Physics of the Aurora

Equivalent forms

\mu_m = \frac{m v_\perp^2}{2B} = \text{const}

r_L = \frac{m v_\perp}{qB}

\sin^2\alpha_{mirror} = B_{eq}/B_{max}

The same adiabatic invariant that traps fusion plasma in a magnetic bottle paints the auroral oval — confinement and curtain-of-light are one idea seen at two scales.

Symbol	Name	SI unit	Dimension	Range
$α$	Pitch angle Angle between velocity and the field line (at the equator)	deg	1	5 – 85
$B_ratio$	Mirror ratio B at the pole divided by B at the equator	-	1	1 – 10
$r_L$	Larmor radiusoutput Radius of the particle's gyration	m	L	1 – 1000

Unit systems

Symbol	SI	CGS	Imperial
$α$	deg	deg	deg
$B_ratio$	-	-	-
$r_L$	m	cm	m

Where it holds

The adiabatic-invariant picture holds when the field changes slowly over one gyro-orbit (r_L ≪ scale length, slow time variation). It breaks near field reversals, in strong wave turbulence, and during fast substorm reconfigurations.

Dimensional analysis

$\sin ^{2}\alpha /B$ is the invariant ratio; $r_L = mv$ ⊥/ $(qB) = (kg\cdot m/s)/(C\cdot T) = m \checkmark$

Discovery

Kristian Birkeland · 1908

Birkeland's terrella experiments — a magnetized sphere bombarded by cathode rays in a vacuum — reproduced auroral rings around the model's poles, showing the aurora is charged particles guided by Earth's magnetic field. The trapping physics was formalized later as the magnetic mirror.

Try this

Why do the northern lights glow in rings around the poles — and almost never over the equator?

Charged solar-wind particles spiral along Earth's field lines and reflect from the converging field near the poles. Use the magnetic-mirror invariant to explain where and why they finally crash into the atmosphere.

Research status: stable

Real-world applications

Space-weather and satellite-hazard forecasting
Magnetic-mirror and tokamak plasma confinement
Radiation-belt (Van Allen) modeling for spacecraft shielding
Ionospheric radio-propagation effects

Common misconceptions

The aurora is sunlight reflecting off ice — it's atmospheric gas excited by precipitating particles
Particles travel straight down the field — they spiral and bounce many times before precipitating
Color is random — green is atomic oxygen $at \sim 558\,\mathrm{nm}$ , red oxygen $at \sim 630\,\mathrm{nm}$ , blue/purple nitrogen

Experimental verification

Birkeland's terrella, Van Allen's 1958 satellite discovery of the trapped radiation belts, and modern THEMIS/Cluster spacecraft measuring pitch-angle distributions all confirm mirror trapping and loss-cone precipitation driving the aurora.

Derivation

The magnetic moment µ_

m = mv

⊥

^{2}/2B

is conserved when the field varies slowly.

As B rises, v⊥

^{2}

rises to keep µ_m constant, while total

v^{2} is

fixed (B does no work), so v∥ falls.

The particle reflects where v∥

= 0

, i.e.

\sin ^{2}\alpha = B/B_\max

.

Particles with equatorial pitch angle below

\alpha _lc

with

\sin ^{2}\alpha _lc = B_eq/B_\max

reach the atmosphere — the loss cone.

Limiting cases

\alpha = 90^{\circ}

⟶ reflects at the equatorAll energy is perpendicular; the particle mirrors immediately and stays trapped.

\sin ^{2}\alpha

< B_eq/B_max⟶ falls into atmosphereInside the loss cone the mirror point lies below the atmosphere — the particle precipitates and makes auroral light.

B_ratio

\to 1

⟶ no mirroringA uniform field can't reflect particles — confinement needs converging field lines.

What if…

What if a solar storm strengthens the solar wind?

More particles are injected and scattered into the loss cone, brightening and expanding the auroral oval toward lower latitudes.

What if Earth had no magnetic field (like Mars)?

No mirror trapping or guiding — the solar wind would strip the upper atmosphere and there'd be no organized auroral oval.

What if the mirror ratio were 1 everywhere?

No reflection: particles would stream straight along field lines with no trapping and no bouncing belts.

1

Loss-cone angle for mirror ratio 4

Given ·

B ratio:: 4

Find ·

\alpha _lc

Steps

Loss-cone condition: $\sin ^{2}\alpha _lc = 1/B$ _ratio $= 1/4$
$\sin \alpha _lc = 0.5$
$\alpha _lc = 30^{\circ}$ — particles with equatorial pitch angle below $30^{\circ}$ precipitate and glow

Result ·

\sin ^{2}\alpha _lc = B_eq/B_\max = 1/4

,

so \alpha _lc = \arcsin (0.5) = 30^{\circ}

Cyclotron Frequency→Lorentz Force Law→Magnetic Dipole Moment & Torque→