 Hello everyone and welcome to our six web seminars of the series of the Latin American webinars on physics. My name is Nicolas Bernal from the new ICTP Safer, South American Institute for Fundamental Research in Sao Paulo, Brazil and it will be your host today. We have a special webinar today because we have two speakers. On the one hand, we have Andrea Albert from the SLAC working, SLAC National Acceleratory Laboratory. She's her PhD in the Ohio State University and now she has a positive position at the SLAC working on the Furby Large Area Telescope. On the second hand, we have Miguel Ángel Sanchez Conde from the Oscar Klein Center. He is his PhD in the Instituto Astrophysicale de la Lucia in Spain. After a couple of positive in the Instituto Astrophysicale de Canarias and the Stanford University, he moved to the Stockholm and he is now a fellow at the Oscar Klein Center. Andrea and Miguel, we talked today about search for gamma rays from dwarf galaxies with the Fermi Large Area Telescope. And we are glad to have them as our speakers today. To remind you that we can be part of the discussion, writing questions and comments using the Q&A system, the Google Plus Q&A system. And on Twitter with the hashtag L-A-W-O-P. Now I will hand you over with Miguel and then Andrea. So thank you, thank you, Nicolas. So I guess that it's time to share my screen, right? Right, please go ahead. Okay. So is everything okay out there? Excellent. Sure. Okay, great. Okay, thank you for this invitation, guys. Of course, we are very happy to be here, both Andrea and myself, on behalf of the Lab Collaboration, by the way. So today we would like to summarize what's the current status of our search for gamma rays from dwarf space faraday galaxies and satellites of the Milky Way with Fermilat data. So I will start first introducing you to this, to the field, essentially giving a broad overview and put everything into context. And then Andrea will follow in the second half of the talk, essentially summarizing our lab search and the kind of analysis that we perform on these objects. Okay, so you can go to slide two. Of course, it's not my intention to go so much into the details of this. You all know that today we have an overwhelming evidence to believe that most of the universe is actually, most of the matter in the universe, is actually composed by a non-varyonic form, exotic form of dark matter. So these evidences have been reported actually at different scales, and that's essentially the main message that I want to make here. You can see, you know, I just give a few samples at different scales, from galactic scales to galaxy cluster scales, cosmological scales. So as I said, we have an overwhelming evidence, also a national evidence to believe on this kind of dark matter. So go to slide number two. This is again, a reminder about the beautiful cosmological framework that we have at the moment, that we call Lambda CDM, that in principle, you know, give us a good explanation for all this, also a national evidence I was talking about. Just a brief reminder about this, the main thing is that it's settling this big planet scenario, of course, then we need this non-varyonic form of dark matter. Mainly to explain, you know, all this kind of observation that we have at different scales in a consistent way. Then it must be called, and this is mainly due to the fact that we need these dark matter particles to move at non-relativistic velocities, in order to explain the kind of large scale structure that we observed in the current, in today's universe. And also to understand how the obvious form and evolve, or at least the kind of picture that we have from observations. And then, of course, you have this Lambda term, this enigmatic energy, in order to explain the measure of cosmic acceleration. As I said, you know, the framework is not perfect, but it's the best tool that we have at the moment in order to explain all these observations at different scales. So you can go to slide number three, actually I'm doing it, so no need to do it. Yeah, so despite all this also national evidence at different scales, we actually don't know what's the true nature of the dark matter particle itself. Unfortunately, there's no good candidate in the standard model of particle physics, and actually the neutrino, the only one that we know in the sixth, is actually excluded just because it cannot account for the total matter content that we measure. And of course, if you go beyond the standard model of particle physics, a huge plethora of possible candidates arise. This plot is just an example of this. Essentially, it doesn't really matter the kind of formative space that we have, but in the exercise you have the mass of the matter particle, in the y-axis you have something which has to do with the kind of interactions that you expect from these particles, and then you see that essentially, you know, you can have a lot of possibilities and that spread over orders of magnitude, actually, that come different by orders of magnitude. But we still, at least we know, you know, some of the requirements that this dark matter particle should fulfill from the cosmological and astrophysical point of view, mainly. They must be neutral, otherwise, of course, we should have detected them already. They have to be stable, or at least long live enough to be present today since the early universe, right? They must be cold. Again, this is mainly because we need them, this cold matter particles, in order to explain their latches' structure. And of course, you must reproduce the measure of the matter amount that we are measuring. And I highlight this fourth point just because as you will see later on, Andrea actually will be reporting a lot of our limits, right? And we are starting to touch the relevant part of the parameter space where we are actually proving testing this fourth point for the first time. Okay, so next slide, just trying to focus a bit on all these discussions. Of course, there are different ways, different approaches to try to detect this dark matter particle, at least to know more about the nature of this dark matter particle. I will be focusing here on the indirect detection technique, and in particular, we will be talking about gammas. Of course, you know that there can be also other annihilation products, like neutrinos, positrons, and so on. But again, we will be talking about gammas here just because we want to report on the Fermilab results, right? And there are good reasons to focus on gammas. I don't want to go really into the details of this, but essentially, you have a good knowledge on the industry spectrum, just because in principle, you don't, these gamma rays do not suffer from attenuation. This is true at least in the local universe. Then also, of course, you have gammas just because in the preferred dark matter particle physics models, the mass of this dark matter particle is typically in the GB to TB energy range. So essentially, this gives you, in another way, gammas, of course. And there's also a very good thing, which is the fact that the gamma rays just travel following straight lines, so you can track back these gamma rays and know the meeting source. This is something that you cannot do, for instance, with positrons or charged particles, right? Okay, so now talking about the kind of gamma reflux that you may expect in your telescopes, above a particular energy threshold. Essentially, it's essentially given by the product of two terms. On one hand, you have what we call the J factor term, which encloses all the statistical considerations, essentially, you know, information about how the matter is distributed in the object, for instance. And, you know, if you have some kind of structure there, and so on, so all the statistics is enclosed by this guy. And then the second ingredient here, the second term is the particle physics term, which essentially glutinates all the particle physics considerations, like the mass of the matter particle itself, the cross-sectional interaction, the channel, where you expect these simulations to occur, and so on. So, of course, if you take a look at the J factor, since you have simulations of the matter particles, this guy is proportional to the square of the matter density along the line of sight, so you have to integrate all the matter density distribution to the square, right? And this gives you a good hint about where to search. So essentially, you want to point your telescope just to analyze data from those places in the universe where you expect the largest concentrations of that matter, like the galaxy center, dwarf galaxies, nearby dwarf galaxies, local galaxy clusters, and so on, right? Because also we are dealing with fluxes, so we need the sources to be close. Okay. Then, of course, there are different matter search strategies that one can envision. So each of these has, you know, I would say its own pros and cons. And so, again, I don't want to spend too much time on this, but, in principle, you know, you can take different targets into account, like the galaxy center, which should be, in principle, the brightest objects, object in the gamma-ray sky for the matter inhalations. You can have different things here, like, as I said, galaxy clusters. You can also look for spectral lines and so on. But in this talk, we are going to talk about these dark matter satellites. Essentially, you can have, well, you expect, in Lambda CDM, you expect a good number of satellites orbiting the Mickey way. Some of them will be hosting gas and stars, and therefore, we can see them in other wavelengths, right, like in optical and so on. In some other cases, we have to expect them to be completely dark, why not? And maybe the only way to detect them will be in gamma rays due to annihilations. But in any case, this is something that we are not going to touch here today, so we are only going to focus on the dwarf galaxies that we know exist, and, as I said, host gas and stars. Okay. Then, talking a bit more about these dwarf galaxies, they are actually very good targets for immediate detection. They are the most armored-dominated systems that we know in the universe, so the mass-to-light ratio that we infer from the movement of stars is particularly high, a few hundreds, typically, for the most recent discovered objects, which is great. We know, right now, we know, I would say, like 2,000 of them, more or less, and we are starting to know more. I will talk a bit more about this in a second. Then, you are interested in annihilation fluxes, so it also means that you need objects which are close to us, and they are also, you know, we know dwarf galaxies which are actually quite close from us, like a few dozens of kiloparsecs. That's very good, too. Probably, more importantly than all these, is that they are expected to be free from any other astrophysical gamma resource. Essentially, this means that if you detect some gamma-ray emission coming from one of these objects, you can really start thinking about exotic physics, like gamma annihilation, because in principle, you don't expect these objects to host gamma-ray emitters, astrophysical gamma-ray emitters, that's what I mean. Finally, let me tell you that from the observational point of view, they have been typically detected as a dim stellar over-density, I would say, in a particular region of the sky, using typically optical surveys, and this is the case, for instance, of the 14 ultra-fane dwarfs that were discovered in the last decade using a slow and digital sky survey data. Okay, this is just for completion. I would like to show you where these guys are typically located, so this is in galactic coordinates. For those of you that are not familiarized, this means, essentially, that you have the galactic center in the center of this map, then you have the galactic plane along this horizontal line crossing the galactic center, and then the galactic poles, essentially, are 90 degrees plus or minus 90 degrees in this map. You can see immediately that many things actually lie at high latitudes, and this is also very good, because in principle, you don't expect too much contamination from the galactic plane, many different astrophysical processes that can give you gamma rays in the galactic plane region. That's also very good. The different colors just mean, well, red one corresponds to the old dwarfs, let's say, before 2006, and then the blue ones correspond to the new objects detected by the strong, using strong data from 2006 to 2002%. Okay, so just a brief note about how to measure the damage contained in the dwarfs. What you really need is to determine the stellar velocity dispersion using spectroscopy. So these in classical dwarfs means that you typically have like hundreds of stars that will have good spectroscopic measurements. That's very good, because, essentially, you can infer the underlying damage distribution in a quite precise and accurate way, but in the case of the ultra-famed dwarf galaxies recently discovered by Sloan or by other means, you know, with other surveys, we'll talk a bit more about this in a second. So in this case, you only have typically a few tens of stars, and that gives you essentially, you know, you have a big uncertainty in the damage distribution or the best damage density profile, the parameters of the damage density profile that fits the data. So this is something that you can see, just as an example, something that you can see in the plot on the right for the old classical dwarfs. So the stellar spheroid, I put in the stellar spheroid and we have an arrow there. So this is what you actually expect from just from the visible model, let's say. But then what you see, what you infer is actually a dispersion profile, which remains generally flat up to a large radii. So every dot here, by the way, every data point here corresponds typically to a few tens of stars. So in this case, as I said, for these guys, you have plenty of information if you compare to the ultra-famed dwarfs. So in the case of the Jake factor, I mean, we are interested in the Jake factor, which gives you a measure of the, you know, how bright are these sources in terms of damage and annihilation. So essentially, you have to take this velocity dispersion profile and then assume a particular damage density profile for the underlying damage distribution, and then you come up with an estimate of the Jake factor, which again will be much more accurate in the case of the old dwarfs and less accurate, substantially less accurate for the ultra-famed dwarfs. So these are the Jake factors for the non-dwarf galaxies. So you can see essentially, again, the y-axis corresponds, it's like a measure of how good is this particular target for our search, right? And then something that you realize is that, of course, since we are dealing with fluxes, this typically goes as the square, you know, goes as the inverse of the distance squared, right? So that's the main trend that you can see. Given the robust, that's not, it's not much that you can do apart from that. But Andrea, we will talk a bit more about this in a second. Okay, so I was talking about just like two dozens of these dwarf galaxies at present, but actually, you know, if you go to land a CDM, you expect a lot more of these kind of objects orbiting the Milky Way. So some studies like the one that Elly Ptollerud did with James Bullock and others a few years ago. So they seem to point to something like 500 of these guys inside the billion radius of our galaxy. Some of them, of course, will be maybe, you know, not visible at all, or maybe they are too far to be detected by, you know, by our surveys and so on. But still, you may expect many of them, much more than two dozens to be detected in the near future. And I would like to remind you that we don't have a complete coverage of the whole sky with any of the surveys that we have at present. And this explains actually why we only have two dozens, essentially because we didn't cover the whole sky. We don't have data from the whole sky, right? Not yet. And this is something, this lets me to the next thing. Essentially, as I said, people discovered 14 ultrafine dwarfs using Sloan data. So here on the top, you can see a map of the portion of the sky that was mapped by Sloan, right? As you can see in the gray area, the gray area corresponds to the regions that were not mapped. As you can see, there's still a huge portion of the sky that was not mapped by Sloan. And then, you know, the idea here is that if you map these uncovered areas with enough sensitivity, you may expect some new ultrafine dwarfs to be detected, right? And this is something that people recently did using the DACA energy survey. So you can see at the bottom, you can see the coverage of this particular survey. Essentially, they will cover 5000 square degrees of the something hemisphere in five years. And according to these estimates from some people, you may expect something, a discovery of between 5 to 20 new dwarfs, more or less. So they did the first release just a couple of months ago. So it's called the DSY1A1. And it's essentially 1800 square degrees that you can see in this map here. And the cool thing is that they actually detected a bunch of new candidates for further galaxies. So you can see actually where these galaxies are located. I put in the title eight or nine, just because there's a work that was submitted in parallel to the DS Collaboration Official Work, where the authors were actually reporting on the discovery of nine satellites, while the DSD Collaboration only reported officially on eight. But this is again, this is a field which is evolving very rapidly. And you may expect a lot of more of these kind of objects to appear to show up in the next year. Okay, so this is just a table showing the kind of characteristics that these candidates for the galaxies has. So each system is, in principle, identified just as a statistically significant over-density of stars. And then what you typically do is to perform the so-called HR diagram that you can see on the bottom, right? So it's essentially you plot the color versus the magnitude in some particular bands. And then you infer from that the stars that have a common origin. And typically, in this case, in the case of dwarf galaxies, they exhibit very, very low metallicities, which correspond to a very old stellar population with a common origin, as I said. Okay. And then there's still one more discovery that was reported also very recently by a Panstar's collaboration in Triangulum. In this case, we are talking about an object, which is also very close. It's something like 30 kiloparsecs. This kind of distances you can infer from this color magnitude diagrams I was talking about. And in principle, the properties of this guy should be also quite similar to the ones of Willman, Sege, Sege II, Votes, essentially the ultra-framed dwarf discovered by Sron. And then, okay, as I said, Andrea will be now reporting on the analysis that the Fermilac collaboration did on these objects. So I stop here. Okay. Okay. Does everything look okay? Sure. Cool. So let's keep this train moving. So like Miguel said, if we see a gamma ray signal from these dwarf galaxies, that's considered a smoking gun for exotic physics like dark matter because there's really no standard astrophysical processes happening in these dwarf galaxies that are expected to give gamma rays. And so we can look for gamma rays with the Fermi Largeria Telescope. This is on board the Fermi Gamma Ray Space Telescope, which has been in orbit for over six years now. And this has been a beautiful instrument that has just opened a new window on the gamma ray sky and we've learned so much. And it's also a wonderful instrument for following up with these new discovered dwarf candidates because the Fermi is a survey instrument. So it has already been observing these objects, but we just didn't know where to look in the data before. And so Fermi will see the entire sky every three hours when it's in standard survey mode. And so the Fermilac collaboration is a team of about 400 international scientific members and all of our data is public. And so we do have a new data release that will be made public very soon called Pass 8. Currently the public data is using the reconstruction and event classification scheme we call Pass 7 reprocessed. And I just want to motivate here the improvements with Pass 8 because this is the data set that the LAT and the DES teams used in our analysis that I'll be discussing later. And I think the bottom line for dwarf search is that in Pass 8 our point source sensitivity improved by about 40% in the relevant energy ranges and this comes from improving both the point spread function and also the acceptance or the effective area. And like I said that will be made public very soon. And so we have performed a search on those previously known before you know pre DES dwarfs that Miguel mentioned and this paper has also come out on the same day as the paper we did with the DES collaboration. And so the results shown on this slide here are from the pre DES dwarfs that have pretty well measured J factors. And the bottom line here is we searched we did a joint analysis using 15 dwarfs we did not see any significant gamma ray emission and therefore by doing a joint likelihood we're able to take the combined power of a null detection in all 15 of these objects to derive very very robust and also constraining limits. And that's what's shown on the bottom. And so you can see our solid black line is our current limits from the most recent Pass 8 analysis using these known dwarfs. And the dashed horizontal line that slightly increases at lower dark matter masses is that thermal relic cross section Miguel mentioned at the beginning. And so I remind you that if you just assume a weak scale particle annihilating with this cross section, you get out the observed relic abundance of dark matter in the universe. And so I always remind everyone this is an incredibly well motivated hypothesis for indirect dark matter searches. And we are testing this hypothesis and for me it was really the first experiment to hit the thermal relic threshold with our sensitivity. And so you can see here that we in this particular plot we're excluding cross sections at or below or yeah at the thermal relic cross section for masses below 100 GeV and we're starting to cut deeper in at lower masses as well. And so we haven't seen any dark matter but we are starting to test very viable dark matter hypotheses with the combined dwarf analysis. And this is particularly interesting if you want to use it as an independent check of other dark matter claims. For example there's an excess of gamma rays around a GeV in the galactic center that many people have interpreted to be from dark matter annihilation. And so the cool thing with the dwarfs is you would expect the same dark matter in the galactic center as you would expect in the dwarfs. So if you see a signal in the galactic center then you would expect to also see a signal in the dwarfs. And mind you the individual dwarfs are going to be fainter than the galactic center because their chief factors are smaller. However when you do the combined joint analysis this gives you additional power. And so what's shown here is our current limits from the dwarf analysis and then also shown on this plot are the contours from recent papers claiming a dark matter interpretation of the GeV access. And so what you can see is that our limits from the dwarfs are starting to challenge some of these signal claims of the excess in the galactic center. And so the dwarfs are an incredibly important independent check of a dark matter interpretation of the galactic center access. And the sensitivity of the dwarfs is going to improve in the future and so they will be an important check to help make the situation in the galactic center more clear. So one of the ways that we are going to be able to improve the sensitivity of the dwarf analysis is by finding new dwarfs. And like I said Fermi has already been observing the entire sky. And so when we find new dwarfs we're going to be able to use the entire exposure of the Fermi-Lat over it's over six year mission so far. And so what we did is we had an agreement with the DES collaboration so the Latin, the DES collaborations collaborated and they told us where these new candidates were and then we could follow up with the gamma ray analysis quite quickly. And I will point out that I try to be very careful to call these objects that were found in the DES candidates until we get spectroscopic data telling us the stellar kinematics that we need to determine if these are in fact dark matter dominated and what their J factors are until that time these are simply candidates. And so we have reticulum two has been confirmed as a dwarf serial galaxy but the other seven I think at this point are still candidates. Anyway so what's shown here are the counts maps above a GEV from the eight candidates that we searched in the their right snack in the middle and what's sort of maybe obvious by eye but then also what we found in a statistical analysis is that there's no significant gamma ray emission coming from any of these objects. We did see our largest positive deviation from our background model in reticulum two which was the dwarf galaxy that was confirmed to be a dwarf galaxy and is also rather close at about 32 kiloparsecs. The test statistic in our search was 6.7 and then the best fit point was for a mass of 25 GEV for dark matter annihilating into a pair of teleptons. And so then since we scanned in mass and channel you do need to apply a trials factor to convert that TS into a p-value or significance. And so when you take into account the scan in mass and channel that TS corresponds to 1.5 sigma and then if you also take into account the fact that we looked in eight different dwarf candidates that significance goes down to 0.26 sigma. And so clearly not a significant detection which is why we set upper limits. Now in order to set upper limits we need a J factor and without spectroscopic data we don't have a measured J factor. However what we did in this paper is we did a very simple first order estimate of what we could expect the J factor to be given the distance. And so what we noticed like Miguel said is with these dwarfs that we've been using in previous line analyses they seem to follow a 1 over D squared distribution. And so what we did is we took those dwarfs we fit a line to it and so then from this you can sort of get a from the distance you can get an estimate of the J factor making the assumption that the candidate that you are talking about is one a dwarf and then two has similar properties to the dwarfs that we already know about. And so like I said this is lots and lots of caveats to this assumption. However you need to make this assumption in order to interpret the flux limits from these objects in a dark matter context. And so when you make that assumption these are the limits that you could expect to get from these new candidates given our assumed J factors. And then also the dashed red line shows what you could get if you combined all of those predicted limits. And you can see again that you get an increase in the power by combining these objects instead of looking at them individually. And additionally the closest objects like reticulum 2 are the ones that are primarily driving these limits. And so on March 10th which I dubbed Dwarf Day, happy dwarf day everyone, there was an additional paper by an independent group that was also checking out these new dwarf candidates and they reported an excess. And in reticulum 2, which again I remind you is the dwarf where that's the closest and therefore you expect to have the brightest J factor from this set of candidates. And so the two plots that I have up here, the top one shows the significance as a function of mass in all of the dwarfs so they took the dwarf candidates in DES and also the known dwarfs and they did a whole search. And a point that I like to make here is that in this analysis they were seeing negative 3 sigma fits, which to me, I guess I'm very conservative, the negative 3 sigma fit suggests that your sort of systematic uncertainty from your background modeling uncertainties is around the 3 sigma level. And so for me the way I interpret this is I say well if you're getting negative 3 sigma fits then I am less confident about a positive 3 sigma fit truly being a real signal, however I realize that that statement is not universally agreed upon. And so anyway they were using pass 7 reprocess data because like I said pass 8 is not public yet and really the big difference between their analysis and our analysis is that with pass 7 reprocessed after taking into account the trials factor from scanning and mass they had 2.3 sigma as their significance and that is best compared to our significance of 1.5 sigma using pass 8. And so really the big difference here is pass 7 versus pass 8. I remind you that pass 8 we expected the point source sensitivity to improve actually we know the point source sensitivity will improve by 40 percent so really the bottom line here is that with a more sensitive data set the significance has gone down from 2.3 sigma to 1.5 sigma. And so this is something to keep an eye on but at this point I would say that it's not looking like there is a very strong convincing signal coming from reticulum 2. And then there was an additional paper by Dan Hooper and Tim Linden that was also confirming that they see an excess in reticulum 2 and using a different analysis they had a local significance of 3.2 sigma and everyone seems to agree that the about the properties of this excess is happening for a dark matter mass around 25 GeV for an annihilation in TAS. Alright so like I mentioned we need spectroscopic data to confirm that these are in fact dwarfs and we have this for reticulum 2 now so it is confirmed new dwarf discovered by DES which I think is amazing. And then using the spectroscopic data different groups have been able to actually measure the J factor of this object and so then you can say well depending like Miguel was saying depending on the dark matter profile assumptions that you make you can conclude that the J factor is somewhat different and what's shown here are the different J factors that are determined by other groups but I will point out that within error bars these reported J factors are pretty are in line with our simple 1 over d squared estimation that we had for reticulum 2 so we weren't completely far off with that. And so then well what else is next? We like Miguel said this was only looking in year 1 of the DES data there is going to be year 2 available quite soon and so hopefully we'll find even more and then DES is an optical survey in the Southern Hemisphere but we have an even more powerful one that will cover more sky coming up called LSST and so we can really I think we can be optimistic to find a lot more dwarfs with these Southern Hemisphere optical surveys and what's shown on this plot here is the expected limits for the Fermilat dwarf analysis using 10 years of exposure and 3 times 15 so 45 dwarfs and you can see and then also some signal contours from the interpretation of the galactic center accesses dark matter and so you can see that with more dwarfs and with the extended exposure from the Fermilat we're really going to be able to cut deep below the thermorella cross section and really start to test a lot of these models that people are claiming for the galactic center access so I think that's awesome so just it's been we've released many dwarf analyses but we're there's still reason to be optimistic and and expect them to just keep getting better and they're an amazing clear test of dark matter like I said these are seeing dark matter or seeing gamma rays from these guys would be a smoking gun they're in relatively clean regions of the sky so this really is a really robust dark matter search that'll just keep getting better and so that's the end and I guess at this point we go on and take questions and Miguel and I thank you for your attention so thank you both of you guys for this interesting seminar now we should pass to the round of questions so remember that you can ask questions to Andrea Miguel via the Google Q&A system at the webinars page and via our Twitter with the hashtag LAWP so I'd like to do their questions so we have a question here I have many can you hear me yes please go ahead okay my first one is how are these new dwarfs compared with the angular resolution of Fermi in the sense that do you expect that at some point it's gonna appear kind of miracle dwarf galaxy that Fermi can resolve in a better way than the previous dwarfs that are already analyzed I don't know who can answer who can answer if Andrea or Miguel Angel yeah I can take that one no problem yes so you okay so you are asking essentially about how the the PSF of the instrument right the angular resolution of the instrument compares to the kind of annihilation the extension of the nation flags in the dwarf right and so I would say you know it's quite unlikely that we are going to solve this with Fermi because essentially you know 90% of the relation flags for instance for an FW profile is coming from the region within the scale the so-called scale radius and the scale radius sustains an angle in the sky of typically 0.2 to 0.3 more or less degrees right so essentially you know we just everything it's just within the PSF of the instrument and of course we pass it we have a better angular solution but still I would say you know like generally speaking I would expect these ultra faint dwarfs and the new ones to come to be you know very small for Fermi and essentially we will be integrating the whole profile the whole nation secret in these objects but that's actually that's actually good in my opinion because essentially you are not sensitive the you know you are not so sensitive to the kind of annihilation the kind of that matter this is the profile that you assume for these objects but it doesn't really matter if you have a core or a cast profile essentially when you integrate the whole signal within the scale radius it doesn't really matter you will have you have essentially the same total annihilation flags so that's you know in some sense that's that's actually good okay thank you I have a another one that is more or less related with the same stuff is kind of how this dwarf history I mean the how they were merging with the with the dark matter halo of the Milky Way may affect this kind of analysis in the sense of to compare what is happening in the galactic center with what could happen in the dwarfs let's see how if it is some caveats or some problem in which to compare both measurement in the same annihilation projection versus mass of that matter for wings as I can say something about that so there is a a point to be made to take into consideration when you compare the galactic center with the dwarfs is that the the dark matter density profile distribution in the galactic center has much much larger uncertainties than in the so you'll remember that the j factor goes as the integral density squared and so how cuspid things are in the galactic center if you get a larger increase towards the the center versus a more chord profile then the cuspier profiles are going to give you a brighter predicted gamma ray signal and so then when you interpret your your flux limits or your flux signal in terms of dark matter you have to make some assumption about the cuspiness in the galactic center and so we show the signal contours that have been published when we compare them with our dwarf limits however there has been work that's been shown that those contours can move down and escape our limits if you consider different levels of cuspiness in the galactic center and the levels of cuspiness pretty much come mostly from n body simulations not from actual observations now for the dwarfs the j factor is much more robustly determined since it's based on actual stellar kinematics from the dwarfs themselves and so the dwarf limits I'd say are rather robust and the signal contours for the galactic center excess interpreting as a dark matter should probably be somewhat larger taking into account uncertainties in the cuspiness in the galactic center okay thank I have a question actually with respect to j factors so so the j factor usually are estimated using these stellar kinematics but for these new dwarfs you're just extrapolating the using the the other j factor for other non dwarfs right for most of them with reticulum 2 at this point we do actually have a measured j factor for reticulum 2 yes but once you can you measure the this stellar kinematics with new dwarfs how much this could could be improved the j factor the uncertainty of the j factor right and that's actually still a lot of work that's being done and maybe Miguel can speak more to this but there there is a lot because you I think people will mostly agree about the dark matter content where you actually have the stars but sort of extrapolating that out beyond the stars requires an assumption on your dark matter profile and that assumption can lead to different j factors that come from the measurement so there is some uncertainty on the j factors even though they're in the in the dwarfs as well yeah there's this actually a some kind of the generality between the isotropy of these stellar you know velocity dispersions and the kind of client that you are so assuming and it's difficult to break us actually this kind of generality but I will say you know I will say that of course you know we were quite conservative in our analysis as Andrea was mentioned we actually assumed a point for the x for the ascended d and you know in in our day factor estimate and this is perfectly in line you know with the kind of j factors ascended this that we did that we are measuring known to us so in that in that regard I think we are you know we did a safe you know we took a safe approach so yeah thanks so are there more questions yeah I have another question also please go ahead one is I mean I guess also Andrea she showed some improvement in the sense if we want to to kind of exclude our one TV WIMP how long we have to wait I mean or how many dwarf galaxy we have to to I mean has to be found in order to to push the bounds so so much to higher energies and a second question just to take advantage of the time in the microphone how does he stand this same analysis for decaying that matter instead of annihilating WIMPs right so the to answer the question like how long will it take us to probe a TV with Fermi I think that probing a TV mass with Fermi can will be quite challenging you can see that even with our predictions with 10 years and 45 dwarfs we're hitting a few hundred GB and so barring some very very clever idea to dramatically increase the sensitivity of our of our dark matter searches with Fermi I think that our best shot of getting a you know testing the TEV thermorella cross-section will be future gamma experiments like CTA the Trenkov telescope array which is going to if you look at the Hess limits which I think are somewhere in one of those plots CTA is going to improve the sensitivity by roughly a factor of 10 and you can see if you extrapolate the Hess limits to a factor of 10 we hit thermorella around a TV and so with CTA is going to come on board in the next few years but then we'll have to make sure that the exposure in the galactic center in order to test a dark matter hypothesis needs to be quite deep with CTA and so depending on how people prioritize the dark matter search there I hope we won't have to work way too long I'm actually quite excited for CTA in order to probe a TEV mass thermorella myself and I can't really think of another way to do it in the near future well I can maybe I would like to do I can add something to to that discussion and so if we restrict ourselves to what you can do with dwarfs then I mean I fully agree with Andrea of course and because that's the reality right but there are actually other dark matter proofs that you can use with Fermi to reach you know to be actually more competitive in the TV regime of sub TV regime and so you are starting to be better somehow with other approaches compared to what you can do with dwarfs and I'm referring basically to the what you can do with cosmological dark matter annihilation so in that case you are using the isotopic gamma ray background measurement by the Fermilat and so you essentially have information and data almost up to the TV so that actually allows you to put registering and a very competitive limits using Fermilat data in the sub TV and TV energy regime much better actually than the dwarfs and yeah but again if we restrict ourselves to what you can do with dwarfs only dwarfs then yeah I fully agree that the future then is cta and ground based telescopes and I believe the second question was asking about the decay results and so all of the annihilation results are based sort of intrinsically on a flux upper limit and so using a J factor where you integrate over density as opposed to density squared then you can sort of use that to rescale and derive for a lifetime upper limit I guess for me my personal experience with the lifetime of limits comes from the line analysis that I've worked on where for example lines I think are some of the best slash possibly only ways to test something like a gravatino where it decays to a gamma ray and a neutrino gives you a monochromatic signal and so we do also produce limits on on dark matter decay although as far as I as I know the dark matter didn't hey doesn't really have this sort of benchmark uh level like the thermal relic cross section in the annihilation case which I think is why people focus on that yeah in this case Roberto I mean it's better to go for galaxy clusters for instance right because you have a linear dependence with the mass uh the matter mass and then you you prefer actually to go for most massive objects essentially so yeah even if it is a flux it doesn't really matter galaxy cluster will be you know will be actually more competitive that's my guess and I guess that because there's big uncertainties with annihilation and galaxy clusters from um a substructure uncertainties right so you don't really know how much substructure there is and if there's really really dense if there's lots of really dense uh sub halos then you expect to see a brighter dark matter signal but those uncertainties are um relaxed somewhat when you do the integral over density linear as opposed to density squared yeah so then that would also be really a good news for for galaxy clusters is being a more robust probe of dark matter decay okay so you can see a comment coming from team linden so I don't know if you guys can see it so if you say that just a quick comment I'm not sure that I comparison between that one point five sigma xs in the Fermi past analysis and the 2.3 sigma xs in the Gehringer summit that all analysis is straightforward since they are not assuming the spectral trials factor that's right so right the Granger summit um 2.3 didn't take into account uh the scan in channels um but the if you look at the dn de uh the expected gamma ray distribution for say a bb bar or a tau channel they overlap quite a bit and so um and I don't know the numbers exactly but if you folding in the the channel component um is really not going to give a big big change in the significance because of that overlap right we can't treat the different channels as independent that would be way overestimating your trials factor um and so I agree that they aren't directly comparable because also both of those significances uh took a data-based way to convert the ts into a p value so in a perfect world where the model perfectly described background model perfectly describes your your data then having one degree of freedom difference between you know looking for a signal or not looking for a signal then your your significance is simply the square root of the ts so then you take our 6.7 square root of that you know that's two point something and we're claiming 1.5 and this is something we this is a detail but we something we discuss in in the tort paper is that we can tell when we look in blank sky fields that our ts distribution from the blank sky fields based on the data is different than you expect from like a 1 over chi squared and we think that this is due to things like unresolved sources uh that are not being put into our background model and so gringo smith all also use a data driven um approach to rescale the ts to a significance uh but they used the area more around reticulum to itself as opposed to a large scale random sky thing like we did so anyway like i say that there are some details that make those two numbers not directly comparable but i still think that roughly those are quite comparable and we did um look and rerun our analysis using past seven reprocessed the data set that they used and we came up with a similar significance to that 2.3 sigma using this the past seven and so then it goes down to 1.5 sigma and past eight i think that's this the bottom line okay cheers there's some other comments uh again by team so he's saying i think both are warned depending on the question you're asking if you ask what is the significance of a dark matter signal in the dwarf your you take a trial factor hit if you ask the dwarves confirm or ruled out the galactic galactic center you don't right i don't know if you want to comment something well with the galactic center i guess if you say does it rule out the teeny tiny contour that was published or that was posted in the dial-in at all paper uh our 95 percent confidence limits in the bb bar channel do roll out the contour presented in that work however i think what tim might be alluding to is that the contour in that work can move up or down depending on what you assume about the local dark matter density or the cuspiness right so that i would say that there is an uncertainty that is that should be folded into that contour that isn't right now and so i would i would like really very much like to see um an updated work with the the signals in the the claim signals the galactic center with the dark matter interpretation that folds in the uncertainty on the cuspiness in the contour itself so that it'll be more easy for us to compare with the dwarf limits okay yes and the next comment say that he completely agrees the numbers for both analysis very comparable yeah okay so i are there more questions comments so uh so let's let me thank andrea miguel again and all of our viewers then we'll meet next week for another lefie america webinars on physics so thank you all guys thank you so much yeah thank you it's fun what's down thank you okay see you ciao