It is very common that we meet cosmic ray particles in optical images recorded by CCDs. You can see the "snowflakes" in the Hubble Space Telescope images below: snowflake

Generally we should try to remove them in order to get photometry or spectroscopic information.

However, there may be useful information we can derive; for example, can we get their directions or energies statistically before they arrive at a CCD?

Is there some correlation between the ADU value of a cosmic ray point and its real energy?

  • $\begingroup$ Please clarify your question, ask one question at time. The topic is very confused and too broad in my opinion. $\endgroup$
    – Py-ser
    Sep 16, 2014 at 5:11
  • $\begingroup$ Are you talking about background issues? Could you give an example of a CCD hit by a cosmic ray? Usually CCD detectors are "protected" against such a radiation. $\endgroup$
    – Py-ser
    Sep 16, 2014 at 6:40
  • $\begingroup$ @Py-ser If they are, I'd hate to use one that isn't protected. All CCDs that I have ever used suffer from cosmic ray hits, which dependent on the purpose, make long exposures useless. $\endgroup$
    – ProfRob
    Jul 13, 2015 at 13:58
  • $\begingroup$ @RobJeffries, yes, it is true, detectors usually suffer of cosmic rays noise. What I meant is that images are usually cleaned from such events. Instead of "protected" I should have used "veto shielded". $\endgroup$
    – Py-ser
    Jul 13, 2015 at 20:38
  • $\begingroup$ @Py-ser I don't think this technique works for optical astronomical CCDs. The HST picture shown is quite typical for a say 1 hour exposure with an optical CCD. $\endgroup$
    – ProfRob
    Jul 13, 2015 at 21:52

1 Answer 1


Cosmic ray "hits" are artifacts caused when high energy particles (cosmic rays, often muons) slam into atoms in the CCD itself and liberate large numbers of electrons, which then show up as bright spots and streaks when the CCD data is read out. They can also occur as a result of radioactive decay processes in the material in the detector and instrument itself and also the immediate surroundings.

They are a tremendous nuisance that can be removed to a certain extent by median stacking images - though this also fails if you get into a situation where cosmic ray hits are found in the same pixels of $>1$ images in your stack. This limits the lengths of exposures that can safely be made with astronomical CCDs. Depending on exact usage, this limit is usually between 30 minutes and an hour.

The flux of muons and the distribution of angles at which they reach the detector is actually very well known already. It is largely unaffected by local shielding (unless your telescope is buried far underground!)

It is possible to distinguish different types of events (muons, locally produced gamma rays etc.) from looking at the pattern made on the detector (e.g. Groom 2004), but I don't think you can tell much about their energies, unless it could be by looking at the flux as a function of how much shielding (lead?) or how far underground the detector was, since often only a small fraction of the cosmic ray energy ends up deposited in the CCD.


You must log in to answer this question.

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