I looked at a lot of high-quality images of solar eclipses and noticed the following thing: In all the photographs I've seen, the lunar disk has a completely clear outline, in which it is impossible to see any irregularities due to the presence of the lunar relief. Although one should expect them to be visible?

enter image description here

Source: Times of India - IndiaTimes

  • 11
    $\begingroup$ en.wikipedia.org/wiki/Baily%27s_beads $\endgroup$
    – John Doty
    Commented Mar 6 at 17:28
  • 8
    $\begingroup$ What is the expected height (in pixels) of the mountains you expect to see? $\endgroup$
    – JohnHoltz
    Commented Mar 6 at 18:06
  • 3
    $\begingroup$ @JohnHoltz: doing the maths, the biggest mountain on the Moon would be about 1 pixel high in this photo. If it was cube sticking out of the surface, then it should be just about visible. Of course, mountains have slopes and the light from the sun is likely to max out the sensor even if partially occluded by lunar terrain, so we shouldn't expect to see it here. But for a significantly higher resolution camera, it can't be completely ruled out based on scale alone. $\endgroup$ Commented Mar 6 at 18:18
  • 2
    $\begingroup$ @JohnHoltz, I understand what you mean. Give a look to here : cs4.pikabu.ru/post_img/big/2015/11/23/6/… $\endgroup$ Commented Mar 6 at 18:27
  • 4
    $\begingroup$ Where is that image from? It looks like it could be CG. $\endgroup$
    – Matthias
    Commented Mar 7 at 16:16

2 Answers 2


In short, you would need a very high resolution photo -- most likely taken through a telescope -- to identify any surface features on the moon by their effect on the limb (that is, the edge of the moon against the sun).

In the photo you posted, the moon is about 500 pixels wide. The moon's diameter is around 1,738 km, so in this photo, each pixel represents a bit less than 3.5 km. The tallest mountain on the moon is Mons Huygens, which has about 4.7 km of topographic prominence (that is, its height above its surroundings). So the biggest mountain on the moon is about 1.3 pixels tall -- and most of the moon is much flatter than that. Surface relief features of the moon just aren't big enough to be identifiable at this scale.

You probably could see some of the larger features of the moon's limb if you took a photo of the moon with enough magnification that it completely fills the frame of a really good 20 megapixel camera, where you'd be looking at something like 350 meters per pixel.

However, there is a phenomenon that shows the surface features of the moon even if we can't see them directly. In the few seconds as the edge of the moon exactly overlaps the very edge of the sun, we can see tiny specks of light with dark between them. This is known as "Baily's Beads":

Baily's Beads from the solar eclipse of August 21, 2017

The dark spots are higher points that block the view of the sun, while the lights are low spots or craters that allow the sun to peek through. They are named for astronomer Francis Baily, vice president of the British Royal Astronomical Society, who in 1836 explained what causes the effect.

  • 1
    $\begingroup$ Observing lunar occultations and also looking at the lunar limb at 50 times magnification, it is very apparent the lunar limb isn't smooth (both the illuminated and dark part (still illuminated by earth shine, unless close to full moon)). This is more representative of what it looks like. $\endgroup$ Commented Mar 8 at 2:15
  • 1
    $\begingroup$ Or this (though JPEG always distorts). $\endgroup$ Commented Mar 8 at 2:21
  • $\begingroup$ Those are some really cool photos! $\endgroup$ Commented Mar 9 at 22:17

The picture in the OP is most likely a drawing. You can tell by a cartoon-like look of the corona as well as a similarly sterile solar flare on the right.

If you search for actual high-resolution photos of a solar eclipse, you'll find e.g. ref. 1 or ref. 2, where the lunar relief, while being very subtle (due to the mountains being small relative to lunar diameter), is still noticeable.

Also note that in some cases atmospheric distortion (particularly near the horizon) will give you a false relief, as can be seen e.g. in ref. 3.

To illustrate visibility of relief, below I used the image form ref. 1 and a simulation of this eclipse in Stellarium, with a custom 3D model of the Moon. The location and time are not exactly the same, since the source web page doesn't specify them. The two white lines mark the corresponding parts of the lunar rim where you can see similar structures in the bumps.

Stellarium + 3D Moon with relief:

Rotated and marked image from ref. 1:


  1. "Annular Solar Eclipse at High Resolution", Stefan Seip, 2005-10-05.
  2. "High resolution image of inner corona", Miloslav Druckmüller, 2008-08-01.
  3. "A Partial Solar Eclipse", NASA/Bill Ingalls, 2021-06-10.
  • 3
    $\begingroup$ the picture in the OP is pretty close to the huge number of photos produced by amateurs of total eclipses using ordinary cameras, rather than using the specialist equipment needed for 1 or 3 of your references $\endgroup$
    – Tristan
    Commented Mar 7 at 9:48
  • $\begingroup$ The random photo I posted is not the only one, it is one of many. I simply didn’t and couldn’t publish a hundred photos at once. It is impossible to exclude that this photo was retouched, like any other. although others differ little in appearance. There are some minor defects in the images from your links, but it is impossible to say certainly that they are a consequence of the relief, they are more similar to image defects. $\endgroup$ Commented Mar 7 at 10:15
  • 1
    $\begingroup$ @Tristan please give a couple examples that are really confirmed to be photos. If you've just looked at Google Image Search, then you must have gotten lots of drawings, not photos. $\endgroup$
    – Ruslan
    Commented Mar 7 at 11:44
  • 2
    $\begingroup$ @Ruslan I'm thinking primarily of one of the ones my dad took of the 2017 eclipse in the US using his DSLR. The one in the OP is definitely a bit more "perfect" than it in a way that might suggest it's been drawn or altered, but that photo of my dad (and the thousands of other such photos produced by amateur photographers with their ordinary cameras)'s is far closer to the OP's than either your examples 1 or 2 $\endgroup$
    – Tristan
    Commented Mar 7 at 14:12
  • 1
    $\begingroup$ the main reason is that the sort of camera accessible to an ordinary amateur photographer doesn't have anything close to the dynamic range necessary for those sorts of photos $\endgroup$
    – Tristan
    Commented Mar 7 at 14:14

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .