WEBVTT 1 00:00:03.399 --> 00:00:07.719 Welcome to Bedtime Astronomy. Explore the wonders of the cosmos 2 00:00:07.759 --> 00:00:12.279 with our soothing Bedtime Astronomie podcast. Each episode offers a 3 00:00:12.359 --> 00:00:16.320 gentle journey through the stars, planets, and beyond, perfect for 4 00:00:16.399 --> 00:00:20.239 unwinding after a long day. Let's travel through the mysteries 5 00:00:20.239 --> 00:00:22.440 of the universe as you drift off into a peaceful 6 00:00:22.480 --> 00:00:23.800 slumber under the night sky. 7 00:00:26.960 --> 00:00:30.600 Imagine pointing an antenna up at a completely dark, quiet 8 00:00:30.640 --> 00:00:35.439 patch of the night sky and catching a literal laser beam, 9 00:00:35.960 --> 00:00:39.200 and not just any laser beam, but one fired when 10 00:00:39.240 --> 00:00:41.679 the universe was just this chaotic toddler. 11 00:00:42.039 --> 00:00:44.320 Yeah, it's pretty wild to think about, right. 12 00:00:44.079 --> 00:00:47.159 I mean, we are talking about an undeniable signal that 13 00:00:47.200 --> 00:00:50.600 has been traveling toward Earth for over eight billion years. 14 00:00:51.000 --> 00:00:54.079 This thing originated from an era long before our solar 15 00:00:54.119 --> 00:00:57.000 system even you know, coalesced exactly. 16 00:00:57.079 --> 00:00:59.359 It's ancient history hitting our receivers right now. 17 00:01:00.000 --> 00:01:03.679 Hey, we're unpacking a record breaking natural space laser. It 18 00:01:03.719 --> 00:01:06.760 was detected by the Meerkat radio telescope in South Africa, 19 00:01:07.000 --> 00:01:09.159 and we're going to look at the mechanical, the quantum 20 00:01:09.640 --> 00:01:13.359 and the astrophysical realities of how you even capture a 21 00:01:13.400 --> 00:01:14.680 signal from the deep past. 22 00:01:14.840 --> 00:01:17.560 And we'll get into what a hydroxyl megamaser actually is too, 23 00:01:17.599 --> 00:01:19.200 because that's a huge part of the puzzle. 24 00:01:19.560 --> 00:01:23.959 Yes, definitely, and ultimately we want to figure out what 25 00:01:24.239 --> 00:01:28.079 the violent collision of two ancient galaxies reveals about the 26 00:01:28.120 --> 00:01:32.560 structural evolution of the cosmos. So you know, for anyone 27 00:01:32.599 --> 00:01:35.599 listening who has ever felt totally overwhelmed by the sheer 28 00:01:35.640 --> 00:01:38.159 scale of the universe, you're about to get a front 29 00:01:38.239 --> 00:01:40.680 row seat to an ultimate cosmic time machine. 30 00:01:41.079 --> 00:01:43.359 It really is a time machine. I mean, the detection 31 00:01:43.480 --> 00:01:49.319 of this specific signal fundamentally alters our timeline for observational astronomy. Also, well, 32 00:01:49.519 --> 00:01:52.000 we're looking at a system situated at a redshift of 33 00:01:52.200 --> 00:01:53.159 roughly z equals one. 34 00:01:53.200 --> 00:01:54.760 Okay, let's unpack that for a second. 35 00:01:54.799 --> 00:01:57.359 Sure, so at that distance, we aren't just observing a 36 00:01:57.400 --> 00:02:00.799 distant object. We're observing the universe and it was less 37 00:02:00.799 --> 00:02:02.560 than half its current age, which is just. 38 00:02:02.560 --> 00:02:04.599 A mind boggling scale of time. 39 00:02:04.480 --> 00:02:05.120 It really is. 40 00:02:05.439 --> 00:02:08.000 And the fact that we can capture the specific quantum 41 00:02:08.039 --> 00:02:10.719 signature of a molecule from that epoch and do so 42 00:02:10.919 --> 00:02:14.000 in a fraction of the time previously thought possible, it 43 00:02:14.080 --> 00:02:16.680 proves that our transition into this new era of wide 44 00:02:16.719 --> 00:02:20.759 field radio astronomy is already yielding structural insights into the 45 00:02:20.759 --> 00:02:21.479 cosmic dawn. 46 00:02:21.680 --> 00:02:24.719 Let's start right there, actually the concept of redshift and 47 00:02:24.800 --> 00:02:27.759 looking back in time, because I mean, for anyone who 48 00:02:27.800 --> 00:02:30.800 follows astrophysics, the basic idea that light has a speed 49 00:02:30.800 --> 00:02:33.719 limit is pretty well understood, right, Well, it takes time 50 00:02:33.719 --> 00:02:36.879 to travel. Yeah, so we know that looking across space 51 00:02:37.080 --> 00:02:39.039 is looking back in time. It's like, well, it's like 52 00:02:39.039 --> 00:02:41.639 a letter in the mail. If a friend overseas sends 53 00:02:41.680 --> 00:02:44.240 you a letter, the news is old by the time 54 00:02:44.280 --> 00:02:45.159 you actually read it. 55 00:02:45.240 --> 00:02:46.639 That's a great analogy. Yeah. 56 00:02:46.439 --> 00:02:48.879 Yeah, in space, the light is the letter exactly. 57 00:02:49.439 --> 00:02:52.199 But what I want to dig into is the physical 58 00:02:52.240 --> 00:02:56.199 reality of what an eight billion year transit does to 59 00:02:56.280 --> 00:02:59.639 that letter. Because when a galaxy at z equals one 60 00:03:00.120 --> 00:03:03.520 fires off a burst of radio waves, that wave has 61 00:03:03.560 --> 00:03:07.759 to travel through a universe that is actively expanding. Right, Yes, 62 00:03:07.800 --> 00:03:09.960 the space itself is expanding, So by the time it 63 00:03:10.000 --> 00:03:12.400 hits the Meerkat telescope, we aren't just dealing with a 64 00:03:12.439 --> 00:03:15.400 faint signal. We were dealing with a stretched one. 65 00:03:15.639 --> 00:03:19.000 That is the crucial mechanical difference between local astronomy and 66 00:03:19.080 --> 00:03:23.439 cosmological astronomy. It's called cosmological red shift. The expansion of 67 00:03:23.439 --> 00:03:27.560 the universe is a metric expansion of space itself. So 68 00:03:27.680 --> 00:03:31.759 as the radiophotons emitted by this ancient galaxy travel toward Earth, 69 00:03:32.199 --> 00:03:35.120 the physical space they're traveling through stretches. 70 00:03:34.840 --> 00:03:37.319 And so the photon gets stretched with it exactly. 71 00:03:37.879 --> 00:03:41.639 Consequently, the actual physical wavelength of the photon is stretched out, 72 00:03:42.000 --> 00:03:43.400 which lowers its frequency. 73 00:03:43.759 --> 00:03:47.719 Okay, So if a molecule in that distant galaxy emits 74 00:03:47.879 --> 00:03:51.479 a radio wave at a specific resting frequency, say like 75 00:03:51.800 --> 00:03:54.759 sixteen hundred and sixty five megahertz, it. 76 00:03:54.759 --> 00:03:57.000 Absolutely does not arrive at Earth at sixteen sixty five 77 00:03:57.039 --> 00:03:59.639 mega it says the stretching right, Because it has traveled 78 00:03:59.639 --> 00:04:02.639 for eight billion years. The expansion of the universe is 79 00:04:02.719 --> 00:04:06.159 stretched that wave down to around half of its original frequency. 80 00:04:06.479 --> 00:04:09.800 Wow. So the engineers and astronomers running meerkat they had 81 00:04:09.840 --> 00:04:13.400 to tune their receivers to a completely different frequency band 82 00:04:13.439 --> 00:04:15.000 than they would use if they were looking for that 83 00:04:15.080 --> 00:04:17.360 exact same molecule in our own Milky Way. 84 00:04:17.439 --> 00:04:18.040 Yes, they did. 85 00:04:18.439 --> 00:04:21.079 They had to mathematically predict how much the universe had 86 00:04:21.079 --> 00:04:25.160 expanded over those eight billion years, calculate the resulting frequency drop, 87 00:04:25.519 --> 00:04:26.759 and literally just listen. 88 00:04:26.560 --> 00:04:29.000 There that is. I mean, that's just an insane level 89 00:04:29.000 --> 00:04:29.600 of precision. 90 00:04:29.800 --> 00:04:33.720 It's incredible, And this tells us something profound about the 91 00:04:33.839 --> 00:04:38.439 environment we are observing too, Like what, well, when we 92 00:04:38.480 --> 00:04:41.519 look at a galaxy at a redshift of one, we 93 00:04:41.639 --> 00:04:44.560 are looking at an era often referred to as the 94 00:04:44.639 --> 00:04:46.279 tail end of cosmic noon. 95 00:04:46.600 --> 00:04:48.759 Cosmic noon. I love that term. 96 00:04:48.839 --> 00:04:51.519 It's a great descriptor. See in the local universe, the 97 00:04:51.519 --> 00:04:55.519 one immediately around us. Today, galaxies are relatively. 98 00:04:55.120 --> 00:04:57.319 Quiet, like our Milky Way exactly. 99 00:04:57.800 --> 00:05:00.360 The Milky Way forms a few solar masses worth of 100 00:05:00.399 --> 00:05:04.199 stars per year. The gas fractions are lower. The metallicity, 101 00:05:04.240 --> 00:05:07.680 which is the amount of elements heavier than helium, is higher. 102 00:05:07.480 --> 00:05:10.560 Right, because generations of stars have already lived and died 103 00:05:10.680 --> 00:05:12.800 and seated the galaxy with heavy elements. 104 00:05:13.000 --> 00:05:16.600 Spot On and major galaxy mergers are relatively infrequent now. 105 00:05:16.920 --> 00:05:19.720 But at z equals one, the universe was structurally different. 106 00:05:19.759 --> 00:05:21.560 It was that chaotic Toddler phase. 107 00:05:21.639 --> 00:05:24.399 Yeah, the gas fractions and galaxies were significantly higher. They 108 00:05:24.399 --> 00:05:27.759 were churning out stars at incredible rates and massive violent 109 00:05:27.800 --> 00:05:31.879 mergers between galaxies were a dominant driver of cosmic evolution. 110 00:05:31.639 --> 00:05:35.240 Which perfectly sets the stage for the specific signal they found. 111 00:05:35.639 --> 00:05:38.000 So they detected a hydroxyl megamaser. 112 00:05:38.160 --> 00:05:38.519 It did. 113 00:05:38.720 --> 00:05:42.319 Now, wait, when I hear the word laser or maser, 114 00:05:42.720 --> 00:05:46.160 the visual is usually a highly collimated beam like a 115 00:05:46.199 --> 00:05:50.000 sci fi blaster or a synthetic optical cavity with mirrors 116 00:05:50.079 --> 00:05:52.800 on either end, bouncing photons back and forth to build 117 00:05:52.879 --> 00:05:53.439 up intensity. 118 00:05:53.519 --> 00:05:55.399 Right, that's the artificial way we build them on Earth. 119 00:05:55.480 --> 00:05:59.240 But galaxy isn't a manufactured machine. So how does a massive, 120 00:05:59.360 --> 00:06:03.600 uncontained cloud of hydroxyl gas maintain the population and version 121 00:06:03.680 --> 00:06:06.800 needed for a laser without the photons just scattering in 122 00:06:06.879 --> 00:06:07.879 random directions. 123 00:06:08.079 --> 00:06:10.720 It's a really good question. It requires a highly specific 124 00:06:10.800 --> 00:06:14.600 set of environmental conditions that only exist in these extreme environments. 125 00:06:14.680 --> 00:06:15.439 Okay, laid on me. 126 00:06:15.720 --> 00:06:17.800 To understand the mechanism, we really have to look at 127 00:06:17.800 --> 00:06:21.240 the quantum mechanics of the hydroxyl radical, which is just 128 00:06:21.319 --> 00:06:24.199 one oxygen atom bonded to one hydrogen atom. 129 00:06:24.240 --> 00:06:25.399 Okay, oh, which got it? 130 00:06:25.639 --> 00:06:28.519 In its ground state, the hydroxyl molecule has four very 131 00:06:28.519 --> 00:06:31.959 close energy levels. We call this hyperfine. 132 00:06:31.399 --> 00:06:32.879 Structure hyperfine structure. Yeah. 133 00:06:32.920 --> 00:06:35.800 Yeah, These energy levels are determined by the interaction between 134 00:06:35.800 --> 00:06:38.639 the magnetic field of the molecule's rotation and the spin 135 00:06:38.720 --> 00:06:41.800 of its unpaired electron and the hydrogen nucleus. 136 00:06:41.959 --> 00:06:45.759 So it's like this incredibly delicate magnetic balancing act inside 137 00:06:45.759 --> 00:06:47.079 the molecule exactly. 138 00:06:47.199 --> 00:06:50.240 And when the molecule drops from a higher hyperfine state 139 00:06:50.319 --> 00:06:53.639 to a lower one, it emits a microwave photon at 140 00:06:53.680 --> 00:06:55.639 a very specific frequency. 141 00:06:55.199 --> 00:06:57.319 And those are the sixteen sixty five and sixteen sixty 142 00:06:57.319 --> 00:06:59.720 seven megahertz frequency, as you mentioned earlier. 143 00:06:59.399 --> 00:07:00.519 Those are the most dominant ones. 144 00:07:00.600 --> 00:07:03.360 Yes, But just having the gas there isn't enough to 145 00:07:03.480 --> 00:07:07.399 create a maser, right, because under normal thermal conditions wouldn't 146 00:07:07.399 --> 00:07:09.959 most of the molecules just sit in the lowest possible 147 00:07:10.079 --> 00:07:11.160 energy state, right. 148 00:07:11.040 --> 00:07:12.839 They just be lazy. You need something to push him up. 149 00:07:12.879 --> 00:07:14.560 You need a pumping mechanism. 150 00:07:14.240 --> 00:07:17.399 A pump, Okay, what acts as a pump in outer space? 151 00:07:17.560 --> 00:07:19.800 In the case of a hydroxyl megamaser, the pump is 152 00:07:19.879 --> 00:07:21.879 usually far infrared. 153 00:07:21.480 --> 00:07:24.120 Radiation infrared so heat basically. 154 00:07:24.199 --> 00:07:28.519 Yeah, in a violently merging galaxy, you have colossal bursts 155 00:07:28.560 --> 00:07:32.720 of star formation and often an actively feeding supermassive black hole, 156 00:07:33.240 --> 00:07:36.040 an active galactic nucleus or agn. 157 00:07:35.759 --> 00:07:38.879 Right, because everything is crashing together and igniting. 158 00:07:38.639 --> 00:07:41.879 Exactly, and these intense energy sources heat up the massive 159 00:07:41.879 --> 00:07:46.480 clouds of surrounding dust. That warm dust then radiates profusely 160 00:07:46.519 --> 00:07:47.480 in the far infrared. 161 00:07:47.759 --> 00:07:51.319 Okay, so the dust is glowing with infrared light. 162 00:07:51.480 --> 00:07:55.839 And the hydroxyl molecules absorb those specific infrared photons, which 163 00:07:55.839 --> 00:07:58.680 excites them into higher rotational energy states. 164 00:07:58.879 --> 00:07:59.879 They soak up the heat. 165 00:08:00.120 --> 00:08:03.199 Yeah, but they don't stay there, as they quickly decay 166 00:08:03.240 --> 00:08:06.360 back down. They don't just return to a random distribution. 167 00:08:06.959 --> 00:08:11.079 The quantum mechanics of the decay preferentially populate the upper 168 00:08:11.079 --> 00:08:13.920 states of that ground level hyperfine structure. 169 00:08:14.000 --> 00:08:16.800 Oh I see, so it artificially loads up the higher 170 00:08:16.920 --> 00:08:17.600 energy levels. 171 00:08:17.680 --> 00:08:21.519 Yes, this creates the population inversion. You now have more 172 00:08:21.560 --> 00:08:24.399 molecules in the high energy state than the low energy state. 173 00:08:24.600 --> 00:08:27.120 So the trap is set. You've got this entire molecular 174 00:08:27.160 --> 00:08:31.000 cloud just sitting there artificially elevated, waiting for a trigger. 175 00:08:31.199 --> 00:08:34.360 Yes, and the trigger is a single photon passing through 176 00:08:34.360 --> 00:08:35.759 the gas at exactly. 177 00:08:35.399 --> 00:08:36.159 The right frequency. 178 00:08:36.320 --> 00:08:39.159 Just one photon, just one to start the avalanche. When 179 00:08:39.200 --> 00:08:43.159 that initial photon interacts with an excited hydroxyl molecule, it 180 00:08:43.240 --> 00:08:45.919 stimulates a molecule to drop to the lower energy. 181 00:08:45.679 --> 00:08:47.679 State, and then it releases its own. 182 00:08:47.480 --> 00:08:51.399 Photon, right but releasing a second photon that is perfectly 183 00:08:51.440 --> 00:08:54.279 in phase, moving in the exact same direction, but the 184 00:08:54.320 --> 00:08:56.360 exact same frequency is the first one. 185 00:08:56.480 --> 00:08:58.480 Ah, So one becomes two. 186 00:08:58.440 --> 00:08:59.840 Two become four, four become eight. 187 00:09:00.039 --> 00:09:02.759 It is an exponential cascade of stimulated emission. 188 00:09:02.799 --> 00:09:04.519 That is so cool. But wait, I want to address 189 00:09:04.519 --> 00:09:08.159 the mirror question. How does it become a directional beam 190 00:09:08.200 --> 00:09:11.240 that we can actually detect from eight billion light years away? 191 00:09:11.320 --> 00:09:12.320 Right without the mirrors? 192 00:09:12.440 --> 00:09:15.000 Yeah, without mirrors to bounce the light back and forth. 193 00:09:15.039 --> 00:09:17.960 Doesn't the amplification just kind of happen in every direction 194 00:09:18.159 --> 00:09:19.440 equally like a light bulb. 195 00:09:19.679 --> 00:09:22.080 You'd think so, But it comes down to the geometry 196 00:09:22.120 --> 00:09:25.159 and velocity of the gas clouds. While there are no 197 00:09:25.240 --> 00:09:30.159 physical mirrors, amplification only happens along paths where the gas 198 00:09:30.360 --> 00:09:31.759 is velocity coherent. 199 00:09:32.080 --> 00:09:34.960 Velocity coherent meaning it's all moving at the same speed. 200 00:09:35.120 --> 00:09:38.519 Essentially, if the gas is swirling chaotically, the Doppler shift 201 00:09:38.519 --> 00:09:41.159 from the relative motion of the molecules will change the 202 00:09:41.200 --> 00:09:44.200 frequency of the photons from the perspective of the molecules themselves. 203 00:09:44.200 --> 00:09:47.240 Oh right, The pitch shifts like an ambulance siren. 204 00:09:47.279 --> 00:09:51.440 Exactly, and that shift breaks the resonance and stops the 205 00:09:51.480 --> 00:09:53.120 cascade dead in its tracks. 206 00:09:53.440 --> 00:09:56.080 So it can't amplify if the gas is turbulent. 207 00:09:56.320 --> 00:09:56.559 Right. 208 00:09:57.320 --> 00:09:59.639 However, if you look at a massive disc of gas 209 00:09:59.759 --> 00:10:03.559 edge on, or if there's a long linear filament of 210 00:10:03.639 --> 00:10:06.600 gas moving at a uniform velocity relative to the line 211 00:10:06.600 --> 00:10:07.639 of sight, then. 212 00:10:07.519 --> 00:10:11.120 The photons can travel immense distances through the cloud without 213 00:10:11.120 --> 00:10:13.159 being Doppler shifted out of resonance. 214 00:10:13.279 --> 00:10:13.799 You got it. 215 00:10:13.840 --> 00:10:17.279 So the maser beam naturally collimates along the paths of 216 00:10:17.320 --> 00:10:21.799 greatest velocity coherence. It's essentially beaming its signal along very 217 00:10:21.840 --> 00:10:24.360 specific geometric sidelines. 218 00:10:24.039 --> 00:10:26.639 And we just happen to be sitting exactly on one 219 00:10:26.639 --> 00:10:29.039 of those sitelines. That is incredible cosmic. 220 00:10:28.679 --> 00:10:29.480 Luck, it is. 221 00:10:29.960 --> 00:10:32.039 But keep in mind, the scale of this emission is 222 00:10:32.039 --> 00:10:34.879 what justifies the mega prefix in megamser. 223 00:10:35.120 --> 00:10:37.840 Right, because standard masers exist in our own milky way, 224 00:10:37.879 --> 00:10:40.240 don't They usually around individual forming stars? 225 00:10:40.279 --> 00:10:43.559 They do, But a megamser is millions of times more 226 00:10:43.639 --> 00:10:44.960 luminous than those local ones. 227 00:10:44.960 --> 00:10:46.679 So this isn't just one star forming. 228 00:10:47.080 --> 00:10:50.320 No, this is the collective scream of an entire galaxy 229 00:10:50.440 --> 00:10:54.799 undergoing a catastrophic transformation. The energy required to pump a 230 00:10:54.799 --> 00:10:57.159 cloud of hydroxyl gas to the point where becomes a 231 00:10:57.159 --> 00:10:58.919 megamaver is staggering. 232 00:10:59.039 --> 00:11:00.799 It must take a galaxy ending event. 233 00:11:00.919 --> 00:11:04.679 It requires the systemic disruption of the entire galaxy. This 234 00:11:04.799 --> 00:11:08.720 is why hydroxyl megamasers are almost exclusively found in ultra 235 00:11:08.840 --> 00:11:12.600 luminous infrared galaxies or ulargies eulargies. 236 00:11:13.120 --> 00:11:17.080 Astronomers and their acronyms I swear, hey it's efficient, but 237 00:11:17.279 --> 00:11:21.360 ulargies are overwhelmingly the result of major galaxy mergers. 238 00:11:21.759 --> 00:11:25.759 The detection of this hydroxyl signature is an unequivocal indicator 239 00:11:25.919 --> 00:11:29.240 that we are looking at a site of masses cosmic trauma. 240 00:11:29.120 --> 00:11:31.600 Which brings us to the operational reality of how this 241 00:11:31.720 --> 00:11:34.960 was actually detected. The signal was caught by the Meerkat 242 00:11:35.080 --> 00:11:38.679 telescope in South Africa in just five hours of observing. 243 00:11:38.279 --> 00:11:40.240 Time, which is astonishingly fast. 244 00:11:40.320 --> 00:11:42.879 I am really stuck on that five hour timeframe, even 245 00:11:42.879 --> 00:11:45.960 with massive bandwidth in the inherent brightness of a mega maser. 246 00:11:46.279 --> 00:11:49.320 Gathering enough photons from eight billion light years away in 247 00:11:49.360 --> 00:11:51.759 a single afternoon seems physically impossible. 248 00:11:51.840 --> 00:11:54.559 Certainly challenges our traditional expectations. 249 00:11:54.399 --> 00:11:57.600 Right, because the inverse square law dictates that as the 250 00:11:57.600 --> 00:12:02.240 signal travels, its energy is distributed over an ever expanding sphere. 251 00:12:02.639 --> 00:12:05.159 By the time it reaches Earth, the flux density should 252 00:12:05.200 --> 00:12:07.919 be buried so deep in the thermal noise floor that 253 00:12:07.960 --> 00:12:11.000 a five hour integration shouldn't even register a blip. 254 00:12:11.360 --> 00:12:11.919 That's the math. 255 00:12:12.000 --> 00:12:16.080 Yeah, So what is Meerkat doing differently that breaks that expectation? 256 00:12:16.679 --> 00:12:19.080 How is it drinking from a fire hose of cosmic 257 00:12:19.120 --> 00:12:21.080 noise and actually finding something? 258 00:12:21.480 --> 00:12:21.639 Well? 259 00:12:21.639 --> 00:12:25.559 Meerkat is a marvel of modern radio interferometry, and to 260 00:12:25.600 --> 00:12:27.879 really understand its efficiency we need to look at both 261 00:12:27.919 --> 00:12:31.000 its physical architecture and its computational back end. 262 00:12:31.080 --> 00:12:32.320 Okay, let's start with the physical. 263 00:12:32.559 --> 00:12:35.320 Meerkat is not a single giant dish. It is an 264 00:12:35.360 --> 00:12:39.279 array of sixty four individual antennas, each thirteen point five 265 00:12:39.399 --> 00:12:42.879 meters across, spread out over distances of up to eight kilometers. 266 00:12:42.919 --> 00:12:45.080 Oh wow, eight kilometers yeah, And. 267 00:12:45.039 --> 00:12:48.080 By linking these antennas together, astronomers use a technique called 268 00:12:48.120 --> 00:12:51.360 aperture synthesis to create a virtual telescope that has the 269 00:12:51.399 --> 00:12:54.480 resolving power of a single dish eight kilometers wide. 270 00:12:54.720 --> 00:12:58.519 Right, Because in radio astronomy, the resolution is dictated by 271 00:12:59.000 --> 00:13:02.519 the maximum dist between your antennas, which are called the baselines. 272 00:13:03.080 --> 00:13:05.960 But resolving power just gives you a sharper image, right, 273 00:13:06.399 --> 00:13:09.840 It doesn't inherently give you the raw sensitivity to pull 274 00:13:09.879 --> 00:13:12.960 a hyper faint signal out of the noise in five hours. 275 00:13:13.200 --> 00:13:13.639 You're right. 276 00:13:14.120 --> 00:13:17.000 Sensitivity is a function of the total collecting area, the 277 00:13:17.000 --> 00:13:19.320 system temperature of the receivers. 278 00:13:19.039 --> 00:13:19.960 And the bandwidth. 279 00:13:20.399 --> 00:13:23.360 And this is where Meerkat absolutely excels. 280 00:13:23.600 --> 00:13:24.720 Okay, break that down for me. 281 00:13:24.919 --> 00:13:28.840 Well, first, the receivers on these antennas are cryogenically cooled, 282 00:13:28.960 --> 00:13:31.759 like sub zero cool, very sub zero. This is done 283 00:13:31.799 --> 00:13:35.600 to reduce the internal thermal noise, literally the random jitter 284 00:13:35.679 --> 00:13:38.559 of electrons in the electronics to an absolute minimal. 285 00:13:38.720 --> 00:13:39.039 Got it. 286 00:13:39.120 --> 00:13:41.120 So the equipment itself is whisper. 287 00:13:40.840 --> 00:13:43.879 Quiet, right, But the real leap forward is the correlator 288 00:13:43.879 --> 00:13:48.200 and the bandwidth. Meerkat can digitize and process an enormous 289 00:13:48.240 --> 00:13:50.440 swath of the radio spectrum simultaneously. 290 00:13:50.759 --> 00:13:52.840 I really want to focus on the data processing side 291 00:13:52.840 --> 00:13:55.279 of this, because capturing the deity is only half the battle. 292 00:13:55.320 --> 00:13:57.879 It's arguably the easier half, honestly, right. 293 00:13:57.919 --> 00:14:01.639 If meerkat has a massive bandwidth and is incredibly sensitive, 294 00:14:01.679 --> 00:14:05.039 it means it is picking up everything everything. It's capturing 295 00:14:05.039 --> 00:14:08.879 the ancient hydroxyl megamaser, sure, but it's also capturing the 296 00:14:08.960 --> 00:14:13.039 microwave background, the synchrotron radiation from our own galaxy, and 297 00:14:13.080 --> 00:14:18.240 critically all the terrestrial radio frequency interference, the RFI. 298 00:14:18.399 --> 00:14:20.480 Bane of radio astronomers everywhere. 299 00:14:20.519 --> 00:14:25.720 Seriously, we are talking about satellite communications, GPS, cell towers, 300 00:14:25.919 --> 00:14:29.720 airplane radar. How do you isolate an eight billion year 301 00:14:29.759 --> 00:14:31.879 old windsbur when you are sitting in the middle of 302 00:14:31.879 --> 00:14:33.600 a planetary scale rock concert. 303 00:14:33.759 --> 00:14:37.399 That is the defining challenge of modern radio astronomy. And 304 00:14:37.440 --> 00:14:39.720 the raw data rate from the meercat correlator is on 305 00:14:39.759 --> 00:14:40.279 the order of. 306 00:14:40.279 --> 00:14:43.279 Terabytes parabytes, just constantly flowing in constantly. 307 00:14:43.639 --> 00:14:45.480 To process this, the data is sent to the inter 308 00:14:45.639 --> 00:14:50.240 University Institute for Data Intensive Astronomy, or Ida, where high 309 00:14:50.279 --> 00:14:53.679 performance computing clusters run complex calibration pipelines. 310 00:14:54.039 --> 00:14:56.840 So if we use late noise canceling headphones as a model, 311 00:14:57.360 --> 00:15:00.440 is the algorithm basically generating an inverse way of the 312 00:15:00.559 --> 00:15:05.440 terrestrial noise the satellites, the cell phones and subtracting it 313 00:15:05.440 --> 00:15:08.679 from the raw feed, leaving only the ancient cosmic whisper. 314 00:15:09.000 --> 00:15:11.720 That's well, it's closer to the reality of the software 315 00:15:11.720 --> 00:15:14.799 pipeline than a simple filtering analogy, but it's more complex 316 00:15:14.840 --> 00:15:18.799 than that. How So, the correlator and the subsequent software 317 00:15:18.799 --> 00:15:23.240 pipelines have to identify the specific phase and amplitude signatures 318 00:15:23.240 --> 00:15:24.480 of terrestrial. 319 00:15:24.080 --> 00:15:25.840 RFI phase and amplitides. 320 00:15:26.360 --> 00:15:30.440 Because terrestrial sources are very nearby, their wavefronts are curved 321 00:15:30.840 --> 00:15:33.840 when they hit the array across that eight kilometer distance, 322 00:15:33.879 --> 00:15:35.120 and they appear localized. 323 00:15:35.519 --> 00:15:38.360 Ah okay, like ripples in a pond radiating from. 324 00:15:38.240 --> 00:15:39.840 A rock you just throw exactly. 325 00:15:40.159 --> 00:15:43.200 Conversely, a plane wave arriving from deep space from eight 326 00:15:43.240 --> 00:15:46.480 billion light years away hits the array with a completely 327 00:15:46.480 --> 00:15:48.679 different flat phase signature. 328 00:15:48.279 --> 00:15:50.679 Because it's traveled so far that the curve has essentially 329 00:15:50.759 --> 00:15:51.360 flattened out. 330 00:15:51.399 --> 00:15:55.840 Precisely, the algorithms analyze the data in the time frequency domain, 331 00:15:56.279 --> 00:15:59.440 searching for spikes in power that violate the expected fringe 332 00:15:59.480 --> 00:16:01.159 patterns of celestial. 333 00:16:00.679 --> 00:16:03.840 Source, and then it just flags them, It flags. 334 00:16:03.480 --> 00:16:08.159 Them or mathematically subtracts that interference. It requires trillions of 335 00:16:08.200 --> 00:16:11.360 floating point operations per second just to clean the data 336 00:16:11.440 --> 00:16:13.759 before you can even begin to look for the science target. 337 00:16:14.000 --> 00:16:16.480 It's like a high tech car wash for data, a high. 338 00:16:16.360 --> 00:16:21.440 Tech car wash running on supercomputers. That's wild. And the 339 00:16:21.480 --> 00:16:25.440 wide bandwidth allowed them to do something unprecedented during this observation. Right, 340 00:16:25.720 --> 00:16:28.200 they didn't just find the red shifted hydroxyl line. 341 00:16:28.279 --> 00:16:29.279 No, they didn't. 342 00:16:29.679 --> 00:16:32.320 Because they were capturing such a massive chunk of the spectrum, 343 00:16:32.519 --> 00:16:36.799 they simultaneously detected the red shifted neutral hydrogen absorption line. 344 00:16:36.799 --> 00:16:38.440 Simultaneous detection. Wow. 345 00:16:38.519 --> 00:16:41.879 Yes, And the simultaneous detection of both the neutral hydrogen 346 00:16:41.960 --> 00:16:46.679 or HI and the hydroxyl OH is a massive observational victory. 347 00:16:46.720 --> 00:16:48.039 Why is it such a big deal. 348 00:16:48.200 --> 00:16:51.200 Well, neutral hydrogen emits and absorbs radio waves at a 349 00:16:51.240 --> 00:16:54.480 resting frequency of fourteen hundred and twenty megahertz. Hydroxyl as 350 00:16:54.480 --> 00:16:57.200 we discussed as around sixteen sixty five and sixteen sixty 351 00:16:57.240 --> 00:16:58.360 seven megahertz, right. 352 00:16:58.279 --> 00:17:00.799 So they are pretty far apart on the dial exactly. 353 00:17:01.039 --> 00:17:04.079 In older narrowband telescopes, you would have to run a 354 00:17:04.079 --> 00:17:08.359 dedicated observation tuned to the red shifted HI frequency, and 355 00:17:08.640 --> 00:17:12.640 a completely separate observation tuned to the red shifted oage frequency. 356 00:17:12.240 --> 00:17:15.200 Which takes twice as long, and the conditions might change. 357 00:17:15.279 --> 00:17:19.440 Right, But Meerkat's fractional bandwidth is so wide that both 358 00:17:19.440 --> 00:17:23.880 of these vastly separated, heavily red shifted lines fell perfectly 359 00:17:23.960 --> 00:17:25.759 within a single observing window. 360 00:17:26.039 --> 00:17:28.279 But why is capturing both of them at the exact 361 00:17:28.359 --> 00:17:29.960 same time so critical? 362 00:17:30.519 --> 00:17:30.599 Like? 363 00:17:30.720 --> 00:17:34.079 What does the combination of hydrogen absorption and hydroxyl emission 364 00:17:34.400 --> 00:17:36.559 tell us that we couldn't figure out from just one 365 00:17:36.640 --> 00:17:37.039 or the other. 366 00:17:37.279 --> 00:17:40.559 It breaks the degeneracy of the galaxy's internal kinematics. 367 00:17:40.680 --> 00:17:42.119 Okay, English, please. 368 00:17:41.880 --> 00:17:42.440 Right, sorry. 369 00:17:42.599 --> 00:17:45.640 The hydroxyl megamaser is an emission line. It is adding 370 00:17:45.720 --> 00:17:48.359 energy to the spectrum. It tells us where the dense, 371 00:17:48.480 --> 00:17:52.440 highly excited molecular gas is, which usually traces the intense 372 00:17:52.480 --> 00:17:55.920 starburst regions or the immediate vicinity of the act of 373 00:17:55.920 --> 00:17:56.920 galactic nucleus. 374 00:17:56.960 --> 00:17:59.640 So the emission shows us where the fire is exactly. 375 00:18:00.039 --> 00:18:01.720 Neutral hydrogen, however, was seen. 376 00:18:01.559 --> 00:18:03.480 In absorption, meaning it's blocking light. 377 00:18:03.960 --> 00:18:08.279 Yes, this means there's a massive cloud of cold, unexcited 378 00:18:08.319 --> 00:18:11.559 atomic gas sitting in front of the radio continuum source 379 00:18:11.599 --> 00:18:13.960 of the galaxy absorbing the background light. 380 00:18:14.200 --> 00:18:16.480 Oh wow, so you have a cold cloud casting a 381 00:18:16.519 --> 00:18:18.400 shadow over the glowing core. 382 00:18:18.279 --> 00:18:21.400 Spot on and by measuring the Doppler shift of the 383 00:18:21.400 --> 00:18:24.519 emission line against the Doppler shift of the absorption line, 384 00:18:24.680 --> 00:18:27.960 astronomers can map the physical geometry of the merger. 385 00:18:28.039 --> 00:18:30.039 Because they can see what parts are moving toward us 386 00:18:30.039 --> 00:18:31.079 in which are moving away. 387 00:18:31.359 --> 00:18:35.319 Yes, they can literally see the cold gas falling inward 388 00:18:35.400 --> 00:18:39.480 toward the core, fueling the massive starburst that is simultaneously 389 00:18:39.519 --> 00:18:40.559 firing off the maser. 390 00:18:40.839 --> 00:18:43.640 So it provides a three dimensional anatomical map of a 391 00:18:43.720 --> 00:18:46.240 galaxy in the throes of a catastrophic collision. 392 00:18:46.359 --> 00:18:47.400 That's exactly what it does. 393 00:18:47.480 --> 00:18:49.640 That is just phenomenal. But wait, I have to go 394 00:18:49.720 --> 00:18:53.039 back to the five hour thing, because even with cryogenically 395 00:18:53.039 --> 00:18:57.680 cooled receivers, an eight kilometer virtual aperture, and supercomputers running 396 00:18:57.759 --> 00:19:01.559 terabytes of phase referencing algorithms, the math for a five 397 00:19:01.599 --> 00:19:05.000 hour detection of a z equals one megamser still feels 398 00:19:05.039 --> 00:19:05.920 pushed beyond. 399 00:19:05.680 --> 00:19:07.319 The limit your right to be skeptical. 400 00:19:07.480 --> 00:19:10.400 Yeah, the inverse square law is unforgiving. There had to 401 00:19:10.400 --> 00:19:14.440