10
$\begingroup$

There was an earlier question on the diffraction pattern in the Mar 16 test image that was mainly explained by the hexagonal shape of the individual mirror segments. Today's image has similar diffraction patterns on some of the bright stars. However some of the 'redder' objects have more complicated patterns. This picture is cropped from the larger image that was released today.:JWST July 11 2022 image

This is a composite false-color image. It appears that the diffraction pattern on the 'red' image is rotated 30 or 90 degrees compared to the blue image. Is it that the short wavelengths are affected more by the shape of the individual segments while long wavelengths are affected more by overall shape of the mirror? Or are these from two totally different sensors with different optical paths? Can anyone explain what's going on here?

Edit: I don't know if this is a clue, but this is a twitter image from NASA

NASA JWST Twitter

It's a different sensor, but from the varying diffraction patterns, the primary mirror is clearly being rotated with respect to the starfield. It's a large angle but less than 30 degrees.

Improve this question
$\endgroup$
8
  • 1
    $\begingroup$ How about: the two colors correspond to channels taken by different instruments. The instruments are not all aligned but turned with respect to each other. $\endgroup$
    – asdfex
    Jul 12, 2022 at 18:05
  • 1
    $\begingroup$ @asdfex it looks to me like the red and blue spikes have orientations with respect to the stars that differ by 90° or 30° which would suggest that the entire telescope has rotated around its optical axis, assuming that the spikes are due to the hexagonally-segmented primary and struts. (some simple simulations may or may not be helpful) The problem is that it's not easy to imagine (sunshield orientation and all) how that would have happened, unless a lot of time passed between the exposures. $\endgroup$
    – uhoh
    Jul 12, 2022 at 22:05
  • 1
    $\begingroup$ @uhoh Yes, I agree that that would be an answer, but, as you say, it would be difficult without the telescope losing its cool. $\endgroup$
    – Roger Wood
    Jul 12, 2022 at 23:18
  • 1
    $\begingroup$ @RogerWood I'm thinking about the telescope rotating around its own optical axis while its heat shield stays within the allowed range of orientation. But I can't do the 3D stuff in my head. $\endgroup$
    – uhoh
    Jul 13, 2022 at 5:57
  • 1
    $\begingroup$ @uhoh I have trouble with 3D stuff too. The telescope is rigidly attached to the sunshield. It can rotate freely in the plane of the sunshield perpendicular to the Sun. So if it were looking North, then a month later the image would have rotated 30 degrees with respect to the mirror. But it looks like exactly 30 degrees which seems too much of a coincidence. Maybe there's another hexagonal aperture somewhere that we don't know about? $\endgroup$
    – Roger Wood
    Jul 13, 2022 at 15:07

2 Answers 2

Reset to default
8
$\begingroup$

The James Webb Space Telescope website has a very detailed info graphic with explanations that explain the diffraction patterns.

As far as I understand, the large blueish spikes in the image in the question are due to the mirror segment's shapes, while the smaller reddish spikes are due to the struts in the view path.

However, the pattern shown in the info graphic does not match the pattern in the question's image exactly: note that in the info graphic, the red star-pattern partially overlaps the yellow star-pattern, while in the image from the question, the red star-pattern doesn't seem to have this overlap. So it's not a complete explanation for what we're seeing in the actual photo.

Here's the last part of the info graphic:

"Webb's eight-pointed stars" info graphic

Improve this answer
$\endgroup$
3
  • 1
    $\begingroup$ The red pattern does appear to overlap in the photo - note that the four white spikes where the infographic indicates overlap should occur are brighter and redder near the center than the other two. $\endgroup$
    – Skyler
    Jul 12, 2022 at 18:33
  • 2
    $\begingroup$ @DarkDust thanks, I hadn't seen that infographic. Thats good. But there's still no explanation of the rays at +/- 30 degrees from north. I'm thinking that these probably correspond to the overall hexagonal shape of the primary mirror (rather than the individual segments) and that this feature becomes more pronounced as the wavelength gets longer (I wonder what the gap between the segments is? - it must be much bigger than the longest wavelength still) $\endgroup$
    – Roger Wood
    Jul 12, 2022 at 20:32
  • 1
    $\begingroup$ @RogerWood Good point! See comment. $\endgroup$
    – uhoh
    Jul 12, 2022 at 22:10
3
$\begingroup$

The JWST infographic linked by DarkDust explains the NIRCam spikes, but the MIRI images also have additional spikes halfway between the main 6 spikes. You can see this by comparing the point spread functions for the different cameras and wavelengths: NIRCam PSF MIRI PSF.

The 30 degree rotated diffraction pattern is generated by the overall outline hexagon of the mirror, which only affects long wavelengths. For short wavelengths the diffraction pattern is caused by the shape of the individual small hexagons, and for intermediate wavelengths you see a mixture of both. This makes physical sense, unlike my previous explanation, because the diffraction pattern spikes should be perpendicular to the straight edges of the aperture shape, and shorter wavelengths are more sensitive to fine details in the shape of the aperture.

The outline of the mirror, marked in blue

Previous explanation: I believe the cause of the 30 degree rotated spikes is the central aperture of the telescope. Probably that aperture is large enough to not affect short wavelengths, so those images only show the pattern caused by the outline of the mirrors, whereas long wavelength images show a combination of both. Since the central aperture is also a hexagon, but rotated by 30 degrees, it will produce a similar diffraction pattern with a 30 degree rotation.

Improve this answer
$\endgroup$
7
  • 2
    $\begingroup$ thanks for the great links! The MIRI instrument tends to show the 30 degree rotated symmetry which could explain the image. Its still not clear why this is so. The central 'hole' in the primary mirror has the same orientation as the individual segments which cause the six-pointed star. The secondary and tertiary mirrors seem to be circular, so I don't know where the 30 degree rotation comes from. But it seems to be a feature of MIRI - thanks for pointing that out $\endgroup$
    – Roger Wood
    Jul 16, 2022 at 16:59
  • 1
    $\begingroup$ @RogerWood Look at the outline of the mirrors compared to the central aperture, the individual mirrors have the same rotation, but the outline is also approximately a hexagon that is rotated 30 degrees. $\endgroup$
    – Kramin
    Jul 16, 2022 at 23:10
  • 2
    $\begingroup$ the major star-shaped pattern arises from the edges of the individual segments and the gaps between them. The overall hexagonal outline of the primary mirror does not seem to contribute - at least at short wavelengths. $\endgroup$
    – Roger Wood
    Jul 17, 2022 at 1:04
  • 1
    $\begingroup$ Hmm, I might be wrong about the central hole shape mattering, and instead it's the small hexagons of individual mirrors that generate the short wavelength diffraction pattern and the large outline hexagon that generates the long wavelength pattern, with a mixture at intermediate wavelengths. $\endgroup$
    – Kramin
    Jul 17, 2022 at 7:57
  • 2
    $\begingroup$ +1 I didn't get twelve spikes here but I didn't try very hard, maybe it's possible. Also, we don't know know how the image is rotated with respect to the primary mirror so can't a priorI say which spikes come from "the big hexagon" and which from "the little hexagons". I never noticed that 30° rotation, excellent point! $\endgroup$
    – uhoh
    Jul 17, 2022 at 19:52

Your Answer

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

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