This answer to How big will Apophis appear? points out that the near Earth asteroid Apophis will likely be close to 2 arcseconds in diameter as seen from Earth during its close approach in 2029. I speculate that if the Hubble Space Telescope were still operational then, it could potentially image the asteroid in visible light at a few dozen pixels in diameter.

This leads me to wonder if the Hubble has ever been used to image or at least spatially resolve in some way an asteroid during a close pass to the Earth before.

Here the verb "image" should be taken to mean the act of producing a resolved image of an object so that different pixels correspond to intensity from different parts of the body being imaged. For the purposes of this question please don't consider telescope images in which an asteroid happens to appear but is too far away to be resolved. Thanks!


1 Answer 1


[rewritten to address the revised question]

Maybe, depending on how fussy you want to be about "resolved".

This is a study from 1995, using observations of asteroid 4179 Toutatis made in 1992 with HST. They reported marginal resolution of the asteroid, as suggested by this figure comparing a deconvolved image of a star (observed with the same filter and imager location) and a similarly deconvolved image of the asteroid itself (each pixel corresponds to about 450 m at the distance of the asteroid):

Figure 4 from Noll et al. 1995

The appearance of the asteroid is pretty clearly not a point source, but it's also fair to say it's only partly resolved, and mostly just in one direction.

(My admittedly vague impression is that this is one of the best, if not the best, case of HST "resolving" a near-Earth asteroid.)

Most observations of near-Earth objects with HST are, I think, aimed at getting optical information on compositions not possible from other wavelengths, and sometimes refining estimates of rotation rates, as was done this year (using data from 2012) for the asteroid Bennu, currently being visited by OSIRIS-REx.

In practice, you get much better spatial resolution using radar (including line-of-sight distance variations due to the structure of the asteroids from time-of-return measurements, allowing you to construct 3D models of them), so there's not much point in trying to resolve them with HST.

  • $\begingroup$ In order to support your "Yes" I think it would be good to find at least once instance of something that is spatially resolved, even if it turns out to only be a few pixels. The first reference is paywalled so I won't be able to check it this week; the second does not seem to have anything resolved (no images are shown in the paper). $\endgroup$
    – uhoh
    Commented Jun 25, 2019 at 9:55
  • $\begingroup$ The first reference does say in its abstract: "Toutatis was marginally resolved by the HST. There is evidence for an extended object with a projected illuminated dimension less than 2.8 km while the best fit symmetric model has a maximum illuminated size of 2.0 km and deconvolutions suggest an irregular object with a maximum dimension of 1.7 km" $\endgroup$ Commented Jun 25, 2019 at 10:07
  • $\begingroup$ (Somewhat pedantically, I would argue that if you observing something with one of HST's cameras -- as opposed to one of its spectrographs -- you are "imaging" it...) $\endgroup$ Commented Jun 25, 2019 at 10:09
  • $\begingroup$ The title of the 1995 paper is "Imaging of Asteroid 4179 Toutatis with the Hubble Space Telescope".... $\endgroup$ Commented Jun 25, 2019 at 10:15
  • $\begingroup$ Being published in Icarus means that the paper has likely passed substantial peer-review scrutiny, but language like "There is evidence for an extended object..." sounds somewhat equivocated and I find it only mildly convincing. I think it will be better to have a little more information about the results so we can be convinced ourselves though. I'd like to be able to balance the evidence for with the evidence against, and the abstract alone doesn't provide an opportunity to do that. $\endgroup$
    – uhoh
    Commented Jun 25, 2019 at 10:17

You must log in to answer this question.

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