GE 153 EARTH’S GRAVITY AND GEOID MODELING

ROSALIE B. REYES, Dr. Eng.

Gravity reduction
 Gravity g is measured on the surface of the earth.
 Normal gravity γ is referred to the surface of the ellipsoid.
 To refer g to mean sea level or geoid, a reduction is
necessary.
 3 main purposes of gravity reduction:
 Determination of the geoid
 Interpolation and extrapolation of gravity
 Investigation of the earth’s crust

Geoid determination requires that:

1. g must refer to the geoid; and
2. There must not be masses outside the geoid
Gravity reduction

Gravity reduction consists of the ff steps:

1. Complete removal or shifting of the topographic
masses below sea level; then
2. The gravity station is lowered from the earth’s surface
(point P to the geoid (point P0)
Auxiliary formulas
 Let U be the potential and A the vertical
attraction of a homogeneous circular
cylinder of radius a and height b at a
point P situated on its axis at a height c
above its base.
 P outside cylinder:

See derivation on Hofmann-Wellenhof and Moritz

 P on cylinder: c=b

 P inside cylinder: c<b; U is the sum of 2 parts

 Circular disk: let b go to zero such that the product of κ=bρ
remains finite. The quantity κ may then be considered as
the surface density with which matter is concentrated on
the surface of a circle of radius a. Thus:

 Sectors and compartments: for a sector of radius a and

angle α=2π/n.
Free-air reduction
 To reduce g that is measured at the earth’s surface the
vertical gradient of gravity must be known.

 The value of g0 at the geoid may be obtained as a Taylor

expansion:

where H is the height between P and Po.

Suppose there are no masses above the geoid then:

where is the Free-air reduction to the geoid

Free-air reduction
 The assumption of no masses above the geoid means
that such masses have been mathematically removed
beforehand, so that this reduction is carried out “in the
free air”
 For practical purposes it is sufficient to use instead of
the normal gradient of gravity (associated
with the ellipsoidal height h) , thus:
Bouger reduction
The objective of the Bouger reduction of gravity is the
complete removal of the topographic masses, that is, the
masses outside the geoid.
The Bouger plate
Assume that the area around the gravity station P to be
completely flat and let the masses between the geoid and
the earth’s surface have a constant density ρ. Then the
attraction A of this plate regarded as circular cylinder of
thickness H and infinite radius is given by:

that is the attraction of an infinite Bouger plate. With

standard density this becomes
Bouger reduction
 Removing the plate is equivalent to subtracting its
attraction from the observed gravity. This is called the
incomplete Bouger reduction.
 To complete the gravity reduction, the free-air
reduction is applied hence:

 This is now called the complete Bouger reduction.

Its result is Bouger gravity at the geoid.
The Bouger gravity at the geoid
gravity measured at P g
minus Bouger plate -0.1119 H
plus free air reduction +0.3086 H
-------------------------------------------------------------
Bouger gravity at P0 gB =g+0.1967 H

The gravity anomalies is now computed by:

where γ is the normal gravity referred to the ellipsoid.

The Normal gravity at the ellipsoid
 The γ, which is also known as normal gravity, is
determined by using the International Gravity
Formula 1980 (Vanicek and Krakiwsky, 1986):
γ = 978.0327(1+0.0052790414 sin2φ + 0.0000232718
sin4 φ + 0.0000001262 sin6φ) Gal.

 For the WGS84 ellipsoid the normal gravity is given by

Blakely, (1995 p. 136) in the following closed-form
formula:
Normal gravity values

GRS 1980 WGS84

γa (at the equator) 9.780 326 7715 9.780 325 3359
γb (at the pole) 9.832 186 3685 9.832 184 9378
Terrain correction
The Bouger anomalies are refined by taking account the
deviation of the actual topography from the Bouger plate at P.

At A the mass surplus Δm+, which attracts upward, is removed

causing g at P to increase. At B the mass deficiency Δm- is
made up causing g at P to increase again. The terrain
correction is always positive.
Terrain correction
The terrain correction or topographic correction is
given as:

By adding the terrain correction the refined Bouger

gravity is given as:

The Bouger anomalies are called refined or

simple depending on whether the terrain correction
has been applied or not.
Unified Procedure
It is possible to compute the total effect of the topographic
masses,
in one step by using columns with base at sea level by
subdividing the terrain by means of a template.
Poincare’ and Prey reduction
Suppose we need the gravity g’ inside the earth. Since g’
cannot be measured it must be computed from surface
gravity. This is done by reducing the measured gravity
using Poincare and Prey.
 gQ can be computed if the actual gravity gradient
∂g/∂H inside the earth were known.
 It can be obtained by Brun’s formula:

if the mean curvature J of the geopotential surfaces

and the density ρ are known between P and Q.
 The normal free-air gradient is given by

Jo is the mean curvature of the spheropotential

surfaces.
If then

with ρ=2.67 g/cm3 and G=6.67.10-11m3kg-1s-2

so that

This simple formula, although being rather crude is

often applied in practice.
Alternative way of computing gQ
3 steps of computing gQ
1. Remove all masses above the geopotential surface
W=WQ which contains Q, and subtract their
attraction from g at P.
2. Since the gravity station P is now “in free-air”, apply
the free-air reduction, thus moving the gravity
station from P to Q.
3. Restore the removed masses to their former position,
and add algebraically their attraction to g at Q.
 If the terrain correction s neglected and only the
Bouger plate between P and Q of the normal density
ρ=2.67 g cm-3 :
Gravity measured at P gP
1. Remove Bouger plate -0.1119(HP-HQ)
2. Free-air reduction from P to Q +0.3086(HP-HQ)
3. Restore Bouger plate -0.1119(HP-HQ)
-------------------------------------------------------------------
together gravity at Q gQ=gP+0.0848(HP-HQ)
The Poincare’ and Prey or just Prey reduction
cannot be used for the determination of the
geoid but to obtain orthometric heights.
 Isostatic reduction
 One might be inclined to assume that the topographic masses
are simply superposed on an essentially homogeneous crust.
 If this were true, the Bouger reduction would remove the main
irregularities of the gravity field. However, just the opposite is
true.
 Bouger anomalies in mountainous areas are systematically
negative and may attain large values.
 It increases in magnitude on the average of 100 mgal per 1000 m
of elevation.
 The only explanation possible is that there is some kind of mass
deficiency under the mountains.
 This means that topographic masses are compensated in some
way.
Isostatic reduction
 There is a similar effect for the deflections of the vertical
(DOV).
 The actual deflections are smaller than the visible
topographic masses.
 J.H. Pratt observed such effect in Himalayas
 At one station he computed a value of 25” for the DOV
from the attraction of the visible masses of the mountains.
 But the value obtained through astrogeodetic
measurements was only 5”.
 Some kind of compensation is needed to account for this
discrepancy.
Isostatic reduction
Two different theories
 J.H. Pratt – the mountains have risen from the
underground somewhat like a fermenting dough.
 Airy – the mountains are floating on a fluid lava of
higher density (somewhat like an iceberg floating on
water), so that the higher the mountain, the deeper it
sinks.

For the mathematical formulation refer to Physical

Geodesy by Hofmann-Wellenhof and Moritz, pp. 141-155
Pratt-Hayford model

(D+H) ρ = D ρ0

ρ0 = 2.67 g/cm3
Airy-Heiskanen model
ρ0 = 2.67 g/cm3

ρ1 = 3.27 g/cm3

Δρ = 3.27 - 2.67= 0.6 g/cm3

t Δρ= H ρ0

t = 4.45 H
 Vening Meinesz
 Pratt and Airy models assume that
the compensation is strictly local;
that they take place along vertical
columns.
 Vening Meinesz modified Airy
floating theory by introducing
regional instead of local
compensation.
 In his theory, the topography is From gravimetric
considered as a load of unbroken evidence alone, it is
but yielding elastic crust. difficult to decide which
model best accounts for
 Although more realistic, it is more
this compensation.
complicated and is, therefore,
seldom used by geodesist.
Topographic-isostatic reduction
 The objective of the topographic-isostatic reduction of
gravity is the regularization of the earth’s crust.
 Regularization means making the earth’s crust as
homogeneous as possible.
 In other terms, the topography is removed together with its
compensation and the final result is ideally a homogeneous
crust of density ρ0 and constant thickness D (Pratt-
Hayford) or T (Airy-Heiskanen).
 Thus, we have 3 steps:
1. Removal of topography
2. Removal of compensation
3. Free-air reduction to the geoid
Topographic-isostatic reduction
The attraction of the (negative compensation) is computed by:

Pratt-Hayford:

Airy-Heiskanen:
Total reduction
The topographic-isostatically reduced gravity on the geoid
becomes:

Attraction of Topography = effect of

Attraction of the Compensation
Bouger plate + terrain correction

Topographic-isostatic anomalies:
If any of these systems were true, then the reduction would fulfill
its goal of complete regularization of the earth’s crust then the
topographic-isostatic anomalies would be zero.
In reality this is not true, there will be topographic-isostatic
anomalies that will be left, but they will be small, smooth, more or
less randomly positive and negative and independent of elevation.
Hence, they are better suited for interpolation or extrapolation.
The indirect effect
 The removal or shifting of masses underlying the gravity
reductions change the gravity potential and, hence, the
geoid.
 This change of the geoid is an indirect effect of the gravity
reductions.
 Thus, the surface computed by Stokes’ formula from
topographic-isostatic gravity anomalies, is not the geoid
but a slightly different surface, the cogeoid.
 Then the undulation N is obtained from N=Nc+δN and
change in potential
at the geoid
Helmert’s condensation reduction
Here the topography is condensed so as to form a surface
layer (somewhat like a glass sphere of very thin but very
heavy and robust glass) on the geoid so that the total mass
remains unchanged.

Helmert’s condensation may

be considered as limiting
case of an isostatic
reduction according to the
Pratt-Hayford system as D
goes to zero.
Helmert’s condensation reduction
We have
is to be computed using Ac
with c = HP and κ = ρH; Hp is the height of the station P
and H is the height of the compartment.
The indirect effect is δW = UT – UC
Uc
with c = 0 since it refers to the geoidal point P0 and κ = ρH
The corresponding δN is very small, amounting to about 1 m
per 3 km of average elevation and may be neglected so that
the cogeoid of the condensation reduction practically
coincides with the actual geoid.
Helmert’s condensation reduction
Even the direct effect –AT+AC, can usually be neglected, as the
attraction of the Helmert’s layer compensates that of
topography.
There remains that is, the simple free-air
reduction.
In this sense, the simple free-air reduction may be considered
as giving approximate boundary values at the geoid to be used
in Stokes’ formula. To the same degree of approximation, the
“free-air cogeoid” coincides with the actual geoid.
Hence, free-air anomalies may be
considered as approximations of “condensation anomalies”