‹ Night of 3 October 2026

The Moon at last quarter

The Moon at last quarter

4 October 2026 · 41 percent lit

Along the line between day and night the Sun is setting, so every crater rim and mountain throws a long shadow. That is why the detail is richest there.

What you are looking at

The Moon at last quarter, with its parts named, a scale bar and north marked

Swipe sideways to see the rest. Tap the picture for full size.

Distance
365,000 km, measured from its size in this picture
Across
3,475 km
Light took
1.2 seconds
For scale
Copernicus is 93 km wide; Clavius, 231 km

For photographers

Lucky imaging over a 2 x 2 mosaic in morning twilight, with cloud about.

Exposures
1/60 s at ISO 100
Frames
78 taken over five aims in two passes, 39 kept. 17 saw only the night side; the rest were lost to cloud
Lucky imaging
Frames registered on the lunar surface itself (12,000 matched features, 0.8 px). In each patch only the sharpest 30 percent of the clear frames are averaged
Restoration
Blur measured at the limb: 1.8 arcsec. Wiener and Richardson-Lucy with the gain capped at 3 x and held at the limb. No ringing found at limb, cusps or terminator
Sampling
0.388 arcsec per pixel, the sensor's own: 0.7 km per pixel on the Moon
Field
Disc 32.7 arcmin across, taller than one 38.8 x 25.9 frame: a 2 x 2 mosaic and a centre frame. One panel saw only the night side
When
4 October 2026, in morning twilight, over half an hour. 62 to 70 degrees up
Calibration
Flat from 7 twilight-sky frames: corners at 74 percent of the centre, 340 dust shadows divided out. Cloud glow fitted and subtracted per frame
Orientation
Lunar north up, from 17 craters of known position (32.2 degrees from the sensor's way up)
Telescope
Celestron 8SE Schmidt-Cassegrain with dew shield: 203 mm aperture, 2,084 mm focal length (measured by plate solve), f/10.3
Camera
Sony a6000, stock, at prime focus: APS-C 23.5 x 15.6 mm, 6000 x 4000, 3.9 micron pixels, RAW
Mount
Sky-Watcher EQ6-R, unguided. Set down 27 degrees off the pole; both axes driven from a pointing model fitted to plate solves. Residual drift about 0.1 arcsec per second

© 2026 Brad Gessler · Celestron 8SE (8 inch, 2,080 mm) and Sony a6000 on an EQ6-R · San Francisco Bay Area · 4 October 2026
Capture and mount: Observatory (Elixir on a Raspberry Pi 3). Stacking: numpy, OpenCV, rawpy. Plate solving: astrometry.net. Stacked from the camera's RAW frames. Nothing generated.

© 2026 Brad Gessler. The pictures and the words are his own; all rights reserved.