1
4
{ }
T *i+1, j + T *i−1, j +T *i+1, j+1 + T *i−1, j+1 = T *i, j+1
(39.30)
Equation (39.30) states that equation (39.11) can be thought of as saying

the value of T at any node is equal to the average value of the two
adjacent nodes at the same time step and the two nodes at the preceding
time step! This is a startlingly simple way to view the second order
partial differential equation we began with. No derivatives are involved
(at least not explicitly) in (39.30). To see this point graphically, examine
the above diagram. Two other key points are: (1) one can see how the
state of the system at one step will effect the state at a subsequent time
step, and (2) one can see how the boundary conditions effect the interior.

GG 454 April 17, 2002 8
Two closing comments. First, the finite-difference solution of the partial

differential equations presented here turns out to be essentially an
averaging procedure, and this is a useful way of viewing the way the
system is trying to respond (i.e., how "information" propagates within the
system). Second, a side-by-side comparison of the finite-difference
solutions for the Laplace equation and the heat equation is interesting:
Laplace Equation 1-D Heat (Diffusion) Equation

∂ 2T ∂ 2T ∂ 2T ∂ T
+ =0 =
∂ x 2 ∂ y2 ∂x2 ∂t
∆y ∆x
∆x ∆t
In both cases, the value of T at the node marked by a diamond is

approximately the average of the T values at the circled nodes.
The expressions on the first page of this lecture would be used to convert
the dimensionless solutions to dimensioned solutions:
x = x* xmax,
t = t* xm a x 2 / K ,
and
T = T* Tmax

GG 454 April 24, 2002 1
TERZAGHI’S 1-D CONSOLIDATION EQUATION (40)
I Main Topics
A The one-dimensional consolidation equation analog to heat flow
C Calculating consolidation for double-sided drainage
I I The one-dimensional consolidation equation analog to heat flow
Sand
H0
z + dz Fluid flow CLAY Z-z dz = h0 } Clay slice
z Z
LAYER Midplane of
Fluid flow
clay layer
Sand
Saturated clay layer (of thickness H0) with double drainage

At time t=0, a pressure P is applied to the top of our sand-clay-sand
sandwich. The increase in load will be initially born by the water; the
water pressure in all the layers goes up. This excess pore pressure (water
pressure above the hydrostatic [equilibrium] level) will dissipate quickly
in the sand layers because the water flows rapidly through the high-
permeability sand (it flows sideways in a "violation" of our one-
dimensional assumptions). The excess pore pressure will dissipate slowly
in the clay. The water in the clay has to flow vertically because of the
long, slow horizontal flow path in the clay. Experience shows that the
sand will consolidate little. Individual sand grains are stiff, and their
volumes change little with the applied loads. Unless the sand grains
change their packing, the collective volume of the sand won’t change
significantly either. The clay, however, will consolidate significantly.

GG 454 April 24, 2002 2
We start by looking at how the water flows, using Darcy’s Law.

Q = - kiA (40.1)
Q=discharge; k=conductivity; i=head gradient; A=Area
Dimensions
Q: L3 /t k: L/t i: L/L (dimensionless) A: L2
q = Q/A = flux = - ki (40.2)
q = unit discharge (discharge/unit area). Dimensions of velocity.
Now we investigate the head gradient, which drives fluid flow:
i= = =
( [
∂H ∂ ( z + [u / ρg]) ∂ z + (u hydrostatic + uexcess ) / ρg ])
∂z ∂z ∂z (40.3)
The excess pore pressure (uexcess ) in (40.3) is the difference between the
actual pore pressure (u) and hydrostatic pressure (uhydrostatic ). The rate
of change in the elevation head (z) and the hydrostatic pressure head
cancel each other out exactly:
( [
∂ z + u hydrostatic / ρg ]) = ∂ (z + [ρg(Z − z) / ρg]) = ∂ (z + (Z − z)) = ∂Z = 0
∂z ∂z ∂z ∂z
Without this cancellation, then water at the bottom of a still swimming
pool might flow to the top of the pool! So equation (40.3) simplifies:
i= =
( [
∂H ∂ z + (u hydrostatic + uexcess ) / ρg
=
])
1 ∂ue
∂z ∂z ρg ∂z (40.4)
The head gradient at the base of the slice (i.e., at elevation “z”) is:
1 ∂ue
i1 =
ρg ∂z (40.5)
The head gradient at the top of the slice (i.e., at elevation “z+dz”) is:
1 ∂ 1 ∂  ∂ue  1  ∂ue ∂ 2 ue 
i2 = (
ρg ∂z e
u + ∆ u e ) ρg ∂z  e
=  u + dz  =  + dz 
∂z  ρg  ∂z ∂z 2
 (40.6)
The change in unit discharge [i.e, the net flow of water out of the clay
slice (per unit area)] reflects the water loss in the slice (per unit area):
−k ∂ 2 ue
∆q = q2 − q1 = −k (i2 − i1) = dz
ρg ∂z 2 (40.7)

GG 454 April 24, 2002 3
Two comments. First, we have held k, the hydraulic conductivity, constant

- is that OK? Second, the right hand side of (40.7) looks somewhat like
one side of the heat equation. What about the left side? The change in
unit discharge of the slice times the area (A) of the slice gives the water
volume loss with respect to time:
L L3 
∂Vwater  L =
2

∆q A = T T 
∂t  (40.8)
The water volume loss is also the void volume loss in the clay. If both
sides of (40.8) are divided by the area A, then the right side is the change
in void volume/unit area with respect to time, or in other words, the
height change of the slice with respect to time. The height change, in
turn, is the product of the vertical strain (ε z ) and the original slice height
h 0 = dz, so
∂ (Vwater A) ∂ (∆h ) ∂ (∆h h0 ) dz (∂ε z ) dz
∆q = = = =
∂t ∂t ∂t ∂t
(40.9)
Let's substitute this into the left side of equation (40.7):
(∂ε z )dz = −k ∂ 2ue dz
∂t ρg ∂z (40.10a)
∂ε z −k ∂ 2 ue
=
∂t ρg ∂z 2 (40.10b)
This looks even more like the heat equation, but note that ε z ≠ ue .
Equation (40.10b) expresses how the vertical strain changes with time
relative to the second partial derivative of the excess pore pressure with
respect to position. We seek to find how the vertical strain changes as a
function of the effective stress. The coefficient of compressibility (m v ) ,
also known as the coefficient of volume change, is defined as the change
in volumetric strain divided by the change in effective stress. For our 1-D
case:
∆V /V0 ∆h / h0
mv = =
∆σ ' ∆σ ' (40.11)
In our case here, the change in effective stress is exactly opposite to the
change in the excess pore pressure (i.e., the increase in load picked up by
the soil skeleton equals the decrease in the excess water pressure). So:

GG 454 April 24, 2002 4
∆h / h0 −∆h / h0 −εz
mv = = =
∆σ ' ∆ue ue − u0 (40.12)
Solving for the strain, which appears on the left side of (40.10b), gives
εz = −mv ( ue − u0 ) (40.13)
Inserting (40.13) into (40.10b) yields
∂ ( mv ( ue − u0 )) k ∂ 2 ue
=
∂t ρg ∂z 2 (40.14)
Now differentiate the left side of (40.14), noting that u 0 is a constant:
∂ ( ue − u0 ) ∂ue ∂u0  ∂ue k ∂ 2 ue
mv = mv − = mv =
∂t  ∂t ∂t  ∂t ρg ∂z 2 (40.15)
This can be simplified by grouping all the constants:
∂ue 2 ∂ 2 ue
k ∂ ue
= = Cv
∂t m v ρg ∂z ∂z (40.16)
The term Cv is called the coefficient of consolidation. This equation is
exactly analogous to the heat equation:
∂T ∂ 2T
=κ
∂t ∂z (40.17)
So the diffusion of heat in an insulated bar is analogous to the diffusion of
excess pore pressure in a soil. A key difference between the two
phenomena is that the hydraulic conductivity of earth materials generally
decreases as the porosity or void ratio decreases; this effect would
increase the consolidation time.
I I I Calculating consolidation for double-sided drainage

At this point we return to some practical questions: How much will a
clay layer will consolidate? What is the ultimate consolidation? How
does the consolidation relate to the pore pressure? We start with our
definition for strain, and focus on the thickness of the clay layer (H).
∆H ∆Vvoids
ε vertical = =
H0 Vtotal (40.18)
The change in void volume can be expressed in terms of the void ratio e:

GG 454 April 24, 2002 5
Vvoids( final ) Vvoids(initial )

−
∆Vvoids Vsolids Vsolids ∆e ∆e
= = =
Vtotal Vtotal Vsolids + Vvoids(initial ) 1 + e 0
Vsolids Vsolids (40.19)
So putting (40.18) and (40.19) together:
∆e
∆H = H 0
1 + e0 (40.20)
This provides a way to limit the ultimate consolidation. We know that ∆ e
cannot exceed the initial void ratio, so
e0
∆H max = H 0
1 + e0 (40.21)
If we know the ultimate change in effective stress, then the coefficient
of volume change can be used to get the thickness change. In our 1-D
situation, volume changes occur as a result of height changes, so
∆H / H 0
mv =
∆σ ' (40.22)
Solving for Ho in the above expression yields
H 0 m v ∆σ ' = ∆H (40.23)
This not only can be used to give the ultimate consolidation, but because
we can calculate the change in effective stress in a layer from the change
in excess pore pressure, we can calculate how the consolidation varies as
a function of time. The excess pore pressure will vary through a layer, so
the average excess pore pressure (with the overbar) in the layer is what is
used:
∆H ( t) H m ∆σ ' ∆σ ' uexcess,t=0 − uexcess( t) ∆uexcess( t)
= 0 v = = = =U
∆H ( t = ∞) H0 mv ∆σ '∞ ∆σ '∞ uexcess,t=0 uexcess,t=0 (40.24)
This ratio is known as the consolidation ratio U. One note of caution in

the use of equations (40.23) and (40.24): they yield unrealistically large
values of settlement for large values of ∆ σ ’. For example, if (∆ σ ’)( mv ) >1,
then ∆H > H0. The layer obviously cannot achieve a negative thickness, so
m v should only be considered a constant for a finite range of ∆ σ ’.

GG 454 April 24, 2002 6
At a given instant in time, this consolidation ratio can be visualized in the

following plot, where the horizontal axis is position across a layer, and
the vertical axis is the excess pore pressure.
1.00
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
1 2 3 4 5 6 7 8 9 10 11 12 13
These curves show how the excess pore pressure decays with time. The
ratio of the area above a curve to the area of the entire box is the
consolidation ratio U at a given time. The lower the curve, the longer the
time for the excess pore pressure to diffuse.

GG454 May 1, 2002 1
SUBSIDENCE MECHANICS REVIEW (41)
I Main Topic: Review of key points in consolidation theory and examples

(From Lambe, T.W., and Whitman, R.V.,1969, Soil mechanics, Wiley, New York, 553p.)
A Main Point 1: three parameters control consolidation time

"The time [for consolidation] should be directly proportional to the
volume of water which must be squeezed out of the soil. This
volume of water must in turn be related to the product of the
[effective] stress change, the compressibility [coefficient of volume
change] of the mineral skeleton, and the volume of the soil."
t ∝ ∆σ ©mv H (this is the time for a given height change) (41.1)

t= time
∆σ ' = change in effective stress
m v = coefficient of volume change = [∆ V/V 0 ]/ ∆σ ’
H = height of soil column
"The time [for consolidation] should be inversely proportional to how

fast the water can flow through the soil. From fluid mechanics we
know that the velocity of flow is related to the product of the
permeability and the hydraulic gradient, and that the gradient is
proportional to the [excess] fluid pressure lost within the soil
divided by the distance through which the pore fluid must flow."
The excess fluid pressure loss ∝ ∆σ ’, and the relevant distance is H, so
1
t∝
k ( ∆σ ' / H ) (41.2)
k= hydraulic conductivity of the porous, permeable solid
From (41.1) and (41.2)

∆σ ©mv H
t∝
k (∆σ ©/H ) (41.3)
hence
m H2
t∝ v (41.4)
k
From (41.4) the consolidation time t:
1 Increases with increasing compressibility (i.e., with increasing m v )
2 Increases rapidly with increasing volume (height) of soil mass
2 Decreases with increasing permeability
4 Is independent of the magnitude of the effective stress change(!)

GG454 May 1, 2002 2
B Main Point 2: dimensional analysis illuminates consolidation time

The one-dimensional consolidation equation for saturated clayey
layers is:
∂ue ∂ 2 ue pressure pressure length 2
= cv = cv ⇒ cv =
∂t ∂x 2
time length 2 time
(41.5)
where
k
c v = coefficient of consolidation =
ρ water gmv
m v = coefficient of volume change
The dimensionless time used in the dimensionless form of (41.5) is found

in the same way as in lecture 39. The underlined terms below are
dimensioned variables; they are multiplied by dimensioned “scaling
constants” to yield dimensionless (starred) variables:
L e t x* = x/xm a x , t* = cv t/xm a x 2 , and ue * = ue / ue max
or x = x* xmax , t = t* xm a x 2 / cv , and ue = ue * ue max.
Now H = xmax , so the second relationship in the line above becomes

ρ gm
t = t * H 2 / cv = t * H 2 water v
k (41.6)
Compare (41.6) with (41.4) - they have the same form.
C Main Point 3: the key length scale is the flow path length
1 For double drainage, H = half the layer thickness

2 For single drainage, H = the whole layer thickness
consolidation at time t
3 Average consolidation ratio U =
ultimate consolidation
a U(t*=1) ≈ 92% (using the dimensionless time defined above)
b U(t*=3) ≈ 99%
c About 92% of the ultimate primary consolidation occurs in the
time given by xmax 2 / cv

GG454 May 1, 2002 3
D Examples
Consider layers of clay and sand, each 10’ thick, and suppose that
the coefficient of volume change (“compressibility”) of the sand is
1/5 that of the clay, and the permeability of the sand is 10,000
times that of the clay.
What is the ratio of the consolidation times of the sand and clay?
Using (41.4)
tclay Hclay 2 mv (clay ) kclay mv (clay ) kclay 51
= = = = 50, 000
tsand H sand 2 mv ( sand ) ksand mv ( sand ) ksand 1 10, 000
If a 10’-thick layer of clay reaches 90% consolidation in 10 years,

how long would it take for a clay layer 40’ thick to reach that level
of consolidation?
Using (41.4)
2 2 402
t40© mv 40©
H 40© k40© H 40©
= = = = 16 16 x 10 years = 160 years
t10© mv H10© 2 k10© H10©2 102
10©
Quadrupling the thickness increases the time for consolidation by a

factor of 16.
How long will a layer of clay take to reach 90% consolidation if the
initial excess pore pressure distribution is constant across the
layer and the layer is drained from its top and bottom? One
dimensionless time unit (t*) is given by t * = c v t/xmax2 = 1.
We need to define our space-time grid to use the numerical

procedure of lecture 39.
1 We start by dividing our layer up into slices of equal thickness.
Let’s pick 10 slices. These 10 slices are defined by 11 points.
2 We need to pick a length scale (xmax ). Let’s pick a length scale
that corresponds to the longest distance fluid has to flow. For
our example here, that is half the thickness of the layer. So xm a x
is 5 increments.
3 How much time does 1 dimensionless time interval correspond
to?
t = t * xmax2 /c v = 52 /c v = 25 /c v . These 25 time intervals are defined
by 26 points.

GG454 May 1, 2002 4
Clay Layer Position-time grid
10
8
5
Position
0
0 5 10 15 20 25
Time

GG454 May 1, 2002 5
% Matlab script transient_heat3.m

% Solves for 1-D transient heat flow by finite differences
% H = length of longest flow path
a = 11; % Number of rows in the T matrix (# of points in space)
b = 26; % Number of columns in the T matrix (# of points in time)
numit = 21; % Number of iterations at each time step
T0 = 0; % Temperature (excess pore pressure) at one end of rod
T1 = 1; % Temperature (excess pore pressure) at other end of rod
% Initialize the "Temperature matrix"

T=zeros(a,b); % First index is position, second is time step
% Set the intial temperature (or excess pore pressure distribution)

T(:,1) = (linspace(T1,T1,a))' % Constant initial excess pore pressure
% Solve by finite difference method

for j=2:b; % j is the index number for the time step
for k = 1:numit;
for i = 2:a-1; % i is the index number for the position
T(i,j) = 0.25*( T(i+1,j-1)+T(i-1,j-1)+T(i+1,j)+T(i-1,j) );
end
end
end
% Plot the figures

figure(1)
clf
subplot(2,1,1)
plot(T); % This plots the columns of T versus their index
title('Excess pore pressure distribution at various times')
xlabel('Position')
ylabel('Excess pore pressure/u0')
axis([1 a 0 1.05])
subplot(2,1,2)
plot([0:b-1]/(((a-1)/2)^2),mean(T),'-.b')
%plot([0:b-1]/(a-1)^2,mean(T),'-.b')
hold on
plot([0:b-1]/(((a-1)/2)^2),1-mean(T),'r')
%plot([0:b-1]/(a-1)^2,1-mean(T),'r')
title('Mean excess pore pressure (blue dash) and U (red solid) at various times')
xlabel('Dimensionless time')
ylabel('Mean excess pore pressure/u0 (or consolidation ratio)')

GG454 May 1, 2002 6
T=
Columns 1 through 7
1.0000 0 0 0 0 0 0
1.0000 0.7320 0.4226 0.3468 0.2930 0.2583 0.2307
1.0000 0.9282 0.7621 0.6249 0.5471 0.4860 0.4369
1.0000 0.9807 0.9132 0.8170 0.7317 0.6609 0.5978
1.0000 0.9945 0.9681 0.9130 0.8417 0.7688 0.6995
1.0000 0.9972 0.9813 0.9406 0.8773 0.8052 0.7341
1.0000 0.9945 0.9681 0.9130 0.8417 0.7688 0.6995
1.0000 0.9807 0.9132 0.8170 0.7317 0.6609 0.5978
1.0000 0.9282 0.7621 0.6249 0.5471 0.4860 0.4369
1.0000 0.7320 0.4226 0.3468 0.2930 0.2583 0.2307
1.0000 0 0 0 0 0 0
Columns 8 through 14
0 0 0 0 0 0 0
0.2078 0.1879 0.1701 0.1541 0.1397 0.1266 0.1148
0.3945 0.3570 0.3234 0.2931 0.2657 0.2409 0.2184
0.5415 0.4908 0.4449 0.4033 0.3657 0.3315 0.3006
0.6352 0.5764 0.5227 0.4740 0.4298 0.3897 0.3534
0.6674 0.6058 0.5495 0.4984 0.4519 0.4098 0.3715
0.6352 0.5764 0.5227 0.4740 0.4298 0.3897 0.3534
0.5415 0.4908 0.4449 0.4033 0.3657 0.3315 0.3006
0.3945 0.3570 0.3234 0.2931 0.2657 0.2409 0.2184
0.2078 0.1879 0.1701 0.1541 0.1397 0.1266 0.1148
0 0 0 0 0 0 0
0 0 0 0 0 0 0
0.1041 0.0944 0.0856 0.0776 0.0704 0.0638 0.0578
0.1980 0.1795 0.1628 0.1476 0.1338 0.1213 0.1100
0.2725 0.2471 0.2240 0.2031 0.1842 0.1670 0.1514
0.3204 0.2905 0.2634 0.2388 0.2165 0.1963 0.1780
0.3369 0.3054 0.2769 0.2511 0.2277 0.2064 0.1872
0.3204 0.2905 0.2634 0.2388 0.2165 0.1963 0.1780
0.2725 0.2471 0.2240 0.2031 0.1842 0.1670 0.1514
0.1980 0.1795 0.1628 0.1476 0.1338 0.1213 0.1100
0.1041 0.0944 0.0856 0.0776 0.0704 0.0638 0.0578
0 0 0 0 0 0 0
0 0 0 0 0
0.0524 0.0475 0.0431 0.0391 0.0354
0.0997 0.0904 0.0820 0.0743 0.0674
0.1373 0.1245 0.1129 0.1023 0.0928
0.1614 0.1463 0.1327 0.1203 0.1091
0.1697 0.1539 0.1395 0.1265 0.1147
0.1614 0.1463 0.1327 0.1203 0.1091
0.1373 0.1245 0.1129 0.1023 0.0928
0.0997 0.0904 0.0820 0.0743 0.0674
0.0524 0.0475 0.0431 0.0391 0.0354
0 0 0 0 0

GG454 May 1, 2002 7
»mean(T)
ans =
Columns 1 through 7
1.0000 0.7516 0.6467 0.5767 0.5186 0.4685 0.4240
0.3841 0.3482 0.3156 0.2861 0.2594 0.2352 0.2133
0.1934 0.1753 0.1590 0.1441 0.1307 0.1185 0.1074
0.0974 0.0883 0.0801 0.0726 0.0658
»1-mean(T)
ans =
Columns 1 through 7
0 0.2484 0.3533 0.4233 0.4814 0.5315 0.5760
0.6159 0.6518 0.6844 0.7139 0.7406 0.7648 0.7867
0.8066 0.8247 0.8410 0.8559 0.8693 0.8815 0.8926
0.9026 0.9117 0.9199 0.9274 0.9342

GG454 May 1, 2002 8
Excess pore pressure distribution at various times

1
t* = 0
0.8
Excess pore pressure/u0
0.6
0.4
0.2
t* = 1
0
1 2 3 4 5 6 7 8 9 10 11
Position
Mean excess pore pressure (blue dash) and U (red solid) at various times
1
Mean excess pore pressure/u0
0.8
Consolidation ratio U
0.6
0.4
0.2
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Dimensionless time
So after one dimensionless time step, about 92% (i.e., roughly 90%) of all
the consolidation that will occur will have occurred.

GG454 May 1, 2002 9
An analogous procedure could be used to solve for the problem where the
drainage is single sided and the initial excess pore pressure distribution
is not constant but varies linearly.
Water Table Level and Subsurface Layers
P Initial water table level
Sand
Lowered water table level

CLAY
LAYER
Sand layer with constant

Sand hydraulic pressure
Saturated clay layer (of thickness H0) with double drainage
Pore Pressure Levels and Excess Pore Pressure

Depth
Initial equilibrium pore pressure level

Final equilibrium pore pressure level
Triangular excess pore pressure distribution
Pore pressure

GG 454 April 29, 2002 1
SUBSIDENCE IN THREE DIMENSIONS: CENTER OF DILATION (42)
I Main Topics
A Center of dilation in full-space
B Center of dilation in a half-space
C Gravitational and elastic analogs
D Effects of a center of contraction
I I Center of dilation in full-space
A What is a center of dilation?
1 A point at which the displacements radiate from
2 An infinitely small spherical hole containing fluid at an infinite
pressure (p dV = 1)
3 A so-called nucleus of strain obtained by superposing, integrating
or differentiating the effect of a force at a point
B Radial displacements vary as 1/r 2
C Radial normal strains vary as 1/r 3
D Radial normal stresses vary as 1/r3
III Center of dilation in half-space
A Construction of the solution by sources and images

GG 454 April 29, 2002 2
Center of Dilation
Suppose a pressure energy souce
(or a spherical volume) is inserted
into a tiny spherical void in an infinte 2
Area of a sphere = 4πR
body, and this pressure (or inserted 3
Volume of a sphere = (4/3)πR
volume) does work on the body by
displacing material radially away from the
source. The material is homogeneous,
isotropic, and continuous.
Energy = work = Force•displacement

= pressure•area•radial displacement
R
W = PAur
2 R1
ur = W/PA = W/(P4πR )
So the radial displacements ur
at the wall of the void go as 1/R2.
Now consider a concentric spherical

surface outside the walls of the void - how
do they displace? Consider the dashed
circle in the diagram. Work is done on the
inside of this surface by the expanding
material inside it. Let the pressure exerted
on the inside of this surface be P(1). Then
W(1) = P(1)A(1)ur(1)
ur(1) = W(1)/P(1)A(1) = W(1)/(P(1)4πR(1)2)
So at any arbitrary distance from the center

of the void, the radial displacements
uroutside the void also go as 1/R2.

GG 454 April 29, 2002 3
Construction of a Center of Dilation

in a Half-Space
Image
center
Source
center
Image
= doublet
(Inifinitesimal
dipole)
Image
double
force

GG 454 April 29, 2002 4
B Displacements for pressure energy E at a point source

Full-space Half-space
ux c
E m
x c  1 3 − 4ν 6 z( z + c ) 
4π R 3 E m x + + 
4 π  R13 R2 3 R25 
uy c
E m
y c  1 3 − 4ν 6 z( z + c ) 
4π R 3 E m y + + 
4 π  R13 R2 3 R25 
c  z − c ( 3 − 4ν )( z + c ) 6 z( z + c )2 2z 
uz c z
E m
4π R 3 E m − − + 
4 π  R13 3 5 3
 R2 R2 R2 
c m = poroelastic strain coefficient, dimension of pressure -1 .
ν = Poisson’s ratio
c = depth of center of dilation
R = x 2 + y 2 + z2 distance from strain nucleus placed at the origin
R1 = x 2 + y 2 + ( z − c )2 distance from a source strain nucleus at z = +c
R2 = x 2 + y 2 + ( z + c )2 distance from an image strain nucleus at z = -c

Note that the half-space solutions yield the full space solutions if c is set
to 0, R1 is replaced by R, and the terms with R2 are eliminated.
C Displacement field at the surface (z = 0) for a center of dilation

Full-space Half-space
ux c
E m
x c  1 3 − 4ν  c  4 (1 − ν ) 
4π R 3 E m x +  = E m x 
4 π  R13 R13  4 π  R13 
uy c
E m
y c  1 3 − 4ν  c  4 (1 − ν ) 
4π R 3 E m y +  = E m y 
4 π  R13 R13  4 π  R13 
c  −c ( 3 − 4ν )(c )  c  4 (1 − ν )(c ) 
uz c z
E m
4π R 3 E m −  = E m − 
4 π  R13 R13  4 π  R 1
3

The half-space displacements at the surface of a half-space are
proportional to the corresponding full-space displacements:
ui half − space = ui full− space [ 4 (1 − ν )] ≈ 3ui full− space (42.1)

GG 454 April 29, 2002 5
IV Gravitational and elastic analogs (from Wang, H., 2000, Theory of linear
poroelasticity: Princeton University Press, Princeton, New Jersey, 287
p.)
Gravity Elasticity Theory
Poisson’s ∇ 2U = −4 πGρ ∂u ∂u y ∂uz
∇ 2Φ = c m p = x + + =ε
equation U = grav. potential ∂x ∂y ∂z
( U = mgz ) Φ = displacement potential
G = universal c m = uniaxial poroelastic
gravitational const. expansion coefficient
−4πρ = “density” p = pressure change
Gradients of Gravitational force Displacement
∂U ∂Φ
potential gi = ui =
∂xi ∂xi
Analogs Acceleration g Displacement u
Potential U Potential Φ
Vertical Gravitational force Displacement per unit
component per unit mass pressure energy amount
c c m(1−ν ) c
(z=0) g *z ( r, 0) = G u *z HS ( r, 0) = −
 r2 + c 2 
3 /2 π  r2 + c 2 
3 /2
   
positive g z = down positive u z = down
Horizontal Gravitational force Displacement per unit
(radial) per unit mass pressure energy amount
c c m(1−ν ) r
component g *r ( r, 0) = −G u *r HS ( r, 0) =
(z=0)  r2 + c 2 
3 /2 π  r2 + c 2 
3 /2
   
positive g r = out positive u r = out
The factor of 4(1-ν ) comes from the difference between the solutions of
displacements in a half-space vs. a full-space
V Effects of a center of dilation or suction

Center of dilation Center of suction
Nearby radial stress Compression increases Compression decreases
Nearby hoop stress Compression decreases Compression increases
Horizontal tensile Decreases for r > c 2 Increases for r > c 2
stress σ xx at z=0 Increases for r < c 2 Decreases for r < c 2

GG 454 April 29, 2002 6
Displacement field associated with a source of suction
0.1
0.2
0.3
0.4
0.5
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
x
* For a center at a depth of 0.15.

* Note that positive z is down.
* The arrowhead that points up just above the center represents the
upward displacement of a point beneath the center. Similarly, the
arrowhead that points down just beneath the center represents the
downward displacement of a point above the center.
* Note that the displacements at the surface are radial, but those
between the surface and the center are not.

GG 454 April 29, 2002 7
Normalized horizontal tensile stress at surface

1
-1
-2
-3
Sxx/Smax
-4
-5
-6
-7
-8
-9
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
x
For a center at a depth of 0.15
r
ur = K
3 /2
 r2 + c 2 
 
∂ur  −5 / 2 −3 / 2 
= K ( r)(−3 / 2) r2 + c 2  (2r) +  r2 + c 2  
∂r  
∂ur
= 0 if 0 = −3r  +  r + c  = −2 r + c
2 2 2 2 2
∂r
∂ur
> 0 if 2 r2 > c 2 or r > c / 2
∂r
∂ur
< 0 if 2 r2 < c 2 or r < c / 2
∂r

GG 454 April 29, 2002 8
Trajectories of most compressive induced stress

due to a source of suction
0.1
0.2
z
0.3
0.4
0.5
0.6
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
x
* For a center at a depth of 0.15

* The induced compressive stresses are “roughly horizontal” within the
hourglass shaped region above and below the center of suction, and they
are “roughly vertical” elsewhere

GG 454 April 29, 2002 9
Stresses acting on the surface z = 0
Szz
1
0.5
Szz
-0.5
-1
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Syz
x
1
0.5
Syz
-0.5
-1
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Szx
x
1
0.5
Szx
-0.5
-1
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
x
* This is a check on the tractions at the surface. They all equal zero
as they should.

GG 454 April 24, 2002 1
REVIEW OF SUBSIDENCE - ASSESSMENT AND RISK MITIGATION (43)
I Main Topics
A Recognition of subsidence hazards
C Evaluation of subsidence hazards (Subsidence mechanics)
D Mitigation of subsidence hazards
I I Recognition of subsidence hazards
http://water.usgs.gov/ogw/pubs/fs00087/
http://water.wr.usgs.gov/subsidednce/ls_3.html
III Characterization
A Monitoring of subsidence (surveying, GPS, InSAR)
B Characterization of subsurface units, voids, and strain
C Tracking fluid withdrawal
D Mapping of surficial fissuring and faulting
IV Evaluation of subsidence hazards (Subsidence mechanics)
1 Consolidation theory
2 Empirical experience
A Typical maximum subsidence/Head loss ≈ 0.01-0.02
B "Extreme" maximum subsidence/Head loss ≈ 0.1-0.02
(From Costa and Baker, p. 289, Santa Clara Valley)
V Mitigation (National Academy of Sciences)
A Education
1 "Recognize and avoid the problem"
2 Adopted by California and Texas for fluid withdrawal
3 Adopted by California and Louisiana for drainage of organic soil
B Geologic mapping
1 "Identify the areas to avoid" (recognition and characterization)
2 Most popular mitigation method today for all forms of subsidence
C Regulation of resource development
1 "Reduce the risk"
2 Adopted by California and Texas for fluid withdrawal
3 Adopted by Florida for drainage of organic soil
4 Adopted by Pennsylvania for mining

GG 454 April 24, 2002 2
D Land-use management (and construction codes)

1 "Reduce and avoid the risk"
2 Adopted by California for fluid withdrawal, hydrocompaction, and
drainage of organic soils
3 Adopted by Florida for drainage of organic soil
E Penalties ("Market-based methods")
1 "Accept the risk and pay for the damage caused"
2 Adopted by California for fluid withdrawal
3 Adopted by Pennsylvania for mining
F Insurance
1 "Share the risk/pay for damage that might be incurred"
2 Mostly used in Midwest U.S. for mining-induced subsidence

GG 454 May 1, 2002 1
NUCLEAR WASTE DISPOSAL (44)
I Main topics
A Types of nuclear waste
B Disposal options
C Experience of different countries in nuclear waste disposal
D U.S. approach to nuclear waste disposal
I I Types of nuclear waste
http://www.nrc.gov/waste.html
A High-level (HLW): waste from spent or reprocessed fuel
1 Require heavy shielding
2 10.2 x 106 ft3 + 0.085 x 106 ft3 (Since 1980)
3 10 0 -10 2 m3 per 1000 MW reactor/year (Milnes, 1985)
4 In form of liquids, solids, sludge, and "cakes" (powder)
5 Main concern: 239 Pu; τ ≈ 24,000 years
B Transuranic (TRU): man-made radioactive elements (e.g., plutonium)
1 26 x 106 ft3 + 0.025 x 106 ft3 (Since 1980)
2 α -particle emitters (particularly injurious to cell tissue)
3 Main concern: 239 Pu; τ ≈ 24,000 years
C Low-level (LLW) not high-level, not uranium mill tailings;
less than 10 nanocuries of TRU (e.g., contaminated clothes)
1 Do not require heavy shielding
2 10 3 -10 5 m3 per 1000 MW reactor/year (Milnes, 1985)
3 Main concern: 90 Sr and 137 Cs; τ ≈ 30 years
4 Mostly β -particle emitters
5 Main disposal options
a Shallow burial
b Dumping at sea
c Liquid injection
d Grout injection (Oak Ridge)

GG 454 May 1, 2002 2
III High-level waste disposal options

Pros Cons
Geologic Disposal • Detailed Political factors:
characterization • Unstable funding
• Many rock types • Bureaucratic control
theoretically OK • Expediency can
outweigh science
• "NIMBYism"
Crystalline basement Rocks are strong Fracture flow
rocks (e.g., granites Permeability is low Nearby mineral wealth
and gneiss)
Volcanic rocks (e.g., Rocks strength Fracture flow
basalt and tuff*) Permeability is low High variability of
Chemical sorption basalt
Shale or clay Low permeability Rock strength
S a l t Low permeability Can dissolve
Fractures heal
Deep borehole Relatively inexpensive Liquefied waste
Injection On-site disposal Fractures
Deep Underground Relatively inexpensive Before solidifying,
melting On-site disposal melt is highly mobile
Fractures
Icebed disposal Remoteness Climate change
Antarctic Treaty of
1959
Seabed disposal Low permeability rock Access to biosphere
Dilution Biologic concentration
Remoteness of radioactivity
Law of the Sea
Extra-terrestrial Permanent disposal Rocket failure
disposal Loss of an energy
resource

GG 454 May 1, 2002 3
IV Experience of different countries in nuclear waste disposal

A Need to be self-dependent requires use of a variety of rocks
B Sweden
1 Decouple politics from site selection
C Switzerland
1 Great openness after initial concealment of program
2 Variety of sites being examined
D Canada
1 Crystalline rock sites being examined
2 Has Underground Rock Laboratory
V U.S. approach to nuclear waste disposal
A Definition of acceptability
1 No more than 1,000 cancer deaths in the next 10,000 years;
emphasis on travel time to biosphere
2 Legal criterion, not a "scientific" or "engineering" criterion
3 10,000 year standard
a How the human race will evolve in 10,000 years?
b Geologic time frame for an "engineering" problem
c Pushes limits of our predictive ability
B Detailed standards set; Earth treated like an engineering material
C Quality assurance
1 Provides detailed “paper trail” of work
2 Tends to put technical control in hands of bureaucrats
D Yucca Mountain; DOE site
1 Selection process
2 Key geotechnical issues:
a Tectonic activity (seismicity and volcanism)
b Fracture hydrogeology and hydrogeochemistry of tuff

GG 454 May 1, 2002 4
E Alternative sites
1 Hanford, Washington (basalt); DOE site
Key geotechnical issues: fractures in basalt, catastrophic floods,
high variability in sedimentary sequences
2 Deaf Smith County, Texas (salt)
Key geotechnical issues: Ogallala aquifer
3 WIPP (Carlsbad, New Mexico)
Key geotechnical issues: gas pressure, salt dissolution, fractured
dolomite above repository
References
Milnes, A.G., 1985, Geology and Radwaste: Academic Press, New York, 328
p.
National Research Council, 1990, Rethinking high-level radioactive waste
disposal: National Academy Press, Washington, D.C., 38 p.
Deese, D.A., 1978, Nuclear power and radioactive waste: D.C. Heath and
Company, Lexington, Massachusetts, 206 p.
Murray, R.L., 1989, Understanding radioactive waste: Battelle Press,
Columbus, Ohio, 167 p.
National Research Council, 1976, The shallow land burial of low-level
radioactively contaminated solid waste: National Academy of Sciences,
150 p.
Shapiro, F.C., 1981, Radwaste: Random House, New York, 288 p.
National Research Council, 1962, Disposal of low-level radioactive waste
into Pacific coastal water: National Academy Press, Washington, D.C.,
87 p.

GG 454 May 1, 2002 5
Selected Country Programs for High-Level Waste Burial

http://www.platts.com/features/nukewastedisposal/index.shtml
Country Earliest Planned Status of Program
Year
Argentina 2040 Granite site selected at Gastre, Chubut
Canada 2020 Independent commission conducting four-year study of
government plan to bury irradiated fuel in granite at yet-
to-be-identified site
China none Irradiated fuel to be reprocessed; Gobi desert sites under
investigation
Finland Construction to The Finnish Parliament May 18 2001 decided on a
begin in 2003- permanent disposal site for HLW in Olkiluoto, Eurajoki.
2004. Operation of the facility will start in 2020.
France 2010 Three sites to be selected and studied; final site not to be
selected until 2006
Germany 2008 Gorleben salt dome sole site to be studied
India 2010 Irradiated fuel to be reprocessed, waste stored for 20
years, then buried in yet-to-be-identified granite
site
Italy 2040 Irradiated fuel to be reprocessed, and waste stored for
50-60 years before burial in clay or granite
Japan 2020 Limited site studies. Cooperative program with
China to build underground research facility
Netherlands 2040 Interim storage of reprocessing waste for 50-100 years
before eventual burial, possibly in another country
Sweden 2020 Granite site selected in 1997; evaluation studies under
way at Aspo site near Oskarshamn nuclear complex
United 2010 Yucca Mountain site being studied. Will receive 70,000
States tons of waste if approved
United 2030 Fifty-year storage approved in 1982; exploring options
Kingdom for permanent disposal

GG 454 May 1, 2002 6

GG 454 May 1, 2002 7

Geo 454 2/13/02 1
An Incomplete and Evolving List of Supplemental References
Perhaps the best overall introductory book

Costa, J.E., and Baker, V.R., 1981, Surficial geology: Wiley, New York, 498 p.
General Earth Science
Press, F., and Siever, R., 1978, Earth: Freeman, San Francisco, 649 p.
Structural Geology
Suppe, J., 1985, Principles of structural geology: Prentice-Hall, Englewood Cliffs, New
Jersey, 537 p.
Mechanics
Means, W.D., 1976, Stress and strain: Springer-Verlag, New York, 339 p.
Johnson, A.M., 1970, Physical Processes in Geology: Freeman, San Francisco, 577 p.
Timoshenko, S.P., and Goodier, J.N., 1970, Theory of elasticity, McGraw-Hill, New York, 567
p.
Middleton, G.V., and Wilcock, P.R., 1994, Mechanics in the Earth and Environmental Sciences,
Cambridge University Press, 459 p.
Environmental Geology, Hazards, and Risk
Lundgren, L., 1986, Environmental geology: Prentice-Hall, Englewood Cliffs, New Jersey, 576
p.
Engineering properties of geologic materials
Costa, J.E., and Baker, V.R., 1981, Surficial geology: Wiley, New York, 498 p.
Turcotte, D.L., and Schubert, G., 1982, Geodynamics: Wiley, New York, 450 p.
Carmichael, R.S., 1989, Practical handbook of physical properties of rocks and minerals: CRC
Press, Boston, 741 p.
Kehew, A.E., 1988, General geology for engineer: Prentice-Hall, Englewood Cliffs, New Jersey,
447 p.
Earthquakes and seismic hazards
Bolt, B. A. , et al., 1975, Geological hazards : earthquakes, tsunamis, volcanoes, avalanches,
landslides, floods: Springer-Verlag, 328 p.
Bolt, B. A., 1976, Nuclear explosions and earthquakes: the parted veil: W. H. Freeman, San
Francisco, 309 p.
Bolt, B. A., 1978, Earthquakes: a primer: W. H. Freeman, San Francisco, 241 p.
Bolt, B. A., 1982, Inside the earth: evidence from earthquakes: Freeman, San Francisco, 191 p.
Bolt, B. A., 1988, Earthquakes: Freeman, New York, 282 p.
Kanamori, H., 1983, Earthquakes: observation , theory, and interpretation: Elsevier,
Amsterdam, 608 p.
Lawson, A.C., 1908, Report of the state earthquake investigation commission: Carnegie
Institution, Washington, D.C.
National Research Council, 1986, Active tectonics: National Academy Press, Washington, D.C. ,
266 p.
Richter, C.F., 1958, Elementary seismology: Freeman, San Francisco, 768 p.
Scholz, C., H., 1990, The mechanics of earthquakes and faulting: Cambridge University Press,
Cambridge, 439 p.
Ziony, J., 1985, Earthquake hazards of the Los Angeles region: U.S. Geological Survey
Professional Paper 1360, 505 p.
Wiegel, R.L., 1970, Earthquake engineering: Prentice-Hall, Englewood Cliffs, New Jersey, p.
253-306.
Geo 454 2/13/02 2
Dams
Wahlstrom, E.E., 1974, Dams, dam foundations, and reservoir sites: Elsevier, Amsterdam, 278
p.
St. Francis Dam, California, 1928
Outland, C.F., 1977, Man-made disaster -- the story of St. Francis dam: Arthur Clark,
Glendale, CA, 275 p.,
Koyna Dam, India
Gupta, H.H., and Rastogi, B.K., 1976, Dams and earthquakes: Elsevier, Amsterdam, 229 p.
Teton Dam, Idaho
Independent panel to review cause of Teton Dam failure, 1976, Failure of Teton Dam: Report to
the U.S. Department of Interior and State of Idaho.
U.S. Department of the Interior, 1980, Failure of Teton Dam, final report: U.S. Goverment
Printing Office, Washington, D.C.
Seed, H.B., and Duncan, J.M., 1981, The Teton Dam - a retrospective review: Proceedings of the
Tenth international Conference on Soil Mechanics and Foundation Engineering, Stockholm,
Sweden, June 1981, p. 1-20.
Vaiont Reservoir, Italy
Kiersch, G.A., 1964, Vaiont reservoir disaster: Civil Engineering, v. 34, p. 32-39.
Hendron, A. J., and Patton, F.D., 1985, The Vaiont Slide : a geotechnical analysis based on new
geologic observations of the failure surface: National Technical Information Service,
Springfield, Virginia, U.S. Army Engineer Waterways Experiment Station Technical Report
GL-85-5.
Malpasset Dam, France
Baldwin Hills, California, 1963
Jessup, W.E., 1964, Baldwin Hills reservoir: Civil Engineering, v. 34, p. 62-64.
James, L.B., 1968, Failure of the Baldwin Hills reservoir, Los Angeles, California: in
Engineering Geology Case Histories, no.6, Geological Society of America, p. 1-11.
Castle, R.O., Yerkes, R.F., and Youd, T.L., 1973, Ground rupture in the Baldwin Hills -- an
alternative explanation: Association of Engineering Geologists Bulletin, v. 10, p. 21-46.
Vega de Tera Dam, Spain, 1959
Johnstown flood, Pennsylvania, 1889
Pagan, A.R., 1974, The Johnstown flood revisited: Civil Engineering, v. 44, p. 60-62.
Video tape 6357 in Hamilton Library
Landslides
Schuster, R.L., and Krizek, R.J., 1978, Landslides: analysis and control: National Academy of
Sciences, Washington, D.C., National Transportation Research Board Special Report 176,
234 p.
Voight, B., 1978, Rockslides and Avalanches: Elsevier, Amsterdam, 2 volumes.
Reviews in Engineering Geology, volume III, 1977, Coates, D.R., ed., Geological Society of
America, 278 p.
Landslide Dams: Processes, Risk, and Mitigation, 1986, Schuster, R.J., ed., Geotechnical
Special Publication No. 3, American Society of Civil Engineers, 164 p.
Fleming, R.W., and Johnson, A.M., 1990, Structures associated with strike-slip faults that
bound landslide elements: Jahns Memorial volume, Journal of Engineering Geology, v. 27, p.
39-114.
Vonder Linden, K., 1990, The Portuguese Bend Landslide: Jahns Memorial volume, Journal of
Engineering Geology, v. 27, p. 301-374.
Committee on Natural Disasters, 1984, The Utah Landslides, debris flows, and floods of May and
June 1983: National Academy Press, 96 p.
Crandell, D.R., Miller, C.D., Glicken, H.X., Christiansen, R.L., and Newhall, C.G., 1984,
Catastrophic debris avalanche from ancestral Mount Shasta volcano, California: Geology, v.
12, p. 143-146.
Geo 454 2/13/02 3
Moore, J.G., Clague, D.A., Holcomb, R.T., Lipman, P.W., Normark, W.R., and Torresan, M.E.,
1989, Prodigious submarine landslides on the Hawaiian Ridge: Journal of Geophysical
Research, v. 94, p. 17,465-17,484.
Waves & Coastal Processes
Wiegel, R.L., 1970, Tsunamis: in Earthquake engineering, Wiegel, R.L., ed., Prentice-Hall,
Englewood Cliffs, New Jersey, p. 253-306.
Sorensen, R.M., 1978, Basic coastal engineering: Wiley, New York, 227 p.
Bascom, W., 1980, Waves and beaches: Doubleday, New York, 366 p.
Tricker, R.A.R., 1964, Bores, breakers, waves and wakes: Mills and Boon, London, 250 p.
Bigelow, H.B., and Edmondson, W.T., 1947, Wind waves at sea, breakers and surf: U.S. Navy
Hydrographic Office, Washington, D.C.,, 177 p.
Stoker, J.J., 1957, Water waves: Interscience Publishers, New York, 567 p.
Hydrogeology
C.W. Fetter, Jr, 1980, Applied Hydrogeology: Merrill, Columbus, Ohio, 488 p..
Freeze, R.A., and Cherry, J.A., 1979, Groundwater: Prentice-Hall, Englewood Cliffs, New
Jersey, 604 p.
Land subsidence
Varnes, D.J., and Kiersch, G.A., 1968: Reviews in Engineering geology, v. 2, Geological Society
of America, 350 p.
Holzer, T.L., 1984, man-induced subsidence: Reviews in Engineering geology, v. 6, Geological
Society of America, 221 p.
Davies, P.B., 1986, Deep-seated dissolution and subsidence in bedded salt deposits: Ph.D. thesis,
Stanford University, Stanford, California, 379 p.
Rock Mechanics
Jaeger, J.C., and Cook, N.G.W., 1979, Fundamentals of rock mechanics: Chapman and Hall,
London, p. 593.
Soil Mechanics
Lambe, T.W., and Whitman, R.V., 1969, Soil mechanics: Wiley, New York, 553 p.
Sowers, G.F., 1979, Introductory soil mechanics and foundations: geotechnical engineering:
Macmillan, New York, 621 p.
Rock Fracture
Pollard, D.D., and Segall, P., 1987, Theoretical displacements and stresses near fractures in
rock: with applications to faults, joints, veins, dikes, and solution surfaces, in Atkinson,
B.K., ed., Fracture mechanics of rock: London, Academic Press, p. 277-349.
Lawn, B.R., and Wilshaw, T.R., 1975, Fracture of brittle solids: Cambridge University Press,
Cambridge, 204 p.
Degraff, J., and Aydin, A., 1987, Surface morphology of columnar joints and its significance to
mechanics and direction of joint growth: Geological Society of America, v. 99, p. 605-617.
Martel, S.J., Pollard, D.D., and Segall, P., 1988, Development of simple fault zones in granitic
rock, Mount Abbot quadrangle, Sierra Nevada, California: Geological Society of America
Bulletin, v. 100, p. 1451-1465.
Olson, J., and Pollard, D.D., 1989, Inferring paleostress from natural fracture patterns: a new
method: Geology, v. 17, p. 345-348.
Segall, P., and Pollard, 1980, Mechanics of discontinuous faults: Journal of Geophysical
Research, v. 85, p. 4337-4350.
Segall, P., and Pollard, D.D., 1983, Joint formation in granitic rock of the Sierra Nevada:
Geological Society of America Bulletin, v. 94, p. 563-575.
Fluid flow in fractured rock
National Academy of Sciences, 1996, Rock fractures and fluid flow: contemporary understanding
and applications: National Research Council, 551 p.
Long, J.C.S., Remer, J.S., Wilson, C.R., and Witherspoon, P.A., 1982, Porous media equivalents
for networks of discontinuous fractures: Water Resources Journal, v. 18, p. 645-6589.
Geo 454 2/13/02 4
Long, J.C.S., 1985, The relationship of the degree of interconnection to permeability in

fracture networks: Journal of Geophysical Research, v. 90, p. 3087-3097
Long, J.C.S., Majer, E.L., Martel, S.J., Karasaki, K., Peterson, J.E., Jr., Davey, A., and Hestir,
K., 1990, Hydrologic characterization of fractured rocks - an interdisciplinary method:
Lawrence Berkeley Laboratory report LBL-27863, 183 p.
Nuclear waste disposal
Milnes, A. G., 1985, Geology and radwaste: Academic Press, London, 328 p.
Glacial and Quaternary Geology/Geomorphology
Flint, R.F. , 1971, Glacial and Quaternary Geology: Wiley, New York, 892 p.
Sharp, R.P., 1960, Glaciers: University of Oregon Press, Eugene, Oregon, 78 p.
Paterson, W.S.B., 1981, The Physics of Glaciers: Pergamon, Oxford, 380 p.
Ritter, D.F., 1978, Process Geomorphology: William Brown, Dubuque, Iowa, 603 p.
Kamb, B., 1964, Glacier geophysics: Science, v. 146, p. 353-365.
GG454 January 25, 2002 1
One Possible General Format For Engineering Geology Reports
Table of Contents
Summary (abstract; abstract by Landes)
Introduction
Statement of problem as presented to you
Description of information you were presented with (maps, cross sections, tables)
Conditions/deadlines that you operated under
(This covers you and gives the reader an appropriate starting point)
Topography
Description
Interpretation
Relevance
Geology
Description
Interpretation
Relevance (if any) to engineering and hydrology
Hydrology
Description
Interpretation
Relevance (if any) to geology and engineering
Engineering
Options
Requirements of different options
Relevance (if any) to geology and hydrology
Costs of different options (if possible)
Pros and cons of different options
Conclusions
Key findings/interpretations from geology and engineering analyses
Key recommendations
Things you were asked to do
Things you weren't asked to do but you think your client should know
(e.g., possible follow-up investigations)
References
Almost invariably some graphics/illustrations will need to be included (e.g., maps and cross
sections, engineering design plans, etc).
Remember to use the four step approach as best you can:

1. Recognition of potentially hazardous/beneficial situations
2. Characterization of potentially hazardous/beneficial situations
3. Evaluation of risk associated with potential hazards and benefits
4. Assessment: Is the level of risk acceptable?
Stephen Martel 1 University of Hawaii

GG 454 1/25/02 1
Project Evaluation Guidelines
Technical Content
Was the four-step approach followed as far as it could/should be?
Are relevant conditions and processes recognized adequately?
Are relevant conditions and processes characterized adequately?
Is risk evaluated (at least qualitatively)?
Is the level of risk tolerable?
Are conclusions firmly supported by evidence?
Are alternatives adequately explored?
Are good recommendations made?
Are the key goals of the project addressed adequately?
Is the report well organized? (e.g., intro, plan of attack, data, interpretation, analysis, concl.)
Is it clear that the geologists and engineers collaborated?
Is the engineering significance of the geology accounted for?
Is the geological effect of the engineering accounted for?
Are the geology and engineering sections coordinated?
Are illustrations used effectively?
Is the report neat? (Sloppy presentation is a tip-off to sloppy thinking)
Writing Style
Is the writing clear?
Is the writing concise?
Is the writing crisp?
Is the grammar correct?
Are the topic sentences of paragraphs good?
Are references handled properly?
Is the abstract or project (summary) done well?

GG 454 February 11, 2002
Homework 1: Characterization of Potential Earthquake Sources

No late assignments will be accepted
5% of total grade; 100 pts total
Before you get started: Draw the plots called for here on good graph paper, not on regular
notebook paper. A green engineering calculation pad is very good for this type of assignment
(and you will see two more like this); I highly recommend getting such a pad. I will consider
neatness in my grading, so be neat.
In describing the stresses here, I will use the on-in convention described in class. Tension
is positive.
Suppose that measurements of stress in the earth indicate on a vertical plane oriented
north-south the horizontal normal traction is -17 MPa, and the shear traction has a magnitude
of +17 MPa. On a vertical plane oriented east-west the normal traction is -17 MPa, and the
shear traction has a magnitude of -17 MPa. Two vertical faults are in the region affected by
this stress field. Fault A strikes 14° (i.e., N14°E), and fault B strikes 346° (i.e., N14°W).
Geologists measure “strikes” relative to true north.
1 Draw a picture that neatly illustrates the quantities above (13 pts total)
a Neatly draw a square box in a map view with sides 3"-4" long, with the sides trending
north-south and east-west, and put a north arrow in the box (1 pt).
b Show normals to the box sides in light lines. Label the x-axis as pointing east and the y-
axis as pointing north (2 pts).
c Label the "far-field" stresses on the sides of the box (4 pts).
d Inside the box draw the traces of the fault planes (i.e., the azimuth of the vertical fault
planes) in fairly heavy lines (2 pts). Make the box big enough so that it is useful and
make sure that the box is in equilibrium.
e Label the normal to fault A as the x'-axis and a line parallel to the fault as the y'-axis;
make the x' and y ' axes be right-handed. (2 pts).
f Label the normal to fault B as the x''-axis and a line parallel as the fault be the x''-axis;
make the y'' and y'' axes be right-handed (2 pts).
2 Fill in the following table as you go (you can’t fill this out entirely at the start, but you can
have it filled out completely by the end): (16 pts total)
Normal stress Shear stress Normal traction Shear traction
(MPa) (MPa) (MPa) (MPa)
North-south plane σ xx σ xy σ xn τ xs
East-west plane σ yy σ yx σ yn τ ys
Fault A σ x ′x ′ σ x ′y ′ σ x ′n τ x ′s
Fault B σ x ′′x ′′ σ x ′′y ′′ σ x ′′n σ x ′′s
Howework 1-1
3 Prepare a Mohr circle diagram using the convention described in class. A positive angle is a
counterclockwise angle in this part of the exercise. (30 pts total)
a Plot the point that depicts the tractions acting on a north-south plane, and label the point
( σ xn , τ xs ) . (2 pts) .
b Plot the point that depicts the tractions acting on an east-west plane, and label the point
( σ yn , τ ys ) . (2 pts) .
c Draw the Mohr circle that defines the state of stress in the region containing the faults
( 4 pts) .
d Plot the point that depicts the normal and shear tractions acting on fault A by using the
negative double angle between the x-axis and the x'-axis; label the point ( σ x ′n , τ x ′s ) and
the negative double angle -2θ xx' (2 pts). Write the stresses in the table (2 pts).
e Plot the point that depicts the normal and shear stresses acting on fault B by using the
negative double angle between the x-axis and the y''-axis; label the point ( σ x ′′n , σ x ′′s ) and
the negative double angle -2θ xx'' (2 pts). Write the stresses in the table (2 pts).
f Determine the magnitude of the most tensile principal stress (σ 1 ) and the least tensile
principal stress (σ 2 ) from the Mohr circle plot. Write your answers below:
σ1 = σx’”n =_____________. (2 pts) σ 2 = σ y’”n =_____________. (2 pts)
g Determine the orientation relative to true north (i.e., find the trend) of the two
principal horizontal stresses (4 pts) by using the negative double angle on the Mohr
circle plot (i.e., find the orientation of σ 1 and σ 2 relative to the y-axis), and draw a
new box with the sides normal to the orientations of these two stresses (2 pts). Draw
faults A and B inside this box (2 pts). Neatly label the stresses acting on the box sides
(e.g., σ x’”n = 55 MPa) (2 pts) .
4 Determination of the sense of slip (9 pts total)

a From your answers above, what is the sense of slip that you would expect for fault A?
By sense I mean right-lateral or left-lateral. (2 p t s )
b What is the sense of slip that you would expect for fault B? (2 pts)
c If the shear strength of the fault is governed by friction (so that the shear strength of
the fault is proportional to the compressive stress acting across the fault) and each fault
has the same coefficient of friction, which fault is most likely slip? Explain below (5
pts).
Howework 1-2
5 Assume that one of the faults does slip. Geologists had collected information before the
earthquake and estimated a possible rupture length of 50 km, a rupture height of 10 km
(this is typical for the San Andreas fault), and an average amount of slip of 2.0 meters, with
the relative displacement being purely horizontal. These turn out to precisely describe the
fault rupture. If the shear modulus of the rock surrounding the fault is 3.0 x104 MPa (and
this is a typical value) make the following calculations (9 pts total):
a Calculate the seismic moment in Newton-meters. (2 pts + 1 pt)
b Using the relations in the class notes, estimate the likely moment magnitude of the quake
( 2 pts + 1 pt) .
c Using the Bikini Atoll atomic blast as a comparison (energy release of 1012 joules),
how many bomb blasts would the earthquake be equivalent to? (2 pts + 1 pt)
6 Suppose that the geologic estimates were off, and that the actual relevant dimensions were:
rupture length = 40 km, rupture height = 8 km, average slip = 1.6 meters (13 pts
total).
a Calculate the seismic moment in Newton-meters. (2 pts + 1 pt)
b Using the relations in the class notes, estimate the likely moment magnitude of the quake
( 2 pts + 1 pt) .
c Using the Bikini Atoll atomic blast as a comparison (energy release of 1012 joules),
how many bomb blasts would the earthquake be equivalent to? (2 pts + 1 pt)
d Comment on how different the answers are for question 8 relative to those of question 7,
and indicate whether you think the differences are likely to be significant in terms of
engineering design considerations (4 pts).
7 Neatness counts for 10 points
Howework 1-3
Geo 454 March 21, 2002 1
Slope Stability Exercise

(Due April 5, 2002)
SHOW ALL YOUR WORK FOR THE FOLLOWING PROBLEMS
1 ) Solve parts 2, 3, and 4 for the attached problem. Use part 1 as a guide. The term γ is the
unit weight, which equals the density time gravitational acceleration. Do you expect a
failure for any of the steps? Comment on the results, specifically the effect of the water,
the effect of cutting the slope back to a safer(?) angle, and the effect of the concrete toe wall
designed to strengthen the toe of the slope. Fill out the table on the last page to make this
easier for me to evaluate.
2 ) Solve for the case of slip surface AB'C'D' on page 184-186 of Rahn using the method of
slices - with one change. Assume that at the base of slice 5 there is a horizontal layer of
quick clay and we've just had an earthquake. The shear strength of the clay has gone to zero.
What is the calculated factor of safety for slip surface AB'C'D'? Is it likely to fail? You
might note that the scale on Figure 6.17 of Rahn is not perfectly consistent - do the best you
can. Create a table like that in Table 6.1 to show your work.
3 ) The above exercises, together with the block-on-a-slope exercise we went through in class,
should provide some mechanical insight into some of the key aspects of slope failures,
namely:
•Why are dip slopes of around 30° particularly susceptible to failure;
•Why rain is an important factor in triggering landslides;
•Why cutting into slopes helps trigger landslides.
Comment on these three points.
Geo 454 March 21, 2002 2
A highway cut is made in horizontal sandstone beds as shown below. The cut face is sloped
1(horizontal) to 6(vertical) for appearance. An open vertical joint that strikes parallel to the
cut face is 9m from the cut toe. A second set of joints also exists; these vertical joints are
spaced 10 m apart and extend all the way through the sandstone and strike perpendicular to the
road. The joints and the bedding thus divide the sandstone into blocks. Consider the 10 m deep
block shown above. Just above the base of the cut is a highly weathered cohesionless seam of
silty sandstone 0.5 m thick. A small concrete wall is poured at the toe of the cut to restrain this
bed. The sandstone throughout has a unit weight of 24 kN/m3 , and the angle of internal friction
of the basal weathered sandstone is 30°.
Case 1: No water in joint
1 The weight (i.e. force) of the upper sandstone block on the weathered bed is (See figure):
Fw = ρgV = γ V
F w = {24 kN/m3 }{15m x 10 m [(9m+6.5m)]/2} = 27,900 kN
2 The average normal vertical stress on the top of the weather bed is:
σ v = weight / area of base
σ v = 27900 kN/m)/(9m x 10m) = 310 kN/m2
3 The frictional strength of the weathered bed is:
τ f = σ v tan 30° = 310kN/m2 tan 30° = 179 kN/m2
There is no cohesion in this case.
4 The weathered sandstone seam can resist a horizontal force of:
( τ f )(basal area) = (179 kN/m2 ) (90 m2 ) = 16,110 kN
The factor of safety is infinite because there is no driving force so far.
Joint 6.5 m
6 15 m
(Bulldozer) 6.5 m
10 m
Concrete Road
9m 15 m
Wall
Weathered sandstone seam
9m
Geo 454 March 21, 2002 3
Case 2: Water rises to 10 m above the top of the weathered bed because the drainage in the
concrete wall gets blocked.
5 Assume the pore pressure µ at the top of the weathered seam is uniform. The pore pressure
µ can be determined by the height of the water column, g, and the density of the water.
µ =_______________kN/m 2
6 The effective stress at the top of the weathered sandstone seam is the total stress (calculated
in part 2) minus the pore pressure: σ v ' = __________
7 The weathered bed now provides a lateral shear resistance of τ f = σ v ' tan 30°:__________
9 The shear force available for resisting sliding depends on the shear strength (τf) and the
area of the base of the block. It is:______________
9 The 10-m-deep water in the joint produces a lateral force. This lateral force is obtained by
integrating the water pressure over the height of the water column and is given by:
F L = (Average pressure)(wetted area of the back of the joint-bounded block)
F L = ({pressure at top of joint + pressure at bottom}/2) x
(the part of the area of the back of the block that feels the water pressure)
F L = ({ρ water g h}/2)(wetted area of the back of the joint-bounded block) = __________
10 The factor of safety against sliding is:________________
Case 3: Repeat steps of case 2, but assume water has risen to 14m above the top of the weathered
bed.
Case 4: Assume that the slope is cut back to the top of the joint and that the water rises to 10 m
above the top of the weathered bed. You will have to recalculate the weight of the block.
Geo 454 March 21, 2002 4
Case 1 Case 2 Case 3 Case 4

Block.length (m)
Block.height (m)
Block.top.width (m)
Block.bottom.width (m)
Block.volume
Block.density (kg/m^3)
Unit.weight (N/m^3)
Block.weight (N)
Basal.area (m^2)
Total.basal.normal.stress (Pa)
Water.height (m)
Basal.water.pressure (Pa)
Effective.basal.normal.stress (Pa)
Basal.cohesion (Pa)
Friction angle (phi), (deg)
Coefficient of friction (mu)
Shear.resisting.stress (Pa)
Basal.area (m^2)
Shear.resisting.force (N)
Water.pressure.top (Pa)
Water.pressure.bottom (Pa)
Average.water.pressure (Pa)
Wetted.area.of.back (m^2)
Driving.force (N)
Factor.of.safety
Developments at Bed Springs
Early one Friday evening, when you are just about ready to leave your consulting firm,
Heraldo Riviera, the recently elected mayor of Bed Springs, knocks on your door. He needs
advice, and your firm has been highly recommended to him by a friend of his, Troy Mann.
Bed Springs needs to develop a new source of water. For years, whale watching at Marble
Point, the beautiful quartz sand beaches adjacent to Bed Springs, and fishing from the old
Prezzure Pier have been the mainstay of the town's economy, but Heraldo expects the economy
to change. "Bed Springs is entering a new high-tech era" claims Heraldo. A new campus for the
College of Reason, Intelligence, Science, and Educational Studies is to be constructed, and
eyesores like the old pier will be torn down. Projections are for the population to increase
four-fold in the coming decade. A 200 acre-foot increase in the annual water supply seems
necessary. Two water options have been suggested to Heraldo: 1) develop a series of
groundwater wells on the south side of town, and 2) build a dam on the Liver River. The mayor
wants you to take a look at the geology and get your opinion of the pros and cons of the proposed
options. He also welcomes any other input you might provide.
Heraldo heard about how well your firm did with a similar sort of projects. "I have much
more information than you were provided with for those other cases! First, I have a map
prepared by the renowned well logger Harold ("Hap") Hazard, and second, I have ten well logs
prepared by the firm of Trump & Maples." The map and the well logs are attached.
Heraldo will be leaving for an important conference hosted by the Western Institute of
Mayors and Politicians at 5:00 on Friday, May 7 and he needs to have your report with him on
the plane so that he can discuss it at the conference. Failure on your part to provide the report
by the due date will result in your firm not getting paid (if this hint is too subtle see me).
Although you emphasize that your firm specializes in highly detailed, extremely thorough
projects and that 4 weeks is not much time for the type of work you are accustomed to doing,
Heraldo convinces you to accept his offer.
Heraldo is about to leave when you notice the facility labeled "Proposed injection wells."
Says Heraldo, "That is where the Unforgiven Nasty Waste Injection Site Experiment was done."
Heraldo has discussed this with another friend of his, Clint Westwood. "Clint says that the site
has a series of deep injection wells that will be used to inject toxic aqueous chemicals deep
underground. It will keep the stuff out of the two aquifers we plan to use if we go for the
groundwater wells. The wells north of the site extend to 500 meters, and the ones on the south
side extend to 300 meters."
In this project, provide your evidence for your conclusions so that Heraldo can defend the
final position he takes as best he can, and indicate who in your firm is responsible for the
different sections of your report.
Possibly useful additional information obtained from drill cores
Limestone and sandstone Clays

ρ saturated = 2400 kg/m3 ρ saturated = 2200 kg/m3
ρ unsaturated = 2168 kg/m3 ρ unsaturated = 1867 kg/m3
e o = initial void ratio = 0.25 e o = initial void ratio = 0.50
k = hydraulic conductivity = 1 x 10-6 m/sec k = hydraulic conductivity = 6 x 10-9 m/sec
mv = coeff. of volume change mv = coeff. of volume change
Unavailable = 3.4 x 10-7 Pa- 1
MAP OF THE BEDSPRINGS ENVIRONS
BY HAROLD HAZARD, 1992
Marble Point Prezzure Pier

Wotztha Point
JOSEKANYOU SEA
o Well 1
BED SPRINGS
x 3m Sandy Beach Proposed injection wells
xxxxxxxxx
Well 2 o x x x x x x x x x x Proposed groundwater wells o Well 10
Sunova Beach
o Well 9
Proposed chemical xxxxxxxxx
The Bluffs waste pipeline
Well 3 o Th Processing
e Facilities
Bl + Flat Toe
uf (Proposed)
Leakyleaky fs
Tunnel Plateau
o Well 8
Liv
er
Ri
Rou
ve
r
te 6
6
N x 39m 10m x x 10m
Flat Toe
20°
Plateau
0 500 m
Proposed site
of Quayle Dam
o Well 4 Well 5 o o Well 6 o Well 7

DRILL LOGS FOR WELLS IN THE BED SPRINGS AREA (COURTESTY OF TRUMP & MAPLES)
1 3 5 7 9 2 4 6 8 10
10 m 40m 40m 110m 110m 10m 40m 110m 110m 100m
Logger's Notes Vertical Scale

1 Logs listed by even and odd well numbers
2 100% recovery of core
3 All units are horizontal where intersected by boreholes 100m
4 Elevation of ground surface is listed below well number
5 Description of units in well #10:
A Clay. Late Pleistocene based on fossil snails at top of unit.
B Sand.
C Clay. Black.
D Limestone. Vuggy.
E Clay. Brown.
F Sylvite. Very pale red.
G Anhydrite, gypsum, and halite
6 All wells intersect the water table 5m below the ground surface.
GG 454 May 3, 2 0 0 2 Name ___________________1
PLEASE BE SENSITIVE TO THE HOMEOWNERS AND DO NOT TRESPASS!!!!

a = 3003 Kalawao St.
What features do you see at this address (which usually has a boat parked in front of it)
that suggests deformation of the ground?
b = 3047 Kalawao St
What features do you see at this address that suggests deformation of the ground?
c = 3017 Lanikaula St
Is the number of water pipes at this address typical?
d = 3049 Lanikaula St
What is unusual about the garage here?
e = 3055 Lanikaula St
What is unusual about the driveway here?
f = 3059 Lanikaula St
What is unusual about the garage here?
g = 3066 Woolsey Pl
What is unusual about the cut in the sidewalk for the driveway here?
h = 3056 Woolsey Pl
What is unusual about the property on the upslope side of the street here?
What is unusual about the sidewalk?
i = 3002 Woolsey Pl (end of the street)

This is the head of the scarp. What is unusual about the street drains and the pipes on
the valley side of the $500,000 concrete retaining wall on the valley-side of the street
here?
What is the condition of the houses on the valley-side of the street here?

GG 454 May 3, 2 0 0 2 Name ___________________2
j = 3084 Hulu Pl
What is unusual about the driveway here?
k = 3071 Hulu Pl
What is the condition of the house on the property northeast of 3071 Hulu Pl?
l = Lanikaula St & Kahaloa Dr

This is the northeast margin of the slide. Draw a plan view sketch showing what the
curb along the NW side of the road does here (include a scale).
m = 3103 Alani Dr (the house on the uphill side of the hairpin)

What conditions at this address seem unusual?
What is unusual about the drain in the street near Alani Dr? How long is it?
On the north side of the hairpin is a drain that goes all the way across the street. Draw
a plan view sketch showing what the curb along the NW side of the road does here
(include a scale).
n = 3081 Paty Dr
This is another part of the head of the slide. Where will the water drain here?

GG 454 May 3, 2 0 0 2 Name ___________________3
Map of Roads in the area of the Alani-Paty Landslide,

Manoa Valley, Oahu
Dr
n
d law
oo
W W
ai
d
pu
er R
na
R
Low
is
e
Lo
wr
Pu
ey
ha
Av
r
iD
la
Rd
e
Ka
an
Ri
o
loa
an
Al
se
s Pl
lui
Ka
ki
ood
E
ha
St
loa
nt W
Lo
iS
Pl
St
Dr
t
umo
nu no
Ka
Lo
Bea
l
St
Pl
e b m
St
i
eh
loa
Hi
ao
Dr
ha
law
Lo
St ty
Ka
wr
Ka
Pa
a
ul j
ey
Dr ka g k Pl
ni f
Av
n h lu
aw La e Hu
e
dl a
oo c
d
Pl
W y N
lse
oo
i W 0 100 m
Approximate boundary of slide
a = 3003 Kalawao St g = 3066 Woolsey Pl m = 3103 Alani Dr

b = 3047 Kalawao St h = 3056 Woolsey Pl n = 3081 Paty Dr
c = 3017 Lani kaula St i = 3002 Woolsey Pl
d = 3049 Lanikaula St j = 3084 Hulu Pl
e = 3055 Lanikaula St k = 3071 Hulu Pl
f = 3059 Lanikaula St l = Lanikaula St & Kahaloa Dr

Engineering Geology

Uploaded by

Copyright:

Available Formats

Engineering Geology

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Engineering Geology

Uploaded by

Copyright:

Available Formats

ENGINEERING GEOLOGY

Copyright © 2001 by S. J. Martel

http://www.soest.hawaii.edu/martel/Courses/GG454/ (1 of 3) [9/2/03 6:46:56 AM]

15-Characterizing ground response

REPORT ORGANIZATION SUGGESTIONS AND EVALUATION CRITERIA

http://www.soest.hawaii.edu/martel/Courses/GG454/ (2 of 3) [9/2/03 6:46:56 AM]

MANOA VALLEY FIELD TRIP QUESTIONS

http://www.soest.hawaii.edu/martel/Courses/GG454/ (3 of 3) [9/2/03 6:46:56 AM]

Day Date Topic Reading

Day Date Topic

IMPORTANT ENGINEERING & HYDROGEOLOGIC PROPERTIES OF ROCKS (02)

I Main Topic: Variety of qualities and quantities relevant to the

Stephen Martel 2-1 University of Hawaii

Linear elastic behavior

Stephen Martel 2-2 University of Hawaii

Components of stress at a point (the "on-in rule")

If i = j: normal stress: σ11, σ22, σ33

If i = j:shear stress: σ12, σ13, σ21, σ23, σ31, σ32

σ11 σ12 σ13

Stephen Martel 2-3 University of Hawaii

2 For seismology: Young's modulus affects seismic velocity

Stephen Martel 2-4 University of Hawaii

Conversions between the elastic moduli of rocks

Elastic moduli describe how a material will deform elastically under a

There are a few common ways strain is described. Longitudinal strain (ε )

Stephen Martel 2-5 University of Hawaii

Now consider the three-dimensional loading arrangement below, where the

As one might guess, K and E are related. For isotropic materials:

Stephen Martel 2-6 University of Hawaii

The Mechanical Energy in Flowing Water

Water flows from high potential energy to low potential energy.

How is the mechanical energy partitioned in the water?

Consider the mechanical energy contained by a small mass of water m.

Total energy = kinetic energy + elevation potential energy (1)

E total = 1/2 mv2 + mgh + Epressure . ( 2 )

Note that the dimension of pressure (Force/area) is the same as

E total /V = 1/2ρ v 2 + ρ gh + P. (3)

Dividing both sides of equation (3) through by the product ρ g, which in

E total /V ρ g = v2 / 2 g + h + P/gh. (4)

Stephen Martel 2-7 University of Hawaii

Stephen Martel 2-8 University of Hawaii

ENGINEERING & HYDROGEOLOGIC CHARACTERISTICS OF ROCKS (03)

Stephen Martel 3-1 University of Hawaii

B Crushed stone (for road fill, dams, riprap; $3 B in 1973)

Stephen Martel 3-2 University of Hawaii

Crystalline Rocks ("Hard" rocks)

Marble Carbonate 55-274/7-20 48-89 0.17-0.30 Soluble mm Fractures Fractures Sheets

Sedimentary Rocks ("Soft" rocks)

STRUCTURES AND DISCONTINUITIES ROCKS (04)

Stephen Martel 4-1 University of Hawaii

Folds in Cross Section

Stephen Martel 4-2 University of Hawaii

FOUR PLANAR GEOLOGIC STRUCTURES

For shear zones and faults, the relative displacement of neighboring

Deformation (displacement) is discontinuous across a fault

Deformation (displacement) is continuous across a shear zone

Stephen Martel 4-3 University of Hawaii

ORIENTATION OF LINES AND PLANES

Horizontal Line Horizontal Line

Plane Dip angle