A glow with a memory.: sources & model

Begin with a real view from the space station. Then prepare two glowing populations, make collisions matter, and discover how atmospheric light, particle energy and a camera’s record fit together.

What supports the explanation?

What this model assumes

1. The curtain geometry, folds, viewing angles and displayed color are authored. They do not reconstruct the ISS event or provide altitude, radiance or a forecast.
2. Both samples start at 100% of their own initial brightness. This normalization compares fading; it does not establish equal absolute red and green intensities.
3. The prepared-state calculation omits ongoing excitation, cascade feeding, ionization, transport and full atmospheric chemistry. It counts total upper-state exits, not a fixed number of photons in one named line.
4. Collisional loss is selected independently. Mapping it to altitude would require density, temperature and state-specific coefficients that this lesson does not fit.
5. The source figure is a published image with axes and color bars, not a raw numerical dataset. The inspection cursor supplies no invented pixel-derived precision.
6. The energy-region geometry is a magnified teaching example. Position is a display coordinate, not measured flight time or a global plasma solution.
7. The historical video is a time-lapse; its frame-to-UTC mapping is not inferred from endpoint times. A clip alone cannot measure the radiative lifetime.
8. Independent space-physics review and learner trials remain pending. Source checking and analytical tests do not constitute that review.
9. Competing rates, with an exact solution: For a prepared population with no further input, dN/dt = −(A + Q)N. A is the total radiative loss rate and Q the collisional loss rate. The remaining fraction is exp[−(A+Q)t]; total radiative and collisional fractions are A/(A+Q) and Q/(A+Q) times the departed fraction. They sum with the remaining fraction to one.
10. The rounded time scales are declared: The selected oxygen examples use A = 1/0.7 s⁻¹ and 1/110 s⁻¹. These rounded upper-state time scales come from the cited RENU2 study. Other published parameter sets yield slightly different values. Q is a selected rate, not calculated from an altitude.
11. Faster fading, fewer radiative exits: For the red-state example with Q = 0.1 s⁻¹, the effective lifetime is about 9.17 s and the eventual radiative share is 1/12, about 8.33%. Without quenching, the rounded lifetime is 110 s. The results are model expectations, not measurements extracted from the footage.
12. A lower state can still be excited: The green oxygen transition goes from ¹S₀ to ¹D₂. The latter is still above the ground ³P₂ level. Our two population comparisons are prepared independently and omit cascade feeding; they cannot predict an absolute red-to-green sky brightness ratio.
13. Air and vacuum wavelength labels differ: The familiar 557.7 nm and 630.0 nm labels are rounded air wavelengths. Converting the cited NIST level separations gives vacuum wavelengths about 557.889 and 630.205 nm and photon energies about 2.22 and 1.97 eV. Those conventions should not be silently interchanged.
14. Total radiative loss is not one particular line: The population budget counts all radiative exits from the selected upper state. A specific line’s rate is its own branching rate times N. Its brightness relative to its own starting value still follows the remaining population in this isolated model.
15. Turning and energizing differ: For an ideal point charge, F = q(E + v × B). The magnetic term is perpendicular to velocity, so its instantaneous work is zero. An electron moving through an energizing increase of 1 kV in electric potential gains 1 keV. This does not mean magnetic-field energy is absent from the coupled space system.
16. Several mechanisms can work together: A 2021 laboratory study measured energy transfer from Alfvén waves to electrons. A 2026 paper interpreted an observed auroral event through wave energy and a quasi-static potential drop. They address particular conditions; they are not mutually exclusive explanations for every aurora.
17. A wavelength band can mix contributions: Original Tromsø spectra showed that rapid variation in a nominal 630 nm band could involve nitrogen-band emission. A broad color channel alone does not prove an isolated oxygen-line response.

What has been checked

Analytical reference cases, conservation or transition invariants, finite drawing commands, bounded setup parsing, discovery and route integrity are checked automatically. These checks do not establish anatomical fidelity, learner outcomes or browser/device compatibility. Independent subject review, learner trials, comprehensive accessibility review and browser video encoding checks remain pending.

Each source supports the associated claim. Sources do not certify this implementation or its visuals.

About the cover illustration

Actual ISS Aurora Australis source poster, photographed September 17, 2011. Credit: Earth Science and Remote Sensing Unit, NASA Johnson Space Center. Complete image resized and converted to WebP without cropping. Educational source media; no NASA endorsement implied.