Professional Documents
Culture Documents
Solved Problems in Geophysics (E - Buforn - Carmen - Pro - Agustin - Udias) PDF
Solved Problems in Geophysics (E - Buforn - Carmen - Pro - Agustin - Udias) PDF
Solving problems is an indispensable exercise for mastering the theory underlying the
various branches of geophysics. Without this practice, students often find it hard to
understand and relate theoretical concepts to their application in real-world situations.
This book is a collection of nearly 200 problems in geophysics, which are solved in
detail showing each step of their solution, the equations used and the assumptions made.
Simple figures are also included to help students understand how to reduce a problem to its
key elements. The book begins with an introduction to the equations most commonly used
in solving geophysical problems. The subsequent four chapters then present a series of
exercises for each of the main, classical areas of geophysics – gravity, geomagnetism,
seismology and heat flow and geochronology. For each topic there are problems with
different degrees of difficulty, from simple exercises that can be used in the most elemen-
tary courses, to more complex problems suitable for graduate-level students.
This handy book is the ideal adjunct to core course textbooks on geophysical theory. It is
a convenient source of additional homework and exam questions for instructors, and
provides students with step-by-step examples that can be used as a practice or revision aid.
Carmen Pro is an Associate Professor at the University of Extremadura, Spain, where she has
taught geophysics and astronomy for over 20 years. She has participated in several
geophysical research projects and is involved in college management.
Agustín Udías is an Emeritus Professor at UCM and is the author of a large number of papers
about seismicity, seismotectonics, and the physics of seismic sources, as well as the
textbook Principles of Seismology (Cambridge University Press, 1999). He has held
positions as Editor in Chief of Física de la Tierra and the Journal of Seismology and as
Vice President of the European Seismological Commission.
Solved Problems in Geophysics
ELISA BUFORN
Universidad Complutense, Madrid
CARMEN PRO
Universidad de Extremadura, Spain
AGUSTÍN UDÍAS
Universidad Complutense, Madrid
cambridge university press
Cambridge, New York, Melbourne, Madrid, Cape Town,
Singapore, São Paulo, Delhi, Mexico City
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
www.cambridge.org
Information on this title: www.cambridge.org/9781107602717
A catalogue record for this publication is available from the British Library
1 Introduction 1
Gravity 1
Geomagnetism 4
Seismology 6
Heat flow 10
Geochronology 11
2 Gravity 13
Terrestrial geoid and ellipsoid 13
Earth’s gravity field and potential 25
Gravity anomalies. Isostasy 53
Tides 95
Gravity observations 116
3 Geomagnetism 121
Main field 121
Magnetic anomalies 142
External magnetic field 156
Main (internal), external, and anomalous magnetic fields 174
Paleomagnetism 201
4 Seismology 208
Elasticity 208
Wave propagation. Potentials and displacements 211
Reflection and refraction 224
Ray theory. Constant and variable velocity 243
Ray theory. Spherical media 277
Surface waves 307
Focal parameters 324
Bibliography 352
v
Preface
This book presents a collection of 197 solved problems in geophysics. Our teaching
experience has shown us that there was a need for a work of this kind. Solving problems
is an indispensable exercise for understanding the theory contained in the various branches
of geophysics. Without this exercise, the student often finds it hard to understand and relate
the theoretical concepts with their application to practical cases. Although most teachers
present exercises and problems for their students during the course, the hours allotted to the
subject significantly limit how many exercises can be worked through in class. Although
the students may try to solve other problems outside of class time, if there are no solutions
available this significantly reduces the effectiveness of this type of study. It helps, there-
fore, both for the student and for the teacher who is explaining the subject if they have
problems whose solutions are given and whose steps can be followed in detail. Some
geophysics textbooks, for example, F.D. Stacey, Physics of the Earth; G.D. Garland,
Introduction to Geophysics; C.M. Fowler, The Solid Earth: An Introduction to Global
Geophysics; and W. Lowrie, Fundamentals of Geophysics, contain example problems, and,
in the case of Stacey’s, Fowler’s, and Lowrie’s textbooks, their solutions are provided on
the website of Cambridge University Press. The main difference in the present text is the
type of problems and the detail with which the solutions are given, and in the much greater
number.
All the problems proposed in the book are solved in detail, showing each step of their
solution, the equations used, and the assumptions made, so that their solution can be
followed without consulting any other book. When necessary, and indeed quite often, we
also include figures that allow the problems to be more clearly understood. For a given
topic, there are problems with different degrees of difficulty, from simple exercises that can
be used in the most elementary courses, to more complex problems with greater difficulty
and more suitable for teaching at a more advanced level.
The problems cover all parts of geophysics. The book begins with an Introduction
(Chapter 1) that includes the equations most used in solving the problems. The idea of
this chapter is not to develop the theory, but rather to simply give a list of the equations
most commonly used in solving the problems, at the same time as introducing the reader to
the nomenclature. The next four chapters correspond to the division of the problems into
the four thematic blocks that are classic in geophysics: gravity, geomagnetism, seismology,
and heat flow and geochronology. We have not included problems in geodynamics, since
this would depart too much from the approach we have taken, which is to facilitate
comprehension of the theory through its application to specific cases, sometimes cases
which are far from the real situation on Earth. Indeed, some of the problems may seem a bit
artificial, but their function is to help the student practise with what has been seen in the
vii
viii Preface
theory. Neither did we want to include specific problems of geophysical prospecting as this
would have considerably increased the length of the text, and moreover some of the topics
that would be covered in prospecting, such as gravimetric and geomagnetic anomalies, are
already included in other sections of this work.
Chapter 2 contains 68 problems in gravity divided into five sections. The first section is
dedicated to the terrestrial geoid and ellipsoid, proposing calculations of the parameters
that define them in order to help better understand these reference surfaces. The second
corresponds to calculating the gravitational field and potential for various models of the
Earth, including the existence of internal structures. Gravity anomalies are dealt with in the
third section, with a variety of problems to allow students to familiarize themselves with
the corrections to the observed gravity, with the concept of isostasy, and with the Airy and
Pratt hypotheses. The fourth section studies the phenomenon of the Earth’s tides and their
influence on the gravitational field. The last section is devoted to the observations of
gravity from measurements made with different types of gravimeters and the corrections
necessary in each case. We also include the application of these observations to the
accurate determination of different types of height.
Chapter 3 contains 42 problems in geomagnetism divided into five sections. The first is
devoted to the main (internal) field generated by a tilted dipole at the centre of the Earth. It
includes straightforward problems that correspond to the calculation of the geomagnetic
coordinates of a point and the theoretical components of the magnetic field. This section
also introduces the student to the use of the principal units used in geomagnetism. The
second considers the magnetic anomalies generated by different magnetized bodies and
their influence on the internal field. The third section is devoted to the external field and its
variation with time. In the fourth section, we propose problems of greater complexity
involving the internal field, the external field, and anomalous magnetized bodies at the
same time. The last section is devoted to problems in paleomagnetism.
Chapter 4 contains 69 problems in seismology divided into seven sections. The first
presents some simple exercises on the theory of elasticity. The second addresses the problem
of the propagation of seismic energy in the form of elastic waves, resolving the problems on
the basis of potentials, and calculating the components of their displacements. We study the
reflection and refraction of seismic waves in the third section. The fourth is devoted to the
problem of wave propagation using the theory of ray paths in a plane medium of constant and
variable velocity of propagation. The fifth studies the problem of the propagation of rays in a
spherical medium of either constant or variable propagation velocity, with the calculation of
the travel-time curves for both plane and spherical media. The sixth section contains
problems in the propagation of surface waves in layered media. The seventh section is
devoted to problems of calculating the focal parameters and the mechanism of earthquakes.
Chapter 5 includes 11 problems in heat flow with the propagation of heat in plane and
spherical media, and seven problems in geochronology involving the use of radioactive
elements for dating rocks.
Finally, we provide a bibliography of general textbooks on geophysics and of specific
textbooks for the topics of gravity, geomagnetism, and seismology. We have tried to
include only those most recent and commonly used textbooks which are likely to be found
in university libraries.
ix Preface
In sum, the book is a university text for students of physics, geology, geophysics,
planetary sciences, and engineering at the undergraduate or Master’s degree levels. It is
intended to be an aid to teaching the subjects of general geophysics, as well as the specific
topics of gravity, geomagnetism, seismology, and heat flow and geochronology contained
in university curricula.
The teaching experience of the authors in the universities of Barcelona, Extremadura,
and the Complutense of Madrid highlighted the need for a work of this kind. This text is
the result of the teaching work of its authors for over 20 years. Thanks are due to
the generations of students over those years who, with their comments, questions, and
suggestions, have really allowed this work to see the light. We are also especially grateful
to Prof. Greg McIntosh who provided us with some problems on paleomagnetism, to
Prof. Ana Negredo for her comments on heat flow and geochronology problems, and to
Dr R.A. Chatwin who worked on translating our text into English.
The text is an extension of the Spanish edition published by Pearson (Madrid, 2010).
Gravity
As a first approximation the Earth’s gravity is given by that of a rotating sphere. The
gravitational potential of a sphere of mass M is:
GM
V ¼
r
where r is the position vector (Fig. A) and G the universal gravitational constant.
If the sphere is rotating with angular velocity o the centrifugal potential at a point on the
surface is given by
1
F ¼ o2 r2 sin2 y
2
where y is the angle that r forms with the axis of rotation.
The gravity potential is their sum U ¼ V þ F.
The value of the acceleration due to gravity (the gravity ‘force’) is given by the gradient
of the potential:
g ¼ rU
The radial component of the gravity force is given by
GM
gr ¼ þ ro2 sin2 y
r2
The potential of the Earth to a first-order approximation corresponds to that of a rotating
ellipsoid, and is given by
GM a J2 a3 2
m r 2 2
U¼ 3sin ’ 1 þ cos ’
a r 2 r 2 a
where ’ ¼ 90º y is the geocentric latitude and a the equatorial radius.
The coefficient m is the ratio between the centrifugal and gravitational forces on the
sphere of radius a at the equator:
a3 o2
m¼
GM
1
2 Introduction
North Pole
P
q
r
a
Equator
Fig. A
g ¼ ge 1 þ bsin2 ’
The geocentric latitude of a point is the angle between the equator and the radius vector of
the point. The geodetic latitude is defined as the angle between the equatorial plane and the
normal to the ellipsoid surface at a point. Astronomical latitude is the angle between the
equatorial plane and the observed vertical at a point.
The normal or theoretical gravity at a point of geocentric latitude ’ referred to the
GRS1980 reference ellipsoid is
The effect of the Sun and Moon on the Earth is to produce the phenomenon of the tides.
If one considers more generally the tidal effect due to an astronomical body of mass M at a
distance R from the centre of the Earth, one must add the corresponding potential, which, in
the first-order approximation, is given by
GMr2
3cos2 #
c¼ 1
2R3
where r is the geocentric radius vector of the point, and # is the angle the position vector r
forms with the distance vector R.
Gravity anomalies, defined as Dg ¼ g –g, are the effects of the existence of anomalous
masses inside the Earth. The gravity anomaly along the Z (vertical) axis at a point distance
x along the horizontal axis produced by a sphere of radius R, density contrast Dr, and
buried at a depth d, is given by
@Va GM ðz þ dÞ
gðx; zÞ ¼ ¼h i3=2
@z
x2 þ ðz þ dÞ2
where Va is the potential produced by the anomalous spherical mass DM ¼ 4/3pR3 Dr.
For problems in two dimensions, one uses the anomaly produced by an infinite horizon-
tal cylinder at depth d, perpendicular to the plane under consideration. The anomalous
potential is given by
0 1
1
Va ¼ 2pGra2 ln@qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiA
B C
x2 þ ðz þ dÞ2
gFA ¼ g g þ 3:086h
4 Introduction
where g is the observed gravity, h the height in metres, and the anomaly is obtained in
gu (gravity units) mm s 2.
The Bouguer anomaly is
g B ¼ g g þ ð3:086 0:419rÞh
with r being the density of the plate of thickness h.
To account for isostatic compensation at height in mountainous areas, one adds an
isostatic correction which can be calculated assuming either the Airy or Pratt hypotheses.
With the Airy hypothesis, the root t of a mountain is given by
rc
t¼ h
rM rc
where rc and rM are the densities of the crust and mantle, and h is the height of the
mountain. For an ocean zone, with water density ra, the anti-root is
rc ra 0
t0 ¼ h
rM ra
With the Pratt hypothesis, the density contrast in a mountainous area is
h
r ¼ r r0 ¼ r
Dþh 0
where D is the level of compensation, h the height of the mountain, and r0 the density at
sea level. For an oceanic zone of depth h0 :
r0 D ra h0
r0 ¼
D h0
0
r ¼ r r0
The isostatic correction can be calculated using a cylinder of radius a and height b, whose
base is located at a distance c beneath the point, and with density contrast Dr:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ 2pGr b þ a2 þ ðc bÞ2
I
a2 þ c 2
Geomagnetism
To a first approximation, the internal magnetic field of the Earth can be approximated by a
centred dipole inclined at 11.5 to the axis of rotation. The potential created by a magnetic
dipole at a point distant r from its centre and forming an angle y with the axis of the dipole is
Cm cos y
F¼
r2
5 Geomagnetism
GNP
90º – fB
l – lB
180º – l∗ 90º – f
GMNP
q = 90º – f∗ D∗
Fig. B
The vertical and horizontal components of the field, the geomagnetic constant B0, and the
total field are given by:
Z ¼ 2B0 sin f
H ¼ B0 cos f
Cm
B0 ¼ 3
a qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F ¼ H 2 þ Z 2 ¼ B0 1 þ 3sin2 f
The units used for the components of the magnetic field are the tesla T and the nanotesla
nT ¼ 10 9 T. The NS (X*) and EW (Y*) components are
X ¼ H cos D
Y ¼ H sin D
and the declination and inclination are given by
6 Introduction
cos fB sinðl lB Þ
sin D ¼
cos f
tan I ¼ 2 tan f
r ¼ r0 cos2 f ¼ r0 sin2 y
where r0 is the radius vector of the point of the line of force located at the geomagnetic
equator.
Magnetic anomalies are produced by magnetic materials within the Earth. The anomal-
ous potential due to a vertical dipole buried at depth d is
Cmcosy Cmðz þ dÞ
FA ¼ ¼h
r2
i3=2
x2 þ ðz þ dÞ2
The vertical (z) and the horizontal (x) components of the magnetic anomaly at the surface
(z ¼ 0) produced by a vertical magnetic dipole at depth d are:
Cmð2d 2 x2 Þ
Z ¼
ðx2 þ d 2 Þ5=2
3Cmxd
X ¼
ðx2 þ d 2 Þ5=2
The Earth is affected by an external magnetic field produced mainly by the activity of the
Sun. This field is variable in time, with distinct periods of variation. The most noticeable is
the diurnal variation (Sq) with a maximum at 12 noon local time. The most important non-
periodic variations are the so-called magnetic storms.
Seismology
Earthquakes produce elastic waves which propagate through the interior and along the
surface of the Earth. Using the plane-wave approximation, the displacements of the
internal P- and S-waves (uiP and uiS) can be obtained from a scalar potential and a vector
potential:
ui ¼ ui P þ ui S ¼ ðr’Þi þ r cj
i
’ ¼ A exp ika gj xj at
cj ¼ Bj exp ikb gj xj bt
where A and Bj are the amplitudes, xj the coordinates of the observation point, ka and kb the
wavenumbers, gj are the direction cosines defined from the azimuth az and angle of
incidence i of the ray as:
7 Seismology
g1 ¼ sin i cos az
g2 ¼ sin i sin az
g3 ¼ cos i
and a and b are the P- and S-wave velocities of propagation, respectively, defined from
the Lamé coefficients (l and shear modulus m) and the density r:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi
l þ 2m
vP ¼ a ¼
r
rffiffiffi
m
vS ¼ b ¼
r
3
Units used are: displacement amplitudes (u) in µm; potential amplitudes (A, Bi) in 10 m 2;
wavenumber (k) in km 1; and wave velocity (a, b) in km s 1.
Poisson’s ratio is defined in terms of the Lamé coefficients as
l
s¼
2ð l þ m Þ
The angle of polarization of S-wave e is defined as
1 uSH
e ¼ tan
uSV
where uSH is the amplitude of the SH component, and uSV that of the SV component. SH
and SV are the horizontal and vertical components of the S-wave on the wavefront plane.
The coefficients of reflection V and transmission W are given by the respective
ratios between the amplitudes of the reflected or transmitted potentials and the incident
potential:
A
V¼
A0
A0
W ¼
A0
where A0 is the amplitude of the incident wave potential, A that of the reflected potential,
and A0 of the transmitted potential.
Snell’s law for plane media is expressed as
sin i
p¼
v
and for spherical media
r sin i
p¼
v
where p is the ray parameter, i the angle of incidence, v the propagation velocity of the
medium, and r the position vector along the ray.
8 Introduction
In the case of plane media with propagation velocity varying with depth v(z), the
epicentral distance and the travel time of a ray for a surface focus are given by
ðh
pdz
x ¼ 2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0 2 p2
ðh
2 dz
t ¼ 2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0 2 p2
where ¼ v 1 and h is the depth of maximum penetration of the ray. The variation of the
epicentral distance x with the ray parameter p is given by
dB
dx 2
ðB dz
¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi þ 2 pdz
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
dp B0 20 p2 0 B2 2 p2
where
1 dv
B¼
v dz
In spherical media with velocity varying with depth v(r), the epicentral distance,
trajectory along the ray, and travel time are given by
ð r0
p dr
¼2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
rp r 2 r2
ð r0
dr
s ¼ 2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
rp 2 r2
ð r0
dr
t¼2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
rp v 2 r2
where ¼ rv 1, r0 is the radius at the surface of the Earth, and rp is the radius at the point
of maximum penetration of the ray.
The variation of the distance from the epicentre D with the ray parameter p in a spherical
medium is
dB
d 2
ðB dr
¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi þ 2 drpffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
dp ð1 B0 Þ 20 p2 0 ð1 B 2 Þ 2 p2
where
r dv
B¼
v dr
The radial and vertical components (u1 and u3) of surface waves can be obtained from
the potentials ’ and c. The transverse component (u2) is kept apart
@’ @c
u1 ¼ ¼ ’;1 c;3
@x1 @x3
u2 ¼ C exp½ iksx3 þ ikðx1 ctÞ
@’ @c
u3 ¼ þ ¼ ’;3 þ c;1
@x3 @x1
9 Seismology
For surface waves, c < b < a, and hence r and s are imaginary.
For dispersive waves, the relationship between the phase velocity c and the group
velocity U is
dc
U ¼cþk
dk
where k is the wavenumber.
The position of the seismic focus is given by the coordinates of the epicentre (’0, l0)
and the depth h. The time is that of the origin of the earthquake t0. The size is given by
the magnitude which is proportional to the logarithm of the amplitude of the recorded
waves. For surface waves this is:
A
Ms ¼ log þ 1:66 log þ 3:3
T
where A is the amplitude of ground motion in microns, T is the period in seconds, and ∆ the
epicentral distance in degrees.
The magnitude of the moment is given by
2
Mw ¼ log M0 6:1
3
where M0 is the seismic moment in N m (newton metres). The seismic moment is related to
the displacement of the fault ∆u and its area S:
M0 ¼ muS
The mechanism of earthquakes is given by the orientation of the fracture plane (fault)
defined by the angles ’ (azimuth), d (dip), and l (slip angle or rake), or by the vectors n
(the normal to the fault plane) and l (the direction of slip).
The elastic displacement of the waves produced by a point shear fault is
@Gki
uk ðxs ; t Þ ¼ muðt ÞS li nj þ lj ni
@xj
where Gki is the medium’s Green’s function which, for an isotropic, homogeneous, infinite
medium, and P-waves in the far-field regime, is given by
1 r
GPki ¼ 2
gi gk d t
4pra r a
10 Introduction
u_ ðt ÞS
uPk ðxs ; t Þ ¼
3
m l i nj þ l j ni g i g j g k
4pra r
This equation can be expressed also in terms of the moment tensor Mij
M_ ij ðt Þ
uPk ðxs ; t Þ ¼ ggg
4pra3 r i j k
Mij is a more general representation of a point source.
Heat flow
The Fourier law of heat transfer by diffusion states that the heat flux q_ is proportional to the
gradient of the temperature T:
q_ ¼ KrT
where K is the thermal conductivity coefficient. The units of heat flow are W m 2.
The heat diffusion equation, assuming that K is constant, is given by
e @T
kr2 T þ ¼
rCv @t
where Cv is the specific heat, r the density, e the heat generated per unit volume and unit
time (heat sources), and k the thermal diffusivity:
K
k¼
rCv
If there are no heat sources, the diffusion equation is
@T
kr2 T ¼
@t
In the case of one-dimensional flow with periodic variation of temperature over time,
one has:
rffiffiffiffiffiffi rffiffiffiffiffiffiffiffi
o o
T ð z; t Þ ¼ T0 exp zþi z þ ot
2k 2k
where z is the vertical direction (positive towards the nadir) and o the angular frequency.
In the case of stationary one-dimensional solutions (T constant in time) one obtains from
the diffusion equation:
e 2 q_ 0
T¼ z þ z þ T0
2K K
11 Geochronology
where T0 and q_ 0 are the temperature and flow at the surface (z ¼ 0).
For a spherical Earth, assuming that the thermal conductivity is constant, and that
the amount of heat per unit volume depends only on time, the diffusion equation takes the
form:
2
@ T 2 @T @T
K þ þ eðtÞ ¼ rCv
@r2 r @r @t
1 d 2 dT e
r ¼
r2 dr dr K
Integrating twice, one has
e 2
r2
T ¼ T0 þ R
6K
where T0 is the temperature at the surface (r ¼ R).
Geochronology
Geochronology is based on determining the age of a rock by measuring the decay of its
radioactive elements. In a sample of radioactive material, the number of atoms that have
yet to disintegrate after time t is given by
lt
nt ¼ n0 e
where n0 is the initial number of atoms, and l the decay constant. The rate of decay dn/dt is
the activity R, so that
lt
R ¼ R0 e
where R0 is the initial activity (at t ¼ 0).
The half-life (or period) of the sample is the time it takes for the activity R to fall to half
its initial value. It is given by:
0:693
T1= ¼
2 l
The mean life-time t of one of the atoms that existed at the start is given by:
1
t ¼
l
If a sample consists of NR radioactive nuclei and NE stable nuclei, the time to arrive at the
propotion NE/NR is given by
12 Introduction
1 NE
t¼ 1þ
l NR
If the rubidium–strontium (Rb-Sr) method is used to date a sample, a correction must be
made for the contamination of the stable 86Sr isotope relative to the radioisotope 87Sr:
87 87 87
Sr Sr Rb lt
86 Srtotal
¼ 86 Sr initial
þ 86 Sr
e 1
o 2 a3
m¼
GM
where G ¼ 6.671011 m3 kg1 s2 is the gravitational constant, M the Earth’s mass, and
the angular velocity is o = 2p/T, where T is the rotation period (T ¼ 24 h). We obtain for
the semi-axis a the value
1=3
mGMT 2
a¼ ¼ 6378:127 km
4p2
The Earth’s flattening a can be obtained directly since we already know a and c so
a c 6378:127 6356:742
a¼ ¼ ¼ 3:3529 103
c 6378:127
The radius vector to the point P is given by the equation r ¼ a(1 a sin2 ’ )
From this equation we can calculate the geocentric latitude ’:
C r
j jd
Fig. 1
Data
The normal or theoretical gravity at a point can be expressed in terms of the normal gravity
at the equator ge, the gravity flattening b, and the latitude of the point ’:
g1 ¼ ge 1 þ b sin2 ’1 g2 ¼ ge 1 þ b sin2 ’2
g1 1 þ b sin2 ’1
¼
g2 1 þ b sin2 ’2
15 Terrestrial geoid and ellipsoid
From this expression we can obtain the gravity flattening, since we already know g1, g2, ’1, ’2:
g1 g2
b¼ ¼ 5:297 103
g2 sin2 ’1 g1 sin2 ’2
The distance r from the centre of the ellipsoid to points on its surface can be given as a
function of the flattening a, the equatorial radius a, and the latitude ’:
r1 ¼ að1 a sin2 ’1 Þ
r2 ¼ að1 a sin2 ’2 Þ
r1 1 asin2 ’1
¼
r2 1 asin2 ’2
a ¼ 3:353 103
From this value we find the equatorial radius,
r1
a¼ ¼ 6378:137 km
1 asin2 ’1
The polar radius c can be found from this value and the flattening:
ac
a¼ and c ¼ að1 aÞ ¼ 6356:751 km
a
The coefficient m is obtained from a and b:
5 2
aþb¼ m m ¼ ða þ bÞ ¼ 3:460 103
2 5
From this value we can obtain the value of the Earth’s mass M from
o 2 a3
m¼
GM
2p
with o ¼ p where T ¼ 24 hours.
T
Therefore
4p2 a3
M¼ ¼ 5:946 1024 kg
T 2 Gm
3. Obtain the value of the terrestrial flattening in the first-order approximation, given
that the normal gravity values for two points of the ellipsoid are:
Point 1: w1 ¼ 42º 200 g1 ¼ 980.389 063 Gal
Point 2: w2 ¼ 47º 300 g2 ¼ 980.854 830 Gal
Take the equatorial radius to be 6378.388 km.
16 Gravity
g1 ¼ ge ð1 þ bsin2 ’1 Þ
g2 ¼ ge ð1 þ bsin2 ’2 Þ
g1 1 þ bsin2 ’1
¼
g2 1 þ bsin2 ’2
o 2 a3 4p2 a3
m¼ ¼ 2 ¼ 3:442 5698 103
GM T GM
And finally the Earth’s flattening is
5
a ¼ m b ¼ 3:318 1575 103
2
o 2 a3 4p2 a3
m¼ ¼ 2 ¼ 3:4429 103
GM T GM
The gravity flattening b is given by
5
b ¼ m a ¼ 5:2072 103
2
The dynamic form factor J2 is found from the relation
2a m
J2 ¼ ¼ 1:1190 103
3
5. At a point P on the ellipsoid at latitude 50 ºS, the value of normal gravity is 9.810
752 m s2 and the distance to the centre of the Earth is 6365.587 km. Given that
the mass of the Earth is 5.976 1024 kg and the ratio between the minor and major
semi-axes is c/a ¼ 0.996 6509, calculate:
(a) The flattening, equatorial radius, gravity flattening, dynamic form factor, and
coefficient m.
(b) The normal gravity at the equator.
(c) The centrifugal force at P.
Data
g ¼ ge ð1 þ b sin2 ’Þ
The value of b is given by
g ge
b¼ ¼ 5:302 103
ge sin2 ’
The coefficient m, taking T ¼ 24 hours, is
4p2 a3
m¼ ¼ 3:442 103
T 2 GM
a and J2 are determined from the equations
5 2 m
a ¼ m b ¼ 3:303 103 J2 ¼ a ¼ 1:055 103
2 3 2
19 Terrestrial geoid and ellipsoid
o 2 a3
m¼ ¼ 3:4425 103
GM
From a and m we find the gravity flattening b:
5
b ¼ m a ¼ 5:2535 103
2
Normal gravity at the equator is found from the value of the gravity at point P:
g
ge ¼ ¼ 9:780 72 m s2
1 þ b sin2 ’
From this value we find the normal gravity at the Pole:
gp ¼ ge ð1 þ bÞ ¼ 9:832 10 m s2
10. Let two points of the ellipsoid be of latitudes w1 and w2, with radius vectors
6372.819 km and 6362.121 km, respectively. The ratio of the normal gravities is
0.997 37, the flattening 3.3529 103, and the gravity flattening 5.2884 103.
Calculate:
(a) The Earth’s mass.
(b) The latitude of each point and the dynamic form factor.
(a) The equatorial radius a can be obtained from the ratio of the two normal gravities
a r
1
g1 1 þ bsin ’12 1 þ b
¼ ¼ aa ¼l
g2 1 þ bsin2 ’2 1 þ b a r2
aa
where l ¼ 0.997 37. We solve for a and obtain
bðlr2 r1 Þ
a¼ ¼ 6382:94 km
ð a þ bÞ ð l 1 Þ
We calculate the mass of the Earth from the flattening and gravity flattening as in Problem 2:
2 4p2 a3
m ¼ ða þ bÞ ¼ 3:4565 103 M¼ ¼ 5:9653 1024 kg
5 T 2 Gm
(b) The latitudes at each point are calculated from the radius vectors
a r1
sin2 ’1 ¼ ! ’1 ¼ 43 270 5800
aa
a r2
sin2 ’2 ¼ ! ’2 ¼ 80 330 4600
aa
The dynamic form factor is obtained from the values of a, b, and m:
9
b ¼ 2a þ m J2
2
2
J2 ¼ ð2a þ m bÞ ¼ 1:0831 103
9
11. Let a point A have a value of gravity of 9793 626.8 gu and a geopotential number
of 32.614 gpu. Calculate the gravity at a point B, knowing that the increments in
dynamic and Helmert height over point A are 271.116 m and 271.456 m, respectively.
Take g45 ¼ 9.806 2940 m s2. Give the units for each parameter.
The dynamic heights at points A and B are given by:
CA
HDA ¼
g45
C B
HDB ¼
g45 !
N
P
where C is the value of the geopotential at each point gj dhj and g45 the normal
j¼1
gravity for a point on the ellipsoid at 45º latitude.
22 Gravity
If heights are given in km and normal gravity in Gal, geopotentials are in gpu (geopotential units)
1 gpu ¼ 1 kGal m ¼ 1 Gal km
The Helmert orthometric height H is given by
C
H¼ ð11:1Þ
g þ 0:0424H
where C is in gpu, g in Gal, and H in km.
Solving for H:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
g g2 þ 4 0:0424C
H¼
2 0:0424
Since the point A is above the geoid (CA > 0), we take the positive solution,
HA ¼ 33:301 m
Then, the Helmert height at point B is
HB ¼ HA þ HAB ¼ 304:757 m
between that point and the geoid along the direction of the plumb-line, and
the point’s geopotential number.
The mean value of gravity between a height H and the surface of the geoid is given by
1 H
ð
g ¼ gðzÞdz
H 0
where g(z) is the value of gravity at a distance z from the geoid along the vertical path to a
point of height H. This value can be obtained using the Poincaré and Prey reduction from
the value of g observed at the Earth’s surface at a point of height H,
gðzÞ ¼ g þ 0:0848ðH zÞ
Then
1 H 1 H
ð ð
g ¼ gðzÞdz ¼ ½g þ 0:0848ðH zÞdz
H 0 H 0
1
H
¼ gz þ 0:0848 Hz 0:0424z2 0
H
g ¼ g þ 0:0424 H ¼ 979:6659 Gal
The geopotential at B can be obtained from the gross increment in elevation between A and B,
g þ g
A B
CB ¼ CA þ hBA
2
and, solving for gB,
2ðCB CA Þ
gB ¼ gA ¼ 980:103 08 Gal
hBA
A 979.88696 – 664.982 ? ?
B ? 0.541 ? 677.577 ?
C 979.88665 ? ? ? 657.134
Station A
Dynamic height:
CA
HDA ¼ ¼ 678:118 m
g45
Helmert height:
C
H¼
g þ 0:0424H
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð15:1Þ
g g2 þ 4 0:0424 C
H¼ ¼ 678:611 m
2 0:0424
Station B
From this value and the difference in height with respect to station A we find the gravity at B:
g þ g
A B
CB ¼ CA þ hA B
2
from which we get gB = 979.877 84 Gal
The Helmert height of B is found as in station A:
HB ¼ 678:077 m
Station C
From the known values of gravity and Helmert height we find the geopotential number
(Equation 15.1)
To calculate the difference in height of C with respect to B we begin with the expression
g þ g
B C
CC ¼ CB þ hCB
2
from which
2ðCC CB Þ
hCB ¼ ¼ 2:937 m
gB þ gC
The dynamic height is found directly from the geopotential number:
CC
HDC ¼ ¼ 674:642 m
g45
The complete table is:
16. Suppose an Earth is formed by a sphere of radius a and density r, and within it
there are two spheres of radius a/2 with centres located on the axis of rotation. The
density of that of the northern hemisphere is 5r and of that of the southern hemi-
sphere is r /5. The value of the rotation is such that m ¼ 0.1. Determine:
(a) The potential U in the r3 approximation.
(b) The values of gr and gu for a point on the equator in the r2 approximation.
26 Gravity
(c) The error made in (b) with respect to the exact solution.
(d) The deviation of the vertical from the radial at the equator.
(a) The gravitational potential is the sum of the potentials of the three spheres
GM GM1 GM2
V ¼ V1 þ V2 þ V3 ¼ þ þ 0 ð16:1Þ
r q q
where r is the distance from a point P to the centre of the sphere of radius a and mass M,
where M is given by
4
M ¼ pra3
3
q and q0 are the distances to the centres of the two spheres in its interior in the northern and
southern hemispheres which have differential masses M1 and M2, respectively (Fig. 16).
The differential masses are those corresponding to the difference in density in each case
with respect to the large sphere:
4 a3 M
M1 ¼ pð5r rÞ ¼
3 8 2
differential mass of the sphere in the northern hemisphere
4 r a3 M
M2 ¼ p r ¼
3 5 8 10
differential mass of the sphere in the southern hemisphere
The distance q can be calculated using the cosine law
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a2 ffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
a a 2 a
q¼ r þ 2 r cos y ¼ r 1 þ 2 cos y
2 2 2r 2r
Considering this expression, 1/q corresponds to one of the generating functions of the
Legendre polynomials. Then 1/q, in the first-order approximation, is given by
1 1 a 1 a 2 2
¼ 1 þ cos y þ 3cos y 1
q r 2r 2 2r
Since cos y0 = cos y, 1/q0 is given by
1 1 a 1 a 2 2
¼ 1 cos y þ 3cos y 1
q0 r 2r 2 2r
P
5ρ q
r
r2 a/2
α θ q'
a
θ'
r3 ρ
ρ/5
Fig. 16
27 Earth’s gravity field and potential
If we substitute in Equation (16.1), the potentials for each sphere are given by
a2
GM GM 1 a
þ 2 cos y þ 3 3cos2 y 1
V1 ¼ V2 ¼
r 2 r 2r 8r
a2
GM 1 a 2
V3 ¼ cos y þ 3 3cos y 1
10 r 2r2 8r
Then, the total gravity potential is the sum of the three gravitational potentials plus the
potential of the centrifugal force due to the rotation:
1 1 1 1 1 a
U ¼ GM 1 þ þ þ cos y
2 10 r 4 20 r2
1 a2
1 2
1 2 2 2
þ 3 cos y 1 þ r o sin y
16 80 r3 2
o2 a3
In terms of the coefficient m, given here by m ¼ ,
GM
GM 7 a 3 a2 1 a3
2
1 r 2 2
U¼ þ cos y þ 3cos y 1 þ m sin y
a 5 r 10 r2 20 r3 2 a
(b) Using this first-order approximation of the potential, the radial and tangential
components of gravity at the equator, r = a and y = 90 , putting m ¼ 0.1, are
@U GM
gr ¼ ¼ 1:3 2
@r a
1 @U GM
gy ¼ ¼ 0:3 2
r @y a
(c) To calculate exactly the value of gr at the equator we have to calculate
the exact contribution of each of the three spheres plus the centrifugal force (m ¼ 0.1):
GM
gr ¼ gr1 þ gr2 þ gr3 m
a2
GM
gr1 ¼
a2
GM
gr2 ¼ 2 cos a
2r2
GM
gr3 ¼ cos a
10r32
where
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a2
r 2 ¼ a2 þ ¼ r 3
4
and a is the angle which forms r2 and r3 with the equator (Fig. 16)
28 Gravity
a=2 1
sin a ¼ ¼ pffiffiffi
r2 5
Then
GM 4 4 GM
gr ¼ 2 1 þ pffiffiffi pffiffiffi m ¼ 1:19 2
a 5 5 25 5 a
0 1
B GM 1 GM 1 C
gy ¼ gy1 þ gy2 ¼ @ pffiffiffi þ pffiffiffiA
5 2 5 5
2 a 10 a2 5
4 4
GM 2 2 GM
¼ 2 pffiffiffi þ pffiffiffi ¼ 0:14 2
a 5 5 25 5 a
The error made in the first-order approximation with respect to the exact solution is
GM GM
gr ¼ ð1:19 þ 1:3Þ ¼ 0:11 2
a2 a
GM GM
gy ¼ ð0:3 þ 0:14Þ 2 ¼ 0:16 2
a a
(d) The deviation of the vertical with respect to the radial direction is given by the angle i
which is determined from the gravity components gr and gy. At the equator this angle is:
17. A spherical planet is formed by a sphere of radius a and density r, and inside it a
sphere of radius a/2 and density 5r centred at the midpoint of the radius of the
northern hemisphere. There is no rotation.
(a) Determine J0, J1, and J2.
(b) What is the deviation of the vertical from the radial at the equator?
(a) The total gravitational potential is the sum of the potentials of the two spheres
(Fig. 17) where g, is the attraction due to the potential V1 and g2 that due to the
potential V2:
GM GM 0
V ¼ V1 þ V2 ¼ þ
r q
where r and q are the distances from a point P to the centres of the large and small spheres,
respectively.
29 Earth’s gravity field and potential
P
5r q
a /2
q g2
i
a g1
Fig. 17
As we did in Problem 16, for the small sphere of radius a/2 we take the differential
mass M 0
4 a3 16 M
M 0 ¼ p ð5r rÞ ¼ pra3 ¼
3 8 24 2
where M is the mass of the sphere of radius a and density r.
For 1/q we take the first-order approximation of the Legendre polynomial, as we did in
Problem 16:
1 1 a 1 a 2 2
¼ 1 þ cos y þ 3cos y 1
q r 2r 2 2r
Then, the expression for the gravitational potential V is:
1 a2
GM GM 1 a 2
V ¼ þ þ 2 cos y þ 3cos y 1
r 2 r 2r 4 4r3
a2
3 a 2
¼ GM þ cos y þ 3cos y 1
2r 4r2 32r3
GM a a 2 a3 1
J2 3cos2 y 1
V ¼ J0 þ J1 cos y þ
a r r r 2
3
J0 ¼
2
1
J1 ¼
4
1
J2 ¼
16
The components of gravity at the surface of the large sphere (r ¼ a) are:
@V GM 3 1 3
3cos2 y 1
gr ¼ ¼ 2 cos y
@r a 2 2 32
1 @V GM 1 3
gy ¼ ¼ 2 sin y cos y sin y
r @y a 4 16
and at the equator, y ¼ 90 :
45GM
gr ¼
32a2
GM
gy ¼ 2
4a
At the equator the deviation of the vertical with respect to the radial direction is
gy 8
tan i ¼ ¼
gr 45
i ¼ 10:08
18. Suppose an Earth is formed by a sphere of radius a and density r, and within it
there are two spheres of radius a/2 and density 2r with centres located on the axis of
rotation in each hemisphere. If M is the mass of the sphere of radius a, calculate:
(a) The potential U(r,u) and the form of the equipotential surface passing through the Poles.
(b) The component gr of gravity in the first-order approximation for points on the surface.
(c) Calculate gr directly at the Pole and the equator, and compare with the first-order
approximation.
(a) This problem is similar to Problem 16, but now the density of the two spheres is the
same. The total gravity potential is the sum of the gravitational potentials of the
three spheres (V, V1 and V2) plus the potential due to the rotation F:
U ¼ V þ V1 þ V2 þ F ð18:1Þ
where
GM
V¼
r
GM 0
V1 ¼
q1
GM 0
V2 ¼
q2
1 2 2 2
F ¼ o r sin y
2
31 Earth’s gravity field and potential
P
q1
2r
r
r
q2
a/2
q
q
a
a
q’
2r
Fig. 18
and where M is the mass of the large sphere of radius a and M 0 the differential mass of each
of the small spheres of radius a/2, r is the distance from a point P to the centre of the large
sphere, and q1 and q2 the distances from P to the centres of the small spheres (Fig. 18). As
in Problem 16 the differential mass is given by the difference in density between the large
and the small spheres:
4 a 3 M
M 0 ¼ pð2r rÞ ¼
3 2 8
GM 1 1 a3 o2 3
GM m r 3 2
F¼ 3
r sin2 y ¼ sin y
r 2 a GM r 2 a
Substituting in Equation (18.1)
GM 10 1 a 2 2
r 3 m 2
U¼ þ 3cos y 1 þ sin y ð18:2Þ
r 8 8 2r a 2
32 Gravity
3
m¼0)r ¼a 1 < a ) prolate ellipsoid
42
(b) For the gravity at the Pole, in the first-order approximation, we take the derivative
of the potential (18.2) and substitute y ¼ 0 ) rp ¼ a:
@U GM 10 6 GM
gr ¼ ¼ 2 ¼ 1:4375 2
@r a 8 32 a
(c) The exact solution for the gravity at the pole is the sum of the attractions of the
three spheres:
GM GM GM GM
gr ¼ 2 ¼ 1:5555 2
a2 2a 18a2 a
33 Earth’s gravity field and potential
At the equator we take the derivative of the potential and substitute r = a and
y = 90 :
@U GM 10 3
gr ¼ ¼ 2 þ þm
@r a 8 32
GM 37 GM
¼ 2 m ¼ 2 ½1:1562 m
a 32 a
For the exact solution we write
GM 2GM
gr ¼ cos a þ o2 a
a2 8q2
From Fig. 18 the distance q is given by
a2 5
q2 ¼ þ a2 ¼ a2
4 ffiffiffi
r 4
4
cos a ¼
5
Therefore
" rffiffiffi #
GM 8 4 GM
gr ¼ 2 1 þ m ¼ 2 ½1:1789 m
a 40 5 a
Then, J2 = 0.05.
The flattening is obtained from the relation
3 m
a ¼ J2 þ
2 2
3 2:3 103
a¼ 50 103 þ ¼ 0:0765
2 2
34 Gravity
X3
2r
r
a/2
X1
a
2r
Fig. 19a
2 2 M a2
IC ¼ Isph a þ 2Isph a2 ¼ Ma2 þ 2 ¼ 0:425Ma2
5 58 4
For the A-axis (x1) the moment of inertia of each of the small spheres is given by
(Fig. 19b)
35 Earth’s gravity field and potential
a b
Fig. 19b
I ¼ ICM þ Mh2
since in this case the A axis does not coincide with the centre of mass, where R ¼ a/2
and h ¼ a/2:
2 M a2 M a 2
Isph a=2 ¼ þ ¼ 0:044Ma2
58 4 8 4
2
IA ¼ Isph a þ 2Isph a=2 ¼ Ma2 þ 2 0:044Ma2 ¼ 0:488Ma2
5
Finally
C A IC IA 0:425 0:488
H¼ ¼ ¼ ¼ 0:147
C IC 0:425
20. Suppose an Earth is formed by a sphere of radius a and density r, and within it
there is a sphere of radius a/2 and density 5r centred at the midpoint of the northern-
hemisphere polar radius. If m ¼ 1/8 and M is the mass of the sphere of radius a,
determine:
(a) The form of the equipotential surface passing through the North Pole.
(b) For latitude 45º, the astronomical latitude and the deviation of the vertical from
the radial.
(a) The gravitational potential is the sum of the potentials for the sphere of radius a
and that of the sphere of radius a/2 (Fig. 20):
GM GM1
V ¼ V1 þ V2 ¼ þ
r q
As in the previous problems the potential of the small sphere is given in terms of
differential mass M1:
36 Gravity
gq i
a/2
q P
g
5r
gr
q j
fa
a
Fig. 20
4 a3 M
M1 ¼ pð5r rÞ ¼
3 2 2
and for the inverse of the distance 1/q we use the approximation
1 1 a a 2 1
2
¼ 1 þ cos y þ 3cos y 1
q r 2r 2r 2
Then, the total gravitational potential is
GM 3 a 1 a2
3cos2 y 1
V ¼ þ cos y þ
r 2 4r 16 r
The total potential U is the sum of the gravitational potential V plus the potential of rotation
F, where
1
F ¼ r2 o2 sin2 y
2
and using the coefficient m = o2a3/GM = 1/8, we have
GM 3 a 1 a2 r 3 m 2
U¼ þ cos y þ 3cos2 y 1 þ sin y
r 2 4r 16 r a 2
At the North Pole, y = 0 and r ¼ a, and the value of the potential is
GM 3 1 1 15GM
Up ¼ þ þ ¼
a 2 4 8 8a
37 Earth’s gravity field and potential
(b) The deviation of the vertical with respect to the radial direction is given by the angle i:
gy
tan i ¼
gr
To find this value we have to calculate the two components of gravity
3a2 1
@U 3 1 2a 1 2
2r 2
gr ¼ ¼ GM 2 cos y 3 cos y 1 þ sin y
@r 2r 4 r3 16 r4 16a3
a2 r2
1 @U GM a
gy ¼ ¼ 2 sin y 6 cos y sin y þ sin y cos y
r @y r 4r 16r3 8a3
For a point on the surface we put r ¼a:
GM 19 1 11 2
gr ¼ 2 þ cos y þ cos y
a 16 2 16
GM 1 1
gy ¼ 2 sin y þ sin y cos y
a 4 4
and for latitude 45º
GM
gr ¼ 1:88
a2
GM
gy ¼ 2 0:30
a
0:30
Then the angle i is given by tan i ¼ ) i ¼ 9:0.
1:88
The astronomical latitude is
fa ¼ 90 y i ¼ 36:0
21. If the internal sphere of Problem 20 is located on the equatorial radius at
longitude zero, find expressions for the components of gravity: gr , gu , gl.
As in the previous problem the differential mass of the small sphere M1 is (Fig. 21a):
M
M1 ¼
2
The total potential U is the sum of the gravitational potentials V and V1, and the potential
due to rotation F. According to Fig. 21b, using the relations of spherical triangles, if ’ and
l are the coordinates of the point where the potential is evaluated, then
38 Gravity
q r
y
j
l = 0° A
a/2
Fig. 21a
90° 90°– j
l
Fig. 21b
a2
GM a 2 2
V1 ¼ 1 þ cos ’ cos l þ 2 3cos ’cos l 1
2r 2r 8r
The total potential U is given by:
a2
GM 3 a 2 2
1 2 2 2
U¼ þ cos ’ cos l þ 3cos ’cos l 1 þ r o cos ’
r 2 4r 16r2 2
39 Earth’s gravity field and potential
a2 3a2
@U 3 2 2
r 2
gr ¼ ¼ GM 2 3 sin y cos l 3sin ycos l 1 þ msin y
@r 2r 2r 16r4 a3
2
a2
1 @U a r
gy ¼ ¼ GM cos y cos l þ 6 cos l sin y cos y þ m sin y cos y
r @y 4r3 16r4 a3
a2 3a2
1 @U
gl ¼ ¼ GM 3 sin l 4 sin y cos l sin l
r sin y @l 4r 8r
22. A planet is formed by a sphere of radius a and density r, with a spherical core of
density 5r and radius a/2 centred on the axis of rotation in the northern hemisphere
and tangential to the equator. The planet rotates with m ¼ 1/4. For the point at
coordinates (45º N, 45º E), calculate:
(a) The astronomical latitude.
(b) The deviation of the vertical from the radial.
(c) The angular velocity of rotation that would be required for this deviation to be zero.
(a) The gravitational potential is the sum of the potentials of the two spheres (Fig. 22):
GM GM 0
V ¼ V1 þ V2 ¼ þ ð22:1Þ
r q
gq i
a/2
q P
5r
g
gr
Fig. 22
40 Gravity
As in Problem 16, the inverse of the distance from a point P to the centre of the small
sphere, 1/q, can be approximated by
1 1 a 1 a 2
3cos2 y 1
¼ 1 þ cos y þ
q r 2r 2 2r
As in previous problems we use the differential mass of the small sphere,
4 a3 2
M 0 ¼ p ð5r rÞ ¼ pra3
3 8 3
4
and since M ¼ pra3 , then M 0 = M/2.
3
Substituting in Equation (22.1) we obtain
1 a2
GM GM 1 a 2
V ¼ þ þ 2 cos y þ 3cos y 1
r 2 r 2r 2 4r3
a2
3 a 2
¼ GM þ 2 cos y þ 3cos y 1
2r 4r 16r3
The total potential U is the sum of V plus the potential due to rotation F:
a2
GM 3 a 1
3cos y 1 þ r2 o2 sin2 y
2
U ¼V þF¼ þ cos y þ
r 2 4r 16r2 2
The components of gravity gr and gy are
3a2
@U 3 a 2
þ ro2 sin2 y
gr ¼ ¼ GM 2 3 cos y 3cos y 1 ð22:2Þ
@r 2r 2r 16r4
a2
1 @U a
gy ¼ ¼ GM 3 sin y 6 cos y sin y þ ro2 sin y cos y ð22:3Þ
r @y 4r 16r4
For a point on the surface of the large sphere and coordinates 45º N, 45º E, we have that
y = 45 , r ¼ a, and
a3 o 2 1
m¼ ¼
GM 4
Putting these values in (22.2) and (22.3), we obtain
GM GM
gr ¼ 1:82 and gy ¼ 0:24 2
a2 a
To find the astronomical latitude we first have to find the deviation of the vertical with
respect to the radial:
gy
tan i ¼ ¼ 0:13 ; i ¼ 7:5
gr
The astronomical latitude is, then, f = ’ i = 45 7.5 = 37.5 .
(b) The deviation of the vertical with respect to the radial direction, as already found, is i = 7.5 .
(c) If we want the deviation of the vertical to be null, i ¼ 0, this implies gy = 0.
41 Earth’s gravity field and potential
a3 o2
m¼ ð22:4Þ
GM
we have
pffiffiffi
GM 2 2þ3 m
gy ¼ 0 ¼ 2
a 16 2
and solving for m gives m ¼ 0.73.
Substituting in Equation (22.4) we obtain
rffiffiffiffiffiffiffiffi
GM
o¼ 0:85
a3
23. A planet consists of a very thin spherical shell of mass M and radius a, within
which is a solid sphere of radius a/2 and mass M 0 centred at the midpoint of the
equatorial radius of the zero meridian. The planet rotates with angular velocity v
about an axis normal to the equatorial plane. Calculate:
(a) The potential at points on the surface as a function of latitude and longitude.
(b) The components of the gravity vector.
(c) If M 0 ¼ 10 M, what is the ratio between the tangential and radial components of
gravity at the North Pole?
(a) The potential U is the sum of the gravitational potentials due to the spherical shell V1, and
to the interior sphere V2, plus the potential due to the rotation of the planet F (Fig. 23):
U ¼ V1 þ V2 þ F
1
F ¼ o2 r2 cos2 ’
2
GM ð23:1Þ
V1 ¼
r
GM 0
V2 ¼
q
where r is the distance from a point P on the surface of the planet to its centre, q is the
distance from point P to the centre of the interior sphere, and ’ the latitude of point P.
Using the cosine law,
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a2
q ¼ r2 þ ar cos c
4
where c is the angle between r and the equatorial radius, and its inverse can be approxi-
mated by (Problem 16)
a2
1 1 a 2
¼ 1 þ cos c þ 2 3cos c 1 ð23:2Þ
q r 2r 8r
Using the relation for spherical triangles
cos a ¼ cos b cos c þ sin b sin c cos A
42 Gravity
90°
–j
P y
r q
y j
x
a a/2
Fig. 23
putting b ¼ 90º ’, c ¼90º, A ¼ l, and a ¼ c, where l is the longitude of P, then
cos c = cos ’ cos l
Substituting in (23.1), the potential due to the small sphere is
GM 0 a2
a 2 2
V2 ¼ 1 þ cos ’ cos l þ 2 3cos ’cos l 1
r 2r 8r
The total potential U is
GM GM 0 a2
a 2 2
U¼ þ 1 þ cos ’ cos l þ 2 3cos ’cos l 1
r r 2r 8r
ð23:3Þ
1
þ o2 r2 cos2 ’
2
(b) The components of the gravity vector are obtained from Equation (23.3):
@U
gr ¼
@r
3a2
GM 0 1 a 2 2
¼ 2 þ GM 2 3 cos ’ cos l 4 3cos ’cos l 1
r r r 8r
þ o2 rcos2 ’
1 @U 1 @U
gy ¼ ¼
r @y r @’
0
a2
GM a 2
¼ sin ’ cos l þ 6 cos ’cos l sin ’
r 2r2 8r3
þ o2 r cos ’ sin ’
43 Earth’s gravity field and potential
GM 0 a2
1 @U a 2
gl ¼ ¼ 2 cos ’ sin l 3 6cos ’ cos l sin l
r cos ’ @l r cos ’ 2r 8r
(c) At the North Pole, ’ = 90 and r ¼ a. Putting M 0 ¼10M and substituting in the
previous equations we obtain
GM GM 0 3GM 0 a2 GM
gr ¼ 2 þ ¼ 7:25 2
a2 a 8a4 a
GM 0 5GM
gy ¼ ¼ 2
2a2 a
The ratio between the radial and the tangential components of gravity at the North
Pole is
gr
¼ 1:45
gy
24. An Earth consists of a sphere of radius a and density r, within which there are two
spheres of radius a/2 centred on the axis of rotation and tangent to each other. The
density of that of the northern hemisphere is 4r and that of the southern hemisphere
is r/4.
(a) Express the gravitational potential in terms of M (the mass of the large sphere) up
to terms of 1/r3.
(b) What astronomical latitude corresponds to points on the equator (without
rotation)?
(c) What error is made by using the 1/r3 approximation in calculating the value of gr
at the equator?
(a) The total gravitational potential V is the sum of the potentials of the sphere of
radius a (V0) and of the two spheres of radius a/2 situated in the northern (V1) and
southern (V2) hemispheres (Fig. 24):
V ¼ V0 þ V1 þ V2
4r
x
r r1
a/2
a
a
r2
r/4 x
Fig. 24
GM
V0 ¼
r
a2
GM1 3GM 1 a
þ 2 cos y þ 3 3cos2 y 1
V1 ¼ ¼
r1 8 r 2r 8r
a2
GM2 3GM 1 a
2 cos y þ 3 3cos2 y 1
V2 ¼ ¼
r2 32 r 2r 8r
where r1 and r2 have been calculated as in Problem 16. Then, the total gravitational
potential in the 1/r3 approximation is
9 a2
41 15 a
3cos2 y 1
V ¼ GM þ 2
cos y þ 3
32r 64 r 256 r
(b) The components of the gravity vector, taking into account that there is no rotation, are
27a2
@V 41 15a 2
gr ¼ ¼ GM cos y 3cos y 1
@r 32r2 32r3 256r4
ð24:1Þ
27a2
1 @V 15a
gy ¼ ¼ GM sin y cos y sin y
r @y 64r3 128r3
GM
gr ¼ 1:175
a2
GM
gy ¼ 0:243 2
a
The astronomical latitude (’a) is the angle between the vertical and the equatorial plane. In
our case at the equator this is given by the deviation of the vertical from the radial
direction:
gy
tan ’a ¼ ¼ 0:207
gr
Then
’a ¼ 11:68 N
(c) If we want to calculate the exact value of gr at the equator, we calculate the exact
attractions of each sphere and add them:
GM 3GM 3GM
gr0 ¼ gr1 ¼ cos a gr2 ¼ cos a ð24:2Þ
a2 8r12 32r22
where r1 and r2 are the distances from the centre of each of the two interior
spheres (Fig. 24):
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffi
a2 a 5
r 1 ¼ r 2 ¼ a2 þ ¼
4 2
and a is the angle which r1 and r2 form with the equatorial plane:
a=2 1
sin a ¼ ¼ pffiffiffi
r1 5
The radial component of gravity is given by
GM
grT ¼ gr0 þ gr1 þ gr2 ¼ 1:335
a2
The error we make using the approximation is
GM
gapprox gexact ¼ 0:160 ; that is; 16%:
a2
25. An Earth consists of a sphere of radius a and density r within which there are
two spheres of radius a/2 centred on the axis of rotation and tangent to each other.
The density of that of the northern hemisphere is 2r and that of the southern
hemisphere is r/2.
(a) Express the potential V in terms of M (the mass of the large sphere), G, and r up to
terms in 1/r3.
46 Gravity
q1 q2
gq
a/2
r g
2r
x gr
r
q
j r
ja
a b
x
r/2
Fig. 25
(b) According to the value of this potential V, which astronomical latitudes corres-
pond to the geocentric latitudes 45º N and 45º S?
(c) What must the rotation period be for the astronomical and geocentric latitudes to
coincide?
(d) What error is made by the 1/r3 approximation in calculating the value of gr at the
equator? And at the North Pole?
(a) As in previous problems the effect of the interior spheres is given in terms of their
differential masses (Fig. 25):
4 a3 M
M1 ¼ pð2r rÞ ¼
3 2 8
4 r
a 3
M
M2 ¼ p r ¼
3 2 2 16
The distances q1 and q2 from the centre of each sphere to an arbitrary point P are found
using the cosine law:
a2 ar
q1 ¼ r 2 þ 2 cos y
4 2
2
a ar
q2 ¼ r2 þ þ 2 cos y
4 2
47 Earth’s gravity field and potential
Using the approximation for 1/q (Problem 16), the total gravitational potential V is the sum
of the potentials of the three spheres:
GM GM1 GM2
V ¼ þ þ
r q1 q2
GM 17 3 a 1 a2 2
¼ þ cos y þ 3cos y 1
r 16 32 r 128 r
3 a2
@V 1 17 6 a 2
gr ¼ ¼ GM 2 þ cos y þ 3cos y 1
@r r 16 32 r3 128 r4
ð25:1Þ
1 a2
1 @V GM 3 a
gy ¼ ¼ sin y þ 6 cos y sin y
r @y r 32 r2 128 r3
(c) If we want the astronomical and geocentric latitudes to coincide, then the
deviation of the vertical must be null, i ¼ 0º. This implies that gytotal must be
zero. To do this by means of the rotation, we have to make the tangential
component of the centrifugal force gyR be equal and of opposite sign to that of
the gravitational potential gyV :
gy total ¼ gy V þ gy R ¼ 0 ) gy V ¼ gy R
(d) The value of the radial component of gravity at the equator, r ¼ a, y = 90 , by
substitution in (25.1), is
GM
gr ¼ 1:04
a2
If we calculate the exact value by adding the contributions of the three spheres
(Fig. 25)
gr2 ¼ g2 cos b
where
49 Earth’s gravity field and potential
a 2
cos b ¼ rffiffiffi ¼ pffiffiffi
5 5
a
4
GM 4
g1 ¼
8 5a2
GM 4
g2 ¼
16 5a2
GM
grexact ¼ 0:96
a2
The error in the approximation is:
GM GM GM
grerror ¼ 0:96 2 1:04 2 ¼ 0:08 2
a a a
In a similar way, for a point at the North Pole, r ¼ a, y ¼ 0º:
gr ¼ 1:30 GM =a2
GM GM GM
grerror ¼ 1:47 2 1:30 2 ¼ 0:17 2
a a a
26. A spherical Earth of radius a has a core of radius a/2 whose centre is displaced a/2 along
the axis of rotation towards the North Pole. The core density is twice that of the mantle.
(a) What should the period of rotation of the Earth be for the direction of the plumb-
line to coincide with the radius at a latitude of 45º S?
(b) What are the values of J0, J1, J2, and m?
(a) As in previous problems we calculate the gravitational potential by the sum of the
potentials of the two spheres, using for the core the differential mass (Fig. 26):
GM GM 0
V ¼ V1 þ V2 ¼ þ
r q
2
4 a M
M 0 ¼ pð2r rÞ ¼
3 8 8
50 Gravity
a/2 P
2r q
x r
q j
45°
g
r
i
gr
gq
Fig. 26
1 a2
GM GM 1 a 2
V ¼ þ þ 2 cos y þ 3cos y 1 ð26:1Þ
r 8 r 2r 2 4r3
The total potential U is the sum of the gravitational potential V and the potential due to
rotation
1
F ¼ r2 o2 sin2 y
2
a2
91 a 1
2
þ r2 o2 sin2 y
U ¼ GM þ cos y þ 3cos y 1
8 r 16r2 64r3 2
In order that the direction of the plumb-line coincides with the radial direction, the
tangential component of gravity, gy, must be null:
a2 1
1 @U GM a 1
gy ¼ ¼ sin y 6 cos y sin y
r @y r 16 r2 64 r3
þ ro2 sin y cos y
For a point on the surface at latitude 45º S, the tangential component of gravity is, with
r = a, y = 135 ,
GM ao2
gy ¼ 0:003
a2 2
51 Earth’s gravity field and potential
Putting this component equal to zero, we find the value of the period of rotation T:
rffiffiffiffiffiffiffiffi
GM ao2 2 GM 2p a3
gy ¼ 0:003 2 ¼ 0 ) o ¼ 0:006 3 ) T ¼
a 2 a 0:077 GM
a2
GM 9 a 2
V ¼ 1þ cos y þ 3cos y 1
r 8 18r 72r2
0
M ¼ p ðr rÞ ¼ p ðr rÞ ¼ 1 ð27:1Þ
3 8 3 8 r 8 r
The potential U is
GM GM 0
a 2
2
U¼ þ 2þ 3cos y 1
r r 2r
The radial components of gravity at the equator and the Pole are found by taking the
derivative of the potential U:
2 3a2
@U GM
¼ 2 þ GM 0 2 4 3cos2 y 1
gr ¼
@r r r 4r
52 Gravity
GM GM 0 5
gre ¼ 2
a2 a 4
p GM GM 0 14
gr ¼ 2 2
a a 4
The gravity flattening is given by
gp ge 1
b¼ ¼
ge 8
By substituting the values of gravity we find the relation between M and M 0 :
7 0 5 0
1 M 2 M þ M þ 4 M 67
¼ ) M ¼ M0
8 5 0 4
M M
4
Putting M 0 in terms of M from Equation (27.1) we find the ratio of the densities:
M r0 r0
4
M0 ¼ M ¼ 1 ) ¼ 1:48
67 8 r r
(b) For a rotating planet we add to the potential U the rotational potential, F:
GM GM 0
a 2
2
GM r 3 m 2
U¼ þ 2þ 3cos y 1 þ sin y
r r 2r r a 2
The radial and tangential components of gravity are now
2 3a2
@U GM 0 2
gr ¼ ¼ 2 þ GM 2 4 3cos y 1
@r r r 4r
rm 2
þ GM 3 sin y
a
GM
¼ 1:11 2
r
1 @U GM 0
a 2
gy ¼ ¼ 2 6 cos y sin y
r @y r 2r
GM r 3
þ 2 m sin y cos y
r a
GM
¼ 0:013 2
r
From Fig. 27 we see that the relation between the geocentric and astronomical latitudes is
’a ¼ ’ i
where i is the deviation of the vertical with respect to the radial direction, which is given by
gy
tan i ¼ ¼ 0:012 ) i ¼ 0:7
gr
53 Gravity anomalies. Isostasy
a/2 gq
r’
g
x gr
r
j = 45°
ja
x
r⬘
Fig. 27
2 1
V ¼ 2pGra ln
r
where r is the distance measured perpendicular to the axis. Assume that a horizontal
cylinder is buried at depth d as measured from the surface to the cylinder’s axis.
(a) Calculate the anomaly along a line of zero elevation on the surface perpendicular
to the axis of the cylinder.
(b) At what point on this line is the anomaly greatest?
(c) What is the relationship between the distance at which the anomaly is half the
maximum and the depth at which the cylinder is buried?
(d) For a sphere of equivalent mass to produce the same anomaly, would it be at a
greater or lesser depth?
54 Gravity
x P
∆g
gr
gz
∆gmax
d
r
(1/2)∆gmax
a x
x1/2
x
r
Fig. 28
(a) The gravity anomaly produced by an infinite horizontal cylinder buried at depth d, with
centre at x ¼ 0 (Fig. 28), is given by the derivative in the vertical direction (z-axis) of
the gravitational potential V:
0 1
1
V ¼ 2pGra2 ln@qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiA
B C
x2 þ ðz þ dÞ2
@V 2prGa2 d
g ¼ gz ¼ ¼ ð28:1Þ
@z x2 þ ðz þ dÞ2
2prGa2 d
g ¼
x2 þ d 2
(b) To find the point at which the anomaly has its maximum value, we take the
derivative with respect to x and put it equal to zero:
@g
¼ 0 ) 2prGa2 d2x ¼ 0 ) x ¼ 0
@x
Substituting x ¼ 0 in (28.1):
2pGra2
gmax ¼
d
(c) The distance at which the anomaly has a value equal to half its maximum value
gives us the depth d at which the cylinder is buried:
55 Gravity anomalies. Isostasy
GM GM
V ¼ ¼ 1=2
r
x2 þ ðz þ dÞ2
(b) The Bouguer anomaly is calculated from the free-air anomaly using the Bouguer
correction with a crust density of 2.65 g cm3:
30. An anomalous mass is formed by two equal tangent spheres of radius R, with
centres at the same depth d ( d R ) and density contrast Dr.
(a) Calculate the Bouguer anomaly at the surface (z ¼ 0) produced by the mass
anomaly along a profile passing through the centres of the two spheres.
(b) Represent it graphically for x ¼ 0 (above the tangent point), 500, 1000, and 2000 m
taking R ¼ 1 km, d ¼ 3 km, and Dr ¼ 1 g cm3.
(a) For one sphere the anomaly for points on the surface (z ¼ 0) is (Problem 28)
GMd
g ¼
ð x2 þ d 2 Þ3=2
For two spheres the anomaly is the sum of the attractions of the two spheres (Fig. 30a):
GMd GMd
g ¼ þ
r13 r23
where
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r1 ¼ ðx RÞ2 þ d 2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r2 ¼ ðx þ RÞ2 þ d 2
Then
GMd GMd
g ¼ h i3=2 þ h i3=2
ð x RÞ2 þd 2 ð x þ RÞ2 þd 2
x P(x,0)
d r2
r1
R
x x
Fig. 30a
57 Gravity anomalies. Isostasy
(b) To represent graphically the curve of the anomaly (Fig. 30b), we first find the point
at which it is a maximum:
0 1
@g @ B GMd GMd
¼ @h i3=2 þ h i3=2 A ¼ 0
C
@x @x
ð x RÞ2 þd 2 ð x þ RÞ2 þd 2
h i5 h i5
ðx þ RÞ2 ðx RÞ2 þ d 2 ¼ ðx RÞ2 ðx þ RÞ2 þ d 2 ) x ¼ 0 maximum
Using the data given in the problem, we find the values of the anomaly for the five points,
with Dr = 1 g cm3, R = 1 km, d = 3 km
4 4
M ¼ pR3 r ¼ p 109 103 ¼ 4:19 1012 kg
3 3
x (m) Dg (gu)
0 53.0
500 52.0
1000 48.9
1500 43.9
2000 37.5
55
50
Anomaly (gµ)
45
40
Fig. 30b
58 Gravity
31. At a point at geocentric latitude 45º N and height 2000 m the observed value of
gravity is g ¼ 6690 000 gu. Taking the approximation that the Earth is an ellipsoid of
equatorial radius a ¼ 6000 km, density ¼ 4 g cm3, J2 ¼ 103, and m ¼ 103,
calculate for that point:
(a) The free-air and the Bouguer anomalies.
(b) The distance from the free surface to that of the sphere of radius a (precision 1 gu).
(a) The volume of an ellipsoid is:
4
V ¼ pa3 ð1 þ 2aÞ
3
The flattening is
3J2 m
a¼ þ ¼ 2 103
2 2
and the mass is
4
Me ¼ V r ¼ pa3 ð1 þ 2aÞr ¼ 3:624 1024 kg
3
Using G = 6.67 1011 m3 kg1 s2
GM 3
ge ¼ 2 1 þ J2 m ¼ 6:736 361 m s2
a 2
For a point at latitude 45º N the radial component of gravity is
gFA ¼ g g þ C FA
¼ 6 690 000 6 738 045 þ 2:244 2000 ¼ 43 557 gu
In order to calculate the Bouguer anomaly, we first calculate the Bouguer correction
N
h
Fig. 31
(b) If we call N the distance at the given point between the free surface and the surface
of the sphere of radius a (Fig. 31), this is given by:
N ¼arh
where r is the radius of the ellipsoid at latitude 45º N which to a first approximation is
2
3 1
r ¼ a 1 asin ’ ¼ 6000 1 2 10 ¼ 5994 km
2
Then,
N ¼ 6000 5994 2 ¼ 4 km
32. Beneath a point A at height 400 m there exists an anomalous spherical mass of
radius 200 m, density 3.5 g cm3, whose centre is 200 m below the reference level.
A point B is located at a height of 200 m and a horizontal distance of 400 m from A,
and a third point C is at a height of 0 m and at a horizontal distance of 800 m from
A. The density of the medium above the reference level is 2.6 g cm3, and below the
reference level it is 2.5 g cm3. The theoretical value of gravity is 980 000 mGal.
Calculate:
(a) The values of gravity at A, B, and C.
(b) The Bouguer anomalies at these points.
60 Gravity
Precision 1 gu.
(a) The gravity at each point is given by
g ¼ g C FA þ C B þ C am
where
Normal gravity: g ¼ 9800 000 gu
Free-air correction: CFA ¼ 3.086 h
Bouguer correction: CB ¼ 0.419 r1 h
r1 ¼ 2.6 g cm3 is the density of the material above the reference level
C am is the anomaly produced by the buried sphere at a point at height h and a horizontal
distance x from its centre:
4
GM ðh þ dÞ G pR3 ðrsph r2 Þðh þ dÞ
am
¼ 3
C 3=2 ¼ 3=2 ð32:1Þ
2
2
x þ ðh þ dÞ x2 þ ðh þ dÞ2
where d is the depth to the centre from the reference level; and rsph and r2 are the densities
of the sphere and of the medium where it is located, respectively. In our case: d ¼ 200 m,
rsph ¼ 3.5 g cm3, and r2 ¼ 2.5 g cm3.
For point A, x ¼ 0, we obtain
C FA ¼ 1234 gu C B ¼ 436 gu
4
GM G pR3 ðrsph r2 Þ
C am
¼ ¼ 3 ¼ 6 gu
ðh þ d Þ2 ðh þ d Þ2
The value of gravity is gA ¼ 9 799 208 gu.
At point B:
CFA ¼ 617 gu
CB ¼ 218 gu
The anomaly produced by the sphere is calculated by Equation (32.1), substituting
x ¼ xB ¼ 400 m and h ¼ hB ¼ 200 m
C am ¼ 5 gu
We obtain gB ¼ 9 799 606 gu.
At point C:
The free-air and Bouguer corrections are null, because the point is at the reference level.
The anomaly due to the sphere, by substitution in Equation (32.1), x ¼ xC ¼ 800 m, and
h ¼ hC ¼ 0, is
C am ¼ 1 gu
The value of gravity is: gC ¼ 9800 001 gu.
(b) The Bouguer anomaly is given by
DgB ¼ g þ CFA – CB – g
61 Gravity anomalies. Isostasy
By substitution of the values for each point we obtain that the anomalies correspond to
those produced by the sphere:
gAB ¼ 6 gu
gBB ¼ 5 gu
gCB ¼ 1 gu
33. For a series of points in a line and at zero height which are affected by the
gravitational attraction exerted by a buried sphere of density contrast 1.5 g cm3, the
anomaly versus horizontal distance curve has a maximum of 4.526 mGal and a point
of inflexion at 250 m from the maximum. Calculate:
(a) The depth, anomalous mass, and radius of the sphere.
(b) The horizontal distance to the centre of the sphere of the point at which the
anomaly is half the maximum.
(a) We know that the inflection point of the curve of the anomaly produced by a
sphere buried at depth d corresponds to the horizontal distance d/2. Then
d
xinf ¼ ) d ¼ 2xinf ¼ 2 250 ¼ 500 m
2
The maximum value of the anomaly at x ¼ 0 is
GM
gmax ¼ ¼ 4:526 mGal ¼ 45:26 gu
d2
and solving for DM
45:26 106 m s2 5002 m2
DM ¼ ¼ 1:6964 1011 kg
6:67 1011 m3 s2 kg1
From this value we calculate the radius of the sphere:
1=3
3M 1=3 3 1:6964 1011 kg
4
M ¼ pR3 r ) R ¼ ¼ ¼ 300 m
3 4pr 4 3:14 1:5 103 kg m3
34. At a point at height 2000 m, the measured value of gravity is 9.794 815 m s2.
The reference value at sea level is 9.8 m s2. The crust is 10 km thick and of density
2 g cm3, and the mantle density is 3 g cm3. Calculate:
(a) The free-air, Bouguer, and isostatic anomalies. Use the Pratt hypothesis with a
cylinder of radius 10 km and a 40 km depth of compensation.
(b) If beneath this point there is a spherical anomalous mass of GDM ¼ 160 m3 s2 at
a 2000 m depth, what should the compensatory cylinder’s density be for the
compensation to be total?
62 Gravity
C a
Fig. 34a
To calculate the isostatic anomaly (Fig. 34a) we begin with the calculation of the isostatic
correction assuming Pratt’s hypothesis and using only a vertical cylinder of radius 10 km
under the point and the compensation level at 40 km. In this way, the correction consists
of the gravitational attraction of a cylinder of radius a and height b at a point at distance c
from the base of the cylinder, which is given by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ 2pGr b þ a2 þ ðc bÞ2 a2 þ c2
I
ð34:1Þ
d
rc = 2 g cm–3
10 km
D = 40 km
rM = 3 g cm–3
Fig. 34b
(b) If under the point considered there is an anomalous spherical mass (Fig. 34b) at
depth d ¼ 2 km, the anomaly it produces is
GM 160
g am ¼ ¼ ¼ 40 gu
d2 ð2000Þ2
The total anomaly now is the Bouguer anomaly plus the anomaly due to the sphere:
g ¼ 689 40 ¼ 729 gu
If the isostatic compensation is total (isostatic anomaly equal to zero), this anomaly must
be compensated by the cylinder. Thus, the necessary contrast of densities Dr to do this can
be calculated using expression (34.1):
so
r ¼ 0:25 g cm3
As the mean value (crust–mantle) of the density is 2.75 g cm3, the density of the cylinder
must now be
g FA ¼ g g þ C FA ¼ g g þ 3:086h ¼ 86 gu
(b) To calculate the isostatic anomaly according to the Airy hypothesis we first need to
obtain the value of the root given by the equation
rc
t¼ h ¼ 4500 m
rM rc
gI ¼ g g þ C FA þ C B þ C I ¼ g B þ C I ¼ 637 gu
The negative value of the anomaly indicates that the zone is overcompensated.
65 Gravity anomalies. Isostasy
(c) If we want the compensation to be total, the value of the isostatic correction
must be
DgI ¼ DgB þ CI ¼ 0 ¼> CI ¼ – DgB ¼ 1046 gu
Since the isostatic correction under the Airy hypothesis is
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ 2pGr t þ a2 þ ðh þ H Þ2 a2 þ ðh þ H þ tÞ2
I
36. Gravity measurements are made at two points A and B of altitude 1000 m and
1000 m above the reference level, respectively, 2 km apart along a W-E profile at
latitude 38.80º N. Below a point C located in the direction AB and 1 km from A
is buried a sphere of radius 1 km and centre 3 km below the reference level, of density
r ¼ 1.76 g cm3. Calculate:
(a) The value of gravity at A and B.
(b) Using the Airy assumption and neglecting the sphere, calculate the root at A and B.
Crustal density rC ¼ 2.76 g cm3, mantle density rM ¼ 3.72 g cm3, a ¼ 10 km, and
H ¼ 30 km.
(a) The gravity observed at points A and B is given by
gA ¼ g C FA þ C B þ C am
gB ¼ g þ C FA C B þ C am
where g is the theoretical gravity, CFA the free-air correction, CB the Bouguer correction,
and C am the attraction due to the anomalous mass.
The theoretical gravity at the observation point at latitude 38.80º N is
The attraction due to the spherical anomalous mass (Fig. 36) is given by
GMðz þ dÞ
C am ¼ 3=2
x2 þ ðz þ dÞ2
66 Gravity
A XA
h
XB
C
–h
B
d
r c = 2.76 g / cm3
x R
ρ
Fig. 36
we find
CAam ¼ 16 gu
CBam ¼ 50 gu
(b) To calculate the value of the root under A and B according to the Airy hypothesis we
use the equation
rC
t¼ h
rM rC
67 Gravity anomalies. Isostasy
where rC and rM are the crust and mantle densities. Then we find
2:76
tA ¼ 1000 ¼ 2875 m
3:72 2:76
2:76
tB ¼ ð1000Þ ¼ 2875 m
3:72 2:76
37. At a point at latitude 43º N, the observed value of gravity is 9800 317 gu, and the
free-air anomaly is 1000 gu.
(a) Calculate the Bouguer anomaly. Take rC ¼ 2.67 g cm3.
(b) If the isostatic compensation is due to a cylinder of radius 10 km which is beneath
the point of measurement, what percentage of the Bouguer anomaly is compen-
sated by the classical models of Airy and Pratt?
(c) According to the Pratt hypothesis, what density should the cylinder have for the
compensation to be total?
(a) First we calculate the normal gravity at latitude 43º N:
g ¼ 9804 385 gu
(b) To apply the isostatic compensation using the Airy hypothesis we first calculate the
root corresponding to the height h ¼ 1642 m:
t ¼ 4:45h ¼ 7307 m
where
a ¼ 10 km; b ¼ t ¼ 7307 m; c ¼ t þ 30 000 þ h ¼ 38 949 m
C I ¼ 148 gu
We have to determine again the Bouguer anomaly using the density according to the Pratt
hypothesis
Dr0
r¼ ¼ 2:63 g cm3
Dþh
gB ¼ g AL 2pGrh ¼ 810 gu
C I ¼ g B
Using the Pratt hypothesis in order to calculate the density r of the cylinder under the
point, we have to take into account that this density must also be the density used in the
determination of the Bouguer anomaly. Then we write
C I ¼ gB ¼ gFA þ 2p G r h
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
and putting
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
N¼ b þ a2 þ ð c bÞ 2 a2 þ c 2
gFA þ 2pGr0 N
r¼ ¼ 2:46 g cm3
2pGðh þ N Þ
where we have used the values r0 ¼ 2.67 g cm3 and DgFA ¼ 1000 gu.
38. At a point on the Earth’s surface, a measurement of gravity gave a value of
9795 462 gu. The point is 2000 m above sea level. At sea level the crust is 20 km thick
and of density rC ¼ 2 g cm3. The density of the mantle is rM ¼ 4 g cm3.
(a) Calculate the free-air and Bouguer anomalies.
69 Gravity anomalies. Isostasy
(b) Calculate the isostatic anomaly according to the Airy and Pratt assumptions.
Use cylinders of 10 km radius and compensation depth of 60 km.
(c) Beneath the point, there is an anomalous spherical mass of GDM ¼ 1200 m3 s2.
How deep is it?
Take g ¼ 9.8 m s2.
(a) The free-air anomaly is given by
g FA ¼ g g þ 3:086h ¼ 9 795 462 9 800 000 þ 3:086h ¼ 1634 gu
For the Bouguer anomaly we first calculate the Bouguer correction
(b) To calculate the isostatic anomaly according to the Airy hypothesis we determine
first the value of the root corresponding to the height 2000 m:
rc 2
t¼ h¼ 2000 ¼ 2000 m
rM rc 42
The isostatic correction, using a single cylinder under the point, is given by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C a
Fig. 38a
70 Gravity
a ¼ 10 km; b ¼ t ¼ 2 km; c ¼ H þ t þ h ¼ 20 þ 2 þ 2 ¼ 24 km
Calling A the term inside the brackets in Equation (38.1)
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
A ¼ 2 þ 100 þ 484 100 þ 576 ¼ 0:166 km
The isostatic correction is, then, given by
g I ¼ g g þ C FA C B þ C I ¼ 97 gu
According to the Pratt hypothesis, the regional density is given by
D 60
r¼ r0 ¼ 3:33 ¼ 3:22 g cm3
Dþh 60 þ 2
where D is the compensation depth (in this problem 60 km) and for r0 (Fig. 38b) we have
used the mean value of the density of the crust (2 g cm3) and of the mantle (4 g cm3)
along the compensation depth
1 2
r0 ¼ 2 þ 4 ¼ 3:33 g cm3
3 3
The contrast of densities is
P
x
h
20 km
rC = 2 g cm–3
40 km
rM = 4 g cm–3
Fig. 38b
71 Gravity anomalies. Isostasy
C I ¼ 2 3:1416 6:67 1011 m3 =kg1 s2 0:11 103 kg=m3 7:4 103 m
¼ 341 gu
Since according to the Pratt hypothesis, the density of the compensating cylinder extends
to the surface of the height 2000 m, we have to calculate again the Bouguer anomaly using
this density (3.33 g cm3). We find for the Bouguer and isostatic anomalies the values
CB¼ 2 p G r h ¼ 2699 gu
DgB ¼ DgFA – CB ¼ 1065 gu
DgI ¼ DgB þ CI ¼ 724 gu
(c) If we assume that the isostatic anomaly is produced by a spherical anomalous mass
buried under the point at a depth d under sea level its gravitational effect is given by
GM
gmax ¼ ¼ gI ¼ 724 gu
ðh þ dÞ2
gP ¼ g C FA þ C B þ C am ð39:1Þ
where,
g ¼ 9800 000 gu
C FA ¼ 3:086 h ¼ 3:086 2000 ¼ 6172 gu
C B ¼ 0:419rh ¼ 0:419 3 2000 ¼ 2514 gu
(b) If there is isostatic compensation, according to the Airy hypothesis, we determine first
the depth of the root, using the density of the crust rC and of the mantle rM:
rC 3
t¼ h¼ 2000 ¼ 6000 m
rM rC 43
C ¼ 2pGr b þ a2 þ ðc bÞ2 a2 þ c2
I
ð39:3Þ
C I ¼ 2 3:14 6:67 1011 1103 6 þ 202 þ ð28 6Þ2 202 þ 282 103
¼ 553 gu
¼ 501 gu
73 Gravity anomalies. Isostasy
g I ¼ 0 ¼ g þ gP þ C FA þ C B þ C I þ C am ð40:1Þ
where g is the normal gravity, gP the observed gravity at point P, CFA the free-air correction,
CB the Bouguer correction, CI the isostatic correction and C am the gravitational effect of
the anomalous mass.
The free-air and Bouguer corrections are given by
P
x
d
a
rC
rsph
H
rM
t
Fig. 40
GM 650
C am ¼ 2
¼ 2
¼ 41 gu
ðh þ d Þ ð4Þ 106
We calculate the isostatic correction using the Airy hypothesis and taking into account the
presence of the spherical anomalous mass. Thus, according to Fig. 40, the equilibrium
between the gravity at P and at sea level far from P is given by
4
pa2 rC H þ pa2 rM t ¼ pa2 rC ðh þ H þ t Þ þ pR3 rsph rC
3
and solving for t:
4
a2 hrC þ R3 rsph rC
t¼ 3 ¼ 3011 m
a2 ðrM rC Þ
As in previous problems we calculate the isostatic correction using a cylinder under point P
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C I ¼ 2pGr b þ a2 þ ðc bÞ2 a2 þ c2
75 Gravity anomalies. Isostasy
g I ¼ 0 ¼ g þ gP þ C FA þ C B þ C I þ C am
Solving for CB we obtain
C B ¼ 975 gu
From the definition of the Bouguer anomaly we can find the value gP of gravity at P:
gB ¼ g P g þ C FA C B þ C am
145 ¼ g P 9800 000 þ 3086 975 þ 41 ) g P ¼ 9797 703 gu
(c) Since the density of the sphere is greater than the density of the crust, there is an excess
of gravity at P with respect to other points at sea level far from P, which must be
compensated by a root of crustal material inside the mantle with negative gravitational
influence. In this situation the root can never be null.
41. At 10 km beneath sea level vertically under a point P of height 2000 m there exists
an anomalous spherical mass GDM ¼ 104 m3 s2. At sea level, gravity is 9800 000 gu
and the crustal thickness 20 km. The density of the crust is 2 g cm3, and of the
mantle 4 g cm3. Using the Airy assumption for the isostatic compensation with a
cylinder of 10 km radius, calculate for that point:
(a) The observed gravity.
(b) The free-air, Bouguer, and isostatic anomalies.
(a) For point P the Bouguer correction is
The gravity at point P, if there is no isostatic compensation and other effects, can be
obtained from the normal gravity and the free-air and Bouguer corrections
Since there is an anomalous mass under point P we have to add its gravitational contribu-
tion to the gravity at P. For a spherical mass at depth h þ d under P the gravitational
attraction is
76 Gravity
We calculate the root corresponding to the isostatic compensation, assuming the Airy
hypothesis, and taking into account the presence of the anomalous mass in the same way as
in Problem 40:
4
pa2 rC H þ pa2 rM t ¼ pa2 rC ðh þ H þ t Þ þ pR3 rsph rC
3
¼ pa2 rC ðh þ H þ t Þ þ Ma
pa2 rC h þ Ma
t¼ ¼ 2239 m
pa2 ðrM rC Þ
The isostatic correction using a cylinder is given by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C I ¼ 154 gu
The gravity at point P is the value obtained in (41.1) plus the contribution of the anomalous
mass and minus the isostatic correction:
(b) The free-air anomaly is equal to this observed value minus the normal gravity plus
the free-air correction:
gFA ¼ gP g þ C AL
Substituting the values we obtain
g B ¼ gP g þ C AL C B ¼ 54 gu
Finally the isostatic anomaly is the Bouguer anomaly plus the isostatic correction:
For the compensation, assume cylinders of the same radius as the sphere. Normal
gravity g ¼ 9.8 m s2.
We first calculate the root t, assuming the Airy hypothesis, corresponding to the height
2000 m of point P. If the situation is of total isostatic equilibrium, we have to introduce the
effect produced by the sphere in the determination of the root (Fig. 40):
4
p rsph rC a3 þ pa2 rC h þ pa2 rC H þ pa2 rC t ¼ pa2 HrC þ pa2 rM t
3
so
4
rC h þ a rsph rC
t¼ 3 ¼ 10 000 m ð42:1Þ
r M rC
The negative value of t (anti-root) is due to the deficit of mass produced by the presence of
the sphere (rsph < rC) under point P.
The isostatic correction, as in previous problems, is calculated taking a cylinder under the point:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C I ¼ 2pGðrM rC Þ b þ a2 þ ðc bÞ2 a2 þ c2
gI ¼ g P g þ C FA C B C I C am ð42:2Þ
The effect of the anomalous mass is given by
4
G pa3 rsph rC
C am ¼ 3 ¼ 267 gu
ðh þ d Þ2
By substitution in (42.2)
g I ¼ 9 795 717 9 800 000 þ 3:086 2000 0:419 2:5 2000 þ 267 36
¼ 25 gu
43. A point A on the Earth’s surface is at an altitude of 2100 m above sea level.
Calculate:
(a) The value of gravity at A if the isostatic anomaly is 2.5 mGal. Assume the Airy
hypothesis (rC ¼ 2.6 g cm3, rM ¼ 3.3 g cm3, H ¼ 30 km).
(b) If the previous value had been measured with a Worden gravimeter of constant
0.301 82 mGal/division giving a reading of 630.6, calculate the value of gravity at
another point B at which the device reads 510.1 (both readings corrected for drift).
(c) At what depth is the centre of a sphere of density 4 g cm3 and radius 5 km which
is buried in the crust, given that the anomaly created at a point A, 12 km from the
78 Gravity
centre of the sphere, not in the same vertical, is 321 gu. Also calculate the
horizontal distance from the centre to point A.
Take, for compensation, cylinders of 10 km radius. g ¼ 9.8 m s2
(a) The isostatic anomaly is given by
To calculate the isostatic correction CI, assuming the Airy hypothesis, we must first
calculate the root t that corresponds to the height h
rC h
t¼ ¼ 7800 m
rM rC
As in other problems the isostatic correction is calculated using
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C I ¼ 2pGðrM rC Þ b þ a2 þ ðc bÞ2 a2 þ c2
b ¼ t ¼ 7800 m; c ¼ h þ H þ t ¼ 39900 m;
a ¼ 10 km; r ¼ rM rC ¼ 700 kg m3
we obtain
C I ¼ 84 gu
Solving for gA in Equation (43.1) we obtain
(b) For a Worden gravimeter the increment in gravity between two points (Dg) is propor-
tional to the increment in the values given by the instrument (DL) corrected by the
instrumental variations
g ¼ KL
gB gA ¼ KðLB LA Þ ) gB ¼ gA þ KðLB LA Þ
¼ gA 364 gu ¼ 9795 384 gu
4
G pa3 rsph rC ðh þ dÞ
C am ¼ 3 3=2 ð43:2Þ
x2 þ ðh þ dÞ2
79 Gravity anomalies. Isostasy
d
rC
x x
rsph
Fig. 43
For point A (Fig. 43) if r is the distance from the centre of the sphere to the point A,
x2 þ (h þd)2 ¼ r2 and solving for d in Equation (43.2) gives
C am r3
d¼ h
4
G pa3 rsph rC
3
44. In a gravity survey, two points A and B on the Earth’s surface gave the values 159
and 80 mGal for the free-air anomaly, and 51 and 25 mGal for the Bouguer
anomaly, respectively. Given that B is at an altitude 1000 m lower than A, and that the
density of the mantle is 25% greater than that of the crust, calculate:
(a) The value of gravity at A and B, and the densities of the crust and mantle.
(b) The isostatic anomaly according to the hypotheses of Airy (H ¼ 30 km) and Pratt
(D ¼ 100 km, r0 the value determined in the previous part) at point A. Take, for
compensation, cylinders of 10 km radius. g ¼ 980 Gal.
(a) The free-air anomaly at point A is given by
Changing values from mGal to gu, we write for the Bouguer anomalies at points A and B
(b) For the isostatic anomaly at point A, according to the Airy hypothesis, we first
calculate the value of the root t corresponding to its height:
rC h A hA
t¼ ¼ ¼ 8000 m
1:25rC rC 0:25
C ¼ 2pGðrM rC Þ b þ a2 þ ðc bÞ2 a2 þ c2
I
b ¼ t ¼ 8000 m; c ¼ h þ H þ t ¼ 40 000 m;
a ¼ 10 km; r ¼ rM rC ¼ 626 kg m3
we obtain
C I ¼ 77 gu
81 Gravity anomalies. Isostasy
If we use the Pratt hypothesis, we first calculate the density corresponding to the material
under point A:
Dr0 100 2:505
r¼ ¼ ¼ 2:456 g cm3
Dþh 100 þ 2
and the contrast of density
C I ¼ 2pGðrM rC Þ b þ a2 þ ðc bÞ2 a2 þ c2
C I ¼ 158 gu
The isostatic anomaly will be the Bouguer anomaly plus the isostatic correction
In both cases the anomaly is negative, but using the Airy model the value is greater than
using the Pratt model.
45. At a point P on the Earth’s surface, the observed value of gravity is 9.795
636 m s2, and the Bouguer anomaly is 26 mGal. Assuming the Airy hypothesis
(rC ¼ 2.7 g cm3, rM ¼ 3.3 g cm3, H ¼ 30 km), calculate:
(a) The height of the point.
(b) The isostatic anomaly.
(c) The value of gravity that would be observed at the point if beneath it were a
sphere at a depth of 10 km below sea level, with a density of 2.5 g cm3 and a
radius of 5 km, such that the compensation was total.
Compensation with cylinders of 5 km radius; g ¼ 9.8 m s2.
(a) We calculate the height of point P from the Bouguer anomaly:
gB gP þ g
gB ¼ g P g þ 3:086h 0:419rC h ) h ¼
3:086 0:419rC
so
260 9 795 636 þ 9 800 000
h¼ ¼ 2099:8 m ffi 2100 m
3:086 0:419 2:7
82 Gravity
(b) To calculate the isostatic anomaly, using the Airy hypothesis, we first calculate the
value of the root t corresponding to the height of the point:
rC 2:7
t¼ h¼ 2100 ¼ 9450 m
rM rC 3:3 2:7
a ¼ 5 km; b ¼ t ¼ 9450 m;
c ¼ h þ H þ t ¼ 2100 þ 30 000 þ 9450 ¼ 41 550 m;
r ¼ 3:3 2:7 ¼ 0:6 g cm3
and obtain
C I ¼ 22 gu
The isostatic anomaly is then
(c) If the compensation is total then the isostatic anomaly must be zero. But now we have
to include the gravitational effect Cam produced by the presence of the anomalous mass
of the sphere.
g I ¼ 0 ¼ gP g þ 3:086h 0:419rC h C am þ C I
x a
H rsph
rC
t rM
Fig. 45
We substitute this value of t in Equation (45.1) together with the other values
a ¼ 5 km; b ¼ 7 ¼ 7228 m;
c ¼ h þ H þ t ¼ 39 328 m; r ¼ 3:3 2:7 ¼ 0:5 kg m3
and obtain
C I ¼ 18 gu
By substitution in (45.2) we find the value of the gravity at P under the given conditions:
g FA ¼ g P g þ 3:086h
ð46:1Þ
g B ¼ gP g þ 3:086h 0:419rC h
gFA g B
h¼ ¼ 2193 m
0:419rC
The observed gravity at P can be obtained from either of the two equations (46.1):
(c) The isostatic correction is found from the known Bouguer and isostatic anomalies:
The isostatic correction, using the Airy hypothesis, is given, as in previous problems, as a
function of the root t, by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
where we substitute
47. In an oceanic region, gravity is measured at a point on the surface of the sea,
obtaining a value of 979.7950 Gal. Calculate, using the Airy hypothesis (H ¼ 30 km,
rC ¼ 2.9 g cm3, rM ¼ 3.2 g cm3, rW ¼ 1.04 g cm3):
(a) The isostatic anomaly if the thickness of the crust is 8.4 km.
85 Gravity anomalies. Isostasy
(b) The thickness that the water layer would have to have if 15 km vertically below
the point there was centred a sphere of 10 km radius such that the anti-root is null
and the compensation total. Also calculate the density of the sphere.
For the compensation, take cylinders of 10 km radius. g ¼ 9.8 m s2.
(a) We are in an oceanic region, therefore in the calculation of the root for the isostatic
compensation according to the Airy hypothesis we have to consider the layer of water
of density rW. The value of the root is now given by
rC rW 0 2:9 1:04 0
t0 ¼ h ¼ h ¼ 6:2h0 ð47:1Þ
rM r C 3:2 2:9
From the value of the root we calculate the isostatic correction using a cylinder of 10 km
radius,
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ 2pGr b þ a2 þ ðc bÞ2 a2 þ c2
I
C I ¼ 269 gu
To calculate the isostatic anomaly we first have to apply the Bouguer correction which in
this case consists of two terms: the first to eliminate the attraction of the water layer
h⬘ rW
rC H
e
t⬘
rM
Fig. 47a
86 Gravity
(2pGrWh0 ) and the second to replace this layer by one of density equal to the crustal
density (þ2pGrCh0 ). Since the point is at sea level the free-air correction is null:
DgI ¼ g g þ C B C I
Dg I ¼ 19 gu
(b) If the anti-root is null and there is total compensation, then we have
t0 ¼ 0
ð47:2Þ
g I ¼ 0 ¼ g g 2pGrW h0 þ 2pGrC h0 C am
Since the isostatic anomaly must be null, then the anomalous spherical mass and the water
layer must compensate each other. The attraction of the anomalous mass is Cam ¼ GDM/d2
where d is the depth of its centre below sea level. Then we can write
GM
g g þ 2pGðrC rW Þh0 ¼0
d2
where the mass of the sphere is
4
M ¼ pa3 rsph
3
If the point P is totally isostatically compensated and the anti-root is null, then (Fig. 47b)
4
pa2 h0 rW þ pa2 ðH h0 ÞrC þ pa3 ðrsph rC Þ ¼ pa2 HrC
3
Solving for h0 gives
4a rC rsph
h0 ¼ ð47:3Þ
3ðrW rC Þ
h⬘ rW
d
H
rC
x a
e
rsph
Fig. 47b
87 Gravity anomalies. Isostasy
g I ¼ gP g 3:086h0 þ 0:419rC h0 C I
The free-air correction (3.086h0 ) is negative because the point is below sea level. For the
same reason the Bouguer correction (0.419 rC h0 ) is positive. The isostatic correction,
using the Airy hypothesis, is calculated as in previous problems using a cylinder under the
point of 5 km radius and density contrast D r ¼ 3.2 2.7 ¼ 0.5 g cm3. The value of the
anti-root t 0 has already been calculated, so
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ 2pGr b þ a2 þ ðc bÞ2 a2 þ c2
I
h⬘ P
rC H
e
t⬘
rM
Fig. 48
88 Gravity
C I ¼ 9 gu
Then the isostatic anomaly is DgI ¼ 5 gu.
49. In an oceanic region where the density of the crust is 2.90 g cm3 and that of the
mantle 3.27 g cm3, the value of gravity measured at a point P on the sea floor at
depth 4000 m is 9.806 341 m s2.
Calculate, according to the Airy hypothesis:
(a) The thickness of the crust.
(b) The isostatic anomaly in gravimetric units.
Data: rw ¼ 1.04 g cm3, H ¼ 30 km, g ¼ 9.8 m s2. Take, for compensation, cylinders
of 10 km radius.
(a) First we calculate the value of the root according to the Airy hypothesis
rC rW 0
t0 ¼ h ¼ 20 108 m
rM rC
The thickness of the crust under the point is found by (Fig. 49)
(b) Because the point is located at the bottom of the sea, to reduce the observed value of
gravity to the surface of the geoid (sea level) we eliminate first the attraction of the
water layer. Then we apply the free-air and the Bouguer corrections, to take into
account the attraction of a layer of crustal material which replaces the water. Finally we
apply the isostatic correction:
C I ¼ 2pGr b þ a2 þ ðc bÞ2 a2 þ c2
h⬘ rW P
rC
e H
t⬘
rM
Fig. 49
89 Gravity anomalies. Isostasy
C I ¼ 598 gu
The isostatic anomaly is:
gI ¼ 9 806 341 9 800 000 3:086 4000 þ 0:419 ð1:04 þ 2:9Þ 4000 598
¼ 4 gu
50. At a point with coordinates 42.78º N, 0.5º E and height 1572 m, the observed value
of gravity is 980.0317 Gal.
(a) Calculate the free-air and Bouguer anomalies.
(b) If cylinders of 10 km radius beneath that point are used for the isostatic compen-
sation, calculate the gravimetric attraction of the mass defect corresponding to
the altitude of the point according to the Airy and Pratt hypotheses. Take, for
the crust, H ¼ 30 km, rC ¼ 2.67 g cm3, for the mantle, rM ¼ 3.27 g cm3, and
D ¼ 100 km for the Pratt level of compensation.
(c) How deep should the root of the Airy model be for the compensation to be
total?
(a) We calculate first the normal gravity at the point where gravity has been observed
using the expression
where Dr is the density contrast, b the height of the cylinder, and c the distance from the
base of the cylinder to the observation point.
Airy: We calculate first the root given by the equation
t ¼ 4:45 h ¼ 4:45 1572 ¼ 6995 km
For the isostatic correction we substitute in (50.1) the values
90 Gravity
and obtain
so
C I ¼ 68 gu
Pratt: The contrast of densities is now given by
h 1575
r ¼ r ¼ 2:67 ¼ 0:04 g cm3
D þ h 0 100 000 þ 1572
and substituting in Equation (50.1) with the values
a ¼ 10 km, b ¼ 100 km, c ¼ D + h ¼ 101 572 m, we have
C I ¼ 135 gu
(c) If the isostatic compensation is total (isostatic anomaly null) the isostatic correc-
tion, according to the Airy hypothesis, coincides with the value of the Bouguer
anomaly (834 gu):
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
so
gB
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
tþ a2 ðH þ hÞ2 a2 þ ðt þ H þ hÞ2 ¼
2pGr
gB
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a2 þ ðt þ H þ hÞ2 ¼ t þ a2 ðH þ hÞ2 ¼tþN
2pGr
a2 þ ðt þ H þ hÞ2 ¼ t 2 þ N 2 þ 2tN
N 2 a2 ðH þ hÞ2
t¼
2ðH þ h N Þ
91 Gravity anomalies. Isostasy
t ¼ 58 875 m
Because this root has a negative value greater than the thickness of the crust, total
compensation is not possible.
51. Calculate the free-air anomaly observed on a mountain of height 2000 m which
is fully compensated by a root of depth t ¼ 10 km. The compensation is by a
cylinder of radius 20 km, the density of the crust is 2.67g cm3, and that of the
mantle is 3.27g cm3.
The free-air anomaly is given by
gFA ¼ g g þ 3:086h
Since the point is isostatically compensated, we calculate the isostatic correction using a
cylinder as in previous problems:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C I ¼ 2pGr b þ a2 þ ðc bÞ2 a2 þ c2
where we substitute
b ¼ t ¼ 10 km, c ¼ h þH þ t ¼ 42 km, a ¼ 20 km, Dr ¼ 600 kg m3
and obtain
CI ¼ 306 gu
Since the point is totally compensated the isostatic anomaly must be zero:
gI ¼ g g þ C FA C B þ C I ¼ 0
The free-air anomaly can now be written as
gFA ¼ g g þ C FA ¼ C B C I ð51:1Þ
We can calculate the Bouguer correction:
gB ¼ g g þ C FA C B þ T
92 Gravity
a2
P
a1 Px
a2 a1
h hm
rC
Fig. 52
C FA ¼ 3:086h ¼ 6481 gu
and the Bouguer correction by
C B ¼ 0:419rC h ¼ 2200 gu
The topographic correction is introduced in order to correct for the topographic masses not
included in the Bouguer correction, that is, in this case those above the height h (Fig. 52).
Remember that the Bouguer correction corrects for an infinite layer or plate of thickness h
and doesn’t consider the additional masses above h or the lack of masses below h. The
topographic correction is always positive because the masses above height h produce on
point P an attraction of negative sign which must be added and the lack of mass under h
must also be taken into account with a positive sign, since it has been subtracted in the
Bouguer correction.
To calculate the attraction of the mass above height h we use the attraction of
concentric cylinders (in our case two) with axis passing through point P and with
height equal to the difference between the height h of the point P and the height of
the mass of the topography hm above h. The cylinder is divided into n sectors with
radius a1 and a2 to approximate the topography (Fig. 52). Then the topographic
correction is given by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2pGrC
T¼ a22 þ ðc bÞ2 a22 þ c2 a21 þ ðc bÞ2 þ a21 þ c2
n
In our case we substitute the values
and obtain
T ¼ 2 gu
The refined Bouguer anomaly is
gB ¼ g g þ C FA C B þ T ¼ 298 gu
53. Calculate the topographic correction for a terrestrial compartment of inner radius
a1 ¼ 5240 m, outer radius a2 ¼ 8440 m, n ¼ 20, mean height 120 m, with 2000 m being
the height of the point P. Take r ¼ 2.65 g cm3.
In this problem we consider the topographic correction for the case of the lack of mass in
the topography at heights below that of the point P. Since in the Bouguer correction we
have subtracted an infinite layer of thickness h, we have to correct for the places where the
mass was not present (Fig. 53).
The topographic or terrain correction T in this case is calculated in the same way as in
the previous problem. Thus we take n sectors of cylinders with axis at point P and height
equal to the difference between h and hm (Fig. 52b). The correction is then given by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2pGrC
T¼ a22 þ ðc bÞ2 a22 þ c2 a21 þ ðc bÞ2 þ a21 þ c2
n
where we substitute a2 ¼ 8440 m, a1 ¼ 5240 m, b ¼ c ¼ h – hm, n ¼ 20 to obtain
T ¼ 0:67 mGal
54. Calculate the topographic correction for an oceanic sector or compartment
of inner radius a1 ¼ 5240 m, outer radius a2 ¼ 8440 m, n ¼ 20, mean depth
525 m, with 600 m being the height of the point P. Take rC ¼ 2.67 g cm3, rW ¼
1.03 g cm3.
In this problem we have to correct for the lack of mass in the oceanic area near the point P,
between the sea level and height h (column 1 in Fig. 54). Also we have to take into account
the attraction produced by the water layer between sea level and the bottom of the sea
(column 2 in Fig. 54).
a2
a1
P
hm rC
Fig. 53
94 Gravity
a2
a1 P
h 1 h
P 2
rw
Fig. 54
2pGr
T¼ a22 þ ðc bÞ2 a22 þ c2 a21 þ ðc bÞ2 þ a21 þ c2
n
where
a2 ¼ 8440 m; a1 ¼ 5240 m; n ¼ 20
For the correction corresponding to the attraction of column 1, between height h and sea
level, we substitute the values:
b ¼ h ¼ c; r ¼ rC
and obtain
T1 ¼ 0:07 mGal
For the correction of the attraction of column 2 between sea level and the bottom of the sea
we use the difference between the densities of the crust and of water:
b ¼ p ¼ 525 m
c ¼ p þ hs ¼ 1125 m
c b ¼ hs
r ¼ rC rW
T2 ¼ 0:11 mGal
Tides
55. Two spherical planets A and B of radii 2a and a and masses 3m and m are
separated by a distance (centre to centre) of 6a. The only forces acting are gravita-
tional, and the system formed by the two planets rotates in the equatorial plane.
(a) Calculate the value of the components of the acceleration of the tides at the Pole of
each planet directly and using the tidal potential. On which planet are they greater?
(b) If each planet spins on its axis with the same angular velocity as the system, what,
for each planet, is the ratio between the centrifugal force and the maximum of the
tidal force at the equator? On which planet is this ratio greater?
(a) From Fig. 55a we can deduce that at the Pole of planet A, the radial component of
the acceleration of the tides produced by planet B is
Gm
grT ¼ cos a
q2
where q is the distance from the Pole of planet A to the centre of planet B and a the angle
formed by q and the radius at the Pole of planet A.
By substitution of the required values we obtain
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffi
q ¼ ð6aÞ2 þ ð2aÞ2 ¼ 40a
2a
cos a ¼ pffiffiffiffiffi ) a ¼ 71:6
40a
Gm Gm
grT ¼ 2
cosð71:6Þ ¼ 0:008 2
40a a
The tangential component gyT is given by (Fig. 55a)
Gm Gm Gm Gm
gyT ¼ þ 2 sin a ¼ þ sinð71:6Þ
36a2 q 36a2 40a2
Gm Gm
¼ 2 ð0:028 þ 0:024Þ ¼ 0:004 2
a a
A
a q
2a B
q⬘
b a
6a m
3m
Fig. 55a
96 Gravity
GMr2
3cos2 # 1
c¼ 3
2R
where R is the distance between the centres of planets A and B (Fig. 55a), and # is the
angle the position vector r forms with the distance vector R (in this case it is equal to the
colatitude, # ¼ y)
GMr2
3cos2 y 1
c¼ 3
2R
The radial and tangential components of the acceleration are given by
@c @ Gmr2 Gmr
grT ¼ 3cos2 y 1 ¼ 3cos2 y 1
¼ 3 3
@r @r 2ð6aÞ 216a
2
1 @c 1 @ Gmr Gmr
gyT ¼ 3cos2 y 1 ¼
¼ 3ðcos y sin yÞ
r @y r @y 2ð6aÞ3 216a3
G3mr2
3cos2 y 1
c¼ 3
2ð6aÞ
@c Gmr
grT ¼ 3cos2 y 1
¼ 3
@r 72a
1 @c 9 Gmr
gyT ¼ ¼ ðcos y sin yÞ
r @y 216 ðaÞ3
2a B
(3/2)a (9/2)a
6a m
3m
Fig. 55b
(b) First we calculate the centre of gravity of the system formed by the two planets
measured from the centre of planet A (Fig. 55b):
3m 0 þ m 6a 3
x¼ ¼ a
3m þ m 2
The rotation radius for planet A is 3/2a and for planet B
3 9
6a a ¼ a
2 2
In the rotating system the centrifugal force equals the force of gravitational attraction,
which at the equator (y ¼ 0º) is
Gm3m 9
fg ¼ ¼ fC ¼ o2 r ¼ mo2 a
ð6aÞ2 2
From this expression we obtain the value of the angular velocity o of the rotation system:
3Gm2 9 Gm
2
¼ mo2 a ) o2 ¼
36a 2 54a3
Since the angular velocity of the spin of each planet is equal to that of the system, the spin
centrifugal force at the equator of planet A is
Gm Gm
fC ¼ o 2 r ¼ 3
2a ¼
54a 27a2
The tidal force is
2Gmr 2Gm2a
fT ¼ grT ¼ ¼ 3 3
R3 6 a
98 Gravity
GMr2
3cos2 # 1
c¼ 3
2R
where R is the centre-to-centre distance between the planets, and # the angle formed by the
vector r to a point and R (Fig. 56a). From this potential we calculate the radial component
of the tidal force:
@ GMr2
@c GMr
frT ¼ 2
¼ 3 3cos2 # 1
¼ 3
3cos # 1
@r @r 2R R
w w
gr
gλ
θ gθ
a
x
R
M M
Fig. 56a
fC ¼ o 2 a
Equating these two expressions we find the value of the spin angular velocity,
rffiffiffiffiffiffiffiffi
T 2 GM 1 GM
fC ¼ fr ) o a ¼ )o¼
256a2 16 a3
At a point on the surface of one of the planets the total potential is the sum of the
gravitational potential V, the spin potential F, and the tidal potential c:
GM 1 2 2 2 GMr2
3cos2 # 1
U ¼V þFþc¼ þ o r cos ’ þ 3
r 2 2R
For a point P at latitude ’ and longitude l (Fig. 56b)
GM 1 2 2 2 GMr2
3cos2 ’cos2 l 1
U¼ þ o r cos ’ þ 3
r 2 2R
P
ϑ
ϑ
j
j
a
Fig. 56b
GM GM 1 GMa 11 GM
gr ¼ 2 þ 3
a þ 3
3 1 ¼ 0:9993 2
a 256a 2 ð8aÞ 24 a
GM 1 1 GMa 1 1 1 GM
g’ ¼ a pffiffiffi pffiffiffi þ 3 pffiffiffi pffiffiffi ¼ 0:0027 2
256a3 2 2 ð8a3 Þ 4 2 2 a
pffiffiffi pffiffiffi
GMa 21 3 GM
gl ¼ 3 ¼ 0:0018 2
ð8aÞ3 2 2 2 a
(b) To obtain the angular velocity of the rotation of the system (O) we take into account
that the centrifugal force due to the rotation of the system at the equatorial plane is
equal to the gravitational attraction between the two planets:
MM
M
2 p ¼ G
ð8aÞ2
where p is the distance from the centre of one planet to the centre of gravity of the system.
Then we find
MM GM GM
M
2 4a ¼ G )
2 ¼ ¼
ð8aÞ2 4 64a3 256a3
and finally O ¼ o.
101 Tides
57. Consider two planets of equal mass m and radius a separated by a centre-to-centre
distance of 8a. Only gravitational forces act.
(a) Calculate the tidal force at the equator on one of the planets directly, using the
formula of the tidal potential (do so at l ¼ 0º, i.e. the point in front of the other
planet). Express the result in mGal given that Gm/a2 ¼ 980 000 mGal.
(b) Compare and comment on the reason for the difference between the results of the
direct calculation and using the tidal potential.
(c) What relationship must there be between the angular velocities of the planet’s
spin and of the system’s rotation for the centrifugal force due to the planet’s spin
to be equal to the tidal force at the equator and l ¼ 0º?
(a) The exact calculation of the tidal force at a point located at the equatorial plane in
front of the other planet is (Fig. 57)
Gm Gm Gm 1 1
fT ¼ ¼ 2 ¼ 46 875 gu
ð7aÞ2 ð8aÞ2 a 49 64
(b) The difference between the value obtained by the exact calculation and by using
the tidal potential is 8594 gu, that is, 18%. This is explained because the tidal
w w
a
x x
8a
m m
Fig. 57
102 Gravity
Gm 1 1
fT ¼ 2
a 49 64
If the spin angular velocity is op, the spin centrifugal force at the equator is given by
fC ¼ o2p a
Gm 1 1 Gm 1 1
fC ¼ fT ) o2p a ¼ 2 ) o2p ¼ 3 ð57:1Þ
a 49 64 a 49 64
The centrifugal force due the rotation of the system with angular velocity os is equal to the
gravitational attraction between the two planets:
Gm Gm2
mo2s 4a ¼ 2
) o2s ¼ ð57:2Þ
64a 256a3
From (57.1) and (57.2) we obtain the relation between the two angular velocities:
Gm 1 1
o2p a3 49 64 op
¼ ¼ 1:22 ) ¼ 1:10
o2s Gm os
256a3
58. Two planets of equal mass m and radius a are separated by a distance R. The spin
angular velocity of each planet is such that the centrifugal force at the equator is
equal to the maximum of the tidal force. If the sum of the two forces at the equator
cancels the gravitational force, what is the distance R?
The tidal potential is given by
Gmr2 1
3cos2 # 1
c¼ 3
R 2
where R is the centre-to-centre distance between the planets, r the radius to the point where
the tide is evaluated, and # the angle between r and R. The maximum value is at a point at
the equator in front of the other planet, # ¼ 0º and r ¼a. Then
@c Gmr 2Gma
fT ¼ ¼ 3 3cos2 # 1 ¼ ð58:1Þ
@r R R3
The spin centrifugal force for a point at the equator is
fC ¼ o 2 a ð58:2Þ
Equating (58.1) and (58.2) we obtain the spin angular velocity,
103 Tides
2Gma 2Gm
3
¼ o2 a ) o2 ¼ 3
R R
If, at the point considered, the sum of the spin centrifugal force and the tidal force cancel
the gravitational force of the planet, then
F þ fC þ fT ¼ 0
where F ¼ Gm/a2.
The value of R must be
4Gma Gm p3
ffiffiffiffiffi
3
¼ 2 ) R ¼ 4a
R a
59. Two spherical planets A and B of radii 2a and a and masses 5M and M spin on
their axes with equal angular velocities. They are separated by a centre-to-centre
distance of 8a, and form a system that rotates in the equatorial plane of both planets
with an angular velocity that is equal to that of the spin angular velocity of each one.
(a) Determine the total potential for points on planet A.
(b) Determine the expression for the three components of the total gravity, including
the tide, for a point on the surface of planet A at longitude 0º.
(c) If the Love number h on planet A is 0.5, determine the height of the terrestrial tide
as a multiple of a at the equator, at local noon with respect to planet B, in the case
that the system’s rotational angular velocity is the same as that of the spin of the
two planets about their axes.
(a) We calculate the centre of gravity of the system, measured from the centre of planet
A (Fig. 59a):
5M 0 þ M 8a 4
X ¼ ¼ a
5M þ M 3
2a B
(4/3)a
x x
8a M
5M
Fig. 59a
104 Gravity
P
ϑ
ϑ
j
j
2a
Fig. 59b
The total potential U at a point on the surface of planet A at latitude ’ is given by the sum
of the gravitational potential V, plus the spin potential Ф, plus the tidal potential c
produced by planet B:
U ¼V þFþc
5GM 1 2 2 2 GMr2 ð59:1Þ
3cos2 # 1
U¼ þ o r cos ’ þ 3
r 2 ð8aÞ 2
where according to Fig. 59b
cos # ¼ cos ’ cos t
where t is the local time of planet B with respect to planet A (hour-angle), at a point of l ¼ 0º,
the geographical longitude at planet A. For t ¼ 0, the planet B is in front of the point, so
5GM 1 2 2 2 GMr2
3cos2 ’cos2 t 1
U¼ þ o r cos ’ þ 3
r 2 2ð8aÞ
(b) The components of gravity are
@U 5GM GMr
¼ 2 þ o2 rcos2 ’ þ 3cos2 ’cos2 t 1
gr ¼ 3
@r r ð8aÞ
1 @U GMr
gy ¼ ¼ o2 r cos ’ sin ’ þ 3 cos ’ sin ’cos2 t
r @’ ð8aÞ3
1 @U GMr
gl ¼ ¼ 3 cos ’ cos t sin t
r cos ’ @t ð8aÞ3
105 Tides
(c) At the equator ’ ¼ 0º, at 12 h with respect to B, t ¼ 180º, h ¼ 1/2, and o ¼ O. The
height of the equilibrium terrestrial tide is given by
c
B¼h
g
2GM ð2aÞ2 GM
c¼ ¼
2ð8aÞ3 128a
If we approximate g by gr
5GM GM 2a 5GM GM
gr ¼ þ o2 2a þ 3
2¼ þ þ o2 2a
4a2 ð8aÞ 4a 2 128a 2
159GM
¼ þ 2o2 a
128a2
and the height of the equilibrium tide is
GM
1 128a
B¼
ð59:2Þ
2 159MG 2
þ 2ao
128a2
We know that the angular velocity of the rotation of the system is equal to the spin angular
velocity of both planets, so the spin angular velocity is given by
4 G5M 2 3GM
5M o2 a ¼ 2
) o2 ¼
3 ð8aÞ 256a2
By substitution in (59.2)
GM
1 128a a
B¼
¼
2 159MG 3GM 312
þ 2a
128a2 256a2
60. The Earth is formed by a sphere of radius a and density r, and a core of radius a/2
and density 2r, in the northern hemisphere, centred on the axis of rotation and
106 Gravity
tangent to the equatorial plane. The Moon has mass M/4 (where M ¼ (4/3)pra3), is at
a distance (centre-to-centre) of 4a, and orbits in the equatorial plane. Determine:
(a) The total potential and the components of gravity including the tidal forces.
(b) The total deviation of the vertical from the radial at lunar noon, and the deviation
due to the tide at the same hour for latitude 45º N, with m ¼ 1/8.
(a) The total potential U is equal to the gravitational potential of the planet V1 with
uniform density plus that of the core V2 using the differential mass, the spin
potential F, and the tidal potential c. The gravitational potentials are given by
(Fig. 60):
GM
V1 ¼
r
GM 0
V2 ¼
q
where the differential mass of the core is
4 a3 M GM
M 0 ¼ pð2r rÞ ¼ ) V2 ¼
3 8 8 8q
and q is the distance to the centre of the core. Its inverse can be approximated by
1 1 a a 2 1
2
¼ 1 þ cos y þ 3cos y 1
q r 2r 2r 2
The spin potential is
1
F ¼ o2 r2 sin2 y
2
and the total potential is
" ! #
GM 1 a a2 1 2
r 3 m
2
U¼ 1þ 1 þ cos y þ 3cos y 1 þ sin y þ c
r 8 2r ð2rÞ2 2 a 2
2r
a/2 q P
ϑ
q r
j L
4a
t – 180°
r
Fig. 60
107 Tides
where m ¼ o2 a3/GM.
The tidal potential c due to the Moon is given by
GML r2
3cos2 # 1
c¼ ð60:1Þ
2R3
According to Fig. 60
GMr2
3cos2 ’cos2 ðt 180Þ 1
c¼ 3
512a
The potential U is
a2
5 a
3 sin2 ’ 1
U ¼ GM þ sin ’ þ
8r 16r2 64r3
r2 m 2 r2
2 2
þ 3 cos ’ þ 3 cos ’ cos t 1
a 2 512a3
The components of gravity are found by taking the derivatives of U with respect to r and ’:
3a2
5 2a
3 sin2 ’ 1
gr ¼ GM 2 3
sin ’ 4
8r 16r 64r
r 2r
þ 3 m cos2 ’ þ 2 2
3 cos ’ cos t 1
a 512a3
1 @U
gy ¼
r @’
a2
GM a
¼ 2
cos ’ ð6 cos ’ sin ’Þ
r 16r 64r3
r2 m r2
2
þ 3 2 sin ’ cos ’ þ 6 sin ’ cos ’ cos t
a 2 512a3
gyM 1 1 @c 1 GMr2
tan i0 ¼ ¼ ¼ 6 sin ’ cos ’cos2 t
gr gr r @’ gr r512a3
and by substitution of the same values
i0 ¼ 0:3
The greater part of the deviation is due to the core.
61. Two spherical planets, planet A of radius 2a and mass 3m and planet B of radius a
and mass m, are separated by a centre-to-centre distance of 6a. The system rotates in
the equatorial plane and each planet spins on its axis with the same angular velocity.
What, for each planet, is the ratio between the force of gravity and the maximum of
the tidal force at the equator in front of the other planet?
The centre of gravity of the system measured from the centre of planet A is (Fig. 61)
0 3m þ 6am 3
¼ a
4m 2
The tidal force (radial component) can be calculated from the tidal potential
GMr2 @c GMr
3cos2 # 1 ) frT ¼ ¼ 3 3cos2 # 1
c¼ 3
2R @r R
where R is the centre-to-centre distance between the planets. For a point on the equator in
front of the other planet, # ¼ 0º, and
2GMr
frT ¼
ð6aÞ3
2Gm2a Gm
Planet A : frT ¼ ¼ 0:018 2
216a3 a
T 2G3ma Gm
Planet B : fr ¼ ¼ 0:028 2
216a3 a
Gravity without tides is the sum of the gravitational and centrifugal forces:
Gm
g¼ o2 r
r2
G3m 3Gm Gm Gm
Planet A : g ¼ o2 2a ¼ 2a ¼ 0:71 2
4a2 4a2 54a3 a
Gm 2 Gm Gm Gm
Planet B : g ¼ 2 o a ¼ 2 a ¼ 0:98 2
a a 54a3 a
109 Tides
2a B
(3/2)a (9/2)a
x x a
6a
m
3m
Fig. 61
62. The Earth is of radius a and density r, with a core of radius a/2 and density 3r
on the axis of rotation in the southern hemisphere tangent to the equatorial
plane. The Moon has mass M/2 and its centre is at 4a from the centre of the Earth
(M ¼ 4/3pra3).
(a) Write down the total potential.
(b) What is the value of the angular velocity of the Earth if at the point 30º N, 30º E at
06:00 lunar time the radial component of gravity is equal to GM/a2?
(c) In this case, what is the ratio between the angular velocity of the Earth’s rotation
and that of the system?
(a) As in previous problems the total potential U is given by
U ¼ V1 þ V2 þ F þ c
4p a3 M
M1 ¼ ð3r rÞ ¼
3 8 4
110 Gravity
P
ϑ
q
r 90° – q
q 4a L
t
a/2 x
3r
Fig. 62
Using the approximation for 1/q, where q is the distance from a point on the surface of the
Earth to the centre of the core, the gravitational potential due to the core is
a2
GM GM 1 a
2 cos y þ 3 3cos2 y 1
V2 ¼ ¼
4q 4 r 2r 8r
GM r2 1 2
GMr2
3sin2 ycos2 t 1
c¼ 3
3cos # 1 ¼ 3
2 64a 2 256a
where t is the hour-angle of the Moon (Fig. 62)
Then, the total potential U is
a2
5 a
3 cos2 y 1
U ¼ GM 2 cos y þ 3
4r 8r 32r
3
r2
r m
þ 3 sin2 y þ 2 2
3 sin y cos t 1
a 2 256a3
(b) The radial component of gravity is given by
@U
gr ¼
@r
3a2
5 2a
3 cos2 y 1
¼ GM 2 þ 3 cos y
4r 8r 32r4
rm sin2 y
2r 2 2
þ þ 3 sin y cos t 1
a3 256a3
111 Tides
GM 141 3 1 GM
gr ¼ þm ¼ 2
a2 128 4 128 a
(c) The centre of gravity r0 of the Earth–Moon system, measured from the centre of the
Earth, is given by
M
M r þ M r 4a 8a
¼ 2
¼
T 1 L 2
r0 ¼
MT þ ML 5 1 7
þ M
4 2
where the mass of the Earth MT includes that of the core,
M M
MT ¼ M þ and ML ¼
4 2
63. Consider two planets of equal mass m and radius a separated by a centre-to-
centre distance of 8a. The planets revolve around their centre of mass and
spin around their own axes. Their spin angular velocity is such that the value
of the centrifugal force is equal to the maximum tidal force of the two at the
equator.
112 Gravity
(a) Calculate, for a point on the equator of one of the planets and longitude l ¼ 90º, the
value of the vector g including all the forces acting at that point (l ¼ 0º corresponds
to the point on the line joining the two centres of the planets) at t ¼ 0.
(b) What is the deviation of the vertical from the radial at the point w ¼ 45º, l ¼ 0º?
(a) If o is the spin angular velocity of the two planets, the centrifugal force at the
equator y ¼ 90 , r ¼ a, only has radial component:
frC ¼ o2 a ð63:1Þ
The radial component of the tidal force can be obtained from the tidal potential c which, in
the first-order approximation, is given by (Fig. 63)
Gmr2
3cos2 # 1
c¼ 3
2R
Gmr2
3cos2 ðt lÞ 1
c¼
2R3
The radial component of the tidal force is
@ Gmr2
@c Gmr
frT ¼ 2
¼ 3 3cos2 ðt lÞ 1
¼ 3
3cos ðt lÞ 1
@r @r 2R R
w w
a 8a a
x x
90°
m m
Fig. 63
113 Tides
Then, as the centrifugal force is equal to the tidal force, we put (63.1) equal to (63.2) and
solve for o:
Gm Gm
o2 a ¼ ) o2 ¼ ð63:3Þ
256a2 256a3
We know that m O2 4a ¼ Gm2/R2, so solving for O,
Gm
2 ¼
256a3
and then using (63.3), we obtain o/O ¼ 1.
The total potential U is the sum of the gravitational, spin, and tidal potentials, which for
t ¼ 0, is given by
Gm 1 2 2 2 Gmr2
3sin2 ycos2 l 1
U ¼V þFþc¼ þ o r sin y þ 3
r 2 2ð8aÞ
The components of gravity including the tidal forces are
@U Gm Gmr
¼ 2 þ o2 rsin2 y þ 3sin2 ycos2 l 1
gr ¼ 3
@r r ð8aÞ
1 @U Gmr
gy ¼ ¼ o2 r sin y cos y þ 3 sin y cos ycos2 l ð63:4Þ
r @y ð8aÞ3
1 @U Gmr
gl ¼ ¼ 3 sin y cos l sin l
r sin y @l ð8aÞ3
By substitution of r ¼ a, l ¼ 90º, and y ¼ 90º we have
Gm Gm
gr ¼ þ o2 a
a2 512a2
gy ¼ 0
gl ¼ 0
Gm 1 Gm 3
gr ¼ 2 þ o2 a þ 1
a 2 512a2 2
Gm Gm 1 Gm 1 1021 Gm
¼ 2 þ 2
þ 2
¼
a 256a 2 512a 2 1024 a2
1 Gm 3 Gm 1 Gm 3 5 Gm
gy ¼ o 2 a þ ¼ þ ¼
2 512a 2 256a 2 512a 2 1024 a2
2 2 2
gl ¼ 0
114 Gravity
64. Two spherical planets of radii 2a and a and masses 8M and M separated by a
centre-to-centre distance of 4a spin on their own axes and rotate in the equatorial
plane with the same angular velocity.
(a) Determine all the forces acting at a point on the smaller planet at geocentric
coordinates w ¼ 60º N, l ¼ 0º (00:00 h local time corresponds to passage of the
other planet through the zero meridian).
(b) For this same point, calculate the astronomical latitude and the tidal deviation of
the vertical.
(a) First we determine the centre of gravity of the system, putting the origin at the
centre of the small planet (Fig. 64):
0 M þ 4a 8M 32a
x¼ ¼
9M 9
Because the spin angular velocity of each planet is equal to the angular velocity of the
system (o ¼ O), we can write, for the small planet, putting the gravitational attraction of
the two planets equal to the centripetal force:
G8MM 32a
2
¼ M o2
ð4aÞ 9
Ω
w
a q r q
2a
x x
4a
M
8M
Fig. 64
115 Tides
fr ¼ o2 rsin2 y
fy ¼ o2 r cos y sin y
Gm 9 GM 1 247GM
grGC ¼ 2
þ 2
¼ 2
a 64
pffiffiffi a 4 pffiffi256a
ffi
9 GM 3 GM 9 3
gyGC ¼ a ¼ 2
64 a3 4 a 256
To add the tidal force we use the tidal potential in the first-order approximation,
G8Mr2
3cos2 # 1
c¼ 3
2R
where cos # ¼ sin y cos (t l).
The tidal force for the point considered, r ¼ a, y ¼ 30º, t ¼ l ¼ 0º, and R ¼ 4a, is
given by
@c GM 1
frT ¼ ¼ 2
@r a 32
pffiffiffi
T 1 @c GM 3 3
fy ¼ ¼ 2
r @y a 32
The total force acting at the point is the sum of the three forces, gravitational, centrifugal,
and tidal:
255GM
grtotal ¼ 2
256a
pffiffiffi
GM 12 3
gytotal ¼ 2
a 256
The maximum deviation of the vertical due to the tide at the point considered is
i0 given by
116 Gravity
pffiffiffi
3 3
fyT
tan i0 ¼ ¼ 32 ¼ 0:163 ) i0 ¼ 9:3
grtotal 255
256
Gravity observations
65. Determine the values of gravity at the following series of points belonging to a
gravimetric survey with a Worden gravimeter, specifying the drift correction for each
of them.
where K is the instrument constant and DL is the difference between the readings at points A and B.
The gravity at point A is
gA ¼ gB þ DgAB ¼ 971 266:6 mGal
(b) The geopotential number at B can be calculated from the value at A in the form
g þ g
A B
CB ¼ CA þ hBA ¼ 82:661 59 gpu
2
where gravity is given in Gal and increments in height in km, because the geopotential
units are, 1 gpu ¼ 1 kGal m ¼ 1 Gal km.
The dynamic height is given by
CB
HDB ¼ ¼ 84:294 m
g45
The Helmert height can be calculated from the dynamic height by
C
H¼
g þ 0:0424H
where C is given in gpu, g in Gal, and H in km. Solving for H, we obtain
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
g g2 þ 4 0:0424C
H¼
2 0:0424
Taking the positive solution because point B is above the geoid (CB > 0) we obtain
HB ¼ 85:107 m
118 Gravity
Station Gravimeter reading (ru) Time Gravity (gu) Height difference (m)
Calculate the gravimeter readings corrected for drift, and the gravimeter constant
The instrument drift is given by
LAe LAb
d¼
tAe tAb
where LAb and LAe are the readings at the base A at the beginning and end of the
measurements taken at times tAb and tAe, respectively. By substitution we obtain,
d ¼ 1:07 ru=hour
A reading corrected at station j is given by
Lcj ¼ Lj dðtj tAb Þ
A 3614.351 10:10
B 3650.242 10:25
C 3610.633 10:37
A 3614.414 11:02
The gravimeter scale factor is 1.000 65, and the equivalence between reading units
and the relative value of gravity in mGal is given by
Given that the value of gravity at point A is 9.794 6312 m s2, calculate the values at
B and C.
First we correct the readings by the instrument drift:
LAe LAb
d¼
tAe tAb
where LAe and LAb are the readings at the base A at the end and the beginning of the survey
at times tAe and tAb. Then
d ¼ 0:0727 ru=hour
The corrected reading at each station j is given by
Lcj ¼ Lj dðtj tAb Þ
LcB ¼ 3650:224
LcC ¼ 3610:600
These readings are converted into relative gravity values Rj using the conversion table.
The reading at station A is
Main field
69. Assume that the geomagnetic field of the Earth is a geocentric dipole with a North
Pole at 80 N, 45 E and a magnetic moment 8 1022 A m². Calculate for a point with
geographical coordinates 45 N, 30 W the components NS, EW, and Z of the Earth’s
magnetic field, the declination and inclination, and the geomagnetic longitude.
Earth’s radius: 6370 km and the constant C ¼ 107 H m1 (this value is used in all
problems).
We calculate first the geomagnetic latitude and longitude (f , l ) from the geographical
coordinates (f, l) of the point and the geographical coordinates of the geomagnetic North
Pole (fB, lB) by the equation
fB ¼ 80 N
lB ¼ 45 E
f ¼ 45 N
l ¼ 30 W ¼ 330
we obtain
f ¼ 46:70
l ¼ 84:82
In the geocentric magnetic dipole model, the vertical (Z ) and horizontal (H ) components
of the magnetic field can be obtained from
Z ¼ 2B0 sin f
H ¼ B0 cos f ð69:1Þ
Cm
B0 ¼ 3
a
121
122 Geomagnetism
In these equations B0 is the geomagnetic constant, m the magnetic moment of the dipole,
a the Earth’s radius, and the constant C ¼107 H m1.
In this case we are given that
m ¼ 8 1022 A m2
a ¼ 6370 km ¼ 6.37 106 m
By substitution in Equations (69.1) we obtain:
B0 ¼ 30 951 nT
Z ¼ 45 051 nT
H ¼ 21 227 nT
The geomagnetic declination is given by
cos fB sinðl lB Þ
sin D ¼
cos f
D ¼ 14:16
X ¼ H cos D ¼ 20 582 nT
Y ¼ H sin D ¼ 5193 nT
f ¼ f ð90 fB Þ ¼ 50
l ¼ 180
The expressions for the geomagnetic vertical and horizontal components and for the total
geomagnetic field are
Z ¼ 2B0 sin f
H ¼ B0 cos f
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F ¼ H 2 þ Z 2
Cm
B0 ¼
a3
123 Main field
GMNP GNP
P
x
f∗ f
fB
l=0
l∗
Fig. 70
D ¼ 0
X ¼ H cos D ¼ H ¼ 19 895 nT
Y ¼ H sin D ¼ 0
(b) The equation of the line of force passing through a point with geomagnetic
co-latitude ỵ is
r ¼ r0 sin2 y
In this equation r0 is the distance from the Earth’s centre to a point on the line of force with
y ¼ 90º. The distance r0 is different for each line of force.
124 Geomagnetism
For the point with geomagnetic latitude f ¼ 50º located in the Earth’s surface (r ¼ a),
y ¼ 90º – f ¼ 40º, so
a
r0 ¼ 2 ¼ 15 417km
sin y
71. Assume that the geomagnetic field is produced by a geocentric dipole of magnetic
moment 7.5 1022A m², with North Pole at 75 N, 65 W, and that the Earth’s radius
is 6372 km. Calculate:
(a) The NS and EW components for a point on the Earth’s surface at which the
inclination is 67 and the geomagnetic longitude is 120 .
(b) The geographical coordinates of that point.
(c) The geomagnetic coordinates, field components, declination, and inclination of the
point on the geographical equator of zero geomagnetic longitude.
(a) The geomagnetic latitude f is obtained from
tan I ¼ 2 tan f ) f ¼ 49:7
The horizontal component, H , can be calculated from the geomagnetic constant, B0, and
the geomagnetic latitude:
Cm
B0 ¼ 3 ¼ 28 989 nT
a
H ¼ B0 cos f ¼ 18 761 nT
To obtain the geomagnetic declination we apply the sine law in the spherical triangle of
Fig. 71a:
D ¼ 23:1
It is important to note that we have added a minus sign in the last equation in order for the
declination be positive toward the east.
125 Main field
GNP
90º– fB
l – lB
180º– l∗ 90º – f
GMNP
q = 90º– f∗
D∗
Fig. 71a
The NS (X ) and EW (Y ) components are
X ¼ H cos D ¼ 17 262 nT
Y ¼ H sin D ¼ 7 349 nT
f ¼ 55:1
The geographical longitude is obtained by applying the cosine law to the spherical triangle
of Fig. 71a:
so
l lB ¼ 101:6
To take the inverse cosine in the correct quadrant we bear in mind that l < 0 implies
that the point is to the west of the Geomagnetic North Pole, that is l – lB < 0. So we
obtain
l ¼ 166:6 W
(c) If this point is on the geographical equator (f ¼ 90 ) and has zero geomagnetic
longitude (l ¼ 0º), it is on the same geographical meridian as the Geomagnetic
North Pole. Then from Fig. 71b
126 Geomagnetism
GMNP GNP
fB
f∗
x
Fig. 71b
f ¼ 90 fB ¼ 15:0
l ¼ 0
Z ¼ 2B0 sin f ¼ 15 006 nT
H ¼ B0 cos f ¼ 28 001 nT
D ¼ 0
X ¼ H
Y ¼ 0 nT
tan I ¼ 2 tan f ) I ¼ 28:2
72. Assume that the geomagnetic field is that of a dipole with North Pole at 75 N, 0 E.
What is the conjugate point of that of geographical coordinates 30 N, 30 E?
First, we calculate the geomagnetic coordinates (f , l ) (Problem 71; Fig. 71a):
fB ¼ 75 lB ¼ 0
f ¼ 30 l ¼ 30
127 Main field
GMNP GNP
fB
Px
f∗
– f∗
P1
x
Fig. 72
f ¼ 42:6
l ¼ 36:0
A magnetic conjugate point is a point on the Earth’s surface that is located on the same line
of force and in the opposite hemisphere (Fig. 72, P and P1). Then, its geomagnetic
coordinates (f1 , l1 ) are
To calculate the geographical coordinates for this point (f1, l1) we use the spherical
triangle of Fig. 71a. We calculate the geographical latitude applying the cosine law:
73. Assume the centred dipole approximation, with the coordinates of the Geomag-
netic North Pole being 65 N, 0 E, and the magnetic moment of the dipole 8 1022
A m². Calculate, for a point on the Earth’s surface at geographical coordinates 30 N,
30 E:
(a) The geographical coordinates of the conjugate point.
(b) The declination, inclination, and vertical and horizontal components of the field
at both points. Compare and contrast the results.
Earth’s radius: 6370 km.
(a) First, we calculate the geomagnetic coordinates (f , l ) for point P with geograph-
ical coordinates f ¼ 30 N, l ¼ 30 E using the equations (Problem 71; Fig. 71a)
sin f ¼ sin fB sin f þ cos fB cos f cosðl lB Þ
sinðl lB Þ cos f ð73:1Þ
sin l ¼
cos f
The geographical coordinates of the Geomagnetic North Pole are
fB ¼ 65 ; lB ¼ 0
Substituting in Equations (73.1) results in
f ¼ 50:4
l ¼ 42:7
The geomagnetic coordinates (f1 , l1 ) of the magnetic conjugate point P1 satisfy
(Problem 72):
f1 ¼ f ¼ 50:4
l1 ¼ l ¼ 42:7
To calculate the geographical coordinates for this point (f1, l1) we use the spherical
triangle of Fig. 71a. We calculate the geographical latitude applying the cosine law
cosð90 f1 Þ ¼ cosð90 fB Þ cosð90 f1 Þ
þ sinð90 fB Þ sinð90 f1 Þ cosð180 l1 Þ
sin f1 ¼ sin fB sin f1 cos fB cos f1 cos l1
f1 ¼ 63:7
129 Main field
To calculate the geographical longitude we apply the cosine law again and, solving for l1,
P P1
cos fB sinðl lB Þ cos fB sinðl1 lB Þ
sin D ¼ sin D1 ¼
cos f cos f1
D ¼ 19.33 D1 ¼ 40:2
tan I ¼ 2 tan f ) I ¼ 67.5 tan I1 ¼ 2 tan f1 ) I1 ¼ I1 ¼ 67:5
Z ¼ 2B0 sinf ¼ 47 662 nT Z1 ¼ 2B0 sin f1 ¼ Z ¼ 47 662 nT
H ¼ B0 cosf ¼ 19 750 nT H1 ¼ B0 cos f1 ¼ H ¼ 19 750 nT
74. Assume the centred dipole approximation, with the coordinates of the Geomag-
netic North Pole 78.5 N, 70.0 W, and the magnetic dipole moment being 8.25 1022
A m². Calculate, for a point on the surface with coordinates 60.0 S, 170.0 W:
(a) Its geomagnetic coordinates, declination, inclination, and vertical and horizontal
components of the field.
(b) The potential at that point.
(c) The declination and inclination at the point diametrically opposite to it.
Earth’s radius: 6370 km.
(a) We calculate the geomagnetic coordinates (f , l ) using the equations (Problem 71,
Fig. 71a)
fB ¼ 78:5 lB ¼ 70:0
f ¼ 60:0 l ¼ 170:0
130 Geomagnetism
f ¼ 60:0
l ¼ 80:0
The geomagnetic declination is given by (Problem 71, Fig. 71a)
cos fB sinðl lB Þ
sin D ¼
cos f
D ¼ 23:1
B0 ¼ 31 918 nT
Z ¼ 2B0 sin f ¼ 55 279 nT
H ¼ B0 cos f ¼ 15 963 nT
GNP GMNP
P1 fB
f1
f f∗
l∗
P
Fig. 74
So, we can notice that the two points P and P1 (Fig. 74) that are diametrically opposite
are not magnetic conjugate points because the geomagnetic longitudes are different
by 180º.
75. Consider a point P on the Earth’s surface at coordinates 30 S, 10 W at which
the NS component of the geomagnetic field is 27 050 nT and the EW component is
5036 nT, with the geomagnetic inclination being negative. Assuming the centred
dipole hypothesis with magnetic moment 7.8 1022 A m², calculate:
(a) The geographical coordinates of the Geomagnetic North Pole.
(b) The geomagnetic coordinates of P’s conjugate point.
(a) We calculate first the geomagnetic constant B0:
Cm
B0 ¼ ¼ 30 177 nT
a3
The geomagnetic declination D is obtained from the NS (X ) and EW (Y ) components
of the geomagnetic field:
Y 5036
tan D ¼ ¼ ) D ¼ 10:5
X 27 050
132 Geomagnetism
tan I ¼ 2 tan f
I < 0 ) f < 0
f ¼ 24:3
With these results we calculate the geographical coordinates of the Geomagnetic North
Pole (fB, lB) using the spherical triangle in Fig. 71a. Applying the cosine rule for the angle
90º – fB:
To calculate the longitude lB of the Geomagnetic North Pole, we apply the cosine law for
the angle 90º – f :
D < 0 ) l lB > 0
lB ¼ 61 10 ¼ 71:0 W
(b) The geomagnetic coordinates (f1 , l1 ) of P’s conjugate point verify that
76. At a point P on the Earth’s surface with coordinates 45 N, 30 W, the value of the
total geomagnetic field is 49 801 nT, the horizontal component is 21 227 nT, and the
EW component is 5171 nT, with the magnetic inclination being positive. Calculate:
(a) The geographical coordinates of the Geomagnetic North Pole.
(b) The value of the geomagnetic potential at P.
(c) The distance from the Earth’s centre to the point at which the line of force passing
through P intersects the geomagnetic equator.
Earth’s radius: 6370 km.
(a) We calculate first the geomagnetic inclination, latitude, and declination by
H
cos I ¼ ) I ¼ 64:8
F
tan I ¼ 2 tan f ) f ¼ 46:7
Y
sin D ¼ ) D ¼ 14:1
H
With these results we calculate the geographical coordinates of the Geomagnetic North
Pole solving the spherical triangle (Fig. 71a) in the same way as in Problem 71:
(b) The geomagnetic potential at point P on the Earth’s surface (r ¼ a ¼ 6370 km) is given by
Cm cos y
F¼ ¼ B0 a sin f ð76:1Þ
a2
We calculate the geomagnetic constant B0 from the horizontal component H :
21 227
H ¼ B0 cos f ) B0 ¼ ¼ 30 951 nT
cosð46:7 Þ
Substituting in the potential equation (76.1) we obtain
F ¼ 143 T m
(c) The equation of the line of force passing through a point with geomagnetic co-
latitude y is
r ¼ r0 sin2 y
134 Geomagnetism
In this equation r0 is the distance from the Earth’s centre to the point at which the line of
force passing through P intersects the geomagnetic equator. Substituting r ¼ a ¼ 6370 km
gives
a a
r0 ¼ 2
¼ ¼ 13 543 km
sin y cos2 f
77. Assume the centred dipole approximation, with the coordinates of the Geo-
magnetic North Pole being 75 N, 65 W, and the magnetic moment of the dipole
7.5 1022 A m2. For a point on the Earth’s surface at which the inclination is 67 and
the geomagnetic longitude is 120 , calculate:
(a) The NS and EW components.
(b) Its geographical coordinates.
The Earth’s radius: 6372 km.
(a) We calculate first the geomagnetic constant B0, latitude f , and horizontal H
component:
Cm
B0 ¼ ¼ 28 989 nT
a3
tan I ¼ 2 tan f ) f ¼ 49:7
H ¼ B0 cos f ¼ 18 761 nT
cos fB sin l
sin D ¼
cos f
D ¼ 23:1
X ¼ H cos D ¼ 17 262 nT
Y ¼ H sin D ¼ 7349 nT
To calculate the geographical longitude we apply the cosine law for the angle 90º – f
(Fig. 71a):
To choose the correct solution we notice that the declination is positive and then the point
must be to the west of the Geomagnetic North Pole
D > 0 ) l lB < 0
l ¼ 101:7 65 ¼ 166:7 W
78. Assume a spherical Earth of radius 6370 km, with magnetic field produced by a
centred dipole whose northern magnetic pole is at 70 N, 60 W. Given that for a point
on the surface with coordinates 50 S, 80 W the horizontal component is 24 890 nT,
calculate:
(a) The magnetic dipole moment.
(b) The geographical coordinates of the conjugate point.
(a) We calculate first the geomagnetic latitude by (Fig. 71a)
To obtain the magnetic dipole moment m we need the geomagnetic constant B0, which is
related with the horizontal component H by
H
B0 ¼ ¼ 29 007 nT
cos f
Cm B0 a3
B0 ¼ 3
)m¼ ¼ 7:5 1022 A m2
a C
(b) Let us obtain first the geomagnetic longitude by (Problem 71, Fig. 71a)
sinðl lB Þ cos f
sin l ¼
cos f
l ¼ 14:8
The geomagnetic coordinates of the conjugate point (f1 , l1 ) are (Fig. 78)
GNP
GMNP
fB
P1
f∗1
l∗
f∗
P
Fig. 78
Solving again the spherical triangle (Fig. 71a) we calculate the geographical coordinates
(f1, l1):
79. If the Earth’s geomagnetic field is produced by a centred dipole, tilted 15 away
from the axis of rotation, of magnetic moment 7.6 1022 A m2, and the Geomagnetic
North Pole is at longitude 65 W, calculate:
(a) The geomagnetic constant in nT.
(b) The geographical coordinates of a point on the Earth’s surface at which the
declination is D ¼ 14 15.50 and the inclination is I ¼ 65 23.50 . Discuss the
possible solutions.
(c) The geographical and geomagnetic longitude of the agonic line.
Assume a spherical Earth of radius 6370 km.
137 Main field
(a) We calculate the geomagnetic constant from the magnetic moment m, the Earth’s
radius a, and the constant C¼107 H m1, by
Cm
B0 ¼ ¼ 29 403 nT
a3
(b) Let us obtain first the geomagnetic latitude from the inclination I by
tan I ¼ 2 tan f ) f ¼ 47:5
If the dipole is tilted 15 away from the axis of rotation the latitude of the Geomagnetic
North Pole will be
fB ¼ 90 15 ¼ 75:0
With these results we calculate the geographical latitude f solving the spherical triangle
(Fig. 71a). Applying the cosine rule,
cosð90 fB Þ ¼ cosð90 f Þ cosð90 fÞ þ sinð90 f Þ sinð90 fÞ cos D
ð79:1Þ
sin fB ¼ sin f sin f þ cos f cos f cos D
To obtain the geographical latitude from this equation we can carry out a change of
variables, introducing two new variables (m, N ) such that
sin f ¼ m cos N
ð79:2Þ
cos f cos D ¼ m sin N
From these equations we can calculate P and N:
cos f cos D cos D
tan N ¼ ¼ ) N ¼ 41:6
sin f tan f
cos N
P¼ ¼ 1:02
sin f
Substituting Equations (79.2) in Equations (79.1) we obtain
sin fB ¼ P cos N sin f þ P cos N cos f ¼ P sinðf þ N Þ
sin fB sin fB cos N
sinðf þ N Þ ¼ ¼
m sin f
f þ N ¼ 78:4 ) f ¼ 36:8
80. The Earth’s magnetic field is produced by two dipoles of equal moment
(M ¼ Cm ¼ 9.43 109 nT m3) and polarity, forming angles of 30 and 45 with the
axis of rotation, and contained in the plane corresponding to the 0 meridian. Find
the potential of the total field and the coordinates of the resulting magnetic North
pole, taking the Earth’s radius to be 6000 km.
The total potential at a point is the sum of the potentials of the two dipoles. If M is the
magnetic moment (M ¼ Cm), r is the distance from the dipole’s centre, and y1 and y2
are the geomagnetic co-latitude relative to each dipole (Fig. 80), the total potential F is
given by
The geographical coordinates of the two Geomagnetic North Poles are given by
GNP
45° 30°
GMNP1
GMNP2 q2
q1
Fig. 80
139 Main field
M ½ðsin fB1 þ sin fB2 Þ sin f þ ðcos fB1 cos fB2 Þ cos f cos l
F¼
r2 ffiffiffi
pffiffiffi pffiffiffi p
M 3 þ 2 sin f þ 1 2 cos f cos l
F¼
2r2
If we call # the geographic co-latitude, # ¼ 90 f, then
pffiffiffi pffiffiffi pffiffiffi
M 3 þ 2 cos # þ 1 2 sin # cos l
F¼
2r2
The resulting magnetic North Pole, the point where the inclination I ¼ 90º, due to the
combined effect of the two dipoles is given by
Z
tan I ¼
H
and therefore at the magnetic Pole, H ¼ 0.
We derive the component H by taking the gradient of the potential F
pffiffiffi pffiffiffi pffiffiffi
1 @F M 3 þ 2 sin # þ 1 2 cos # cos l
X ¼ B# ¼ ¼
r @# 2r3
pffiffiffi
1 @F M 1 2 sin l
Y ¼ Bl ¼ ¼
r sin # @l 2r3
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
H ¼ X þY 2 2
Since the magnetic North Pole is contained in the plane corresponding to the 0 geograph-
ical meridian, then its longitude is either 0º or 180º.
If the longitude is 0º
M h pffiffiffi pffiffiffi pffiffiffi i
l ¼ 0 ) H ¼ 3 3 þ 2 sin # þ 1 2 cos # ¼ 0
2r
# ¼ 8 ¼ 172
But this result doesn’t correspond to the north hemisphere. Then we must take the
geographical longitude 180º:
M h pffiffiffi pffiffiffi pffiffiffi i
l ¼ 180 ) H ¼ 3 3 þ 2 sin # 1 2 cos # ¼ 0
2r
# ¼ 8 ) fB ¼ 82
140 Geomagnetism
This is the correct result and the coordinates of the magnetic North Pole are
fB ¼ 82 ; lB ¼ 180
81. The Earth’s magnetic field is produced by one dipole in the direction of the axis
of rotation (negative pole in the northern hemisphere) and another with the same
moment in the equatorial plane which rotates with differential angular velocity v
with respect to the points on the surface of the Earth (consider that the Earth doesn’t
rotate). Its negative pole passes through the 45 E meridian at time t ¼ 0 and
completes a rotation with respect to that point in 24 hours. Consider a point of
geographical coordinates 45 N, 45 E.
(a) Calculate the magnetic field components (Br, Bu, Bl) at that point.
(b) Illustrate graphically how each of them varies with local time.
(a) The total potential at a point on the surface of the Earth is the sum of the potentials
of the two dipoles (Problem 80, Equation 80.1):
M cos y1 M cos y2 M ðcos y1 þ cos y2 Þ
F ¼ F1 þ F2 ¼ ¼
r2 r2 r2
Dipole 1 is in the direction of the axis of rotation and so the geomagnetic co-latitude of the
point with respect to this dipole (Fig. 81a) is equal to the geographical co-latitude,
y1 ¼ 90 f
cos y1 ¼ sin f
GMNP1
GNP
q1
P
q2
GMNP2
Fig. 81a
141 Main field
Dipole 2 is on the equatorial plane (fB2 ¼ 0) and rotates with respect to the points of
the surface. Owing to this rotation its geographical longitude lB2 changes with time t in
the form
lB2 ¼ ot þ 45
where o is the angular velocity, o ¼ 360 /T, T being the rotation period of 24 h.
The co-latitude y2 is
We obtain the magnetic field components (Br , By, Bl) at the point (#, l) by taking the
gradient in spherical coordinates of the potential F:
2B0
Br ¼ pffiffiffi ð1 þ cos ot Þ
2
B0
B# ¼ pffiffiffi ð1 cos ot Þ
2
Bl ¼ B0 sin ot
(b) The variation of each component with local time is shown in Fig. 81b
142 Geomagnetism
1
Bl
0
Br
Br,q,l
B0
–1
Bθ
–2
0 5 10 15 20
t (h)
Fig. 81b
Magnetic anomalies
82. Calculate the magnetic anomaly created by a magnetic dipole buried at depth d,
arbitrarily oriented, at an angle to the vertical of a. The negative pole is upwards.
Consider a point P with coordinates (x, z), where x is measured along the horizontal from
the projection of the centre of the dipole and z is the vertical from the reference level (the
Earth’s surface). The position vector r forms an angle b to the vertical (Fig. 82). The
anomalous magnetic potential created by the dipole for this point is
Cm cosða þ bÞ Cmðsin b cos a þ cos b sin aÞ
F ¼ ¼ ð82:1Þ
r2 r2
x P
z
X
r
d
b
–
Fig. 82
B ¼ rðFÞ
The vertical component of the magnetic field anomaly, taking the z-coordinate positive
downward, is
h i
2 2
@ðFÞ Cm x þ ðz þ dÞ cos a 3ðz þ dÞ ½ðz þ dÞ cos a x sin a
Z ¼ ¼ i5=2
@z
h
x2 þ ðz þ dÞ2
The component of the magnetic anomaly in an arbitrary horizontal direction x for the
Earth’s surface points (z ¼ 0) is given by
Z ¼ 0
Substituting
Cm ¼ 5 105 T m3
d ¼ 10 m
we obtain
Y ¼ 50 nT
jBj ¼ Y ¼ 50 nT
X ¼ 0
P
E
a
– +
Fig. 83a
145 Magnetic anomalies
GNP
GMNP
fB
f
f∗
Fig. 83b
To calculate the components of the magnetic anomaly in the direction of the Earth’s
magnetic field, F, and their horizontal component, H, we need to determine the magnetic
declination and inclination at the point:
H ¼ X cos D þ Y sin D
ð83:1Þ
F ¼ H cos I þ Z sin I
Because the point has the same longitude as the Geomagnetic North Pole (Fig. 83b),
H ¼ 0
F ¼ 0
The total value of the field in the NS, EW, and vertical directions, and total field F are
XT ¼ X þ X
YT ¼ Y þ Y
ZT ¼ Z þ Z
FT ¼ F þ F
146 Geomagnetism
The vertical Z and horizontal H components of the geomagnetic field are given by
X ¼ H cos D ¼ H ¼ 17 894 nT
Y ¼ H sin D ¼ 0
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F ¼ H 2 þ Z 2 ¼ 55 994 nT
XT ¼ 17 894 nT
YT ¼ 50 nT
ZT ¼ 53 058 nT
FT ¼ 55 994 nT
The variations in magnetic declination and inclination due to the presence of the buried
dipole are
YT
tan D0 ¼ ) D0 ¼ 0:02
XT
D D0 ¼ 0:02
ZT
tan I 0 ¼ ) I 0
71:4 ¼ I
HT
84. Buried at a point with magnetic latitude 30 N and a depth of 50 m is a horizontal
magnetic dipole with Cm ¼ 107 nT m3 with the positive pole to the geographical
north.
(a) Calculate DF if B0 ¼ 30 000 nT and the declination at that point is 15 . Find the
ratio DF /F.
(b) How far from the dipole’s centre along the north–south line will the dipole field
strength be in the same direction as that of the Earth (take D ¼ 0 ).
(a) The component of the magnetic anomaly DF in the direction of the Earth’s
magnetic field (the total field anomaly) is given by
P
N
a
– +
Fig. 84
Cm
X ¼
d3
Y ¼ 0
Z ¼ 0
Cm ¼ 107 nT m3
d ¼ 50 m
we obtain
X ¼ 80 nT
Substituting D ¼ 15º, the component of the magnetic anomaly in the direction of the
horizontal component H of the Earth’s magnetic field is
H ¼ X cos D ¼ 77 nT
At a point of magnetic latitude f ¼ 30 the magnetic inclination is
F ¼ 50 nT
To calculate the geomagnetic field F we first obtain the components H and Z :
The ratio of the total field anomaly and the Earth’s total magnetic field is
F
¼ 1:26 103
F
(b) If the dipole field strength is in the same direction as that of the Earth then the
inclination I 0 due to the dipole is equal to that of the Earth’s field I , where
Z
tan I 0 ¼
H
Z
tan I ¼
H
Assuming D ¼ 0º, then
H ¼ E
If we substitute in Equations (82.2) and (82.3) of Problem 82, the angle a ¼ 90º because
the dipole is horizontal, we obtain
3Cmdx
Z ¼
5=2
x þ d2
2
Cmð2x2 d 2 Þ
X ¼
5=2
x2 þ d 2
We have changed the sign of the vertical component because the negative pole is toward
the south.
Applying the condition, tan I 0 ¼ tan I , we obtain
Z Z Z
¼ ¼
H X H
3Cmdx
5=2
x þ d2
2 3dx Z
¼ ¼
Cm 2x d 2 Þ 2x2 d 2 H
2
5=2
x2 þ d 2
2Z x2 3dH x Z d 2 ¼ 0
We have two solutions: a point 80 m to the north from the surface projection of the dipole’s
centre and another 15 m to the south.
149 Magnetic anomalies
P
N
d
+
45°
Fig. 85a
150 Geomagnetism
To obtain the Earth’s main field we eliminate the buried dipole contribution from the
observed values:
F ¼ F F ¼ 56 369 nT
H ¼ H H ¼ 11 866 nT
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Z ¼ ðF Þ2 ðH Þ2 ¼ 55 106 nT
X ¼ H cos D ¼ 9182 nT
Y ¼ H sin D ¼ 7516 nT
D0 D ¼ 2:1
(c) We calculate first the geomagnetic latitude of the point from the vertical and
horizontal components:
Z ¼ 2B0 sin f
H ¼ B0 cos f
Z
tan f ¼
2H
f ¼ 66:7
With this value, the declination D and the geographical coordinates of the point (f, l), we
can solve the spherical triangle (Fig. 85b) and obtain the geographical coordinates of the
Geomagnetic North Pole:
GNP
90º – fB
l – lB
180º – l∗ 90º – f
GMNP
D∗
q = 90º– f ∗
Fig. 85b
151 Magnetic anomalies
The correct solution is the negative one because a positive value of the declination implies
that the point is to the west of the Geomagnetic North Pole:
D > 0 ) l lB < 0
lB ¼ 0
86. Located at a point with geographical coordinates 45 N, 30 W, at a depth of 100 m,
is a dipole of magnetic moment Cm ¼ 1 T m3, inclined 45 to the vertical towards the
south, with the positive pole upwards, and in the geographical north–south vertical
plane. The Earth’s dipole has its north pole at 60 N, 0 E and B0 ¼ 30 000 nT.
Calculate:
(a) The values of Z, H, F at the given point.
(b) Where does the compass point to at that same point?
(a) We calculate first the geomagnetic latitude corresponding to the point by
From this value we obtain the geomagnetic components Z , H and the total main field F :
cos fB sinðl lB Þ
sin D ¼
cos f
D ¼ 39:2
The magnetic anomaly created by the dipole buried at depth d is given by Equations (85.1)
of Problem 85. Substituting Cm ¼ 1 T m3, d ¼ 100 m, and a ¼ 45º, we obtain
2Cm cos a
Z ¼ ¼ 1414 nT
d3
Cm sin a
X ¼ ¼ 707 nT
d3
Y ¼ 0
H ¼ X cos D ¼ 548 nT
F ¼ H cos I þ Z sin I ¼ 1266 nT
Z ¼ Z þ Z ¼ 53 693 nT
F ¼ F þ F ¼ 55 104 nT
H ¼ H þ H ¼ 12 414 nT
(b) To calculate in what direction the compass points we need the value of the observed
declination D’ including the effects of the geomagnetic field and the buried dipole:
Y Y þ Y
tan D0 ¼ ¼
X X þ X
Y ¼ H sin D ¼ 7500 nT
X ¼ H cos D ¼ 9195 nT
D0 ¼ 37:1
87. Located at a point on the Earth with geographical coordinates 45 N, 30 E, at a
depth of 100 m, is a dipole of magnetic moment Cm ¼ 107 nT m3, tilted 45 to the
vertical towards the south, with the positive pole downwards, and contained in the
plane of true north. The Earth’s field is produced by a centred dipole tilted 30 from
the axis of rotation in the plane of the 0 meridian, with B0 ¼ 30 000 nT. Calculate
the total values of F, Z, and H observed at the point of the surface above the centre of the
buried dipole.
We first calculate the geographical coordinates of the Geomagnetic North Pole and the
geomagnetic latitude
The geomagnetic field components Z , H and the total main field F are given by
H ¼ X cos D ¼ 5 nT
F ¼ H cos I þ Z sin I ¼ 13 nT
Z ¼ Z þ Z ¼ 55 121 nT
F ¼ F þ F ¼ 56 383 nT
H ¼ H þ H ¼ 11 861 nT
88. Buried at a point with geographical latitude 20 N and the same longitude as the
geomagnetic pole, at a depth of 200 m, is a sphere of 50 m radius of material with
magnetic susceptibility 0.01. The Earth’s field is produced by a centred dipole tilted
10 from the axis of rotation and magnetic moment M ¼ 1030 g cm3 (Earth’s radius:
6000 km). Calculate:
(a) The anomaly produced by induced magnetization in the sphere at a point on the
Earth’s surface above the centre of the sphere. Give the vertical and horizontal
components in units of nT.
(b) The total anomaly for a point on the Earth’s surface 100 m south of the above
point.
154 Geomagnetism
GNP GMNP
q = 90° – f ∗
10°
20°
Fig. 88a
(a) We first calculate the geomagnetic co-latitude (y) and latitude (f ) of the point,
knowing that it is in the same meridian as the Geomagnetic North Pole (Fig. 88a):
y ¼ 90 f ¼ 90 10 20 ¼ 60
f ¼ 30
F∗
49°
GN P
MN
41°
Fig. 88b
To calculate Cm we use the magnetic susceptibility w and the volume V of the sphere:
Cm ¼ wF V ¼ 3:2 107 nT m3
Substituting this value in the equation of the components of the magnetic anomaly we
obtain
(b) The anomaly created by the sphere at a point at a distance x ¼100 m to the south of
the above point is given by
Cm ðx2 2d 2 Þ cos a þ 3dx sin a
Z ¼
5=2 ¼ 0:8 nT
x2 þ d 2
Cm ð2x2 d 2 Þ sin a 3dx cos a
H ¼
5=2 ¼ 3:3 nT
x2 þ d 2
89. Buried at a point on the Earth at magnetic latitude 45 N, at a depth d, is a vertical
dipole of magnetic moment M (Cm) with negative pole upwards. If M/d3 ¼ 10B0 (B0 is
the geomagnetic constant of the main field) calculate how far along the magnetic
meridian the direction of the buried dipole’s field will coincide with that of the Earth
(the terrestrial dipole field).
First we calculate the geomagnetic field components and the inclination by
pffiffiffi
Z ¼ 2B0 sin f ¼ 2B0 nT
pffiffiffi
2
H ¼ B0 cos f ¼ B0 nT
2
Z
tan I ¼ ¼ 2 ) I ¼ 63:4
H
The components of the magnetic field created by the dipole are given by Equations (82.2)
and (82.3) of Problem 82, putting a ¼ 0º:
Cm x2 2d 2
Z ¼ 5=2
x2 þ d 2
Cm3dx
H ¼ 5=2
x þ d2
2
If the buried dipole’s field coincides with that of the Earth the magnetic inclinations due to
both have to be equal and so
Z
¼ tan I ¼ 2
H
pffiffiffiffiffi
x2 þ 6xd 2d 2 ¼ 0 ) x ¼ d 3 28
Of the two solutions, x ¼ 2.3d and x ¼ 8.3d, only the positive corresponds to the equal
direction of the two fields.
90. The Earth’s magnetic field is produced by two dipoles of equal moment and
polarity that are at an angle of 60 to each other, with the bisector being the axis of
rotation. The dipoles are contained in the plane of the 0 geographical meridian.
(a) Calculate the potential on points of the Earth’s surface as a function of geograph-
ical coordinates f and l.
(b) At what points on the surface are the magnetic poles located?
(c) What form would the external field have in order to annul the internal field at the
magnetic equator?
157 External magnetic field
30° GNP
GMNP2 GMNP1
q2
q1
Fig. 90
(a) The total potential at a point is the sum of the potentials of the two dipoles (see
Equation 80.1 of Problem 80):
M cos y1 M cos y2 Mðcos y1 þ cos y2 Þ
F ¼ F1 þ F2 ¼ ¼
r2 r2 r2
We calculate the geomagnetic co-latitudes by Equation (71.3) of Problem 71:
The geographical coordinates of the North Pole of each dipole are given by (Fig. 90)
Br ¼ 0
pffiffiffi
3M
Bf ¼
r3
Bl ¼ 0
The external magnetic field to annul out the internal field is therefore
pffiffiffi
3M
Be ¼ 0; 3 ; 0
r
91. The Earth’s magnetic field is formed by a centred dipole with northern geomag-
netic pole at 60 N, 0 E and B0 ¼ 32 000 nT and a uniform external field from the Sun
of 10 000 nT parallel to the equatorial plane.
(a) For a point at coordinates 60 N, 60 W, calculate the components X, Y, Z of the
total field, and the values of D and I.
(b) How do D and Z of the total field vary throughout the day with local time t?
159 External magnetic field
The external field is parallel to the equatorial plane and has a diurnal period (o ¼ 2p/24)
because it comes from the Sun. We assume that at local time t ¼ 0 the Sun is at the point’s
meridian. If we denote by N the modulus of the external field (N ¼ 10 000 nT) and bearing
in mind Fig. 91a (representation of the plane parallel to the equator that contains the point)
we have at time t 6¼ 0
Y ¼ Bel ¼ N sin ot
(a)
P
N coswt
B eλ
wt
Fig. 91a
160 Geomagnetism
GNP
X = Bf
Ncosωt
P
z = –Br
r
Fig. 91b
(b) To see how D and Z vary during the day with local time t we substitute several
values for t, obtaining the values in the table:
0 60976 41
6 55976 28
12 50976 83
18 55976 73
161 External magnetic field
92. The Earth’s magnetic field is formed by a centred dipole of moment m and
north pole 60 N, 0 E, and a uniform external field of magnitude N ¼ B0/4 (B0 is
the geomagnetic constant of the internal field) parallel to the axis of rotation.
Determine:
(a) The total potential at any point.
(b) The coordinates of the boreal magnetic pole.
(c) The magnetic declination at the point 45 N, 45 E.
(d) The angle along the meridian between that point and the magnetic equator.
(a) The total potential (F) is the sum of two contributions: the main (internal) field
(Fi) and the external field (Fe):
F ¼ Fi þ Fe
The main field is formed by a centred dipole of moment m so the potential is given by
Cm cos y
Fi ¼
r2
We calculate the geomagnetic co-latitude y ¼ 90 – f at a point with geographical
coordinates (f, l) by (71.3):
pffiffiffi
3 1
cos y ¼ sin fB sin f þ cos fB cos f cosðl lB Þ ¼ sin f þ cos f cos l
2 2
GNP
P
ze
Xe
f Fe
Fig. 92a
162 Geomagnetism
The external field is parallel to the axis of rotation; therefore its components are in the
vertical and NS directions (Fig. 92a). If we call # ¼ 90 f the geographical co-latitude,
the components are given by
Bearing in mind that Be ¼ ∇Fe the potential for the external field is
Fe ¼ Nr cos # ¼ Nr sin f
Therefore the total potential is given by
pffiffi
Cm 23 sin f þ 12 cos f cos l
F¼ þ Nr sin f
r2
(b) At the magnetic boreal pole the inclination is I ¼ 90º and the horizontal field
components H, X, Y are given by
Z
tan I ¼ )H ¼0
H
X ¼0
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
H ¼ X þY ) 2
Y ¼0
We obtain the components X and Y by taking the gradient of the potential
Be ¼ rFe
We bear in mind the relations
@F @F
¼
@y @f
X ¼ By
and obtain that the components are given by
pffiffi
3 1
1 @F Cm 2 cos f 2 sin f cos l
X ¼ ¼ N cos f ¼ 0
r @f r3 ð92:1Þ
1 @F Cm
Y ¼ ¼ 3 sin l ¼ 0 ) l ¼ 0
r cos f @l 2r
Substituting in the equation of component X (92.1) the values
l¼0
r¼a
Cm
B0 ¼
a3
N ¼ B0 =4
163 External magnetic field
GNP GMNP
fB
f
l
feq
Fig. 92b
we obtain the geographical latitude of the boreal magnetic pole
pffiffiffi
2 31
tan fBM ¼ ) fBM ¼ 51
2
Therefore the geographical coordinates of the magnetic boreal pole are 51 N, 0 . Notice
that this is different from the geomagnetic pole.
(c) The declination is given by
B0
Y sin l
tan D ¼ ¼ pffiffiffi 2
X
3 1 B0
B0 cos f sin f cos l cos f
2 2 4
sin l
¼ pffiffiffi
1
3 cos f sin f cos l
2
Substituting the geographical coordinates of the point (45 N, 45 E) we obtain
D ¼ 62
(d) We call feq the angle between the geographical and magnetic equators at longitude
l ¼ 45º (Fig. 92b). Then the angle to calculate will be feq þ 45 .
To calculate feq we take into account that at the magnetic equator the vertical component is Z ¼ 0:
pffiffiffi
@F 2CM 3 1
Z ¼ Br ¼ ¼ 3 sin feq þ cos feq cos l þ N sin feq ¼ 0
@r r 2 2
4 cos l
tan feq ¼ pffiffiffi
4 3þ1
164 Geomagnetism
feq ¼ 20
M cos y1 M cos y2
F ¼ F1 þ F2 þ Fe ¼ þ Fe
r2 r2
Mðcos y1 þ cos y2 Þ
¼ þ Fe
r2
In this equation r is the distance from the dipole’s centre (the Earth’s centre), y1 and y2 are
the co-latitudes relative to each dipole, and M ¼ Cm.
Dipole 1 is on the direction of the axis of rotation and so the geomagnetic co-latitude of
the point with respect to this dipole is
y1 ¼ 90 f
cos y1 ¼ sin f
Dipole 2 is in the equatorial plane (fB2 ¼ 0) and contained in the 0 meridian so the
geomagnetic co-latitude y2 is given by (71.3)
fB2 ¼ 0
lB2 ¼ 0
Fe ¼ þNr sin f
(b) We obtain the component Z by taking the vertical component of the gradient of the
potential (B ¼ ∇F)
@F 2M ðsin f þ cos f cos lÞ
Z ¼ Br ¼ ¼ þ N sin f
@r r3
To calculate the maximum of Z on the 0 meridian we substitute l ¼ 0 and apply the
condition that the first derivate with respect to the latitude is null:
@Z 2M
¼ 3 ðcos f sin fÞ þ N cos f ¼ 0
@f r
At the Earth’s surface r ¼ a and we know that
M
B0 ¼
a3
Substituting this constant and solving the equation we determine the latitude at which the
Z component is maximum:
B0
2B0 ðcos f sin fÞ þ cos f ¼ 0 ) f ¼ 48
4
94. The internal field has its northern geomagnetic pole at the coordinates 60 N, 0 E,
and B0 ¼ 30 000 nT. At a point with coordinates 30 N, 45 W, one observes an
increase of 7.7 in the value of the declination from 00:00 h to 09:00 h. There is
known to be an external field parallel to the Earth’s axis of rotation in the direction
from N to S which is null at 00:00 h and maximum at 12:00 h local times. Calculate:
(a) The components of the internal and external fields.
(b) The difference in the inclination at 00:00 h and 09:00 h.
(c) The maximum value of the declination during the day.
(a) To calculate the geomagnetic main field intensity components we obtain first the
geomagnetic latitude (Equation 71.3):
cos fB sinðl lB Þ
sin D ¼ ) D ¼ 31:7
cos f
tan I ¼ 2 tan f ) I ¼ 65:5
The external field is parallel to the axis of rotation so its components are in the vertical and
NS directions. This field is null at 00:00 h and maximum at 12:00 h local time (period T ¼
24 h). Its components are given by
N
Ze ¼ ð1 cos otÞ sin f
2
N
X e ¼ ð1 cos otÞ cos f
2
Ye ¼ 0
He ¼ X e
2p 2p
o¼ ¼
T 24
(b) To calculate the difference in the inclination we obtain first the value of N, bearing
in mind the time variation of the declination. The observed declination as a
function of time is given by
Y Y þ Ye Y
tan D ¼ ¼ ¼
X X þ Xe N
X ð1 cos otÞ cos f
2
For t ¼ 0 h:
Y
tan D1 ¼ ¼ tan D ) D1 ¼ D ¼ 31:7
X
Since we know the change in declination between 0 h and 9 h, we find the declination at
9 h, D2:
D2 D1 ¼ 7:7 ) D2 ¼ 39:4
We know that at t ¼ 9 h
Y
tan D2 ¼ ¼ 0:82
N
X ð1 cos otÞ cos f
2
Solving for N we obtain
Y
2
N¼ X ¼ 5766 nT
ð1 cos otÞ cos f tan D2
The magnetic inclination is given by
N
Z Z þ Ze Z þ ð1 cos otÞ sin f
tan I ¼ ¼ ¼ 2
H H þ He
N
H ð1 cos otÞ cos f
2
At t ¼ 0 h:
I1 ¼ I ¼ 65:5
At t ¼ 9 h:
I2 ¼ 71:2
167 External magnetic field
I2 I1 ¼ 5:7
(c) The declination is given by
Y
tan D ¼ ð94:1Þ
N
X ð1 cos otÞ cos f
2
The maximum value is at 12 h because at that time the external field has the maximum
value. Substituting ot ¼ p and the values obtained for X , Y , N, and f:
Dmax ¼ 41:0
95. At a point on the Earth with coordinates 45 N, 45 E, measurements are made of
the magnetic field components at 00:00 h and 12:00 h in nT with a 2 nT precision:
0h X ¼ 20 732 Y ¼ 2500 Z ¼ 57 768
12 h X ¼ 24 267 Y ¼ 2500 Z ¼ 54 232
It is known that the modulus of the magnetic field intensity has a harmonic diurnal
variation, and that the geomagnetic pole is on the zero meridian. Calculate:
(a) The moment and coordinates of the main field dipole.
(b) Expressions for the potential and components of the external field.
(Earth’s radius a ¼ 6400 km, m0 ¼ 4p107 kg m s2 A2).
(a) To calculate the moment and coordinates of the main field dipole we need to obtain the
geomagnetic main field intensity components. The observed values are equal to the
sum of the geomagnetic main field (X , Y , Z ) and the external field (X e, Y e, Z e):
X ¼ X þ Xe
Y ¼ Y þ Ye
Z ¼ Z þ Ze
The geomagnetic main field is constant but the external field changes with time. So if we
denote by (X0, Y0, Z0) and (X12, Y12, Z12) the observed values at 0 h and at 12 h respectively
then the differences are due to the variations of the external field:
e
X12 X0 ¼ 3535 nT ¼ X12 X0e
e
Y12 Y0 ¼ 0 nT ¼ Y12 Y0e
e
Z12 Z0 ¼ 3536 nT ¼ Z12 Z0e
We notice that the Y component doesn’t vary, which implies that the component Y e is zero,
and so the external field is parallel to the axis of rotation. We also notice that the NS
component increases in the time interval between 0 h and 12 h, while the vertical
component diminishes, which implies that the polarity of the external field is inverted
with respect to that of the main field.
The modulus of the magnetic field intensity has a harmonic diurnal variation and
increases with time. Therefore the components of the external field are (Fig. 95a)
168 Geomagnetism
GNP Fe
Xe Ze
Fig. 95a
X ¼ X12 X12
e
¼ 20 732 nT
Y ¼ Y12 Y12
e
¼ 2500 nT
Z ¼ Z12 Z12 e
¼ 57 768 nT
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
H ¼ ðX Þ2 þ ðY Þ2 ¼ 20 882 nT
If we consider the centred magnetic dipole model, the vertical and horizontal field
components are given by the equations
Z ¼ 2B0 sin f
H ¼ B0 cos f
169 External magnetic field
GNP
90º – fB
l – lB
180º – l ∗ 90º – f
GMNP
q = 90º – f ∗ D∗
Fig. 95b
From these equations we obtain the geomagnetic latitude (f ) and the geomagnetic
constant B0:
Z
tan f ¼ ) f ¼ 54:1
2H
Z
B0 ¼ ¼ 35 657 nT
2 sin f
From B0 we calculate the magnetic moment:
m0 m
B0 ¼ ) m ¼ B0 107 a3 ¼ 9:3 1022 A m2
4p a3
The longitude of the Geomagnetic North Pole is lB ¼ 0 and we calculate the latitude fB
from the spherical triangle (Fig. 95b), but obtaining first the declination from the X and
Y components:
Y
tan D ¼ ) D ¼ 6:9
X
Applying the cosine rule:
(b) We obtain the radial and transverse components of the external field from the
vertical and NS components:
@Fe
Ber ¼ Z e ¼ N ð1 cos otÞ sin f ¼
@r
e e 1 @Fe
Bf ¼ X ¼ N ð1 cos otÞ cos f ¼
r @f
170 Geomagnetism
+ GNP
P
q
q2
Fig. 96
Cme cosðy þ eÞ
Fe ¼
r2
and the total potential is given by
M cos y Cme cosðy þ eÞ
FT ¼ þ
r2 r2
We calculate the components of the main field intensity by taking the gradient of the
potential:
@F 2M cos y
Br ¼ ¼
@r r3
1 @F M sin y
By ¼ ¼
r @y r3
The magnitude of the external field is given by
m0 J
Be ¼ ¼N
4 10a
The radial and tangential components are
Ber ¼ N cosðy þ eÞ
Bey ¼ N sinðy þ eÞ
172 Geomagnetism
B er
q – 90° +
r
B eθ
q – 90°
a a
4a
θ
Fig. 97
pffiffiffiffiffi
At the North Pole r ¼ 17a ¼ 4:12a and we calculate the geomagnetic co-latitude y by
(Fig. 97)
a 1
tanðy 90 Þ ¼ ¼
4a 4
y ¼ 14 þ 90 ¼ 104
Substituting these values in the equations for the radial and tangential components with
respect to the planet producing the external field:
2Cm cos y Cm cos y B0 cos y
Ber ¼ ¼ ¼ ¼ 69 nT
r3 35a3 35
Cm sin y Cm sin y B0 sin y
Bey ¼ ¼ ¼ ¼ 139 nT
r3 70a3 70
From this value we calculate the vertical and horizontal components (Fig. 97):
Z e ¼ Ber cosð180 yÞ þ Bey cosðy 90 Þ ¼ Ber cos y þ Bey sin y ¼ 118 nT
H e ¼ Ber sinð180 yÞ þ Bey sinðy 90 Þ ¼ Ber sin y Bey cos y ¼ 33 nT
X e ¼ H e ¼ 33 nT
Ye ¼ 0
The components of the total field are finally
Z ¼ Z þ Z e ¼ 19882 nT
H ¼ H þ H e ¼ 33 nT
X ¼ X þ X e ¼ 33 nT
Y ¼ Y þ Ye ¼ 0
Y
tan D ¼ ) D ¼ 0
X
Z
tan I ¼ ) I ¼ 89:9
H
174 Geomagnetism
98. At a point with geographical coordinates 40 N, 45 E, measurements are made of
the magnetic field components, obtaining the values (in nT):
At 06:00 h: X ¼ 19 204; Y ¼ 0; Z ¼ 38 195
At 12:00 h: X ¼ 11 544; Y ¼ 0; Z ¼ 44 623
Buried at a depth of 20 m below this point is a dipole of magnetic moment Cm ¼ 0.01
T m3, oriented in the NS plane at an angle of 45 with the vertical towards the south,
and the positive pole upwards. Given that the external field at 12:00 h is twice that at
06:00 h, determine:
(a) The geomagnetic constant Bo and the coordinates of the northern geomagnetic pole.
(b) The magnitude and direction of the external field. How does the magnitude of the
external field vary with time?
(a) The observed magnetic field is composed of three parts: the geomagnetic main
(internal) field, the anomalous field (magnetic anomaly) created by the buried
dipole, and the external field.
We determine first the magnetic anomaly produced by the buried dipole, applying Equa-
tions (82.2) and (82.3) in Problem 82, substituting a ¼ 225º and x ¼ 0. The horizontal
component is in the NS direction (DX), because the buried dipole is in the NS-vertical
plane (Fig. 98a):
2Cm cos a
Z ¼ ¼ 1768 nT
d3
Cm sin a
X ¼ ¼ 884 nT
d3
Y ¼ 0
P
N
d
+
45°
Fig. 98a
175 Main (internal), external, and anomalous magnetic fields
X1 ¼ X þ X þ X e ¼ 19 204 nT
Y1 ¼ Y þ Y þ Y e ¼ 0
Z1 ¼ Z þ Z þ Z e ¼ 38 195 nT
X2 ¼ X þ X þ 2X e ¼ 11 544 nT
Y2 ¼ Y þ Y þ 2Y e ¼ 0
Z2 ¼ Z þ Z þ 2Z e ¼ 44 623 nT
X2 X1 ¼ X e ¼ 7660 nT
Y2 Y1 ¼ Y e ¼ 0
Z2 Z1 ¼ Z e ¼ 6428 nT
X ¼ X1 X X e ¼ 25 980 nT
Y ¼ Y1 Y Y e ¼ 0
Z ¼ Z1 Z Z e ¼ 33 535 nT
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
H ¼ ðX Þ2 þ ðY Þ2 ¼ X ¼ 25 980 nT
Y
tan D ¼ ) D ¼ 0
H
We calculate the geomagnetic latitude of the point f and the geomagnetic constant B0
from the vertical and horizontal geomagnetic main field components by
Z ¼ 2B0 sin f
H ¼ B0 cos f
Z
tan f ¼ ) f ¼ 32:8
2H
Z
B0 ¼ ¼ 30 953 nT
2 sin f
We obtain the coordinates of the Geomagnetic North Pole by (Fig. 98b)
GNP
GMNP
f∗
f
90° – fB
Fig. 98b
The direction of the external field is in the NS-vertical plane because the EW component is
null, forming with the horizontal an angle Ie (Fig. 98c). This direction is the same at 06:00
h and at 12:00 h. We calculate the angle Ie by
Ze
tan Ie ¼
Xe
Ie ¼ 40
Xe
N
Ie
Ze
Be
Fig. 98c
177 Main (internal), external, and anomalous magnetic fields
Because at
06 : 00 h ðt ¼ p=2Þ ! N ¼ 10 000
12 : 00 hðt ¼ pÞ ! N ¼ 20 000
the variation of the magnitude of the external field with time is given by
N ¼ 10 000ð1 cos tÞ
99. Buried at a depth of 100 m at a point with geographical coordinates 45 N, 45 W is
a dipole anomaly of Cm ¼ 0.1 T m3, inclined 45 from the horizontal northwards in the
vertical plane with the negative pole downwards. Measurements gave the following
results (in nT):
09:00 h X ¼ 27 759; Y ¼ 0; Z ¼ 30 141
12:00 h X ¼ 28 052; Y ¼ 0; Z ¼ 30 141
Find:
(a) The coordinates of the magnetic dipole’s North Pole.
(b) The value of B0.
(c) An expression for the variation Sq knowing that it is zero at 00:00 h and
maximum at 12:00 h.
(a) As in Problem 98, the observed field is the result of three parts: the main (internal)
field, the buried dipole field, and the external field. To calculate the coordinates of
the magnetic dipole’s North Pole we need to obtain the components of the
geomagnetic main field from the components of the total field. With this aim we
begin by calculating the magnetic anomaly created by the buried dipole, applying
Equations (82.2) and (82.3), and substituting a ¼ 225º and x ¼ 0. The horizontal
component is in the NS direction (DX) given that the dipole is on the NS-vertical
plane (Fig. 99a).
2Cm cos a
Z ¼ ¼ 141 nT
d3
Cm sin a
X ¼ ¼ 71 nT
d3
Y ¼ 0
X1 ¼ X þ X þ X1e
Y1 ¼ Y þ Y þ Y1e ¼ 0 ) Y ¼ Y1e
Z1 ¼ Z þ Z þ Z1e
∆X P
N
∆Z
d
+
45° a
Fig. 99a
We assume that the time variation of the observations is due to the diurnal Sq variation
which is zero at 00:00 h and maximum at 12:00 h. Therefore the only possible values
for the components Ye and Ze are zero because these components have the same values at
09:00 h and at 12:00 h:
Y2e ¼ Y1e ¼ Y ¼ 0
Z2e ¼ Z1e ¼ 0
Then, the intensity of the external field is given by
X e ¼ N ð1 cos otÞ
2p
o¼
24
The NS components of this field at 09:00 h (X1) and at 12:00 h (X2) are
pffiffiffi
e 3p 2
X1 ¼ N 1 cos ¼N 1þ
4 2
X2e ¼ N ð1 cos pÞ ¼ 2N
Subtracting the two values and using Equation (99.1) we obtain
pffiffiffi
pffiffiffi
e
e e 2 2 X2
X2 X1 ¼ 1 N ¼ 1 ¼ 293 nT ) X2e ¼ 2001 nT
2 2 2
X1e ¼ 1708 nT
179 Main (internal), external, and anomalous magnetic fields
GNP
GMNP
f∗
f
90° – fB
Fig. 99b
X ¼ X1 X X1e ¼ 26 122 nT
Y ¼ Y1 Y Y1e ¼ 0
Z ¼ Z1 Z Z e ¼ 30 000 nT
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
H ¼ ðX Þ2 þ ðY Þ2 ¼ X ¼ 26 122 nT
Y
tan D ¼ ) D ¼ 0
H
The geomagnetic latitude of the point f is determined from the vertical and horizontal
geomagnetic main field components:
Z ¼ 2B0 sin f
H ¼ B0 cos f
Z
tan f ¼ ) f ¼ 29:9
2H
We obtain the coordinates of the Geomagnetic North Pole by (Fig. 99b)
D ¼ 0 ) lB ¼ 180 þ l ¼ 135 E
90 fB ¼ f f ) fB ¼ 74:9
180 Geomagnetism
H ¼ B0 cos f
H
B0 ¼ ¼ 30 133 nT
cos f
(c) We have obtained above that
F e ¼ X e ¼ N ð1 cos otÞ
To calculate N we take into account that the Sq variation is maximum at 12:00 h:
X2e ¼ 2N ) N ¼ 1000 nT
HT ¼ FT cos IT ¼ 21 390 nT
ZT ¼ FT sin IT ¼ 43 280 nT
XT ¼ HT cos DT ¼ 21 363 nT
YT ¼ HT sin DT ¼ 1082 nT
XT
X
YT DT
HT
ZT IT
Y
FT
Fig. 100a
181 Main (internal), external, and anomalous magnetic fields
P
X60
– +
Fig. 100b
The observed field is the result of three parts: the main field (X , Y , Z ), the buried dipole
field (DX, DY, DZ), and the external field (Xe, Ye, Ze):
XT ¼ X þ X þ X e
YT ¼ Y þ Y þ Y e
ZT ¼ Z þ Z þ Z e
We determine the magnetic anomaly created by the buried dipole from Equations (82.2)
and (82.3) substituting a ¼ 90º and x ¼ 0. If we call X60 the direction N 60º E the horizontal
component of this anomaly is DX60 (Fig. 100b):
2Cm cos a
Z ¼ ¼ 0 nT
d3
Cm sin a
X60 ¼ ¼ 1250 nT
d3
The NS and EW components will be given by (Fig. 100c)
X ¼ X60 cos 60 ¼ 625 nT
Y ¼ X60 sin 60 ¼ 1083 nT
The external field is parallel to the Earth’s axis of rotation in the southwards direction so its
components are in the vertical and NS direction and are given by (Fig. 100d)
Z e ¼ 1000 sin f ¼ 866 nT
X e ¼ 1000 cos f ¼ 500 nT
We calculate the main field components from these values:
X ¼ XT X X e ¼ 21 238 nT
Y ¼ YT Y Y e ¼ 1082 nT
Z ¼ ZT Z Z e ¼ 42 414 nT
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
H ¼ ðX Þ2 þ ðY Þ2 ¼ 21 266 nT
Y
tan D ¼ ) D ¼ 2:9
X
182 Geomagnetism
60°
∆Y
E
∆X
∆X60
Fig. 100c
GNP
P
Ze
Xe
f Fe
Fig. 100d
The geomagnetic latitude of the point f and the geomagnetic constant B0 are found from
the vertical and horizontal geomagnetic main field components by
Z ¼ 2B0 sin f
H ¼ B0 cos f
Z
tan f ¼ ) f ¼ 44:9
2H
Z
B0 ¼ ¼ 30 953 nT ¼ 30 044 nT
2 sin f
183 Main (internal), external, and anomalous magnetic fields
GNP
90º – fB
l – lB
90º – f
180º – l∗
GMNP
q = 90º – f∗
D∗
Fig. 100e
To calculate the geographical coordinates of the Geomagnetic North Pole we use the
corresponding spherical triangle (Fig. 100e). We obtain the latitude fB by applying the
cosine law for the angle 90º – fB:
cosð90 fB Þ ¼ cosð90 f Þ cosð90 fÞ þ sinð90 f Þ sinð90 fÞ cos D
sin fB ¼ sin f sin f þ cos f cos f cos D
fB ¼ 75:0
We obtain the longitude lB by applying the cosine rule for the angle 90º – f :
101. At a point with geographical coordinates 30 N, 30 E, the observed geomagnetic
field components are (in nT): X ¼ 15 364, Y ¼ 7660, Z ¼ 48 980. The northern
geomagnetic pole is at 60 N, 0 E, and B0 ¼ 30 000 nT. There is also a constant external
magnetic field normal to the equatorial plane, with a southwards direction, of 1000 nT
intensity. Buried 10 m below the observation point is a magnetic dipole. Calculate:
(a) The magnetic anomalies DX, DY, DZ, DH, DF.
184 Geomagnetism
(b) The orientation and magnetic moment (Cm, in nT m3) of the buried dipole.
(a) The observed values are the sum of the geomagnetic main field, the magnetic field
due to the buried dipole, and the external field:
X ¼ X þ X þ X e
Y ¼ Y þ Y þ Y e
Z ¼ Z þ Z þ Z e
To obtain the magnetic anomalies from these equations we first calculate the geomagnetic
latitude and the vertical and horizontal components of the geomagnetic main field by (71.3):
X ¼ H cos D ¼ 16 007 nT
Y ¼ H sin D ¼ 7498 nT
The external field is parallel to the axis of rotation directed southwards so its components
are in the vertical and NS direction and are given by (Fig. 101a)
X ¼ X X X e ¼ 223 nT
Y ¼ Y Y Y e ¼ 162 nT
Z ¼ Z Z Z e ¼ 1 nT
H ¼ X cos D þ Y sin D ¼ 271 nT
F ¼ H cos I þ Z sin I ¼ 94 nT
(b) We call b the angle between the geographical north and the buried dipole direc-
tions. Then using Equations (82.2) and (82.3) we obtain
2Cm cos a
Z ¼ ð101:1Þ
d3
185 Main (internal), external, and anomalous magnetic fields
GNP
P
Ze
Xe
f Fe
Fig. 101a
Cm sin a
X ¼ cos b ð101:2Þ
d3
Cm sin a
Y ¼ sin b ð101:3Þ
d3
To solve this system of three equations in three unknowns (Cm, a, b) we divide Equation
(101.3) by (101.2):
Y
tan b ¼ ) b ¼ 36 þ 180 ¼ 144
X
This value of the angle b implies that the dipole is oriented in the N 144º E direction.
To calculate the angle a between the buried dipole and the vertical we divide Equation
(101.2) by (101.1):
X 1
¼ tan a cos b
Z 2
2 X
tan a ¼ ) a
90
cos b Z
Therefore the dipole is practically horizontal (Fig. 101b) and this explains the small value
of the vertical component DZ.
Finally we calculate the magnetic moment of the buried dipole from Equation (101.2):
d3
Cm ¼ X ¼ 2:8 105 nT m3
sin a cos b
186 Geomagnetism
P
N 144° E
a
– +
Fig. 101b
H ¼ F cos I ¼ 17 476 nT
Z ¼ F sin I ¼ 49 352 nT
X ¼ H cos D ¼ 15 707 nT
Y ¼ H sin D ¼ 7661 nT
The magnetic anomalies are found by subtracting from the observed values the main and
external field contributions:
X ¼ X X X e
Y ¼ Y Y Y e
Z ¼ Z Z Z e
First we determine the geomagnetic latitude and declination using (71.3) and (71.2):
sin f ¼ sin fB sin f þ cos fB cos f cosðl lB Þ
f ¼ 53:9
cos fB sinðl lB Þ
sin D ¼ ) D ¼ 25:1
cos f
The vertical and horizontal components are given by
Z ¼ 2B0 sin f ¼ 48 479 nT
H ¼ B0 cos f ¼ 17 676 nT
X ¼ H cos D ¼ 16 007 nT
Y ¼ H sin D ¼ 7498 nT
187 Main (internal), external, and anomalous magnetic fields
GNP
P
60°
Xe
Ze
f Fe
Fig. 102a
60°
E
f
Xe
Ze
Fe
Fig. 102b
The external field is on the plane containing the vertical and NS directions (Figs 102a
and 102b):
X ¼ X X X e ¼ 200 nT
Y ¼ Y Y Y e ¼ 163 nT
Z ¼ Z Z Z e ¼ 7 nT
(b) We call b the angle between the positive pole of the buried dipole and the geographical
north. Then applying Equations (82.2) and (82.3) we obtain
2Cm cos a
Z ¼ ð102:1Þ
d3
Cm sin a
X ¼ cos b ð102:2Þ
d3
Cm sin a
Y ¼ sin b ð102:3Þ
d3
(a) The components of the total field are the sum of the geomagnetic main field and
the external field:
X ¼ X þ Xe
Y ¼ Y þ Ye
Z ¼ Z þ Ze
Let us first calculate the geomagnetic latitude, declination, and inclination using (71.3)
and (71.2):
sin f ¼ sin fB sin f þ cos fB cos f cosðl lB Þ
f ¼ 75:1
cos fB sinðl lB Þ
sin D ¼ ) D ¼ 76:5
cos f
tan I ¼ 2 tan f ) I ¼ 82:4
The external field is parallel to the equatorial plane and to the zero meridian. If its
magnitude is N ¼ 10 000 nT, its components, from Fig. 103a (plane through the point
parallel to the equator) and Fig. 103b (plane through the geographical meridian of the
point), are given by
Z e ¼ Ber ¼ N cos l cos ’ ¼ 4330 nT
X e ¼ Bef ¼ N cos l sin ’ ¼ 7500 nT
Y e ¼ Bel ¼ N sin l ¼ 5000 nT
(b) The buried vertical dipole is in the direction of the compass, that is, in the direction
of the magnetic north (Fig. 103c). We calculate the anomaly DZ produced at
P using Equations (82.2) and (82.3) substituting a ¼ 0º and x ¼ 30:
B eλ = Nsinl
P
Be
Ncosλ
Plane parallel
to the equator
Fig. 103a
GNP
B ef
Ncosλ P
Ber
Fig. 103b
191 Main (internal), external, and anomalous magnetic fields
P x
NM
–
d
Fig. 103c
104. The Geomagnetic North Pole is at 60 N, 150 W, with B0 ¼ 30 000 nT, and there
is an external magnetic field of intensity 3000 nT parallel to the axis of rotation
pointing away from the North Pole. Buried 10 m below a point with coordinates
30 N, 30 E there is a horizontal dipole with Cm ¼ 40 000 nT m3 and the negative
pole pointing in the direction N 45 E.
(a) What are the components X, Y, Z of the total field?
(b) Calculate the total field anomaly DF.
(c) What is the angle between the direction of the compass and geographic north?
(a) The components of the total field are the sum of the geomagnetic main field, the
magnetic field due to the buried dipole, and the external field:
X ¼ X þ X þ X e
Y ¼ Y þ Y þ Y e
Z ¼ Z þ Z þ Z e
We determine first the geomagnetic latitude, declination, and inclination using (71.3)
and (71.2):
cos fB sinðl lB Þ
sin D ¼ ) D ¼ 0
cos f
tan I ¼ 2 tan f ) I ¼ 0
192 Geomagnetism
P
X45
a
+ –
Fig. 104
The external field is parallel to the axis of rotation directed southwards, so its components
are in the vertical and NS direction and are given by
F ¼ X ¼ 28 nT
(c) The direction of the compass is affected by the three fields. The angle D between
the direction of the compass and the geographic north is obtained from the
horizontal components of the total observed field
Y 28
tan D ¼ ¼ ) D ¼ 0:06
X 27430
105. The geomagnetic field is that of a dipole in the direction of the axis of rotation
and B0 ¼ 30 000 nT. There is also a constant external field of 2500 nT normal to the
equatorial plane in the direction of the South Pole.
(a) Calculate the value of the inclination observed at a point P with coordinates 45 N,
45 E given that, at 10 m below it, there is a vertical dipole with the negative pole
upwards and moment Cm ¼ 40 000 nT m3.
(b) For a point 20 m north of P, calculate the observed inclination and declination,
and the total field anomaly DF.
(a) The components of the total observed field are the sum of the geomagnetic main
field, the magnetic field due to the buried dipole, and the external field:
X ¼ X þ X þ X e
Y ¼ Y þ Y þ Y e
Z ¼ Z þ Z þ Z e
The magnetic dipole is oriented in the direction of the axis of rotation and therefore
f ¼ f ¼ 45
D ¼ 0
P
X
–
d
Fig. 105a
GNP
P
Ze
Xe
f Fe
Fig. 105b
The external field is parallel to the axis of rotation directed southwards so its components
are in the vertical and NS direction (Fig. 105b) and are given by
Z e ¼ 2500 sin f ¼ 1768 nT
X e ¼ 2500 cos f ¼ 1768 nT
Ye ¼ 0
195 Main (internal), external, and anomalous magnetic fields
X ¼ X þ X þ X e ¼ 19 445 nT
Y ¼ Y þ Y þ Y e ¼ 0
H ¼X
Z ¼ Z þ Z þ Z e ¼ 44 274 nT
X ¼ X þ X þ X e ¼ 19 441 nT
Y ¼ Y þ Y þ Y e ¼ 0
H ¼X
Z ¼ Z þ Z þ Z e ¼ 44 193 nT
From these values we calculate the observed inclination and declination by the expressions
Z
tan I ¼ ) I ¼ 66:2
H
Y
tan D ¼ ) D ¼ 0
X
The total field anomaly DF is given by
F ¼ X cos I þ Z sin I ¼ 4 cos 66 1 sin 66 ¼ 3 nT
106. Consider a point with coordinates 30 N, 30 E under which is buried at a depth
of 100 m a horizontal dipole of moment m0 m/4p ¼ 1 T m3, with the positive pole in
the direction N 60 E. The terrestrial field is formed by a centred dipole in the
direction of the axis of rotation and a constant external field of 10 000 nT from the
Sun, B0 ¼ 30 000 nT.
(a) Calculate F, D, and I at that point on December 21 at 12 noon.
196 Geomagnetism
X ¼ X þ X þ X e
Y ¼ Y þ Y þ Y e
Z ¼ Z þ Z þ Z e
We first calculate the geomagnetic latitude and the declination and inclination. The
magnetic dipole is oriented in the direction of the axis of rotation and therefore
f ¼ f ¼ 30
D ¼ 0
We calculate the magnetic anomaly produced by the buried dipole from Equations (82.2)
and (82.3) substituting a ¼ 90º and x ¼ 0. We call X60 the direction N 60º E and DX60 the
horizontal component of the anomaly (Fig. 106a):
2Cm cos a
Z ¼ ¼ 0 nT
d3
Cm sin a
X60 ¼ ¼ 1000 nT
d3
P
X 60
a
– +
z
Fig. 106a
197 Main (internal), external, and anomalous magnetic fields
60°
∆Y
E
∆X
∆X60
Fig. 106b
GNP
N = Be
P
f +e
f
e
Sun
Fig. 106c
The external field has a diurnal period (o ¼ 2p / 24 h) and we know that the Sun is at the
meridian point at 12:00 h (solar time). This field changes through the year because the Sun
moves on the ecliptic plane (apparent motion) which is tilted with respect to the equatorial
plane by an angle e ¼ 23º. Therefore the solar declination (d), the angle from the Sun to the
celestial equator, changes through the year. On December 21 (winter solstice) this angle is
d ¼ e ¼ 23º (Fig. 106c). If we call N the magnitude of the external field (N ¼ 10 000nT)
its components on December 21 at 12:00 are (Fig. 106d)
GNP
X = Bf
N
P
f+e
Z = – Br
Fig. 106d
X ¼ X þ X þ X e ¼ 25 481 nT þ N sinðf dÞ
Y ¼ Y þ Y þ Y e ¼ 866 nT
Z ¼ Z þ Z þ Z e ¼ 30 000 nT þ N cosðf dÞ
107. The internal field of the Earth corresponds to a centred dipole with the negative
pole in the northern hemisphere at coordinates 80 N, 130 W and B0 ¼ 30 000 nT.
There is a uniform external field from the Sun of 1000 nT. Buried at 500 m depth
under a point P with geocentric coordinates 40 N, 50 E there is a positive magnetic
pole of strength CP ¼ 0.5 T m². Calculate:
(a) The total field components X, Y, and Z, and the magnetic and geomagnetic
declination at P on March 21 at 12:00 h.
199 Main (internal), external, and anomalous magnetic fields
(b) The same parameters for a point 200 m north of P, assuming that neither the
internal nor the external fields change (precision 1 nT).
(a) The components of the intensity of the total field are the sum of the geomagnetic
main field, the external field, and the magnetic field due to the buried pole.
To calculate the main field we determine first the geomagnetic latitude and the declination
by (71.3) and (71.2):
cos fB sinðl lB Þ
sin D ¼ ) D ¼ 0
cos f
From these values we obtain the vertical and horizontal components of the geomagnetic
main field:
Z ¼ 2B0 sin f ¼ 30 000 nT
H ¼ B0 cos f ¼ 25 981 nT
X ¼ H cos D ¼ 25 981 nT
Y ¼ H sin D ¼ 0
To calculate the external field we notice that it comes from the Sun which on March 21
(spring equinox) is on the equatorial plane so that the external field is parallel to this plane.
In addition this field changes during the day as a function of local time t with a diurnal
period (o ¼ 2p/24 h). At t ¼ 12 h, the external field is maximum given that at this time
the Sun is at the meridian point (Fig. 107a). Calling N its magnitude (N ¼ 1000 nT), the
components of the external field are given by
GNP
Xe
N P
Ze
Fig. 107a
200 Geomagnetism
The magnetic field anomaly created by the buried pole is derived from its potential DF,
given by
CP
F ¼
r
Applying the gradient, we obtain
B ¼ F
but the only component is the vertical:
@ðFÞ CP
Z ¼ Br ¼ ¼ 2
@r r
Substituting r ¼ d ¼ 500 m in this equation we obtain
Z ¼ 2000 nT
Therefore the components of the total field and the observed declination are given by
X ¼ X þ X e þ X ¼ 26 624 nT
Y ¼ Y þ Y e þ Y ¼ 0
Z ¼ Z þ Z e þ Z ¼ 28 766 nT
Y
tan D ¼ ) D ¼ 0
X
(b) The radial component of the magnetic field anomaly created by the buried pole for
a point 200 m north of P (x ¼ 200 m) is given by
@ðFÞ CP
Br ¼ ¼ 2
@r r
From Fig. 107b the vertical and NS components are
CP d CPd
Z ¼ Br cos a ¼ ¼
r2 r ðx2 þ d 2 Þ3=2
CP x CPx
X ¼ Br sin a ¼ 2 ¼
r r ðx þ d 2 Þ3=2
2
Y ¼ 0
Z ¼ 1601 nT
X ¼ 640 nT
Y ¼ 0
201 Paleomagnetism
x P
X
r a
d
α
Fig. 107b
The components of the observed total field and the declination are the sum of the three
contributions:
X ¼ X þ X e þ X ¼ 27 264 nT
Y ¼ Y þ Y e þ Y ¼ 0
Z ¼ Z þ Z e þ Z ¼ 29 165 nT
Y
tan D ¼ ) D ¼ 0
X
Paleomagnetism
108. At a point with geographical coordinates 60 N, 60 W a 1 cm3 sample was taken
of a rock with remanent magnetism, age 10 000 years, specific susceptibility
0.01 cm3. The magnetization components of the rock were:
X ¼ 40, Y ¼ 30, Z ¼ 50 nT (N, E, nadir).
The current field is B0 ¼ 30 000 nT and the geomagnetic pole coincides with the
geographical pole. Calculate:
(a) The coordinates of the virtual geomagnetic pole which corresponds to the
sample.
(b) The magnetic moment of the terrestrial dipole 10 000 years ago.
(c) The secular variation of F, D, and I in nT and minutes per year assuming that the
variation since that time has been constant.
202 Geomagnetism
GNP
90º – fB
l – lB
180º – l∗ 90º – f
VP
90º – f ∗ D∗
Fig. 108
(a) First we determine the declination D and the geomagnetic co-latitude y, corres-
ponding to the virtual pole, from the magnetization components of the rock X, Y,
and Z:
Y
tan D ¼ ) D ¼ 36:9
X
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
H ¼ X 2 þ Y 2 ¼ 50 nT
Z
tan I ¼ ) I ¼ 45
H
tan I ¼ 2 cot y ) y ¼ 63:4 ; fvirtual ¼ 26:6
Since at present the geomagnetic pole coincides with the geographical pole, the geograph-
ical latitude of the point coincides with the present geomagnetic latitude:
f ¼ fpresent ¼ 60 N
To determine the coordinates of the virtual Geomagnetic North Pole (VP), corresponding
to the magnetization of the rock, we solve the spherical triangle of Fig. 108 for ’B and lB
using the obtained values of y and D. The latitude fB applying the cosine rule is given by
To choose between the positive and negative solution we bear in mind that the declination
is negative and so the point is to the east of the virtual magnetic North Pole:
D < 0 ) l lB > 0
lB ¼ 173:6 E
(b) To obtain the magnetic moment we first calculate the constant B0. The susceptibility
w relates the magnetization and the magnetic field. If we call F the magnitude of the
paleomagnetic field and F0 the remanent magnetization, the relation between them is
F 0 ¼ wF ð108:1Þ
w ¼ 0:01
We calculate F0 from its components
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F 0 ¼ X 2 þ Y 2 þ Z 2 ¼ 71 nT
The field F of the virtual pole is given by
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F ¼ B0 1 þ 3 cos2 y
Substituting in Equation (108.1) we obtain
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F 0 ¼ wB0 1 þ 3 cos2 y
F0
B0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 5610 nT
w 1 þ 3 cos2 y
From this value we calculate the magnetic moment of the virtual pole taking a ¼ 6370 km
for the Earth’s radius
Cm
B0 ¼ ) m ¼ 1:45 1022 A m2
a3
(c) The magnetic field, the declination, and inclination 10 000 years ago were
F0
F¼ ¼ 7100 nT
w
D ¼ 36:9
I ¼ 45
At present the values of these parameters are
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F p ¼ Ba0 1 þ 3 sin2 fpresent ¼ 54 083 nT
D p ¼ 0
tan I p ¼ 2 tan f ) I p ¼ 73:9
204 Geomagnetism
Demagnetization
temperature (°C) Declination (D, °E) Inclination (I, °) NRM Intensity (J, mA/m1)
20 32 33 0.056
100 36 22 0.056
200 38 12 0.057
300 39 4 0.058
400 41 5 0.058
500 41 5 0.050
600 41 5 0.016
650 41 5 0.009
700 300 55 0.000
H ¼ J cos I
X ¼ H cos D
Y ¼ H sin D
Z ¼ J sin I
In Fig. 109 plotting X versus Y gives the projection of the demagnetization vector
onto the horizontal plane, while plotting X versus Z gives the projection onto the vertical
plane.
A stable component of NRM is represented by collinear points on the vector component
diagrams, so that two stable components can be identified in the range 20–300 C and in
the range 400–700 C.
The declination of a stable component is determined by measuring or by calculating the
angle between the north axis and the trajectory of the stable component in the horizontal
plane.
205 Paleomagnetism
0.06 0.06
0.05 0.05 γ
a
0.04 0.04
0.03 0.03
X
X
0.02 0.02
b d
0.01 0.01
0.00 0.00
0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 –0.01 0.00 0.01 0.02 0.03
Y Z
Fig. 109
1 Y400 0:0379
1
D ¼ b ¼ tan ¼ tan ¼ 41:0
X400 0:0436
Note: this value can be obtained directly from the declination of the observations between
400 and 650 C.
The apparent inclination, Iap, of a stable component is determined by measuring or by
calculating the angle between the north axis and the trajectory of the stable component in
the vertical plane. Iap is related to the true inclination, I, by:
Z400 0:0051
Iap ¼ d ¼ tan1 ¼ tan1 ¼ 6:7
X400 0:0436
I ¼ tan1 ðtanð6:7Þj cosð41ÞjÞ ¼ 5:0
Note: this value can be obtained directly from the inclination of the observations between
400 and 650 C.
Therefore the stable component isolated in the range 20–300 C has D ¼ 203.9 and
I ¼ 78.1 and the stable component isolated in the range 400–700 C has D ¼ 41.0 and
I ¼ 5.0 .
206 Geomagnetism
30 43
28 39
34 44
25 45
32 38
35 44
26 40
Use unit vector addition to calculate the mean direction of the primary remanence.
Calculate the direction cosines of each direction, the resultant total field vector, F, and
then the mean direction using:
X ¼ cos I cos D
Y ¼ cos I sin D
Z ¼ sin I
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!2ffi
u N !2 N
!2
N
u X X X
F¼t X þ i i Y þ i Z ¼ 6:98542
i¼1 i¼1 i¼1
N
P
Xi
i¼1
Xmean ¼ ¼ 0:6446;
F
N
P
Yi
Ymean ¼ i¼1 ¼ 0:37186;
F
PN
Zi
Zmean ¼ i¼1 ¼ 0:66799
F
Ymean
Dmean ¼ tan1 ¼ 30:0
Xmean
Imean ¼ sin1 ðZmean Þ ¼ 41:9
The mean direction of the primary remanence has a declination of 30.0 and an inclination
of 41.9 .
207 Paleomagnetism
Next, calculate the expected field direction at the site using the reference palaeomagnetic
pole. The first step is to determine y ¼ 90 f , (Equation 71.3), from the pole (fp, lp) to
the site (fs, ls) using spherical triangles:
Elasticity
111. Determine the principal stresses and principal axes of the stress tensor:
0 1
2 1 1
@ 1 0 1 A
1 1 2
Find the invariants I1, I2, I3, the deviator tensor, its eigenvalues, and the invariants
J2 and J3.
To calculate the principal stresses (s1, s2, s3) and principal axes (n1i, n2i, n3i), we
calculate the eigenvalues and eigenvectors of the matrix. They are found through the
equation
The eigenvalues are the roots of the cubic equation for s resulting from putting the
determinant of the matrix in (111.1) equal to zero:
2 s 1 1
1 s 1 ¼0
1 1 2 s
) ð2 sÞðsÞð2 sÞ 1 1 þ s ð2 sÞ ð2 sÞ ¼ 0
s3 4s2 þ s þ 6 ¼ 0
s1 ¼ 1
s2 ¼ 2
s3 ¼ 3
s3 I1 s2 þ I2 s I3 ¼ 0
208
209 Elasticity
I1 ¼ 4 ¼ s1 þ s2 þ s3
I2 ¼ 1 ¼ s1 s2 þ s1 s3 þ s2 s3
I3 ¼ 6 ¼ s1 s2 s3
The principal axes of stress are the eigenvectors ni associated with the three eigenvalues.
For s1 ¼ 1
0 10 1
3 1 1 n1
@ 1 1 1 A@ n2 A ¼ 0 ) n1 ; n1 ; n1 ¼ ð1; 2; 1Þ
1 2 3
1 1 3 n3
For s2 ¼ 2
0 10 1
0 1 1 n1
1 A@ n2 A ¼ 0 ) n21 ; n22 ; n23 ¼ ð1; 1; 1Þ
@ 1 2
1 1 0 n3
For s3 ¼ 3
0 10 1
1 1 1 n1
1 A@ n2 A ¼ 0 ) n31 ; n32 ; n33 ¼ ð1; 0; 1Þ
@ 1 3
1 1 1 n3
s3 J1 s2 þ J2 s J3 ¼ 0
210 Seismology
calculate:
(a) The principal stresses.
(b) The angles formed by the greatest of these stresses with the axes 1, 2, 3.
(a) As in the previous problem to find the principal stresses we calculate the eigen-
values of the stress matrix
0 1
3=2
1 5 1
B 4s
B C
B 4 2 C
C
B
3=2 C
5 1 1
C ¼ 0 ) s3 2s2 s þ 2 ¼ 0
B C
B s
B
B 4 4 2 C
C
B
3=2
3=2 C
@ 1 1 3 A
s
2 2 2
s1 ¼ 2
s2 ¼ 1
s3 ¼ 1
The largest is s1. The associated eigenvector corresponds to the axis of greatest stress
whose direction cosines are (n1, n2, n3). They are found by solving the equation
0
3=2 1
7 5 1
B C
B
B 4 4 2
C0 1
C n1
B 5 7 1 C
B 3=2 C@ n2 A ¼ 0
B
B 4 4 2 C
C n3
B
3=2
3=2 C
@ 1 1 1 A
2 2 2
211 Wave propagation. Potentials and displacements
Solving this equation with the condition that n21 þ n22 þ n23 ¼ 1, we obtain,
1
n1 ¼ n2 ¼
2
1
n3 ¼ pffiffiffi
2
(b) From these values we obtain #, the angle with the vertical axis (x3) and ’, the angle
which forms its projection on the horizontal plane with x1:
1 >
9
n1 ¼ sin # cos ’ ¼ >
2 >
>
>
1=
n2 ¼ sin # sin ’ ¼ ) ’ ¼ 315 ; # ¼ 45
2>
1 >
>
n3 ¼ cos # ¼ pffiffiffi
>
>
;
2
113. The stress tensor t ij in a continuous medium is
0 1
3x1 x2 5x22 0
@ 5x2 0 2x3 A
2
0 2x3 0
Determine the stress vector Tin acting at the point (2, 1, √3) through the plane
tangential to the cylindrical surface x22 1 x23 at that point.
First we calculate the value of the stress tensor at the given point:
0 1
pffiffiffi 6 5 0 ffiffiffi
p
tij ð2; 1; 3Þ ¼ @ 5 p0 ffiffiffi 2 3 A
0 2 3 0
A unit vector normal to the surface f ¼ x22 þ x23 4 ¼ 0 at the given point is
@f @f @f
0 1
pffiffiffi
Then, the stress vector acting at the point through that surface is given by
0 1
1 0
1 C
0
6 5 0 ffiffiffi B
p pffiffiffi
Tin ¼ tij nj ¼ @ 5 AB 2 C ¼ 5 ; 3; 3
B C
0
pffiffiffi 2 3 B pffiffiffi C
0 2 3 0 @ 3A 2
2
4 4 4
uP ¼ pffiffiffi ; pffiffiffi ; pffiffiffi
3 2 3 2 3
pffiffiffi pffiffiffi pffiffiffi
S
u ¼ 3; 3; 2
212 Seismology
and their speeds of propagation are 6 km s1 and 4 km s1, respectively. Find the
scalar and vector potentials. Displacements are always given in µm.
The displacements of P-waves can be deduced from a scalar potential function ’ such that
uPi ¼ ðr’Þi . The general form of the potential function for P-waves for harmonic motion is
’ ¼ A exp ika ðni xi atÞ ð114:1Þ
where A is the amplitude, ni the direction cosines of the ray or propagation direction, a the
velocity of propagation, and ka the wavenumber. If uj is given in µm and ka in km1, then A
is given in 103 m2. Taking the derivatives in (114.1) we obtain for the components of the
displacement
@’
uPj ¼ ¼ Aika nj exp ika ðnk xk at Þ
@xj
Their amplitude is
uPj ¼ Aka nj ð114:2Þ
and the wavenumber is
4
o 2p 2p 2
ka ¼ ¼ ¼ km1
a 6 3
By substitution in (114.2), we obtain for the two horizontal components
4
uP1 ¼ Aka n1 ¼ pffiffiffi mm
3 2
4
uP2 ¼ Aka n2 ¼ pffiffiffi mm
3 2
Dividing these two expressions and writing the direction cosines of the ray in terms of the
incident angle i and azimuth az,
n1 ¼ sin i cos az
v2 ¼ sin i sin az ð114:3Þ
n3 ¼ cos i
we have
uP1 n1 sin i cos az
P ¼1¼ ¼ ) a ¼ 45
u2 n2 sin i sin az
Using the uP3 and uP1 components,
uP3 Aka n3 cos i pffiffiffi
¼ ¼ ¼ 3 ) i ¼ 30
uP1 Aka n1 sin i cos az
From the values of the direction cosines and the amplitude of the displacement we calculate
the amplitude of the potential A:
uP1
A¼ ) A ¼ 4 103 m2
ka n1
213 Wave propagation. Potentials and displacements
2 2 2 3
’ ¼ 4 exp i x1 þ x2 þ x3 6t
3 4 4 2
Displacements of the S-wave are obtained from a vector potential ci of null divergence,
whose general form is
ci ¼ Bi exp ikb nj xj bt
where b is the velocity of propagation and kb the wavenumber. The displacements are
given by
uS ¼ r c ð114:4Þ
The wavenumber is kb ¼ o/b ¼ 1 km1. According to (114.4) the relation between the
components of the displacement and of the amplitude of the potential is
pffiffiffi pffiffiffi
S 2 3 pffiffiffi
u1 ¼ B3 B2 ¼ 3 mm
4 2
pffiffiffi pffiffiffi
3 2 pffiffiffi
uS2 ¼ B1 B3 ¼ 3 mm
2 4
pffiffiffi pffiffiffi
2 2 pffiffiffi
uS3 ¼ B2 B1 ¼ 2 mm
4 4
The potential must have null divergence,
pffiffiffi pffiffiffi pffiffiffi
2 2 3
rc¼ B1 þ B2 þ B3 ¼ 0
4 4 2
From these equations we obtain, in units of 103 m2,
B1 ¼ 2
B2 ¼ 2
B3 ¼ 0
2 2 3
cj ¼ ð2; 2; 0Þ exp i x1 þ x2 þ x3 4t
4 4 2
Note: These units will be used for all problems but not explicitly given.
115. The components of the S-wave with respect to the axes (x1, x2, x3) are (6, 3.25, 3)
where x2 is the vertical axis, the azimuth is 60 , and the angle of incidence is 30 .
Determine the amplitude and direction cosines of the SV and SH components.
From the azimuth and incident angles we calculate the direction cosines (x2 is the
vertical axis)
214 Seismology
X2
SV
r
30°
SVH
X1
60°
SH
90°
X3 R
Fig. 115a
1
n1 ¼ sin i cos az ¼
4
pffiffiffi
3
n2 ¼ cos i ¼
2
pffiffiffi
3
n3 ¼ sin i sin az ¼
4
Since the SV and SH components are on a plane normal to the direction of the ray r
(Fig. 115a) unit vectors in the direction of SV (a1, a2, a3) and of SH (b1, 0, b3) must satisfy
the equations
pffiffiffi pffiffiffi
a1 a2 3 a3 3
SV r ¼ 0 ) þ þ ¼0
4 2 4
pffiffiffi
b1 b3 3 ð115:1Þ
SH r ¼ 0 ) þ ¼0
4 4
SH SV ¼ 0 ) a1 b1 þ a3 b3 ¼ 0
The projections on the horizontal plane R of the ray r and SH are perpendicular
(Fig. 115b). Then SH forms an angle of 180 – 30 with the x1 axis. The direction
cosines of SH are
pffiffiffi
3
b1 ¼
2
1
b3 ¼
2
215 Wave propagation. Potentials and displacements
SVH
30°
X1
30°
60°
SH
R
X3
Fig. 115b
SV forms an angle of 60 with the vertical axis x2 (Fig. 115a). Then a2 ¼ sin i ¼ 12
From a2 using Equations (115.1) and a21 þ a22 þ a23 ¼ 1, we calculate a1 and a2:
pffiffiffi
3
a1 ¼
4
3
a3 ¼
4
pffiffiffi pffiffiffi
pffiffiffi
7 5 7 7
116. Given the potential ci ¼ pffiffiffi ; pffiffiffi ; 6 exp 4i p1ffiffiffi x1 þ p1ffiffiffi x2 þ pffiffiffiffiffi x3 4t ,
5 5 5 3 15
calculate the polarization angle.
First we calculate the amplitudes of the components of the displacement of the S-wave
from the vector potential
13
8
>
> uS1 ¼ c3;2 c2;3 ¼ 4 pffiffiffi ¼ 30:02
3
>
>
>
>
>
7 6
>
<
S S p p
ui ¼ r c i ) u 2 ¼ c 1;3 c3;1 ¼ 4 ffiffi
ffi þ ffiffi
ffi ¼ 13:97
> 5 3 5
pffiffiffi
>
>
pffiffiffi
>
7
>
>
S
: u3 ¼ c2;1 c1;2 ¼ 4 7 pffiffiffiffiffi ¼ 7:85
>
>
15
The modulus is
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uS ¼ u21 þ u22 þ u23 ¼ 34:03
From the vertical component uS3 we calculate the SV component (Fig. 116) knowing that
pffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffi
7 7 8
n3 ¼ cos i ¼ pffiffiffiffiffi ) sin i ¼ 1 ¼ pffiffiffiffiffi
15 15 15
Then, uS3 ¼ uSV cosð90 iÞ ) uSV ¼ 10:75.
216 Seismology
X3
i
SV
90° – i
Fig. 116
uSH
tan e ¼ ) e ¼ 71:6
uSV
117. Given the amplitudes of the P- and S-waves (ka ¼ 1):
uP1 ¼ 4 uS1 ¼ 8
pffiffiffi
uP2 ¼ 4 uS2 ¼ 2 2
pffiffiffi
uP3 ¼ 8 uS3 ¼ 4 þ 2
determine the angle of incidence i, azimuth az, polarization angle « of the S-wave,
and apparent polarization angle g.
Given that the displacements of the P-wave are on the incident plane, in the direction of the
ray r, the angle of incidence i can be obtained from the modulus and the vertical component:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffi
uP ¼ 42 þ 42 þ 82 ¼ 4 6
uP3 8
cos i ¼ P
¼ pffiffiffi ) i ¼ 35:3
u 4 6
The azimuth, the angle between the horizontal projection of the ray and the north (x1), is
obtained from the horizontal components, uP1 and uP2 :
uP2
tan az ¼ ) az ¼ 45
uP1
217 Wave propagation. Potentials and displacements
90° – az uS
1
X1
az
u S2 uS
H
az γ
SH
R
X2
Fig. 117
We calculate the SV component from the vertical component uS3 , as in the previous
problem:
uS3 uS
uSV ¼ ¼ 3 ¼ 9:37
cosð90 i0 Þ sin i0
The SH component is found from the horizontal components (Fig. 117)
uSH 3:66
tan e ¼ ¼ ) e ¼ 21:3
uSV 9:37
To calculate the apparent polarization angle g, the angle between the horizontal component
of S, (uSH), and the radial direction R (Fig. 117), we use the relation
SV
SH
Fig. 118a
X3
SV
uS
3
i
i
R
SV
uH
Fig. 118b
u SV
H
az
X1
uS
H az
u SH az
SH
X2 R
Fig. 118c
2 1 1 3
’ ¼ 3 exp i p ffiffi
ffi x1 þ p ffiffi
ffi x2 þ x3 6t
3 2 2 2 2 2
pffiffiffi ð119:1Þ
pffiffiffi
2 1 1 3
ci ¼ ð2; 2; 0Þ exp i pffiffiffi pffiffiffi x1 þ pffiffiffi x2 þ x 3 2 3t
3 2 2 2 2 2
To determine the displacements of the P- and S-waves we use the relations
uP ¼ r’
uS ¼ r c
obtaining
1
uP1 ¼ pffiffiffi uS1 ¼ 2
2
1
uP2 ¼ pffiffiffi uS2 ¼ 2
2
pffiffiffi
p ffiffi
ffi 2 2
uP3 ¼ 3 uS3 ¼ pffiffiffi
3
From the components of the displacement of the S-wave we can obtain the SV component.
The values of the angles of incidence and azimuth are found from the direction cosines,
1
n1 ¼ sin i cos az ¼ pffiffiffi
2 2
1
n2 ¼ sin i sin az ¼ pffiffiffi ) i ¼ 30 ; az ¼ 45
2 2
pffiffiffi
3
n3 ¼ cos i ¼
2
and from the angle i we obtain the SV component from uS3:
pffiffiffi
S SV SV 4 2
u3 ¼ u cosð90 iÞ ) u ¼ pffiffiffi ¼ 3:27
3
From the horizontal components of the S-waves and the azimuth we calculate the SH
component:
120. In an elastic medium of density r ¼ 3 g cm3 and Poisson ratio 1/3 there
pffiffiffi
1 1 7
propagate P- and S-waves of frequency 1 Hz in the direction ; pffiffiffi ; pffiffiffi . Given
3 2 3 2
that the pressure of the P-wave is 5000 dyn cm2, the magnitude of its displacement,
10 mm, is twice that of the S-wave, and the angle g ¼ 45 , find all the parameters
involved in the expression of the potentials ’ and ci. (It is not necessary to solve the
equations to obtain the coefficients Bi)
221 Wave propagation. Potentials and displacements
Given that Poisson’s ratio is 1/3 the relation between the elastic coefficients l and m is,
l 1
s¼ ¼ ) l ¼ 2m
2ðl þ mÞ 3
and between the velocities of P (a) and S (b)-waves is
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffi
l þ 2m 4m
a¼ ¼ ) a ¼ 2b
r r
8 a2 2
rffiffiffi
m 8
b¼ ) m ¼ rb2 ) K ¼ rb2 ¼ r ¼ ra2
r 3 3 4 3
Taking into account that the bulk modulus K is defined as the applied pressure divided by
the change in volume per unit volume y,
2
P ¼ Ky ¼ ra2 y
3
y ¼ r ’ ¼ Aka2
2
ð120:1Þ
3P 3P
y¼ ¼ Aka2 ) a2 ¼
2ra2 2rAka2
where
P ¼ 5000 dyn cm2
r ¼ 3 g cm3
f ¼ 1 Hz
uP ¼ 10 mm
222 Seismology
1 1 7
’ ¼ 10 exp i1:58 x1 þ pffiffiffi x2 þ pffiffiffi x3 3:98t
3 2 3 2
The vector potential of the S-waves is given by
ci ¼ Bi exp ikb ðn1 x1 þ n2 x2 þ n3 x3 bt Þ ð120:3Þ
We calculate kb:
o
kb ¼ ¼ 3:16 km1
b
The displacements are given by
uS ¼ r c
and
@c3 @c2
uS1 ¼ ¼ kb ðB3 n2 B2 n3 Þ
@x2 @x3
@c1 @c3
uS2 ¼ ¼ kb ðB1 n3 B3 n1 Þ ð120:4Þ
@x3 @x1
@c2 @c1
uS3 ¼ ¼ kb ðB2 n1 B1 n2 Þ
@x1 @x2
The incidence angle i is found from n3 and, using tane ¼ cosi tang, we find the polarization
angle e:
pffiffiffi
7
n3 ¼ cos i ¼ pffiffiffi ) i ¼ 31:95 ) e ¼ 31:95
3 2
The azimuth is
n2
az ¼ tan1 ¼ 64:76
n1
Since the amplitude of the S-wave displacement is 5 mm, knowing the value of e we can
find the values of the SV and SH components:
Using the values found for the displacements and Equations (120.4) and ∇ · c ¼ 0 (n1 B1 þ
n2 B2 þ n3 B3 ¼ 0) we find the values of B1, B2, B3. Substituting all the values in (120.3) we
obtain for the vector potential
pffiffiffi
1 1 7
ci ¼ ð1:17; 2:5; 3:44Þ exp 3:16i x1 þ p ffiffi
ffi x2 þ p ffiffiffi x3 1:99t
3 2 3 2
121. At the origin in an infinite medium in which s, Poisson’s ratio, is 0.25, and the
density is 3 g cm3, there is an emitter of elastic plane waves of frequency 0.5 cps.
Calculate:
(a) The equation of the P- and S-waves in exponential form and with arbitrary
amplitudes for the wave arriving at the point A(500, 300, 141) km.
(b) The arrival time.
(a) First we calculate the distance to point A and the direction cosines of the direction
of the ray (r) (Fig. 121):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r¼ 5002 þ 3002 þ 1412 ¼ 599:90
600 km
500 5
n1 ¼ ¼
600 6
300 1
n2 ¼ ¼
600 2pffiffiffi
141 2
n3 ¼ ¼
600 6
X3
141
X1
300
500
X2
Fig. 121
224 Seismology
Leaving the amplitudes A and Bi in arbitrary form and substituting the obtained values we
have for the potentials,
pffiffiffi
pffiffiffiffiffi
p 5 1 2
’ ¼ A exp i pffiffiffiffiffi x1 þ x2 þ x3 30t
30 6 2 6
pffiffiffi
pffiffiffiffiffi
p 5 1 2
ci ¼ Bi exp i pffiffiffiffiffi x1 þ x2 þ x3 10t
10 6 2 6
(b) The travel times for P- and S-waves from the origin to the given point are
r 600
ta ¼ ¼ pffiffiffiffiffi ¼ 109:5 s
a 30
r 600
t b ¼ ¼ pffiffiffiffiffi ¼ 189:7 s
b 10
x1 x2 x3
w ¼ 4 exp 0:25i p ffiffiffi þ p ffiffi
ffi þ p ffiffi
ffi 4t
6 3 2
is incident on the surface x3 ¼ 0 separating two liquids of densities 3 g cm3 and
4 g cm3. If the speed of propagation in the second medium is 2 km s1, write the
expressions for the potentials of the reflected and transmitted waves.
225 Reflection and refraction
2 8
’refl ¼ 1:07 exp i x3 þ x1 pffiffiffi t
8 2
pffiffiffi
p
2 ffiffi
ffi 8
’trans ¼ 2:01 exp i 7x3 þ x1 pffiffiffi t
8 2
pffiffiffi pffiffiffi pffiffiffi
123. A P-wave of amplitude 5 2; 5 6; 10 2 and frequency v ¼ 12 rad s1 in a
semi-infinite medium of speed of propagation a ¼ 6 km s1 and Poisson’s ratio 0.25 is
incident on the free surface. Calculate:
(a) The potential of the incident P-wave.
(b) The potential of the reflected S-wave.
(c) The components u1, u2, u3 of the reflected S-wave.
(a) The displacements of the P-wave can be deduced from its scalar potential:
’ ¼ A exp ika ðn1 x1 þ n2 x2 þ n3 x3 at Þ
ð123:1Þ
uP ¼ r’
where A is the amplitude, ka is the wavenumber (P), ni are the direction cosines, and a is the
P-wave velocity. The wavenumber is found from the given angular frequency and velocity:
o 12
ka ¼ ¼ ¼ 2 km1
a 6
Since we know the amplitudes of the components of the displacements we can find the
incidence angle i and the azimuth az:
@’ pffiffiffi
uP1 ¼ ¼ Aka n1 ¼ A2 sin i cos az ¼ 5 2
@x1
@’ pffiffiffi
uP2 ¼ ¼ Aka n2 ¼ A2 sin i sin az ¼ 5 6
@x2
@’ pffiffiffi
uP3 ¼ ¼ Aka n3 ¼ A2 cos i ¼ 10 2
@x3
pffiffiffi
Dividing the two first equations we obtain 3 ¼ tan az ) az ¼ 60 , and dividing the
last two,
pffiffiffi
5 2 1
pffiffiffi ¼ tan i cos az ¼ tan i ) i ¼ 45
10 2 2
P qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 2 2 j uP j
u ¼ uP1 þ uP2 þ uP3 ¼ Aka ) A ¼ ¼ 102 m2
ka
Finally the potential is given by
pffiffiffi pffiffiffi pffiffiffi
2 6 2
’inc ¼ 102 exp i2 x1 þ x2 þ x3 6t m2
4 4 2
227 Reflection and refraction
If we express the potential referred to the plane (x1, x3) as the incidence plane, as we did in
Problem 122, then
’ ¼ A exp ik ðx1 þ tan e x3 ct Þ
where
1 pffiffiffi
e ¼ 90 i ¼ 45 ; k ¼ ka cos e ¼ 2 pffiffiffi ¼ 2
2
a 6 pffiffiffi
c¼ ¼ ¼ 6 2 km s1
cos e 1
pffiffiffi
2
and the potential is
pffiffiffi pffiffiffi
’inc ¼ 102 exp i 2 x1 þ x3 6 2t m2 ð123:2Þ
(b) Since Poisson’s ratio is 1/4, l ¼ m, and
a pffiffiffi
b ¼ pffiffiffi ¼ 2 3 km s1
3
The angle f of the reflected S-wave is obtained by Snell’s law:
pffiffiffi rffiffiffi
cos e cos f b 2 1 5
¼ ) cos f ¼ cos e ¼ pffiffiffi ¼ pffiffiffi ) sin f ¼
a b a 2 3 6 6
From the values of e and f we calculate the P-to-S reflection coefficient VPS, using equation
4að1 þ 3a2 Þ
VPS ¼
4ab þ ð1 þ 3a2 Þ2
where we substitute
pffiffiffi
a ¼ tan e ¼ 1 and b ¼ tan f ¼ 5
so
4ð1 þ 3Þ
VPS ¼ pffiffiffi ¼ 0:64
4 5 þ ð 1 þ 3Þ 2
From this coefficient we calculate the proportion of the incident P-wave which is reflected
as an S-wave (only with SV component; the negative sign indicates the opposite sense of
the reflected ray):
(c) To calculate the amplitudes of the total displacements in terms of the two scalar
potentials, we remember that for this orientation of the axes the displacements are
given by
@’ @c
u1 ¼ ¼ uP1 þ uSV
1
@x1 @x3
@’ @c
u3 ¼ þ ¼ uP3 þ uSV
3
@x3 @x1
The displacements of the SV reflected wave in this case are
@c pffiffiffi
uSV
1 ¼ ¼ 6:4 5
@x3
@c pffiffiffi
uSV
3 ¼ ¼ 6:4 2
@x1
If we want to determine the components 1 and 2, referred to the original system of axes, we
project uSV SV
1 ¼ uR using the azimuth 60 :
pffiffiffi
uSV SV
1 ¼ uR cos az ¼ 3:2 5
pffiffiffiffiffi
uSV SV
2 ¼ u2 sin az ¼ 3:2 15
x1 3 3
ci ¼ 10 3; 2; 4 exp 5i þ x2 þ x3 4t
4 4 2
is incident on the free surface x3 ¼ 0. Find the SV and SH components of the reflected
S-wave referred to the same coordinate system as the incident wave, and the coeffi-
cient of reflection. Poisson’s ratio is 3/8.
According to the value of Poisson’s ratio the relation between l and m is
l 3
s¼ ¼ ) l ¼ 3m
2 ð l þ mÞ 8
and the relation between the velocities of the P-waves and S-waves is
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffisffiffiffiffiffiffi
l þ 2m 5m pffiffiffi
a¼ ¼ ¼ 5b
r r
The incidence angle i and the azimuth az are obtained from the direction cosines:
pffiffiffi
3
n3 ¼ cos i ¼ ) i ¼ 30
2
1
n1 ¼ sin i cos az ¼ ¼ sin 30 cos az ) az ¼ 60
4
Using Snell’s law we find the value of the critical angle
sin ic 1 b 1
¼ ) sin ic ¼ ¼ pffiffiffi ) ic ¼ 26:5
b a a 5
229 Reflection and refraction
X3
i
u SV
60°
Fig. 124a
Since i > ic, there is no reflected P-wave. The components of the incident S-wave are
obtained from the potential 8
< uS1 ¼ 0
S
ui ¼ r c ) uS2 ¼ 80
: S
u3 ¼ 40
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffi
The modulus of the displacement is uS ¼ ð80Þ2 þð40Þ2 ¼ 40 5:
The SV component is given by (Fig. 124a)
u3
uSV ¼ ¼ 80
cosð90 i0 Þ
and the SH component is
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uSH ¼ ðuS Þ2 ðuSV Þ2 ¼ 40
3 1 1 3 3
uSH
i ¼ 40 ; ; 0 exp 5i x 1 þ x 2 x 3 4t
2 2 4 4 2
For total reflection the amplitude of the reflected SV is equal to that of the incident one, but
with a phase shift d. The components are given by (Figs 124a and 124b)
uSV
1 ¼ u
SV
cos i cos az
uSV
2 ¼ u
SV
cos i sin az
uSV
3 ¼ u
SV
sin i
230 Seismology
X1
R
u SH 30°
X2
60°
SV
uH
Fig. 124b
where, using (x1, x3) as the plane of incidence, the scalar potential ’ of the reflected
P-waves is
’ ¼ A exp ik ðax3 þ x1 ct Þ
and the scalar potential of the incident and reflected S-wave is
1 4^ab
d ¼ tan
ð1 b2 Þð3 5^a2 Þ
231 Reflection and refraction
1 3
c ¼ 10 exp i3 x1 þ x3 4 2 t
2 2
is incident from an elastic medium with l ¼ 0 onto a liquid with velocity a0 ¼ 4 km s1
(the two media have the same density). Derive the equations relating the amplitudes of
the potentials of the incident, reflected, and transmitted waves.
Given that the wave is incident from an elastic medium onto a liquid medium, there are reflected
S- and P-waves in the elastic medium and transmitted P-waves in the liquid (Fig. 125).
If l ¼ 0, the P-wave velocity in the elastic medium is
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffi
l þ 2m 2m pffiffiffi pffiffiffipffiffiffi
a¼ ¼ ¼ b 2 ¼ 4 2 2 ¼ 8 km s1
r r
Assuming (x1, x3) is the incidence plane, we use the scalar S-wave potential which, for the
incident and reflected waves in the solid medium, is given by
X3
P
e⬘
a⬘= 4
M⬘ b⬘= 0
X1
f e a=8
M f
b = 4√2
S
S
Fig. 125
232 Seismology
By substitution of the potentials we obtain the equations (in units of 103 m2)
pffiffiffi
A þ 10 þ B ¼ A0 7
A þ B þ 10 ¼ 0
pffiffiffi pffiffiffi
2A0 ¼ A þ 10 3 3B
4bð1 þ 3a2 Þ
VSP ¼ ð126:1Þ
4ab þ ð1 þ 3a2 Þ2
234 Seismology
X1
f f e
S S
X3
Fig. 126
where a ¼ tan e and b ¼ tan f, and f is the emergence angles of the incident S-wave and e is
that of the reflected P-wave (Fig. 126). The relation between f and e according to Snell’s law is
cos f cos e a pffiffiffi 1
¼ ) cos e ¼ cos f ¼ 3 ) e ¼ 30
b a b 2
Substituting in Equation (126.1):
pffiffiffi
4 3ð 1 þ 1Þ pffiffiffi
VSP ¼ 2
¼ 3
4 þ ð 1 þ 1Þ
We can write the potential of the incident S-wave referred to the incidence plane (x1, x3) by
means of the rotation matrix
0 1
cos az sin az 0
@ sin az cos az 0 A
0 0 1
and substituting
0 pffiffiffi 1
1 3
0 C0 10pffiffi3ffi 1 0 B1 1 pffiffiffi
8
B 2 2 <B1 ¼ 4 3
B pffiffiffi
2 A ¼ @ B2 A ) B2 ¼ 16 ¼ B0
C@
@ 3 1
B C
0A 4 B3
:
B3 ¼ 4
2 2
0 0 1
the potential is
c0i ¼ ðB1 ; B2 ; B3 Þ exp ikb ðcos fx1 þ sin fx3 4tÞ
pffiffiffi
ð126:2Þ
pffiffiffiffi 1 3
¼ 4 3;16; 4 exp 5i x1 þ x3 4t
2 2
The scalar potential of the SV component is
cSV ¼ B0 exp ikb ðcos fx1 þ sin fx3 4tÞ
pffiffiffi
1 3
¼ 16 exp 5i x1 þ x3 4t
2 2
235 Reflection and refraction
5 3 3 1
’ ¼ 16 3 exp i pffiffiffi x 1 þ x 2 x 3 4 3t
3 4 4 2
The components of the displacement in mm are
8
P
pffiffiffi
< u1 ¼ 20 3
>
>
uP ¼ r’ ) uP2 ¼ 60
>
>
: P
u3 ¼ 40
(b) For the reflected S-wave we have to separate the SV and SH components. The SV
component can be deduced from the scalar potential
pffiffiffi
r 1 3
cSV ¼ B exp ikb ðcos f x1 sin f x3 4tÞ ¼ B exp 5i x1 x3 4t
2 2
We obtain B by means of the reflection Vss:
2
4ab ð1 þ 3a2 Þ B
VSS ¼ ¼0¼ )B¼0
4ab þ ð1 þ 3a2 Þ2 B0
For the reflected SH component we use the displacement of the u2 instead of the
potential. The displacements of the SH component of the incident wave are obtained from
(126.2):
uiSH ¼ c01;3 c03;1 ¼ 40
Referred to the reference of the plane of incidence, the displacement is given by
pffiffiffi
i 1 3
uSH ¼ 40 exp 5i x1 þ x3 4t
2 2
The amplitude of the reflected SH wave is equal to that of the incident SH wave. Referred
to the incidence plane system of reference,
pffiffiffi
r 1 3
uSH ¼ 40 exp 5i x1 x3 4t
2 2
The displacement of the reflected S-wave referred to the original system of axes is
pffiffiffi pffiffiffi
1 3 3
uri ¼ ðBr1 ; Br2 ; Br3 Þ exp i5 x1 þ x2 x3 4t
4 4 2
Since the SV component is zero, Br3 ¼ 0; Br1 and Br2 are found using the equations
x1 x2 x3
w ¼ 4 exp 0:25i pffiffiffi þ pffiffiffi þ pffiffiffi 4t
6 3 2
is incident on the surface x3 ¼ 0 of separation between two liquids. If the speed of
propagation in the second medium is 2 km s1, the pressure exerted by the incident
wave on the surface of separation is 5 109 Pa, and the transmitted energy is four
times greater than the reflected energy, calculate:
(a) The energy transmitted to the second medium.
(b) The potentials of the transmitted and reflected waves referred to the same
coordinate system as the incident potential.
(a) The intensity or energy per unit surface area of the wavefront of an incident
P-wave is given in units of J m2 by
A0 ¼ 4 103 m2
ka ¼ 0:25 km1
a ¼ 4 km s1
o
ka ¼ ) o ¼ 1 s1
a
We need to know the value of the density r. Since the medium is liquid l ¼ K (bulk
modulus) and then l ¼ P/y, where P is the pressure and y the cubic dilatation (change of
volume per unit volume). For liquids the shear modulus m is zero and from the velocity of
P-waves we obtain,
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffi
l þ 2m l
a¼ ¼ ) l ¼ a2 r
r r
and from Snell’s law the emergence angle of the transmitted wave e0 is
rffiffiffiffiffiffiffiffiffiffiffi pffiffiffi
cos e cos e0 0 a0 2 1 1 0 1 7
¼ 0 ) cos e ¼ cos e ¼ pffiffiffi ¼ pffiffiffi ) sin e ¼ 1 ¼ pffiffiffi
a a a 4 2 2 2 8 2 2
Given that the transmitted energy is four times the reflected energy,
W 2 A20 r0 o4
Itras 0 V 2 2r0 a
¼ 4 ¼ 2 a2 4 ) 2 ¼
Iref V A0 ro W 5ra0
a
If we substitute in (127.4)
5 pffiffiffi
r0 7
V ¼ 4
5 pffiffiffi
r0 þ 7
4
5
2
W ¼ 4
5 pffiffiffi
r0 þ 7
4
We have three equations for r’, V, and W. The solution for positive values of the variables
is
W ¼ 0:23
r0 ¼ 7:7 )
V ¼ 0:40
(b) The potential of the reflected P-wave is
’ref ¼ VA0 exp ik a ðn1 x1 þ n2 x2 n3 x3 at Þ
1 1 1 1
¼ 1:6 exp i pffiffiffi x1 þ pffiffiffi x2 pffiffiffi x3 4t
4 6 3 2
To determine the potential of the transmitted wave we have to calculate the direction
cosines of the transmitted ray. The azimuth is the same as that of the incident wave which
can be deduced from the direction cosines and the value of i:
1 1
n1 ¼ sin i cos az ¼ pffiffiffi ) cos az ¼ pffiffiffi
6 3
pffiffiffi
1 2
n2 ¼ sin i sin az ¼ pffiffiffi ) sin az ¼ pffiffiffi
3 3
The direction cosines of the transmitted ray in the medium M0 are
1 1 1
n01 ¼ sin i0 cos az ¼ pffiffiffi pffiffiffi ¼ pffiffiffi
2 2 3 2 6
1
n02 ¼ sin i0 sin az ¼ pffiffiffi
2 3
pffiffiffi
7
n03 ¼ cos i0 ¼ pffiffiffi
2 2
239 Reflection and refraction
1 1 1 7
’tras ¼ 0:92 exp i pffiffiffi x1 þ pffiffiffi x2 þ pffiffiffi x3 2t
2 2 6 2 3 2 2
128. Two liquid media are separated at x3 ¼ 0, the first of volumetric coefficient
1
K ¼ 109 Pa and density 1 g cm3. The amplitudes of the components of an
2 pffiffiffi
incident wave of frequency 3 Hz ffi ui ¼ 18p 1;1; 6 mm and those of the wave
pffiffiare
63 2p pffiffiffiffipffiffiffi pffiffiffi
transmitted to medium 2 are 2; 2; 3 mm. Given that the amplitude of
7
the transmitted potential is twice that of the reflected potential, find expressions for
the incident, reflected, and transmitted potentials.
In liquids only P-waves are propagated and their displacements can be deduced from the
scalar potential
’ ¼ A0 exp ika ðn1 x1 þ n2 x2 þ n3 x3 at Þ ) uP ¼ r’
Then in our case the components of the displacement in mm are
@’
uP1 ¼ ¼ A0 ka n1 ¼ A0 ka sin i cos az ¼ 18p
@x1
@’
uP2 ¼ ¼ A0 ka n2 ¼ A0 ka sin i sin az ¼ 18p
@x2
@’ pffiffiffi
uP3 ¼ ¼ A0 ka n3 ¼ A0 ka cos i ¼ 18p 6
@x3
and we find az ¼ 45 ; i ¼ 30 ; e ¼ 60 , A0 ¼ 6 103 m2.
The emergence angle of the refracted wave, e0 , can be found from its displacements,
pffiffiffi pffiffiffi pffiffiffi pffiffiffi
pffiffiffi
63p 2 2 2 3 3 2
P
utras ¼ pffiffiffi pffiffiffi ; pffiffiffi ; pffiffiffi ) n3 ¼ sin e ¼ pffiffiffi ) cos e0 ¼ pffiffiffi
0 0
7 7 7 7 7 7
The P-wave velocity in the medium of the incident wave is
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffi
l þ 2m K 1
a¼ ¼ ¼ pffiffiffi km s1
r r 2
Using Snell’s law we find the velocity of the medium of the refracted wave,
pffiffiffi
cos e a 0 2 2
¼ ) a ¼ pffiffiffi
cos e0 a0 7
From the values of the velocities in the two media we calculate their densities:
1 0
9
a 2
7 Kr 0 r >
>
2
>
¼ ¼ ¼
>
>
a 02 16 K r 0 K 0
= 3 3
) r0 ¼ r ¼ g cm3
qffiffiffiffi 2 2p ffiffi
ffi > 2 2
0 K0 0 8 0>
>
a ¼ r0 ¼ pffiffiffi ) K ¼ r ;
>
>
7 7
240 Seismology
The reflection V and transmission W coefficients are found using their expressions and from
them we get the relation between the amplitude A0 of the incident wave potential and those
of the reflected A and refracted A0 waves, and substituting the value for A0 ¼ 6, we obtain
A0 ¼ 6 ) A ¼ 3 103 m2
From these values we can write the potentials of the incident, reflected, and transmitted
waves:
pffiffiffi
pffiffiffi 1
1 3 1
’inc ¼ 6 exp i6p 2 pffiffiffi x1 þ pffiffiffi x2 þ x3 pffiffiffi t
2 2 2 2 2 2
pffiffiffi
pffiffiffi 1
1 3 1
’ref ¼ 3 exp i6p 2 pffiffiffi x1 þ pffiffiffi x2 x3 pffiffiffi t
2 2 2 2 2 2
pffiffiffi pffiffiffi pffiffiffi pffiffiffi pffiffiffi
7 2 2 3 2
’tras ¼ 6 exp i3p pffiffiffi pffiffiffi x1 þ pffiffiffi x2 þ pffiffiffi x3 2 pffiffiffi t
2 7 7 7 7
129. Two liquids in contact have speeds of propagation of 4 and 6 km s1. The density
of the first is 2 g cm3 and is less than that of the second. For waves of normal
incidence, the reflected and transmitted energies are equal. A wave of v ¼ 1 s1 and
with a potential of amplitude A0 ¼ 2103 cm2 is incident from the first onto the
second at an angle of 30 . Calculate:
(a) The transmitted and reflected energies.
(b) An expression for the transmitted potential.
(a) For normal incidence, the reflection and transmission coefficients in terms of the
refractive index m ¼ a/a0 and the density contrast m ¼ r0 /r are given by
mn
Vn ¼
mþn
2
Wn ¼
mþn
If the reflected energy is equal to the transmitted energy, then
ðm nÞ2 4mn
Vn2 ¼ mnWn2 ) ¼ ð129:1Þ
ðm þ nÞ2 ðm þ nÞ2
Substituting n ¼ a/a0 ¼ 2/3, from (129.1) we obtain the value of m:
2 4 m ¼ 3:9
4
m 4m þ ¼ 0 )
9 m ¼ 0:1
r0
m¼ ¼ 0:1 ) r0 ¼ 0:2 g cm3
r
m ¼ 4 ) r0 ¼ 8 g cm3
241 Reflection and refraction
i = 30°
r = 2 g cm–3 M
e
a = 4 km s–1
e⬘
a ⬘ = 6 km s–1
M⬘
Fig. 129
1 7
’tras ¼ 920 exp i x3 þ x1 8t
8 3
130. An SV wave is incident on the free surface of an elastic medium of Poisson ratio
0.25. If the potential of the wave is (in units of 103m2)
pffiffiffi
5 5 1 1 3
c i ¼ pffiffiffi ; pffiffiffi ; 0 exp i pffiffiffi x1 þ pffiffiffi x2 þ x3 4t
2 2 2 2 2 2 2
find the components of the amplitude of the reflected P-wave referred to this set of axes.
From the direction cosines we find the incidence angle i, the emergence angle f, and the
azimuth az of the incident SV wave (Fig. 130):
pffiffiffi
3
n3 ¼ cos i ¼ sin f ¼ ) i ¼ 30 and f ¼ 60
2
1 1
n1 ¼ sin i cos az ¼ cos az ¼ pffiffiffi ) az ¼ 45
2 2 2
Bearing in mind that Poisson’s ratio is 0.25, from Snell’s law we find the emergence angle
e of the reflected P-wave:
pffiffiffi pffiffiffi
s ¼ 0:25 ) l ¼ m ) a ¼ 3b ¼ 4 3 km s1
pffiffiffi
cos f cos e 3
¼ ) cos e ¼ ) e ¼ 30
b a 2
X3
e
f
f
P
i
SV S
Fig. 130
243 Ray theory. Constant and variable velocity
The reflection coefficient for the reflected P-wave gives us the relation between the
amplitude of the potential, B0, of the incident SV wave and A, that of the reflected P wave:
To find B0 we write the potential of the incident SV wave referred to the (x1, x3) plane of
incidence using the rotation matrix
0 5 1
0 1 pffiffiffi 0 1
cos az sin az 0 B 2 C 0
B C
@ sin az cos az 0 AB B p5ffiffiffi C ¼ @ 5 A
B C C B C
0 0 1
@ 2A 0
0
pffiffiffi
B0 ¼ 5 103 m2 ) A ¼ 5 3 103 m2
Referred to this system of axes the potential of the reflected P-wave is given by
kb 1
k¼ ¼ 1 ¼ 2 km1
cos f 2
pffiffiffi pffiffiffi
1
’ ¼ 5 3 exp i2 x3 pffiffiffi þ x1 4 3t
3
The amplitudes of the displacements of the reflected P-wave referred to this set of axes are
@’ pffiffiffi
u1 ¼ ¼ 10 3 mm
@x1
@’
u3 ¼ ¼ 10 mm
@x3
Referred to the original set of axes the horizontal components are
pffiffiffi
10 3
u1 ¼ u1 cos az ¼
0 p ffiffiffi mm
2
pffiffiffi
10 3
u20 ¼ u1 sin az ¼ pffiffiffi mm
2
131. Assume that the Earth’s crust consists of a single layer of thickness H and a
constant speed of propagation of seismic waves of v1 on top of a mantle of velocity of
propagation 20% greater than the crust. Given that a focus on the surface produces a
reflected wave that takes 17.2 s to reach a distance of 99 km, and that this is the
244 Seismology
xc
F S
H
ic v1
ic
v2
Fig. 131a
critical distance, calculate the values of H, v1, and v2. Plot the travel-time curve (t, x)
for this specific case with numerical values.
The critical distance xc is the distance at which a ray that is reflected with the critical angle
at the top of the mantle arrives at the surface and is given by the equation (Fig. 131a)
xc ¼ 2H tan ic ¼ 99 km ð131:1Þ
where H is the thickness of the crust. Since we know the relation between the velocities in
the crust and the mantle, we can calculate the critical angle
sin ic 1 1
v2 ¼ 1:2v1 ) ¼ ) sin ic ¼ ) ic ¼ 56:44
v1 v2 1:2
If we substitute in Equation (131.1) we obtain the thickness of the crust, H:
99 ¼ 2H tan 56:44 ) H ¼ 32:8 km
The travel time of the critically reflected ray is
H H
t¼2 ¼2 ¼ 17:2
v1 cos ic v1 cos 56:44
32:84
) v1 ¼ 2 ¼ 6:9 km s1 ) v2 ¼ 8:3 km s1
17:2 cos 56:44
To draw the travel-time curve for different distances of the direct, reflected, and critically
refracted waves we use the equations
x
t1 ¼
v1
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 x2
t2 ¼ þ H2
v1 4
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
x 2H v22 v21
t3 ¼ þ
v2 v1 v2
245 Ray theory. Constant and variable velocity
0 0 9.5 –
30 4.3 10.4 –
60 8.7 12.9 –
90 13.0 16.1 –
99 14.3 17.2 17.2
120 17.4 19.8 19.7
150 21.7 23.7 23.4
40
30
t (S)
20
2 1
10
ti
0
0 50 xc 150 200 250 300
x (km)
Fig. 131b
246 Seismology
x
P
Fig. 132
132. In a seismogram recorded at a regional distance, the S-P time lag is 5.5 s, and the
focus is at a depth x/2, where x is the epicentral distance. The
pffiffiffi model1Earth has a single
layer of Poisson ratio 0.25 and constant S-wave velocity 3 km s . Calculate:
(a) The depth of the focus.
(b) The epicentral distance.
(a) For a direct wave from point F to point P (Fig. 132) the difference of the arrival
times of the P- and S-waves (the S-P interval) is
FP FP
t S-P ¼ 5:5 ¼
b a
The distance FP can be expressed in terms of x as (Fig. 132)
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
x 2 pffiffiffi
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 5
FP ¼ x2 þ h2 ¼ x2 þ ¼x
2 2
The S-P interval is given by
pffiffiffi
5ab
5:5 ¼ x
2 ab
Since Poisson’s ratio is 0.25 and knowing the S-wave velocity we obtain
pffiffiffi pffiffiffi
pffiffiffi 5 31
s ¼ 0:25 ) a ¼ b 3 ) 5:5 ¼ x pffiffiffipffiffiffi
2 3 3
x ¼ 21 km
x
h ¼ ¼ 10:5 km
2
133. The Earth consists of a layer of thickness 20 km and seismic wave velocity
6 km s1 on top of a medium of speed of propagation 8 km s1. A seismic focus is
247 Ray theory. Constant and variable velocity
x S
h
F
H
ic ic
ic i v1
v2
Fig. 133
located at a depth of 10 km. Calculate the difference in travel times between the
reflected and the critical refracted waves observed on the surface at a distance of
150 km from the epicentre.
This problem is similar to Problem 131, but now the focus is at depth h ¼10 km.
The critical distance in this case is given by (Fig. 133)
xc ¼ ð2H hÞ tan ic
v1 1 6
sin ic ¼ ) ic ¼ sin ¼ 48:6
v2 8
Since the distance 150 km is greater than the critical distance there arrive critically
refracted rays. The travel times of the reflected (t2), and critically refracted (t3) rays at that
distance are
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
x2 þ ð2H hÞ2 1502 þ ð2 20 10Þ2
t2 ¼ ¼ ¼ 25:5 s
v1 6
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
x ð2H hÞ v22 v21 150 ð2 20 10Þ 82 62
t3 ¼ þ ¼ þ ¼ 22:1 s
v2 v1 v2 8 86
The time difference between the travel times of the two rays is
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
x ð2H hÞ v22 v21
t3 ¼ þ
v2 v1 v2
H
xc ¼ ð2H hÞ tan ic ) 51:09 ¼ 2H tanð48:59Þ
2
) H ¼ 30 km; h ¼ 15km
Knowing the depth and thickness of the crust, using Snell’s law, the value of the critical
angle, and the delay time ti, we find the velocities v1 and v2:
sin ic 1
¼ ) v1 ¼ 0:75v2
v1 v2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ð 2 30 15 Þ v22 ð0:75v2 Þ2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
ð2H hÞ v2 v1 2
ti ¼ ) 4:96 ¼
v1 v2 0:75v22
) v2 ¼ 8 km s1
v1 ¼ 6 km s1
135. In a seismogram, the S-P time difference is equal to 5.31 s, and corresponds to a
regional earthquake that occurred at a depth h ¼ 2H, where H is the thickness
of the crust. Given that the crust is formed by a layer of constant P-wave velocity
of 3 km s1, that below it there is a semi-infinite mantle of double that speed of
propagation, and that Poisson’s ratio is 0.25, determine:
(a) An expression for the travel-time of the P- and S-waves.
(b) The epicentral distance for an emerging P-wave with a take-off angle of 30 at the
focus.
(a) The travel time corresponding to the ray given in Fig. 135 is given by
FA AS
t¼ þ
2v v
F⬘ S
x
H i0 i0
v
2H
A S⬘
ih
2v
Fig. 135
249 Ray theory. Constant and variable velocity
where v ¼ a for P-waves and v ¼ b for S-waves. Using Snell’s law we find the relation
between the incidence angle at the focus ih and at the station i0:
sin ih sin i0 1
¼ ) sin i0 ¼ sin ih
2v v 2
From Fig. 135 we obtain
H H
cos ih ¼ ) FA ¼
FA cos ih
H H
cos i0 ¼ ) AS ¼
AS cos i0
From these equations we deduce the expression for the travel time:
H H
t¼ þ ð135:1Þ
2v cos ih v cos i0
For the epicentral distance we obtain
E S
x
H1
F∗
n1
2 (H1 – h)
i1 i1
F⬘
H2 n2
ic ic
n3
Fig. 136
251 Ray theory. Constant and variable velocity
critical angle ic and from this value the incidence angle at the focus i1 for the critically
refracted ray:
sin i1 sin ic 1 6
¼ ¼ ) sin ic ¼ ) ic ¼ 48:6
v1 v2 v3 8
7
sin i1 ¼ ) i1 ¼ 61:0
8
The travel time t3 of the critically refracted ray at the bottom of the second layer is given by
FA AB BC CD DS
t3 ¼ þ þ þ þ ð136:1Þ
v1 v2 v3 v2 v1
If the epicentral distance x is 100 km, the different segments of (136.1) are
H1 h H1
cos i1 ¼ ¼ ) FA ¼ 12:4 km; DS ¼ 24:8 km
FA DS
H2
cos ic ¼ ) AB ¼ CD ¼ 27:2 km
AB
BC ¼ x 2H2 tan ic ðH1 hÞ tan i1 H1 tan i1 ¼ 26:8 km
Finally, by substitution in (136.1) we obtain,
t3 ¼ 17:7 s
137. Consider a two-layered structure of thickness H and speed of propagation v and
3v on top of a half-space medium of speed of propagation 2v. At a depth 3H below the
surface there is a seismic focus. Write the expressions (as functions of H, v, and ih) for
the travel times of waves that reach the surface without being reflected. Give the
range of values of ih.
In this problem the focus is located at the half-space medium at depth h¼3H under its
boundary. Applying Snell’s law we can find the relation between the velocities, the
incidence angles at the focus and at the bottom of each layer, and the critical angle at the
boundary between the second layer and the half-space (Fig. 137):
sin ih sin i2 sin i1
¼ ¼
2v 3v v ð137:1Þ
sin ic 1
¼ ) ic ¼ 41:8
2v 3v
The rays which leave the focus and arrive at the surface at a distance x are only those with
angles less than the critical angle (Fig. 137). The travel time for these rays is
FA AB BS H H H
t¼ þ þ ¼ þ þ ð137:2Þ
2v 3v v 2v cos ih 3v cos i2 v cos i1
According to Equation (137.1) we have the relation between the incidence angles:
3 1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
sin i2 ¼ sin ih ) cos i2 ¼ 1 sin2 ih ¼ 4 9 sin2 ih
2 2
1 1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
sin i1 ¼ sin ih ) cos i1 ¼ 1 sin2 ih ¼ 4 sin2 ih
2 2
252 Seismology
E x S
i1
H i0 v
H
3v
i2
A
ic
ih 2v
H
Fig. 137
Substituting in (137.2) we write the travel time as function of the take-off angle ih:
!
H 1 2 2
t¼ þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v 2 cos ih 3 4 9 sin2 ih 4 sin2 ih
The range of values of the take-off angle for rays which arrive at the surface is
0 < ih <42 .
138. A semi-infinite medium consists of two media of velocities v and 3v separated by a
vertical surface. In the first medium there is a focus of seismic waves at a depth a
below the free surface and at the same distance a from the surface separating the two
media. Write the expressions for the direct, reflected, and transmitted waves arriving
at the free surface, and plot the travel time curve (t, x) in units of a/v and a (neglecting
waves with more than a single reflection).
In this situation we have the following rays arriving at the surface: direct in the first
medium, reflected at the boundary, and critically refracted and refracted to the
second medium. We consider two cases for rays arriving at distances 0 < x < a and
distances x > a.
253 Ray theory. Constant and variable velocity
E S
x
i P
a i
i A
F
a a
v 3v
Fig. 138a
E S
x
ic B
a
A
ic
ic
F
v 3v
Fig. 138b
(a) For 0 < x <a, the travel time t1 of the direct wave is (Fig. 138a)
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
FS x 2 þ a2
t1 ¼ ¼
v v
The travel time t2 of the reflected ray is
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
FP þ PS AP þ PS AS ð2a xÞ2 þa2
t2 ¼ ¼ ¼ ¼
v v v v
The reflected rays exist also for negative distances, but we will not consider them.
The travel time t3 of the critically refracted ray (Fig. 138b) is
FA AB AS
t3 ¼ þ þ ð138:1Þ
v 3v v
254 Seismology
S E
xc
ic
a
ic
ic
F ic
a
v 3v
Fig. 138c
xc ¼ a SB ð138:2Þ
From Fig. 138c we obtain
BP a a tan ic
tan ic ¼ ¼ ) SB ¼ 1:83a
SB SB
E S
X
a
A
e
i
F∗
v 3v
Fig. 138d
2.0
2
3
1.5
t (a/v)
1
1.0
0.0
0.0
0.0 0.2 0.4 0.6 0.8 1.0
x (a)
Fig. 138e
(b) For distances x > a, we have the rays refracted at the boundary between the two
media when the incidence angle is less than the critical angle (Fig. 138d), that is,
i <19.47 :
256 Seismology
FA AS
t4 ¼ þ
v 3v
a
FA ¼
cos i
xa
AS ¼
cos e
Using Snell`law
sin i sin e
¼ ) sin e ¼ 3 sin i
v 3v
The travel-time is given by
a xa
t4 ¼ þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v cos i 3v 1 9 sin2 i
The travel-time curves for direct (1), reflected (2), critically refracted (3) and transmitted
(4) waves are given in Fig. 138e.
139. Given the structure in the diagram, calculate the arrival times of the direct and
(non-reflected) transmitted waves for x 0, where x ¼ 0 is a point on the free surface
in the vertical above the focus.
At x ¼ 0 )ih ¼ 0 , the travel-time of the vertical ray is (Fig. 139a)
a a 3a
t¼ þ ¼
2v v 2v
For rays arriving at x > 0 and leaving the focus with take-off angles 0 < ih < 45 , the
travel-times are given by
FA AS a a
t¼ þ ¼ þ
2v v 2v cos ih v cos r
a
cos ih ¼ ð139:1Þ
FA
a
cos r ¼
AS
a
2a
2v a
Focus
Fig. 139
257 Ray theory. Constant and variable velocity
X
E A⬘ S
2a A
F⬘
2v v
a
ih
Fig. 139a
The relation between the epicentral distance x and the incidence angle ih is
x ¼ F0 A þ A0 S
F0 A ¼ a tan ih
A0 S ¼ a tan r
!
sin ih
x ¼ aðtan i þ tan rÞ ¼ a tan i þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4 sin2 ih
x
E A⬘ S
n
2a
r
2n
a A
ih
ih
F F⬘
a
Fig. 139b
By substitution in (139.3),
2x 3a
t¼
v sin ih 2v cos ih
For ih ¼ 45 the epicentral distance is
1
x ¼ a 1 þ pffiffiffi
7
This is the limit of the epicentral distance at which these rays arrive. The corresponding
time limit is
a 1 4
t¼ pffiffiffi þ pffiffiffiffiffi
v 2 14
For angles ih > 45 (Fig. 139b), the travel time and epicentral distance, as a function of the
take-off angle ih, are
FA AS a xa a xa
t¼ þ ¼ þ ¼ þ
2v v 2v sin ih v sin r 2v sin ih sin ih
v
2
x ¼ a þ A0 S
A0 S
tan r ¼
2a F0 A
a
F0 A ¼
tan ih
a sin ih
x ¼ a þ 2a pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
tan ih 4 sin2 ih
259 Ray theory. Constant and variable velocity
Using
cos ih cos r 1
¼ ) cos r ¼ cos ih
2v v 2
1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
sin r ¼ 4 cos2 ih
2pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4 cos2 ih
tan r ¼
cos ih
we obtain for x and t,
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a 4 cos2 ih
x ¼ a þ 2a
tan ih cos ih
a 2ðx aÞ
t¼ þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2v sin ih v 4 cos2 ih
For i ¼ 90 , as expected the ray doesn’t arrive at the free surface.
140. For the structure in Fig. 140a, write the equations of the travel times of the
direct, reflected, and transmitted waves (neglecting waves with more than a single
reflection) as a function of the epicentral distance. Determine the times of intersection,
and the minimum and maximum distances in each case in terms of a/v and v. Plot the
travel-time curves.
The travel-time of the direct wave for distance 0 < x < 1 is (Fig. 140a)
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4x2 þ a2
t¼
2v
For the ray reflected on the horizontal surface at depth a the travel time is
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
3a
x2 þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 4x2 þ 9a2
t¼ ¼
v 2v
x
S
a /2
F
∗
a
ν
a
2ν
Fig. 140a
260 Seismology
x
S
a/2
F v
a
a
2v
Fig. 140b
xmax ¼ 1
The critically refracted ray on the surface at depth a, using the general expression, is given by
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffi
x ð2H hÞ v22 v21 x a3 3
t¼ þ ¼ þ
v2 v 1 v2 2v v 4
The minimum distance for this ray corresponds to the critical distance:
sin ic 1
¼ ) ic ¼ 30
v 2v pffiffiffi
a 3
xc ¼ tan ic þ a tan ic ¼ a
2 2
and the maximum distance is
pffiffiffi
a a 3þ 3
xmax ¼ a þ pffiffiffi ¼
3 3
261 Ray theory. Constant and variable velocity
10 3
6
t (a/v)
2 1
2
4
0 2 4 6 8 10 12
X (a)
Fig. 140c
On the surface at depth 2a there is no critically refracted ray, since the minimum take-off
angle ih at that surface is
a
tan ih ¼ a ¼ 2 ) ih ¼ 63:4
2
greater than the critical angle 30 .
The travel-time curves are drawn after rewriting the equations in units of a/v and a, and
are represented in Fig. 140c
tv2 x 2 x
1. Direct ray: ¼ 14 ; 0
< 1 (a hyperbola)
a a a 2
tv x 2 9 x
2. Reflected ray on the surface at depth a: ¼ ; 0
3 (a hyperbola)
a a 4 a
tv2 x2 49 x
3. Reflected ray on the surface at depth 2a: ¼ ; 7
1 (a hyperbola)
a a 4 pffiffiffi a
tv 1 x 3 3 x
4. Critically refracted ray on surface at depth a: ¼ þ ; 0:87
1:58 (due
a 2a 4 a
to the short range of distances this is not noticeable in the figure)
262 Seismology
141. For the structure in the diagram, assume a seismic focus at the surface, and
calculate the travel time of the direct, reflected, and critical refracted waves for
epicentral distances between 0 and a. Calculate the critical distance, and the expres-
sion for the transmitted wave.
Since the focus is at the free surface, the travel time of the direct ray is simply given by
(Fig. 141a)
x
t¼ 0<x<a
v
a
F
*
a ν 3ν
45°
Fig. 141
F S
ν 3ν
45°
a√ 2
45°
F⬘
Fig. 141a
263 Ray theory. Constant and variable velocity
x
F S D
*
β 45°
45° ic
ic
B
A
a a
√2
2
45°
Fig. 141b
FP PS F0 P PS 1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
t¼ þ ¼ þ ¼ x2 þ 2a2 2ax
v v v v v
The critical angle is given by
pffiffiffi
sin ic 1 1 2 2
¼ ) sin ic ¼ ) cos ic ¼ ) ic ¼ 19:47
v 3v 3 3
The travel-time of the critically refracted ray is (Fig. 141b)
FA AB BS
t¼ þ þ
v 3v v
a
pffiffiffi
2 3
FA ¼ ¼ a
cos ic 4
The distance BS can be found using the sine law in triangle SBD:
sinð90 ic Þ sin 45 3
¼ ) BS ¼ ða xÞ
ax BS 4
and the distance AB (calling d ¼ BD) is
pffiffiffi pffiffiffi pffiffiffi
pffiffiffi 2 2 2
AB ¼ a 2 a cos 45 a tan ic d ¼ a a tan ic d
2 2 2
264 Seismology
xc
F s⬘ D
45°
β
45° ic
ic
A
2
a a
2
45°
Fig. 141c
142. A medium consists of a flat crust of thickness H and constant speed of propagation
v1 on a semi-infinite mantle of constant speed of propagation v2. For a focus at the
surface, at a distance x the direct wave arrives at a time t1 ¼ x/a, the critical distance is
2a pffiffiffi
xc ¼ pffiffiffi, and the direct and critical refracted waves intersect at the distance x ¼ 2a 3.
3
(a) Calculate the crust’s thickness, its speed of propagation, the mantle’s speed of
propagation, and the critical angle.
265 Ray theory. Constant and variable velocity
(b) Assume now that this is a layer which dips downwards at 45 with the parameters
of the model being those determined in the previous part. Calculate the travel
times of the reflected and critical refracted waves at x ¼ a, 3a, and 5a.
(a) We determine the velocity of the crust from the travel time of the direct ray:
x x
t1 ¼ ¼ ) v1 ¼ a
v1 a
The critical distance is given by
v1 a a
xc ¼ 2H tan ic ¼ 2H pffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi ¼ 2 pffiffiffi
¼ 2H pffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð142:1Þ
2
v2 v1 2 2
v2 a 2 3
xc
F S
H *
θ
ic
xc sinθ
Fig. 142
266 Seismology
x t2 (a/v) t3 (a/v)
a 2.80 –
3a 4.64 –
5a 6.57 6.56
a b
R
ν0 / k
E S
V0
h
r
i0
F ∗
ih
Fig. 143
The problem is solved if we can express r as a function of x, v0, h, and k. The epicentral
distance x is the sum of a and b (Fig. 143):
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v 2ffi
0
b ¼ R2
k
But we know that for a distribution of velocities which increases linearly with depth the
rays are circular and their radius R is
r ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 v 2ffi
v0 v0 0
R¼ þr )b¼ þr
k k k
and
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v 2 v 2ffi
0 0
a¼ þr þh
k k
Therefore,
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r 2 v 2ffi r
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v0 0 v0 2 v
0
2
b¼xa) þr ¼x þr þh
k k k k
Solving for r we obtain
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u x 2 h2 2 v 0 h
u0 12
k A þ v0 þ h 2 v0
u
r ¼ t@
2x k k
144. Consider a semi-infinite medium in which the velocity increases linearly with depth
according to the expression v ¼ 4 þ 0.1 z. There is a seismic focus at a depth of 10 km.
Calculate the epicentral distance reached by a wave leaving the focus at an angle of 30 .
The velocity at the focus is found directly by putting in the equation for the distribution of
velocity, z ¼ h:
vh ¼ 4 þ 0:1 10 ¼ 5 km s1
According to Snell’s law we find the velocity at the point of greatest depth penetration
(i ¼ 90 ) of the ray (Fig. 144):
sin ih 1 5
¼ ) vm ¼ ¼ 10 km s1
vh vm sin 30
From this value we find the depth to that point:
vm ¼ 10 ¼ 4 þ 0:1 r ) r ¼ 60 km
Knowing that the rays are circular of radius R,
v0
R ¼ r þ ¼ 60 þ 40 ¼ 100 km
k
As in the previous problem, the epicentral distance (from point E to S) is (Fig. 144):
x¼aþb
where
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v 2
0
a ¼ R2 þ h ¼ 86:60 km
k
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
v 2ffi
0
b ¼ R2 ¼ 91:65 km
k
so
x ¼ 178:25 km
a b
R
ν0/k
E S
V0
h
r
F
i0
ih
Fig. 144
269 Ray theory. Constant and variable velocity
145. A flat medium consists of a layer of thickness H and constant speed of propaga-
tion v on top of a medium of variable speed of propagation v ¼ v0 þ k(z H) where z
is the depth and k is a constant. If there is a focus at the surface:
(a) Write expressions for the epicentral distance x and the travel time t as functions of
the angle of incidence i0 at the surface.
(b) If H ¼ 10 km, k ¼ 0.1 s1, and v0 ¼ 6 km s1, calculate the angle of incidence of a
wave that reaches an epicentral distance of 140 km.
(a) As we saw in Problem 143, for a distribution with a linear increase of velocity with
depth, now in the medium under the layer, v ¼ v0 þ k(z H), the rays are circular with
v0
radius R ¼ þ r where r is the maximum depth of penetration of the ray (Fig. 145).
k
The travel-time of the ray that crosses the layer and penetrates the medium is given by
FP 2 kx0
t¼2 þ sinh1 ð145:1Þ
v0 k 2v0
In the layer of constant velocity the path is a straight line and in the medium it is circular.
The epicentral distance x (from F to S) is (Fig. 145)
x ¼ x0 þ 2H tan i0
The length of the straight ray in the layer is
H
FP ¼
cos i0
Substituting in (145.1):
0
2H 2 1 kx
t¼ þ sinh ð145:2Þ
v0 cos i0 k 2v0
Since the layer has constant velocity the angle i0 is the same at the focus as at the bottom of
the layer at the boundary with the medium. According to Snell’s law
F S
i0 i0 ν0
H x⬘
i0
ν0/k
P Q
r
Fig. 145
270 Seismology
sin i0 1 v0
¼ ) vm ¼ v0 þ kr ¼
v0 v0 þ kr sin i0
where r is the maximum depth reached by the ray in the medium and vm the velocity at that
depth. According to Fig. 145,
0
2 2 v2
x v0 2 v0 v2 v0
þ ¼ þ r ¼ m2 ¼ 2 0 2 ) x0 ¼ 2 cot i0 ð145:3Þ
2 k k k k sin i0 k
The epicentral distance x is given by
2v0
x ¼ x0 þ 2H tan i0 ¼ cot i0 þ 2H tan i0 ð145:4Þ
k
Substituting in (145.2) the expression for x 0 in terms of i0 (143.3) we obtain
2H 2
t¼ þ sinh1 ðcot i0 Þ
v0 cos i0 k
(b) By substituting the given values in (145.4), we obtain
i0 ¼ 45
26
140 ¼ cot i0 þ 2 10 tan i0 )
0:01 i0 ¼ 80:5
146. Beneath a layer of thickness H of velocity distribution v ¼ v0 þ kz there is a semi-
infinite medium of speed of propagation v1 ¼ 2(v0 þ kH).
(a) Determine expressions (as functions of the above parameters) for the critical
distance, the time of intersection of the reflected wave, and the maximum distance
of the direct wave.
(b) For H ¼ 10 km, v0 ¼ 1 km s1, and k ¼ 0.1 s1, calculate these parameters and
plot the travel-time curves.
(a) In a layer of thickness H with variable velocity the epicentral distance x for a
reflected ray is given by
ðH
x ¼ 2 tan i dz
0
For a ray incident at the bottom of the layer at the critical angle, we have
sin ic 1
p¼ ¼
v0 þ kH 2ðv0 þ kHÞ
271 Ray theory. Constant and variable velocity
xmax
xc
n0/k
F S
H
ic
n1
Fig. 146a
Substituting this expression in (146.1) and evaluating the integral, making the change of
variable u ¼ v0 þ kz, we obtain the critical distance
"pffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi#
4ðv0 þ kH Þ 3 v20
xc ¼ 1 ð146:2Þ
k 2 4ðv0 þ kH Þ2
The intercept time for x ¼ 0, corresponding to the time of the reflected vertical ray (p ¼ 0),
is given by
ðH
dz 2 v0 þ kH
ti ¼ 2 ¼ ln ð146:3Þ
0 v0 þ kz k v0
The maximum distance xmax corresponds to the last ray propagated inside the layer
v0
without penetrating into the medium and has a circular path of radius R ¼ H þ
k
(Fig. 146a):
x 2 v 2 rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
max 0 v0 2 2v0
þ ¼ Hþ ) xmax ¼ 2H 1 þ ð146:4Þ
2 k k kH
(b) For the particular case with the values, H ¼10 km, v0 ¼ 1 km s1, and k ¼ 0.1 s1,
the velocity at the bottom of the layer H is
vH ¼ v0 þ kH ¼ 1 þ 10 0:1 ¼ 2 km s1
30
25
20
t (s)
15
10
0 10 20 30 40
x (km)
Fig. 146b
2 v0 þ kH 2 1 þ 0:1 10
ti ¼ ln ¼ ln ¼ 13:9 s
k v0 0:1 1
and the maximum distance for the ray in the layer (146.4) is
pffiffiffi
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
21
xmax ¼ 2 10 1 þ ¼ 20 3 ¼ 34:6 km
0:1 10
The travel-time curve for rays inside the layer is calculated using the expression
2 1 kx
t ¼ sinh ¼ 20 sinh1 ð0:05xÞ
k 2v0
and is represented in Fig. 146b.
273 Ray theory. Constant and variable velocity
F S
Fig. 147
147. A medium has a distribution of velocity with depth of the form v ¼ v0 eaz, with
0 < a <1. Write as functions of the epicentral distance x the expressions for the ray
parameter, travel-time, and maximum depth reached.
If r is the maximum depth reached for a ray with ray parameter p (Fig. 147), the epicentral
distance x is given by
pv0 eaz dz
ðr ðr
pvdz
x ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0 1 p2 v 2 0 1 p2 v20 e2az
r
2 2
¼ sin1 ðpv0 eaz Þ ¼ ½sin1 ðpv0 ear Þ sin1 ðpv0 Þ
a 0 a
1
and, as p ¼ , we have
v0 ear
2 hp i
x¼ sin1 ðpv0 Þ
a 2
From this expression we obtain
1 ax
p ¼ cos
v0 2
The travel-time is given by
ðx
1 x
ð
ax 2 ax
t ¼ pdx ¼ cos dx ¼ sin
0 v0 0 2 v0 a 2
To find the maximum depth of penetration r of a ray arriving at distance x, we write
1 1 ax 1 ax
p¼ ¼ cos ) r ¼ ln cos
v0 ear v0 2 a 2 z
148. In a semi-infinite medium of speed of propagation v ¼ 6 exp , the P-wave
2
emerges with an angle of incidence of 30 . Calculate the difference in arrival times at
a given station of the P-wave and the PP-wave (the wave reflected once at the free
surface). At what angle of incidence does the PP-wave emerge?
For a velocity distribution increasing with depth of the type v ¼ v0ea z (in our case with
v0 ¼ 6, a ¼ 1/2) rays follow a curved path. For a focus on the free surface the ray parameter
p and the travel times t are given by (Problem 147)
1 ax
p ¼ cos
v0 2
ax ð148:1Þ
2
t¼ sin
av0 2
274 Seismology
F S
Fig. 148
For a ray with incidence angle at the surface i0 ¼ 30 , the ray parameter of the direct
P-wave is given by
sin i0 1 1
p¼ ¼ ¼
v0 2 6 12
2 1 4p 1
tP ¼ sin ¼ 0:58 s
1 2 3 2
6
2
The travel time of the reflected PP-wave (Fig. 148) is double that of the direct P-wave
arriving at the distance x/2:
2 1 4p 1
t PP ¼ 2 sin ¼ 0:67 s:
1 2 6 2
6
2
The difference between the two times is
To calculate the incidence angle of the PP-wave, we determine first the ray parameter
p corresponding to the distance x/2:
x 4p 4p
¼ ¼
2 32 6
so
pffiffiffi
1 1 4p 1 3
p ¼ cos ¼
6 2 6 2 12
275 Ray theory. Constant and variable velocity
xmax
xc
S
F
ic
n1
Fig. 149
276 Seismology
u ¼ v0 eaz
du ¼ v0 eaz adz
we find
2 1
sin pv0 eaH sin1 pv0
x¼
a
For the critical angle ic at the bottom of the layer we have
v0 eaH
2 v0 2 vH v0 2
xc ¼ sin1 sin1 ¼ sin1 sin1 ¼ ðic i0 Þ
a v1 v1 a v1 v1 a
(b) The intercept time of the reflected ray corresponding to the vertical ray (x ¼ 0 and p ¼ 0) is
ðH ðH
dz 1 az 2
1 eaH
ti ¼ 2 ¼2 e dz ¼
0 v 0 v0 av0
(c) The maximum distance of a ray contained in the layer is given by (149.2)
ðH
vp 2
xmax ¼ 2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi dz ¼ sin1 pv0 eaH sin1 pv0
ð149:3Þ
0 1 p2 v 2 a
At the point of greatest depth penetration the incidence angle is 90 and, according to
Snell’s law,
1 e0:110
ic ¼ sin1 ¼ 30
5:62
2
1 e0:110 ¼ 12:6 s
ti ¼
0:1 1
2
xmax ¼ cos1 e0:110 ¼ 23:9 km
0:1
2 1 1 1 1
xc ¼ sin sin ¼ 6:8 km
0:1 2 5:62
277 Ray theory. Spherical media
150. Assume that the Earth consists of two concentric regions of constant velocity: the
core of radius R/2 and the mantle. The speed of propagation in the core is twice that of
the mantle. Calculate:
(a) The maximum angular distance of the direct ray in the mantle.
(b) The critical angular distance of the refracted ray in the core.
(c) Plot the paths of the waves that propagate through the Earth’s interior, and the
travel-time curves of these waves in units of R/v, where v is the speed of propaga-
tion in the mantle.
(a) The travel time of the direct ray in the mantle in terms of the angular distance is
given by
R
t1 ¼ 2 sin
v 2
where 0
D
Dmax and Dmax is the maximum distance for a ray contained in the mantle.
According to Fig. 150a the last ray which propagates in the mantle without entering the
core corresponds to angular distance Dmax which in our case is
R
max 2
cos ¼ ) max ¼ 120
2 R
(b) The critical angle for a ray incident at the core is
sin ic 1
¼ ) ic ¼ 30
v 2v
To calculate the critical distance Dc we consider the relation (Fig. 150b)
S
F∗
R
n Δmax
F∗
2n R
Δ/2
R/2
Fig. 150a
278 Seismology
F S
α
ic
P
n
β
2n R
q
R/2
Δc
Fig. 150b
S
F
P
n
R
2ν
Δ/2
R/2
Fig. 150c
y þ b þ a ¼ 180
sin a sin b
¼
R R
2
ic þ b ¼ 180 ) b ¼ 150 ) a ¼ 14:5 ) y ¼ 15:5
c ¼ 2y ¼ 31
(c) The travel time of a ray reflected at the mantle–core boundary is (Fig. 150c)
FP PS FP
t2 ¼ þ ¼2 ð150:1Þ
v v v
279 Ray theory. Spherical media
S
i1
Q
P i2
i1 Δ
i2 α
F
2n n
R/2 R
Fig. 150d
2 R R a
FP ¼ R2 þ 2R cos
4 2 2
R a
PQ ¼ 2 sin
2 2
Because a ¼ 180 – 2i2, we obtain
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2R 5 R
t3 ¼ sin þ i2 þ cos i2 ð150:3Þ
v 4 2 2v
for values of the incidence angle 0 < i < ic, corresponding to distances Dc < D < 180 .
The relation between the incidence angle i1 and angular distance D is given by
280 Seismology
i 1 ( ) i 2 ( ) D ( ) t3 (R/v)
0 0 180.0 1.50
10 20.3 169.9 1.53
20 43.2 153.4 1.60
30 90 89.0 1.46
ð150:4Þ
sin i1
sin i1 ¼ 2 cos i1 i2 ) ¼ 2 i1 i2 cos1
2 2
Using Equations (150.2), (150.3), and (150.4) we can calculate the travel times of the direct
ray in the mantle, the reflected ray, and the transmitted ray through the core. Some values
for the transmitted rays in the core are given in the table.
The travel-time curves for direct rays (1), reflected rays (2), and rays refracted in the core
(3) are shown in Fig. 150e.
2.0
1.5
2
t (R/v)
1.0
0.5
0.0
0 50 100 150
Δº
Fig. 150e
281 Ray theory. Spherical media
151. Assume that the Earth consists of two concentric regions of constant velocity: the
core of radius R/2 and the mantle. The speed of propagation in the core is half that
of the mantle. Plot the travel-time curves of the waves that propagate in the interior of
the Earth in units of R/v where v is the speed of propagation in the mantle.
This problem is similar to Problem 150, but now the velocity of the core is less than that of
the mantle. In the mantle we have direct and reflected rays. As in Problem 150 the
maximum angular distance for the direct wave is 120 . The travel times for the direct
(t1) and reflected (t2) rays are
R
t1 ¼ 2 sin ð151:1Þ
v 2
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R
t2 ¼ 5 4 cos ð151:2Þ
v 2
Since the velocity of the core is less than that of the mantle there is no critical angle. All
rays incident at the core are refracted into it. According to Snell’s law the refracted angle i2
is less than the incident angle i1 (Fig. 151a):
sin i1 sin i2
¼ v
v
2
1
sin i2 ¼ sin i1
2
P
i1
R
i2
a
F∗ i2
R/2
Q
Δ
n n/2 i1
Fig. 151a
282 Seismology
The travel-time for a ray crossing the mantle and the core is (Fig. 151a)
FP PQ QS 2FP 2PQ
t3 ¼ þ v þ ¼ þ
v v v v
2
where
R2 R2
2 R
FP ¼ R2 þ 2R cos a ¼ R2 þ R2 cos 90 þ i2
4 2 4 2
2
2 5R
FP ¼ þ R2 sin i2
4 2
R
PQ ¼ 2 cos i2 ¼ R cos i2
2
so
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2R 5 2R
t3 ¼ þ sin i2 þ cos i2 ð151:3Þ
v 4 2 v
3.5
3
3.0
2.5
2.0
t (R / n )
1.5
1.0
0.5
1
0
0 25 50 75 100 125 150 175
Δ (º)
Fig. 151b
283 Ray theory. Spherical media
The relation between the incidence angle at the mantle–core boundary, i1, and the angular
distance, D, of a ray which crosses the core is
sin i 1 i 2 þ 90
sinð180 i1 Þ 2
¼
R R=2 ð151:4Þ
sin i1 1
sin i1 ¼ 2 cos i1 i2 þ ) ¼ 2 cos1 i1 þ sin1 sin i1
2 2 2
i 1 ( ) i2 ( ) D ( ) t3 (R/v)
0 0 180.0 1.50
10 20.3 149.4 1.48
20 43.2 114.0 1.40
30 90 31.0 1.07
The travel-time curves for rays that are direct (1), reflected (2), and refracted in the core (3)
are shown in Fig. 151b.
152. Consider a spherical Earth of radius R formed by two hemispherical media of
constant velocities of propagation v and 2v. For a focus on the surface of the
hemisphere of velocity v at the point of intersection of the diameter perpendicular
to the plane that separates the two media, calculate the travel times and travel-time
curves of the direct, reflected, and critical refracted waves at the surface of separation
of the two media, in units of R/v. Calculate the expression for the travel time of waves
that propagate through the medium of speed of propagation 2v.
The travel time for angular distances D
90 are given by (Fig. 152a)
FS R
t1 ¼ ¼ 2 sin ð152:1Þ
v v 2
The travel time of the ray reflected at the plane boundary between the two hemispheres is
(Fig. 152a)
FP PS F0 S
t2 ¼ þ ¼
v v v
According to Fig. 152a (triangle OSF0 ) the relation between the angles a and D is
ð180 Þ þ 2a ¼ 180 ) a ¼
2
0
SS SS0
sin a ¼ 0
) F0 S ¼
FS sin a
R sin
SS0 ¼ R sin ) F0 S ¼ ¼ 2R cos
sin a 2
284 Seismology
S
S⬘
∆ α
O P
2ν
R
α
F⬘
Fig. 152a
FA AB BS
t3 ¼ þ þ ð152:3Þ
v 2v v
According to Fig. 152b
R
FA ¼
cos ic
SS0 R cos
BS ¼ ¼
cos ic cos ic
0 0
AB ¼ OS OA BS
0
OS ¼ R sin
OA ¼ R tan ic
0
BS ¼ BS sin ic
285 Ray theory. Spherical media
F
∗
S
R ic
ic ic
Δ
n ic
P
O A B S⬘
2n
F⬘
Fig. 152b
Substituting ic ¼30 :
2R 2R cos R R cos
FA ¼ pffiffiffi ; BS ¼ pffiffiffi ; AB ¼ R sin pffiffiffi pffiffiffi
3 3 3 3
and substituting in (152.3) we obtain, for Dc
D
90 ,
Rh pffiffiffi i
t3 ¼ sin þ 3ð1 þ cos Þ ð152:4Þ
2v
The critical distance can be calculated from ic ¼ 30 using Fig. 152a (triangle OSP) and a ¼ D/2
n Δ
O P
2n
i⬘
R S
P⬘
F⬘
Fig. 152c
2.0
2
1.5 3
t (R / n )
1.0
1
0.5
0
0 20 40 60 80
Δ (°)
Fig. 152d
287 Ray theory. Spherical media
3R
FP ¼ þ R2 2R cos
4 4
Then, we obtain
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R 25 3
t1 ¼ cos ð153:1Þ
v 16 2
The maximum distance for the direct ray is
max ¼ 1 þ 2
R
cos 1 ¼ 2 ) 1 ¼ 48:2
3R
4
R
cos 2 ¼ 2 ) 2 ¼ 60:0
R
max ¼ 48:2 þ 60:0 ¼ 108:2
P S
Δ2
ΔMAX
Δ1 Δ 2n n
F
∗
R/4 R/2 O R
Fig. 153a
288 Seismology
i⬘
r
b
P
r Δ2
a 2n n
F i Δ1
∗
R /4 R/2 O R
Fig. 153b
FP PS
t2 ¼ þ
v v
The distances FP and PS are expressed in terms of R, D1, and D2 using the cosine law in
triangles FOP and SOP (Fig. 153b):
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
2
3 R R 3R
FP ¼ R þ 2 cos 1
4 2 2 4
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 ffi
R R
PS ¼ R2 þ 2 cos 2
2 2
Then, we obtain
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
13 3 5
R cos 1 R cos 2
16 4 4
t2 ¼ þ ð153:2Þ
v v
Now we need to express D1 and D2 in terms of the take-off angle i at the focus (F). Using
Snell’s law for a spherical medium, we relate i and i0 , the incidence angle at the station (S):
3R R
sin i sin r R sin i0 3
4 ¼2 ¼ ) sin i0 ¼ sin i ð153:3Þ
v v v 4
According to Fig. 153b for triangle FOP we have
i þ a þ 1 ¼ 180
sin a sin i
¼
3R R
4 2
and we obtain
2 sinð1 þ iÞ ¼ 3 sin i ð153:4Þ
289 Ray theory. Spherical media
1.4
0.8
2
0.8
t (R / n )
0.8
1
0.6
0.4
0 20 40 60 80 100
Δ (⬚)
Fig. 153c
i ( ) i 0 ( ) D1 ( ) D2 ( ) D ( ) t2 (R/v)
0 0 0 0 0 0.75
10 7.5 5.1 7.6 12.7 0.76
30 22.0 18.6 26.6 45.2 0.92
40 28.8 34.6 45.8 80.4 1.19
41.8 30.0 47.1 58.9 106.0 1.41
For angular distance D greater than 108.2 there are no reflected rays. The travel-time
curves corresponding to the direct (1) and reflected (2) rays are shown in Fig. 153c.
290 Seismology
154. Consider a spherical Earth of radius R ¼ 3000 km and constant P-wave speed of
propagation of 4 km s1. Within it there is a core of radius R/2 and constant velocity
v1. At a station at epicentral distance D from an earthquake with focus at the surface,
the observed time interval is tS-P ¼ 547.0 s. Given that Poisson's ratio is 1/6, and that
the arrival of the P-wave is at 12 h 23 m 20.4 s, calculate:
(a) The epicentral distance.
(b) The time of the earthquake.
(a) For a spherical Earth of constant velocity the travel time of the direct ray is given by
FO0 R
t¼2 ¼ 2 sin ð154:1Þ
v v 2
Taking into account the presence of the core the maximum distance for the direct ray is (Fig. 154)
R
max 2
cos ¼ ) max ¼ 120
2 R
Since Poisson’s ratio is 1/6 we have
1 l
s¼ ¼ ) m ¼ 2l
6 2ðl þ mÞ
and we can calculate the relation between the velocities of the P-wave (a) and the S-wave (b),
rffiffiffiffiffiffiffiffiffiffiffiffiffiffi 9
l þ 2m >
a¼ > rffiffiffi
5
rr
=
)a¼ b ) b ¼ 2:53 km s1
m >
ffiffiffi
2
b¼ >
r
;
Using (154.1) and assuming the same path for P- and S-waves, from the time interval S-P
we obtain the distance:
S-P 1 1
t ¼ 2R sin ) ¼ 77:7
2 b a
O⬘
S
Δmax
n1 n
F∗ Δ
R/2 O R
Fig. 154
291 Ray theory. Spherical media
Δ2
Δ1 ν2 ν1
F Δ
h R/2 O R
Fig. 155
292 Seismology
For a point on the surface at distance D, the S-P time interval implies, assuming the same
path for P- and S-waves,
FP FP FP
t S-P ¼ ¼ ða b Þ ð155:1Þ
b a ab
From the value of Poisson’s ratio the relation between the P and S velocities is given by
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffi rffiffiffi
1 l m l þ 2m 7m 7
s¼ ¼ )l¼ )a¼ ¼ ¼ b
8 2ð l þ m Þ 3 r 3r 3
Substituting in (155.1) the S-P interval equal to 600 s we obtain the length of the ray:
ab
FP ¼ 600 ¼ 5212 km
ð a bÞ
Using the cosine law for triangle FOP
2
FP ¼ ðR hÞ2 þ R2 2RðR hÞ cos
2
FP ðR hÞ2 R2
cos ¼
2RðR hÞ
2
If h ¼ R then D ¼ 106 , but this result is not possible because the maximum distance of
5 R
the direct ray for that depth is 94 . If h ¼ then D ¼ 86 , this result is possible because
10
this distance is less than the maximum distance. The depth is, then, 400 km.
156. Consider the Earth of radius R and constant velocity v with a core of radius 6R/10
and constant speed of propagation 2v. An earthquake occurs with focus at 8R/10 from
the centre of the Earth. A wave emerges from that focus with a take-off angle of 15 .
(a) Will it pass through the core?
(b) What epicentral distance will it reach?
(c) What will be the travel time of the wave (in units of R/v)?
(a) First we calculate the maximum epicentral distance for a ray which doesn’t
penetrate the core. According to Fig. 156a the maximum distance is
max ¼ 1 þ 2
6
R
cos 1 ¼ 10 ) 1 ¼ 41:4
8
R
10
6
R
cos 2 ¼ 10 ) 2 ¼ 53:1
R
max ¼ 41:4 þ 53:1 ¼ 94:5
From this value we calculate the take-off angle ih for this ray:
1 þ ih ¼ 90 ) ih ¼ 48:6
293 Ray theory. Spherical media
P S
Δ2
Δ1 2n n
F ih Δ
6R/10 O R
8R/10
Fig. 156a
S
i
B
i0
Δ3
A i2
i i2
ih Δ1 Δ2
F 2n n
6R/10 O R
8R/10
Fig. 156b
For take-off angles less than 48.6 the rays travel through the core.
Since the velocity in the core is greater than in the mantle, to find out which rays
penetrate into the core, we also need to know the critical angle. Rays with incidence angle
at the core–mantle boundary with i > ic are totally reflected and don’t penetrate into the
core. According to Snell’s law the critical angle is given by
sin ic 1
¼ ) ic ¼ 30:0
v 2v
We calculate, using Snell’s law, the angle of incidence i corresponding to the take-off angle
of 15 (Fig. 156b):
8 6
R sin ih R sin i
10 ¼ 10 ) i ¼ 20:2
v v
Since the incidence angle i (20.2 ) is less than the critical angle (30 ) and less than the
angle corresponding to the maximum distance (48.6 ), the ray with take-off angle of 15
penetrates into the core.
294 Seismology
(b) Applying Snell’s law we find the angle of the transmitted ray in the core i2 (Fig. 156b):
8 6
R sin ih R sin i2
10 ¼ 10 ) i2 ¼ 43:7
v 2v
By consideration of triangles FOA and AOB, we determine D1 and D2 (Fig. 156b):
and using Snell’s law we determine i0 the incidence angle at the surface and D3:
6 6
R sin i2 R sin i R sin i
10 ¼ 10 ¼
o
) io ¼ 11:9
2v v v
3 ¼ i io ¼ 8:3
The epicentral distance of the ray is
¼ 1 þ 2 þ 3 ¼ 106
(c) The travel time is
FA AB BS
t¼ þ þ
v 2v v
where
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
2 ffi
8R 6R 8 6
FA ¼ þ 2 R R cos 1 ¼ 0:21 R
10 10 10 10
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
2
6 6 6 6 2
AB ¼ R þ R 2 R cos 2 ¼ 0:87 R
10 10 10 10
s
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 ffi
6 6
BS ¼ R þ R2 2 RR cos 3 ¼ 0:42 R
10 10
so
R
t ¼ 1:07
v
157. Assume a spherical Earth of radius R ¼ 6000 km and constant S-wave speed of
propagation 4.17 km s1. Poisson's ratio is 1/4. At a station at epicentral distance 60
an earthquake is recorded with a time interval t S-P ¼ 554 s. Calculate the depth of the
earthquake.
Given that Poisson’s ratio is 0.25, the P-wave velocity is
1 l pffiffiffi
s¼ ¼ ) l ¼ m ) a ¼ 3b ¼ 7:22 km s1
4 2ðl þ mÞ
295 Ray theory. Spherical media
R–h
Δ
Fig. 157
From the time interval S-P we can calculate the length of the ray FS (Fig. 157):
FS FS ab
t S-P ¼ ¼ FS
b a ab
ð157:1Þ
t S-P ab
FS ¼ ¼ 5469 km
ab
The distance along the ray in terms of the angular epicentral distance D, using the cosine
law, is
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
FS ¼ R2 þ ðR hÞ2 2RðR hÞ cos
We substitute in (157.2) the values R ¼ 6000 km and D ¼ 60 and solve for h, finding two
possible solutions:
h1 ¼ 4706 km
h2 ¼ 1294 km
158. Assume a spherical Earth of radius R and P-wave velocity which can be
expressed by the equation v(r) ¼ a br2. The speed of propagation at the surface
of the Earth is v0 and at the centre of the Earth it is 2v0. What angular distance D does
a wave reach which penetrates to a depth equal to half the Earth’s radius?
If the velocity distribution inside the Earth is v(r) ¼ a br2, the ray paths are circular with
radius given by (Fig. 158)
296 Seismology
7/4 R
7/4 R
F S
Δ/2 Δ/2
R/2 O R
Fig. 158
r
r¼ ð158:1Þ
dv
p
dr
From the conditions of the problem
r ¼ R ) v ¼ v0 ¼ a bR2
r ¼ 0 ) v ¼ 2v0 ¼ a
and
v0 )
r2
b¼ 2
R ) v ¼ v0 2 2
a ¼ 2v0 R
The radius of curvature of the ray which penetrates to r ¼ R/2 is that corresponding to the
ray parameter
r0 R
p0 ¼ 0 ¼ 0 ð158:2Þ
v 2v
The velocity at depth R/2 is
0 0 12 1
R
0
B 2C C 7
@2 @RA A ¼ v0 4
v ¼ v0 B
B C
2 1 p
¼ cos ð159:1Þ
1þa 0
where
r R
¼ ) 0 ¼
v v0
The maximum distance for a ray which travels only through the mantle, that is that reaches
depth R/2, can be calculated from the velocity at that depth, vm:
0 11
2 R R
BRC pffiffiffi 1 R
vm ¼ v0 @ A ¼ v0 2 ) p ¼ 2 ¼ 2pffiffiffi ¼ pffiffiffi ¼ p
R vm v0 2 2 2 v0
2
298 Seismology
Δc
F ∗ S
ic i0
Δ/2 Δ/2
R/2 O R
Fig. 159a
By substitution in (159.1) we obtain, for the maximum distance,
1 R
1 0
pffiffiffi
2 B2 2 v0 C
m ¼ cos1 B
@ R A ¼ 92:4
C
1
1þ
2 v0
(b) The critical angle of a reflected ray at the mantle-core boundary applying Snell’s law is
sin ic 1
pffiffiffi ¼ ) ic ¼ 20:7
v0 2 4v0
The take-off angle at the surface i0 for this ray is found by again applying Snell’s law:
R
R sin i0 2 sinð20:7 Þ
¼ pffiffiffi ) i0 ¼ 7:2
v0 v0 2
(c) To find the critical distance we use the expression
ð r0
p dr
c ¼ 2 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð159:2Þ
rp r p2
2
where
R
rp ¼
2
r0 ¼ R
r0 sin i0 R
p¼ ¼
v0 8v0
r r r3=2
¼ ¼ 1 ¼ pffiffiffi
v v R 2 v0 R
0 r
1.2
1.0
0.8
t (R/n)
0.6
0.4
0.2
0 Δc Δm
0 20 40 60 80
Δ (⬚)
Fig. 159b
160. A spherical medium of radius R has a constant speed of propagation v0 from the
surface down to R/2, and from R/2 to the centre a core of variable speed of propaga-
1=2
R
tion v ¼ v0 .
r
(a) What value should i0 have for the waves to penetrate into the core?
(b) Calculate the epicentral distance reached by a wave leaving a focus at the surface
at angle i0.
(a) The velocity at the top of the core (r ¼ R/2) is
300 Seismology
F
∗ l
P
i0
Δ1
Δ1 Δ2
n0
R/2 O R
Fig. 160
0 11
2
BRC pffiffiffi
v1 ¼ v0 @ A ¼ v0 2
R
2
Applying Snell’s law we find the critical angle ic for incident rays at the core (Fig. 160):
R
R sin ic
¼ pffiffi2ffi ) ic ¼ 45
v0 2v 0
The take-off angle i0, for a focus at the surface corresponding to the critical angle, using
Snell’s law, is
R
R sin i0 2 sin ic
¼ ) i0 ¼ 20:7
v0 v0
The rays that penetrate into the core must leave the focus with take-off angles less than
20.7 .
(b) For a ray with take-off angle i0 which penetrates the core the epicentral distance is the
sum of that corresponding to the part that has travelled through the mantle, D1, plus the
part that has travelled through the core, D2:
¼ 21 þ 2
Since in the core the velocity varies with depth with the law given in the problem, the
epicentral distance is given by
2 1 p
2 ¼ cos
1þa 1
where a is the exponent of the velocity distribution
1
v ¼ v0 ra ) a ¼
2
301 Ray theory. Spherical media
and p is the ray parameter, which can be obtained using Snell’s law:
R
R sin i0 2 sin i1
p¼ ¼ pffiffiffi
v0 2v 0
and
R
r
¼ ) 1 ¼ pffiffi2ffi
v 2v 0
Then, we find
0 1
R sin i0
2 1 B v0 C
B C 4 1 pffiffiffi
2 ¼ cos @ R A 3 ¼ cos 2 2 sin i0
1 þ 12 pffiffiffi
2 2v 0
The distance D1 can be determined using the sine and cosine laws for the triangle FOP
(Fig. 160):
R
2 ¼ l ) ¼ sin1 2l sin i0
1
sin i0 sin 1 R
R2 1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
l 2 þ R2 2Rl cos i0 ¼ ) l ¼ R cos i0 4R2 cos2 i0 3R2
4 2
and we find the expression in terms of i0:
ffi
1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4 pffiffiffi
¼ 2sin1 2 cos i0 4cos2 i0 3 sin i0 þ cos1 2 2 sin i0
2 3
161. Consider a spherical Earth of radius 6000 km and surface velocity of 6 km s1, with
pffiffi
a velocity distribution of the type vðrÞ ¼ a= r. At the distance reached by a wave
emerging at a take-off angle of 45 from a focus on the surface, calculate the interval
between the arrival times of the direct P- and reflected PP-waves (the PP-wave is one that
is reflected at the surface at the midpoint between
the focus and the point of observation).
a
R
The velocity distribution is of the type v ¼ v0 where
r
a pffiffiffi
r ¼ R ) v ¼ v0 ¼ pffiffiffi ) a ¼ v0 R
R
12
R
v ¼ v0
r
and the ray parameter p for a ray with take-off angle of 45 is
1
6000 pffiffiffi
R sin i0 2
p¼ ¼ ¼ 707 s
v0 6
302 Seismology
For this type of velocity distribution the relation between the ray parameter and the
epicentral distance is
1þa
p ¼ 0 cos ð161:1Þ
2
In this problem the value of 0 is
r R
¼ ) 0 ¼ ¼ 1000 s
v v0
Substituting the values in (161.1) we obtain the distance for the ray with take-off angle
of 45 :
3
707 ¼ 1000 cos ) ¼ 60
4
The PP-wave which arrives at D ¼ 60 travels twice the distance which a P-wave does for a
distance of 30 . For this type of velocity distribution the travel time for a distance D is
given by
20 ð1 þ aÞ
t¼ sin
1þa 2
In our case for the P- and PP-waves at distance 60 we substitute the values of the problem
and find
tP ð60 Þ ¼ 943 s
tPP ð60 Þ ¼ 2tP ð30 Þ ¼ 1021 s
162. In an elastic spherical medium of radius r0, the velocity increases with depth
according to v ¼ arb. If v0 ¼ 6 km s1, r0 ¼ 6000 km, and, at a point at distance
D ¼ 90 , the slope of the travel-time curve is 500 s, determine:
(a) The value of b.
(b) The value of rp and of vp of the wave reaching an epicentral distance of 90 .
(a) For this type of velocity distribution the travel time in terms of the epicentral
distance is given by
20
t¼ sin ð1 þ bÞ ð162:1Þ
1þb 2
As we know the velocity at the surface,
r r0
¼ ) 0 ¼ ¼ 1000 s
v v0
303 Ray theory. Spherical media
The ray parameter p is known, because it is equal to the slope of the travel-time curve
which for D ¼ 90 is given as 500 s. Using the relation between p and D for this type of
velocity distribution,
dt
p¼ ¼ 0 cos ð1 þ bÞ
d 2
h pi h pi 1 p p 1
500 ¼ 1000 cos ð1 þ bÞ ) cos ð1 þ bÞ ¼ ) ð1 þ bÞ ¼ ) b ¼
4 4 2 4 3 3
(b) At the point of greatest penetration rp for D ¼ 90 , we have the relation
rp sin 90 rp
p¼ ¼ ) rp ¼ pvp
vp vp
and also
1
r0 3
vp ¼ v0
rp
From these two equations we obtain rp and vp:
1 1 3 1 3
rp ¼ v0 pr03 rp 3 ) rp ¼ v40 r04 p4 ¼ 3564 km
rp 3564
vp ¼ ¼ ¼ 7:1 km s1
p 500
163. Consider a spherical Earth of radius R, the northern hemisphere with a constant
speed of propagation v0, and the southern hemisphere with a speed of propagation of
1
R 2
v ¼ v0 .
r
(a) Calculate the travel time of seismic waves for a focus on the equator and stations
on the same meridian.
(b) In which hemisphere does the wave at a distance of 60 arrive first?
(a) In the northern hemisphere the velocity is constant and the rays have straight paths
and their travel time is (Fig. 163)
2R
tN ¼ sin 0 < < 90 ð163:1Þ
v0 2
In the southern hemisphere the velocity increases with depth and the rays have curved
paths. Their travel time is given by
20
tS ¼ sinð1 þ bÞ ð163:2Þ
1þb 2
where
b
R R 1
0 ¼ and v ¼ v0 )b¼
v0 r 2
Substituting in (163.2) we obtain for the travel time in the southern hemisphere
304 Seismology
S1
Δ n0
F ∗
S2
Fig. 163
4R 3
tS ¼ sin 0 < < 90 ð163:3Þ
3 v0 4
(b) The travel times for waves in the northern and southern hemisphere are given by
Equations (163.1) and (163.3). By substitution of D ¼ 60 we obtain
R
tN ¼
v0
pffiffiffi
S R2 2 R
t ¼ ¼ 0:47
v0 6 v0
The waves arrive first in the southern hemisphere.
164. Consider a spherical medium of radius R consisting of two concentric regions
(mantle and core), the core of radius R/2. The speeds of propagation are v ¼ ar1=2
for the mantle and v ¼ aR1=6 r1=3 for the core. The surface velocity is v0. For a wave
leaving a focus with angle of incidence 14.5 , calculate the angular distance D at
which it reaches the surface.
We calculate a by applying the boundary conditions
1 1
r ¼ R ) v ¼ v0 ) v0 ¼ aR 2 ) a ¼ v0 R2
1
R 2
Mantle : v ¼ v0
r
1
R 3
Core : v ¼ v0
r
We determine the ray parameter corresponding to the ray with take-off angle i0 ¼ 14.5 ,
using Snell’s law (Fig. 164a):
r sin i R sin 14:5 R
p¼ ¼ ¼
v v0 4v0
305 Ray theory. Spherical media
F i0
∗
i
i0
i2 Δ1
Δ1 Δ2
R/2 O R
Fig. 164a
¼ 21 þ 2
306 Seismology
F Δ3
∗
Δ1 Δ1
Δ4
O R
Fig. 164b
2 p
2 ¼ cos1
1þb 0
where
R
p¼
4v0
R
0 ¼ 2 1
v0 23
and where p is the ray parameter and ¼ r/v. Substituting the values we obtain
2 ¼ 76:4
To calculate D1 we suppose that there is no core and a ray with take-off angle i0 ¼ 14.5
would arrive at distance D3 which is related with D1 by (Fig. 164b)
21 ¼ 3 4
The distances D3 and D4 can be determined using the equation
2 1 p
¼ cos
1þb 0
where for D3
R
0 ð RÞ ¼
v0
R
3
R 2 R 2
¼ ¼
0 11 v 0 2
2 2
BR C
v0 @ A
R
2
307 Surface waves
and we obtain
0
1
R
2 B4v0 C
3 ¼ 1
cos1 B
@ R A ¼ 100:6
C
1þ 2
v
and by similar substitutions for D4
0 1
B R C
2 1 B 4v0 C
B C
4 ¼ cos B C ¼ 60:0
1 3
1þ @ R A
2 2 2
v0
Then, 2D1 ¼ 100.6 – 60 ¼ 40.6 and the epicentral distance is
¼ 40:6 þ 76:4 ¼ 117:0
Surface waves
Then,
u1 ¼ 0 ) ikA expð0:85kx3 Þ 0:39ikB expð0:39kx3 Þ ¼ 0 ð165:1Þ
so
2p 2p
k¼ ¼ ffi 0:1 km1
l TcR
We can write B in terms of A using the boundary condition of zero stress at the free surface:
t31 ¼ 0jx3 ¼0 ) 2rA 1 s2 B ¼ 0
so
B ¼ 1:47iA
Substituting in (165.1) we obtain the value of x3:
At 12 km depth u1 is null and for greater values of depth the particle motion is prograde
while for lesser values of depth it is retrograde.
166. Given a layer of thickness H and shear modulus m ¼ 0 on top of a half-space or
semi-infinite medium in which l ¼ 0, study (without expanding the determinant)
whether there exist surface waves that propagate in the x1-direction. Are they
dispersive waves?
In the liquid layer ( m ¼ 0) the P- and S-velocities are
sffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffi
m 0 l0 þ 2m0 l0
b0 ¼ 0 ¼ ) a0 ¼ ¼
r r r
and in the solid half-space
sffiffiffiffiffiffi
2m pffiffiffi
l¼0)a¼ ¼ 2b
r
The relation between the stress and strain is
tij ¼ lydij þ 2meij
1
where eij ¼ 2 ui; j þ uj;i .
t0ii ¼ l0 ðe011 þ e022 þ e033 Þ
In the layer: m0 ¼ 0 )
t0ij ¼ 0
In the half-space: l ¼ 0 ) tij ¼ 2 meij
If there are surface waves propagating in the x1-direction, their displacements in terms of
the potentials are given by (Fig. 166)
u1 ¼ ’;1 c;3
u2 ¼ u 2
u3 ¼ ’;3 þ c;1
309 Surface waves
x3
H x1
μ⬘ = 0
λ=0
Fig. 166
The boundary conditions at the free surface are null normal stresses:
0
t33 ¼ 0
x3 ¼ H )
t031 ¼ 0
and at the boundary between the liquid layer and the solid half-space continuity of the
normal component of the displacement and stress and zero tangential stresses,
u3 ¼ u03
8
>
>
< t ¼ t0
>
33 33
x3 ¼ 0 ) 0
t
> 32
>
> ¼ t 32 ¼ 0
: 0
t31 ¼ 0
In the liquid layer there is only the P-wave potential ’. Taking (x1, x3) as the incidence
plane
c ¼ C expðiksx3 þ ik ðx1 ct ÞÞ
u2 ¼ E expðiksx3 þ ik ðx1 ct ÞÞ
’ ¼ D expðikrx3 þ ik ðx1 ct ÞÞ
rffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
r¼ 1
a2
sffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
s¼ 1
b2
In the layer we have only guided P-waves and r0 is real, while in the half-space for surface
waves, r and s must be imaginary. Then a > b > c > a0 must be satisfied.
310 Seismology
x3 ¼ H
0 0
t033 ¼ 0 ) A 1 þ r02 eikr H þ B 1 þ r02 eikr H ¼ 0
x3 ¼ 0
t31 ¼ 0 ) 2Dr C þ Cs2 ¼ 0
u03 ¼ u3 ) Ar0 Br0 ¼ Dr þ C
t033 ¼ t33 ) l0 1 þ r02 ðA þ BÞ ¼ 2m Dr2 þ Cs
AeikrH þ BeikrH ¼ 0
AB¼0)A¼B
311 Surface waves
x3
liquid layer
0 x1
rigid half-space
Fig. 167a
x3
O
H
0 x1
P
Fig. 167b
Then, 2A cos krH ¼ 0. For waves propagating in the layer, r must be real and c > a. The
solution is given by
1
krH ¼ n þ p; n ¼ 0; 1; 2; . . . ð167:2Þ
2
The solution can also be found by the method of constructive interference. The condition
of constructive interference implies that waves coinciding at a given wavefront (AB) are in
phase, that is, the distance along the ray must be an integer multiple of the wavelength,
taking into account possible phase shifts (Fig. 167b). In our case on the free surface,
x3 ¼ H, there is a phase shift of p (l/2) and we write the condition as (Fig. 167b)
la
AP þ PQ þ QB ¼ nla
2
or
2p
ðA P þ P Q þ Q BÞ p ¼ 2pn
la
312 Seismology
Substituting
A P þ P Q þ Q B ¼ 2H cos i
we obtain
2p
2H cos i p ¼ 2pðn þ 1Þ ð167:3Þ
la
According to Snell’s law,
rffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffi
a a2 a c 2 a
sin i ¼ ) cos i ¼ 1 2 ¼ 2
1¼ r
c c c a c
a
and ka ¼ k.
c
Substituting in (167.3), we obtain the same solution obtained in (167.2):
a 1
ka Hr ¼ n þ p
c 2
This expression can also be written as
rffiffiffiffiffiffiffiffiffiffiffiffi " #12
1 2 p2
1 oH c2 1 1
krH ¼ n þ p ) 1 ¼ nþ p)c¼ 2 nþ ð167:4Þ
2 c a2 2 a 2 H 2 o2
The fundamental mode (FM) corresponds to n ¼ 0, and n 1 to the higher modes (HM).
In the FM and the higher modes, the frequency oc corresponding to the zero in the
denominator in (167.4) is called the cut-off frequency, as there are no values of c for o <
oc. For a mode of order n the cut-off frequency is given by
1
p nþ a
2
oc ¼
H
The dispersion curve is shown in Fig. 167c.
c
FM 1 HM 2 HM
ω
πα 3πα 5πα
2H 2H 2H
Fig. 167c
313 Surface waves
iksH iksH 1
Eðe þe Þ ¼ 0 ) cosðksHÞ ¼ 0 ) ksh ¼ nþ p
2
Substituting s and putting k ¼ o/c:
sffiffiffiffiffiffiffiffiffiffiffiffiffi ! 1
1 2 p2 2
Ho c2 1 1
1 ¼ n þ p)c¼ n þ ð168:1Þ
c b2 2 b2 2 H 2 o2
This equation give us, for the SH component, the frequency dependence of the velocity c(o).
The boundary conditions for P and SV are similarly
x3
H x1
α, β, µ
rigid medium
Fig. 168a
314 Seismology
x3 ¼ H ) t31 ¼ 0; t33 ¼ 0
x3 ¼ 0 ) u1 ¼ 0; u3 ¼ 0
where
t33 ¼ lðe11 þ e33 Þ þ 2me33
t31 ¼ 2me31
u1 ¼ ’;1 c;3
u3 ¼ ’;3 þ c;1
Substituting the expression for the potentials we obtain
ðl þ 2mÞðr2 AeikrH þ r2 BeikrH Þ þ lðAeikrH þ BeiksH Þ þ 2msðCeiksH DeiksH Þ ¼ 0
2rðAeikrH BeikrH Þ þ ð1 s2 ÞðCeiksH þ DeiksH Þ ¼ 0
ðA þ BÞ sðC DÞ ¼ 0
rðA BÞ þ C þ D ¼ 0
For a solution we put the determinant of the system of equations equal to zero:
1 1 s s
r r 1 1
2 iksH ¼ 0 ð168:2Þ
ikrH ikrH 2 iksH
2re 2re ð1 s Þe ð1 s Þe
½l þ r2 ðl þ 2mÞeikrH ½l þ r2 ðl þ 2mÞeikrH 2mseiksH 2mseiksH
Expanding the determinant and putting it equal to zero, we obtain the dependence with
frequency of the velocity c(o) which gives us the dispersion curve.
For the wave with SH component the dispersion curve is given in Fig. 168b:
!12
1 2 p2
1
c¼ nþ
b2 2 H 2 o2
c
FM 1 HM 2 HM
ω
πβ 3πβ 5πβ
2H 2H 2H
Fig. 168b
315 Surface waves
For n ¼ 0 the curves correspond to the fundamental mode and for 1
n to the higher
modes. For all modes, including the fundamental mode, there is a cut-off frequency oc ¼
(nþ1)pb/2H, with n ¼ 0 for the fundamental mode and n 1, for higher-order modes.
169. For a liquid layer of thickness H with a rigid medium above and below, derive
the dispersion equation c(v) of the fundamental and higher modes. For the FM, at
what height above the layer is the motion circular?
Given that the medium is a liquid, motion is represented only by the scalar potential f:
’ ¼ ðA exp ikrx3 þ B expðikrx3 ÞÞ exp ik ðx1 ct Þ ð169:1Þ
rffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
where r ¼ 1.
a2
The boundary condition at the two boundaries between the liquid and rigid solid is that
the normal component of the displacement is null (Fig. 169):
x 3 ¼ 0 ) u3 ¼ 0
x 3 ¼ H ) u3 ¼ 0
Substituting u3 ¼ ’;3 we have
AB¼0
ð169:2Þ
AeikrH BeikrH ¼ 0
which leads to the equation
A eikrH eikrH ¼ 0
ð169:3Þ
Consider first that r is real, that is, c >a. Then, from (169.1)
2iA sin krH ¼ 0 ) krH ¼ np; n ¼ 0; 1; 2; :::
with n ¼ 0, fundamental mode (FM), and n 1 for higher modes.
For the FM, n ¼ 0 and r ¼ 0, and then
rffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
Hk 1¼0)c¼a
a2
The displacements from (169.1) and (169.2) are
x3
rigid medium
H
liquid layer
0 x1
rigid medium
Fig. 169
316 Seismology
@’
u3 ¼ ¼ Aikrðexp ikrx3 exp ikrx3 Þ exp ikðx1 ctÞ
@x3
@’
u1 ¼ ¼ Aik ðexp ikrx3 þ exp ikrx3 Þ exp ikðx1 ctÞ
@x1
For the FM r ¼ 0, then u3 ¼ 0 and this is a P-wave, with only a u1 component, which
propagates in the x1-direction. For all HM the displacements have both components
For the first higher mode (1HM), n ¼ 1:
rffiffiffiffiffiffiffiffiffiffiffiffiffi
o c2
H 1¼p
c a2
1
c2 ¼
1 p2
a2 o2 H 2
If
1 p2 ap
¼0 then o¼ )c!1
a2 o 2 H 2 H
The cut-off frequency is oc > ap/H. For each higher mode there is a cut-off frequency onc
> nap/H.
_ _
If r ¼ ir is imaginary, then c < a and this implies that 2 sinhðkr HÞ ¼ 0 which is
impossible (1< sinhx <1).
The particle motion inside the layer is circular when
u1 ¼ u3 ð169:4Þ
so
u1 ¼ u3 ) ð1 rÞ exp ikrx3 þ ð1 þ rÞ exp ikrx3 ¼ 0
Taking only the amplitudes of the displacements,
ð1 rÞ cos krx3 þ ð1 þ rÞ cos krx3 ¼ 0 ) cos krx3 ¼ 0
1 1 p
krx3 ¼ n þ p ) x3 ¼ n þ
2 2 kr
For the FM, we have seen that u3 ¼ 0, so there is no circular motion. For the 1HM, the
height in the layer at which the motion is circular is
3p
x3 ¼
2kr
The height inside the layer at which the motion is circular depends in each higher mode of
the frequency.
170. In the hypothetical case of a layer of thickness H and speed of propagation b0 on
top of a semi-infinite medium of speed of propagation b, the phase shifts at the free
surface and the contact plane are
2 sffiffiffiffiffiffiffiffiffiffiffiffiffi 3
c2
6 b0 1 7
6 7
p 1 6
and tan 6sffiffiffiffiffiffiffiffiffiffiffiffiffiffi7:
7
4 27
4 1 c 5
6
b2
317 Surface waves
Determine:
(a) The dispersion equation using constructive interference.
(b) The cut-off frequency of the fundamental mode and first higher mode.
(c) Plot the dispersion curve of the FM and 1HM using units of c/b and H/l for
b ¼ 2b0 .
(a) The distance from A to B along the ray path is (Fig. 170a)
AB ¼ 2H cos i
According to Snell’s law
b0
sin i ¼
c sffiffiffiffiffiffiffiffiffiffiffiffiffiffi
b0 c2
cos i ¼ 1
c b02
b0
kb0 sin i ¼ kb0 ¼k
c
As explained in Problem 167, the condition for constructive interference is that the distance
AB along the ray path be an integer multiple of the wavelength, taking into account the
phase shift at the free surface and the boundary surface between the two media:
0
p s
2kB0 H cos i þ tan1 ¼ 2pn ð170:1Þ
4 s
where
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
s0 ¼ 1
b02
sffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
s¼ 1 2
b
B
H
A
i
β⬘
Fig. 170a
318 Seismology
vffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u 2
u c
u
ub02 1
1
tan 2kHs0 2n p ¼ u ð170:2Þ
4 u c2
1 2
t
b
s0
0 7
tan 2kHs p ¼
4 s
The tangent function is positive for the range [0, p/2]:
7p p
0
2kHs0
4 2
For c ¼ b0
319 Surface waves
7p 7p
tan 2kHs0 ¼ 0 ) kHs0 ¼ )k¼1
4 8
For c ¼ b
7p 7p p 9p
tan 2kHs0 ¼ 1 ) 2kHs0 ¼ )k¼
4 4 2 8Hs0
(c) Taking b’¼ b/2, the dispersion equation for the FM is
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4c2
" sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi # 1
b2
2
4c2 p c 2
tan 2kH 2
1þ ¼ s ffiffiffiffiffiffiffiffiffiffiffiffiffi ) ¼ "
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi#
b 4 c 2 b H
2
c
1 2 5 3 sin 8p 4 1
b l b
H/l c/b
0.0 0.63
0.01 0.68
0.02 0.78
0.036 1
For the first higher mode (1HM), we arrive at the same equation, given that tan(a þ p/4)
¼ tan(a þ 7p/4), but vary the intervals of H/l and kHs0 :
7p 9p
kHs0
8 8
7 H 9
16s0 l 16s0
H 7
For ¼ at one limit we have the value s0 ¼ 0 which corresponds to c ¼ b0 and c/b ¼
l 16s0
0.5.
H 9 H 9
For ¼ 0
we have s ¼ 0 and c ¼ b, and consequently ¼ pffiffiffi.
l 16s l 16 3
The dispersion curves for the fundamental mode and the first higher mode are shown in
Fig.170b.
320 Seismology
c
β
1.0
FM 1 HM
0.5
H
1 0.10 0.20 0.30 9 λ
16 3 16 3
Fig. 170b
Calculate:
(a) The dispersion equation of the Love wave.
(b) For the fundamental and first higher mode, and the minimum and maximum
frequencies as functions of H, b, and b0 .
(c) For this mode, given b0 ¼ b/2, the maximum and minimum frequencies, and the
corresponding values of c.
(a) As in Problem 170, the condition of constructive interference with the phase shifts
given in this problem results in (Fig. 171)
4p p s
H cos i sin1 0 ¼ 2pn ð171:1Þ
l 2 s
where
321 Surface waves
B
H
A i
β⬘
Fig. 171
sffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
s¼ 1 2
b
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi
c2
s0 ¼ 1
b0 2
p s 1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
k¼ ) ¼ 0 ) s ¼ 0 ) c ¼ b ) s 0
¼ b2 b02
4Hs0 s0 b0
322 Seismology
qffiffiffiffiffiffiffiffiffiffiffiffi2
1 bc 2
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi)
m
(
c 2
tan kH 1 ¼ ð172:1Þ
b02
q ffiffiffiffiffiffiffiffiffiffiffiffi
2
ffi
m0 c02 1
b
In this problem,
m ¼ 4m0
r ¼ r0
then,
rffiffiffi rffiffiffiffiffiffi
m 0 m b
b¼ !b ¼ ¼
r 4r 2
We now introduce a and b:
c c
a¼ ) 0 ¼ 2a
b b
H 2p 2pb
b¼ )k¼ ¼
l l H
Substituting in (172.1):
pffiffiffiffiffiffiffiffiffiffiffiffiffi
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 a2
tan 2pb 4a 1 ¼ 4 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
4a2 1
(b) For the FM
a ¼ 1; b ¼ 0 and c ¼ b
1
a ¼ ; b ! 1 and c ¼ b0
2
rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
9
1
" rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi#
3 9 16
a ¼ ) tan 2pb 4 1 ¼ 4 rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ) b ¼ 0:17
4 16 9
4 1
16
For the 1HM,
1
a¼ ; b!1
2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffi 1
a ¼ 1 ) tan 2pb 4a2 1 ¼ 0 ) 2pb 3 ¼ p ) b ¼ pffiffiffi ¼ 0:29
2 3
0 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi1 rffiffiffiffiffi!
2
pffiffiffi
rffiffiffiffiffi
3 3 7 1 7
a ¼ ) tan@2pb 4 1A ¼ 4 ) pb 5 ¼ tan 4 þ p ) b ¼ 0:61
4 4 20 20
(c) Inside the layer the amplitude of the displacements of the Love wave are given by
h x3 i
u02 ¼ 2A0 cos ks0 H 1 cos k ðs0 H þ x1 ct Þ
H
324 Seismology
The nodes are the points where the amplitude is zero. For the 1HM the node is located at
the value of x3 which satisfies the relation
x3 3p
ks0 H 1 ¼
H 2
If we want a node located at x3 ¼ H/2 then
3p 3
k¼ ) b ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s0 H 2 4a2 1
If we substitute the values of a, ½, 1, and ¾ we obtain for b infinity, 0.86, and 1.34. The
infinite value of b corresponds to l¼ 0.
Focal parameters
40°N
39°N
St3
38°N
E2
E1
37°N St2
St1
36°N
35°N
1°E 2°E 3°E 4°E 5°E 6°E
Fig. 173
ð37:40 N; 3:83 EÞ
ð xe ; ye Þ ¼
ð37:63 N; 4:71 EÞ
jh ih F⬘
h
F A
ih
A⬘
Fig. 174
An sP-wave leaves the focus upward as an S-wave is reflected at the Earth’s surface and
converted into a P-wave that travels to the station. Its travel time is (Fig. 174)
0
FS SF L
t sP ¼ þ þ
b a a
If we consider the length along the ray L from F and F0 to the station to be the same for both
waves, then the sP-P time interval is (Fig. 174)
0
FS SF
t sP t P ¼ t 0 ¼ þ ð174:1Þ
b a
At the focus at depth h, the take–off angle of the direct P-wave is ih and the take-off angle
of the sP-wave is jh. We can write
h
FS ¼
cos jh
0
SF 0 h
cosðjh þ ih Þ ¼ ) SF ¼ ðsin jh sin ih cos jh cos ih Þ
FS cos jh
a sin jh
sin ih ¼
b
By substitution in (174.1) we obtain
cos jh cos ih
t 0 ¼ h þ
b a
175. The displacement vector l of an earthquake is (0, 1, 0) and the vector normal to
the plane of displacement n is (0, 1, 0). Determine:
(a) The components of the P-wave displacement at the point of azimuth 45 and angle
of incidence 30 .
327 Focal parameters
where Green’s function corresponding to the P-wave in the far field for an infinite medium
is given by
1 r
Gki ¼ g i g k d t
4pra2 r a
and its derivative is
1 _ tr
Gki;j ¼ g g g d
4pra3 r i k j a
where gi are the direction cosines of the line from the focus to the observation point.
The amplitude of the displacement is then
u
uPk ¼
3
lns ls dij þ m li nj þ lj ni gi gk gj ð175:1Þ
4pra r
In our problem the direction cosines of the ray of the waves arriving at the point are
pffiffiffi
2
g1 ¼ sin i cos az ¼
4
pffiffiffi
2
g2 ¼ sin i sin az ¼
pffiffiffi 4
3
g3 ¼ cos i ¼
2
X3
P
r
i
li= ni e
X1
az
X2
Fig. 175
328 Seismology
The orientation of the source is given by li ¼ (0, 1, 0) and ni ¼ (0, 1, 0), and substituting in
(175.1) gives
uP1 ¼ A l þ 2mg22 g1
uP2 ¼ A l þ 2mg22 g2
uP3 ¼ A l þ 2mg22 g3
u
A¼
4pa3 rr
Substituting the direction cosines of the ray we obtain,
pffiffiffi
1 2
uP1 ¼ A l þ m
4 4
pffiffiffi
P 1 2
u2 ¼ A l þ m
4 4
pffiffiffi
P 1 3
u3 ¼ A l þ m
4 2
(b) The mechanism corresponds to a fault on the (x1, x3) plane which opens in the
direction of its normal, x2, under tensional forces in that direction.
176. The focal mechanism of an earthquake can be represented by a double-couple
(DC) model. The orientation of the fault plane is azimuth 30 , dip 90 , and slip angle
0 . Calculate:
(a) What kind of fault it is. Sketch it, indicating the direction of motion.
(b) The auxiliary plane.
(c) The azimuth of the stress axis.
(d) A wave incident at a station has azimuth 180 and angle of incidence at the focus
of 90 . Calculate the amplitude of the components of the P-wave at that station.
(a) Given that the dip of the plane is 90 , the fault plane is vertical, and since the slip
angle is 0 , the motion is horizontal. Thus, it corresponds to a right lateral strike-
slip fault (Fig. 176).
ϕA
Δu
δA l
n
Fig. 176
329 Focal parameters
(b) From the azimuth (’ ¼ 30 ), dip (d ¼ 90 ), and slip (l ¼ 0 ) we can calculate the
unit vectors li and ni which give the direction of the fracture and of the normal to
the fault plane:
1
n1 ¼ sin d sin ’ ¼ ¼ sin Yn cos Fn
pffiffiffi 2
3
n2 ¼ sin d cos ’ ¼ ¼ sin Yn cos Fn
2
n3 ¼ cos d ¼ 0 ¼ cos Yn
pffiffiffi
3
l1 ¼ cos l cos ’ þ cos d sin l sin ’ ¼ ¼ sin Yl cos Fl
2
1
l2 ¼ cos l sin ’ cos d sin l cos ’ ¼ ¼ sin Yl cos Fl
2
l3 ¼ sin l sin d ¼ 0 ¼ cos Yl
’A ¼ Fn þ 90 ) Fn ¼ 120
dA ¼ Yn ¼ 90
cos Yl
lA ¼ sin1 ¼ 0 ) Yi ¼ 90
sin Yn
where Y and F are the spherical coordinates for the vectors n and l (r ¼ 1, unitary vectors)
and for the auxiliary plane,
’B ¼ 30 þ 90 ¼ 120
dB ¼ 90
cos Yn
lB ¼ sin1 ¼ 0
sin Yl
(c) The T-axis is on the same plane as ni and li at 45 between them, and the direction
cosines are
0 1 0 10 1 1
T1 n1 n2 n3 pffiffi
2
@ T2 A ¼ @ l1 l2 l3 A@ p1ffiffi A ð176:1Þ
2
T3 Z1 Z2 Z3 0
where Zi is the axis normal to ni and li, that is, Zi ¼ ni li which results in Zi ¼ (0, 0, 1).
pffiffiffi pffiffiffi
31 3þ1
Substituting in (176.1) we obtain Ti ¼ pffiffiffi ; pffiffiffi ; 0 . The azimuth of the T-axis is
2 2 2 2
pffiffiffi
1 T2 1 3þ1
FT ¼ tan ¼ tan pffiffiffi ¼ 75
T1 31
(d) The direction cosines of the direction from the focus to the station are
g1 ¼ sin ih cos az ¼ 1
g2 ¼ sin ih sin az ¼ 0
g3 ¼ cos ih ¼ 0
330 Seismology
The amplitude of the displacements for a shear fracture or double-couple (DC) source in an
infinite medium is given by
1 cos Yl
lA ¼ sin ¼ 135:68
sin Yn
’B ¼ Fl þ 90 ¼ 35:96
dB ¼ Yl ¼ 50:09
1 cos Yn
lB ¼ sin ¼ 31:12
sin Yl
qp
X l
n qn
T
P
fn
Fig. 177
331 Focal parameters
To calculate the T and P axes, we calculate first the direction cosines of the l and n axes
from the given angles:
x1 ¼ sin Y cos F n1 ¼ 0:50; l1 ¼ 0:45
x2 ¼ sin Y sin F ) n2 ¼ 0:77; l2 ¼ 0:62
x3 ¼ cos Y n3 ¼ 0:40; l3 ¼ 0:64
Given that the T and P axes are on the same plane as n and l at 45 between them, we can
write, as in Problem 176,
1 p1ffiffiffi
0 1
0 1 0
T1 l1 n1 Z1 B 2 C
@ T2 A ¼ @ l2 n2 Z2 AB p1ffiffiffi C ð177:1Þ
@ A
T3 l3 n3 Z3 2
0
where the Z-axis is normal to n and l and is found by Z ¼ n l. Its direction cosines
are (Z1, Z2, Z3) ¼ (0.72, 0.14, 0.66). By substitution of ni, li, and Zi in (177.1) we
obtain
9
T1 ¼ 0:67 ¼ sin YT cos FT >
=
T2 ¼ 0:11 ¼ sin YT sin FT ) T ðYT ¼ 42:27 ; FT ¼ 9:32 Þ
>
T3 ¼ 0:74 ¼ cos YT
;
Find the values of the principal stresses, and the orientation of the tension and
pressure stress axes.
First we calculate the eigenvalues of Mij. Since Mij is a symmetric tensor its eigenvalues are
real and the corresponding eigenvectors mutually orthogonal (Problem 111):
0 1
2 s 1 1 3
1 0s 1 ¼ 0 ) s ¼ @ 2 A
1 1 2 s 1
In this form Mij is referred to the coordinate system formed by the eigenvectors or principal
axes. Given that the sum of the elements of the principal diagonal is not zero, the source
has net volume changes. Then, we can separate Mij into two parts: an isotropic part with
volume changes (ISO) and a deviatoric part without volume changes. The second part can
be separated into two parts: a part corresponding to a double-couple or shear fracture (DC)
and a part corresponding to a non-double-couple source usually expressed as a compen-
sated linear vector dipole (CLVD). Thus the moment tensor is separated into three parts,
namely
M ¼ M ISO þ M DC þ M CLVD
The isotropic part is given by
1 4
M ISO ¼ s0 ¼ ðs1 þ s2 þ s3 Þ ¼
3 3
The deviatoric part (DCþCLVD) is given by
1
0 s2 1 0 1
ðs s3 Þ 0 0 0 0
B2 1 C B 2 C
0
Mij ¼ B 0 0 0 CþB 0 s2 0 C
B C B C
@
1
A @ s2 A
0 0 ðs1 s3 Þ 0 0
2 2
0 1 1
0 0
3
0 1
2 0 0 B C
2
B C
Mij0 ¼ @0 0 0 AþB 0 0 C
B C B C
B 3 C
0 0 2 @
1
A
0 0
3
333 Focal parameters
The orientation of the P and T axes is calculated from the double-couple part MDC:
0 1
1 0 0
MijDC ¼ 2@ 0 0 0 A
0 0 1
MijDC ¼ M0 ðli nj lj ni Þ
1 1
n : pffiffiffi ; 0; pffiffiffi
~ ) nðYn ¼ 45 ; Fn ¼ 0 Þ
2 2
~ 1 1
l: p ffiffi
ffi ; 0; pffiffi
ffi ) lðYl ¼ 45 ; Fl ¼ 0 Þ
2 2
In the same way as in Problems 176 and 177, we determine T and P from n and l, finding
first Z ¼ n l:
1 p1ffiffiffi
0 1
0 1 0
T1 l1 n1 Z1 B 2 C
@ T2 A ¼ @ l2 n2 Z2 AB B 1 C
C
@ pffiffiffi A
T3 l3 n3 Z3 2
0
T1 ¼ 1 ¼ sin YT cos FT
T2 ¼ 0 ¼ sin YT sin FT ! T ðYT ¼ 90 ; FT ¼ 0 Þ
T3 ¼ 0 ¼ cos YT
where A is the ground motion amplitude, T is the period of the wave, and D is epicentral
distance in degrees. Knowing the magnitude and period of the waves we can calculate the
wave amplitude:
A 3000
6:13 ¼ log þ 1:66 log þ 3:3
T 111:11
A
) log ¼ 0:454 ) A ¼ 2:84 20 1500 ¼ 8:5 cm
T
We have reduced the ground motion to the amplitude of the seismogram using the
amplification of the instrument (1500).
Knowing the magnitude we can calculate the seismic energy:
M0 2:19 1018 Nm
u ¼ ¼ ¼ 0:52 m
mS 4:4 1010 N m2 9:6 107 m2
5 Heat flow and geochronology
Heat flow
180. Assume that the temperature variation within the Earth is caused by gravita-
tional forces under adiabatic conditions. Knowing that the coefficient of thermal
expansion at constant pressure is aP ¼ 2105 K1 and the specific heat at constant
pressure is cP ¼ 1.3 kJ kg1 K1, determine an expression for the gradient of the
temperature with depth. Compare it with the value observed at the surface which is
30 K km1, knowing that, at 200 km depth, T ¼ 1600 K.
Under adiabatic conditions, there is no heat flow and the variation of pressure with depth z
is a function of gravity g and density r:
dP ¼ rgdz ð180:1Þ
Using the first and second laws of thermodynamics
dU ¼ dQ PdV
dQ ¼ TdS
where Q is the heat, U is the internal energy, S is the entropy, T is the absolute temperature,
P is the pressure, and V is the volume.
If we use the specific variables (variables divided by mass) we can write
du ¼ dq pdv
dq ¼ Tds
Considering that
@S @S
dS ¼ dT þ dP
@T P @P T
we can write
@s @s
dq ¼ Tds ¼ T dT þ T dP
@T P @P T
According to the definition of specific heat at constant pressure, cP and the increase in heat
dq are given by
335
336 Heat flow and geochronology
@S
cP ¼ T
@T P
ð180:2Þ
@s
dq ¼ cP dT þ T dP
@P T
G ¼ u Ts þ pv
Taking the differential in this expression, and taking into account the second law of
thermodynamics
du ¼ Tds pdv
we obtain
dG ¼ vdp sdT
If we compare this expression to the differential of the Gibbs function
@G @G
dG ¼ dp þ dT
@p T @T p
@v @s
¼
@T P @P T
1 @v
ap ¼
v @T p
In our case the process is adiabatic and in consequence using this equation and Equation
(180.2) we obtain
dT T ap g
¼ ð180:3Þ
dz cp
where we have taken into account that the variables are by unit mass so rv ¼ 1, and
substituting the values we obtain:
181. If the Earth’s temperature gradient is 1 C/30 m, calculate the heat loss per
second due to conduction from its core. Compare this with the average power
received from the Sun.
Data:
Thermal conductivity K ¼ 4 W m1 C1.
Earth’s radius R ¼ 6370 km.
Solar constant: 1.35 kW m2.
The heat flow is given by
dT
q_ ¼ KA
dr
where A is the area of the Earth’s surface
Using the temperature equation for a flat Earth for one-dimensional heat-flow and the
stationary case we can write
e 2 q_ 0
T ¼ z þ z þ T0
2K K
q_ 0
e¼
H
For z ¼ H:
q_ 0 H ðTH T0 Þ2K
TH ¼ þ T0 ) H ¼
2K q_ 0
338 Heat flow and geochronology
H ¼ 133:3 km
and the heat production by unit of volume is
q_ 0
e¼ ¼ 4:5 104 mW m3
H
183. Consider the crust to be H ¼ 30 km thick and the heat flow at the surface to be
60 mW m2.
(a) If all the heat is generated in the crust, what is the value of the heat generated per
unit volume? (Take K ¼ 3 W m1 K1)
(b) If all the heat is generated in the mantle with a distribution Aez=H mW m3 , what
is the value of A? What is the temperature at 100 km depth?
(a) We solve the heat equation for a stationary one-dimensional case, assuming a flat
Earth with one-dimensional flow in the z-direction (vertical) positive downward.
In this case the solution of the heat equation is given by
e 2 q_ 0
T ¼ z þ z þ T0
2K K
where e is the heat generated by unit volume and time, K is the thermal conductivity, q0 and
T0 are the heat flow and temperature at the surface of the Earth, respectively
If all the heat is generated at the crust we can write (Fig. 183)
z ¼ H ! q_ 0 ¼ 0
e ¼ Aez=H
the heat equation is
d2 T e A
¼ ¼ ez=H
dz2 K K
and the solution is given by
A 2 z=H
T ¼ H e þ Cz þ D ð183:1Þ
K
where C and D are constants of integration. They may be estimated from the boundary
conditions at the surface
339 Heat flow
•
q0
T0
•
qH
TH
Fig. 183
A 2 A
z ¼ 0 ! T ¼ T0 ¼ H þ D ! D ¼ T0 þ H 2
K K
A
z ¼ 0 ! q_ ¼ q_ 0 ¼ 0 ! C ¼ H
K
Substituting in (183.1) we obtain
A 2 z=H AH A AH
T ¼ H e z þ T0 þ H 2 ¼ Hez=H z þ H þ T0 ð183:2Þ
K K K K
If the heat has its origin in the mantle, the flow at the base of the crust is
dT q_ 0
_ z¼H ¼ q_ o ¼ K
qj )A¼
dz z¼H H ðe1 1Þ
60 103 3 100=30 3 3
T ðz ¼ 100Þ ¼ 30 10 e 100 10 þ 30 10 ¼ 2249 K
3ðe1 1Þ
184. Calculate the thickness of the continental lithosphere if its boundary coincides
with the 1350 C geothermal, knowing that the surface temperature is 15 C,
the heat flow at the surface is q_ 0 ¼ 46 mW m2 , the lithospheric mantle’s thermal
conductivity is K = 3.35 W m1 K1, and the radiogenic heat production is
P = 0.01 103 mW m3
The geothermal equation at depth z is given by:
q_ 0 P0
Tz ¼ T0 þ ðz z0 Þ ðz z0 Þ2 ð184:1Þ
K 2K
340 Heat flow and geochronology
where K is the thermal conductivity, T0 is the temperature at the surface of the Earth (in K), q_ 0
is the heat flow at the surface, and P0 is the radiogenic heat production at the Earth’s surface.
At the Earth’s surface z0 = 0, and Equation (184.1) becomes:
q_ 0 P0 2
Tz ¼ T0 þ z z
K 2K
P0 2 q_ 0
z z þ ðTz T0 Þ ¼ 0
2K K
So
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 ffi
q_ 0 q_ 0 P0
4 ðTz T0 Þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
K K 2K q_ 0 q_ 20 2P0 KðTz T0 Þ
z¼ ¼
P0 P0
2
2K
Substituting the data given in the problem:
K ¼ 3.35 W m1 K1
q_ 0 ¼ 46 103 W m2
P0 ¼ 0.01106 mW m3
Tz ¼ 1623 K
T0 ¼ 288 K
we obtain two solutions, but only z = 98.27 km is realistic (the second one gives a depth
larger than the Earth’s radius).
185. On the surface of an Earth of radius 6000 km, the temperature is 300 K, the heat
flow is 6.7 mW m2, and the thermal conductivity is 3 W m1 K1. If the heat produc-
tion per unit volume inside the Earth is homogeneously distributed, what is the
temperature at the centre of the planet?
We begin solving the problem of heat conduction inside a sphere with constant internal
heat generation per unit volume e and conductivity K. The differential equation for heat
conduction with spherical symmetry is
1 d 2 dT e
2
r ¼ ð185:1Þ
r dr dr K
Integrating twice and using the boundary conditions:
Surface: r = R ! T = T0
Center: r = 0 ! T finite
we obtain the solution
e 2
R r2
T ¼ T0 þ ð185:2Þ
6K
The heat flow is given by
dT
q_ ¼ K ) q_ 0 ðr ¼ RÞ ¼ 6:7 mW m2
dr
d e 2 e r
¼ K T0 þ R r2 ¼
dr 6K 3
341 Heat flow
3:35 109
Tr¼0 ¼ 300 þ 62 1012 ¼ 7000 K
63
186. Consider a spherical Earth of radius R = 6000 km and a core at R/2, in which
there is a uniform and stationary distribution of ε heat sources per unit volume. The
heat flow at the surface is 5 mW m2, the thermal conductivity is 3 W m1 K1, and
the temperature at the core–mantle boundary is 4000 K. Calculate the temperature at
the Earth’s surface.
We consider the problem as one of heat conduction inside a sphere with conductivity K and
constant heat generation per unit volume e inside the core (radius R/2). We begin with
Equation (185.1)
1 d 2 dT e
r ¼
r2 dr dr K
The boundary conditions at the core–mantle boundary and its centre are
r ¼ R=2 ! T ¼ TN
r ¼ 0 ! T finite
where A ¼ 4pr2 is the area in the normal direction to the heat flow. Integrating this
equation:
ð R0 ð T0
q_ dr
¼ K dT
4p r1 r2 T1
where the conditions at the Earth’s surface are, r = R0 !T = T0 and at the base of the crust,
r = r1 !T = T1.
Solving Equation (187.1), assuming that K is constant, we obtain
4pK 4pKr1 R0
q_ ¼ 1 1
ðT0 T1 Þ ¼ ðT0 T1 Þ ð187:2Þ
r1 R0
R0 r1
(b) The temperature distribution inside of the Earth may be obtained by integration of
Equation (187.1):
ðT
q_ r dr
ð
¼ K dT
4p r1 r2 T1
R0 ðr r1 Þ
T ð r Þ ¼ T1 þ ðT0 T1 Þ
ðR0 r1 Þr
(c) The radial distance to the base of the crust is r2 = 5500 km, so, using expression
(187.2), we obtain
_ 0 r1 Þ
qðR 5:5 1013 500 103
T1 ¼ T0 þ ¼0þ ¼ 16579 C
4pKr1 R0 4p 4 5500 103 6000 103
This result implies a constant increase of temperature from the Earth’s surface
of 1 ºC each 33.2 m similar to the observed gradient in the real Earth of 1 ºC
per 30 m
188. Assume that the heat flow inside the Earth is due to solar heating of the
Earth’s surface. Calculate the maximum penetration of this flow in the diurnal and
annual cycles. Take as typical values for the Earth K = 3 Wm1 K1, r = 5.5 g cm3,
Cv = 1 kJ kg1 K1.
343 Heat flow
We assume the heat propagation inside the Earth coming from the solar radiation on its
surface as unidirectional flow thermal diffusion (inside the Earth) with periodic variation of
surface temperature. The diffusivity equation is
@ 2 T @T
k ¼ ð188:1Þ
@z2 @t
K
where the thermal diffusivity is k ¼ , K is the thermal conductivity, r is the density,
rCv
and Cv is the specific heat at constant volume. We solve Equation (188.1) using the
separation of variables
T ðz; t Þ ¼ Z ð zÞyðt Þ
Substituting in (188.1) we obtain the solution
Z ð zÞ ¼ Aeaz þ Beaz
2
yðt Þ ¼ Ceka t
where a is the constant of separation of variables. Using the boundary condition of periodic
flow and the temperature T0 at the Earth’s surface,
z ¼ 0 ) T ¼ T0 eiot
and as Z(z) exists only inside the Earth, B = 0. At the surface, z = 0, so
2
ACeka t ¼ T0 eiot
Then
AC ¼ T0
ka2 ¼ io
But putting, i ¼ 12 ð1 þ iÞ2 , we have
rffiffiffiffiffiffi
o
a ¼ ð1 þ iÞ
2k
Then, we can write the temperature variation inside of the Earth as:
rffiffiffiffiffiffi rffiffiffiffiffiffiffiffi
o o
T ðz; t Þ ¼ T0 exp zþi z þ ot
2k 2k
This equation corresponds to a periodic wave, with angular frequency o propagating for
positive z values (to the Earth’s interior) and with the amplitude decreasing with depth. The
propagation velocity and wavelength are given by
rffiffiffiffiffiffi
2k
v¼
o
rffiffiffiffiffiffi
8k
l ¼ 2pv ¼ p
o
344 Heat flow and geochronology
The values of l corresponding to the daily and annual cycles give their maximum penetration:
Daily cycle:
2p
o¼ ¼ 7:2 105 s1
24 60 60
K 3 W m1 K1
k¼ ¼ ¼ 0:5 106 m2 s1
rCv 5:5 103 K gm3 103 J Kg1 K1
Then
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
8 0:5 106
l¼p ¼ 0:74 m
7:2 105
Annual cycle:
2p
o¼ ¼ 2 107 s1
365 24 60 60
l ¼ 14 m
The penetration of the solar radiation as periodic heat conduction inside the Earth is very
shallow due to the poor heat conduction.
189. Consider a lithospheric plate of 100 km thickness created from asthenospheric
material originating from a ridge in the asthenosphere with constant temperature Ta and
in which no heat is generated. Given that k = 106 m² s1, that the temperature at the base
of the lithosphere is 1100 C, and in the asthenosphere is 1300 C, calculate the age of the
plate, and, if the velocity of drift is 2 cm yr1, how far it has moved away from the ridge.
The heat propagation inside the plate is given by:
K @2T @2T
@T
þ 2 ¼u ð189:1Þ
rcv @x2 @z @x
where T is the temperature, r is the density, cv is the specific heat at constant volume, and u
is the horizontal velocity of the plate in the x-direction (normal to the plate front). If we
assume that the horizontal conduction of heat is insignificant in comparison with the
horizontal advection and vertical conduction, we can write, using the following change
of variable t = x/u,
K @ 2 T @T
¼
rcv @z2 @t
Integrating this equation and using the boundary conditions at the ridge and surface:
x ¼ 0 ! T ¼ Ta
z¼0!T ¼0
z
T ðz; t Þ ¼ Ta erf pffiffiffiffiffi
2 kt
345 Geochronology
where
K
k¼
rcv
2 x
ð
2
erf ð xÞ ¼ pffiffiffi ey dy
p 0
Substituting the data of the problem,
L L
1100 ¼ 1300erf pffiffiffiffiffi ) erf pffiffiffiffiffi ¼ 0:846
2 kt 2 kt
Values of the error function, erf(x), may be obtained from tables. If erf(x) = 0.846, x = 1.008, then
L2 1010
L
pffiffiffiffiffi ¼ 1:008 ) t ¼ ¼
2 kt 4k 1:0082 4 106 1:0082
¼ 2:5 1015 s ¼ 79 Myr
If the displacement velocity is 2 cm yr1, the plate has moved 1580 km.
190. If the concentrations of 235U and 235Th in granite are 4 ppm and 17 ppm,
respectively, and the respective values of heat production are 5.7 104 W kg1
and 2.7 105 W kg1, respectively, calculate the heat flow at the base of a granite
column of 1 m² cross-section and 30 km height (the density of granite is 2.65 g cm3).
We estimate first the mass of the granite column:
Geochronology
191. The mass of 1 millicurie of 214Pb is 3 1014 kg. Calculate the value of the decay
constant of 214Pb.
The mass of the sample is
346 Heat flow and geochronology
N
3 1014 kg ¼ 3 1017 g ¼ M
N0
where N is the number of atoms in the sample, N0 is Avogadro’s number ¼ 6.02 1023,
and M is atomic number ¼ 214. Solving for N we obtain
N ¼ N0 elt ð193:1Þ
347 Geochronology
1 NE
t ¼ ln 1 þ
l NR
NE, the number of atoms of 40K in the sample, can be estimated from the number of atoms
contained in 1 g of potassium:
N
1¼ 40 ! N ¼ 1:506 1022 atoms g1
6:023 1023
348 Heat flow and geochronology
so
1:792 1036
1
t ¼ ln 1 þ ¼ 7:4 1010 yr
l 4:41 1017
195. At an archaeological site, human remains were found and assigned an age of
2000 years. One wants to confirm this with 14C dating whose half-life is 5730 yr. If the
proportion of 14C/12C in the remains is 6 1013, calculate their age. (Assume that at
the initial time the 14C/12C ratio was 1.2 1012.)
The decay constant l may be obtained from the half-life:
0:693 0:693
T1=2 ¼ !l¼ ¼ 1:2094 104 yr1
l 5730
The activity in a sample is given by
R ¼ R0 elt
where R0 ¼ (14C/12C)t ¼ 0 and R ¼ (14C/12C).
Then the age of the remains is
1:2 1012
1 R0 1
t ¼ ln ¼ ln ¼ 2379 yr
l R 1:2094 104 9 1013
196. Mass spectrometry of the different minerals in an igneous rock yielded the
following table of values for the concentrations of 87Sr originating from the radio-
active decay of 87Rb and of 87Rb, with the concentration expressed relative to the
concentrations of 86Sr of non-radioactive origin.
87
Mineral Sr /86Sr 87
Rb /86Sr
A 0.709 0.125
B 0.715 0.418
C 0.732 1.216
D 0.755 2.000
E 0.756 2.115
F 0.762 2.247
0.80
0.78
87Sr 0.76
86Sr
0.74
0.72
0.70
0.0 0.5 1.0 1.5 2.0 2.5
87Rb
86Sr
Fig. 196
where [87Sr]now and [87Rb]now are the number of atoms of each isotope at time t, [87Sr]0 is the
amount of original number of atoms of the isotope 87Sr [87Sr]now, and l is the decay constant.
Equation (196.1) may be written as
87 87
Sr 87 Rb lt
Sr
86 Sr
¼ 86 þ 86 e 1 ð196:2Þ
now Sr 0 Sr now
87
Sr
This equation corresponds to a line (y ¼ a þ bx) with intercept 86 and slope (elt 1),
Sr 0
which is called an isochron.
If we plot the values given in the problem (Fig. 196) we can obtain the equation of the
line by least-squares fitting:
y ¼ 0:025x þ 0:705
The age of the sample can be obtained from the slope b ¼ 0.025:
el t 1 ¼ b
lnð1 þ bÞ
t¼
l
Substituting the values of b and l:
87
Sample Rb/86Sr
A 1.195
B 2.638
C 4.892
D 5.671
(a) Calculate the proportions of 87Sr/86Sr and 87Rb/86Sr that these rocks will have
after 500 Myr. Take l ¼ 1.421011 yr1.
(b) Express in a 87Sr/86Sr–87Rb/86Sr diagram the isochrons corresponding to t ¼ 0
and t ¼ 500 Myr.
(a) Using the same method as in the previous problem, we can write
87
Sr 87 Sr 87 Rb lt 87
Rb 1:421011 5108
86 Sr
¼ 86 Sr
þ 86 e 1 ¼ 0:709 þ 86 e 1
0 Sr Sr
0.78
0.76
t = 500 Ma
Sr
Sr
0.74
87
86
0.72
t=0
0.70
0 1 2 3 4 5 6 7
87Rb
86Sr
Fig. 197
351 Geochronology
The results for each rock are given in the following table
87
Sample Sr/86Sr 87
Rb/86Sr
A 0.717 1.187
B 0.728 2.619
C 0.744 4.857
D 0.749 5.631
(b) For t ¼ 500 Myr, we carry out a least-squares fitting to obtain the isochron,
which results in
y = 0.007x þ 0.709
In Fig. 197 the isochrones corresponding to t ¼ 0 and t ¼ 500 Myr are shown.
Bibliography
General geophysics
Berckhemer, H. (1990). Grundlagen der Geophysik. Wissenschaftlische Buchgeselschaft,
Darmstadt.
Buforn, E., C. Pro, and A. Udías (2010). Problemas Resueltos de Geofísica. Pearson, Madrid.
Cara, M. (1989). Geophysique. Dunod, Paris.
Coulomb, J. and G. Jobert (1973 and 1976). Traité de Géophysique Interne (I and II).
Masson, Paris.
Fowler, C.M. R. (2005). The Solid Earth: An Introduction to Global Geophysics. (2nd
edn). Cambridge University Press, Cambridge.
Garland, G.D. (1979). Introduction to Geophysics. W.B. Saunders, Philadelphia.
Kaufman, A.A. (1992). Geophysical Field Theory and Method. Academic Press, San Diego.
Larroque, C. and J. Virieux (2001). Physique de la Terre solide. Observations et Théories.
Gordon and Breach, Paris.
Lillie, R.J. (1999). Whole Earth Geophysics. An Introductory Textbook for Geologists and
Geophysicists. Prentice Hall, New Jersey.
Lowrie, W. (2007). Fundamentals of Geophysics (2nd edn). Cambridge University Press,
Cambridge.
Lowrie, W. (2011). A Student’s Guide to Geophysical Equations. Cambridge University
Press, Cambridge.
Officer, C.B. (1974). Introduction to Theoretical Geophysics. Springer, New York.
Poirier, J. P. (2000). Introduction to the Physics of the Earth´s Interior. (2nd edn).
Cambridge University Press, Cambridge.
Schick, R. and G. Schneider (1973). Physik des Erdkörpers. Eine Einfürung für
Naturwissenschafler und Ingenieure. Ferdinand Enke, Stuttgart.
Sleep, N. H. and K. Fujita (1997). Principles of Geophysics. Blackwell Science, London.
Stacey, F. D. (1992). Physics of the Earth (3rd edn). Brookfield Press, Brisbane.
Udías, A. and J. Mezcua (1997). Fundamentos de Geofísica. Alianza Universidad, Madrid.
Gravimetry
Blakely, R. J. (1995). Potential Theory in Gravity and Magnetic Applications. Cambridge
University Press, Cambridge.
Bomford, G. (1980). Geodesy (4th edn). Clarendon Press, Oxford.
Heiskanen, W.A. and H. Moritz (1985). Physical Geodesy. Freeman, San Francisco.
Hofmann-Wellenhof, B. and H. Moritz. (2006). Physical Geodesy. Springer, New York.
Lambeck, K. (1988). Geophysical Geodesy. Clarendon Press. Oxford.
352
353 Bibliography
Geomagnetism
Backus, G., R. Ladislav, and C. Constable (1996). Foundations of Geomagnetism.
Cambridge University Press, Cambridge.
Basavaiah, N. (2010). Geomagnetism: Solid Earth and Atmospheric Perspectives.
Springer, Berlin.
Blakely, R. J. (1995). Potential Theory in Gravity and Magnetic Applications. Cambridge
University Press, Cambridge.
Campbell, W.H. (2003). Introduction to Geomagnetic Fields. Cambridge University Press,
Cambridge.
Campbell, W.H. (2001). Earth Magnetism: A Guided Tour through Magnetic Fields.
Cambridge University Press, Cambridge.
Delcourt, J.J. (1990). Magnétisme Terrestre. Introduction. Masson, Paris.
Jacobs, J.A. (1987–1991). Geomagnetism (4 volumes). Academic Press, London.
Miguel, L. de (1980). Geomagnetismo. Instituto Geográfico Nacional, Madrid.
Parkinson, W.D. (1983). Introduction to Geomagnetism. Scottish Academic Press, Edinburgh.
Rikitake, T. and Y. Honkura (1985). Solid Earth Geomagnetism. Terra Scientific
Publishing, Tokyo.
Wait, J.R. (1982). Geo-electromagnetism. Academic Press, New York.
Seismology
Aki, K. and P.G. Richards (2002). Quantitative Seismology. (2nd edn). University Science
Books, Sausalito, California.
Bullen, K.E. and B.A. Bolt (1985). An Introduction to the Theory of Seismology (4th edn).
Cambridge University Press, Cambridge.
Chapman, C. (2004). Fundamentals of Seismic Wave Propagation. Cambridge University
Press, Cambridge.
Dahlen, F.A. and J. Tromp (1998). Theoretical Global Seismology. Princeton University
Press, Princeton
Gubbins, D. (1990). Seismology and Plate Tectonics. Cambridge University Press,
Cambridge.
Lay, T. and T.C. Wallace (1995). Modern Global Seismology. Academic Press, San Diego.
Kennnett, B.L.N. (2001). The Seismic Wavefield. Cambridge University Press, Cambridge.
Pujol, J. (2003). Elastic Wave Propagation and Generation in Seismology. Cambridge
University Press, Cambridge.
Shearer, P.M. (1999). Introduction to Seismology. Cambridge University Press, Cambridge.
Stein, S. and M. Wyssesion (2003). An Introduction to Seismology, Earthquakes and Earth
Structure. Blackwell Publishers, Oxford.
Udías, A. (1999). Principles of Seismology. Cambridge University Press, Cambridge.
354 Bibliography