1 00:00:02,910 --> 00:00:08,010 I would say one of the most remarkable stories of my my own. 2 00:00:11,050 --> 00:00:21,940 Experience, professional lifetime experiences has been the transformation of cosmology from a subject which was 3 00:00:21,940 --> 00:00:29,890 described in once by John Peoples as an area that really wasn't fit for grown adults to work in, 4 00:00:30,760 --> 00:00:44,620 to becoming, have become a remarkably precise and extremely important domain of theoretical astrophysics in some ways, 5 00:00:45,040 --> 00:00:48,670 perhaps even sort of the crown jewel in the subject of cosmology. 6 00:00:49,600 --> 00:00:55,150 That description I can remember actually as a graduate student in Berkeley. 7 00:00:55,540 --> 00:01:05,619 Some of the early microwave background experiments showing grotesque distortions from simple black bodies and people taking it seriously and putting 8 00:01:05,620 --> 00:01:14,470 considerable amounts of efforts into explaining what enormous source of energy could possibly have knock the universe out of kilter in this way. 9 00:01:15,310 --> 00:01:18,580 In any event, we all know those measurements were in error. 10 00:01:18,880 --> 00:01:26,550 The post-Soviet era paralleled with a whole new dawn on the subject. 11 00:01:26,560 --> 00:01:35,440 The subject really begins after to establish that the universe, the microwave background is nearly perfect. 12 00:01:35,440 --> 00:01:37,450 Blackbody at 2.7 degrees. 13 00:01:38,470 --> 00:01:50,170 So one of the reasons why this subject is so interesting and has become so rich is that I think this is probably David Spergel, who puts it this way. 14 00:01:50,290 --> 00:02:01,750 The universe has had its baby picture taken by looking at the time and the ratio of the time of recombination, the cosmic microwave background itself. 15 00:02:03,160 --> 00:02:10,750 It is possible to, by studying its intricacies and its details and its statistical properties, 16 00:02:10,750 --> 00:02:17,560 there's an incredible amount of information that can be extracted from that source of data, 17 00:02:19,090 --> 00:02:24,580 but only to those still in the methods, by no means straightforward. 18 00:02:25,270 --> 00:02:35,290 And that brings me to today's speaker, Joe Donnelly, who is one of the most important leaders in this field. 19 00:02:36,070 --> 00:02:46,600 Joe was an undergraduate at Cambridge who graduated in 2001, came here to Oxford as a postgraduate postgraduate worked with Peter Ferrara, 20 00:02:47,050 --> 00:03:02,290 took a degree in 2005 and after three years of postdoctoral work with the team at Princeton has been on the staff pretty much continuously since 2008. 21 00:03:03,820 --> 00:03:13,510 Joe, as I say, is one of the great leaders in this field and she's been widely recognised for her work and for her 22 00:03:13,510 --> 00:03:24,880 ability to extract useful information out of this swarm of photons that form the microwave background. 23 00:03:25,840 --> 00:03:28,850 Her work has been widely recognised. 24 00:03:28,860 --> 00:03:40,390 She won the Maxwell Prize and the Fowler Maxwell Medal and the Fowler Prize and as part of her work on W, 25 00:03:42,040 --> 00:03:48,760 she was one of those recognised with the Cosmology Gruber Prize in 2012 as well. 26 00:03:49,630 --> 00:03:59,560 And we're very fortunate to have her today and she is going to tell us about her work cosmology from the microwave background and thanks to you. 27 00:04:06,150 --> 00:04:11,490 So thanks for that very kind introduction. Yeah. I want to tell you about the work that that I've been doing and we're doing here in 28 00:04:11,490 --> 00:04:15,720 Oxford on the microwave background and some experiments that we're involved with. 29 00:04:16,410 --> 00:04:19,350 You can't do good work without amazing students and post-docs, 30 00:04:19,350 --> 00:04:23,940 and I'm lucky that we have many of them here and many of them in the audience and you know who you are. 31 00:04:25,890 --> 00:04:34,500 So, okay, so it's this year is actually the 50th anniversary of the discovery of the microwave background, 1965. 32 00:04:34,740 --> 00:04:42,180 This this faint light was picked up by China Penzias and Robert Wilson in New Jersey when they weren't even looking for it. 33 00:04:42,300 --> 00:04:51,720 They saw it by chance. They saw something that they that they they saw in the empty sky this faint microwave signal that turned out to be light. 34 00:04:52,380 --> 00:04:56,220 The earliest light we have of all. Now, here's a picture of it. 35 00:04:58,140 --> 00:05:01,650 This is a map. This is a picture of the cosmic microwave background. 36 00:05:01,980 --> 00:05:07,050 This is light that's been travelling since 380,000 years after the big bang. 37 00:05:07,650 --> 00:05:12,450 In the hot, big bang model, the universe has been growing, expanding, 38 00:05:12,450 --> 00:05:23,579 cooling down and transitioned at this particular epoch from being this hot, ionised plasma where we have this this fluid of photons, 39 00:05:23,580 --> 00:05:33,570 electrons and dark matter particles around transition from being ionised to being perfectly neutral, where before this epoch, 40 00:05:33,870 --> 00:05:41,580 photons would be bizarrely busily scattering the free electrons in this plasma without being able to travel freely through space. 41 00:05:42,960 --> 00:05:48,240 But as the universe cooled down, it hit a particular temperature that at that point, 42 00:05:48,870 --> 00:05:54,180 from that point onwards, photons could now travel freely through space because atoms were now neutral. 43 00:05:54,390 --> 00:06:03,330 Hydrogen atoms are formed. Helium has been formed. Dark matter particles continue to be around that they don't affect the photons anyway, we think. 44 00:06:03,990 --> 00:06:10,300 And photons could then travel freely through space to us and we get to catch capture some of those photons. 45 00:06:10,530 --> 00:06:16,830 They're hitting us now, today. They've been travelling for the full age of the universe with very little interruption. 46 00:06:17,580 --> 00:06:21,300 And this map here is made. This is the more exciting one I'll show you. 47 00:06:21,350 --> 00:06:25,470 See, it was made by the Kobe satellite in the 1990s. 48 00:06:25,710 --> 00:06:31,770 Of its average temperature. The average temperature of this light. And on average, you can see the orange ness of it. 49 00:06:31,800 --> 00:06:37,470 This is a map of the full sky of the CMB unwrapped onto the flat, the flat oval. 50 00:06:37,770 --> 00:06:45,299 It's all 2.7 degrees Kelvin. And the fact that it's all the same temperature and its distribution in terms of frequency is 51 00:06:45,300 --> 00:06:51,390 a perfect Blackbody was the nail in the coffin of any model but the Big Bang 50 years ago, 52 00:06:51,390 --> 00:06:55,200 and then even even more so when Kobe came out with its results. 53 00:06:56,400 --> 00:07:00,600 So it it's existence tells us a lot, 54 00:07:00,900 --> 00:07:08,640 but the details in it tell us even more and the details in it what we've been measuring over the last 55 00:07:09,240 --> 00:07:16,440 20 years in greater detail after Kobe came in terms of satellite missions to the sea and became a map. 56 00:07:17,130 --> 00:07:20,130 But then most recently, we've had the Planck satellite. 57 00:07:20,940 --> 00:07:27,690 So the Planck satellite is a European mission, European led mission that launched back in 2009. 58 00:07:28,440 --> 00:07:35,069 It's pictured here before launch that the Planck satellite with this it's about a person is about 59 00:07:35,070 --> 00:07:41,340 this size and the mirror is in there and it's about almost two metres or about my arm span. 60 00:07:42,600 --> 00:07:52,140 So it's a big it was a big beast sent up sent up from French go out to this point L2 a million miles from here 61 00:07:52,410 --> 00:07:58,920 where it could quietly and calmly observe and observe the sky and pick up the microwave background signal. 62 00:08:00,390 --> 00:08:07,530 And it did so with greater sensitivity and better angular resolution than any of our previous experiments. 63 00:08:08,370 --> 00:08:10,140 So we've been involved in that here in Oxford. 64 00:08:10,830 --> 00:08:17,940 And it's a it's a significantly large collaboration with many people in the UK, many people in Europe and internationally too. 65 00:08:18,120 --> 00:08:22,680 It's a it's a large a large body. I mean, we showed you some results from Planck, 66 00:08:23,010 --> 00:08:28,680 but at the same time we've also been heavily involved in this experiment called the Atacama Cosmology Telescope, 67 00:08:29,340 --> 00:08:34,320 which is, we think, the highest experiment, high telescope on earth. 68 00:08:34,920 --> 00:08:42,780 Someone may have just put one slightly higher. It's on Cerro Toco in northern Chile right by the Bolivian border and it's 69 00:08:42,780 --> 00:08:48,030 pictured here hidden inside a ground screen that shields it from the mountain. 70 00:08:49,170 --> 00:08:58,620 And inside there we have a six metre telescope designed, custom built to look at the microwave sky to measure the CMB light. 71 00:08:59,340 --> 00:09:05,130 And inside it is a very sensitive camera that's pictured here. 72 00:09:05,360 --> 00:09:10,880 And I'll show you in a bit more detail later on. We had this large group in Oxford to work on it. 73 00:09:12,020 --> 00:09:16,790 Their name, Zahir, and he he'd lead a lot of the analysis for this project. 74 00:09:17,930 --> 00:09:24,450 So the sort of be the observation, doing analysis and doing observation to the C and B means you either are in space 75 00:09:24,450 --> 00:09:28,070 for the satellite or you do your observations as best you can from the ground. 76 00:09:28,550 --> 00:09:33,980 And the best locations we found so far from the ground are northern Chile, where it's incredibly dry. 77 00:09:34,340 --> 00:09:38,420 So up here, we've got you quantify it in terms of perceptible water vapour, 78 00:09:38,690 --> 00:09:48,200 how much water vapour you'd have if you compressed all the water vapour above you into a into a small into water? 79 00:09:48,350 --> 00:09:55,850 And this is point five millimetres. It's really dry, but some of it snows, which is annoying. 80 00:09:56,070 --> 00:10:02,690 And so to there and the South Pole, the optimal places for observing this microwave light, you want it to just be as dry as possible. 81 00:10:04,610 --> 00:10:13,790 Okay. So using Planck, we've now been able to zoom in and make that orange orange map more interesting. 82 00:10:15,200 --> 00:10:21,080 What I'm showing you here is, is the measurements by Planck of the microwave background, the fluctuations about the mean. 83 00:10:21,440 --> 00:10:24,919 So I've subtracted the mean and showing you just show you the deviations about. 84 00:10:24,920 --> 00:10:30,740 That's where the colour scale here is hundreds or up to 100 micro kelvin plus or minus. 85 00:10:32,300 --> 00:10:38,600 So the, the little, the little features you see are typically one part in a 100,000th of a degree. 86 00:10:39,020 --> 00:10:43,429 So the very small. And so it's taken these sensitive instruments to map them out, 87 00:10:43,430 --> 00:10:49,400 although we have seen them previously with this kind of progression of ground based, balloon based and then satellite missions. 88 00:10:50,600 --> 00:11:00,830 So what we're seeing here is the fluctuations in the light where you got the blue is ever so slightly colder than average and the red is hotter. 89 00:11:01,790 --> 00:11:08,329 And these little features in the temperature of the light show us reveal the features in the universe. 90 00:11:08,330 --> 00:11:13,670 At 400,000 years, we're seeing a picture of part of space, as it was at 400,000 years. 91 00:11:14,810 --> 00:11:22,340 And the temperature of the light is tracing out the underlying features in the universe, the density of the underlying space. 92 00:11:22,910 --> 00:11:28,010 So back at that time, the universe was almost completely featureless. We didn't have the large cosmic structures. 93 00:11:28,010 --> 00:11:31,820 Things hadn't collapsed and formed large gravitationally bound objects. 94 00:11:32,060 --> 00:11:39,950 We had an almost completely smooth universe with these tiny density fluctuations about the average, about the average density. 95 00:11:41,210 --> 00:11:45,590 And we're tracing them here with the temperature of the CMB light because on average, 96 00:11:45,590 --> 00:11:54,080 while on the large scales where you have a slight, where you had a slightly denser region in the universe at 400,000 years, 97 00:11:54,620 --> 00:12:02,480 that slightly denser region actually turns out to be a cold spot in the CMB because that denser region will have more light there, 98 00:12:02,690 --> 00:12:06,410 but it has to climb out of a deeper potential well to get out to us. 99 00:12:06,950 --> 00:12:11,360 So a slightly dense region comes out cold in the CMB and vice versa. 100 00:12:12,440 --> 00:12:17,210 So we're basically capturing this picture of the density fluctuations at that early time. 101 00:12:17,390 --> 00:12:21,290 And the reason we've been able to then use that to extract so much science or so many, 102 00:12:21,560 --> 00:12:27,500 so much the precision cosmology that Steve is talking about is because everything was very it was linear. 103 00:12:27,950 --> 00:12:33,080 These fluctuations are tiny. And we think we understand the physics of what was going on before then. 104 00:12:33,620 --> 00:12:39,440 And we can evolve the physical behaviour of the universe from time, zero up to time and be formed. 105 00:12:40,040 --> 00:12:46,310 And that's not the case later on in the universe. We could do an awful lot later in the universe, but things definitely get messier. 106 00:12:46,610 --> 00:12:50,060 Things definitely get more non-linear and harder to model. 107 00:12:50,900 --> 00:12:54,140 And so in a way, it's sort of easy to look at the CMB. 108 00:12:55,550 --> 00:13:00,890 And so, again, these are these are then the seeds of cosmic structure, these tiny fluctuations we're seeing here. 109 00:13:03,590 --> 00:13:14,540 If you have a density at slight over density at this time, then over the next millions of years that would gravitationally collapse to eventually, 110 00:13:14,540 --> 00:13:20,120 after about a million years or so, form the first stars and the first gravitationally significant objects. 111 00:13:20,900 --> 00:13:25,880 So we'd be able to track back the very the very formation of cosmic structure using this. 112 00:13:26,570 --> 00:13:32,300 Now what let me let me then get out some dirty laundry. So when we show you these these nice maps of the CMB. 113 00:13:33,380 --> 00:13:38,930 First of all, we've done a lot to them to get there. There's a huge amount of work that happens to make a map of the CMB. 114 00:13:39,260 --> 00:13:42,290 And one of the most obvious things that happens is that you actually just measure 115 00:13:42,290 --> 00:13:47,030 microwave light and you have to work out what is the CMB and what's everything else. 116 00:13:47,450 --> 00:13:52,309 Because we're sitting here in the Milky Way and we're looking out through it and we're also looking out through the rest of the universe. 117 00:13:52,310 --> 00:13:56,240 And we're picking up this light that's been coming in from from the most distant place possible. 118 00:13:57,020 --> 00:14:00,349 So I just wanted to show you what we actually measure, for example, 119 00:14:00,350 --> 00:14:04,520 with the Planck satellite at a few of the different wavelengths that it gets to see. 120 00:14:05,070 --> 00:14:13,920 Now Planck has nine wavelengths and I'm showing you maps of the sky and three of them centred at around 1 to 200 gigahertz, 121 00:14:14,190 --> 00:14:20,250 which is about two millimetres in wavelength. And this is the sweet spot for measuring the CMB. 122 00:14:21,360 --> 00:14:25,170 It's the wavelength where it peaks and other things don't. 123 00:14:25,620 --> 00:14:31,020 Okay, so if you go to slightly lower wavelength, sorry, sorry, slightly lower frequencies, 124 00:14:32,130 --> 00:14:37,710 then you start getting swamped by emission from synchrotron radiation. 125 00:14:38,190 --> 00:14:45,420 So electrons spiralling the magnetic field of our galaxy emitting synchrotron, and that increases that low at lower frequencies. 126 00:14:46,500 --> 00:14:53,010 If you come up you so you'll see this, this red emission is is light from the galaxy and much of that would be synchrotron emission. 127 00:14:53,970 --> 00:15:01,620 Then as you increase in frequency, you can see the galaxy, this signal from the galaxy, the red stuff is getting brighter. 128 00:15:02,070 --> 00:15:06,530 And what you're seeing there is more emission thermal emission from dust grains, 129 00:15:06,930 --> 00:15:13,470 micron sized dust grains in the in the galactic plane that are being heated up by starlight and are emitting thermally. 130 00:15:13,800 --> 00:15:18,990 And that also sends a signal that we can we can see now, happily, 131 00:15:18,990 --> 00:15:22,920 when we're looking at the temperature of this light, you can see the CMB fluctuations pop up. 132 00:15:23,090 --> 00:15:29,190 Just you can see them here and here. The galaxy obviously isn't it isn't swamping the signal in the temperature. 133 00:15:30,480 --> 00:15:32,430 This map of the temperature of the of the CMB. 134 00:15:33,570 --> 00:15:38,340 But certainly what we have to do is we combine this information to extract out what's the blackbody signal. 135 00:15:39,330 --> 00:15:42,600 And luckily other things that come from the galaxy aren't blackbody. 136 00:15:42,870 --> 00:15:47,610 And we use that fact. We use that information to tease out that the CMB signal. 137 00:15:49,710 --> 00:15:58,510 Okay. So. What? What we've been used to seeing in CMB observations is measurements of the temperature of the light. 138 00:15:59,410 --> 00:16:08,170 But what we're now getting from Planck and from our new ground based instruments are new views of the universe through the CMB. 139 00:16:09,250 --> 00:16:13,090 And one of those views is through the polarisation of the light. 140 00:16:13,100 --> 00:16:22,450 And this isn't very nice. This isn't a very nice image. I'll show you nice one in minutes of you can measure the intensity of the photons of the CMB. 141 00:16:22,750 --> 00:16:30,370 But you can also measure that polarisation. You can measure that linear polarisation in the Q Stokes vector, or they use stokes factor. 142 00:16:30,700 --> 00:16:40,100 So you're measuring. So we just defined the key stokes factor is polarisation in this direction and you in that direction. 143 00:16:41,690 --> 00:16:44,780 And so any measurement you can, if you have a polarisation sensitive detector, 144 00:16:44,780 --> 00:16:50,180 you can measure the polarisation of the light and these two different in these different directions, not just its intensity. 145 00:16:51,710 --> 00:17:01,700 And that's what we've now done with the Planck satellite. And those fluctuations are shown here in this not very nice greyscale. 146 00:17:02,660 --> 00:17:07,580 And the range of fluctuations is very much smaller than the temperature. 147 00:17:07,580 --> 00:17:11,600 This is now a few micro kelvin fluctuations. 148 00:17:12,650 --> 00:17:18,830 Now, I'll show you a little zoom in. That's maybe a little pressure taken from from our telescope. 149 00:17:18,830 --> 00:17:26,090 Acts of a of a a zoom in on like ten degrees at the skies about this speck on the sky showing the 150 00:17:26,090 --> 00:17:34,040 fluctuations in the CMB with this Q type stokes vector and a U type stokes vector the different patterns. 151 00:17:34,670 --> 00:17:43,700 Okay, now why does this arise? So I said before that the CMB temperature kind of traces the density of the universe back at 4000 years. 152 00:17:44,210 --> 00:17:50,810 Now the polarisation roughly traces the velocity with a motion of the the photons in the baryons. 153 00:17:51,410 --> 00:17:58,610 At that time, you can only get you only get CMB polarised, you only get a polarised photon arising. 154 00:18:00,050 --> 00:18:01,760 If you have two things happening you need, 155 00:18:02,450 --> 00:18:10,610 if you have a free electron and you have Thomson's scattering of it and if the radiation pattern incident on that photon has a quadrupole pattern. 156 00:18:11,420 --> 00:18:18,260 So if you imagine a photon scattering out towards you and you wouldn't know whether it's going to be polarised or not if it's got hotter light, 157 00:18:18,260 --> 00:18:23,210 for example, coming in from the two sides and colder light from top and bottom. 158 00:18:23,930 --> 00:18:28,990 When that scatters out, it's going to come out with a net polarisation along this direction. 159 00:18:29,930 --> 00:18:34,490 Whereas if it had, for example, well isotropic or even dipole pattern, 160 00:18:34,880 --> 00:18:40,700 if you had the same amount of radiation coming in from side to side and top and bottom, when it scatters out, it will be polarised. 161 00:18:41,750 --> 00:18:48,080 So you need this particular quadrupole pattern where you have this, you know, sort of hot, hot pattern side to side and cold, top and bottom. 162 00:18:48,440 --> 00:18:59,030 And you get that from the motion of, um, of this tightly coupled photon baryon fluid that, 163 00:18:59,110 --> 00:19:02,690 that it, that existed before the CMB became completely neutral. 164 00:19:03,620 --> 00:19:11,750 So when we, when we map out the anisotropy in the polarisation of the light, we're basically measuring the velocity of the of the fluid at that time. 165 00:19:12,500 --> 00:19:14,420 And so we're getting so you imagine that, you know, 166 00:19:14,420 --> 00:19:23,540 you've got a map of kind of the the density of space for 2000 years and a map of the the velocity of space to lots of stuff in space at that time. 167 00:19:25,040 --> 00:19:29,630 And I'll show you the science we can extract from the moment. Okay. 168 00:19:29,690 --> 00:19:35,450 So these are little maps that we made from from Acts just last year. 169 00:19:36,290 --> 00:19:44,210 And what we do, we extract the information we try and extract from them is we quantify it in terms of the type of polarisation we see. 170 00:19:44,450 --> 00:19:49,490 And you can decompose it into these two types of polarisation and B modes, 171 00:19:49,940 --> 00:19:58,160 where if you take a map of polarisation, if you pull out the pure divergence part of the field. 172 00:19:58,490 --> 00:20:04,370 So at any point, any polarisation which is pure radial or tangential or at any point in space, 173 00:20:04,820 --> 00:20:09,050 that's the emo type of polarisation that looks has this kind of pattern around any point. 174 00:20:09,560 --> 00:20:16,160 Or you can pull out a curl type of polarisation that is a pure B mode. 175 00:20:17,810 --> 00:20:25,590 And this is useful because both because otherwise your polarisation is coordinate dependent and depends how you define your map. 176 00:20:25,940 --> 00:20:29,120 Whereas these things are coordinate and dependent. But more importantly, 177 00:20:29,120 --> 00:20:36,499 it's the physics that we can extract from the different kinds of polarisations the motion of 178 00:20:36,500 --> 00:20:41,630 the photon baryon fluid act recombination when the c b formed should only produce this kind. 179 00:20:42,020 --> 00:20:47,630 It shouldn't produce this kind. This kind comes from a more exciting thing, which is gravitational waves. 180 00:20:48,200 --> 00:20:49,790 We'll come back to that. Okay. 181 00:20:49,800 --> 00:20:59,690 So with Planck, we get the temperature and the polarisation and we get a third thing to now that we didn't used to have, 182 00:20:59,690 --> 00:21:03,680 which is we get some measure the lensing of the CMB. 183 00:21:04,700 --> 00:21:07,310 So you measure its intensity, you measure its polarisation. 184 00:21:07,520 --> 00:21:16,700 But we can also now measure how much it's got bent by the gravitational effect of all the stuff, but it won't pass on the way to us. 185 00:21:18,650 --> 00:21:22,860 It's just simple, you know, but matter there. 186 00:21:22,880 --> 00:21:25,070 It bends light. Einstein tells us that. 187 00:21:26,360 --> 00:21:32,780 And it didn't used to be possible to measure this because CMB photons typically only get bent or distorted by about two or. 188 00:21:32,920 --> 00:21:37,090 Minutes. So if your experiment hasn't got that resolution, you can't see it. 189 00:21:38,830 --> 00:21:46,540 But Planck has got just about enough resolution and set in the ACT, and Chile has definitely enough resolution and can measure it. 190 00:21:47,380 --> 00:21:54,610 And so what we have here, what this map is now, is a picture of the lensing potential where think of it. 191 00:21:54,620 --> 00:21:58,210 So if I measure the temperature of the CMB in some direction in the sky, 192 00:21:58,930 --> 00:22:09,730 then I can think of the lensed CMB temperature as the unknown signal coming from some direction plus a lensing angle. 193 00:22:10,330 --> 00:22:19,030 So every photon gets the photon that started off in some direction appears to come to me from a slightly deflected, deflected direction. 194 00:22:19,570 --> 00:22:22,780 So every point in the sky has a deflection angle. 195 00:22:22,960 --> 00:22:30,070 How much it got bent by. And I can write the angle as the gradient of some lensing potential field. 196 00:22:30,610 --> 00:22:39,160 And that's what's shown here, where, again, our blue is, is slightly less lensing potential and red is slightly more. 197 00:22:40,180 --> 00:22:46,150 And what that corresponds to is an integral of all of the matter between here and our scattering. 198 00:22:46,870 --> 00:22:52,540 If you have more matter between me and where this can be formed, then it's going to bend the light more. 199 00:22:52,720 --> 00:22:54,880 And if you have less matter, it's going to bend the light less. 200 00:22:55,510 --> 00:23:01,780 So this really is it's a weighted it's a weighted sum because how much the light gets bent depends on where the stuff is. 201 00:23:02,380 --> 00:23:07,360 But it is a weighted sum of all of the matter in the universe. From here it's a lot of scattering. 202 00:23:07,840 --> 00:23:11,920 And that's powerful because that includes all of the invisible stuff, that includes all the dark matter. 203 00:23:13,330 --> 00:23:22,360 And so we can use it to look to to understand dark matter in the universe with with Axe in Chile. 204 00:23:22,360 --> 00:23:25,390 We've also managed to zoom in. I haven't got the plot of it here. 205 00:23:26,410 --> 00:23:35,440 We can take part of that sky and zoom in and get an even better, better measurements of that lensing at high resolution in a bit to the sky. 206 00:23:36,490 --> 00:23:44,080 Okay so so as so we use, I think in the last few years we've gone from having measurements of the temperature of the CMB to now having temperature, 207 00:23:44,080 --> 00:23:47,830 polarisation and lensing. And this is really telling us a lot. 208 00:23:48,040 --> 00:23:54,279 So it's helping us answer these questions that we have, that questions we have about the geometry, 209 00:23:54,280 --> 00:23:57,459 the contents of the universe, and the initial conditions. 210 00:23:57,460 --> 00:24:06,460 How did it start? What happened to start the expansion of the universe and what are the properties of the dark sector in particular, 211 00:24:07,720 --> 00:24:15,550 including cold, dark matter, dark energy and also hot dark matter presumed to be in the form of neutrinos. 212 00:24:15,940 --> 00:24:19,720 And we can learn all of this from from from the CMB. 213 00:24:21,070 --> 00:24:24,690 So what do we do? Sydney, these maps we have right? 214 00:24:24,700 --> 00:24:27,279 We have these maps that we very well. 215 00:24:27,280 --> 00:24:34,570 But then we extract statistics from them and at the main statistics that we pull out from all of these three maps, 216 00:24:34,690 --> 00:24:39,760 temperature, polarisation, lensing is their power spectrum. 217 00:24:40,210 --> 00:24:45,190 That two point function pretty much that all the information, 218 00:24:46,630 --> 00:24:53,320 the power spectrum contains all the information in those maps because they're more or less Gaussian fluctuations, 219 00:24:53,620 --> 00:24:58,210 which means if you compute the power spectrum, you capture all of information that's in them. 220 00:24:59,170 --> 00:25:02,440 So essentially what you're doing is taking you know, you're doing a for a transform. 221 00:25:03,100 --> 00:25:06,670 But because we're doing it on the sphere, we take an angular power spectrum. 222 00:25:06,910 --> 00:25:12,460 So we decompose the maps into spherical harmonics and look at there and look at that, that power. 223 00:25:13,180 --> 00:25:17,500 And so I'm going to be showing you a few of these these these plots, these statistics. 224 00:25:19,270 --> 00:25:29,140 So what I'm showing you here is this is this is this statistic, the angular power spectrum as a function of angular scale. 225 00:25:29,500 --> 00:25:36,460 So large scale over here, 90 degrees in the sky, down to degree scale down to some degree down here. 226 00:25:37,090 --> 00:25:41,200 And you can also think of that as multiple moments ranging up to a few thousand. 227 00:25:41,620 --> 00:25:48,090 So large scales through small scales. And what we see is this incredibly rich statistic, right? 228 00:25:48,100 --> 00:25:53,620 It's the power spectrum of the CMB. Sky is by no means featureless. 229 00:25:53,960 --> 00:25:59,450 Right. It's not flat. It's got all these it's got has got all these peaks and troughs. 230 00:26:00,910 --> 00:26:07,840 And what this is capturing is the physics of what's been happening up from time, from time, zero, up to when this could be formed. 231 00:26:09,070 --> 00:26:12,610 And the physics we think is is pretty straightforward. 232 00:26:13,420 --> 00:26:22,840 We think we had an initial input of primordial fluctuations, and we think those were perhaps put in by mechanism. 233 00:26:23,080 --> 00:26:32,140 We call inflation an exponential expansion of space where quantum fluctuations on very small scales could have been imprinted. 234 00:26:32,780 --> 00:26:36,740 Expand it to macroscopic scales. I'll come back to inflation in a bit. 235 00:26:38,540 --> 00:26:45,740 Some mechanism put in primordial fluctuations and they then evolved over the next 400,000 years. 236 00:26:46,910 --> 00:26:53,330 If these fluctuations were on such a large scale that they're actually beyond the horizon of the universe, 237 00:26:53,930 --> 00:26:59,780 then they won't evolve and they won't have evolved at all between time, zero and time recombination. 238 00:27:01,070 --> 00:27:09,520 And actually, this is what we're seeing on these very large scales here that we know that that's the way we plot. 239 00:27:09,530 --> 00:27:16,430 It says that when you see a flat line here, these are fluctuations that have not evolved by the time that CMB formed, 240 00:27:17,450 --> 00:27:22,010 but smaller with smaller wavelengths that are inside the horizon of the universe. 241 00:27:25,310 --> 00:27:30,880 Those fluctuations evolve according to what's in the universe. 242 00:27:31,140 --> 00:27:37,160 So we've got this mixture. We think of a mixture of dark matter photons and baryons and also neutrinos. 243 00:27:37,640 --> 00:27:45,890 The photons in the baryons, a tightly coupled together, the photons, a scattering of the electrons, and that tightly coupled photon baryon fluid. 244 00:27:46,310 --> 00:27:53,570 In that fluid, we set up sound waves because you have this competition between the gravity of the baryons and the pressure of the photons, 245 00:27:53,960 --> 00:27:56,780 and that sets out sound waves in this in this plasma. 246 00:27:57,430 --> 00:28:04,190 At the same time, you have dark matter particles that are just gravitationally collapsing because they're not interacting with the with the baryons. 247 00:28:05,960 --> 00:28:15,590 Okay. And if you then so you let these sound waves evolve and you capture them after 400,000 years and 248 00:28:15,590 --> 00:28:20,690 different length scales will have reached different points in their evolution of their sound wave, 249 00:28:20,930 --> 00:28:27,440 their oscillations. So for example, we find this peak in the CMB at one degree, 250 00:28:27,980 --> 00:28:33,620 and that corresponds to the wavelength of an oscillation that just had time to reach 251 00:28:33,620 --> 00:28:37,670 its maximal compression of the acoustic wave at the time that the CMB formed. 252 00:28:38,570 --> 00:28:46,280 And then a smaller wavelength that might be at the next peak just had time to have one compression in a rare fraction before the CMB formed, 253 00:28:46,850 --> 00:28:50,900 and another one had, you know, compression, RAF action compression when the CMB formed. 254 00:28:51,710 --> 00:29:05,480 And so you have this what we see here is this is this is this rich acoustic peak structure with harmonics of these, uh, of these sound waves. 255 00:29:08,750 --> 00:29:15,079 And then you get this damping effect known as silk damping, where the CMB, 256 00:29:15,080 --> 00:29:19,340 where the structure is actually a bit, a little bit damped or very much damped because of, 257 00:29:20,090 --> 00:29:26,630 because during the time of recombination, while the CMB is transitioning from being neutral of ionised to being neutral, 258 00:29:28,100 --> 00:29:31,580 photons have time to diffuse and wash out this structure. 259 00:29:31,790 --> 00:29:33,200 So we get this damped effect. 260 00:29:34,550 --> 00:29:46,550 So we think we have this incredibly, in a way, this simple model that fits this great curve, fits this data really remarkably well. 261 00:29:46,670 --> 00:29:51,470 What we call Lambda CDM, and it's a flat, geometrically flat universe. 262 00:29:52,010 --> 00:30:02,329 It has these baryons photons called up matter three neutrino species and also a cosmological constant that we don't understand, 263 00:30:02,330 --> 00:30:04,440 of course, yet, but we think it's there. 264 00:30:05,540 --> 00:30:13,070 And if you put so if you put it and two parameters that describe the initial fluctuations just to an amplitude and a scale dependence. 265 00:30:13,520 --> 00:30:20,900 And if you put if you describe the fluctuations by just two numbers, and then you put in just those those components of the universe, 266 00:30:21,290 --> 00:30:25,190 and you evolve that physics through each recombination, you see exactly this. 267 00:30:25,520 --> 00:30:32,200 Right. And so this this grey curve is the theoretical prediction from our numerical codes that, you know, 268 00:30:32,210 --> 00:30:38,840 Einstein, Boltzmann solvers that that numerically integrate these equations from time 0 to 4000 years. 269 00:30:39,470 --> 00:30:41,990 And, you know, it fits. It's kind of amazing. 270 00:30:42,200 --> 00:30:49,310 It didn't we didn't have to have a model that fitted this data, this incredibly rich data with a lot of features. 271 00:30:50,000 --> 00:30:55,890 And, you know, there's a universe that goes right through it and we've kind of got used to that, right? 272 00:30:55,910 --> 00:30:59,060 We're like, Yeah, yeah, Lambda CDM It works, but it didn't have to. 273 00:30:59,780 --> 00:31:04,920 And the fact that it continues to work kind of continues to surprise us or it should do anyway, right? 274 00:31:04,970 --> 00:31:12,230 Because we've seen it. We kind of had this Lambda CDM model in mind for a while and with the map data 275 00:31:12,380 --> 00:31:16,640 we measured these first three peaks that said Peak wasn't mentioned very well. 276 00:31:18,020 --> 00:31:22,429 And then we've been kind of revealing unwrapping this kind of fourth, fifth, sixth, seventh, eighth, 277 00:31:22,430 --> 00:31:28,970 ninth acoustic peak down here with the Planck data shown in blue and then data from our telescope act. 278 00:31:29,210 --> 00:31:32,390 And then the South Pole Telescope, another ground based experiment in the South Pole. 279 00:31:34,040 --> 00:31:39,740 So in the green an orange remarkably consistent and extending this down here. 280 00:31:40,460 --> 00:31:51,080 So these are many decades of scale. And the fact that we can continue to explain it is pretty remarkable. 281 00:31:51,650 --> 00:31:57,889 And this grey curve is fixed, that data. But actually, if we take in the curve, it just fit to the first. 282 00:31:57,890 --> 00:32:05,570 But the data would have from map then the stuff that fits. Now Planck we got for free, you know, it just, it just works. 283 00:32:06,890 --> 00:32:15,590 And so this has allowed us to measure the primordial fluctuations rather well and measure 284 00:32:15,590 --> 00:32:19,430 the properties of the universe that the amounts of stuff in the universe rather well. 285 00:32:19,430 --> 00:32:22,640 These numbers don't and are going to be a measure to 2% level. 286 00:32:23,060 --> 00:32:26,629 But, you know, ultimately, one doesn't just want to measure numbers to tip at a precision. 287 00:32:26,630 --> 00:32:30,800 You want to actually find out something new and answer questions that you don't know the answer to. 288 00:32:31,610 --> 00:32:34,520 And one of the we've had a few things that we didn't know the answer to. 289 00:32:35,420 --> 00:32:43,940 And we thought and by adding in the polarisation map, we've, we've been able to learn or constrain more things. 290 00:32:44,420 --> 00:32:49,270 So what I'm showing you here now is underneath that temperature curve, 291 00:32:49,880 --> 00:32:56,390 I'm showing you the angular power spectrum of the mode of the polarisation pattern measured from Planck. 292 00:32:57,350 --> 00:33:02,240 And this is now you're seeing now these acoustic peaks again down here. 293 00:33:02,690 --> 00:33:05,870 Note the fact that it's, you know, 1 to 2 orders of magnitude smaller. 294 00:33:06,200 --> 00:33:14,660 So these measurements that small and hard to make from, from Planck and from our initial data from our act experiment in Chile. 295 00:33:16,520 --> 00:33:23,599 Now so the and so this the grey curve that beautifully matches that data, 296 00:33:23,600 --> 00:33:28,340 this new data from Planck, the polarisation data, this curve was not fit for that data. 297 00:33:29,390 --> 00:33:35,200 The curve just goes bang through the data points. And this is the prediction. 298 00:33:35,210 --> 00:33:40,220 This is the prediction from Lambda CDM. This is this you take the theoretical curve that fits this data. 299 00:33:41,240 --> 00:33:46,280 You say, I'm going to take this Lambda CDM model. Tell me what I should predict for the polarisation of the light. 300 00:33:46,940 --> 00:33:52,880 And it's this curve down here and it's really doing it's really doing pretty well. 301 00:33:54,710 --> 00:33:59,510 And the fact that we now have this additional probe and say and it's just this as an aside, 302 00:33:59,840 --> 00:34:03,320 it's just kind of neat, although I don't have a ruler long enough, we try and do it. 303 00:34:04,160 --> 00:34:08,390 I told you that the polarisation roughly measures the velocity of this this fluid. 304 00:34:08,930 --> 00:34:16,850 And if you do draw the line, you see that the peak of that matches up with the trough of that and the peak of that with the trough of that. 305 00:34:17,350 --> 00:34:21,110 And so they are offset by that that the derivative of the other. 306 00:34:21,740 --> 00:34:25,190 So you get the peaks in the polarisation where you get the troughs in the sorry, 307 00:34:25,220 --> 00:34:31,340 where you get the minima in the temperature, which is what you'd expect. And we find that correlated to I just haven't plotted it on here. 308 00:34:32,660 --> 00:34:44,120 So this actually the fact that this works so well is again, the, the conclusion we come to is Lambda CDM works. 309 00:34:45,350 --> 00:34:48,379 But again, we didn't need to come to that conclusion and before Planck, 310 00:34:48,380 --> 00:34:55,790 before new experiments and in Chile in the South Pole there was this kind of whole array of possible cosmologies that we thought might be possible. 311 00:34:56,840 --> 00:35:03,650 And the data now really limit those. So one of the things that we thought might be possible was extra relativistic species in the universe. 312 00:35:04,010 --> 00:35:11,030 We normally assume there are three neutrinos and nothing else, no extra particles, no additional neutrino species, 313 00:35:11,030 --> 00:35:21,020 no sterile neutrinos, no other particles that that could be around that could have decoupled early on relativistic ones. 314 00:35:21,980 --> 00:35:30,800 And before before we had these measurements, there was actually scope for having maybe four or five neutrino species, additional relativistic species. 315 00:35:31,070 --> 00:35:37,940 Now, there's very little scope for that. The constraints now from this data that the neutrinos must be the number of relativistic 316 00:35:37,940 --> 00:35:43,310 species that we assume to be neutrinos must be 3.1 plus a minus point three. 317 00:35:44,300 --> 00:35:48,620 So still a little bit of wiggle room in there. But before this data, it could have been like ten. 318 00:35:49,090 --> 00:35:58,250 That would have been fine. It also greatly limits the scope for fluctuations that could be described by any more than just two numbers, 319 00:36:00,200 --> 00:36:04,310 just a power law description of the primordial fluctuations. 320 00:36:05,300 --> 00:36:06,380 And I'll come back to that. 321 00:36:07,010 --> 00:36:13,579 There's no you can't there's not room for putting in additional features in the early universe that could come from cosmic defect or 322 00:36:13,580 --> 00:36:21,290 magnetic fields or even from some models that produced that would have dark matter annihilation that could produce an extra signal. 323 00:36:21,890 --> 00:36:28,100 There were just this whole this whole set of things that could have been possible that we're not seeing. 324 00:36:29,450 --> 00:36:32,450 And so maybe that's. 325 00:36:33,230 --> 00:36:37,730 Or maybe that saying, okay, the universe really does look too simple, 326 00:36:39,410 --> 00:36:42,800 even though now we have these still these kind of big questions that we don't know the answer to. 327 00:36:45,080 --> 00:36:51,860 The Fed the Fed's statistic is the lending of the CMB. 328 00:36:52,220 --> 00:37:03,260 I just want to give you an image of what lending does to the CMB so that top panel shows the lens and lensed CMB. 329 00:37:03,980 --> 00:37:11,450 Okay, so what you see, we should say, is the effect is not ignore the fact that the bottom line is moving. 330 00:37:11,450 --> 00:37:15,110 It shouldn't be of the effect is not strong. 331 00:37:15,300 --> 00:37:21,440 Okay. But it's coherence. You're seeing the large you're seeing this the things coherently shift. 332 00:37:21,440 --> 00:37:27,400 And that's because you're having this background light currently distorted by large cosmic structures and the structure, 333 00:37:27,420 --> 00:37:37,180 the kind of degree scale on the sky. When you work out the lensing and process and then compute, 334 00:37:37,220 --> 00:37:42,860 it's again got a map of the lensing potential and you compute it and get a power spectrum again for a transform. 335 00:37:43,220 --> 00:37:47,810 How much power is down? Different angular scales? This is again a power spectrum we show. 336 00:37:47,810 --> 00:37:49,910 And this past spectrum we're looking at the CMB. 337 00:37:50,930 --> 00:37:56,990 So power structure was a function of angular scale where this is like a degree and this is below that. 338 00:37:57,860 --> 00:38:05,540 And now here again, okay, so the black curve, I think here is the nice old lambda CDM cosmological model that was fixed. 339 00:38:05,540 --> 00:38:09,140 The Planck temperature data not fit to this data. 340 00:38:09,890 --> 00:38:12,770 And then here is the map here, the measurements from Planck, 341 00:38:13,190 --> 00:38:21,020 including either temperature or temperature optimisation to extract the lensing shown here and even the biggest slight dip here. 342 00:38:21,860 --> 00:38:27,410 We don't think it's significant or maybe it's not significant. It certainly fits this model rather well. 343 00:38:32,000 --> 00:38:35,120 And once again, you know, to belabour the point, it didn't have to fit. 344 00:38:35,690 --> 00:38:41,330 There are different universes that could have fit. Both the temperature and the polarisation data are not fit here. 345 00:38:41,920 --> 00:38:45,500 Lambda CDM comes out, comes out, winning again. 346 00:38:46,670 --> 00:38:52,910 And the kind of universe is that it could have been a universe that's a bit geometrically curved, 347 00:38:53,240 --> 00:38:56,690 that's maybe a little bit closer, a little bit open, and it's not completely flat. 348 00:38:57,590 --> 00:39:03,290 I'm measuring the amplitude of this. This spectrum limits that to a couple of percent. 349 00:39:04,370 --> 00:39:10,040 What the curvature could be actually adding in the positions of galaxies constrains like even further. 350 00:39:10,490 --> 00:39:17,540 But this is, this is, this is, uh, this is, you know, new and powerful measurements. 351 00:39:17,900 --> 00:39:24,290 And it also constrains the neutrino mass from the C and B alone to be less than 0.7. 352 00:39:25,610 --> 00:39:30,770 So this is this is interesting. I think this is this is saying this is the sum of the neutrino masses. 353 00:39:31,910 --> 00:39:38,420 This is saying 95% confidence. How much could it possibly be in terms of what we see cosmological? 354 00:39:39,950 --> 00:39:47,120 And we're saying that that if it's more than 0.7, even the mass of the neutrinos, then it won't fit this data. 355 00:39:49,770 --> 00:39:53,940 And the reason. So this is I think this is this is interesting, 356 00:39:54,060 --> 00:40:00,570 but it's also the path to what we think we know we're claiming is a future detection of neutrino mass cosmological. 357 00:40:00,990 --> 00:40:09,210 Well, they will have to do that in the context of of a discussion with particle physicists and hopefully agreement with particle physics experiments. 358 00:40:09,250 --> 00:40:16,620 Okay. So I want to just have a note and I would kind of talk a bit about why why the lensing helps us with neutrinos. 359 00:40:16,620 --> 00:40:31,480 So much so. After we measured the cleansing of the CMB, we're measuring basically an integral of the dark matter out from here to the scattering. 360 00:40:32,950 --> 00:40:42,190 And what I can think of is if I increase the neutrino mass some of the neutrino masses, 361 00:40:42,760 --> 00:40:51,160 then I should think roughly of switching cold, dark matter into matter. 362 00:40:51,430 --> 00:40:55,210 Dark matter that started off hot and then became cold. 363 00:40:56,050 --> 00:41:01,840 So cold, dark matter we think has been cold all the way through the universe. 364 00:41:02,330 --> 00:41:09,040 It's perhaps a wimp decoupled very early on and it's been known relativistic for the whole duration of the universe. 365 00:41:11,100 --> 00:41:18,420 Neutrinos, in contrast, are much lighter, which means that they started off as relativistic particles. 366 00:41:18,450 --> 00:41:27,330 They started off like radiation. And then as the universe cooled down, depending on their mass, they suddenly became not much of a stick. 367 00:41:28,020 --> 00:41:34,020 So if you think of a neutrino particle as basically being a hot, dark matter particle and then a cold, dark matter particle. 368 00:41:34,740 --> 00:41:40,950 So now today it's all cold. All in all, all of the dark matter, including the neutrinos, are cold. 369 00:41:41,700 --> 00:41:49,110 But to begin with, they weren't. And that means that we have we have a way of detecting their effects because logically, 370 00:41:50,580 --> 00:41:58,590 because what they do is cold, dark matter clumps together and collapses gravitationally. 371 00:41:59,430 --> 00:42:03,150 But hot, dark matter free streams. And it doesn't clump. 372 00:42:04,620 --> 00:42:11,459 So during the epoch of the universe, when the neutrinos were behaving relativistic, 373 00:42:11,460 --> 00:42:17,370 only when they were hot, they basically washed out the formation of cosmic, large cosmic structures. 374 00:42:19,020 --> 00:42:24,000 But during the epoch, when they were cold, they were fully contributing to the regular growth of cosmic structure. 375 00:42:25,290 --> 00:42:31,620 And so this has a very visible effect on the clustering of matter in the universe. 376 00:42:32,250 --> 00:42:37,230 I'm showing it, first of all, here, and I'm still using a power spectrum, but now I'm not showing you the lensing. 377 00:42:37,350 --> 00:42:41,310 I'm showing you the power spectrum of just matter. 378 00:42:41,580 --> 00:42:44,820 The two point function of what dark matter should look like, 379 00:42:44,880 --> 00:42:51,570 how the clustering of dark matter looks as a function again of scale from large scales to small scales. 380 00:42:55,650 --> 00:43:00,150 And if I have basically a zero no neutrino mass, 381 00:43:00,510 --> 00:43:08,370 then the kind of lambda CDM zero neutrino mass curve has this particular shape that basically looks like that blue curve here. 382 00:43:10,560 --> 00:43:19,560 Now, if I increase the neutrino mass, then what I do is I dump, I reduce the amount of clustering on these small scales. 383 00:43:19,560 --> 00:43:25,770 And you can look at the most extreme thing where it's one IB the sun and you lose structure on these small scales, 384 00:43:26,100 --> 00:43:28,530 but you don't lose any structure on these large scales. 385 00:43:29,760 --> 00:43:39,660 And these small scales correspond to scales that began to evolve earlier in the universe and earlier was when they were still hot. 386 00:43:40,800 --> 00:43:46,920 So scales in the universe that began to evolve when the neutrinos were relativistic, 387 00:43:47,970 --> 00:43:54,210 their structure gets suppressed compared to the larger scales in the universe that have been evolving. 388 00:43:54,870 --> 00:44:03,390 While the neutrinos were not much of a stick. So you get this, you get this behaviour that the more and the more massive it is, 389 00:44:04,290 --> 00:44:09,720 the more this contrast between, you know, neutrinos that behave like radiation and then matter. 390 00:44:10,470 --> 00:44:16,260 And the more that you get, the more you have this, this, this spectrum damped compared to that compared to the large scales. 391 00:44:20,010 --> 00:44:25,860 And this is what is leading to the current limits on the genomes from from the Planck lensing power spectrum, 392 00:44:25,860 --> 00:44:34,320 which is basically an integral of this dark matter over, over the whole of the history is now a 95% confidence, less than .23 EV. 393 00:44:35,300 --> 00:44:38,070 Okay. And that's I mean, that's a pretty strong constraint. 394 00:44:38,550 --> 00:44:44,580 That's the kind of level that, you know, is hoped to achieve with the captured experiments. 395 00:44:45,420 --> 00:44:51,210 So if we can make a at least a statement about what we think is the case from cosmology, then this should have interesting, 396 00:44:53,460 --> 00:44:57,540 consequent implications for what one would expect to see from particle physics experiments. 397 00:44:58,230 --> 00:45:05,010 And actually, in the next decade, we're hoping to make a detection of non-zero in a treated mass through its cosmological effect, 398 00:45:05,010 --> 00:45:06,930 through its effect on the lensing of the CMB. 399 00:45:07,980 --> 00:45:15,990 We know from oscillation experiments that the minimum mass of neutrinos should be 0.06 IB or 60 maybe electron volts. 400 00:45:17,220 --> 00:45:20,530 And we're looking at its effect on this. 401 00:45:20,610 --> 00:45:27,900 On the CMB lensing power spectrum, we should be able to get to a few sigma detection of that in the next decade. 402 00:45:28,410 --> 00:45:36,030 And this plot is showing the fractional a fractional effect of changing the neutrinos on the CMB lensing power spectrum. 403 00:45:36,390 --> 00:45:41,430 So this is this is a 1% shift, 5% shift, 10% shift. 404 00:45:42,300 --> 00:45:47,610 And this is the curves for, you know, for some of the neutrino masses. It's 50 or 100, 150 electron volts. 405 00:45:48,600 --> 00:45:51,830 So they're small shifts, but they are they got the shape dependent. 406 00:45:51,840 --> 00:45:57,030 They look different. And we do think and this is work that we're doing at the moment, 407 00:45:57,030 --> 00:46:04,140 is trying to establish how similar the effects can be to other cosmological features like the curvature of the universe, like dark energy. 408 00:46:05,460 --> 00:46:09,790 But it appears that we can get to a. 409 00:46:10,130 --> 00:46:15,530 Detection in a decade. But that detection is obviously it's indirect detection. 410 00:46:15,530 --> 00:46:23,570 Right. We're saying with we're saying that we think we're seeing the effects of these these relativistic particles and we think they've got this mass. 411 00:46:23,960 --> 00:46:29,780 So putting that in the context of then direct detection particle experiments will be exciting. 412 00:46:31,010 --> 00:46:38,330 But obviously we'd like to be able to make the statement from the CMB and then see if if we see the same thing in particle physics. 413 00:46:38,900 --> 00:46:50,220 Okay. Let me say something about inflation. Inflation is our best current model or scenario for the universe. 414 00:46:50,640 --> 00:46:55,620 This exponential expansion of the universe driven by the potential of some scalar field, 415 00:46:56,170 --> 00:47:01,110 it's, it's the most favourite model out there, but we don't yet know if it's what happened. 416 00:47:02,040 --> 00:47:06,929 It seems to work so well because it predicts this exponential expansion. 417 00:47:06,930 --> 00:47:08,670 Should it predict the flatness of the universe, 418 00:47:09,180 --> 00:47:15,719 it should also produce the right kind of fluctuations that we see quantum fluctuations imprinted during the 419 00:47:15,720 --> 00:47:27,570 inflationary expansion should will naturally produce this almost scale independent spectrum of fluctuations. 420 00:47:28,110 --> 00:47:35,550 They should be Gaussian and they should be adiabatic, which means that fluctuations in all of the fluids in the universe should of traced each other. 421 00:47:36,000 --> 00:47:43,260 So where you have had an over dense region of dark matter, you'd also have an evidence region of photons and baryons. 422 00:47:43,260 --> 00:47:47,459 At the same time it's called adiabatic. And inflation predicts all these things. 423 00:47:47,460 --> 00:47:53,400 And we're seeing all of those things from Planck and from Planck and from from ACT as well. 424 00:47:54,300 --> 00:47:59,310 We're in particular seeing this very particular prediction of some inflationary models, 425 00:47:59,790 --> 00:48:04,420 which are that the fluctuations are not not precisely scale invariant. 426 00:48:05,070 --> 00:48:09,360 So scale invariant means that there are the injection of fluctuations at the beginning. 427 00:48:10,230 --> 00:48:13,170 You have the same amount on all scales or on all scales. 428 00:48:14,310 --> 00:48:21,960 But many inflationary models predict that you should get slightly smaller fluctuations on small scales because inflation has to end. 429 00:48:22,680 --> 00:48:29,340 And the dynamics of that, the inflationary process, many of the models predict smaller, less power and small scales. 430 00:48:30,180 --> 00:48:36,570 And so we've detected this now where we measure it by using the spectral index of fluctuations, 431 00:48:36,660 --> 00:48:41,880 where scale independent is one and not scale independent is a bit less than one or a bit more than one. 432 00:48:42,810 --> 00:48:47,730 And what I'm showing you here is just the angle of power spectrum, of the temperature fluctuations from Planck, 433 00:48:48,870 --> 00:48:54,860 with the data in black, with the best fit curve, with a special index of one in blue. 434 00:48:54,930 --> 00:48:58,170 And it doesn't fit. It doesn't fit like Six Sigma. 435 00:48:58,800 --> 00:49:05,100 You need slightly less power at small scales shown by this red curve here to fit the data. 436 00:49:05,640 --> 00:49:07,050 And this is new from Planck. 437 00:49:07,230 --> 00:49:14,760 We had kind of hints at Three Sigma before, but this is Six Sigma and we think it's there and it still doesn't mean it's inflation, 438 00:49:15,240 --> 00:49:17,760 but it, it's all looking a lot like inflation. 439 00:49:18,990 --> 00:49:25,470 And so if it's not inflation, it needs to look a lot, a lot like this and make all these predictions as the last thing we're looking for. 440 00:49:26,670 --> 00:49:31,230 My last thing would be if we found it's the beginning of a lot of things, a gravitational waves. 441 00:49:31,710 --> 00:49:39,840 So inflation produces scalar fluctuations that are the things that cause density over densities and gravitational collapse of objects. 442 00:49:40,170 --> 00:49:45,750 But it should also imprint gravitational waves in space and they should propagate. 443 00:49:46,440 --> 00:49:54,240 So these are tensor fluctuations to the metric that just propagate as gravitational waves and do not couple don't produce density fluctuations. 444 00:49:55,020 --> 00:50:02,400 And, you know, we want to find them because inflation happens then they should produce. 445 00:50:02,730 --> 00:50:04,150 They should print it. 446 00:50:04,200 --> 00:50:11,040 So sorry, this this little cartoon is designed to show what gravitational gravitational wave travelling towards you would do to space. 447 00:50:11,190 --> 00:50:17,890 Right? A gravitational wave is going to distort space shrinking and stretching it as it goes. 448 00:50:17,970 --> 00:50:26,630 It can have these two different oscillation patterns. Gravitational waves, if there were any gravitational waves travelling through the universe. 449 00:50:27,110 --> 00:50:32,900 When the CMB formed at recombination, they would polarise the CMB light. 450 00:50:34,610 --> 00:50:39,790 And roughly, you should imagine I told you before that to get a polarised photon you need a quadrupole pattern. 451 00:50:39,800 --> 00:50:42,260 So you'd like more radiation this way than this way. 452 00:50:42,830 --> 00:50:49,760 And actually, if a gravitational wave is passing through, then it distorts space in such a way that you get that exact pattern instant on a photon. 453 00:50:50,270 --> 00:50:55,370 So if it's passing through, then the photon, that CMB photon that comes out, comes out polarised. 454 00:50:55,820 --> 00:51:00,260 So it's an indirect, again, effect of the gravitational wave. 455 00:51:01,550 --> 00:51:07,430 Now gravitational waves produce both patterns of polarisation, both emerge and b mode polarisation. 456 00:51:07,880 --> 00:51:14,480 And so this is the space that we're looking, looking for and looking at once again, because it's our favourite statistic. 457 00:51:14,600 --> 00:51:21,050 This is the angular power spectrum of the CMB polarisation as a function of angular scale up here. 458 00:51:22,280 --> 00:51:28,250 This red curve actually is the curve that I showed you before with the Planck and the pole data on. 459 00:51:28,640 --> 00:51:33,950 This is a thing we've actually now already measured, although we've got some work to do still we've seen it already. 460 00:51:34,880 --> 00:51:40,580 What we're now looking for is this smaller blue signal, the blue curve down here, 461 00:51:41,540 --> 00:51:46,520 that's the prediction of a gravitational wave signal, but whose amplitude is unknown? 462 00:51:47,210 --> 00:51:52,580 I'm showing you two examples here of two different sizes. It could be it's got to now be less than about here. 463 00:51:52,580 --> 00:51:57,290 We would have seen it, but it could be anywhere down here and it could be down through the floor. 464 00:51:58,100 --> 00:52:01,610 It could depend on the size of it depends on the energy scale of inflation. 465 00:52:03,500 --> 00:52:14,840 But its shape is is better known in green is actually a lens and b mode signal that 466 00:52:14,840 --> 00:52:19,460 gets generated because the polarised E mode signal gets lot gravitationally lensed. 467 00:52:20,780 --> 00:52:27,440 It's interesting in its own right, but it's not what we want to look for. We're looking for this this primordial signal from gravitational waves. 468 00:52:28,210 --> 00:52:32,540 The thing we're searching for is this very small B mode signal. 469 00:52:33,720 --> 00:52:38,030 I'm running short time. Let me just say a couple of things about the challenges there. 470 00:52:39,440 --> 00:52:46,790 This is what many of us in the CMB community and are doing is is building experiments to search for these signals. 471 00:52:47,150 --> 00:52:52,250 But this is the tiny signal. You can see it's down and it's below. 472 00:52:52,490 --> 00:52:57,350 It's very much sub micro kelvin, we're talking dense and then a Kelvin range. 473 00:52:57,950 --> 00:53:01,829 It's the kind of technology we have to use and this is what's in it's an act. 474 00:53:01,830 --> 00:53:12,620 And surely thousands of detectors coupled together in array arrays in a telescope where you have thousands of very sensitive detectors. 475 00:53:13,070 --> 00:53:23,360 There's just a little picture of all of our detectors for act where you have this these this array of 500 feet horns collecting C and B light, 476 00:53:23,840 --> 00:53:28,490 splitting it into POLARISATIONS That then get instant on a transition to a sensible ometer. 477 00:53:28,730 --> 00:53:33,320 So these are big emitters that change their resistance when they receive photons 478 00:53:33,740 --> 00:53:38,330 and you couple thousands of those together to get as much sensitivity as possible. 479 00:53:38,900 --> 00:53:42,380 And we've gone from having tens of detectors to hundreds to now thousands. 480 00:53:42,950 --> 00:53:49,189 And you need that many if you want to get sensitive detection of the polarisation of the CMB and where this 481 00:53:49,190 --> 00:53:54,950 is going is going from thousands to tens of thousands is where we need to go in the next in the next decade. 482 00:53:56,060 --> 00:54:04,190 So this technology has had remarkable improvements in the last few years, which means we can now field thousands of detectors in the sky. 483 00:54:05,360 --> 00:54:10,790 The other challenge is to see through the galaxy. The galaxy emits light. 484 00:54:10,790 --> 00:54:18,470 I told you that already. It's also polarised that light and it's polarised because the magnetic field in the galaxy is not isotropic. 485 00:54:18,710 --> 00:54:29,510 It aligns with the plane of the galaxy and that aligns dust grains and dust grains line up along the galaxy through radiative, 486 00:54:29,510 --> 00:54:38,680 radiative talks and they are firmly along their long axis and that comes out polarised and we never knew how polarised it was going to be. 487 00:54:38,750 --> 00:54:43,550 So we measured it with Planck just last year and it turns out it's quite polarised. 488 00:54:44,150 --> 00:54:50,240 We weren't sure how polarised it was and it's about it seems to be up to about 10% or more polarised and that's actually quite polarised. 489 00:54:50,510 --> 00:54:57,950 We optimistically thought it might be like 1% or 2%, but we didn't know and it turns out some of it's a few percent from it's 10%. 490 00:54:58,340 --> 00:55:05,840 And that means it's really going to continue to be one of the main challenges to getting through to see and be a big bang signal. 491 00:55:07,280 --> 00:55:08,360 What that also means, 492 00:55:08,510 --> 00:55:20,420 this fact that the galaxy is quite bright means that it's going to that that the the AP might signal was in fact detected last year by the. 493 00:55:20,460 --> 00:55:30,060 Bicep2 experiments operate in the South Pole, but that B mode signal turns out to have been almost certainly just a beam signal from the galaxy, 494 00:55:30,120 --> 00:55:33,150 not a beam mode signal from the big bang from the CMB itself. 495 00:55:34,530 --> 00:55:44,610 And if we zoom in on parts of the spectrum that it measured, then its measured data points well here in black. 496 00:55:44,730 --> 00:55:52,800 This is the B mode power as a function of scale. But when you subtract off the signal from the galaxy, you get this blue curve. 497 00:55:53,610 --> 00:56:00,030 So these blue data points, and that's completely consistent with this theoretical curve just from a lensing spectrum, 498 00:56:00,030 --> 00:56:04,620 not from any gravitation, not from any gravitational waves. 499 00:56:05,130 --> 00:56:08,160 So right now, we haven't seen any gravitational wave signals. 500 00:56:08,400 --> 00:56:12,120 And in fact, they've pushed down the upper limits on the possible signal, 501 00:56:12,120 --> 00:56:16,110 such that actually one of the simplest models of inflation is now basically disfavoured. 502 00:56:17,610 --> 00:56:25,480 Quadratic potential for the for the inflation field that now is is kind of is is out. 503 00:56:26,280 --> 00:56:30,720 What is disfavoured. So the path to getting to gravitational waves, 504 00:56:30,780 --> 00:56:39,930 this is what we're going to be doing in the next decade is using ground based experiments with more and more detectors, 505 00:56:40,140 --> 00:56:42,750 covering more of the sky with more wavelengths. 506 00:56:43,200 --> 00:56:49,950 So we'll be mapping half the sky in a number of wavelengths with these many thousands of sensitive detectors. 507 00:56:50,430 --> 00:56:56,550 And that can get us down to two to many factors lower than where we are now. 508 00:56:56,700 --> 00:57:01,790 And we hope that we might get to see it. And we'll skip to actually. 509 00:57:02,220 --> 00:57:11,910 Okay. So so all our CMB data really demands this simple model, this very simple lambda CDM cosmological model, 510 00:57:12,330 --> 00:57:18,150 and it holds up to this fifties many new views of the universe that we have from these experiments. 511 00:57:18,810 --> 00:57:24,780 The neutrino sector is fascinating and holds questions that cosmology can help to answer in the coming decades. 512 00:57:26,040 --> 00:57:28,950 And in fact, if inflation is not the thing that began the universe, 513 00:57:29,130 --> 00:57:33,330 then whatever it was has got to look a lot like it's searching for gravitational waves. 514 00:57:33,900 --> 00:57:37,980 That search is still firmly on. So watch this space.