The Loss Cone of the Van Allen Belts

The problem

Earth's dipole field is a natural magnetic mirror. For a particle trapped on the \(L = 4\) field line, compute the mirror ratio, the loss cone angle, and the fraction of an isotropic population that precipitates into the atmosphere — which is to say, compute the aurora.

Dipole geometry

A dipole field line is labelled by its equatorial crossing distance \(L\) (in Earth radii). Along that line, at magnetic latitude \(\lambda\):

\[B(\lambda) = \frac{B_0}{L^3}\frac{\sqrt{1+3\sin^2\lambda}}{\cos^6\lambda}\]

with \(B_0 \approx 3.1\times10^{-5}\) T the equatorial surface field. The line reaches the Earth's surface where \(r = R_E\), i.e. where

\[\cos^2\lambda_0 = \frac{1}{L}\]

For \(L = 4\): \(\cos\lambda_0 = 1/2\), so \(\lambda_0 = 60°\). The field line that crosses the equator at four Earth radii comes down at 60° latitude — roughly Oslo, Anchorage, or the southern tip of Greenland. The auroral oval's latitude is this equation.

Mirror ratio

Take the ratio of the field at the foot point to the field at the equator (\(\lambda = 0\), where the \(L^3\) and the rest reduce to \(B_0/L^3\)):

\[R_m = \frac{B(\lambda_0)}{B(0)} = \frac{\sqrt{1+3\sin^2 60°}}{\cos^6 60°} = \frac{\sqrt{1+3(0.75)}}{(0.5)^6} = \frac{\sqrt{3.25}}{0.015625} = \frac{1.803}{0.015625} = 115.4\]

Note what dropped out: \(B_0\) and \(L^3\) both cancel. The mirror ratio depends only on the geometry of the dipole, not on the field strength — a satisfying consequence of the field line's self-similar shape.

Loss cone

\[\sin\alpha_{\rm LC} = \frac{1}{\sqrt{R_m}} = \frac{1}{\sqrt{115.4}} = 0.0931 \qquad\Longrightarrow\qquad \alpha_{\rm LC} = 5.34°\]

A very narrow cone. Particles with equatorial pitch angle within 5.3° of the field direction plunge into the atmosphere; everything else bounces between hemispheres.

Fraction lost

The solid angle of two cones of half-angle \(\alpha\), as a fraction of \(4\pi\):

\[\frac{\Omega_{\rm LC}}{4\pi} = 1 - \cos\alpha_{\rm LC} = 1 - \cos 5.34° = 0.0043\]

0.43% of an isotropic population is in the loss cone at any instant. That sounds negligible, and it is exactly the point: the belts are very well confined. Trapped particles bounce between hemispheres in a fraction of a second and survive for months to years.

Why there is an aurora at all

If only 0.43% is lost and the belts persist for months, what feeds the aurora? The loss cone refills. Pitch-angle scattering — from wave–particle interactions with chorus and hiss emissions, and from collisions at the top of the atmosphere — continuously scatters trapped particles into that narrow cone. The precipitation rate is set by the scattering time, not by the size of the cone.

That is why the aurora brightens during geomagnetic storms: storms inject fresh particles and drive intense wave activity, so the scattering rate rises while the geometry stays exactly the same.

Two consequences worth knowing

The South Atlantic Anomaly. Earth's field is offset and non-ideal; over the South Atlantic the surface field is anomalously weak, so the mirror points on those lines sit at lower altitude. Trapped particles penetrate deeper into the atmosphere, and satellites crossing the region take a measurable radiation dose. The ISS logs it every orbit; Hubble stops observing during passes.

Starfish Prime. The 1962 high-altitude nuclear test injected a large population of energetic electrons into these same field lines. Trapped by exactly the mechanism computed above, they persisted for years and disabled at least six satellites — an unintentional and very expensive demonstration that the mirror confinement is real.

Loss cone · Magnetic mirror · Adiabatic invariants · Drift motions · Drift orbit lab — find a loss cone experimentally