Observers classify a zoo of distinct aurora forms, several discovered and named by amateur photographers within the last decade. Physics background lives in the field guide.
Each entry below is tagged with what this observatory's renderer does with it:
●rendered ◐partly rendered ○not yet — with the reason
The baseline form: a smooth, thin ribbon stretching east–west along the oval, often sitting quietly for hours. During a substorm's growth phase, arcs sharpen, multiply and drift equatorward before breaking up1. In the model: the base state — the fitted auroral boundary rendered as a discrete ridge.
An arc that has developed folds and kinks, snaking across the sky — the workhorse form of active displays, with folds traveling along it at kilometers per second. In the model: rendered — the boundary carries traveling waviness and folding, tuned against calibrated all-sky research cameras.
Vertical shafts of light tracing individual magnetic field lines, tens to a few hundred kilometers tall, drifting along bands. In the model: the pillar system — every ray gets its own height, width and color transition, field-aligned like the real thing.
Broad, structureless glow washing over large areas of sky behind and between the discrete forms — often the majority of the total light in a display. In the model: a dedicated diffuse layer rendered separately from the discrete ridges, on top of the natural airglow floor.
Not a separate aurora but a viewing geometry: when a rayed band crosses your zenith, perspective makes the parallel rays appear to radiate from a single overhead point2. Many chasers' single most intense memory. In the model: emerges automatically in ground view when a rayed band passes overhead — parallel field-aligned pillars seen end-on.
A pure-red (630.0 nm only), hours-stable band at ~400 km sitting equatorward of the main oval: heat from the ring current conducted down where it overlaps the plasmasphere3. Usually too faint for the eye — and the lowest-latitude "red glow" reports during superstorms are often SAR arcs rather than classic aurora4. In the model: rendered in the 2024 reconstructions — smooth 630 nm glow at ~400 km drawn over the longitude sectors with SAR-classified sighting reports, night side only — and now in the live sky too: a model band placed at the plasmapause computed from live Kp5, with intensity ramped on the real-time ring current and labeled "(model)" in the app.
Strong Thermal Emission Velocity Enhancement: a narrow mauve ribbon far equatorward of the oval, produced not by particle precipitation but by a ~5 km/s channel of westward-drifting ions heating the atmosphere — identified and named with citizen scientists in 20186. Strictly speaking, not aurora at all. In the model: rendered in the May 2024 reconstruction — the mauve arc at ~450 km, anchored to the SAR-classified sighting sectors and gated by the measured ring current. Not in the live sky: no public real-time feed resolves the ion-drift channels, and simulating one without a measured driver would mean simulating what nothing measures.
The row of green vertical stripes that appears below STEVE's mauve band. Recent analysis argues the green comes from electric fields accelerating electrons locally, in the atmosphere, rather than particles raining down from the magnetosphere7. In the model: travels with STEVE — green strokes at 95–150 km beneath the May 2024 arc; absent from the live sky for the same reason.
Even, parallel green stripes rippling through the diffuse glow near 100 km — discovered by Finnish citizen scientists and interpreted as a mesospheric bore (an atmospheric wave) modulating the oxygen emission8. In the model: feasible as a periodic modulation of the veil layer, but no feed tells us when real dunes are present — so it stays out.
Patches of glow blinking on and off every ~2–20 seconds across the morning-side sky after substorm expansions — chorus waves out in the magnetosphere scattering electrons loose9. Easy to miss by eye, unmistakable on camera. In the model: rendered — the model detects each substorm from the AE record, and for the following ~two hours hash-phased patches near dawn blink on and off on second scales, modulating the curtain layer.
Named by the Finnish amateur observers who kept photographing it: during substorms, a green diffuse band and a red arc appear together at subauroral latitudes and drift equatorward; the green fades, the red persists — maturing into a SAR arc. The follow-up science, with the discoverers as co-authors, showed the pair is proton aurora, and that proton precipitation is what initiates the SAR arc10. In the model: both ingredients render in the measured storms — the proton-aurora band and the SAR arc — but the coupled hand-off between them isn't modeled.
Small (a few kilometers), short-lived greenish blobs seen near aurora over Svalbard that don't appear to come from particle precipitation at all — hence "aurora-like"11. In the model: below the grid scale and outside the physics — out of scope.
Precipitating protons from the ring current charge-exchange their way down, painting broad, diffuse, mostly subvisual glow with a tell-tale Doppler-shifted hydrogen line12. What little the naked eye catches reads as structureless green — lit by the secondary electrons the protons knock loose10. In the model: rendered in the measured storms — a distinct diffuse band built from the POES satellites' measured proton flux, equatorward of the electron oval, faint and structureless as in nature. Not in the live sky yet.
Numbered references for the claims above, DOI-verified. Entries marked in the renderer are also implemented — their formulas or thresholds run as code in this observatory's engine (the app's data & methods panel lists the full implementation set).
By The Space Weather Observatory · checked against the data & methods record · Updated August 21, 2026
the sky · the sun · field guide · data & methods · about · privacy · pro · ☁️ seasonmap · ☕ support