Astrophotography diagnosis of M31: Oversaturation, clipped stars and 1 other
ProcessedANTLIA tri band RGB Ultra 269x300s
The Doc examined this image of M31 (processed, ANTLIA tri band RGB Ultra 2, 69x300s). Estimated overall technical quality: 8/10. 3 defects found: Oversaturation (severity 3/5), Clipped stars (severity 2/5), Background chroma noise (severity 2/5).
&w=1920&q=75)
Info
- Cible
- M31
- Site
- Bortle 7 · périurbain pollué (saisi)
- Position
- 0h42m09s · +41°22'24"
Decent conditions for this target: the Moon was at 0%, so no lunar pollution at all, which is the most limiting factor on a broadband galaxy. Bortle 7 is the real handicap: the sky background sits high and the SNR of M31's faint extensions suffers, which explains the residual chroma mottling in the background despite 5h45 of integration. On a galaxy, a light pollution filter works no miracles since the signal is continuum. The right answer here is time: aim for 12 to 15h to genuinely lift the outer arms and halo above the background noise under this sky.
Setup
- Type d'image
- Traitée
- Télescope
- skywatcher 80ED avec reducteur 0.85x
- Caméra
- ZWO ASI2600MC AIR
- Filtre
- ANTLIA tri band RGB Ultra 2
- Monture
- MLAStro SAL-33
- Exposition
- 69x300s
- Phase de lune
- 0%
- FOV
- 2.45°
The setup matches M31 well. The 80ED with the 0.85x reducer gives 2.45 degrees of field width, framing the whole galaxy along with M32 and M110 and leaving margin around it: the classic, successful framing for this target. Sampling is comfortable, and the measured PSF of 2.6-3.0 px FWHM confirms optics and mount keep up without oversampling. The SAL-33 holds 300s perfectly, with center elongation at 1.04 against a field floor of 1.02. One reservation: the tri-band filter is designed for emission nebulae; on a galaxy it cuts part of the continuum and does not really help SNR. For broadband M31, a plain UV/IR cut would give you more signal.
The diagnosis in detail
On the acquisition side there is nothing to criticize. Moffat measurements give FWHM from 2.57 to 2.97 px depending on the zone, elongation between 1.02 and 1.11 against a field floor of 1.02, a corner/center ratio of 0.976 and FWHM asymmetries of 0.058 horizontally and 0.042 vertically. In other words, the corners are as sharp as the center: no tilt, no backfocus error, no coma despite the reducer, and impeccable tracking on 300s subs. The background map confirms the +55% variation corresponds to M31's real halo rather than an instrumental artifact, which the DSS directly validates.
The real subject is post-processing, and it all hangs together. The stretch was pushed to bring out the arms and dust lanes, which succeeds in the outer regions, but without a highlight protection mask it tipped the bulge into a flat neutral white plateau while the DSS still shows structure and tint at the same place. The same excess drained the star cores of their color. That same push also amplified background color noise, already high under Bortle 7.
None of this is fatal and all of it is recoverable from the linear master. The priority is redoing the stretch with highlight protection.
Priority actions
- Redo the stretch from the linear master with a luminance mask protecting the bulge, or apply HDRMultiscaleTransform to restore the core's structure and tint
- Add a short series of brief subs (10-30s) to rebuild the bulge and bright star cores in an HDR composite
- Apply targeted chrominance denoising on the sky background before any saturation boost
- Extend integration toward 12-15h under this Bortle 7 sky to lift the faint extensions above the noise
- Consider a plain UV/IR cut rather than the tri-band on a broadband target like M31
