A comment under this answer to ** links to Hubblesite.org's Hubble Shoots the Moon The image, its description and credits are shown below.

As explained, the purpose of the observation was to record a detailed spectrum of attenuated sunlight via its diffuse reflection from the Moon. I seem to remember a more recent effort to record the spectrum of the reddish brown light reflected from the Moon during a lunar eclipse where sunlight passes through Earth's atmosphere as a way of simulating exoplanet atmospheric analysis during transits of their stars, but I am not sure if this is a related precursor or not.

Anyway. What I find interesting is

The image was taken while the Space Telescope Imaging Spectrograph (STIS) was aimed at a different part of the moon to measure the colors of sunlight reflected off the Moon.

I remember seeing a map Hubble's primary focal plane showing that most of it is empty, wasted space, and each of the instruments -- including the cameras! -- "pick off" a small bit for itself. That means that light from different parts of the sky can simultaneously enter different instruments; we can project the map of the focal plane back to the celestial sphere. With a focal length of 57.6 m that means every centimeter at the focal plane is about 36 arcseconds.

For more about that see answer(s) to How many science instruments can be used in parallel with the Hubble Space Telescope?

Okay so what's my question again?

According to this description, while the Space Telescope Imaging Spectrograph (STIS) was collecting data from one part of the Moon, the Wide Field Planetary Camera 2 was also snapping pics of the lunar surface.

Question: How often and for what main reasons does Hubble use two different instruments at the same time?

About This Image

In a change of venue from peering at the distant universe, NASA's Hubble Space Telescope has taken a look at Earth's closest neighbor in space, the Moon. Hubble was aimed at one of the Moon's most dramatic and photogenic targets, the 58 mile-wide (93 km) impact crater Copernicus.

The image was taken while the Space Telescope Imaging Spectrograph (STIS) was aimed at a different part of the moon to measure the colors of sunlight reflected off the Moon. Hubble cannot look at the Sun directly and so must use reflected light to make measurements of the Sun's spectrum. Once calibrated by measuring the Sun's spectrum, the STIS can be used to study how the planets both absorb and reflect sunlight.

(upper left) The Moon is so close to Earth that Hubble would need to take a mosaic of 130 pictures to cover the entire disk. This ground-based picture from Lick Observatory shows the area covered in Hubble's photomosaic with the Wide Field Planetary Camera 2.

(center) Hubble's crisp bird's-eye view clearly shows the ray pattern of bright dust ejected out of the crater over one billion years ago, when an asteroid larger than a mile across slammed into the Moon. Hubble can resolve features as small as 600 feet across in the terraced walls of the crater, and the hummock-like blanket of material blasted out by the meteor impact.

(lower right) A close-up view of Copernicus' terraced walls. Hubble can resolve features as small as 280 feet across.

CREDITS: John Caldwell (York University, Ontario), Alex Storrs (STScI), and NASA

enter image description here

  • 1
    $\begingroup$ with great self-restraint I was able to not add a new [multi-instrument] tag :-) $\endgroup$
    – uhoh
    Sep 13, 2022 at 23:39
  • 2
    $\begingroup$ Are you more interested specifically with relation to Hubble, or also any telescope that uses multiple instruments at the same time? $\endgroup$
    – Justin T
    Sep 14, 2022 at 0:48
  • 2
    $\begingroup$ @JustinT that is such an interesting topic! Hmm... I'm interested in everything, the question is what's the best way to proceed in order to generate the most good answers for the benefit of future readers. I could change this to "which telescopes are capable of..." but I think that will be better as a new, follow-up question. Since Hubble is so special, being pretty much the first major visible light space telescope and has such a long and complicated instrumentation history itself, I think it deserves its own question. I'll ask the second, broader question shortly and then ping you back here. $\endgroup$
    – uhoh
    Sep 14, 2022 at 0:56
  • 1
    $\begingroup$ Does this answer your question? How many science instruments can be used in parallel with the Hubble Space Telescope? $\endgroup$
    – DialFrost
    Sep 14, 2022 at 9:39
  • 1
    $\begingroup$ @DialFrost obviously not the main part of the question; I asked "How often and for what reasons...?" while that question asks "How many..instruments...?" But it does cover the last bit, so I've deleted that and replaced it with a link that one. Thanks! $\endgroup$
    – uhoh
    Sep 15, 2022 at 1:23

1 Answer 1


I can't answer the first one but I can give a plan for figuring it out:

How often does Hubble use two different instruments at the same time?

Unfortunately the Hubble search tool does not have a way to search for "parallel" exposures. So to check which exposures are parallel and which are regular single-instrument exposures, you would have to check the metadata of every frame. There is a keyword "INSTRUME" that tells you the instrument recording a given exposure, and a keyword "PRIMESI" that tells you the prime science instrument. Any time the value of INSTRUME is different from the value of PRIMESI, you're looking at a parallel exposure. When the keywords match it's the prime exposure. So the ratio of parallel exposures to prime exposures gives you roughly the fraction of time two instruments are working. I would guess the number is like 5% but I could be way off. Staff working at Hubble are probably already tracking this for their operational statistics.

Trick answer: The Fine Guidance Sensor (FGS) is used to control telescope pointing, but it can also be used as a science instrument. So actually most of the time the telescope is using two instruments (FGS + the prime science instrument), and your question becomes how often does Hubble use three instruments at the same time. However, the science mode of FGS is different from the pointing control mode.

For what main reasons does Hubble use two different instruments at the same time?

One main reason is to save time in surveys that cover a large area compared to the area of an individual instrument footprint. This type of parallel observation is what they call coordinated parallel, and it's the same observing team controlling the prime and the parallel instruments. Here is a map of fields in the CANDELS survey: enter image description here

The offset between the blue area (WFC3 instrument data) and pink area (ACS instrument data) is directly because of the parallel exposures. You can match the size of the offset and the angle between the tiny tiles to the offsets between WFC3 and ACS in the focal plane map you mentioned: enter image description here

The other type of parallel observation is pure parallel. Pure parallel observations basically say "take data with a second instrument whenever some conditions are met." They are "free" in a way because they don't require extra pointing to a target, and they let two separate teams take data at the same time. The parallel data are taken while someone else is controlling the pointing and using a different prime science instrument. Pure parallel observations are good for when you want data in some part of the sky, but you don't care precisely where you point. One example is searches for new objects in the Kuiper belt: you could just request parallel exposures be taken whenever Hubble is pointing near the ecliptic plane.

  • 1
    $\begingroup$ Oh this is really fantastic; thanks! $\endgroup$
    – uhoh
    Sep 15, 2022 at 12:38
  • 2
    $\begingroup$ There is a paper by Civeit et al from 2009 which talks about scheduling parallel observing after Servicing Mission 4 in 2008. They quote that 6 proposals were submitted and 3 accepted for parallel observing. Reid 2014 lists 958 submitted proposals and 228 accepted proposals for cycle 17 so as percentage of proposals (not no. of orbits/time), parallel programs are 0.6% of submitted and 1.3% of accepted proposals $\endgroup$ Sep 15, 2022 at 16:16
  • $\begingroup$ @astrosnapper That's helpful. My larger guess of 5% includes the fact that these surveys take a lot of time, and getting something useful out of pure parallel data also takes many orbits. So I suspect the programs that use parallel observations are on the large size. $\endgroup$
    – giardia
    Sep 17, 2022 at 1:23
  • $\begingroup$ @giardia I think this is most likely and the percentage by no. of orbits is likely to be much higher than the number of proposals. A survey, wanting a large number of orbits, is the most logical use of parallel mode since you can't influence the primary observations length, pointing or roll requirements in anyway. $\endgroup$ Sep 17, 2022 at 22:06

You must log in to answer this question.

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