JWST will look for an infrared signal as low as 1 photon per second that comes from the earliest sources of infrared. But there will be infrared from any star in its field of view that is not necessarily the signal that JWST is looking for.

How will JWST extract the desired signal from all the other infrared it will receive from other sources in its field of view?


I want to clarify the assumptions underlying my question.

My assumption is that JWST is looking for a series of spectral lines that are red-shifted into the IR. If the stars of interest have a continuum of velocities (ages/distances) the signal will be smeared. There will also be noise from other sources.

I'm trying to understand the signal extraction problem, but I'm making assumptions about what the signal and noise are. The signal could be as low as one photon per second, but what is the signal? And what will the noise look like?

Does anyone have a reference to what signal the JWST is looking for?

  • 2
    $\begingroup$ Are you talking about the spectrometry or imaging? With imaging it's fairly trivial to distinguish sources, given they are optically resolved... $\endgroup$ Dec 22, 2021 at 17:13
  • 2
    $\begingroup$ Great question! When we think of infrared, especially long wavelength "thermal IR" we sometimes think that it's heat rather than light. But it really is light and can be imaged just like visible light can. This telescope works just like visible light telescopes, it images whatever is out there on a sensor's pixels. So the challenge is no different than asking how a visible light telescope separates visible signals from the desired sources from all the other visible light it will see. $\endgroup$
    – uhoh
    Dec 22, 2021 at 18:21
  • $\begingroup$ Your eyes are bombarded by light from many different sources, yet you have little trouble picking out specific ones. Telescopes can use similar techniques. $\endgroup$
    – Barmar
    Dec 23, 2021 at 16:01

1 Answer 1


Especially for the MIRI (Mid-InfraRed Imaging) JWST has three main image processing tools for this: filtering, dithering and sub-arraying.

Filters allow scientist to select a sub-band of the Mid-Infrared band. If a target galaxy/star is farther (=older), its light will be affected by more redshift (=the wavelength will be longer). Selecting the right filter will "suppress" the lights from closer stars and further stars.

enter image description here


sub-arraying is like filtering but in the spatial domain. It allows removing nearby bright areas in order to not saturate the image sensor.

dithering is a well-known technique in image post-processing, but it's purpose is mainly to reduce artifacts due to (slight) movements.

Another very clever post-processing available is coronapraphing. This technique will be used, for example, to study an exoplanet close to its bright star. To probe further see this source

  • 1
    $\begingroup$ I've edited my question to be more specific. If JWST is looking for series of spectral lines that have been red-shifted into the IR by expansion there will be a continuum of spectral lines due to a continuum of distances/ages. The same set of lines will cover a band of the IR. How will the weak signal from those lines be extracted from a bath of IR noise? $\endgroup$ Dec 23, 2021 at 3:54
  • $\begingroup$ @Dean Schulze, each filter selects a continuous band of IR (its "passband") and suppresses all other. As an example, if you select F2550W, you will see the very old light sources while excluding the younger ones (those that shines at wavelengths <22 micrometer about). Indeed, by definition, you don't know in advance at which IR the sources are shining, but you can select a range of time (ages) to hunt for them. $\endgroup$
    – Ng Ph
    Dec 23, 2021 at 10:50
  • 1
    $\begingroup$ @DeanSchulze every site is different, but for most SE sites outside of the software/programing world we try not to update/modify/refine the question too much after an answer has been posted, even with the best of intentions, unless there's some buy-in from the authors of the existing answer posts. Your question about spectroscopy is really a new and different question than what you'd originally posted, and actually I think it would receive better answers in Astronomy SE than here, though it's certainly on-topic here. $\endgroup$
    – uhoh
    Dec 23, 2021 at 22:15
  • $\begingroup$ @DeanSchulze perhaps consider rolling back to your original version here and asking a new, spectroscopy question in Astronomy? In each question you can link to the other for better continuity. $\endgroup$
    – uhoh
    Dec 23, 2021 at 22:16

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.