Aurora Oval Geometry & Elevation

Knowing whether aurora is happening is one thing. Knowing whether you can actually see it from your location is another. Lumina models the position of the auroral oval in real time and works out the geometry between you and the display.

What is the auroral oval?

The aurora doesn't happen everywhere at once — it forms a roughly circular band centred on the magnetic pole, called the auroral oval. When conditions are quiet, the oval sits near the poles. As activity picks up, the oval expands equatorward. During a strong storm, the equatorward edge can push well into southern Australia and New Zealand.

On the Space Weather Visualiser you can see this as the glowing band around the Earth — it swells and brightens as hemispheric power rises.

How Lumina determines visibility

Lumina uses two approaches to work out whether aurora is visible from your location. The primary approach samples NOAA's OVATION auroral flux at your exact coordinates. The fallback (used when OVATION data is unavailable) estimates an oval boundary from Hemispheric Power.

Primary: flux at your location

NOAA's OVATION model predicts a continuous map of auroral energy flux across the globe — a smooth gradient, not a sharp boundary. Lumina samples this flux field at your geographic coordinates and converts it into a probability using a calibrated sigmoid:

\[ P_{\text{flux}} = \frac{1}{1 + e^{-k(\text{flux} - t)}} \]

Where flux is the OVATION intensity (0–100) at your location, and k and t are tuned from known southern hemisphere visual records. Below about 15 on the intensity scale, aurora is too faint for visual detection. Around 30–40, it's clearly visible at dark sites. Above 60, it's bright and unmistakable.

This is more reliable than a hard geometric boundary because flux values are continuous — a small calibration error in the OVATION model shifts a probability by a few percent rather than flipping a binary inside/outside determination.

Fallback: Hemispheric Power boundary

When the OVATION flux grid isn't available (e.g. a fetch failure), Lumina falls back to estimating the oval's equatorward edge from Hemispheric Power using an empirical fit calibrated against southern AU/NZ observations:

\[ \Lambda_{\text{eq}} \approx 69.5 - 4.2 \log_{10}(\text{HP}) \]

As a rough guide, here's what different HP levels correspond to:

Even in the primary flux-based path, the boundary is still used for geometric reference — it tells us where the emission region sits so we can calculate the elevation angle and atmospheric extinction (see below). It just doesn't gate visibility.

Elevation angle — where to look

Separately from the flux probability, Lumina calculates the elevation angle — how many degrees above the southern horizon the aurora sits from your location. This determines atmospheric extinction (how much air the light travels through), not whether aurora is happening.

The calculation uses proper spherical Earth geometry and accounts for the emission altitude, which depends on your position relative to the oval:

• If you're inside the oval (poleward of the equatorward boundary), the aurora you see is dominated by green 557.7 nm emission at 110–250 km altitude — the full vertical curtain.
• If you're outside the oval (equatorward of the boundary), only the tops of the arcs are visible: red 630 nm emission at 200–400 km altitude. The green emission is below your horizon. This is why Adelaide typically sees red aurora low on the southern horizon.

This red-vs-green altitude distinction is important for borderline locations — an event that's below the horizon at 110 km (green) may be above the horizon at 250 km (red). Lumina selects the correct altitude automatically.

Earth's curvature matters a lot here — a flat-Earth approximation would be off by 10–20° at the distances involved. That's the difference between "aurora is 10° above your horizon" and "aurora is below your horizon."

The visibility factor

Plan Mode visibility is the product of three independent terms:

\[ \text{visibility} = P_{\text{flux}} \times f_{\text{extinction}} \]

• Flux probability ( P_flux ) — the sigmoid-mapped OVATION intensity at your location (0–1). This is the primary driver.
• Atmospheric extinction ( f_extinction ) — how much atmosphere the light travels through, determined by the elevation angle:
  • ≥ 5° elevation → 1.0. No penalty — the aurora is high enough that atmospheric extinction is minimal.
  • 0° to 5° → ramps from 0.15 to 1.0. Aurora is very low on the horizon — you'll need a dead-flat, unobstructed southern view.
  • Below horizon → 0.02–0.15 floor. The oval is geometrically below your horizon. The factor doesn't go to zero because aurora can sometimes be seen at slightly negative elevation angles (refraction, tall displays), but it's heavily suppressed.

This is why Plan mode sometimes shows a lower number than Field mode even when conditions are good — if the flux at your location is low or the geometry puts the display near the horizon, the visibility factor works against you.

Why elevation matters

When you look at something on the horizon, you're looking through a lot more atmosphere than when you look straight up — about 40 times more airmass at the horizon than at zenith. That extra atmosphere scatters and absorbs light, dimming the aurora. Below about 5° elevation, this effect becomes significant. Below 2°, you're really fighting the atmosphere.

This is also why a clear southern horizon is so important for aurora watching in Australia and NZ — even during a good storm, the display might only be a few degrees above the horizon from your location. A hill, trees, or buildings to your south can block the view entirely.

Why flux, not boundaries

The northern hemisphere calibration problem

NOAA's OVATION model — which provides the flux data and the hemispheric power number — was primarily calibrated against northern hemisphere satellite and ground data. The southern hemisphere has different ionospheric conductivity (especially during our winter aurora season), different magnetic field geometry, and different dawn-dusk responses to the solar wind's magnetic field orientation.

A hard geometric boundary amplifies these calibration differences. If OVATION's boundary is off by half a degree (a tiny error in a global model), it can flip a borderline location from "inside the oval" to "outside" — creating a false negative. The boundary is a cliff edge; calibration bias manifests as a cliff you can fall off. Flux values are a slope — a small systematic offset shifts a probability by a few percent rather than flipping a binary determination.

IMF B_y oval asymmetry

The solar wind's east-west magnetic field component (IMF B_y) tilts the auroral oval — pushing it equatorward on one side and poleward on the other. In the southern hemisphere, positive B_y shifts the oval equatorward on the dusk side; negative B_y shifts it equatorward on the dawn side. Lumina applies this correction to the boundary used for geometric elevation, so the extinction factor reflects the actual oval position rather than a symmetric average.

Why this matters for southern hemisphere viewers

1. No cliff edges — borderline locations near the oval edge get smooth probability transitions rather than binary inside/outside.
2. Calibration robustness — flux values are more robust to northern-hemisphere model bias than boundary positions.
3. Physical corrections — BY asymmetry and wavelength-dependent emission altitude are real physical effects that matter at southern latitudes.

The legacy boundary-based approach is retained as a fallback when OVATION flux data is unavailable.