Normal Gravity, Clairaut's Theorem and the Shape of the Earth
CSIR NET Earth Sciences Part C Geophysics: reference ellipsoid, MacCullagh's formula, J2, normal gravity formulas, Clairaut's theorem and moment of inertia.
- 20 min
On this page · 15 sections
- Prerequisites
- The Hook: A Clock That Lost Time in Cayenne
- Gravity, Gravitation and the Geopotential
- The Shape of the Earth: Four Surfaces
- The External Gravitational Potential
- Normal Gravity
- Clairaut’s Theorem
- Moment of Inertia and Dynamic Ellipticity
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- The Part B overview of spheroid, geoid and gravity in Gravity, Geomagnetism, Heat Flow and Isostasy. This page assumes those definitions and goes one level deeper.
- Newtonian potential, Laplace’s equation and Legendre polynomials from Newtonian Potential, Laplace and Poisson Equations.
- Rotational mechanics (moments of inertia, centrifugal force) as revised in Physical Sciences.
The Hook: A Clock That Lost Time in Cayenne
Gravity, Gravitation and the Geopotential
Gravitation is the Newtonian attraction of Earth’s mass. Gravity is what a gravimeter or a plumb line feels on a rotating Earth: gravitation plus the centrifugal acceleration of the rotating frame.
With the geodetic sign convention (potentials negative, acceleration \( \mathbf{g} = -\nabla W \)), the geopotential is
$$W = V + \Phi \qquad V = -G\int \frac{dm}{\ell} \qquad \Phi = -\tfrac{1}{2}\omega^2 r^2\cos^2\phi$$where \( V \) is the gravitational potential, \( \Phi \) the centrifugal potential, \( \ell \) the distance to each mass element, \( \omega \) the angular velocity and \( \phi \) the latitude. (Many geodesy texts use the opposite sign, with positive potentials and \( \mathbf{g} = +\nabla W \). The physics is identical; just stay consistent.)
Three points to fix early:
- Level (equipotential) surfaces are surfaces of constant \( W \). Gravity is everywhere perpendicular to them, so a plumb line is normal to the local level surface.
- Outside the masses \( V \) obeys Laplace’s equation; \( W \) does not, because \( \nabla^2\Phi = -2\omega^2 \) in the sign convention above (the centrifugal term is not harmonic).
- The magnitude of the centrifugal acceleration is \( \omega^2 r\cos\phi \). At the equator, \( \omega^2 a \approx 0.0339 \) m s\(^{-2}\) (about 3,392 mGal), roughly 1/288 of gravity.
Units
| Unit | Value |
|---|---|
| 1 Gal | 1 cm s\(^{-2}\) = \( 10^{-2} \) m s\(^{-2}\) |
| 1 mGal | \( 10^{-5} \) m s\(^{-2}\) |
| 1 μGal | \( 10^{-8} \) m s\(^{-2}\) |
| 1 gravity unit (g.u.) | 1 μm s\(^{-2}\) = 0.1 mGal |
| 1 Eötvös (E) | \( 10^{-9} \) s\(^{-2}\) (gradient unit; 0.1 mGal km\(^{-1}\)) |
The Shape of the Earth: Four Surfaces
| Surface | What it is | Used for |
|---|---|---|
| Sphere | radius 6,371 km (mean) | rough calculations, Part B estimates |
| Reference ellipsoid | oblate ellipsoid of revolution, chosen to be a level surface of a model field | coordinates (GPS), normal gravity |
| Geoid | the level surface of the real field that best fits mean sea level | heights above sea level, gravity anomalies |
| Topography | the physical surface | where measurements are made |
The geoid is treated on the next page. Here we build the smooth reference figure.
The reference ellipsoid (GRS80 / WGS84)
| Parameter | Symbol | Value |
|---|---|---|
| Equatorial radius | \( a \) | 6,378,137 m |
| Polar radius | \( c \) (also \( b \)) | 6,356,752.3 m |
| Flattening | \( f = (a - c)/a \) | 1/298.257 \( \approx 3.3528\times 10^{-3} \) |
| First eccentricity squared | \( e^2 = (a^2 - c^2)/a^2 \approx 2f - f^2 \) | 0.0066944 |
| Geocentric gravitational constant | \( GM \) | \( 3.986005\times 10^{14} \) m\(^3\) s\(^{-2}\) |
| Angular velocity | \( \omega \) | \( 7.292115\times 10^{-5} \) rad s\(^{-1}\) |
| Dynamic form factor | \( J_2 \) | \( 1.0826\times 10^{-3} \) |
| Mean radius | \( R \) | 6,371.0 km |
The equatorial radius exceeds the polar radius by about 21.4 km. That is tiny compared with the radius (one part in 298) but enormous compared with the geoid’s departures from the ellipsoid (about 100 m at most).
GRS80 and WGS84 share \( a \) and \( \omega \); their \( GM \) differs by about one part in \( 10^7 \) and their \( 1/f \) values agree to the sixth decimal place (298.257222 against 298.257224), far below anything CSIR NET will test.
Three latitudes
| Latitude | Defined by |
|---|---|
| Geodetic \( \phi \) | angle between the ellipsoid normal and the equatorial plane (used on maps and by GPS) |
| Geocentric \( \psi \) | angle between the radius vector from Earth’s centre and the equatorial plane |
| Astronomical \( \Phi_a \) | angle between the local plumb line (normal to the geoid) and the equatorial plane |
On an ellipsoid \( \tan\psi = (c^2/a^2)\tan\phi \). The difference \( \phi - \psi \) is zero at the equator and poles and reaches about 11.5 arc-minutes at 45°. The difference between astronomical and geodetic latitude is a component of the deflection of the vertical, a few arc-seconds normally and tens of arc-seconds near large mountains.
Why a degree is longer near the poles
The length of one degree of geodetic latitude is \( M\,\Delta\phi \), where \( M \) is the meridian radius of curvature:
$$M = \frac{a(1 - e^2)}{(1 - e^2\sin^2\phi)^{3/2}}$$The surface is flatter (larger radius of curvature) near the poles, so a degree is longer there: about 110.57 km at the equator and 111.69 km at the pole. This is exactly what the Lapland and Peru arc measurements found.
The External Gravitational Potential
MacCullagh’s formula
Far from a body of mass \( M \) with principal moments of inertia \( A \le B \le C \), the gravitational potential to order \( 1/r^3 \) is
$$V = -\frac{GM}{r} - \frac{G\,(A + B + C - 3I)}{2r^3}$$where \( I \) is the moment of inertia about the line joining the centre to the observation point. For an Earth symmetrical about its rotation axis (\( A = B \)) and with latitude \( \phi \), this becomes
$$\boxed{V = -\frac{GM}{r} + \frac{G\,(C - A)}{r^3}\,P_2(\sin\phi)} \qquad P_2(x) = \tfrac{1}{2}(3x^2 - 1)$$Check the sign: at the equator \( P_2(0) = -\tfrac{1}{2} \), so the second term makes \( V \) more negative there. At a given \( r \) the equatorial bulge pulls harder at the equator, as it must.
Spherical harmonic form and \( J_n \)
For an axially symmetric Earth the full expansion is
$$V = -\frac{GM}{r}\left[1 - \sum_{n=2}^{\infty} J_n \left(\frac{a}{r}\right)^n P_n(\sin\phi)\right]$$- There is no \( n = 1 \) term when the origin is at the centre of mass.
- \( J_2 = \dfrac{C - A}{Ma^2} \approx 1.0826\times 10^{-3} \): the dynamic form factor, the gravitational signature of the equatorial bulge. It is hundreds of times larger than any other coefficient (about 400 times \( J_3 \)).
- \( J_3 \approx -2.5\times 10^{-6} \): a small north–south asymmetry, the origin of the popular phrase “pear-shaped Earth”.
- \( J_4 \approx -1.6\times 10^{-6} \).
Non-axisymmetric terms (\( C_{nm}, S_{nm} \) with \( m \ne 0 \)) describe longitude-dependent structure; together they make up the geoid undulations.
Measuring \( J_2 \) from satellites
The equatorial bulge exerts a torque on an orbit and makes its plane precess. The nodal regression rate of a satellite in a near-circular orbit of semi-major axis \( a_s \), mean motion \( n \) and inclination \( i \) is
$$\dot\Omega = -\frac{3}{2}\,n\,J_2\left(\frac{a}{a_s}\right)^2\cos i$$Prograde orbits (\( i < 90° \)) regress westward; retrograde orbits (\( i > 90° \)) advance eastward. A Sun-synchronous orbit chooses \( i \) so that \( \dot\Omega \) equals the Sun’s apparent motion, \( 360°/365.24 \) days \( \approx 0.986° \) per day; at 700 km altitude this needs \( i \approx 98° \). Early satellite tracking in the late 1950s gave \( J_2 \) with far greater precision than surface surveys, and with it the flattening.
Normal Gravity
The normal gravity field is the field of a model Earth whose surface is the reference ellipsoid and which is itself a level surface (the “level ellipsoid”). Its value on the ellipsoid, \( \gamma(\phi) \), is the reference against which observed gravity is compared.
Somigliana’s closed formula
$$\boxed{\gamma(\phi) = \gamma_e\,\frac{1 + k\sin^2\phi}{\sqrt{1 - e^2\sin^2\phi}}} \qquad k = \frac{c\,\gamma_p}{a\,\gamma_e} - 1$$For GRS80: \( \gamma_e = 9.7803267715 \) m s\(^{-2}\), \( \gamma_p = 9.8321863685 \) m s\(^{-2}\), \( k = 0.001931851353 \), \( e^2 = 0.00669438002 \).
The series forms you will meet
| Formula | \( \gamma \) (m s\(^{-2}\)) |
|---|---|
| GRS80 (1980) | \( 9.780327\,(1 + 0.0053024\sin^2\phi - 0.0000058\sin^2 2\phi) \) |
| GRS67 (1967) | \( 9.780318\,(1 + 0.0053024\sin^2\phi - 0.0000059\sin^2 2\phi) \) |
| International Gravity Formula (1930) | \( 9.780490\,(1 + 0.0052884\sin^2\phi - 0.0000059\sin^2 2\phi) \) |
The 1930 formula is tied to the old Potsdam absolute-gravity datum. Its equatorial value is about 16 mGal higher than GRS80’s, which is why anomalies computed with different formulas cannot be mixed. The GRS80 series agrees with Somigliana’s closed form to better than 0.1 mGal.
| Latitude | GRS80 normal gravity (m s\(^{-2}\)) |
|---|---|
| 0° | 9.78033 |
| 30° | 9.79325 |
| 45° | 9.80620 |
| 60° | 9.81918 |
| 90° | 9.83219 |
Gravity increases by about 5,186 mGal from equator to pole. Three effects combine:
- Rotation removes \( \omega^2 a \approx 3{,}392 \) mGal at the equator and nothing at the poles.
- Distance: polar points are 21 km closer to the centre, which by itself increases gravity.
- Bulge mass: the extra equatorial mass partly offsets effect 2, because it lies beneath equatorial stations.
A first-order check: \( \gamma_e \approx \dfrac{GM}{a^2}\left(1 + \tfrac{3}{2}J_2\right) - \omega^2 a = 9.7983\times 1.00162 - 0.0339 \approx 9.7803 \) m s\(^{-2}\), which matches the formula.
The latitude correction in surveys
Differentiating the GRS80 formula, the north–south gradient of normal gravity is
$$\boxed{\frac{\partial\gamma}{\partial x} = 0.8140\,\sin 2\phi\ \ \text{mGal km}^{-1}}$$where \( x \) is distance along the meridian. It is greatest (0.814 mGal km\(^{-1}\)) at 45° and zero at the equator and poles. In a local survey, a station north of the base (in the northern hemisphere) sits where normal gravity is higher, so that amount is subtracted from its reading.
The normal vertical gradient
Normal gravity also decreases upward. To first order the gradient is \( -2\gamma/R \approx -0.3086 \) mGal m\(^{-1}\): the free-air gradient. The full GRS80 expression for the decrease with ellipsoidal height \( h \) (in metres) is
$$\gamma(h) - \gamma_0 \approx -(0.3087691 - 0.0004398\sin^2\phi)\,h + 7.2125\times 10^{-8}\,h^2\ \ \text{mGal}$$The latitude and \( h^2 \) terms matter only for high-precision or airborne work. For CSIR NET, 0.3086 mGal m\(^{-1}\) is the number.
Clairaut’s Theorem
Statement
Define
- geometric flattening \( f = (a - c)/a \),
- gravity flattening \( \beta = (\gamma_p - \gamma_e)/\gamma_e \),
- \( m = \omega^2 a/\gamma_e \), the ratio of centrifugal acceleration to gravity at the equator (0.003468, or 1/288.4; the form \( \omega^2 a^3/GM \) = 0.003461, or 1/288.9, is the same to first order).
Clairaut’s theorem (1743), correct to first order in the small quantities, states:
$$\boxed{f + \beta = \frac{5}{2}\,m}$$and the companion result from the potential is
$$\boxed{f = \frac{3}{2}J_2 + \frac{m}{2}}$$Eliminating \( f \) gives \( \beta = 2m - \tfrac{3}{2}J_2 \).
What makes it powerful
- It needs only that the outer surface is a level surface that is nearly ellipsoidal (a body in hydrostatic equilibrium satisfies this). It does not depend on how density varies inside.
- So gravity measured at the surface (giving \( \beta \)) and the rotation rate (giving \( m \)) are enough to find the flattening, with no geometric survey at all. This is the route Stokes later generalised: the external field determines the level surface.
Numerical check with GRS80
| Quantity | Value |
|---|---|
| \( m = \omega^2 a^3/GM \) | 0.003461 (1/288.9) |
| \( \tfrac{5}{2}m \) | 0.008653 |
| \( f \) | 0.003353 |
| \( \beta \) | 0.005302 |
| \( f + \beta \) | 0.008655 |
The two sides agree to about one part in 5,000 (the rounded table values differ by roughly 1 in 4,000), the size of the second-order terms. Similarly \( \tfrac{3}{2}J_2 + \tfrac{m}{2} = 0.001624 + 0.001731 = 0.003355 \approx 1/298.1 \), against the true 1/298.26.
Moment of Inertia and Dynamic Ellipticity
The torque of the Sun and Moon on the equatorial bulge makes the rotation axis precess with a period of roughly 26,000 years. The precession rate gives the dynamic ellipticity
$$H = \frac{C - A}{C} \approx \frac{1}{305} \approx 3.274\times 10^{-3}$$Combining it with \( J_2 = (C - A)/Ma^2 \) from satellites:
$$\boxed{\frac{C}{Ma^2} = \frac{J_2}{H} \approx \frac{1.0826\times 10^{-3}}{3.274\times 10^{-3}} \approx 0.3307}$$| Body | \( C/MR^2 \) |
|---|---|
| Uniform sphere | 0.4 |
| Hollow thin shell | 0.667 |
| All mass at centre | 0 |
| Earth | 0.3307 |
A value well below 0.4 is direct evidence that density increases with depth, the key constraint (with mean density 5.51 g cm\(^{-3}\)) on any Earth model before seismology. See Seismology and Earth Structure for how the dense core was then located.
For a body in hydrostatic equilibrium the Radau–Darwin approximation relates \( C/Ma^2 \) to the flattening:
$$\frac{C}{Ma^2} = \frac{2}{3}\left[1 - \frac{2}{5}\sqrt{\frac{5m}{2f} - 1}\right]$$With Earth’s \( m \) and \( f \) it gives about 0.331, close to the observed 0.3307. Turned around, it predicts a hydrostatic flattening a little smaller than the observed 1/298.26 (close to 1/299–1/300). The small excess flattening is usually attributed to mantle density heterogeneity and to the lag of Earth’s shape behind its slowing rotation.
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| Geometric flattening \( f = (a - c)/a \approx 1/298 \) | Gravity flattening \( \beta = (\gamma_p - \gamma_e)/\gamma_e \approx 1/189 \) | |
| \( J_2 = (C - A)/Ma^2 \approx 1.08\times 10^{-3} \) | \( H = (C - A)/C \approx 1/305 \) | |
| Gravitation (attraction only) | Gravity (attraction plus centrifugal) | |
| Geodetic latitude (ellipsoid normal) | Geocentric latitude (radius vector); differ by up to 11.5′ | |
| Astronomical latitude (plumb line, geoid normal) | Geodetic latitude (ellipsoid normal); difference = deflection of the vertical | |
| Normal gravity (on the ellipsoid, from a formula) | Observed gravity (measured, on the topography) | |
| IGF 1930 (Potsdam datum) | GRS80 (modern); about 16 mGal apart at the equator |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper. Use \( 2\pi G = 0.04193 \) mGal m\(^{-1}\) per g cm\(^{-3}\) where needed.
Level 1: Foundation
- State Clairaut’s theorem in its first-order form, defining each symbol.
- Give the physical meaning of \( J_2 \) and its approximate value for Earth.
- At which latitude is the north–south gradient of normal gravity largest?
Answers
- \( f + \beta = \tfrac{5}{2}m \), with \( f = (a - c)/a \) the geometric flattening, \( \beta = (\gamma_p - \gamma_e)/\gamma_e \) the gravity flattening and \( m = \omega^2 a/\gamma_e \) the ratio of equatorial centrifugal acceleration to gravity.
- \( J_2 = (C - A)/Ma^2 \), the second-degree zonal coefficient of the gravitational potential, measuring the equatorial bulge: about \( 1.083\times 10^{-3} \).
- 45°, where \( \sin 2\phi = 1 \) (0.814 mGal km\(^{-1}\)).
Level 2: Part B Style
Practice Q1. Using the GRS80 formula, the difference in normal gravity between latitudes 60° and 30° is closest to:
(a) 1,300 mGal (b) 2,590 mGal (c) 3,390 mGal (d) 5,190 mGal
Answer: (b) 2,590 mGal
\( \sin^2 60° - \sin^2 30° = 0.75 - 0.25 = 0.5 \), and \( \sin^2 120° = \sin^2 60° \), so the \( \sin^2 2\phi \) terms cancel.
\( \Delta\gamma = 9.780327\times 0.0053024\times 0.5 = 0.025930 \) m s\(^{-2}\) \( \approx 2{,}593 \) mGal.
(d) is the full equator-to-pole difference and (c) is the centrifugal acceleration at the equator.
Practice Q2. For Earth, \( J_2 = 1.0826\times 10^{-3} \) and \( m = \omega^2 a^3/GM = 3.461\times 10^{-3} \). To first order the flattening is:
(a) 1/289 (b) 1/298 (c) 1/305 (d) 1/924
Answer: (b) 1/298
\( f = \tfrac{3}{2}J_2 + \tfrac{1}{2}m = 1.624\times 10^{-3} + 1.731\times 10^{-3} = 3.355\times 10^{-3} \approx 1/298 \).
(a) is \( 1/m \), (c) is \( 1/H \) and (d) is \( 1/J_2 \).
Practice Q3. A hypothetical rotating planet in hydrostatic equilibrium has \( m = 0.0046 \). Surface gravity measurements give a gravity flattening \( \beta = 0.0060 \). By Clairaut’s theorem its geometric flattening is about:
(a) 1/100 (b) 1/182 (c) 1/217 (d) 1/385
Answer: (b) 1/182
\( f = \tfrac{5}{2}m - \beta = 0.0115 - 0.0060 = 0.0055 \approx 1/182 \).
Practice Q4. In a gravity survey at latitude 45° N, a station lies 5 km due north of the base. Ignoring all other effects, the latitude correction to be applied to the station reading relative to the base is:
(a) +4.07 mGal (b) −4.07 mGal (c) −2.62 mGal (d) −0.81 mGal
Answer: (b) −4.07 mGal
Gradient \( = 0.814\sin 90° = 0.814 \) mGal km\(^{-1}\); over 5 km, 4.07 mGal. Normal gravity is higher to the north, so this amount is subtracted.
(c) uses \( \sin 2\phi \) for 20° (\( \sin 40° \)), a slip in the angle; (d) is the gradient for one kilometre.
Level 3: Part C Style
Practice Q5. Satellite tracking gives \( J_2 = 1.0826\times 10^{-3} \), and astronomical precession gives \( H = (C - A)/C = 3.2738\times 10^{-3} \). Which of the following statements are correct?
- \( C/Ma^2 \approx 0.331 \).
- The result shows that density increases towards Earth’s centre.
- If Earth were homogeneous, \( C/Ma^2 \) would be 0.5.
- \( J_2 \) alone, without \( H \), is enough to find \( C/Ma^2 \).
(a) 1 and 2 only (b) 1, 2 and 3 only (c) 2 and 4 only (d) 1, 2 and 4 only
Answer: (a) 1 and 2 only
- 1 true: \( C/Ma^2 = J_2/H = 1.0826/3.2738 = 0.3307 \).
- 2 true: it is less than the uniform-sphere value, so mass is concentrated towards the centre.
- 3 false: a uniform sphere gives \( 2/5 = 0.4 \).
- 4 false: \( J_2 \) gives only the difference \( C - A \) (scaled by \( Ma^2 \)). You need \( H \), which contains \( C \) itself, to separate them. (For a planet assumed hydrostatic, Radau–Darwin can stand in for \( H \), but that is an assumption, not a measurement.)
Practice Q6. Gravity observations at sea level on a hypothetical rotating planet give the following (all on the reference surface):
| Latitude | Gravity (m s\(^{-2}\)) |
|---|---|
| 0° | 3.700 |
| 90° | 3.722 |
The planet’s equatorial radius is 3,400 km and its rotation period is 24.6 h. Treating it as a body in hydrostatic equilibrium, the geometric flattening is closest to:
(a) 1/758 (b) 1/178 (c) 1/161 (d) 1/116
Answer: (b) 1/178
\( \omega = 2\pi/(24.6\times 3600) = 7.095\times 10^{-5} \) rad s\(^{-1}\).
\( m = \omega^2 a/\gamma_e = (7.095\times 10^{-5})^2\times 3.4\times 10^{6}/3.700 = 4.626\times 10^{-3} \).
\( \beta = (3.722 - 3.700)/3.700 = 5.946\times 10^{-3} \).
\( f = \tfrac{5}{2}m - \beta = 0.011564 - 0.005946 = 0.005618 \approx 1/178 \).
Distractors: (a) is \( \beta - m \) (a misremembered theorem); (c) uses a 24 h day instead of 24.6 h, which raises \( m \) by about 5%; (d) comes from \( f = \tfrac{5}{2}m - \tfrac{1}{2}\beta \). Compute \( m \) and \( \beta \) separately to four significant figures before subtracting, because the subtraction loses a digit.
Practice Q7. A satellite is in a circular orbit 700 km above Earth’s equatorial radius. Using \( \dot\Omega = -\tfrac{3}{2}nJ_2(a/a_s)^2\cos i \) with \( GM = 3.986\times 10^{14} \) m\(^3\) s\(^{-2}\), \( a = 6{,}378 \) km and \( J_2 = 1.0826\times 10^{-3} \), which inclination makes the orbit Sun-synchronous (node advancing about 0.986° per day)?
(a) 60° (b) 82° (c) 90° (d) 98°
Answer: (d) 98°
\( a_s = 7{,}078 \) km; \( n = \sqrt{GM/a_s^3} = 1.060\times 10^{-3} \) rad s\(^{-1}\).
\( \tfrac{3}{2}nJ_2(a/a_s)^2 = 1.5\times 1.060\times 10^{-3}\times 1.0826\times 10^{-3}\times 0.812 = 1.398\times 10^{-6} \) rad s\(^{-1}\) \( \approx 6.92° \) per day.
Required: \( -6.92°\cos i = +0.986° \Rightarrow \cos i = -0.1425 \Rightarrow i \approx 98.2° \).
The node must advance (eastward), which needs \( \cos i < 0 \): a slightly retrograde orbit. (c) gives no precession; (a) and (b) make the node regress.
Quick Revision Table
| Idea | Fact or formula |
|---|---|
| Ellipsoid | \( a = 6{,}378.137 \) km, \( c = 6{,}356.752 \) km, \( f = 1/298.257 \), \( e^2 = 0.00669 \) |
| Geopotential | \( W = V + \Phi \), \( \Phi = -\tfrac{1}{2}\omega^2 r^2\cos^2\phi \) |
| MacCullagh | \( V = -GM/r + G(C - A)P_2(\sin\phi)/r^3 \) |
| \( J_2 \) | \( (C - A)/Ma^2 = 1.0826\times 10^{-3} \); \( J_3 \approx -2.5\times 10^{-6} \) |
| Nodal regression | \( \dot\Omega = -\tfrac{3}{2}nJ_2(a/a_s)^2\cos i \) |
| Somigliana | \( \gamma = \gamma_e(1 + k\sin^2\phi)/\sqrt{1 - e^2\sin^2\phi} \) |
| GRS80 | \( 9.780327(1 + 0.0053024\sin^2\phi - 0.0000058\sin^2 2\phi) \) |
| Equator / pole | 9.7803 / 9.8322 m s\(^{-2}\); difference 5,186 mGal |
| Centrifugal at equator | \( \omega^2 a = 3{,}392 \) mGal |
| Latitude gradient | \( 0.814\sin 2\phi \) mGal km\(^{-1}\) |
| Free-air gradient | 0.3086 mGal m\(^{-1}\) |
| Clairaut | \( f + \beta = \tfrac{5}{2}m \); \( f = \tfrac{3}{2}J_2 + \tfrac{1}{2}m \) |
| \( m \) | \( \omega^2 a/\gamma_e \approx 1/288 \) (\( \omega^2 a^3/GM \approx 1/289 \)) |
| Dynamic ellipticity | \( H = (C - A)/C \approx 1/305 \) |
| Moment of inertia | \( C/Ma^2 = J_2/H = 0.3307 \) (uniform 0.4) |
| Latitudes | geodetic vs geocentric: up to 11.5′ at 45° |
Key Takeaways
What’s Next
- Geoid; Free-Air, Bouguer and Isostatic Anomalies: compare real gravity with the normal field and read the anomalies.
- Newtonian Potential, Laplace and Poisson Equations: the potential theory behind the expansions used here.
- Gravity and Magnetic Fields of the Earth: back to the chapter overview.