I stumbled upon this answer: https://space.stackexchange.com/a/41535/10334

It states the percentage of NEOs we have already discovered.

NEO survey

Seeing this I've asked myself: How do we know what percentage is already discovered? For that one would have to know, how many objects there are. But how can you if you haven't discovered them?

  • 1
    $\begingroup$ I don't know for sure, but I suspect it's based on the number of NEOs we expect versus the number we've actually tracked. How that expectation came to be though, I don't know. If the expectation is based on the number we've tracked so far, I smell a rat. $\endgroup$
    – Mast
    Commented Feb 20, 2020 at 13:20
  • 1
    $\begingroup$ I'm really curious what's going on with the dip around 4km. Either there's something off with our expectation of how many there should be, or that size is really good at hiding for some reason. $\endgroup$ Commented Feb 20, 2020 at 14:30
  • 1
    $\begingroup$ Looking more closely at the rest of the chart, the 4km dip might not be a problem with the model or our detection at that size. The expected total for that size bucket is only a few dozen; so if the percentage found line calculated on number found/number expected it wouldn't take a large absolute variance for it to just mean our solar system is weird at that datapoint. $\endgroup$ Commented Feb 20, 2020 at 20:57
  • $\begingroup$ you seem to forget that we have - thanks to Kepler, Einstein and other Astronomers - quite exact models of our solar system. We can calculate what mass must circulate a certain orbital to get the model match the reality i.e. paths of known objects. Also we have models for the known bodies - down to their mass and hence forth expected gravitational influence on other bodies. We can therefore with a good measure say "there need to be more bodies of x mass to make the model match the known paths". We know the densities of those bodies and their variance so we can predict their size $\endgroup$
    – eagle275
    Commented Feb 21, 2020 at 11:06

1 Answer 1


NEOs are mostly found as dots in images taken by various telescopes, often those of amateurs (as in not paid, nothing about skill or equipment). By taken repeated images days apart moving dots can be picked out against the static background stars and an orbit plotted.

Then that orbit is matched against known objects, and used to either add a new object to the data base or update the existing one.

The number of new finds to number of existing works like mark and release to allow accurate calculation of the total population pool.

  • 2
    $\begingroup$ The mark and release method might work well for short period NEOs, but what about long period near earth comets? $\endgroup$
    – Uwe
    Commented Feb 20, 2020 at 14:09
  • 7
    $\begingroup$ @Uwe very long period and hyperbolic orbit comets are essentially impossible to detect until they begin outgassing and brighten significantly. This generally requires them to be near to or within Jupiter's orbit. Otherwise they're effectively invisible. $\endgroup$ Commented Feb 20, 2020 at 14:38
  • 10
    $\begingroup$ Virtually none of the NEOs are found by amateurs now, most are too faint. The NEO stats by survey shows that e.g. last year only 99 in the 'All others' out of the 2430 NEOs were found by something that wasn't the big surveys (And this number isn't all amateurs, it also includes smaller surveys such as SONEAR or ASAS-SN which aren't broken out into their own column) $\endgroup$ Commented Feb 20, 2020 at 16:12
  • $\begingroup$ Mark and release only tells you what portion of detectable works you've found, given your current technology. For example, if a certain type of fish is transparent to sonar (impressive fish), and you're using sonar for your catch step, you won't see it on first pass, you won't see it on second pass, and you won't see it on third pass, so you're not calculating the "total" population pool, you're estimating the "currently detectable" population pool, which as this question posits, isn't actually that helpful $\endgroup$ Commented Apr 19, 2021 at 5:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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