1 00:00:00,000 --> 00:00:06,324 We now turn to the study of galaxy formation for stars, galaxies and cosmic 2 00:00:06,324 --> 00:00:10,778 reionization. This is probably the most active area of 3 00:00:10,778 --> 00:00:15,857 observational cosmology today. First some preliminaries. 4 00:00:15,857 --> 00:00:22,321 What do you mean by galaxy formation? Galaxies, kind of keep forming all the 5 00:00:22,321 --> 00:00:29,032 time, there is no really sharp definition between formation and evolution. 6 00:00:29,032 --> 00:00:33,030 And just as with galaxy evolution, there are two components to it. 7 00:00:33,030 --> 00:00:37,470 How do you assemble mass and how do you turn gas into stars, and then their 8 00:00:37,470 --> 00:00:42,620 effects on the remaining gas. Obviously galaxy formation as well as the 9 00:00:42,620 --> 00:00:48,508 evolution have to be related to their large scale environment, because they're a 10 00:00:48,508 --> 00:00:53,572 sort of low end extension of it. But unlike structure formation itself, 11 00:00:53,572 --> 00:00:59,864 galaxy formation includes very dissipative processes, in particular star formation. 12 00:00:59,865 --> 00:01:05,335 We now also know that processes of galaxy formation seem to be somehow closely 13 00:01:05,335 --> 00:01:10,263 coupled to the formation of the central super massive black holes, and we'll 14 00:01:10,263 --> 00:01:13,069 address that in a little more detail later. 15 00:01:13,069 --> 00:01:17,644 But generally when people talk about galaxy formation they usually mean 16 00:01:17,644 --> 00:01:23,212 assembly of large galaxies of today. Some were at large redshifts, certainly 17 00:01:23,212 --> 00:01:26,994 greater than 3 or 4, maybe greater than 5 or 10. 18 00:01:26,994 --> 00:01:32,758 The old dwarf galaxies may be just coming together today, and so we do see some 19 00:01:32,758 --> 00:01:37,996 residual galaxy formation going on. So in the last decade or, or so of, we 20 00:01:37,996 --> 00:01:43,912 discovered large populations of young galaxies, large redshifts young, simply 21 00:01:43,912 --> 00:01:49,210 because universe was young then. And now we have a fairly good idea of how 22 00:01:49,210 --> 00:01:54,356 the whole process works. And indeed, you can think of the frontier 23 00:01:54,356 --> 00:02:00,020 as understanding the very first star formation and formation of first 24 00:02:00,020 --> 00:02:06,644 protogalactic fragments and how they re-ionize the neutral universe, somewhere 25 00:02:06,644 --> 00:02:10,457 around redshifts between 6 and maybe 15 or 20. 26 00:02:10,457 --> 00:02:15,842 To recap the general picture of structure formation, we start with small initial 27 00:02:15,842 --> 00:02:19,370 density perturbations that come out of the Big Bang. 28 00:02:19,370 --> 00:02:25,805 They keep collapsing, driven by the dark matter, and that process begins well 29 00:02:25,805 --> 00:02:31,094 before recombination when micro-background is released. 30 00:02:31,094 --> 00:02:35,720 At that point there are no sources of light in the universe for awhile and 31 00:02:35,720 --> 00:02:40,417 barions gas keeps falling into the potential oils delineated by the dark 32 00:02:40,417 --> 00:02:43,979 matter. At some point the very first stars begin 33 00:02:43,979 --> 00:02:47,208 to form. That probably happens around ratchets of 34 00:02:47,208 --> 00:02:50,134 20 or 30 or so, but nobody really knows for sure. 35 00:02:50,134 --> 00:02:56,840 Why, redshift of six or so, the young stars from, freshly formed proto-galaxies 36 00:02:56,840 --> 00:03:04,040 or their fragments we ionize the universe, and it becomes transparent to UV radiation 37 00:03:04,040 --> 00:03:07,325 again. All through this process and continued to 38 00:03:07,325 --> 00:03:12,435 the present day, there is a hierarchical structure formation that galaxies merge, 39 00:03:12,435 --> 00:03:17,034 or smaller fragments are absorbed into the large ones, so the galaxy growth 40 00:03:17,034 --> 00:03:19,956 continues. As far as stellar populations are 41 00:03:19,956 --> 00:03:25,269 concerned, massive stars will produce heavier chemical elements in their cores, 42 00:03:25,269 --> 00:03:30,120 exploded super novi returning to interstellar medium from which new stars 43 00:03:30,120 --> 00:03:33,482 are formed and so on. And in this way we have ever more 44 00:03:33,482 --> 00:03:38,949 enriched, chemically enriched interstellar material as well as new generations of 45 00:03:38,949 --> 00:03:42,889 stars. An important player in all this may be 46 00:03:42,889 --> 00:03:48,498 energetics related to super massive black holes in galaxies, those responsible for 47 00:03:48,498 --> 00:03:53,870 quasar-like activity, because they can dump substantial amounts of kinetic and 48 00:03:53,870 --> 00:03:58,597 radiative energy in their hosts and modify their evolution as well. 49 00:03:58,597 --> 00:04:04,130 So here is a nice schematic illustration of the early cosmic history. 50 00:04:04,131 --> 00:04:08,588 This one is due to Avi Loeb. In the beginning of this particular 51 00:04:08,588 --> 00:04:12,294 segment we have cosmic micro background recombination. 52 00:04:12,294 --> 00:04:17,158 At which point universe enters what we now call dark ages because there are no 53 00:04:17,158 --> 00:04:21,354 sources of light. Dark holes keep collapsing and eventually 54 00:04:21,354 --> 00:04:25,363 you die substar formation and more and more of those happen. 55 00:04:25,363 --> 00:04:30,137 When the ionized bubbles of gas around those first protogalaxies start to 56 00:04:30,137 --> 00:04:35,065 overlap, this is what we call the reionization era, where the universe which 57 00:04:35,065 --> 00:04:40,224 was filled with neutral hydrogen and helium ever since the microbackground was 58 00:04:40,224 --> 00:04:43,439 released. Now becomes ionized again and that is 59 00:04:43,439 --> 00:04:48,653 probably as good a boundary as any when we say, well this is roughly where galaxy 60 00:04:48,653 --> 00:04:53,346 formation is really occurring. And after that obviously evolution of 61 00:04:53,346 --> 00:04:57,324 galaxies continues. So the assembly of the mass is driven by 62 00:04:57,324 --> 00:05:02,459 the dark matter and those potential wells keep us getting assembled through a 63 00:05:02,459 --> 00:05:07,007 creation of More material. A way to characterize them is by their 64 00:05:07,007 --> 00:05:10,514 mass function, how many there are in a given mass. 65 00:05:10,514 --> 00:05:16,310 Here's a set of theoretical curves of how the mass function of dark halos evolves as 66 00:05:16,310 --> 00:05:21,176 a function of time. There are many small ones, and there are a 67 00:05:21,176 --> 00:05:24,408 few large ones. But if you look at curves, which are 68 00:05:24,408 --> 00:05:29,984 labeled by the red shift, early on at high red shifts there are really very few very 69 00:05:29,984 --> 00:05:33,228 massive hails and the growth keeps occurring. 70 00:05:33,228 --> 00:05:38,646 There isn't so much growth at low mass end, but you start populating the high 71 00:05:38,646 --> 00:05:44,768 mass end with evermore more massive halos, and that process continues until today. 72 00:05:44,768 --> 00:05:50,220 So one of the important predictions of our general theoretical understanding of 73 00:05:50,220 --> 00:05:55,660 structure formation is that there will relatively few massive potential wells, 74 00:05:55,660 --> 00:05:59,390 massive galaxies early on. Well we discussed structure formation. 75 00:05:59,390 --> 00:06:02,319 We talked about essential gravitational processes. 76 00:06:02,320 --> 00:06:07,996 Galaxies follow that as well, but they're also much, much messier dissipative 77 00:06:07,996 --> 00:06:13,930 processes, where gases are converted into stars, stars radiate energy, change to 78 00:06:13,930 --> 00:06:19,932 gas, stars explode and so on and so forth. These processes are impossible to model 79 00:06:19,932 --> 00:06:25,740 analytically, not because with their hard, they truly are impossible to model 80 00:06:25,740 --> 00:06:29,610 analytically. And they can be only studied numerically. 81 00:06:29,610 --> 00:06:33,898 So we do not have a clean-cut theory of star formation, although we do understand 82 00:06:33,898 --> 00:06:38,717 it fairly well in some sense. And that then maps into our relatively 83 00:06:38,717 --> 00:06:44,032 shaky understanding of how initial star formation proceeds. 84 00:06:44,032 --> 00:06:49,744 Nevertheless, much progress was made, and we'll talk more about that later. 85 00:06:49,744 --> 00:06:55,269 So, all of the things that I already mentioned are parts of the overall picture 86 00:06:55,269 --> 00:07:00,980 of galaxy formation and evolution. But the basic paradigm is very simple. 87 00:07:00,980 --> 00:07:06,474 You have potential wells formed by dark matter, which can keep merging, and they 88 00:07:06,474 --> 00:07:12,214 form containers for the gas that will fall in, get sufficiently dense in the middle, 89 00:07:12,214 --> 00:07:16,789 that will start making stars. So, generically we expect the dark halos 90 00:07:16,789 --> 00:07:21,821 will be more extended than the stellar components of galaxies, which was exactly 91 00:07:21,821 --> 00:07:26,804 what we've seen when we discussed global properties and structure of galaxies. 92 00:07:26,804 --> 00:07:31,220 So for a while, protogalaxies, or really young galaxies, were something of a 93 00:07:31,220 --> 00:07:34,844 mythical creature, and the question is how do you define one? 94 00:07:34,844 --> 00:07:39,936 And there are many different ways in which you can define it, as some first X years 95 00:07:39,936 --> 00:07:44,952 or fraction of the age of the galaxy's life, or when certain fraction of the mass 96 00:07:44,952 --> 00:07:49,221 is assembled, or a certain amount of stars is assembled, and so on. 97 00:07:49,221 --> 00:07:53,694 Or we can simply declare that beyond certain redshift everything should be 98 00:07:53,694 --> 00:07:57,377 treated as a young galaxy. It doesn't really matter. 99 00:07:57,378 --> 00:08:03,169 Essentially we've tried to map the whole process from the very first star formation 100 00:08:03,169 --> 00:08:08,956 until galaxy evolution to the present day. But for a variety of reasons, we think 101 00:08:08,956 --> 00:08:14,252 that in most young galaxies, there was very intense star formation. 102 00:08:14,252 --> 00:08:19,364 And there should be very little luminous objects because of that. 103 00:08:19,364 --> 00:08:24,056 Although, obviously, there will be a lot of small ones and very few really massive 104 00:08:24,056 --> 00:08:27,169 ones early on. So it's the energy release from forming 105 00:08:27,169 --> 00:08:30,162 galaxies that makes it possible for us to see them. 106 00:08:30,162 --> 00:08:37,380 And there essentially two major mechanisms for young galaxy's to create energy. 107 00:08:37,380 --> 00:08:42,276 First of all it's the collapse from large destruction. 108 00:08:42,276 --> 00:08:47,580 You'll recall that, cooling plays a crucial role that galaxies are thousands 109 00:08:47,580 --> 00:08:52,416 times denser than they should be if they're just extrapolation of logical 110 00:08:52,416 --> 00:08:57,924 structure. That is, they collapse by an extra factor 111 00:08:57,924 --> 00:09:02,310 of ten. And therefore binding energy change due to 112 00:09:02,310 --> 00:09:08,340 the dissipative collapse overwhelms any changes in binding energy from simple 113 00:09:08,340 --> 00:09:13,247 gravitational collapse. However it happened half the potential 114 00:09:13,247 --> 00:09:18,297 energy today is equal to the kinetic energy in galaxies within a sign. 115 00:09:18,298 --> 00:09:24,766 And therefore we can estimate kinetic energy of the galaxy today, and this much 116 00:09:24,766 --> 00:09:30,706 energy was also released in process of galaxy, proto-galaxy collapse. 117 00:09:30,706 --> 00:09:36,250 So we can take approximately mass of the cooling materials, that will be the 118 00:09:36,250 --> 00:09:42,146 variants, and multiply it with a square of characteristic three D velocities in 119 00:09:42,146 --> 00:09:48,042 galaxies and for typical numbers, that the relevant here for galaxy say like the 120 00:09:48,042 --> 00:09:52,510 Milky Way, the relevant number is roughly n to the 59 ergs. 121 00:09:52,510 --> 00:09:55,710 That's just for the variance part that cools. 122 00:09:55,710 --> 00:10:01,470 There's 10 times that much in dark matter, but there is no visible signal from it. 123 00:10:01,470 --> 00:10:06,247 It's simply a collapse and that sounds like a lot of edge. 124 00:10:06,248 --> 00:10:12,572 It turns out actually that conversion of gas interstellar material to stars also 125 00:10:12,572 --> 00:10:18,989 produces additional amount of energy which is approximately of the same order, so 126 00:10:18,989 --> 00:10:25,313 just Putting things together will release of the order of ten to the 59 or few times 127 00:10:25,313 --> 00:10:30,518 10 to the 59 at most of energy. But it's not the energy that matters but 128 00:10:30,518 --> 00:10:34,393 luminosity or what time interval this is being done. 129 00:10:34,393 --> 00:10:40,170 However, the second source of energy is actually more important. 130 00:10:40,170 --> 00:10:45,670 And this is conversion of hydrogen into helium in massive stars. 131 00:10:45,670 --> 00:10:51,892 The term nuclear fusion in stars, converting hydrogen into helium in the 132 00:10:51,892 --> 00:10:56,554 heavy elements. Releases energy and approximately 0.1% of 133 00:10:56,554 --> 00:11:00,490 the initial mass of hydrogen is converted into energy. 134 00:11:00,490 --> 00:11:06,170 It turns out that subsequent fusion from helium into heavier elements is relatively 135 00:11:06,170 --> 00:11:11,258 minor perturbation to this one. So it's really cooking up of the helium 136 00:11:11,258 --> 00:11:14,979 stars that matters. Observing helium abundance, and 137 00:11:14,979 --> 00:11:20,205 subtracting the primordial one, is actually very, very difficult, especially 138 00:11:20,205 --> 00:11:24,474 for old stellar populations. Instead of that, we can model chemical 139 00:11:24,474 --> 00:11:29,586 evolution of stellar populations and find that indeed there be correlation between 140 00:11:29,586 --> 00:11:34,372 the amount of helium cooked up in stars, as opposed to primordial one. 141 00:11:34,372 --> 00:11:38,790 And the amount of heavier elements which astronomers always call metals, that 142 00:11:38,790 --> 00:11:41,308 includes things like oxygen and carbon and so on. 143 00:11:41,308 --> 00:11:46,984 Now that is something that we can measure from spectroscopy, and so it turns out 144 00:11:46,984 --> 00:11:52,746 that the typcial metallicities of stellar populations today are of the order of 145 00:11:52,746 --> 00:11:58,508 solar, which is approximately two percent by mass or maybe one percent by mass if 146 00:11:58,508 --> 00:12:03,120 you really account for all different stars, but so that order. 147 00:12:03,121 --> 00:12:09,110 And for each gram of heavy elements that's made, approximately five grams of helium 148 00:12:09,110 --> 00:12:14,720 that are made The actual numbers are a little iffy and depend on exact models of, 149 00:12:14,720 --> 00:12:19,835 stellar population evolution, but they're of that order of magnitude. 150 00:12:19,835 --> 00:12:25,907 And so we can estimate the net energy from star formation or, energy released by 151 00:12:25,907 --> 00:12:30,799 stars, by taking the mass of all stars formed, times z squared. 152 00:12:31,800 --> 00:12:37,008 Times the efficiency of thermonuclear reactions, which is 0.1% times the 153 00:12:37,008 --> 00:12:42,048 fraction of the hydrogen that was converted into the helium and heavier 154 00:12:42,048 --> 00:12:45,993 metals. And for typical numbers for galaxies 155 00:12:45,993 --> 00:12:52,883 today, that turns out to be an order of magnitude more than what we got from just 156 00:12:52,883 --> 00:12:59,455 release of the binding edge of the order 10 to the 60 ergs, or galaxy like the 157 00:12:59,455 --> 00:13:03,632 Milky Way. So it's really burning of young stars that 158 00:13:03,632 --> 00:13:09,796 will form the most important observe, observable signature of young galaxies, 159 00:13:09,796 --> 00:13:16,512 and incidentally, active galactic nuclei that super massive black holes that they 160 00:13:16,512 --> 00:13:22,308 create material and convert some fraction of it into energy can contribute 161 00:13:22,308 --> 00:13:26,613 comparable amounts as does all of the star formation. 162 00:13:26,613 --> 00:13:32,143 And right away that tells you that their energetics might be important component in 163 00:13:32,143 --> 00:13:36,367 determining structure of galaxies and their star formation. 164 00:13:36,367 --> 00:13:42,891 So those are the generic expectations and next we will take a look at the 165 00:13:42,891 --> 00:13:47,321 observational evidence for galaxy formation.