Earth radius

Rotation of a planet causes it to approximate an oblate ellipsoid/spheroid with a bulge at the equator and flattening at the North and South Poles, so that the equatorial radius a is larger than the polar radius b by approximately aq. The oblateness constant q is given by

${\displaystyle q={\frac {a^{3}\omega ^{2}}{GM}}\,,}$

where ω is the angular frequency, G is the gravitational constant, and M is the mass of the planet.[lower-alpha 5] For the Earth 1/q ≈ 289, which is close to the measured inverse flattening 1/f ≈ 298.257. Additionally, the bulge at the equator shows slow variations. The bulge had been decreasing, but since 1998 the bulge has increased, possibly due to redistribution of ocean mass via currents.[4]

The variation in density and crustal thickness causes gravity to vary across the surface and in time, so that the mean sea level differs from the ellipsoid. This difference is the geoid height, positive above or outside the ellipsoid, negative below or inside. The geoid height variation is under 110 m (360 ft) on Earth. The geoid height can change abruptly due to earthquakes (such as the Sumatra-Andaman earthquake) or reduction in ice masses (such as Greenland).[5]

Not all deformations originate within the Earth. Gravitational attraction from the Moon or Sun can cause the Earth's surface at a given point to vary by tenths of a meter over a nearly 12-hour period (see Earth tide).

Al-Biruni's (973–1048) method for calculation of the Earth's radius simplified measuring the circumference compared to taking measurements from two locations distant from each other.

Given local and transient influences on surface height, the values defined below are based on a "general purpose" model, refined as globally precisely as possible within 5 m (16 ft) of reference ellipsoid height, and to within 100 m (330 ft) of mean sea level (neglecting geoid height).

Additionally, the radius can be estimated from the curvature of the Earth at a point. Like a torus, the curvature at a point will be greatest (tightest) in one direction (north–south on Earth) and smallest (flattest) perpendicularly (east–west). The corresponding radius of curvature depends on the location and direction of measurement from that point. A consequence is that a distance to the true horizon at the equator is slightly shorter in the north–south direction than in the east–west direction.

In summary, local variations in terrain prevent defining a single "precise" radius. One can only adopt an idealized model. Since the estimate by Eratosthenes, many models have been created. Historically, these models were based on regional topography, giving the best reference ellipsoid for the area under survey. As satellite remote sensing and especially the Global Positioning System gained importance, true global models were developed which, while not as accurate for regional work, best approximate the Earth as a whole.

The Earth's equatorial radius a, or semi-major axis, is the distance from its center to the equator and equals 6,378.1370 km (3,963.1906 mi).[8] The equatorial radius is often used to compare Earth with other planets.

The Earth's polar radius b, or semi-minor axis, is the distance from its center to the North and South Poles, and equals 6,356.7523 km (3,949.9028 mi).

The distance from the Earth's center to a point on the spheroid surface at geodetic latitude φ is:

${\displaystyle R(\varphi )={\sqrt {\frac {(a^{2}\cos \varphi )^{2}+(b^{2}\sin \varphi )^{2}}{(a\cos \varphi )^{2}+(b\sin \varphi )^{2}}}}}$

where a and b are, respectively, the equatorial radius and the polar radius.

• Maximum: The summit of Chimborazo is 6,384.4 km (3,967.1 mi) from the Earth's center.
• Minimum: The floor of the Arctic Ocean is approximately 6,352.8 km (3,947.4 mi) from the Earth's center.[9]

There are two principal radii of curvature: along the meridional and prime-vertical normal sections. The curvatures are the roots of Equation (125) in:[10]

${\displaystyle (EG-F^{2})\,\kappa ^{2}-(eG+gE-2fF)\,\kappa +(eg-f^{2})=0=\det(A-\kappa \,B),}$

where in the first fundamental form for a surface (Equation (112) in [10]):

${\displaystyle ds^{2}=\sum _{ij}a_{ij}dw^{i}dw^{j}=E\,d\varphi ^{2}+2F\,d\varphi \,d\lambda +G\,d\lambda ^{2},}$

E, F, and G are elements of the metric tensor:

${\displaystyle A=a_{ij}=\sum _{\nu }{\frac {\partial r^{\nu }}{\partial w^{i}}}{\frac {\partial r^{\nu }}{\partial w^{j}}}=\left[{\begin{array}{ll}E&F\\F&G\end{array}}\right],}$

${\displaystyle r=[r^{1},r^{2},r^{3}]^{T}=[x,y,z]^{T}}$, ${\displaystyle w^{1}=\varphi }$, ${\displaystyle w^{2}=\lambda ,}$

in the second fundamental form for a surface (Equation (123) in [10]):

${\displaystyle 2D=\sum _{ij}b_{ij}dw^{i}dw^{j}=e\,d\varphi ^{2}+2f\,d\varphi \,d\lambda +g\,d\lambda ^{2},}$

e, f, and g are elements of the shape tensor:

${\displaystyle B=b_{ij}=\sum _{\nu }n^{\nu }{\frac {\partial ^{2}r^{\nu }}{\partial w^{i}\partial w^{j}}}=\left[{\begin{array}{ll}e&f\\f&g\end{array}}\right],}$

${\displaystyle n={\frac {N}{|N|}}}$ is the unit normal to the surface at ${\displaystyle r}$, and because ${\displaystyle {\frac {\partial r}{\partial \varphi }}}$ and ${\displaystyle {\frac {\partial r}{\partial \lambda }}}$ are tangents to the surface,

${\displaystyle N={\frac {\partial r}{\partial \varphi }}\times {\frac {\partial r}{\partial \lambda }}}$

is normal to the surface at ${\displaystyle r}$.

With ${\displaystyle F=f=0}$ for an oblate spheroid, the curvatures are

${\displaystyle \kappa _{1}={\frac {g}{G}}}$ and ${\displaystyle \kappa _{2}={\frac {e}{E}}\,,}$

and the radii of curvature are

${\displaystyle R_{1}={\frac {1}{\kappa _{1}}}}$ and ${\displaystyle R_{2}={\frac {1}{\kappa _{2}}}.}$

Geometrically, the second fundamental form gives the distance from ${\displaystyle r+dr}$ to the plane tangent at ${\displaystyle r}$.

In particular, the Earth's radius of curvature in the (north–south) meridian at φ is:

${\displaystyle M(\varphi )=R_{1}={\frac {(ab)^{2}}{{\big (}(a\cos \varphi )^{2}+(b\sin \varphi )^{2}{\big )}^{\frac {3}{2}}}}={\frac {a(1-e^{2})}{(1-e^{2}\sin ^{2}\varphi )^{\frac {3}{2}}}}={\frac {1-e^{2}}{a^{2}}}N(\varphi )^{3}\,.}$

where ${\displaystyle e}$ is the eccentricity of the earth. This is the radius that Eratosthenes measured.

If one point had appeared due east of the other, one finds the approximate curvature in the east–west direction.[lower-alpha 6]

This radius of curvature in the prime vertical which is perpendicular (normal or orthogonal) to M at geodetic latitude φ is:[lower-alpha 7]

${\displaystyle N(\varphi )=R_{2}={\frac {a^{2}}{\sqrt {(a\cos \varphi )^{2}+(b\sin \varphi )^{2}}}}={\frac {a}{\sqrt {1-e^{2}\sin ^{2}\varphi }}}\,.}$

This radius is also called the transverse radius of curvature. At the equator, N = R. B. R. Bowring [11] gives a geometric proof that this is the perpendicular distance from the surface to the polar axis.

Three different radii as a function of Earth's latitude. R is the geocentric radius; M is the meridional radius of curvature; and N is the prime vertical radius of curvature.

The Earth's meridional radius of curvature at the equator equals the meridian's semi-latus rectum:

b²/a = 6,335.439 km

The Earth's polar radius of curvature is:

a²/b = 6,399.594 km

The Earth's radius of curvature along a course at an azimuth (measured clockwise from north) α at φ is derived from Euler's curvature formula as follows:[12]^:97

${\displaystyle R_{\mathrm {c} }={\frac {1}{{\dfrac {\cos ^{2}\alpha }{M}}+{\dfrac {\sin ^{2}\alpha }{N}}}}\,.}$

It is possible to combine the principal radii of curvature above in a non-directional manner.

The Gaussian curvature is ${\displaystyle K=\kappa _{1}\,\kappa _{2}={\frac {\det \,B}{\det \,A}}}$. The Earth's Gaussian radius of curvature at latitude φ is:[12]

${\displaystyle R_{\mathrm {a} }(\varphi )={\frac {1}{\sqrt {K}}}={\frac {1}{2\pi }}\int _{0}^{2\pi }R_{\mathrm {c} }(\alpha )\,d\alpha \,={\sqrt {MN}}={\frac {a^{2}b}{(a\cos \varphi )^{2}+(b\sin \varphi )^{2}}}={\frac {a{\sqrt {1-e^{2}}}}{1-e^{2}\sin ^{2}\varphi }}\,.}$

The Earth's mean radius of curvature at latitude φ is:[12]^:97

${\displaystyle R_{\mathrm {m} }={\frac {2}{{\dfrac {1}{M}}+{\dfrac {1}{N}}}}\,\!}$

Equatorial (a), polar (b) and mean Earth radii as defined in the 1984 World Geodetic System revision (not to scale)

In geophysics, the International Union of Geodesy and Geophysics (IUGG) defines the mean radius (denoted R₁) to be[2]

${\displaystyle R_{1}={\frac {2a+b}{3}}\,\!}$

For Earth, the mean radius is 6,371.0088 km (3,958.7613 mi).[13]

In astronomy, the International Astronomical Union denotes the nominal equatorial Earth radius as ${\displaystyle {\mathcal {R}}_{\mathrm {eE} }^{\mathrm {N} }}$, which is defined to be 6,378.1 km (3,963.2 mi).[1]^:3 The nominal polar Earth radius is defined as ${\displaystyle {\mathcal {R}}_{\mathrm {pE} }^{\mathrm {N} }}$ = 6,356.8 km (3,949.9 mi). These values correspond to the zero tide radii. Equatorial radius is conventionally used as the nominal value unless the polar radius is explicitly required.[1]^:4

Earth's authalic ("equal area") radius is the radius of a hypothetical perfect sphere that has the same surface area as the reference ellipsoid. The IUGG denotes the authalic radius as R₂.[2]

A closed-form solution exists for a spheroid:[14]

${\displaystyle R_{2}={\sqrt {\frac {a^{2}+{\frac {b^{2}}{e}}\ln {\left({\frac {1+e}{b/a}}\right)}}{2}}}={\sqrt {{\frac {a^{2}}{2}}+{\frac {b^{2}}{2}}{\frac {\tanh ^{-1}e}{e}}}}={\sqrt {\frac {A}{4\pi }}}\,,}$

where e² = a² − b²/a² and A is the surface area of the spheroid.

For the Earth, the authalic radius is 6,371.0072 km (3,958.7603 mi).[13]

Another spherical model is defined by the volumetric radius, which is the radius of a sphere of volume equal to the ellipsoid. The IUGG denotes the volumetric radius as R₃.[2]

${\displaystyle R_{3}={\sqrt[{3}]{a^{2}b}}\,.}$

For Earth, the volumetric radius equals 6,371.0008 km (3,958.7564 mi).[13]

Another mean radius is the rectifying radius, giving a sphere with circumference equal to the perimeter of the ellipse described by any polar cross section of the ellipsoid. This requires an elliptic integral to find, given the polar and equatorial radii:

${\displaystyle M_{\mathrm {r} }={\frac {2}{\pi }}\int _{0}^{\frac {\pi }{2}}{\sqrt {{a^{2}}\cos ^{2}\varphi +{b^{2}}\sin ^{2}\varphi }}\,d\varphi \,.}$

The rectifying radius is equivalent to the meridional mean, which is defined as the average value of M:[14]

${\displaystyle M_{\mathrm {r} }={\frac {2}{\pi }}\int _{0}^{\frac {\pi }{2}}\!M(\varphi )\,d\varphi \,.}$

For integration limits of [0,π/2], the integrals for rectifying radius and mean radius evaluate to the same result, which, for Earth, amounts to 6,367.4491 km (3,956.5494 mi).

The meridional mean is well approximated by the semicubic mean of the two axes,

${\displaystyle M_{\mathrm {r} }\approx \left({\frac {a^{\frac {3}{2}}+b^{\frac {3}{2}}}{2}}\right)^{\frac {2}{3}}\,,}$

which differs from the exact result by less than 1 μm (4×10⁻⁵ in); the mean of the two axes,

${\displaystyle M_{\mathrm {r} }\approx {\frac {a+b}{2}}\,,}$

about 6,367.445 km (3,956.547 mi), can also be used.

The mean curvature in all directions at all points on the surface is given by the weighted mean Gaussian curvature:

${\displaystyle R_{4}={\frac {1}{2}}\int _{-{\frac {\pi }{2}}}^{\frac {\pi }{2}}\!\cos \varphi \,R_{\mathrm {a} }(\varphi )\,d\varphi ={\frac {a}{2}}\,{\sqrt {{\frac {1}{e^{2}}}-1}}\,\ln {\frac {1+e}{1-e}}.}$

For the WGS 84 ellipsoid, the mean curvature equals 6,370.994 km (3,958.752 mi).

Most global mean radii are based on the reference ellipsoid, which approximates the geoid. The geoid has no direct relationship with surface topography, however. An alternative calculation averages elevations everywhere, resulting in a mean radius 230 m larger than the IUGG mean radius, the authalic radius, or the volumetric radius. This average is 6,371.230 km (3,958.899 mi) with uncertainty of 10 m (33 ft).[15]