3
$\begingroup$

I recently read about precovery, where plates or images of an object predating its discovery are used to precisely determine its orbit. It occurred to me that it might be possible that some objects have yet to be identified in open-access astronomical databases, simply as an oversight or maybe as a misidentification (say, mistaking a planet for a star). Is it possible for an amateur astronomer to go through open-access astronomical image data and perhaps use machine learning algorithms to find new celestial bodies? Has it been done? If not, how likely is it that an amateur astronomer could even do so?

$\endgroup$
6
  • $\begingroup$ astronomy.stackexchange.com/questions/40626 might be helpful, I guess. $\endgroup$ – B--rian Dec 28 '20 at 23:24
  • $\begingroup$ Does this answer your question? Why most discovered exoplanets are heavier than Earth? $\endgroup$ – fasterthanlight Dec 28 '20 at 23:38
  • 2
    $\begingroup$ Related: Before the advent of automated supernova searches (~25-30 years ago), practically all of them were discovered by amateurs. Usually only 1 or 2 dozen per year. $\endgroup$ – D. Halsey Dec 28 '20 at 23:44
  • 4
    $\begingroup$ I don't think that is a dupe. It doesn't address the core of the question, though may be useful background reading. $\endgroup$ – James K Dec 29 '20 at 0:21
  • 1
    $\begingroup$ I'll vote to leave-open so that answering is not prematurely blocked. I think fact-based answers are certainly possible, let's give the community a chance. There is no danger of a swarm of opinion-based answers overwhelming this post, so no need to quickly close in order to shut off some firehose of bad answers. $\endgroup$ – uhoh Jan 2 at 1:32
2
$\begingroup$

Back in the "good old days" (the 1980s), a lot of discoveries were made by amateurs. Transient phenomena such as comets or supernovae were spotted by amateurs. This was possible because the bottleneck in making discoveries was the human eye and brain. You needed eyes to look at the sky, or pictures of the sky. Computers couldn't process images well enough to find new transient objects. Only humans could do that. Professionals would be no better than amateurs at looking around the sky trying to spot new stars in galaxies or fuzzy stars that are not known nebulae. So amateurs with eyes, brains, and time had an advantage.

The trouble with "using machine learning" is that lots of people can do that. Are your algorithms really better than the ones that the pros have already used on the data? As computers have gotten better at image processing, the "human bottleneck" has gone. So it has become possible to have automated supernova surveys, automated searchs (using space telescopes) for comets and asteroids. Automated processing of lightcurves to detect exoplanets.

For finding exoplanets, a space telescope is really convenient: you really want several years of data, and it is very convenient if you can get out of the atmosphere as you are looking for very small changes in brightness. The air (and the moon) make measuring this from Earth harder. And you really want a lot of measurements in order to pull the very weak signal from the noise. The Kepler telescope did just that. But the Kepler data is now pretty much processed out.

All this does not mean it is impossible for new amateur discoveries. Amateurs can run supernova discovery surveys. Even so, they still have to compete with the pros now. This makes the entry level very high: you need a lot of cash and a lot of sky, and a lot of luck!

One possible avenue for new discovery is weak meteor showers. Some of the "sporadic" meteors are probably part of very weak streams. Here amateurs have an advantage as meteors are local phenomena and can't be observed by a space telescope. Meteors and meteor showers don't get named after their discoverers (so you don't get your name in the skies. But here is an area at which amateurs can still contribute. See the UK meteor Network for more details

$\endgroup$
5
$\begingroup$

Zooniverse has a citizen science project called Exoplanet Explorers that used volunteers to examine data on exoplanet candidates. It was so successful that it has now run out of data, but there are other space projects that are looking for contributors

$\endgroup$
1
  • 2
    $\begingroup$ For historic purposes, on might also mention the discontinued SETI@home project. Strictly speaking, it is not about exoplanets in general, but "only" for those with inhabitants sending out signals. $\endgroup$ – B--rian Dec 29 '20 at 13:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.