LEARNED IT · 09/26
Declination sets the ceiling, rotation sets the hour
"Airmass 8" is not eight atmospheres. It is one atmosphere crossed at a slant. Two separate things decide how much of it you look through: the Moon's declination caps how high it can ever get from your latitude, and Earth's rotation decides where in that arc you catch it. On 30 Aug I got both wrong and threw away a 6x improvement that was available the same night.
Airmass is a path length, not a count. Straight up is the shortest way out of the
atmosphere and is defined as airmass 1.0. Tilt toward the horizon and the same air gets
crossed on a slant, lengthening roughly as 1 / sin(altitude):
| altitude | airmass |
|---|---|
| 90° | 1.00 |
| 60° | 1.15 |
| 30° | 1.99 |
| 10° | 5.58 |
| 7° | 7.73 |
| 5° | 10.33 |
One atmosphere, ours, traversed obliquely. Nothing is stacked or layered. (Below ~10° the
simple 1/sin overestimates — the atmosphere curves with the Earth and refraction bends
the sightline — which is why 7° gives 7.73 rather than 8.21.)
The 30 Aug session sat at 6.4°, airmass 8.4, and paid for it three separate ways: turbulence (the blur), extinction (the R/B ratio of 2.6 — grey rock reddened by eight air paths), and differential refraction, which warped the field by a measured 0.276% between two runs seven minutes apart and made combining them lose 35–45% of contrast. See [[altitude-beats-everything]] for the dispersion side of the same coin.
Two independent controls
1. Declination sets the ceiling. An object's maximum possible altitude is
max altitude = 90° − |your latitude − its declination|
From Goolwa (−35.50°):
| Moon's declination | ceiling | |
|---|---|---|
| 30 Aug | +5.2° | 49° |
| 18 Sept | −27.8° | 82° |
In August the Moon could not have exceeded 49° at any hour. In September it passes nearly overhead, because its declination almost matches the latitude. The Moon runs its full declination range every 27.3 days, so a fortnight swings it end to end. This is the lever, and no time-of-night choice can beat it.
2. Earth's rotation decides where in the arc you catch it. This is the one that hurt, because it was free. Same Moon, same night of 30 Aug:
| local time | altitude | airmass |
|---|---|---|
| 20:42 — when I shot | 6.4° | 8.37 |
| 22:00 | 20.8° | 2.79 |
| 23:30 | 35.6° | 1.71 |
| 02:02 — culmination | 47.8° | 1.35 |
Waiting would have cut the air path 6x with no change of gear, target or phase. The 94% illumination still needed a different date — flat lighting kills crater shadows regardless of altitude — but blur, reddening and refraction were all fixable that same night.
Phase then decides whether culmination lands at a civilised hour: a waning gibbous culminates after midnight, a first-quarter Moon culminates around sunset. That is the whole reason 18–19 Sept works — 82° at 6pm rather than 48° at 2am. See [[moon-is-an-hour-later-every-night]].
The rule
Plan a session on culmination time and declination, not on phase alone. Phase tells you what the lighting looks like; declination tells you whether the target can get high at all; culmination tells you when to be outside. Compute all three before the night, not after.