Sabine Hossenfelder: Is Dark Matter Real?

[Music] [Music] [Music] good [Music] [Music] [Music] [Music] [Music] so [Music] [Music] [Music] [Music] [Music] [Music] [Music] hello everyone i hope you are all healthy and well welcome to the next talk in our golden webinars in astrophysics series our speaker today is sabine felder who is a research fellow at the fram for institute for advanced studies my name is elisabeth artur de la villarroa and together with thomas puczia we have organized today's webinars for you as in our previous webinars language interpretation is provided by by patricia gonzalez who will be simultaneously translating for us into spanish in susie positivos puenes cuchala interpretation el espanol de la conferencia al pincial but we would like to acknowledge the generous support of the center of astrophysics and related technologies also known as kata for its spanish acronym thank you so much for your feedback and comments we really appreciate your coming back to us with suggestions if you're watching a recording of this talk on youtube please leave your comments below if you would like to support our series please give us feedback or send us an email and get in touch with us over our social media platforms if you have any questions during the talk please type them in the q a window you can also upvote questions and comment on them of course and we will select the best questions for the discussion after the talk the link to the live version of this video will be automatically taken down by youtube shortly after the streaming ends however we will put the high resolution version both in english and spanish of the final product online to the next channel to the channel in the next few weeks right so before we begin let's introduce the other panel members that are with us today one sabine our speaker of course patricia interpreter elizabeth and myself and from the institute of astrophysics at pontificia we visited catholica we have twila zilliotto who is a graduate student at the institute with us who is a postdoctoral fellow at the institute and paula uncle who is also a postdoctoral fellow at the institute with us we also have the great pleasure to welcome our guest panelists today emanuela pompei astronomer at the european southern observatory at la silla federico le research staff at archety astophysical observatory brian miller astronomer at the gemini south observatory stacey magog professor of astronomy at the case western reserve university william van der berg professor of philosophy at california state university at san bernardino paul heinengen juene professor of philosophy institute of philosophy at the leibniz university of hanover germany and lecturer at the department of economics at the university of zurich hansinnaker astronomer emeritus ex-deputy director of the sophia science center at nasa ames pavel gruppa professor of astrophysics at the university of bonn and the charles university in prague and last but not least we have our excellent q a managers ricardo acevedo and simon angel so it is our great pleasure to introduce sabine hosenfelder as our golden webinar speaker today sabine completed her undergraduate degree in 1997 at johannes wolfgang goetz university in frankfurt and maine she remained there for a master's degree which she completed in 2000 and after that she received her doctorate from the same institution in 2003. sabine stayed in germany until 2004 as a postdoctoral researcher at the gsi helmholtz center for heavy iron research in darmstadt she then moved to north america and completed research fellowships at the university of arizona in tucson the university of california in santa barbara and the perimeter institute in canada in 2009 she joined the nordita institute for theoretical physics in sweden as an assistant professor and since 2015 sabine is a research fellow at the frankfurt institute for advanced studies in germany where she leads the super fluid dark matter group sabina is an expert on quantum gravity and superfluid dark matter her work focuses on the role of lawrence invariance and locality which would be altered in the discovery of quantum gravity since 2007 she has been involved in the annual conference series experimental search for quantum gravity and she has also contributed to the foundations of quantum mechanics which she argues in favor of super determinism sabine contributes to the forbes column stars with a bang as well as quantum magazine nude scientists nature scientific american nautilus quarterly and physics today apart from the foundations of physics her interests also extended the philosophy and sociology of science so now we head over to sabine who will tell us about the question is dark matter real sabine the audience is yours well thanks everybody for being here or being there or wherever you are um i've been told that today is the day of astronomy in chile um and this year's topic is light pollution and i i don't have a terrible lot to say about light pollution but i thought i would briefly mention that a few weeks ago they exchanged the street lights in our streets so little city in the middle of germany and they took away all the yellowish light bulbs the old ones and replaced them with downward spacing leds which make a much nicer light and combat light pollution okay um [Music] but today i'll be talking about something completely different uh which is dark matter so dark matter is the stuff that most astrophysicists think make makes up 24 of the meta energy budget of the universe which is about 80 percent of the matter as you see in this pie here where all the stuff that we know and like so the normal matter that we're familiar with like a tiny teeny tiny little sliver of this yellow piece of the pie but that's not what i'll be talking about i'll be talking about um the blue part mostly and i will have to say a few words about dark energy which is this big mysterious uh green bit so um a few years ago i wrote an article for scientific american together with uh stacey magog and this talk is basically my part of the story and uh stacy is on the panel and i suppose you'll hear a little bit from him later so i want to tell you today how my thinking about dark matter has changed over the past 20 years or so um and explain why i changed my mind and what my opinion is today i used to think like 20 years ago that dark matter is probably a particle mostly because well i was trained i was a particle physicist and i had this feeling that well if you can explain it with a particle then why look any further also at the time there were a lot of conjectures for new particles that would fit the bill so that that would have made sense of the observations uh for example there are some supersymmetric particles and there are axions that would have worked and people have looked for observational evidence in direct detection experiments for these particles they began searching for this in the mid 1980s but as you all know they haven't seen them so not a single of these hypothetical particles was ever detected there have been some anomalies in the experiments over the years but they've they've all um gone away and so i gradually um i started thinking whether that can't be quite it uh something's going wrong and also there have been more and more observations that have been increasingly difficult to make sense of with the with the dark matter uh hypothesis so uh before i say anything more i want to make really really sure that we're all on the same page and i'll start with the first things first what is dark matter um so in in the first place dark matter as the name says is matter but when astrophysicists or cosmologists say matter they don't just mean stuff any type of stuff but it's stuff with a particular property and that property is that the energy density which is normally uh written row decreases with the inverse volume so if you have a cube with a side length of a as you see in this little image and the volume increases because the universe expands then the energy density of the stuff with what matter um decrease decreases with uh one over the volume and so um the only thing you need if the stuff is uh homogeneous is really the initial value that's usually called um rho naught so so that's kind of um one of the important properties of dark matter and then the other one uh as you can guess is uh in the word dark though that's a little bit misleading because um dark matter actually neither emits nor absorbs light so light just goes through light at any frequency um all the way from the long to the very short wavelength and that really means it's it's transparent not so much dark so it's not that it swallows light the light just goes through and finally dark matter um rarely interacts both with itself and with normal matter we just know this observational you wouldn't fit on the data but just how ex how rarely it interacts no one really knows we just have some some upper bounds can't be stronger than something and yeah so so then i should add of course well this is only the simplest type of dark matter usually called called cold dark matter and there are lots of variations on that but for the most part like most of the simulations have worked with with that type of dark matter so why do we think that dark matter exists before i come to the observations that are difficult to explain with dark matter i want to say um something about the observations that made astrophysicists think originally that it should exist there there are numerous independent lines of evidence that have accumulated for for more than 80 years and that's with the galaxy clusters which were historically the first evidence for dark matter in the 1930s by zwicky who observed the coma cluster which is this image that you see in the background so galaxy clusters are um several hundreds up to a thousand or so galaxies that are held together by their own gravitational pull and the higher the total mass in the cluster um the higher is the average velocity of the galaxies which you can observe and the observations show that the average velocity of galaxies in clusters not only in the in the coma cluster but in pretty much all clusters that have been observed is much higher than we can explain by the observed matter alone and so tzwiki suggested that there has to be an unknown additional type of matter in the clusters which he called dark matter it's a fairly similar story with galaxies themselves so in galaxies if you look at the velocity of a star in in a galaxy like say the milky way so you have a star that orbits around the center of the galaxy at a particular distance then you have a an expectation for the velocity on that orbit which you can calculate from the normal newtonian gravitational pull of all the matter that's inside that radius and yeah strictly speaking of course we know that newtonian gravity is not correct but we should be using general relativity but it's it's an excellent approximation in that situation to use newtonian gravity and so what newtonian gravity tells you um what the velocity should behave like with distance from from the center of the galaxy is that it should drop um roughly with um one over the square root of uh the distance but that's not what we observe instead what we observe is that the velocity of these stars which you can infer using the red and blue shift of the vector lines um roughly remains constant as you go away from the center of the galaxy and that's not only the case for the milky way but for for hope for a bunch of galaxies so it's a fairly general feature they don't always exactly say constant sometimes they have bumps and wiggles and some continue to go up sometimes they go a little bit down but they definitely don't do what newtonian gravity would tell you should happen and again you can fix that problem by introducing a new type of matter that we just can't see and you distribute it in these galaxies where you need it and that makes up the mismatch and the stars then fit onto the rotation curve then there's gravitational lensing so in general relativity um as you know space-time can be curved there's this whole picture of the trampoline that bounce if you put in a mouse so the masses curve space time around them and if light travels through that space time it has to follow the curvature of space time which mean means that masses act as lenses and from the strength of the deviation of the light that's caused by these masses you can deduce what the mass must have been to um cause what you observe as there's an example here in this image which is called an einstein ring so you have a mouse uh in the center and then there is another object behind the mass uh which becomes and the the light from this object behind the mass which creates the lens gets smeared out and and appears in several dots or sometimes a circle and the observations in this case too show that the normal matter just is not sufficient to explain what we see then there is the cosmic microwave background microwave background that's um radiation which which is left over from the very early phase of the universe when light could travel freely for the first time so that light is still around today it has an average temperature of about 2.7 kelvin so that's degrees above zero but the average temperature is not really the the most important part of the cosmic microwave background but it's that around this average temperature that tiny fluctuations which is what you see in this image so some directions of the sky the microwave background is a little bit hotter these are the red dots or a little bit colder which which are the blue dots and now um well looking at a picture like this doesn't tell you all that much so what astrophysicists do is they they basically count how many dots are there of a particular size and more technically you do a fourier transform that gives you what's called the power spectrum of the cosmic microwave background looks like this so what you have in the um vertical direction that's roughly the number of dots and you see this there's this really big peak here that's a kind of the typical size of the dots that you pick out with your eye immediately and on the horizontal axis you're roughly speaking see the size of the dots where the larger ones are to the left and the smaller ones are to the right and so it has this very characteristic curve um and again you need dark matter to get this powers right have a nice uh visualization for this um so here again you see the the power spectrum and this um pink bar which goes down on the very right um tells you the amount of dark matter so if it's high that's kind of um yeah so half half of all the um the entire matter energy content so and uh what you can see is if this amount goes up is that all the each every second peak in this curve gets suppressed so if if the amount of dark matter is at zero they fall off kind of in a in a line like this and uh if the amount of dark matter goes up um every second one gets suppressed and if you look back at this you you can see that the the second and the third peak are about um the same height and that tells you you need dark matter to get that right and finally so that's the the final item that i want to mention here that's a structure formation so that's the uh the large scale structure that we see in the universe the distribution of galaxies and galaxy clusters and galactic filaments and voids and all that um dark matter alone just doesn't give us the structures the way that we see it today that's because um it doesn't build up um [Music] sorry it's the the normal matter in the early universe creates radiation pressure which resists the formation of structures early on dark matter on the other hand because it doesn't interact with electromagnetic radiation cannot build up this radiation pressure and therefore it starts forming structures sooner than the normal matter and it aids the formation of structures from the normal matter which basically falls down into the gravitational potentials that were created by the dark matter so these are the most important evidence for dark matter and in in all of these cases the way that dark matter solves the problem is the same so what what you see here are uh what's called einstein's field equations um it's not really all that relevant if you know exactly what are these quantities stand for roughly speaking on the left side the the rs and the g's tell you what the curvature of space time looks like and on the right side uh you have all sources of mass and energy that are combined in this thing that's called t which is called the stress energy tensor and these little greek things here the mu and the new are indices that run from zero to three so this isn't only one equation but it's actually 14 except that these things are symmetric so there are only ten different equations so you have these ten equations that are einstein's field equations they're um they're partial coupled non-linear differential equations they're quite difficult to solve in general but the thing is that the way that dark matter resolves this discrepancy between the visible matter that we see and the motion that we infer which should have been caused by this matter is to just say well i fixed this equation by adding a new term here so um if from observations we see that this equation is not fulfilled we can always fix this by adding a term here now that's rather trivial like you can always do this it doesn't really explain anything but the important thing is that of of all these entries in this tensor so that's a more complicated type of matrix basically which could be 10 different ones in principle you really only need one which is the zero zero entry which contains um the energy density which is what what i mentioned earlier that's the important part of cold dark matter is the energy density and and that makes it a really simple explanation which gives this hypothesis its power so it's parametrically very simple you you pretty much only need the initial value for the for the energy density and that's where the appearance of this hypothesis comes from so the the obvious question that always comes up is can dark matter be normal metal like can it be something that we've already seen uh and so that that would be a different talk by itself but the brief answer is no so of the of the particles that we know that are in the standard model of particle physics none really fits the bill they either interact with light so that's the case for pretty much all of them the one exception are the neutrinos so the three neutrinos the problem with the neutrinos is that they don't clump enough they're just too light so that they don't make sufficiently many structures um and then you know people have considered a lot of other hypothesis like could be brown dwarfs or black holes or other exotic compact dark objects um it's not entirely perfectly excluded like it's generally really difficult to totally exclude something but the problem with with this idea is that they would make too many gravitational lenses uh which we have not seen we would see this for example in the in the milky way lensing in the milky way and also it would distort the cosmic microwave background um so it as i said it's not entirely ruled out um but uh it's really difficult to make it work and so the the general feeling is that it has to be something new it's if it's a particle it's it's not something that we've seen before but there are problems that dark matter does not solve um as as i said previously um and these you know when i when i was talking to astrophysicists about this like 10 15 years ago they would always tell me oh that that's you know that is a problem with the computer simulations that are just not good enough we're missing some astrophysical processes that we haven't fully understood and it would all work out fine and so on and so forth and you know 15 years ago it was plausible but over the years it's become less and less plausible so here are a few of these problems that that's not an exhaustive list but the probably most important problem is the the tally fischer law so that's an observed correlation between the brightness of galaxies and the asymptotic rotational velocity which as we previously discussed becomes constant dark matter just doesn't explain it so it's there in the data but why it should be the case dark matter doesn't explain also um if you do simulation for structure formation then dark matter gives you density peaks in the centers of in particular small galaxies and um that fits very badly with observations the observations say if you're inferring the density uh it should actually be flat uh so this is called um the problem of galaxy cos or the cosby halo problem or something like that dark matter also predicts too many dwarf galaxies at least that was the case for a long time for example the milky way should have like 500 500 600 small galaxies circling around it uh satellite galaxies but from observations we only have like something like a dozen so it it that does not fit very well with with observations and the small satellite galaxies are more often aligned in planes with a host host than dark matter simulations predict so so that that's just to give you um a selection of the problems that that large dark matter has the alternative explanation has been around for a long time was proposed in the 1980s by milgram and it's called modified newtonian dynamics i'll explain you in a moment just roughly how this works but i want to put a hat that strictly speaking we already know that mont is wrong so whenever whenever someone tells me that but mont is wrong i'm like yes we know it's wrong okay it we know it's wrong for the same reason that we know newtonian gravity is wrong strictly speaking newtonian gravity is wrong because it fails for certain situations where we have to take into account relativistic effects that's why it was replaced by general relativity and mond relates to the more overarching theory of modified gravity the same way that newtonian gravity relates to general relativity but we don't yet have this full theory of modified gravity to which mond should be an approximation so how's it work briefly that's how it works um in newtonian gravity we have the gravitational potential that's proportional to the mass of the object so in the simplest case that would be a point mass and then the newtonian potential then is just um proportional to 1 over r with the newtonian constant here in modified newtonian gravity the gravitational potential is instead a function of the logarithm of the distance to that point mass so that this is all only about point masses to keep it simple and if you take the gradient of this it gives you the force um which gives you for newtonian gravity the familiar one over r square law uh and if you take the derivative of this it will drop only with one over r um actually i i should uh add here that this is not strictly speaking the force it's the force per mass of whatever is the object you're you're looking at for example a star that orbits around the center of a galaxy and for purely dimensional reason you have reasons you have to introduce a new constant here which has the dimension of an acceleration so what can you do with that well you can you can look as i just said at a star that orbits the center of the galaxy so in that case this mass would be um the the mass of uh the galaxy and n r is the distance to the center of the galaxy and then you can ask what what what is what is the radius um for a stable orbit um so you have to balance this force with uh the centripetal force that depends on the velocity on that orbit which goes with one over r and then you can just solve this equation um for the velocity and you will see that um the square of the velocity falls with the inverse radius or i briefly mentioned this previously the velocity falls with 1 over the square root of r so this gives you falling rotation curves which does not fit with the observations now if you do the same in modified newtonian gravity um you must balance these forces and you see immediately that the r just drops out if you solve it for the velocity so this is why uh asymptotically the velocity just becomes constant so this gives you flat rotation curves you you also see immediately from this if you square this equation then the mass of the galaxy is proportional to the fourth power of the velocity which is the total official law so these these two things uh belong together so of course though you know that uh you you you can't just replace the one over square law uh with the with the logarithm um or the the force with a with a one over r force that just would work here on earth so what you have to do is that you somehow have to interpolate between our observation of the one over r square fourth law here on earth and this one over r force on galactic scales so modified newtonian dynamics has what's called the interpolation function that basically smoothly um goes over from the one over r square to the one over r um though of course you know if it's not spherically symmetric then it's much more complicated but roughly speaking that's that's how it works and the effects of mond become relevant below a certain acceleration scale so that's the scale a naught that appears here and not beyond a certain distance so that's like really really important to understand how this works um it's it's this acceleration that marks where modified newtonian dynamics becomes important and that acceleration scale just empirically uh seems to somehow be related to the cosmological constant so the cosmological constant is this lambda why that is the case uh no one really knows it's just um you know something that people have found out by looking at the data so modified gravity helps with um the problems that i mentioned earlier it explains the total official law it avoids the problem with the galaxy cost it reduces the number of dwarf galaxies it helps with the with the planar arrangements of satellite galaxies it does have difficulties with the cosmic microwave background with the early universe uh and with galaxy clusters situation is somewhat unclear for the solar system uh that's because it depends on the interpolation function okay so so it kind of looks like modified gravity it doesn't really quite fit the build because there's some things that that it can't do whereas dark matter might have some niggly things that make you feel a little bit uncomfortable but at least you can make everything fit uh and indeed if you if you try really really hard you can make it fit um so these computer simulations that i was talking about where they said like 15 years ago well there's something in the numerics we just have to work this out there's some astrophysical processes that we haven't included yet and so on they've pretty much managed to make that work and that's the price you have to pay for that which is that these computer simulations have become increasingly difficult they all introduce what's called subgrid parameters because you can't really resolve at least for now you can't really resolve galaxies on the level of individual stars so you need to kind of average over that and then you have to parameterize everything that happens below that scale so that that makes up these subgrid parameters and that's just there's just a lot of astrophysics going on there which is not constrained from any other observations and where you can play around with the parameters until everything everything fits um which is why i made this little remark earlier about the number of the dwarf galaxies because what's happened is that um the people who work on these simulations have played around with um the with the parameters uh for long enough to bring down the number of dwarf galaxies but meanwhile observational astrophysicists have discovered more dwarf galaxies that are going around the milky way and so now the situation is kind of reversed to what it was previously um now actually the the simulation seemed to predict fewer dwarf galaxies which is a little bit awkward but it goes to show that um these simulations they they don't make predictions you know you you take what you see and then you try to get it right and there's there's nothing wrong with this but it it means they're not they're not good explanations on galactic scales modified gravity is simpler and more predictive and it's just a better scientific explanation and indeed there was uh recently um david merritt wrote a very good book about this which i can really recommend it's called a philosophical approach to mont please don't be put off by the title it's called philosophical approach because he uses criteria that have been put forward for theory evaluation by philosophers of science and he just goes through uh what has mod predicted how did it work out and compares it to dark matter and at least according to merit monde comes out far ahead if you're asking for how predictive is it um so his summary in a nutshell is that one predicts and dark matter accommodates so you can certainly fiddle with these computer simulation and you can adjust the parameters and number cdm and so on and that fits all the data but uh mont has uh repeatedly scored and making predictions for example the baryonyx tally fischer relation the radial acceleration relation which i'll say a little bit more about in a second uh the the ratio between the first and second peak of the cmb though then it struggles with the ratio between the second and third there's something that's called ransom's rule which basically says that if you plot the observed luminosity in galaxies as the function from the center and you compare it with the rotation curves of a galaxy then for every feature that you see in the in the luminosity curve like a wiggle or a bump or something like that there will also be a feature in the rotation curve which if you look at it from the perspective of dark matter makes absolutely no sense because most of the mass in the galaxy should be dark matter so why would these two things be correlated and yeah so so amanda explains it dark matter doesn't and um so i i'm more on the theory side of general relativity and and for me mont has always been unappealing because it's missing this generally covariant completion that would make would make it compatible with uh general relativity and also because it struggles with some observations like the cosmic microwave background so i was really super excited when i came across this paper in in 2015 why curry and berezioni for me it was really eye opening and their idea was that dark matter can have two phases a normal fluid face um that would describe would be very similar to what we normally call cold dark matter and a superfluid face and then the superfluid face it looks like modified gravity i'll explain this to you in a little bit more detail um so uh in in this idea in the you have a normal phase for for the fluid um which is a fairly high temperature so you know for for for normal human beings these temperatures are ridiculously tiny you know we're talking about something like milk and also um and in this phase the fluid acts like normal particle dark matter um and it has mastered off about an electron volt and but then it has a superfluid face which you find um inside galaxies uh and in the superfluid um you have phonons uh in in the medium which carry a new long-range force that looks like modified gravity and i've written down here and the lagrangian um for that which i i don't want to say a terrible lot about this this chi that's that's called the kinetic term um in in the non-relativistic limit but the the important thing that you need to pay attention to is that in the lagrangian you have this weird power in the kinetic term so usually um it would be a square but it has it has a a word fraction here and the other thing is that it couples linearly so so the this field does the from the superfluid that's the face of the superfluid it couples linearly to the energy density of the baryons and so i i found this really eye-opening because it explains why sometimes particle dark matter works better and sometimes modified gravity the other thing that's very interesting about this is that it requires no interpolation function it it works a little bit differently than modified gravity though um for what the observations are concerned it gives you the same thing in a certain limit you add a new force to the newtonian gravitational force and that's the force that comes from the superfluid and that automatically gives you an interpolation um between the newtonian force on short distances and the mont-like law on longer distances so what does any of that have to do with me so far not a terrible lot um but in 20 and 17 i came across a paper by eric verlindel which i thought was really interesting um so where linda put forward his own idea for how to derive mont um which is is a very original way uh and i thought it was really interesting and i spent some time trying to understand this so what he does is he introduces a vector field which he calls the displacement vector that's the thing that's called u here from which he constructs a strain tensor so so um and this uh this vector is supposed to describe how a medium which supposedly fills the whole universe reacts upon the insertion of normal matter so he has this idea that this medium is elastic and it's incompressible and so roughly speaking what this displacement vector describes is if you if you push at the medium how does it react and that goes into the strain tensor so i had some difficulties with the paper because well for one this vector the indices that you see here the eyes they don't drawn from zero to three as you wouldn't have in in general relativity they only run from one to three so they're only for the spatial part of the vector and the other thing is that these derivatives even though verlander writes them with these nablas that are usually used for covariant derivatives they're actually partial derivatives and that makes me a little bit uneasy you know with this gr background um i i yeah i should mention uh this is really a plus if that's not a typo in in case you think it looks a little bit like the field strength tensor that you would find for example in electro dynamics it's not it's really it's really a plus okay so so what i did then was i thought well can can we maybe make this generally covariant because really what we're looking for is the theory that fits general relativity and the answer is uh yes you can it's like the first thing you do is that you you let all these indices run from zero to three not from one to three um you replace all the partial derivatives with covariant derivatives and then you know i i tried around for a little bit to find a lagrangian that would actually give me the equations that are on berlin this paper and after a few tries i hit on something and that's what came out um you have here um two uh a kinetic term that's composed of um the strain tensors quadratic in in the strain uh tensor so it's um it is quadratic in the derivatives of the field as you expect of a kinetic term and um that enters into the lagrangian with a power 3 over 2 which which you've seen before and it couples to the stress energy energy tensor of um or the the other the normal matter so to speak uh linearly okay so you see the class one u and a couples to t so there are these two ingredients that look very much like you have in the korean beneziani approach for the superfluid and i call this vector feed the imposter field because it looks like gravity but it isn't gravity and the reason it looks like gravity is because of that way that it couples to the stress energy tensor which is the way that normally the metric would couple yeah and and i wrote a paper about this and and that kind of you know i thought i would never think about this again but um i had a student who um later became very interested in that um and um so i've more or less been working on this for for the for the past couple of years now i i caught this whole thing covariant emergent gravity um and it has several advantages over the belinda approach um that come just from using this the lagrangian formalism one is that it respects all the symmetries and conservation laws um the same way that you have a general relativity um the equations are guaranteed to be consistent if you derive them um this way you don't end up with too many equations that in the end don't quite agree with each other it can be used beyond the spherical symmetric non-relativistic case which is what were linda looked at in the paper and i actually did a few calculations for that it lends itself to stability analysis though honestly we haven't done that uh in principle you can estimate the limits of the effective description from this and for me one of the most interesting aspects of this is that it does have the visitor solution so that means it it gives rise to cosmological constant so verlander in his paper he puts in the cosmological constant and and then he derives this mont-like law i don't need to put in the the cosmological constant it comes out from uh the lagrangian so that's that's uh really neat um the relations to the superfluid i already said that you have the same power of the kinetic term you have the same coupling to normal matter that's what you need to get uh the amount limit uh basically uh and if you look a little bit at this coupling um to the to the vector field and you know that um for for most practical purposes very often let me say like this very often you you have in the stress energy tensor only one component that really matters which is the zero zero component that carries the energy density then you really only have to worry about the zero component of this vector and that's basically a scalar function um so on galactic scales the two the two approaches are almost identical uh which is why um tobias and i have um so far pretty much exclusively worked with the curry model uh for it for a scale up here just because it's easier and one can actually solve the equations and so on so now let me say a little bit about the radial acceleration relation i already briefly mentioned this previously um so what you see in this figure here that's a double lock plot i said don't get confused um you see on the x-axis you have the acceleration that's caused by only the baryons and on the y-axis you have the the total observation um that you can infer the total acceleration that you can infer from the observations and so in uh you know if there was no dark matter and you just had newtonian gravity then you would expect these two to be identical so that's what this dashed line says me and all these blue dots here that are data points from the rotation curves of various um galaxies from one of stacy's papers and uh what the red line here shows you is that that's the prediction for this relation between the acceleration um for um the baryonic mass and the observed total acceleration and it comes out of the lagrangian it's basically it's basically a three line derivation you can literally do it on the back of an envelope and it just fits very nicely on the data i was really amazed when i saw this the first time there are no free parameters of in this from the parameters that are on the lagrangian two don't matter for this fit and the third one um is related to the hubble rate um according to a a relation derived by berlin in his paper and it just and it just fits on the data i mean you cannot complain you know it's not a perfect fit you know maybe there's a little deviation here but considering you know all the approximations that go into that for example spherical symmetry you know most galaxies are not actually it's very clearly symmetric i find it really amazing and i think is when i when i saw this curve i thought there has to be something to it it can't possibly be that uh my student can derive this curve like in three lines from a lagrangian that's that simple but if you want to do it with particle dark matter you need your i don't know 20 parameter um model and run it on a super computer for for several weeks and even then you have difficulty explaining the observations so i think from for me this was the point when i thought that there's there's something there's something to mind um yeah but we looked a little bit closer at uh what this might predict like if this is correct what does lead to uh one of the predictions that you get from this pretty much immediately from this curve is that these curves should not change all that much with redshift so if you um separate the galaxies into younger and older ones uh it changes a little bit but the only reason it changes is because the harbor rate changes and one of the parameters is fitted by the hublerate whereas in uh cold dark metal simulations the redshift dependence on the radial acceleration relation is much stronger so this is a cold dark matter simulation from from 2017. so this is not something that can presently be inferred from the data um the redshift resolution is just not good enough but maybe in the future it can it can be done um there are a few other observations that we looked at um at some point i became very worried about the galaxy lansing in that kind of super fluid model because if you remember in 2007 i think it was there was an observation of a gravitational wave event with an optical counterpart and the gravitational waves arrived at pretty much the same time as the optical counterpart with you know a delay maybe of a few seconds and now um if the photons would experience this force which comes from the phonons in the superfluid they would be lagging behind by several hundred years or something like this this um has actually ruled out some modified gravity theories because in the modified gravity theories um it's well it's not an interaction uh with with some kind of superfluid but it's it's property of the geometry you can't just say well the photons don't really feel it right it only acts on the baryons but on this superfluid uh model there's no particular reason why this force should act uh on on the photons it's actually it's quite easy um to make this coupling dependent on the mass of the photon so for we know the photons for all we know don't have a mouse so they wouldn't couple and you can do that but but that brings up the problem like what happens with the gravitational lancing um right so so you have one mouse which you would infer from the rotation curves of the baryonic mass basically and that must feel the phonon force in the superfluid but you can also try to infer the mass on the galaxy from uh strong lancing and that would have to give you the same mass like these these uh or the stream strength of the force or the the um you know the the mass of this imposter field whatever where you want to put it so these parameters have to fit together can you actually do that if the photons have a different coupling than the variants and the brief answer is yes you can it's actually it's not all that difficult um you can fit both the strong gravitational lensing uh and the rotation curves with with the same set of parameters and i i what i learned from this is that that's actually also the case for mont so so that's there's a um paper from 2013 from sanders where he points out that actually one can't uh it's really difficult like let me put this way it's really difficult to use strong lensing to distinguish between mont uh and uh dark matter the reason is that the strong lancing uh in galaxies is uh dominated by the mass in in the galactic center which is mostly baryonic mass anyway okay and then there's a recent paper which we did pretty much just for the fun of it where we showed that uh super fluid dark matter uh fits the milky way rotation curve uh very very nicely so i just wanted to show you this because i really like this figure um there are a few other ways to test superfluid dark matter um so i i don't have really good ideas how to go about this but i want to briefly mention it one is to look for particles in local experiment like uh direct detection experiments that's difficult because uh for this you would need the uv completion of the lagrangian so this lagrangian which we have for the super fluid is is a very low temperature approximation but in a laboratory on earth you would not be within the limit where where it's valid uh and since we don't know the completion it's somewhat unclear what to look for uh and also i would be guessing that they would be too weak or too late to measure so but i'm just guessing here you know i would be happy to be proved wrong on that uh then also this phase transition to superfluid should happen somewhere in the evolution of galaxies somewhere in in the past and that should make should leave an observable imprint you know the in the evolution of galaxies in principle it should be observable if you look back at all the galaxies um the problem is it's somewhat unclear what to look for because the equation of state for the superfluid is unknown so um we don't actually know exactly when it would condense or what kind of phrase transition it is like is a first order second order phase transition that kind of thing also it probably requires large-scale uh computer simulations uh because i can't really see that you could analyze this uh analytically and it's not that these simulations are impossible but they take time and money uh and people basically and um finding that there's an idea that was brought up by curry already in i think in the first paper if superfluids collide they can create interference patterns so they're periodic um patterns in in the density and in principle these should be there the problem is that the current observational data just can't resolve such small structures so i i don't think the chances are very good that we will see it so personally for me the conclusion from this has been that the likely reason we haven't made progress on this dark matter problem for for a long time as i said we've known this problem since the 1930s is that they're just the wrong people looking at it this isn't a problem for particle physicists and it's not even a problem for um people who work in general relativity it's really a problem for condensed metaphysicists okay we we need to figure out um what superfluids do in gravitational potentials like that's something that no one really knows we need to find a way to figure out exactly under which circumstances this phase transition happens or we need to figure out how we extract the equation of state from from the data so here's my summary a two-phase system i think is almost certainly the parametrically most simple explanation of our current evidence for dark matter it combines the achievements of both modified gravity and dark matter and i think the hypothesis of dark matter being a superfluid is well motivated and it makes testable predictions and that's why after a long phase of stagnation i think we're finally getting closer to solving this riddle which is more than 80 years old thank you for your attention thank you very much sabina this was absolutely amazing overview um i think arkham would be very happy listening to your talk um so we will take a 120 second break and please everyone send in their questions now who is connected and the panelists have a look at the q a and we will get back after the music is over all right so again thank you very much sabina for this great talk um let's start um with the question and answer and discussion section so um i want to ask the first question there is tension as probably all of us know between the hubble expansion that is measured locally and what the cmb is predicting this tension has been made even worse now with the new gaia results so there is a lot of strain on on the discrepancy now is your super fluid dark matter model somehow contributing to the resolution of that tension as i understood you had proposed that there is some coupling and that the cosmological constant actually comes out of your derivations so would that actually help to use that question uh eased attention so as i briefly mentioned this the the hubble rate comes into this model by an idea from eric verlinder which we used how to fix one of the constants and the parameters it's it's not without that it's not necessarily part of the theory and the the brief answer is that i don't see how it removes the ascension so we we thought about it a little but i i can't really see it i i believe that verlinda at some point may have something to say about it maybe you should you should invite him and and ask him all right so let's go to brian okay find the unmute button thank you thank you for a very interesting talk i was intrigued by the statement give me the prediction of some um interference patterns if to these superfluid cells merge and so if you go over a little more into sort of what that would look like what is the size of the structure we would have to detect and related with the existence of the superfluid affect how galaxies merge the time scales or what that would look like yeah so i suppose the the size of these interference product would be somewhere in in the kiloparsec range just because that's the typical size that you would expect for for that kind of perturbation i've i've actually never looked at it uh into close detail because i can't really see how the perturbation would be large enough to be observable um and sorry oh the other question about the merger um so that that's something that i can't really answer without having done some kind of numerical simulation which i haven't it's like so doing anything analytically in that model is like basically impossible unless you have a significant amount of symmetries it's like this typical problem that you have in general relativity so as long as you have a spherical isometric situation uh you can gain some insights from uh the equations but once you're away from this i so we for the for the thing with the milky way we've we've managed to do axial symmetry so we have a disc uh with a bulge uh and that kind of works uh it's a numerical thing already but now if you want to take two galaxies uh and figure out what goes on with the superfluid um i i think serbians would not be able to run it on his laptop anymore stacy okay thank you uh that was great talk and there are a lot of great questions in the q a um i'm i'm tempted to ask all of them um but emmanuel has pointed out one she wants to ask so i'm gonna highlight two um one of which i will start with i hope is the more focused one and the other is more speculative so uh niels martens uh asks uh specifically about the imposter field that you mentioned and in what sense is that not really gravity um you know saying basically if it walks like a duck and quite quacks like a duck why isn't it a duck um yeah that's a very good point um and actually i have to say that it's not all that clear like the the way that i was thinking about it originally is that it's a modification of gravity and and this field is part of the gravitational side it it changes the gravitational force um but then if you look at it as if it was a kind of a condensed phase of some kind of particle then suddenly it starts looking like it's a particle and it it becomes even more confusing um if you keep in mind that quantum mechanics tells us that fields are particles and particles are fields so what's really the difference and in the end i came to the conclusion that really the difference between the two is the form of the lagrangian like it's it's how it couples how the new stuff that you put in couples couples to gravity so um it's like my personal terminology is if if the force that causes the say the effect that we're looking at like the rotation curves is the normal newtonian force of something or the you know the general relativistic equivalent uh then i call it dark matter if it's not if that's not the only force that there is if it's or if it's a different force than newtonian gravity then i call it modified gravity but i have to i have to add that's not the way how other people would use the word definitely not so so other people would just say that this model is modified gravity so in the end i think that this it's really it's a fantasy you know to think that these are two entirely different things that they're not you know there's a blurry range in the middle where you can look at it this way or that way and it just flips over so the second question for that i wanted to highlight was from lindsay forbes uh who invites you to spec speculate on the numerical coincidence between the acceleration scale of mond and the square root of the cosmological constant and the product of the speed of light and the hubble constant so it's inviting speculation that can be very long or very short look i i've written a whole book in which i explained that we should not pay attention to numerical coincidences and i i think that's the only thing i want to say about it you know unless you can quantify how unlikely it is i'll not pay attention to it manuela please hello thanks sorry uh about the coexistence of a normal fluid and then superfluid in the normal fluid there's still the need for a dark matter particle but all the physics experiment have failed to find one so what could be this particle in your opinion well yeah uh thanks for pointing this out i i actually i think i didn't i didn't say that it all that clearly so um there are certain constraints on the type of particle that um give you the right lagrangian that makes a phase transition roughly in the right range uh one of which is that the mass has to be somewhere in the if if you use the curry model right it it it's very model dependent so if you use the query model then the mouse is about an electron vault uh if you use different models it can be lighter but it can't be a heavy particle and and so so that pretty much explains uh why these detection experiments haven't found anything it also as as i said uh it doesn't couple to photons if it would couple to photons we'd already have seen this with with the delay between the gravitational waves and the optical uh counterparts and that rules out that it's axions uh which would couple uh to the f menu um so what we know is that uh it's a light particle it could be a vector or it could be a scalar at this point you can't really say and you know something about the self-coupling uh because you need that self-coupling to be in a specific range for the condensation to occur in in the right range and so the self-coupling i don't really know if one could infer something from it about the possibilities to directly detect it i i can't immediately see how that would work philosophically speaking could this be the famous graviton i mean the particle equivalent of the gravitational wave philosophically speaking because we really don't know yet well i where you you needed to have a mouse um and for all we know the gravitons are massless um so you will have to introduce a mass for the graviton which you can do and people have done their theories for it but they they have other observational consequences so i'm not an expert on that uh but yeah i can't you know exclude it for any particular reason you would have to think about the uh about the self-interaction though right so i i kind of feel like it would be really difficult oh maybe you can do it i don't know okay and then there is another question by ruiz mermaid who asked is the transition at which temperature between a two super fluid state because it says that you the microwave background is a three kelvin so how can it be that we cannot observe on on earth okay it's a ridiculously low temperature i've forgotten exactly what it is so we're talking about milli micro kelvin something like this it's way way below the cosmic microwave background well i mean basically the least and the person asked how can this temperature exist when the lowest temperature we know is three kelvin oh well it doesn't it doesn't couple to the same b oh okay thanks okay uh federico please hello i would like to ask a question from constantinostasis about your covariant theory of emerging gravity and also other questions myself so the question from constantinos is the following what is the physical intuition behind the epsilon strain tensor that you introduced in your theory and my addition is the following during the talk you mentioned that in this theory you can understand the scepter solutions so you could explain naturally the gravitational the cosmological constant so i was wondering whether those solutions also explain the coincidence between a naught and the square root of lambda since you have later on shown that you can reproduce the radial acceleration relation yeah that's like two different questions right so the first one about the interpretation of the epsilon i don't know uh so as i said uh verlinder has this particular interpretation with uh the strain tensor and that has a geometric interpretation you can look this up on wikipedia if you want uh about what i said if you have an elastic incompressible medium uh you push it uh it then parts of it will go elsewhere right that's what happens like if you imagine you you push into i don't know a rubber or an eraser whatever way you want to call it um so that's the displacement vector and from that you calculate the strain tensor that tells you how the thing deforms but that's for belinda's interpretation and then i i just took this thing and i made a general covariant and at this point i i really don't know what it means i haven't thought about it you know i'm i'm afraid i'm like very much a math person okay so i i write down the logarithm derive something and see if it fits to the data so i'm really bad with interpretation of stuff now about the a naught um yeah you you may have seen that and found a little bit offensive that actually milo grosjean there's no a naught that's just this constant which is called l so i i've called these constants all a little bit uh differently and now i have to remember how they're all related um so the l uh turns out to be related to the cosmological constant which uh turns out to be related to um yeah to the potential so the prefactor in the potential so yeah i think that's right um but um there is a free parameter between the two terms right if you so if you um i mean that you have the lagrangian has three different terms and uh one of them you can throw away because of normalization uh but then in the other ones you you still have freedom so so it's maybe a little bit confusing the brief answer is no i cannot derive this relation i presume i was confused because uh you mentioned that you could reproduce the regular acceleration relation with no three parameters um and so i was wondering you know in that relation there is a scale an emergent scale which is a naught so where does this number come from from the lagrangian and then the same time we know that this number empirically is the square root of lambda at least you know uh the first order um and so you know i guess what i'm saying is that i would be really convinced about a modified gravity theory can sort of explain these type of things at the same time so so yeah i'm sorry i think i gave you about about answer which i haven't actually looked at the paper for a while so the the thing is that so the the a naught which you have in mind um you get this from the relation between the gravitational theory um the gravitational acceleration and um the observed acceleration so this this is where the a naught comes in now in in the theory that i've looked at the relation between the two is different i i mentioned this briefly it's like you don't have um a modification of you you don't have a different function of this gravitational force it's that you have an additional force and so you from from this addition of the two forces you can calculate a constant that plays a similar role as the a naught but it's not actually the a naught so if you ask me can i reproduce a relation between two constants in month the answer is no because this constant doesn't really appear in the lagrangian it's just a i'm not sure if i'm explaining this correctly does it make anything that's what i said it's not clear to me now i mean where you know the problem is okay i i swear if you look at the paper you look at the paper like it's it's only a few lines and you will see immediately what i mean it's kindergarten math thank you shall we continue with the pavel please well thank you for this uh remarkable presentation i really learned uh very much and um also much better now understood the super fluid approach and there are two questions which i thought i might try to merge and and amalgamate with something i also wanted to ask and that is uh the one by xavier hernandez and then by vasu girl so xavier points out that dark matter hasn't been found a lot of time has been spent on it a lot of ex expensive experiments have been uh spent on that and uh so he then writes uh to put it bluntly which funding agency wants to be the first to accept that millions upon millions have been sunk down the rabbit hole looking for the ether so that's severe uh and then related to this is an issue with antimatter i think antimatter um was not mentioned in your presentation and um and and um my and that was my vasu girl so um the um what i was wondering is that we we don't actually understand gravitation right so there are different ideas it can be um a um distortion of space-time it could be um an emergent phenomenon from um different regions of space having different information contents fairlinda and i was just wondering um we and and dark matter is extremely speculative and it's uh almost certainly well certainly doesn't exist given the astronomical evidence but uh since we do not uh know what uh gravitation is and we do and i think there is also no dark matter uh but we do know that antimatter exists and at this point we don't even know whether an anti-apple will fly up or down on earth right uh so why why are people not studying the gravitational effects of antimatter to the same degree experimental degree of an unknown particle or known existence of a substance that they are doing to search for for something which is extremely potato you know the dark matter did you have any uh insights on that and then related to that how would the antimatter fit into the superfluid model perhaps because you have this um a coupling of um of the superfluid uh dark matter um to the baryons and and we know that they are antibarion so do you have any ideas on that um on how antimatter would fit into to this picture and what you might expect about antimatter and its gravitational properties plus the experimental issue and all that money which is being lost so i let me let me start with the with the last question because that's the easiest one so the um the fear that we use doesn't have an electric charge so it's its own anti-particle it just you know um and you don't want it to have an electric charge so for for the superfluid matter it's kind of i i think it's irrelevant i mean um i don't know um maybe other people can try some other model but i i at least for this model i'm pretty sure it doesn't matter now let me say something about your question about antimatter why don't people study uh we we we don't know if anti-matter flies up or down and so on um well for one people do study it it's just really really difficult because an antimatter has this awkward property that it tends to annihilate with with normal matter and but they have done it at least with small things and they they do it at certain there's an experimental actually i think two experiments are devoted to that and to the extent that they have measured it we know that antimatter falls down the same way that normal matter does um as as you expect they can really only do it for really small things um and i i'm not uh you know i i don't exactly know what the current constraints are but you can you can look it up but it's also from the um and and i'm certainly not the first to say that various people have written about this it's theoretically uh it's really really difficult to get dark matter to gravitationally behave differently than uh normal matter because whenever you calculate higher order contributions in quantum field theory you you know the story with the mata antimatter pairs that make virtual contributions um for for example to um the the energy content of bond states uh right and and so if you say that actually the anti-matter has the opposite gravitational charge than the matter it it just comes out wrong it it just doesn't work and you can do violence to the theory and try to make it work really really urgently but i've never encountered anyone who could figure out how to do that so i would say that's like um from both from the theoretical perspective and also from all we know experimentally it's it's not a very promising avenue um and then the thing about the funding you know i don't have a terrible lot to say about that someone once suggested to me there should be something that's like huge cliffs law for scientific discussions that you know like this law that every internet uh discussion eventually uh turns to a conversation about hitler or something as for for science is that every every discussion eventually turns to a conversation about academic funding um i i i don't think actually that the problem is is with the funding agencies because why would the funding agencies have an interest in not making a new discovery i think the problem is more that whenever you want to get funding you have to um you have to go through this whole peer review process and it's just really really hard for a new and small community like the few people who are working uh on a two-phase system superfluid dark matter that kind of thing uh to compete with such a large and well-established community as you have for particle dark matter right now it's like it's really really easy to punch holes into this because there's there's so many open questions like you you've noticed this like in it's like an 80 of the cases the only thing i can say is i don't know i don't know nobody's looked at it we don't know what the equation of state is i don't know what happens if you did this i don't know exactly what's the scale of the interference patterns or what have you um and and it's just getting started getting somewhere is like really really difficult okay thank you paul please yes hi sabine pleased to meet you again um a nice talk thank you i have a potential objection and i would like to have your reaction on that possibly i can predict what your reaction will be because this objection is somehow aesthetic in the following sense i mean we don't know what dark matter is um it's very different from byronic matter that's pretty clear i think but the point is in your theory you now ascribe to dark matter a property that we know exists in bionic matter under certain circumstances superfluidity and somehow isn't that somehow well either ugly or implausible uh to ascribe that to dark matter or would you position rather be it i don't care about that sort of argument i just write it in the lagrange as you said before and then i calculate and then we'll see what the results are and whether you find that implausible that dark matter should be super fluid i don't really care well let me put it this way i think if you look if you look at the data forget about theory for a moment i think what the data have been screaming at us for at least two decades is that the simplest explanation is a kind of system that has a different set of equations below a certain acceleration scale and then in other regimes it does something else now you could say well i just use the one theory in this range and the other theory in that range and you know there's nothing in the scientific method that would forbid you from doing that but we have never seen a fundamental law that works like this you you don't want it you know you want an overall theory that allows you to derive in which circumstances is the one thing a good approximation and which is the other a good approximation and that's what in my opinion this two-phase system does for you it gives you a rational for saying in some ranges it does that and in the other region you use the other equation maybe there are other ways to do it but i don't know of any and i i don't know why you find it ugly you know to me it solves the problem that that's my perspective on it thanks er toilet please hello uh i would like to ask a question from rodrigo freitas how this superfluid dark matter will affect the formation of primordial blackhawks in the future could be touched its existing using gravitational waves from high redshift waves from high redshift so i i don't know what it would do to primordial black holes um my guess is that it would look exactly the same as in cold dark matter just because these holes would have to be formed in the very early universe and we if the superfluid dark matter idea is to work then the early universe should work pretty much exactly the same than it does with normal dark matter so i think this wouldn't change um i generally don't know how you would see evidence how how you would be able to tell apart superfluid dark matter from a cold dark matter say with gravitational waves regardless of what what redshift so i'm i don't know about that sorry okay thanks brian you're muted okay thank you um so i'd like to ask a question by nina mixa there was a paper published earlier this year that claimed to match galaxy rotation curves using straight general relativity with just the visible uh matter i'm using a gravito electromagnetic approach to general relativity uh could you comment on that approach and do you think this is viable well uh okay so let me say a fairly general thing i did not read the paper in great detail this idea has been around for 15 20 years there have been various papers been written about it the thing is that you can estimate the size of this effect it's a relativistic correction that you get in general relativity it's a really tiny effect the way that these effects tend to be it is extremely implausible that a tiny correction would end up making up 85 right i mean that's an obvious problem with that i looked at the paper um it's not explained in the paper i kind of lost interest at that at that point because you know if you don't explain what the obvious problem with the idea something weird is going on um there is um um a blog by name overcoming bias from a guy by name what's his name robin hansen or something um where he claims you know that he has found a problem in in this type of argument generally not only in this paper uh which may or may not be correct honestly as i said i wasn't even interested enough to look at it like even if it's correct it wouldn't it would only work for the rotation curves and as i said there's so much evidence for dark matter from other areas yeah thanks hello hi there this is hans can you hear me i can hear you oh yeah um so i must admit that i have not completely understood this explanation of dark matter with the transition of normal to superfluidity but the question is did you answer the question whether dark matter is real or is that not dark matter so is that just a semantics or or how should we look at this and the other question i have is what are the symmetry of since you're a mathematical person what are the symmetries or the additional symmetries if there are any in your equations uh in my equations uh so far from covariance yeah so um yeah that's an interesting question i haven't i actually haven't thought about this before so about the the title is dark material that was suggested as you can probably guess by the editor at scientific american uh you know i just thought it's it's an eye-catching title that uh addresses the general question um so i mean from from a certain perspective you can say well there's obviously something real going on because we have all these observations how can it possibly not be real but if you ask and i briefly mentioned this um if you ask is there a particle or is it not a thing but part of the geometry like is it stuff in the space time or is it part of the space time i don't know that is my question thank you yeah so so maybe then the answers uh maybe depending on how you look at it i really think it's it's a question for philosophers and and actually we we were at uh at a workshop like i think two years ago before corona uh where this was was discussed to some extent uh and in the end i think we didn't come to any conclusion but so you know i would suggest that philosophers think about this for a little more so about the symmetries so um in in the superfluid uh you have a u1 symmetry uh which you break uh in in this condensed um phase so so there's the symmetry but as i said it's kind of i i don't really think of this particular model as necessarily being the right thing it's kind of like an example for a certain class of models that must somehow have this property of having a phase transition and the benefit of the kuri model is that just very very simple um you know you can integrate the equations on on a laptop and it's possible to work with it so that's the good uh thing about it um and you probably know that uh milgram has been pointing out that there's a certain kind of um scale symmetry uh in the equations and so naturally since all these models reproduce this limit they will also reproduce um this scaling symmetry and i've tried to think about if one could try to turn it around you know could you take the scale symmetry and actually try to construct um the lagrangian from it but i haven't i haven't gotten verified it just didn't work in the end okay thank you thank you great talk really appreciate it um as a philosopher maybe i can just piggyback on hans that was a great uh lucky introduction that he gave us there for this particular question uh and i noticed that tomasula asked a similar thing in the questions um so what i'm getting at is the idea that in the history and philosophy of science we make a distinction between two different attitudes towards scientific theories the realized attitude where we interpret a scientific theory as being an attempt to describe fundamental reality versus an instrumentalist approach which says that scientific theory is just an attempt to get correct predictions about observable entities or observable quantities so i noticed that when we talk about mond we all admit that well it can't be true so we're treating mon instrumentally but then the question of the talk the title uh asked is that dark matter real so i'm asking now about your attitude sabina about superfluid dark matter do you think of it as an attempt to describe fundamental reality or is it merely a tool that's allowing you to make correct predictions about observables i'm afraid i'm an instrumentalist all the way down whenever someone starts talking about reality i get very uncomfortable um also as you mentioned like when it comes to question like how do you interpret um the strain tensor i'm like i don't really know you know it works you put it into the mouth you calculate something with it it gives you the right predictions and for me that's sufficient i understand that for some people that's not sufficient but but that's the way i look at it and maybe it's because i so i originally i have a math background um so so you know you give me an equation i can work with that and that's it thanks emmanuella okay so there was a very interesting question by sriram vaminatan i hope i answered it correctly if if in this theory there is any prediction at what temperature of the universe the phase transition from normal fluid to super fluid occurs so in that way we may try to observe it as some ratio to look for evidence so that's a great question it was one of the points that i had on my second to last slide that you should in principle be able to see evidence for this phase transition if you look back at older galaxies uh because you know there's the all the stuff pools and the gravitational potential at some point the pressure gets high enough to actually cause the condensation and that would should make a rather sudden difference for the for the structure of the whole for the whole gravitationally bound object but um we can't derive even for the simplest lagrangian where and under which circumstances the phase transition takes place and then there's also the question like is this really exactly the right lagoon it should be something more complicated so um this is why you know if if i had all the money that i wanted to spend on research the way that i would go about it would be the other way around i would look at the data and ask where would the phase transition have to be to fit the data the best and that would give you some kind of contours basically in in phase space uh contours of energy density pressure uh temperature and then you could ask well what kind of matter and what kind of lagrangian would give you face transitions that fit to the observations so that's the way that i would say we should go about it i think it's pretty futile to try and to try a top-down approach where you pick up a lagrangian like from out of thin air and then hope that it condenses in just the right way to fit the data i mean there are just too many guests that you could make thank you so what i i was just now wondering um well actually for some time already um that um i'm a little bit worried about the overall approach so um first i'd like to point out that or state that i'm i'm very impressed by the mathematical work which is being done it's it's very very hard and ultimately the price to pace that it might not be correct and yet it's essential for us to see um of course which way one can go and so here's my sort of uh well the problematic issue and that is are we not making potentially an error by by trying to find the theory um like like you're doing the superfluid approach and others are also doing with other approaches where we try to forcefully something which we know works and that is that moon works on the more local scale so local universe um to a large scale structure yeah so you take the cmb as the one boundary condition and then we we try to uh to force that into what we observe along the way and put primarily at the at the current time and that might be a wrong exercise because um the cmb might not be what it what we think it is now i know this is a a um something one one one one should not date to challenge um all hell breaks free if one days to challenge the nature of the cmb and yet there was a very interesting paper recently uh by uh average shook that the cmb might just be cosmological dusty mission the the the reason why i'm i'm not taking this i'm not discarding this is because we know that dust exists between galaxies and it would have been much more in the past since it slowly breaks up and the first stars which were formed would have blown out much dust to large distances because of the energy they are emitting and um and that does then just be between the galaxies that would be uh uh irradiating right and survivorship did the calculation he calculated the photons from that and it seems that it actually accounts for the cmb including the correct temperature and now starkman in a conference recently and he had published on that glenn stalkman pointed out that uh there's a lack of correlation um in the cmb fluctuations beyond a certain scale i forgot exactly how how many degrees it was but if if the cmb were of inflationary origin then we expect correlations on all scales and this has been quantified and the observations seem well indeed will show definitely that there is no correlation above a certain beyond a certain scale so it's it suggests that uh what we see as the cmb um um the those regions were not in causal contact with each other that's what the non-correlation would would mean and that's what the dust model would actually imply because the dust if if the cmb is thermal dust emission from cosmological times it would be exactly that implication and vibration because of points are that you get polar polarization signals and so on so what what i'm trying to indicate is that maybe we we have to be a bit careful with um and trying to develop a theory which which forces fit with the cmb and and the local data it might be safer to to work outwards from the local you know words outwards and you know that of course has implications for for your approach in particular also that the kbc void and the el gordo cluster might suggest that the successes of the lambda cdm model on large scales actually are major like more than five sigma failures uh and so again that is a problem right because if if one claims that this success then that is not really correct because uh the theory has been falsified on those uh scales beyond uh well on a scale of a gigapassic on and a dredge of nearly one where we see these very massive galaxy clusters which are colliding so just like this and i know that there's no answer to this and i think it's it's important to still do what you're doing but i'm just mentioning that one has to be careful with being certain that the higher edge of boundary conditions are the correct boundary conditions that the universe seems to be rather drastically different to what we see it what we think it is so it's not homogeneous it's much more structured with potentially huge density contrasts and globally we see a pretty big one with the kbc world so then if you have any comment on that particular statement from me but um you know i know it's a difficult uh topic to to to deal with mathematically so well that's very interesting but i haven't read the paper so i really i can't say anything about it um let me maybe just say that um it you know you you need this two-phase behavior even if you throw out the cmb um to get the large scale structure uh right and to get the galaxy clusters right uh and so on so it's not it's not only the cmb but that's that's that's the sort of problem it's not only that the cmb is at stake but also um the large-scale structure right so um the kp so the um uh the coenbarg embargo um i've gotten the next the so the kbc would as it's called so we see that launch under density of matter and a scale of something like 500 mega pascal across a gigapass nearly which is uh massively incompatible with any model which is which which works like the c like the lambda cdm model which assumes homogeneity and isotropic right so uh the real universe makes fabulously larger structures than we think it would be making based on that uh cosmological principle and that's and um and so um that's that's that's a problem right which one has to address which the superfluid model has to address because the success of of the model on large scale simply isn't there given the data that's that's the sort of what i'm trying to point at so i'm not sure if one can somehow change the super fluid model to allow for much more structure growth much more quickly maybe that would be something you could try to consider yeah maybe um you know i just think one has to start somewhere right and i think reproducing the achievements uh of lambda cdm together with those of mont on galactic scales would already be a good starting point yeah okay thank you okay so i'm gonna take the last question here from the q a by dietrich baden who is asking can the superfluid dark matter be heated locally so that it is locally no longer superfluid and what would be the observational consequences that you could predict would that be somehow falsifiable um basically it's it's kind of the opposite it's it's pretty much impossible to keep it super fluid uh around here like in the solar system on earth and so on so you would you wouldn't expect to see a super fluid here and and that's a problem because then you don't know what to look for is i i briefly mentioned this uh on my slide where i said other things that you could look for you would expect the particle still to be around like where should it go you know it has to be somewhere there but it's no longer condensed and this means that you don't really know what it does because from the astrophysical cosmological observations you really only know what it does uh at really low temperatures so if if you have it in a in a comparably high temperature environment like we would have here you don't really know what to look for like does it still have the same coupling to the baryons uh does it cover to something else uh does it not couple at all we have no idea really so as a corollary of that i think would you say that it's valid to say that there would be two different regimes in the structure formation in the universe right if you look for instance into the voids that yes on small scales galaxies would still form like they form in in normal filaments or galaxy group environments but on the largest statistical scales the structure formation would progress very differently than you would see in you know the over densities that we generally consider galaxy clusters no it's kind of the opposite way like you see the differences on smaller scales like on larger scales like on the scale of galaxy clusters and so on this stuff just behaves the same way like normal dark matter and you you start see seeing the difference when it's begins to condense which only happens uh with within galaxies like not overall in in in the cluster most of the stuff in the cluster is not condensed so clusters do kind of the same that they always do in in cold dark matter but you start seeing it in in in galaxies so it's a you know i i hesitate to call galaxy's small structures but i i guess if you think of the cosmology as a whole it counts as a small structure so that's where you would see the difference all right so let's wrap this up here thank you so much sarina for a magnificent talk thank you panelists for being around and asking great questions thank you to the audience equally for providing us with fantastic questions um please be so kind fill out the survey after the zoom webinar is over and the next golden webinar will be on march 26th which will be given by karen meach who is an astrobiologist at the university of hawaii and she will be talking about the archaeology of habitable worlds using small bodies to unlock the past remember omar moore so that's going along this direction all right everyone thank you so much again please stay safe stay healthy and until the next golden webinar in astrophysics take care everyone bye bye [Music] [Music] [Music] [Music] you

Channel: Astrofísica UC

Views: 94,956

Keywords: astronomy, Astrophysics, astrofísica, universe, cosmos, goldenwebinars, galaxy, galaxies, dark matter, gravity, stars, star formation, supernovae, black holes, quasars, spacetime, Sabine Hossenfelder

Id: ASjdlUVKmj0

Length: 111min 58sec (6718 seconds)

Published: Sat Aug 07 2021