Sunspotter Talk

CR Hit?

  • Mjtbarrett by Mjtbarrett

    Sub-titled "Newbie Attempts To Justify Wild Theorising From A Position Of No Knowledge To Chief Scientist" ! (insert embarrassed smilie here )

    Voyager1682002 spotted an artifact in Image ASZ00007ms http://talk.sunspotter.org/#/subjects/ASZ00007ms @pahiggins suggested a Cosmic Ray hit on the CCD module as being the likely cause of the artifact, but invited suggestions of other theories to account for the cause of the anomaly.

    When I looked at the image I (in my innocence) thought that the non-linear appearance of the anomaly (in fact curved appearance) was more suggestive of a software or data processing problem. I had thought that a CR hit on the CCD module would produce a more "blocky" appearance, or at least a linear track (assuming that cosmic rays travel in straight lines).

    The appearance of the anomaly changes slightly in different images. In the image presented, the curve is noticeable; on the full disk image it looks linear when you zoom in manually: http://www.solarmonitor.org/full_disk.php?date=20010416&type=smdi_maglc&indexnum=1

    but on the close up image of region 09418 the artifact looks to be linear but thicker at the north-east corner: http://www.solarmonitor.org/region_pop.php?date=20010416&type=smdi_maglc&region=09418 Interestingly there's another linear feature two sectors to the left that looks straight in both the solarmonitor images...

    I had a look for examples of cr hits and found this: http://sungrazer.nrl.navy.mil/index.php?p=guide which shows some of the different types. The images are from LASCO but the artifact in question does look very similar (albeit without the faded "ghost trails). I am happy to concede that this anomaly is a CR hit that didn't look quite as I would've expected.

    Whilst looking around, I found the following explanation on the MDI known calibration problems (but admit to being out of my depth 😃)
    http://soi.stanford.edu/data/cal/mag_data.html#nonlinearity

    There are, however, some other known instrument effects relating to CCD modules used in astronomy such as Charge Transfer Efficiency effect, Pixel Centroid Shift or Cross-talk that may also account for the non-linearity of the appearance of the artifact in AGZ00007ms. (See: http://talk.planethunters.org/discussions/DPH101z5mp?object_id=APHE20031ex&page=2&per_page=10 for an interesting discussion).

    I am a newbie, not a scientist, and am happily taking part in the classifications based on the premise that the Mk I human eyeball is more efficient than a computer algorithm at carrying out judgements quickly; if we can help to train the poor computers to perform better that's great. I am only using intuition and the appearance of the images and this one piqued my interest because it didn't "look right."

    This is a really interesting project and I am learning all manner of new things. I am delighted to have got this wrong, because I have had to look up all sorts of stuff in order to think it through. Thanks very much for your patience @pahiggins. I wasn't being arrogant, just curious, and when you asked for alternative theories you got me wondering about the cause.

    Apologies if I've wasted your time.

    Regards,
    Mike.

    Posted

  • Mjtbarrett by Mjtbarrett

    There's a really good section on Cosmic Rays here if anyone else is interested: http://helios.gsfc.nasa.gov/cosmic.html

    Posted

  • pahiggins by pahiggins

    Wow, great post! Thats a really good observation- while it is likely that the artifact is fundamentally do to a CR hit, the CR could have caused some of these other second order CCD effects that you mention. Perhaps the funny shape is in fact due to pixel bleed, or something else. I confess I am not an expert on CCDs what so ever. Great stuff!

    Posted

  • voyager1682002 by voyager1682002

    @Mjtbarrett thanks for all the links 😃 They are really helpful.

    Posted

  • voyager1682002 by voyager1682002

    Here is an interesting news report on artifact/energetic particle (proton) impacts on the CCD : http://www.foxnews.com/scitech/2012/04/26/ufo-spaceship-orbiting-sun-or-camera-glitch

    Posted

  • Mjtbarrett by Mjtbarrett in response to voyager1682002's comment.

    Thanks V, Ours is definitely a UFO then; probably from the base on Mars!!! 😃

    Posted

  • Mjtbarrett by Mjtbarrett

    I may have found the pilot of our UFO! Looked like an Aladdin film character at first, but now....
    http://talk.sunspotter.org/#/subjects/ASZ0000529
    Zoom right in on the grey area to the left of the dark spot in the linked image. See the face of a character from an Aladdin film? Or is it just me??

    Posted

  • voyager1682002 by voyager1682002

    You are really testing my eyesight, Mike 😃 Oh! bty this may be another UFO that has landed : http://talk.sunspotter.org/#/subjects/ASZ000018b

    Posted

  • Mjtbarrett by Mjtbarrett

    😃

    Just a quick question - given that the other wavelengths show a particle storm, if the "curved CR hit" looks odd, could it be some other type of energetic particle? What type of particles (eg SEPs) can show up on the MDI images? Is there such a thing as a magnetically charged high speed particle that could perhaps be a CR imposter?

    Posted

  • pahiggins by pahiggins

    Well, as far as I am aware, most of the SEPs (solar origin) the instruments encounter are electrons, protons, and alpha particles (helium nuclei; but I confess I don't know the ratios- I believe it depends on the type of source). These are all electrically charged and thus affected by magnetic fields. There are some neutrals as well, which would not be affected by mag. fields.

    CRs are also usually charged. This is why the number of CRs we see decreases during the peak of the solar cycle when the Sun's magnetic field is strongest, and repels them from entering the solar system. According to wikipedia:

    'Of primary cosmic rays, which originate outside of Earth's atmosphere, about 99% are the nuclei (stripped of their electron shells) of well-known atoms, and about 1% are solitary electrons (similar to beta particles). Of the nuclei, about 90% are simple protons, i. e. hydrogen nuclei; 9% are alpha particles, and 1% are the nuclei of heavier elements.[10] A very small fraction are stable particles of antimatter, such as positrons or antiprotons.'
    -all of these particle types are charged.

    Since CRs are so energetic (traveling near C) it is not expected that their radius of curvature due to any conceivable magnetic field present in the instrument. One thing to note is that the magnetogram was taken two hours before the EUV mages showing the particle storm. This radius of curvature is called the Larmor Radius: http://en.wikipedia.org/wiki/Larmor_radius
    The relativistic case applies here. Assume Z=1, and v=0.5*c, where c=speed of light. I haven't done the calculation so I could be wrong about this!

    Some potential effects of electronics: http://en.wikipedia.org/wiki/Cosmic_rays#Effect_on_electronics

    Particle showers occur in Earth's atmosphere when a CR occurs. I wonder if the same could happen if the CR interacted with the metal in an instrument in space? http://en.wikipedia.org/wiki/Cosmic_rays#Secondary_cosmic_rays

    Posted

  • Mjtbarrett by Mjtbarrett

    Wow! Thanks for the comprehensive reply! I was sort of hoping there would be an easy answer so it serves me right... Newbie question: If energetic particles are tiny but have mass, and are travelling at near C, why don't they cause physical damage to the spacecraft I wonder?

    Posted

  • pahiggins by pahiggins

    By the way, consider the timing of the event. Here is the corresponding plot of the particle storm (SEP event): http://www.solarmonitor.org/goes_pop.php?date=20010416&type=proton
    which appears to have begun around 13:30 UT, the day before.

    Here are a few images showing what MDI saw just before the start of the event, and just after:
    http://soi.stanford.edu/production/gif_images/magnetograms/2001/04/MDI_mag_2001.04.15_12:48.gif
    http://soi.stanford.edu/production/gif_images/magnetograms/2001/04/MDI_mag_2001.04.15_14:24.gif
    http://soi.stanford.edu/production/gif_images/magnetograms/2001/04/MDI_mag_2001.04.15_16:00.gif
    http://soi.stanford.edu/production/gif_images/magnetograms/2001/04/MDI_mag_2001.04.15_17:36.gif
    ...
    And here is your image 24hr later:
    http://soi.stanford.edu/production/gif_images/magnetograms/2001/04/MDI_mag_2001.04.16_12:48.gif

    Compare to a couple EUV images ~12 and ~36hrs after event:
    http://solarmonitor.org/data/2001/04/16/pngs/seit/seit_00284_fd_20010416_022759.png
    http://solarmonitor.org/data/2001/04/17/pngs/seit/seit_00284_fd_20010417_010606.png
    Lots of SEP hits are still seen here, unlike the magnetogram.

    This could be due to image processing methods (that are not well described anywhere) that are designed to remove energetic particle hits from the magnetograms. According to SOI: http://soi.stanford.edu/magnetic/Lev1.8/
    'Level 1.8 processing provides corrections for plate scale, zero offset, sensitivity, and the most severe cosmic rays.'

    Posted

  • Mjtbarrett by Mjtbarrett

    From Wikipedia: "The gyroradius (also known as radius of gyration, Larmor radius or cyclotron radius) is the radius of the circular motion of a charged particle in the presence of a uniform magnetic field." and "Thus, the gyroradius is directly proportional to the particle mass and velocity, and inversely proportional to the particle electric charge, and the magnetic field strength." Does the magnetic field in a flare or CME Count as being a uniform magnetic field? If not, what effect would it have on the robustness of the calculation in this context (theoretically at least)?

    Posted

  • Mjtbarrett by Mjtbarrett

    Work in progress, please ignore this post
    http://adsabs.harvard.edu/abs/2010AAS...21641511K Pay wall!! Grrr
    http://eo.nso.edu/sites/eo.dev.nso.edu/files/files/archive/2009_final-report.pdf

    http://pamela.roma2.infn.it/index.php

    http://star.mpae.gwdg.de/gallery/Slides99_3.pdf

    http://en.wikipedia.org/wiki/Cherenkov_radiation

    http://soi.stanford.edu/cgi-bin/mdi/magday2.pl?Date=3026. http://soi.stanford.edu/cgi-bin/mdi/magday2.pl?Date=2001.04.16

    Posted

  • pahiggins by pahiggins

    Well, in the context of trying to figure out if you might see curvature in the path of a SEP or CR in the MDI CCD, you need to know the magnetic field inside of the MDI instrument- I haven't a clue what the magnetic field at a CCD chip surface would be... At a wild guess, somewhere between 0.5 gauss (the magnetic field at earth's surface) and ~10^5 G, the magnetic field of a strong laboratory magnet...

    Perhaps plug in a few numbers, assuming the particle has a mass of 2 protons (alpha particle), is traveling at 0.5c (c=speed of light), and crosses a magnetic field perpendicular to its path of 10^5 G.

    Then maybe change the mass to that of an electron. Assuming a light particle and a strong field will give you the strongest curvature.

    Posted

  • dvdgc13 by dvdgc13 translator, scientist

    @Mjbarrett, one thing to have in mind is how MDI detects the magnetic field. It does not work as an in-situ instrument, i.e. like a compass that let you measure the direction of the magnetic field which is passing through you. MDI (and similar instruments) works out the magnetic field from the light that it comes from the Sun. This is possible because the Zeeman effect. The zeeman effect has two signatures, on one side some spectral lines split, so we could look at the separation of these lines as a function of the magnetic field as shown in the first figure in the wikipedia article; on the other each the polarization components are different. To observe the splitting of the lines we need strong magnetic fields - so the line separation is sufficient for us to distinguish them - however, to analyse the polarization of the lines is easier for weaker fields. MDI measures the magnetic field by observing the right and left circularly polarised light. [ I see a blog post coming to explain all this with beautiful diagrams]

    So, I'm saying this to understand why MDI cannot detect the magnetic field of the particles that pass through. However, there's some in-situ instruments that can, one example is the Advanced Composition Explorer (ACE). It's located also on L1 point and you can see real-time data in solarmonitor too.

    @pahiggins, regarding the numbers you show yesterday, why would the magnetic field in SoHO be that big? The solar wind's magnetic field is on the order of 10^-5 G (1nT), right?

    Posted

  • Mjtbarrett by Mjtbarrett

    Sorry gang, this is my fault. Mea Maxima Culpa. 😦 I should learn to think before I type!
    Up until last week, the only things I knew about the sun were:

    1. Mrs B won't let me outdoors in summer without nearly drowning me in factor 50 and making me wear a baking foil hat or a balaclava (odd fetish - don't ask)

    2. It is generally hotter in June than in December in Yorkshire, and

    3. Errrmmm, I'll have to come back to 3).

    One of the problems for non-scientist enthusiasts of zooniverse projects (me) is that it is incredibly exciting to discover new things. That is even more exciting (sad) when "real" scientists engage us in conversation!

    There's a real sense of curiosity about any new project. The price we pay is that there's also an incredibly steep learning curve. I am old enough, and have taken part in enough zooniverse projects now, to know better than to respond quickly to any post from anyone who is more knowledgeable than me. But I fall for it every time...

    So here's the thing. I spotted an odd looking feature after having had it highlighted by someone else. I wasn't sure about the identification because it didn't look like I expected it to. An invitation was made to propose an alternative theory which I then tried to do (a bit rude but hey..). After I finally got to the point where I had had to put a bit of effort in I was happy to change my opinion. There's just a little niggling doubt left that this artifact may be something other than a bog standard CR hit on the CCD module. From poking around on the internet I discovered that there were different types of CR. I wondered what other particles may mimic the effect of a CR hit and wondered if there might be some way to perhaps differentiate between the particles based on their appearance.

    My question about the energetic particles having mass and travelling at near C was based on a half-remembered scrap of information about micro-meteorites; along the lines of "a wrecking ball with high mass moving slowly will knock a house down but the same amount of energy will be released by a small pebble travelling extremely fast and will therefore significantly damage a spacecraft."

    That's why even a tiny particle striking the mirrors or CCD in SOHO at very high speed seems like it should do some damage. It turns out that the answer may have something to do with the physical size and thin-ness of the pixels and the material that they're made of (energy dissipation being in there somewhere too, but that's a work in progress...

    Similarly the question about the Larmour radius was a reaction to new information. Given that I need an electronic abacus thingy to get beyond the number of fingers and toes of an average human male (having never successfully progressed beyond long division at school) I just wondered if the magnetic field around a bipolar sunspot and the L1 point could be described as stable. I hadn't realised that the calculation would, of necessity, be performed at the point of contact inside a shielded aluminium box on the spacecraft! My bad. 😦

    I was going to try to make a bit of an effort to figure all this out in my mind first and gathered the links above to see if I could better understand some of the processes without having to pester you all for information. It may help if you treat me a bit like the village idiot and assume that any reply of mine that you get within at least an hour of a post that you've created should be disregarded as the ramblings of a madman.

    As far as the differences over time in the fantastic images that pahiggins graciously posted are concerned; I sort of thought that you were helping me to help myself because I,d missed something important. I can see that during the solar storm there are lots of similar curved artifacts to the one originally posted. There are also some long, thin, linear tracks. During the "quiet phase" (ie no storm) there are very few. [I tried to track down more information but a lot of stuff is behind a blasted pay wall and I'm still a bit skint... ] I also spotted that there are two similar artifacts in one image, but one white and one black - I assume that they are of opposite polarity but appear to have a similar appearance.

    That's about as far as I've got so far. I'm really sorry if I have been a pain but it would be great if a chance comment led to a new method of categorising CR hits by morphology or "hey this is the first time we've seen squiggly CRs" or something. I am suspicious that there may just be a normal CR hit that has been altered in appearance by the processing pipeline or in later data manipulation or a weird instrument effect or something.

    I'm sorry if I've caused any confusion and remain enormously grateful that you are all taking such time and effort to help a thick impoverished zooite.

    Thank you. Very much.

    Posted

  • pahiggins by pahiggins in response to dvdgc13's comment.

    @dvdgc13 Yes the megnetic field of the solar wind would be nT, (10^-5 G?) but I was referring to the magnetic field generated by the electronic components in the instrument, specifically near the CCD chip. Perhaps it is small also- I don't have an intuition for the magnetic field within electronics.

    You are right, MDI would not show the magnetic field generated by a CR or SEP, like it does the magnetic field at the solar surface- so I hope I didn't give that impression.

    BTW, good explanation of how to measure magnetograms- I will be incorporating that into my next blog post 😃

    @Mjtbarrett Don't take anything we say the wrong way; we are extremely appreciative of the interest you have shown in this project, and we really value your input!
    BTW, I loathe pay-walls. I think all research funded by the public should be able to be read by the public! I know @dvdgc13 feels the same. This is a good case for citizen science projects to be given special log-ins so that volunteers can read papers. I wonder if anyone has proposed this before.

    Like@dvdgc13 said, the color of a CR or SEP hit in a magnetogram will not necessarily indicate its properties. When looking at an image of one of these events, you are seeing the interaction of a high-energy particle with a silicon chip (CCD camera) that has the ability to hold electrons in place in separate 'pixels', which are really just containers. My impression is that when a CR hits a pixel, it strips electrons off of the surrounding material, which then fill the container. The number of electrons in a pixel corresponds to its brightness, so that is why CR/SEP hits are bright.

    Getting back to the curvature of the (potential) particle path we are seeing in the image in question, the question is whether the magnetic field within the electronics near the CCD chip inside MDI is strong enough to noticeably bend the path of an incoming charged particle.

    My intuition is that it is not, because I have never seen a long particle track in a CCD image that curves.

    One possibility is that the feature we see in this image looks curved because it is the superposition of two simultaneous particle hits coming in from two different directions.

    Posted

  • Mjtbarrett by Mjtbarrett

    I love the idea of free access to papers!
    I found this if it is of any use: http://articles.adsabs.harvard.edu/full/2004ESASP.538..411S Interestingly on p 413 they note a large increase in CR hits during solar flares...

    Posted

  • dvdgc13 by dvdgc13 translator, scientist in response to Mjtbarrett's comment.

    @Mjbarrett, please, do not apology for your questions and comments (and this goes to everyone!). Most of the basic questions we get from the general public make us to to change the perspective in which we see our work, and many times it helps us to refresh our basic knowledge on the matter so we can explain it to you. For example, I had to look up how MDI works because I hadn't used it much since I finished my thesis, and that also helped to refresh my atomic physics knowledge (we should incorporated that in the blog post too!).

    Keep it up! Learning is an awesome experience, and which I wish I will never stop.

    Regarding @pahiggins comment about the journals: Oh yes! I'm really an advocate for open-science and open-access.

    Posted

  • pahiggins by pahiggins in response to Mjtbarrett's comment.

    Hi @Mjbarrett, would you mind listing the CR papers you tried to get access to, and couldn't? I will make an effort to get you access to these (and any future papers you would like to read).

    Posted

  • Mjtbarrett by Mjtbarrett

    Will do thanks! Work and temporary lack of computing power means using my 'phone for a few days. I'll catch up when I stop squinting at the screen 😦

    Posted