Astrophotography diagnosis of M16: Light pollution gradient, oversaturation and 3 others
Processed
The Doc examined this image of M16 (processed). Estimated overall technical quality: 6/10. 5 defects found: Light pollution gradient (severity 3/5), Oversaturation (severity 3/5), Sensor tilt (severity 2/5), Clipped stars (severity 2/5), Background chroma noise (severity 2/5).
&w=1920&q=75)
Info
- Cible
- M16
- Position
- 18h18m50s · -13°51'15"
A 25% waxing crescent is only a marginal nuisance, especially on an emission target like M16 shot in narrowband: the Moon does not explain the 209% measured gradient. That background rise oriented at 42 degrees points instead to a ground light source or a low target altitude. M16 culminates low from European latitudes, which worsens extinction and light pollution at the field edge. Aim for meridian transit and point the tube away from the dominant light source, and you will cut the gradient at the source rather than in post.
Setup
- Type d'image
- Traitée
- Phase de lune
- Premier croissant (25 %)
- FOV
- 49.1'
The 0.82 degree field is well sized for M16: the nebula and the Pillars sit in the center with room for the surrounding clouds and no clipping. The monochrome camera and recombined palette are consistent with narrowband acquisition, which explains the good signal retention despite the Moon. The 3.84 px central FWHM is acceptable, but the 5.16 px bottom-right corner shows the optical train is not perfectly flat: that is the mechanical item to fix before adding more exposure time. Checking the tightening and, if needed, a tilt adjuster would be enough to even out the field.
The diagnosis in detail
The acquisition foundation is sound: 1.04 elongation at center against a 1.01 field floor, 33 degrees of PA dispersion, so round stars and no tracking defect. Mount and guiding are not at fault, which simplifies the diagnosis.
The instrumental problem is mechanical rather than tracking related. FWHM goes from 2.62 px in the top-left corner and 2.66 px bottom-left to 5.16 px bottom-right, a corner/center ratio of 1.36 and a horizontal asymmetry of 0.35 with no associated elongation. That is the signature of a slightly tilted sensor plane, not of symmetric field curvature which would hit all four corners equally. Fix it on the optical train tightening side.
The rest is processing. The 209% plane gradient oriented at 42 degrees dominates the radial component fivefold, which rules out vignetting and points to poorly extracted light pollution. On top of that, heavy saturation flattens the blue-violet core, clips a few star cores and brings out chroma noise in the dark areas. Reworking the processing in the order gradient extraction, then stretch, then moderate saturation would give a far more readable image.
Priority actions
- Extract the background gradient with GraXpert or DBE before any nonlinear stretch
- Check the optical train tightening and correct the tilt degrading the bottom-right corner
- Cut global saturation by 15 to 25% and rebalance the channels before the stretch
- Protect bright stars during the stretch to avoid clipped cores
- Apply targeted chroma noise reduction on the sky background
