GEOL30005 Applied Geophysics: Challenge For Gravity Processing
GEOL30005 Applied Geophysics: Challenge For Gravity Processing
GEOL30005 Applied Geophysics
Lecture 5: Gravity processing
• In the gravity theory lecture, we said that the normal value
of gravity at the earth’s surface:
9.8 m/s2 = 980 Gal = 980,000 mGal = 9,800,000 GU
• Variations in acceleration (g) due to geology are in the
order of 1 mGal
• Need to measure g better than 1 part in 1 million – and
this is just for conventional gravity!
• Today’s lecture is about removing the effect of all other
things with contribute to gravity so we’re left with the
gravity response of the geology.
1
8/15/2013
Gravity corrections = Airborne surveys only
All of these may impart an effect on the response
that are at least as large as the influence of geology
Gravity corrections
• This lecture will step through the process required to take raw gravity
values, and convert them into a Bouguer gravity anomaly map
• We will be doing some of these corrections using land based gravity in the
prac this afternoon
Raw gravity Processed Bouguer gravity
2
8/15/2013
Gravity corrections
• This example uses airborne gravity data collected in early 2012 in the
on/off shore Gippsland Basin. The data was acquired to try to
understand more about the carbon storage potential of this region.
Gravity Correction ‐ Instrument drift
Drift: the gradual and unintentional change in the reference value with
respect to which measurements are made.
• For a gravity meter this can be up to 0.1 mGal per day.
• Because of change in the physical properties of the instrument (pressure, temperature,
spring creep).
• Imagine what would happen if you dropped one or shake one in an aeroplane?
• Drift is due to internal changes in the mechanism of the gravity meter, and is assumed to
be linear over short intervals of time (a few hours).
• However, a linear assumption is not always
justified, since sudden small jumps in the data,
called tares, do occur.
In this example the blue line represents
oscillatory trends in the gravitational
y g
acceleration caused by tidal effects related to
the sun and moon (tidal attraction).
The red line is a straight line least square fit to
the data, which represents the instrument drift.
Wolf, A. Tidal Force Observations, Geophysics, V, 317‐320, 1940
3
8/15/2013
Tidal effect: Variations in the gravity observations resulting from the
attraction of the sun and the moon and the distortion of the Earth
• Tidal effects are real changes in the gravitational acceleration (unlike instrument drift)
• Not related to geology and are therefore considered noise
g gy
• Caused because the gravitational force of the sun and the moon distort the shape of
the Earth – subtly
• Varies from place to place but can be as
great as 0.2 mGal
Wolf, A. Tidal Force Observations, Geophysics, V, 317‐320, 1940
Removal of aircraft accelerations using GPS
• In survey flying, accelerations in an aircraft can reach 0.1 G (or
more), equivalent to 100,000 mGal.
• Data processing must extract gravity data from this very noisy
environment.
• This is achieved by modelling the movements of the aircraft in
flight by extremely accurate GPS measurements (ie. we calculate
the acceleration of the aircraft using GPS, and remove it from the
gravity data).
• Airborne gravity contractors have their own proprietary software
for this correction, so you don’t see how this is performed in their
provided report.
id d
Just be aware that this is a large correction, and airborne
gravity would not be possible without it.
4
8/15/2013
Theoretical gravity correction
(aka latitude correction)
This one is by far the biggest correction because
you are removing the gravitational effect of the
earth.
Correction accounts for 2 things:
• Shape (source‐sensor distance)
• Rotation (centrifugal force opposed to
gravitational force)
• The Earth is shaped as an ellipsoid, widest at
the equator (radius variation up to 22km or
0.3% of the Earth
0 3% of the Earth’ss radius) and flattened at
radius) and flattened at
the poles.
• The ellipsoidal shape of the earth produces
variations in acceleration that could be
confused with geology.
Theoretical gravity correction
(aka latitude correction)
• Because of:
5
8/15/2013
Theoretical gravity correction
(aka latitude correction)
Rotation of the Earth also causes a change in
the gravitational acceleration with latitude.
Centrifugal force: outward directed force of a
rotating body
rotating body.
Force is proportional to the distance from the
axis of rotation and the rate of rotation.
On the Earth the rate of rotation does not vary
with location but the distance does.
Large at the equator and zero at the poles.
Because the force is always away from the axis
of rotation it reduces the gravitational
acceleration.
Theoretical gravity correction
(aka latitude correction)
Assuming the earth is elliptical & rotating at the appropriate rate and
made of homogeneous material (no variation in density distribution
due to geological variation), the correction for latitude taking into
account centrifugal force and sensor‐source distance variation is:
Where λ = latitude.
Correction is greatest at mid‐latitudes. So, at a latitude of 45°, g would vary
by 0.81mgal/km. To reach an accuracy of 0.01mgal, we need to know
our N‐S position within 12m.
6
8/15/2013
Theoretical gravity correction Units in mGal
(aka latitude correction)
Theoretical Theoretical
Raw gravity ‐ gravity
correction
= corrected
gravity
min 973433 05
973433.05 min 979973 75
979973.75 min 6585 38
‐6585.38
max 986453.77 max 980074.61 max 6459.77
range 13020.72 range 100.86 range 13045.15
mean 979924.75 mean 980018.75 mean ‐94
Theoretical gravity correction
This one still gets
(aka latitude correction) used today
Latitude correction formulas:
1930 International Gravity Formula:
9 78049 (1 + 0 0052884 i 2 2 λ – 0.0000059 sin
g0 = 9.78049 (1 + 0.0052884 sin 0 0000059 i 2 2 2
2 λ)
2 λ
1967 International Gravity Formula:
g0 = 9.78031846 (1 + 0.0053024 sin2 λ – 0.0000058 sin2 2
2 λλ)
Most recently IAG developed Geodetic Reference System 1980
Most recently IAG developed Geodetic Reference System 1980, ,
leading to World Geodetic System 1984 (WGS84); in closed
leading to World Geodetic System 1984 (WGS84); in closed
form it is:
7
8/15/2013
Eötvös correction
• Loránd Eötvös (27 July 1848 – 8 April 1919)
• Hungarian physicist known for his work on
gravitation and surface tension
Unit for gravity gradient was named after him as
• Unit for gravity gradient was named after him as
well as the Eötvös correction.
• Loránd Eötvös realised that because the earth
rotated toward the east, the angular velocity of
an observer moving east is greater than for an
observer remaining stationary.
q y, g
• Consequently, gravitational attraction will be
slightly reduced for the moving observer.
• Similarly, gravitational attraction will be slightly
increased for an observer moving in a westerly
direction.
Eötvös correction
• This effect is called the Eötvös effect and must be accounted for in
gravity surveys on a moving platform.
• There are lots of versions of this formula, however this one was used
in 2011 for the Gippsland gravity data.
Where:
• φ is the latitude of the aircraft
is the latitude of the aircraft
• vx and vy are the velocities of the aircraft in the x (east) and y (north) direction
• r is the Earth’s radius at the equator (6,378,137 m)
• e2 is a correction for Earth’s flattening towards the poles (6.69437999013 × 10‐3)
• Ws is the angular velocity of Earth's rotation (7.2921158553 × 10‐5 rad/s)
• h is the altitude of the plane above the GRS‐80 ellipsoid
8
8/15/2013
Theoretical
corrected
gravity
‐ Eötvös
correction = Eötvös corrected
gravity
i
min 6585 38
‐6585.38 i
min 435 55
‐435.55 i
min 4924 28
‐4924.28
max 6459.77 max 697.07 max 4337.7
range 13045.15 range 1132.62 range 9261.98
mean ‐94 mean 16.89 mean ‐77.2
Free Air correction – Free air gravity
• Topography is an additional complication adding mass in regions
of topography (increasing the gravity pull) but also increasing
the sensor‐source distance (decreasing the gravity pull).
• The Free Air correction will take care of sensor‐source distance increase while
terrain correction and Bouguer correction will account for added mass due to
topography
topography.
• If two perfect gravimeters were to collect data from the same position on the
Earth, one from a ladder and one from the ground they would give different
readings.
• The measurement from the ladder would record a smaller gravitational
acceleration. We need to take into account station elevation when measuring
gravity acceleration and correct to a common elevation/datum.
This is known as the free air correction. Gravity measurement
• This is known as the free air correction Gravity measurement
corrected to a common datum is also called Free Air Gravity.
FA = ‐0.3086 x h
• To obtain an accuracy of 0.01mgal, we need to know station
elevation within 3cm
9
8/15/2013
Free‐air correction Units in mGal
Eötvös corrected
gravity ‐ Free‐air
correction = Free‐air
gravity
min ‐4924.28
4924 28 min 45 84
45.84 min ‐4871 06
4871.06
max 4337.7 max 113.99 max 4399.27
range 9261.98 range 68.15 range 9270.33
mean ‐77.2 mean 57.4 mean ‐19.8
Bouguer correction
• Pierre Bouguer (16 Feb 1698 ‐ 15 Aug 1758)
• French mathematician, geophysicist,
geodesist
• Bouguer measured gravity by pendulum at
different altitudes and was the first to
attempt to measure the horizontal
gravitational pull of mountains.
So, the Bouguer
• So the Bouguer correction
correction – which accounts for the gravitational
which accounts for the gravitational
effect of mountains/topography has been named after him.
10
8/15/2013
Bsb = 2πϒρh
• Where ϒ is the gravitational
constant
• ρ is the density (typically 2.67)
• h is the height above a
h is the height above a
common datum
Bsb = 0.04193 ρ h (mgal)
Bouguer correction –
Variations in gravity due to topography
• What if there is rugged topography?
• A peak, excess of mass, will pull g towards the peak thus reducing the
vertical component of g. A valley, mass deficit, also reduces g.
• It is possible to calculate the gravity effect of topography assuming a
homogeneous value of density (e.g. same as for Bouguer) and remove
this effect from the gravity data. This is typically called a ‘Terrain
correction’
11
8/15/2013
Terrane corrections
• A terrain correction is a better alternative to
a Bouguer correction.
• Sometimes called “Complete Bouguer”.
• The terrain correction calculates the
gravitational effect of terrain
everywhere for each point in the
survey
• So unlike the simple Bouguer
correction which assumes a slab
extends to infinity, the terrain
correction actually calculates the
correction actually calculates the
gravitational effect of topography
at some distance away from where
the measurement was made.
TCBA_p360
0
1400
TCBA_p350
TCBA_p340 -100
TCBA_p330 1200
TCBA_p320
-200
TCBA_p310 1000
-300
TCBA_p300
TCBA_290
TCBoug (ums-2)
-400 800
Elev (m)
TCBA_p280
TCBA_p270 -500
600
TCBA_p267_25052004
TCBA_p260 -600
400
TCBA_p250
TCBA p250
-700
TCBA_p240
TCBA_p230 200
-800
TCBA_p220
TCBA_p210 -900 0
2000320835
2000320836
2000320837
2000320838
2000320839
2000320840
2000322506
2000322505
2000322504
2000322503
2000322074
2000322502
2000322501
2000322500
2000322000
2000322001
2000322002
2000322003
2000322512
2000322507
2000322508
2000322509
2000322510
2000322511
TCBA_p200
Elev
Station Number
12
8/15/2013
Free‐air gravity ‐ Bouguer
correction = Bouguer
corrected gravity
min ‐4871 06
4871.06 min ‐11 59
11.59 min ‐4871.48
max 4399.27 max 3.79 max 4395.58
range 9270.33 range 15.38 range 9267.06
mean ‐19.8 mean 0.7 mean ‐19.11
Bouguer
Bouguer gravity
corrected gravity
filtered
min ‐4871.48 min ‐41.67
max 4395.58 max 10.61
range 9267.06
3000m low
range 52.28
mean ‐19.11 pass filter mean ‐19.02
13
8/15/2013
Filtering
• Filtering is not removing noise – it is just suppressing it.
Removing noise is when you can quantify the effect
of something (eg terrain) and extract it out.
of something (eg terrain) and extract it out
• Filtering is not nearly as important for land based gravity
• We’ll do more with filtering during the image
enhancement lecture (10)
Gravity stripping
• Free‐air, Bouguer and terrain corrected gravity can be used for
interpretation / modelling depending on the intended application.
For example, if we’re doing gravity modelling, the software
may require free‐air corrected gravity –
i f i d i i hi
in this case, you would
ld
need to include topography in the model.
• However, there are some further things we can do to gravity data
depending on the circumstances.
• If we know some key characteristics about something we want to
remove from the gravity, its possible to calculate the gravity
response of it, and subtract it out.
• Strictly speaking, this is not really a correction because the signal is
geological – so the process is often referred to as gravity stripping
14
8/15/2013
Terrain corrected Bouguer anomaly
Isostatic correction
• The principle of isostasy states that mass excesses, i.e.
topographic loads at the surface, are compensated by
mass deficiencies at depth, which are referred to as
isostatic roots.
The Moho is therefore deeper belo mo ntains
• The Moho, is therefore deeper below mountains.
• The effect of these mass deficiencies are not accounted
for in the Bouguer or terrain correction.
• The isostatic correction Isostatic correction Isostatic residual bouguer
results in an inverse
correlation between broad
Bouguer anomaly lows and
positive topography.
iti t h
• The isostatic correction
removes the gravity effect
of the isostatic roots.
Gravity stripping of an ice sheet
• Its also possible to calculate the gravitational effect of an
ice sheet, and remove it from the observed gravity
• This works relatively well since we know the density of
ice to be ~0.87g/cm3
Ice sheet surface Sub‐Glacial Topography
Sub‐ice topography
15
8/15/2013
Gravity stripping of an ice sheet
Surface topography
Sub‐ice topography Free Air ‐ Filtered Ice corrected Bouguer Gravity
Temporal Variation Correction:
• Instrument drift
• Tidal Correction
Spatial Variation Correction:
p
• Latitude Correction
• Free Air Correction
• Bouguer Correction/ Terrain Correction
Airborne corrections
• Eötvös Correction
• Aircraft accelerations calculated from GPS
Gravity stripping
• Eg. removal of ice /
• Isostatic effect of mountain roots
The remaining anomalies are due to geological variation in the subsurface.
16