Of Particular Significance

The Black Hole `Photo’: Seeing More Clearly

POSTED BY Matt Strassler

POSTED BY Matt Strassler

ON 04/16/2019


Ok, after yesterday’s post, in which I told you what I still didn’t understand about the Event Horizon Telescope (EHT) black hole image (see also the pre-photo blog post in which I explained pedagogically what the image was likely to show and why), today I can tell you that quite a few of the gaps in my understanding are filling in (thanks mainly to conversations with Harvard postdoc Alex Lupsasca and science journalist Davide Castelvecchi, and to direct answers from professor Heino Falcke, who leads the Event Horizon Telescope Science Council and co-wrote a founding paper in this subject).  And I can give you an update to yesterday’s very tentative figure.

First: a very important point, to which I will return in a future post, is that as I suspected, it’s not at all clear what the EHT image really shows.   More precisely, assuming Einstein’s theory of gravity is correct in this context:

  • The image itself clearly shows a black hole’s quasi-silhouette (called a `shadow’ in expert jargon) and its bright photon-sphere where photons [particles of light — of all electromagnetic waves, including radio waves] can be gathered and focused.
  • However, all the light (including the observed radio waves) coming from the photon-sphere was emitted from material well outside the photon-sphere; and the image itself does not tell you where that material is located.  (To quote Falcke: this is `a blessing and a curse’; insensitivity to the illumination source makes it easy to interpret the black hole’s role in the image but hard to learn much about the material near the black hole.) It’s a bit analogous to seeing a brightly shining metal ball while not being able to see what it’s being lit by… except that the photon-sphere isn’t an object.  It’s just a result of the play of the light [well, radio waves] directed by the bending effects of gravity.  More on that in a future post.
  • When you see a picture of an accretion disk and jets drawn to illustrate where the radio waves may come from, keep in mind that it involves additional assumptions — educated assumptions that combine many other measurements of M87’s black hole with simulations of matter, gravity and magnetic fields interacting near a black hole.  But we should be cautious: perhaps not all the assumptions are right.  The image shows no conflicts with those assumptions, but neither does it confirm them on its own.

Just to indicate the importance of these assumptions, let me highlight a remark made at the press conference that the black hole is rotating quickly, clockwise from our perspective.  But (as the EHT papers state) if one doesn’t make some of the above-mentioned assumptions, one cannot conclude from the image alone that the black hole is actually rotating.  The interplay of these assumptions is something I’m still trying to get straight.

Second, if you buy all the assumptions, then the picture I drew in yesterday’s post is mostly correct except (a) the jets are far too narrow, and shown overly disconnected from the disk, and (b) they are slightly mis-oriented relative to the orientation of the image.  Below is an improved version of this picture, probably still not the final one.  The new features: the jets (now pointing in the right directions relative to the photo) are fatter and not entirely disconnected from the accretion disk.  This is important because the dominant source of illumination of the photon-sphere might come from the region where the disk and jets meet.

Updated version of yesterday’s figure: main changes are the increased width and more accurate orientation of the jets.  Working backwards: the EHT image (lower right) is interpreted, using mainly Einstein’s theory of gravity, as (upper right) a thin photon-sphere of focused light surrounding a dark patch created by the gravity of the black hole, with a little bit of additional illumination from somewhere.  The dark patch is 2.5 – 5 times larger than the event horizon of the black hole, depending on how fast the black hole is rotating; but the image itself does not tell you how the photon-sphere is illuminated or whether the black hole is rotating.  Using further assumptions, based on previous measurements of various types and computer simulations of material, gravity and magnetic fields, a picture of the black hole’s vicinity (upper left) can be inferred by the experts. It consists of a fat but tenuous accretion disk of material, almost face-on, some of which is funneled into jets, one heading almost toward us, the other in the opposite direction.  The material surrounds but is somewhat separated from a rotating black hole’s event horizon.  At this radio frequency, the jets and disk are too dim in radio waves to see in the image; only at (and perhaps close to) the photon-sphere, where some of the radio waves are collected and focused, are they bright enough to be easily discerned by the Event Horizon Telescope.


Share via:


31 Responses

  1. Matt, are BHs the path which the universe reverses back to a singularity, the big bang? Yes, the universe is still expanding but as the BHs grow in number and size there could be a point where the direction reverses and starts to contract back down to a singularity.

    Is it possible to measure Hawking’s radiation from this M87’s BH?

    Apologies, to my out of scope tendencies. Too many questions so little time. 🙂 I’m old! 🙂

    1. I’m afraid there’s one problem with your theory; black holes don’t have any more gravity than the matter that makes them up. That is to say, if our sun were to collapse into a black hole right now, Earth’s orbit would be unchanged.

      Similarly black holes as a whole are not something that is ‘growing stronger’ over time. They can pull more matter into themselves and thus make gravity more extreme locally, but the overall gravity of their system will remain unchanged.

      As it is with the apparent accelerating expansion of our universe we won’t see a collapse (‘Big crunch’) unless something changes in the future. (Even without the accelerating expansion our universe seems quite close to being flat, which would avoid a collapse.)

  2. The image shows no conflicts with those assumptions, but neither does it confirm them on its own.
    Depending on the properties of dark matter, it might indeed glow. Its superposition applies (bend light) to different Spacetime geometries (Fluorescence) becaz, Energy is a form of Space. Penrose “Objective Reduction” refers to idea about quantum gravity.
    At singularity, the impossible Penrose stairs “connect” is broken, there is no particle, (in unitarity violation this “connect” is not broken, thus accommodate the “*Bend”) so no black hole rotation and no time, only Space or different spacetime geometry.
    This *Bend (quasi-silhouette) does not imply fully the properties in different spacetime geometry ??

    1. Black hole makes Equivalence principle inequivalent?
      The “bend” that creates time and thus “appear” as mass again goes to its unequal or another universe. So no mass and no “bend”. Blue shift and red shift balance or implicit quantum effect.

      1. The coordinates of different events (light cone) inside equivalence principle is same because the “bend or time difference (Dirac ball on spring)” which creates mass, again come inside the event as particle. It is like the geodesic on surface of earth. If the “bend” coudn’t go to the future or Past, no time and no particle (black hole?). Blue shift cannot become red shift or “make a bend” in another space into spacetime. ??

  3. Professor Matt, first I want to say thanks for your outstanding posts (it’s hard to find these gems of the web).
    I do appreciate your dislike against the “shadow” term.
    And I really do appreciate your warnings concerning assumptions. I am not an expert at all in this field. But while reading your posts, I am getting a strong feeling, that having the first “photo” is indeed a good thing (while my first reaction on April/10 were some doubts whether the effort pays off).
    The “photo” in combination with your questions increased my sensitivity for the lack of knowledge-by-observation so far.

    Terms are loaded with assumptions. I like to look at these terms as a short abbreviation, useful for conversation. “Dark” matter is also a convenient short word. I would prefer “invisible” matter until somebody finds a better name.
    For an experiment of thought: When we assume that a special type of black hole exists, which completely emerged from dark matter: How could we tell the difference?
    As I understand the results of the theory, we would have a hard time in finding any difference.
    And what about the transformation of matter into energy? If I remember this correctly, the processes near a black hole are the most efficient ones, that we can find in the universe. What if dark matter plays a role in this game? If, finally, dark matter is not completely dark (which is something that could be imagined, if gravity and electromagnetism can be explained by the same laws), then even dark matter could finally glow in the process of transformation into energy.

    I am just letting my thoughts wander. If my naive ruminations don’t ring a bell (maybe I am completely off the road; I am not an expert) don’t bother commenting.

    I like to think about the low energy side of the spectrum, where extremely red-shifted light can be observed. I am wondering: Is there a maximum wavelength at the long side of the scale? A kind of limit for the red-shift?

    Looking forward for your next post…

    1. It is a theorem of general relativity that it is impossible to tell whether a black hole formed from ordinary matter, dark matter, or even pure light (though the latter, while possible in principle, is not likely to happen in our universe.) We have no clear idea how the black holes at the centers of galaxies were formed and how much dark matter went into them.

      Depending on the properties of dark matter, it might indeed glow. But this requires dark matter collide with dark matter; collisions of dark matter with ordinary matter are already known to be very rare. It is hard for this glow to be bright, but it has been looked for; indeed entire conferences and workshops on this subject have been held in the last ten years, for example this one: https://agenda.albanova.se/conferenceDisplay.py?confId=3914 . Look up experiments with names such as HESS and FERMI.

      There is no true maximum wavelength except the size of the entire universe; practically speaking, however, we don’t measure anything close to that.

      1. Thank you for the links.
        I know that this might be getting off-topic now, because the following thought is about things we can’t measure. (On the other hand: Isn’t QM in a nutshell something like “if it can’t be measured, it doesn’t exist”?)
        Concerning the maximum wavelength: If we directly look at the quasi-silhouette, our instruments can’t measure anything above noise. If I understand the theory correctly, there should be some extremely long waves coming out, though our instruments can’t measure these.
        For practical reasons in the first place.
        But if the size of the universe would be a theoretical limit for the emitted wavelenght, what about GR then? Why should GR care about the size of the universe while stretching space and time?

        1. As far as I am aware NO scientific theory says ‘If you can’t measure it, it doesn’t exist.’ The Copenhagen INTERPRETATION of QM suggests that things that are not known to exist in a single state exist in multiple states simultaneously. (Thus Schrodinger’s cat, which is alive an dead at the same time because it hasn’t been measured, but certainly exists.) But there are other interpretations that are (at present) equally valid.

          Philosophy does ask the question of whether something that has no effect on anything else can be said to exist, but that’s a different matter.

          There are two limits to what our instruments can measure. The first is any ‘signal’ weaker than noise. This doesn’t relate to wavelength (except in that some wavelengths will have more noise than others.) You can’t see stars in the daytime because their shortwave visible light is drowned by the noise of the blue sky.

          The second limit is their sensitivity to various wavelengths. The EHT focuses on a very limited range and it misses wavelengths both longer and shorter than this. (For example it picks up no visible light and cannot detect kilometer long radio waves which other instruments can make and detect quite easily.)

          That being said, the question of a longest possible wavelength is essentially the question of whether there’s a lowest possible photon energy; a photon with so little energy that it can’t have any less. I have seen theories (Or rather discussions of them) where energy is quantized, but I suspect they’re far from the mainstream for a reason.

          As far as I am aware neither GR or QM puts a limit on the lowest energy\longest wavelength possible. (Professor Strassler suggests the size of the universe but I have not seen any justification for this and would be interested in the theory behind it.)

          Any ‘hyper redshifted’ light would need to be coming to us from closer to the hole than the photon sphere I believe, since the sphere bends light but is relatively neutral when it comes to making the light climb out of a gravitational well.

          Redshifted light would come from matter falling into the hole that had to take a more direct path towards us, it’d be lighting the black circle in the middle of the image. There shouldn’t be much matter between the edge of the accretion disk and the hole and the total energy of photons emitted will be quite low so I don’t think we’re missing too much at this stage.

          1. You are right, I have to take back my “QM in a nutshell” sentence.
            As far as I know when Heisenberg was on Helgoland, he was led by his intuition that the laws of physics must be based on what can be observed. This is what I had in mind, when I wrote that comment. But I was mixing things up. It could be that there is hyper-redshifted light, even if there exists no structure in the universe to detect these photons.

            Concerning the amount of stuff, that could be the source of some escaping photons: I think everything that has started its journey towards the black hole is still on its way from our distant observer point of view. Because of time dilation we never can observe anything reach the “event horizon” in finite time. Assuming we could detect ultra-low-energy photons and assuming we are willing to wait for eons.

            (And from the infalling particles point of view: Shouldn’t the black hole evaporate before the particle can reach the event horizon? I prefer to think that there is no singularity and no place “beyond” a horizon.)

        2. The interaction of black holes with matter is a complex and subtle thing.

          The first issue is QM itself; if the universe were continuous we might well expect mass falling into a black hole to constantly be emitting energy from our POV. But the fact that photons are discrete puts some severe limits on this. The time dilation we ‘see’ increases exponentially over time (from our POV.) and rapidly becomes near infinite.

          Given the emission rate of a body it’s not hard to calculate that if we start timing when time dilation is less than 10% then after around second from our POV, for stellar mass black hole (A bit longer for a supermassive one.) it’s far more likely than 50-50 that we’ve seen the last photon the body emitted before it fell below the horizon. (That is to say an infalling body not only emits less energy from our perspective but also less photons per second to the point where even given the massive number of photons something like a lightbulb emits this can be rapidly slowed to essentially zero.)

          The second issue is that the classic ‘object seems to sit at the horizon forever’ scenario only works in one, single, very specific and actually impossible situation: It’s true from the point of an observer infinitely far from a hole that does not ever grow or shrink.

          To see this you can imagine a body falling into the hole and leaving a ‘trail’ of photons behind that are slowly escaping due to time dilation and whatnot. We can imagine the ‘last photons’ being right near the edge of the hole almost ‘held in place’.

          If the hole gains mass from another infalling body, or even just feeding off the Cosmic Microwave Background it will expand outwards, the growing horizon ‘swallowing’ those last photons. We’d see the hole grow to encompass the first object. On the other hand if the hole shrinks due to emitting Hawking Radiation the time dilation at the point in space where the last photons are will lessen and they will be emitted at a time less than infinity, letting us see the object fall into the hole. If we approach the hole we’ll run into photons trying to escape it as well as experience time dilation and see the objects fall speed up. (To the point that, if we go to where we think the object is it’ll always fall away from us right up until we ourselves meet the horizon.)

          There’s even an interesting effect with what you’ll see with a single body falling into the hole.We generally imagine it going darker and redder but remaining unchanged; this is inaccurate. Imagine a glowing sphere falling towards the hole, emitting light in all directions.

          In normal space you can see half the sphere’s surface; a collection of points where you can draw a straight line between the sphere’s surface and your eyes. But as the sphere gets closer to the hole light rays bend. The first effect is that the side of the sphere nearest the hole is more time dilated so moves more slowly from our POV… so the far side will fall onto it; the sphere will look like it’s flattening out into a disc.

          Now at the ‘photon sphere’ 1.5x the event horizon away light can orbit the hole; light from the side of the sphere can move right around the hole and hit the opposite side. (If you were there you could see the back of your own head in front of you.) But light from other angles will also be warped; light not heading directly for you (or rather away from the hole) will be increasingly bent. Photons that might have been 1 degree off will become 2, 4,8..90… 360… as the object nears the hole it will seem to spread out until it eventually seems to cover the hole’s entire surface. (At any distance there will be some light sent close to the hole, bent around and sent back, meaning that at any time you should be able to see a distorted view of an object falling towards a black hole in the ‘light ring’ like that seen in the black hole photo discussed here.)

          The end result of this is that only the most recent object to approach a black hole will be visible and it will cover the whole horizon, any newer object will ‘encase’ any older ones and from observations of the hole it can be calculated that previous bodies will now reside inside the event horizon.

          As a side note this also limits the ultra-redshifted light; only light emitted directly away from the hole near the horizon can escape, any with sideways motion will be bent inwards and absorbed. The amount of light that can escape is 50% at the photon sphere (1.5R, where time dilation is essentially zero.) and rapidly shrinks to nothing the closer the object gets to the horizon. So only the last object counts, it has a finite number of photons it can release and essentially zero of them can even theoretically escape near the horizon.

          Now as for a falling body’s view, it should notice no difference and fall past the event horizon in a relatively short and calculable period of time. Why doesn’t the hole seem to evaporate away before it reaches it? The trick is that the falling body is different from the light it emits.

          We see the last photons emitted by an infalling body and note that they’re very time dilated, nearly infinitely so. And indeed, if we were discussing a body moving away from the black hole at light speed from just above its horizon, a significant amount of evaporation could be expected to occur before that body could escape the hole and reach our eyes.

          But the actual body that emitted those photons is moving inwards, blueshifting the black hole. At the horizon it’s falling in at light speed (even if massive.) and this essentially cancels things out.

          As for what’s beyond the horizon, who knows? I doubt singularities are a thing; we get one when we let relativity’s math run until the theory breaks down, but that’s hardly proof. If I get a nonsense result my first instinct isn’t that nonsense must be right but that my math is incomplete. Whenever we’ve had singularities or infinities in the past (Such as, say, the Ultraviolet Catastrophe) we’ve eventually found a bigger, better theory that avoids them. I suspect black holes are the same.

  4. The interesting part for me is the bright region in the lower half which is a consequence of relativistic beaming; I’m assuming this is related to synchrotron beaming in particle accelerators. However, in say the LHC as an example, one sees this radiation as being tangential to the particle’s velocity vector at any instant whereas in the black-hole images, it’s continuous along an arc of the circle. I’m trying to understand this initially as a consequence of the black-hole bending the radiation towards our field of view somehow.

    1. Unfortunately I cannot yet explain this in a simple way, especially since the details of the mechanism depend on the black hole spin, which is still not well known even under reasonable assumptions. An expert told me today he has a hunch that the spin is greater than 0.7 of the maximal allowed spin, but even if he is right, there are big differences between 0.7, 0.95, and 0.999. It may be months before I understand this.

  5. Matt, thanks for the indefatigable and open examination! We bystanders learn a lot (and get some other stuff, like the “mostly BH optics, not so much disk detail”, confirmed).

  6. Hi Matt, that was exactly my point; the claims surrounding the photo of April 10 describe something EHT team insists upon ignoring.

    I agree with your depiction based upon the ‘Kerr Black Hole’ proposition, but lacking rotation, EHT (and 11 papers written since 2008 proposing M87 is a massive Black Hole) fails to demonstrate their claims through observation.

    EHT claims are too vast and unverifiable, in my opinion, going back to the original Austrian Team claims of M87 being a massive black hole back in 2008.

    1. To my reading of the papers the EHT team do not “insist” on ignoring stuff, they ignore some stuff that is not making much of a difference the first time around – and they are open with what they approximate and how they tested it. I can agree with Matt that the press conference had them argue differing opinions not supported in the paper (on the jet mechanism, say), but the papers clear that out and it was their science to discuss.

      Conversely, you ‘insist on ignoring’ that the community – so far – has found that they *have* demonstrated their claims. Oy vey, that observation is (demonstrably) not rocket science.

      1. Also, I note that my formulation is sophistry (though valid claim). Sorry about that, I usually hate it. But I had an aborted discussion with a “bible code” type starting with *this very subject* today – he interjected his crackpot conspiracy theory into a discussion out of the blue (or perhaps “black”) and then started Gish galloping from there – and I may be a bit touchy today.

  7. So the ring we see is light being bend around the photosphere, bend by the black hole, right?

    Cool, that makes a lot of sense.

    So, if we want to see more features around the black hole, we would need a better telescope (e.g. adding a space based component to the EHT)? And I suspect the accretion disc itself would become visible, albeit much dimmer than the photosphere? And could maybe the “small bulges NW and SE” be from the accretion disc? Or am I completely off track here?

    1. Basically right, to the edge of my understanding. I don’t personally know yet what it would take to make the accretion disks and jets and the transition region between them visible in radio waves. Neither do I yet have insights into the extra light outside the photo-sphere; those who understand the modeling of these systems probably have a best guess, but I don’t know what it is.

  8. Really helpful having you do this Matt. Thanks not only from me, but also on behalf of the students I (attempt to) teach.

  9. Matt, thanks for these posts.

    Nice to know that the small bulges NW and SE in the image are not the actual jets.

    Will further images of this black hole over the coming months help figure out what’s going on.

    1. That’s right, the jets are not oriented that way. I’m not sure if anything specific is known about those bulges yet, I haven’t found mention of them in the papers. Those subtle features may be transient… that is, if you looked again a few days later they might have completely changed.

      For further clarify: a few years rather than months, I think… but yes. When more images are taken, study of the time-dependence of the images may clarify what’s transient in the image versus what’s stable and characteristic of the system as whole. Observations at another radio frequency will give some complementary information. And over the longer term, addition of space-based radio telescopes will improve the resolution of the image.

Leave a Reply


Buy The Book

A decay of a Higgs boson, as reconstructed by the CMS experiment at the LHC


I’ll expound below upon the second bullet point, hoping to draw attention to general questions concerning aesthetics in theoretical physics.

POSTED BY Matt Strassler

POSTED BY Matt Strassler

ON 11/06/2023

Once we clear away the hype (see the previous posts 1, 2, 3, 4), and realize that no one is doing anything as potentially dangerous

POSTED BY Matt Strassler

POSTED BY Matt Strassler

ON 12/18/2022