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 00:00:26.879 --> 00:00:29.800 So I want you to imagine standing right in the 8 00:00:29.839 --> 00:00:33.119 middle of a pitch black football stadium, Okay, totally dart 9 00:00:33.240 --> 00:00:35.520 right and at the far into the field. Someone just 10 00:00:35.520 --> 00:00:40.159 flips a switch and every single high intensity floodlight turns 11 00:00:40.200 --> 00:00:43.759 on simultaneously. Oh wow, and they're all aimed directly into 12 00:00:43.759 --> 00:00:47.320 your eyes like it is blinding physical pain. 13 00:00:47.439 --> 00:00:49.079 Right that sounds. 14 00:00:49.799 --> 00:00:53.240 Now, someone hands you a camera and your job is 15 00:00:53.280 --> 00:00:57.719 to take a perfectly clear, like sharply focused photograph of 16 00:00:57.759 --> 00:00:59.520 a single firefly. 17 00:00:59.240 --> 00:01:02.039 A firefly in front of the floodlights. 18 00:01:01.520 --> 00:01:04.640 Exactly buzzing about I don't know, one inch away from 19 00:01:04.640 --> 00:01:05.799 the glass of one of those lights. 20 00:01:05.840 --> 00:01:08.120 That is, I mean, that's literally impossible, right. 21 00:01:08.239 --> 00:01:10.560 You have to capture the faint, delicate glow of this 22 00:01:10.680 --> 00:01:14.120 tiny bug through this actual wall of light that is 23 00:01:14.159 --> 00:01:17.400 actively overwhelming your retinas and your cameras sensors, and. 24 00:01:17.799 --> 00:01:20.239 That right there is exactly why we don't know if 25 00:01:20.239 --> 00:01:21.319 we're alone in the universe. 26 00:01:21.400 --> 00:01:26.560 Yet exactly that impossible blinding task is the reality of 27 00:01:26.599 --> 00:01:30.760 modern astronomy. We've been staring at the night sky for millennia, 28 00:01:31.120 --> 00:01:33.519 you know, asking if there's life out there, but we're 29 00:01:33.519 --> 00:01:35.879 effectively just staring straight into those flood lights. 30 00:01:35.959 --> 00:01:38.519 Yeah, the fireflies are out there, but we just can't see. 31 00:01:38.319 --> 00:01:40.959 Them, right, And that's really what we're getting into today 32 00:01:41.400 --> 00:01:44.120 because the transition happening right now in astronomy is just 33 00:01:44.159 --> 00:01:48.200 it's wild. Wondering about those fireflies is no longer enough. 34 00:01:48.280 --> 00:01:51.120 No, it's not. I mean, for most of human history, 35 00:01:51.239 --> 00:01:54.359 asking you know, are we alone? That was a question 36 00:01:54.400 --> 00:01:56.120 strictly for philosopher. 37 00:01:55.680 --> 00:01:57.200 Or theologians or science fiction writers. 38 00:01:57.319 --> 00:02:00.799 Exactly you'd sit around a campfire, look up and just guess. 39 00:02:01.319 --> 00:02:03.920 But the era of guessing is basically over, which is 40 00:02:03.959 --> 00:02:08.280 so exciting. It really is, that profound existential question has 41 00:02:08.319 --> 00:02:11.039 officially been translated into an engineering problem. 42 00:02:11.080 --> 00:02:13.080 An engineering problem. I love that, right. 43 00:02:13.439 --> 00:02:16.080 We're no longer debating whether life might exist. We are 44 00:02:16.080 --> 00:02:19.599 physically building the machines to filter out those floodlights and 45 00:02:19.639 --> 00:02:23.080 get a definitive, undeniable photograph of the answer. 46 00:02:23.039 --> 00:02:26.960 Which means we really need to understand the sheer brutal 47 00:02:27.080 --> 00:02:28.840 physics standing in our way. 48 00:02:28.919 --> 00:02:30.840 Yeah, the physics are not very forgiving. 49 00:02:31.280 --> 00:02:34.000 No they aren't. Because if we want to find life, 50 00:02:34.199 --> 00:02:37.360 astrobiologists are very clear that we need to look for 51 00:02:37.439 --> 00:02:38.840 planets in the habitable zone. 52 00:02:38.960 --> 00:02:40.759 Right, Yeah, the Goldilock zone. 53 00:02:40.479 --> 00:02:43.120 Exactly, not too hot, not too cold, just right for 54 00:02:43.199 --> 00:02:46.439 liquid water. But when you actually break down the mechanics 55 00:02:46.439 --> 00:02:51.120 of a solar system, that requirement the habitable zone is 56 00:02:51.159 --> 00:02:53.919 the exact thing making this almost impossible. 57 00:02:54.000 --> 00:02:56.719 It really is. To understand the obstacle, you have to 58 00:02:56.759 --> 00:02:58.120 look at the geometry of it all. 59 00:02:58.199 --> 00:02:59.520 Okay, break that down for us. 60 00:03:00.039 --> 00:03:02.280 So if a planet is too far from its host star, 61 00:03:02.680 --> 00:03:05.360 any water on the surface just freezes solid. 62 00:03:05.159 --> 00:03:06.039 Right, you get an ice ball. 63 00:03:06.280 --> 00:03:10.000 Yeah, and biochemistry, as we understand it just grinds to 64 00:03:10.080 --> 00:03:13.639 a halt. But if it orbits too close, the stellar 65 00:03:13.759 --> 00:03:16.520 radiation boils the water away into vapor and. 66 00:03:16.520 --> 00:03:18.520 The atmosphere just gets stripped away exactly. 67 00:03:18.599 --> 00:03:21.479 You're left with a scorched rock. So the habitable zone 68 00:03:21.520 --> 00:03:25.759 is this really narrow? Specific orbital band where a rocky 69 00:03:25.840 --> 00:03:29.000 planet can actually maintain liquid water on its surface. 70 00:03:29.319 --> 00:03:31.879 But because it's so narrow and so close to the star. 71 00:03:32.400 --> 00:03:35.719 Right, because it's relatively close, any planet inside that band 72 00:03:36.000 --> 00:03:39.560 is completely engulfed by the star's glare from our vantage point, 73 00:03:39.680 --> 00:03:40.800 you know, light years away. 74 00:03:41.080 --> 00:03:44.719 So proximity is the prerequisite for life, but it's also 75 00:03:44.759 --> 00:03:46.639 the ultimate barrier to us seeing it. 76 00:03:46.759 --> 00:03:48.759 That's exactly it. And you have to factor in the 77 00:03:48.800 --> 00:03:50.039 scale of brightness too. 78 00:03:50.280 --> 00:03:52.800 Oh man, the brightness scale is just absurd. 79 00:03:52.439 --> 00:03:56.840 It's incomprehensible. A host star like something similar to our Sun, 80 00:03:57.439 --> 00:04:01.879 is typically ten billion times brighter than the earth like 81 00:04:01.919 --> 00:04:02.919 planet orbiting it. 82 00:04:03.080 --> 00:04:03.800 Ten billion. 83 00:04:04.039 --> 00:04:07.120 Yeah, I mean, the star is a raging nuclear furnace right, 84 00:04:07.319 --> 00:04:10.479 constant fusion, and the planet is just rock exactly. The 85 00:04:10.520 --> 00:04:14.360 planet produces zero visible light of its own. It only 86 00:04:14.400 --> 00:04:17.839 reflects a microscoptic fraction of the star's light back out. 87 00:04:17.639 --> 00:04:19.680 Into space, which is its albedo right right. 88 00:04:19.680 --> 00:04:23.000 Albido Earth, for example, reflects roughly thirty percent of the 89 00:04:23.040 --> 00:04:24.079 sunlight that hits. 90 00:04:23.879 --> 00:04:26.079 It, mostly from like clouds and ice and. 91 00:04:26.040 --> 00:04:28.160 Stuff, mostly clouds and ice. Yeah, So you are looking 92 00:04:28.160 --> 00:04:31.120 for an object that is ten billion times dimmer than 93 00:04:31.160 --> 00:04:33.240 the nuclear fireball sitting right next to it. 94 00:04:33.519 --> 00:04:37.079 Ten billion to one. I mean, the human brain literally 95 00:04:37.240 --> 00:04:39.360 cannot comprehend a ratio like that. 96 00:04:39.560 --> 00:04:40.240 No, it can't. 97 00:04:40.560 --> 00:04:44.279 We understand one hundred to one, maybe one thousand to one, 98 00:04:44.560 --> 00:04:47.600 but ten billion to one is a number so vast 99 00:04:47.639 --> 00:04:49.759 it just registers as a wall of noise. 100 00:04:50.480 --> 00:04:53.000 And in the realm of optical engineering, that ratio is 101 00:04:53.040 --> 00:04:54.879 your ultimate adversary. 102 00:04:54.639 --> 00:04:56.560 Because the sensors just get overwhelmed. 103 00:04:56.680 --> 00:04:59.759 Right when you point a highly sensitive space telescope at 104 00:04:59.759 --> 00:05:03.319 a d solar system, you are trying to collect these 105 00:05:03.360 --> 00:05:07.920 incredibly rare, precious photons bouncing. 106 00:05:07.480 --> 00:05:10.160 Off that tiny planet, the firefly. 107 00:05:09.879 --> 00:05:13.639 Firefly, But those few planetary photons are traveling alongside an 108 00:05:13.680 --> 00:05:15.759 absolute tsunami of stellar. 109 00:05:15.360 --> 00:05:16.959 Photons, just a wave of light. 110 00:05:17.279 --> 00:05:20.160 And when that wave hits the telescope sensors, the star's 111 00:05:20.279 --> 00:05:23.240 light instantly saturates the image, it blows out the exposure, 112 00:05:23.240 --> 00:05:25.920 creates massive lens flares, and just buries the planet in 113 00:05:26.000 --> 00:05:26.720 digital noise. 114 00:05:26.879 --> 00:05:28.759 See. I look at a problem like that, and my 115 00:05:28.839 --> 00:05:32.279 brain instantly searches for a brute force, low tech solution, 116 00:05:33.120 --> 00:05:35.160 Like what like if I'm driving down the highway and 117 00:05:35.199 --> 00:05:37.199 the sun is blinding me. I just reach up and 118 00:05:37.199 --> 00:05:40.439 pull down the sun visor, right. Or if I'm outside 119 00:05:40.480 --> 00:05:43.120 trying to watch an airplane fly near the sun, I 120 00:05:43.279 --> 00:05:46.279 just hold my thumb up block the sun, and suddenly 121 00:05:46.360 --> 00:05:49.800 I can see the plane. A physical block exactly. We 122 00:05:49.879 --> 00:05:51.639 know exactly where the star is in the sky. We 123 00:05:51.680 --> 00:05:55.319 have these multi billion dollars space telescopes. Why don't we 124 00:05:55.439 --> 00:06:00.439 just engineer a highly precise mechanical thumb to block the starlight. 125 00:06:00.639 --> 00:06:02.720 Well, it's funny you say that, because that is the 126 00:06:02.720 --> 00:06:06.399 most intuitive approach, and it's actually the foundational technology we've 127 00:06:06.480 --> 00:06:07.480 used for decades. 128 00:06:07.639 --> 00:06:09.600 Wait, really, we do have a mechanical thumb. 129 00:06:09.600 --> 00:06:09.920 We do. 130 00:06:10.000 --> 00:06:13.240 It's called a coronagraph. The coronagraph, Okay, yeah, the concept 131 00:06:13.279 --> 00:06:15.639 was actually invented way back in the nineteen thirties by 132 00:06:15.639 --> 00:06:17.600 a French astronomer named Bernard Leo. 133 00:06:17.920 --> 00:06:18.920 In the thirties. 134 00:06:19.279 --> 00:06:22.600 Wow, yeah, he wanted to study our sun's corona, you know, 135 00:06:22.680 --> 00:06:25.639 it's outer atmosphere. But the main disc of the Sun 136 00:06:25.759 --> 00:06:28.199 was just too bright, so he just blocked it exactly. 137 00:06:28.480 --> 00:06:31.560 He put a physical mask, just a tiny, precisely machine 138 00:06:31.560 --> 00:06:35.399 opaque disc inside his telescope directly in the focal plane 139 00:06:35.439 --> 00:06:36.680 where the image of the sun forms. 140 00:06:36.759 --> 00:06:40.839 Okay, so he basically created a tiny artificial solar eclipse 141 00:06:41.279 --> 00:06:42.680 inside the instrument itself. 142 00:06:42.720 --> 00:06:44.360 You nailed it. That's exactly what it is. 143 00:06:44.639 --> 00:06:47.639 Well, then problem solved. We just need to put a 144 00:06:47.639 --> 00:06:51.639 coronagraph in a space telescope, block the distant star and 145 00:06:51.759 --> 00:06:54.439 photograph the planet's orbiting it. Easy. 146 00:06:54.680 --> 00:06:57.639 If only it were that easy. The physics of light 147 00:06:57.879 --> 00:06:59.279 simply do not allow it. 148 00:06:59.240 --> 00:07:02.680 To work perfect because it's light, and light is complicated. Right. 149 00:07:03.040 --> 00:07:06.399 If light behaved entirely like a stream of tiny, independent 150 00:07:06.439 --> 00:07:10.319 bullets moving in perfectly straight lines, a physical block would 151 00:07:10.319 --> 00:07:11.279 work flawlessly. 152 00:07:11.639 --> 00:07:13.879 The bullets hit the mask and stop, and. 153 00:07:13.959 --> 00:07:17.319 The bullets that miss the mask pass through to the camera. Perfect. 154 00:07:17.720 --> 00:07:20.199 But light doesn't just act like a particle. 155 00:07:20.319 --> 00:07:20.920 It's a wave. 156 00:07:21.079 --> 00:07:24.839 It propagates as a wave, and waves do incredibly frustrating 157 00:07:24.920 --> 00:07:26.759 things when they encounter an obstacle. 158 00:07:26.839 --> 00:07:29.600 Oh right, they bend. I remember learning this in high 159 00:07:29.600 --> 00:07:33.199 school physics. Yeah, diffraction, Right. If a wave hits a barrier, 160 00:07:33.279 --> 00:07:36.199 it doesn't just cut off in a clean straight line exactly. 161 00:07:36.319 --> 00:07:40.360 Think about ocean waves rolling into a harbor and hitting 162 00:07:40.439 --> 00:07:44.160 a concrete breakwater the water doesn't just stop perfectly behind 163 00:07:44.160 --> 00:07:47.720 the wall. The waves curl, they wrap around the edges 164 00:07:47.720 --> 00:07:51.759 of the concrete, and they create these new complex ripple 165 00:07:51.800 --> 00:07:55.839 patterns in the supposedly calm water behind the barrier. 166 00:07:56.040 --> 00:07:58.120 Right, the water still gets turbulent back there. 167 00:07:58.199 --> 00:08:01.399 That's diffraction. So when the star light enters the telescope 168 00:08:01.439 --> 00:08:04.720 and hits the edge of that physical coronagraph mask, the 169 00:08:04.839 --> 00:08:06.879 light waves bend right around the edge of it. 170 00:08:07.000 --> 00:08:09.720 Oh wow, so they bleed into the exact dark zone 171 00:08:09.720 --> 00:08:11.399 where we're desperately trying to find. 172 00:08:11.240 --> 00:08:14.879 The planet exactly, Even with a physical shield perfectly in place, 173 00:08:14.959 --> 00:08:17.360 the star's light is sneaking around the edges. 174 00:08:17.480 --> 00:08:18.920 That is so frustrating. 175 00:08:19.000 --> 00:08:21.839 And the crazy part is diffraction is only half the battle. 176 00:08:22.000 --> 00:08:23.399 There's more. Oh yeah. 177 00:08:23.759 --> 00:08:28.360 The other, arguably more devastating problem comes from the telescope itself, 178 00:08:29.199 --> 00:08:33.279 specifically the microscopic imperfections and the telescope's optics. 179 00:08:33.279 --> 00:08:35.200 Okay, wait, let me push back on that for a second. Yeah, 180 00:08:35.320 --> 00:08:38.840 when we talk about modern space telescopes, we are talking 181 00:08:38.919 --> 00:08:45.039 about the most meticulously engineered, flawlessly polished surfaces ever created 182 00:08:45.080 --> 00:08:45.759 by humanity. 183 00:08:45.799 --> 00:08:46.960 Right, they are incredible. 184 00:08:47.039 --> 00:08:50.000 Yes, I mean the mirrors on. These things take years 185 00:08:50.039 --> 00:08:53.679 to grind and polish. They're tested in vacuum chambers. Are 186 00:08:53.720 --> 00:08:55.200 you telling me they still have flaws? 187 00:08:55.879 --> 00:08:58.919 Well, the concept of flawless changes depending on the scale 188 00:08:58.960 --> 00:08:59.519 you're working at. 189 00:08:59.600 --> 00:09:00.840 Okay, fair point. 190 00:09:01.120 --> 00:09:05.159 When optical engineers polish a primary mirror for a space telescope, 191 00:09:05.279 --> 00:09:08.360 they are smoothing the surface down to the nanimeter level. 192 00:09:08.360 --> 00:09:11.759 And a nanometer is one billionth of a meter. It's 193 00:09:11.840 --> 00:09:13.480 roughly the width of a few atoms. 194 00:09:13.600 --> 00:09:16.080 So practically perfect to the human eye. 195 00:09:16.200 --> 00:09:20.679 Yes, to almost any standard of measurement, these mirrors are perfect, 196 00:09:20.879 --> 00:09:22.759 but they are not mathematically perfect. 197 00:09:22.799 --> 00:09:24.480 Because nothing is right. 198 00:09:24.759 --> 00:09:28.440 There will always be residual surfacers, like a subnanometer bump here, 199 00:09:28.559 --> 00:09:30.240 a microscopic dip there. 200 00:09:30.080 --> 00:09:32.639 A bump the size of a molecule. That actually matters. 201 00:09:32.799 --> 00:09:35.720 It matters massively. When the wavefront of light from the 202 00:09:35.759 --> 00:09:39.480 star hits that primary mirror, it should ideally reflect uniformly, 203 00:09:39.799 --> 00:09:42.039 But when it hits that sub nanimeter bump. 204 00:09:41.759 --> 00:09:42.799 It gets thrown off. 205 00:09:43.000 --> 00:09:46.120 Yeah, that tiny fraction of the light wave is delayed 206 00:09:46.279 --> 00:09:49.039 or advanced by just a fraction of a wavelength, so 207 00:09:49.120 --> 00:09:52.519 it scatters. Oh I see, and this scattered light bounces 208 00:09:52.559 --> 00:09:54.679 through the rest of the telescope, so when it reaches 209 00:09:54.720 --> 00:09:57.799 the final image, it doesn't look like a smooth wash 210 00:09:57.840 --> 00:09:58.200 of light. 211 00:09:58.399 --> 00:09:59.120 What does it look like? 212 00:09:59.519 --> 00:10:03.440 It forms this chaotic, complex pattern of bright and dark spots. 213 00:10:03.480 --> 00:10:07.039 We call them specklesckls. Yeah, and these speckles dance all 214 00:10:07.080 --> 00:10:10.639 around the coronagraph, spilling directly into the area where the 215 00:10:10.679 --> 00:10:11.840 exoplanet should be. 216 00:10:12.320 --> 00:10:14.799 This sounds like an absolute nightmare scenario. 217 00:10:15.080 --> 00:10:17.440 It's an engineer's worst nightmare. Honestly. 218 00:10:17.679 --> 00:10:21.360 I mean, you launched this incredibly expensive piece of machinery 219 00:10:21.440 --> 00:10:25.960 right a million miles into space, and an imperfection the 220 00:10:26.000 --> 00:10:29.440 size of an atom creates a lens flare that completely 221 00:10:29.480 --> 00:10:31.440 blinds you to the planet you're looking for. 222 00:10:31.519 --> 00:10:32.679 That's the reality of it. 223 00:10:32.840 --> 00:10:35.919 But surely the coronograph catches most of it, right, Yeah, Like, 224 00:10:36.120 --> 00:10:38.480 how much light is actually leaking through these speckles. 225 00:10:38.519 --> 00:10:40.879 Let's run the math, because this is where the sheer 226 00:10:40.919 --> 00:10:45.399 brutality of that ten billion to one ratio really reveals itself. 227 00:10:45.440 --> 00:10:46.399 Okay, hit me with the math. 228 00:10:46.639 --> 00:10:50.120 So this scenario comes directly from Steve and Eichenberry. 229 00:10:49.679 --> 00:10:51.879 The professor at the University of Central Florida, Right. 230 00:10:52.000 --> 00:10:56.639 Right, He's a principal investigator specializing and tackling this exact problem. 231 00:10:57.039 --> 00:10:59.639 So let's assume your engineers are absolute geniuses. 232 00:10:59.720 --> 00:11:01.080 Ok, geniuses, got it. 233 00:11:01.120 --> 00:11:04.679 The coronagraph and the mirrors are so incredibly well designed 234 00:11:04.879 --> 00:11:07.960 that they successfully block ninety nine point nine nine nine 235 00:11:08.039 --> 00:11:09.440 nine percent of the starlight. 236 00:11:09.919 --> 00:11:12.080 I mean, if an engineer tells me a system operates 237 00:11:12.120 --> 00:11:15.159 at ninety nine point nine nine nine nine nine percent efficiency, 238 00:11:15.600 --> 00:11:17.039 I'm breaking out the champagne. 239 00:11:17.159 --> 00:11:17.360 Right. 240 00:11:17.399 --> 00:11:19.519 That sounds like a total, absolute victory. 241 00:11:19.559 --> 00:11:22.120 It sounds like a triumph until you remember the baseline. 242 00:11:22.159 --> 00:11:28.080 The baseline. Ooh, the star is ten billion times brighter exactly. 243 00:11:28.399 --> 00:11:30.399 So you let one part in a million of that 244 00:11:30.480 --> 00:11:34.279 starlight leak through. Ten billion divided by one million is 245 00:11:34.320 --> 00:11:35.559 ten thousand. 246 00:11:35.200 --> 00:11:36.240 Ten thousand. Wow. 247 00:11:36.399 --> 00:11:39.960 Yeah, the leaked starlight, that microscopic one in a million 248 00:11:40.120 --> 00:11:43.879 error caused by diffraction and atomic scale bumps, is still 249 00:11:44.120 --> 00:11:46.919 ten thousand times brighter than the faint exoplanet you're trying 250 00:11:46.919 --> 00:11:47.519 to photograph. 251 00:11:47.600 --> 00:11:49.240 So you're still completely blinded. 252 00:11:49.360 --> 00:11:52.559 Completely. Your planet is buried under a mountain of scattered light. 253 00:11:52.720 --> 00:11:55.840 And the worst part is the speckles look exactly like planets. 254 00:11:55.879 --> 00:11:58.120 Oh yeah, and there are thousands of them in the 255 00:11:58.120 --> 00:12:02.360 image I can. Barry is very blunt about this reality. 256 00:12:02.720 --> 00:12:06.159 His exact words regarding this math are, quote, You're doomed. 257 00:12:06.559 --> 00:12:10.360 You're doomed. That is, I mean, we really are fighting 258 00:12:10.440 --> 00:12:13.399 a war on two impossible fronts here, we really are. 259 00:12:13.720 --> 00:12:17.399 On the macroscale, you have a nuclear fireball that is 260 00:12:17.519 --> 00:12:19.840 ten billion times brighter than your target. 261 00:12:20.080 --> 00:12:20.159 Ye. 262 00:12:20.559 --> 00:12:23.519 And then on the microscale, you have a physical shield 263 00:12:23.559 --> 00:12:27.159 that leaks because light waves refuse to travel in straight lines, 264 00:12:27.559 --> 00:12:29.879 and your mirror has a bump on it smaller than 265 00:12:29.879 --> 00:12:30.679 a strand of DNA. 266 00:12:30.759 --> 00:12:31.720 It's a lot to deal with. 267 00:12:31.960 --> 00:12:33.720 How on earth do you fix a leak you can't 268 00:12:33.759 --> 00:12:34.840 even physically see? 269 00:12:35.159 --> 00:12:38.279 Well, if you accept that you cannot build a flawless 270 00:12:38.279 --> 00:12:42.080 physical shield and you can't polish a mirror to mathematical. 271 00:12:41.519 --> 00:12:44.240 Perfection, which we clearly can't, right, then. 272 00:12:44.080 --> 00:12:46.159 You have to change your entire approach. You have to 273 00:12:46.159 --> 00:12:48.440 stop relying entirely on static hardware. 274 00:12:48.519 --> 00:12:49.799 Okay, so what do you do instead? 275 00:12:49.960 --> 00:12:53.399 You must develop a dynamic system, something that can measure 276 00:12:53.440 --> 00:12:56.000 the leaked light and correct it in real time, actively 277 00:12:56.039 --> 00:12:58.720 smoothing out the wavefront before the final picture is taken. 278 00:12:59.000 --> 00:13:03.080 Active correction Okay, I know. Ground based telescopes use something 279 00:13:03.080 --> 00:13:07.120 called adaptive optics to deal with the Earth's atmosphere exactly. Yes, 280 00:13:07.200 --> 00:13:10.279 because the atmosphere is turbulent, it makes stars twinkle, so 281 00:13:10.360 --> 00:13:13.399 they use special mirrors that change shape rapidly to cancel 282 00:13:13.440 --> 00:13:16.559 out the twinkling spot on. But space telescopes don't have 283 00:13:16.600 --> 00:13:19.559 to deal with the atmosphere. They do, however, have these 284 00:13:19.600 --> 00:13:22.600 microscopic mirror bumps and thermal shifts we just talked about. 285 00:13:23.120 --> 00:13:26.039 Don't we already put adaptive optics on space telescopes to 286 00:13:26.080 --> 00:13:27.600 correct for these internal errors? 287 00:13:27.679 --> 00:13:31.840 We do. We use deformable mirrors in modern space telescope designs. 288 00:13:31.879 --> 00:13:33.759 Okay, how does a deformable mirror work. 289 00:13:34.080 --> 00:13:38.519 Basically, it has hundreds or even thousands of tiny actuators 290 00:13:38.679 --> 00:13:41.279 pushing and pulling on the back of a very thin sheet. 291 00:13:41.080 --> 00:13:44.279 Of blasts like little pistons, exactly. 292 00:13:44.000 --> 00:13:46.120 And by changing the shape of the mirror by just 293 00:13:46.200 --> 00:13:49.399 a few nanometers, you can counteract those bumps on the 294 00:13:49.440 --> 00:13:51.960 primary mirror and flatten the wavefront of light. 295 00:13:52.120 --> 00:13:54.039 That's brilliant, so that solves it. 296 00:13:54.279 --> 00:13:57.639 Well, here is the critical flaw. A deformable mirror is 297 00:13:57.879 --> 00:14:00.639 useless if it doesn't know exactly what to take. 298 00:14:00.840 --> 00:14:03.759 Oh, it needs instructions. 299 00:14:03.200 --> 00:14:05.600 Right, It needs a sensor to tell it where the 300 00:14:05.679 --> 00:14:08.559 light is distorted. It needs to know where the leaks are. 301 00:14:09.200 --> 00:14:11.480 And this brings us to one of the most maddening 302 00:14:11.600 --> 00:14:14.039 architectural problems in modern astronomy. 303 00:14:14.360 --> 00:14:16.120 Okay, I'm braced. What is it? 304 00:14:16.720 --> 00:14:18.600 Non common path aberration. 305 00:14:18.320 --> 00:14:21.240 Con common path aberrations. Okay, we need to break that 306 00:14:21.279 --> 00:14:23.159 down because it sounds like the crux of the. 307 00:14:23.159 --> 00:14:26.279 Whole failure it really is. Let's trace the journey of 308 00:14:26.320 --> 00:14:28.639 a single photon through a space telescope. 309 00:14:28.679 --> 00:14:30.200 Okay, Following the photon. 310 00:14:30.080 --> 00:14:33.720 The light enters the aperture, hits the giant primary mirror, 311 00:14:34.000 --> 00:14:37.000 bounces to a smaller secondary mirror, and makes its way 312 00:14:37.039 --> 00:14:38.559 deep into the optical. 313 00:14:38.120 --> 00:14:40.080 Assembly, cloutsing through the machine. 314 00:14:40.200 --> 00:14:44.039 Right. Eventually the light hits a beam splitter. This is 315 00:14:44.080 --> 00:14:46.440 a piece of glass that lets some light pass through 316 00:14:46.480 --> 00:14:47.679 while reflecting the rest. 317 00:14:47.960 --> 00:14:50.080 Okay, it forks the road exactly. 318 00:14:50.799 --> 00:14:54.480 In traditional telescope architectures, a small percentage of the light 319 00:14:54.559 --> 00:14:57.679 is split off here and sent to a wayfront sensor. 320 00:14:57.600 --> 00:14:59.919 And that sensor is the thing checking for air. 321 00:15:00.320 --> 00:15:04.480 Yes, this sensor analyzes the light, detects any distortions caused 322 00:15:04.480 --> 00:15:07.039 by the primary mirror and sends a signal to the 323 00:15:07.080 --> 00:15:09.519 deformable mirror to fix the wavefront. 324 00:15:09.720 --> 00:15:14.480 Okay, so the sensor checks the light, diagnoses the microscopic bumps, 325 00:15:14.960 --> 00:15:18.399 and tells the system to correct itself. That sounds like 326 00:15:18.440 --> 00:15:19.480 it should work perfectly. 327 00:15:19.639 --> 00:15:22.360 The problem is what happens to the rest of the light. 328 00:15:22.639 --> 00:15:24.480 The light that didn't go to the sensor. 329 00:15:24.279 --> 00:15:27.679 Right the main beam of light, the science beam. It 330 00:15:27.720 --> 00:15:31.320 continues past the beam splitter, passes through the coronagraph, and 331 00:15:31.399 --> 00:15:34.519 travels through a final series of lenses and mirrors before 332 00:15:34.519 --> 00:15:36.200 it finally hits the focal plane. 333 00:15:36.240 --> 00:15:39.120 And the focal plane is where the actual camera is, yes. 334 00:15:39.000 --> 00:15:41.320 The main science camera, where the photograph is taken. That 335 00:15:41.399 --> 00:15:44.360 final stretch of the journey, the physical space between where 336 00:15:44.360 --> 00:15:47.039 the sensor split off and where the camera sits. 337 00:15:47.080 --> 00:15:49.679 That's the non common path exactly. 338 00:15:49.440 --> 00:15:51.720 Because the sensor is on one path and the camera 339 00:15:51.759 --> 00:15:54.000 taking the picture is on a completely different path. 340 00:15:54.080 --> 00:15:55.759 They aren't sharing the same road anymore. 341 00:15:55.799 --> 00:15:59.200 Precisely, and in the freezing vacuum of space, things are 342 00:15:59.240 --> 00:16:04.519 never entirely static. A telescope experiences these tiny thermal fluctuations 343 00:16:04.559 --> 00:16:07.080 as it moves in and out of the sun's shadow, So. 344 00:16:07.080 --> 00:16:09.679 The metal is expanding and contracting. 345 00:16:09.480 --> 00:16:11.639 Just by fractions of a millimeter. But yes, and you 346 00:16:11.679 --> 00:16:15.240 have spinning reaction wheels that keep the spacecraft pointed, which 347 00:16:15.320 --> 00:16:17.519 create microscopic. 348 00:16:16.879 --> 00:16:18.840 Vibrations, and all of that adds up. 349 00:16:18.960 --> 00:16:23.320 All of these environmental factors introduce new distortions, new aberrations 350 00:16:23.320 --> 00:16:26.360 into the light as it travels through that final non 351 00:16:26.399 --> 00:16:27.080 common path. 352 00:16:27.320 --> 00:16:30.000 But the sensor has already done its job. It checked 353 00:16:30.000 --> 00:16:32.600 the light way back at the beam splitter. Yes, it 354 00:16:32.679 --> 00:16:35.559 is absolutely no idea what happens to the light during 355 00:16:35.559 --> 00:16:36.519 that final stretch. 356 00:16:36.639 --> 00:16:40.000 It is entirely blind to it. Steve and Eikenberry uses 357 00:16:40.039 --> 00:16:43.559 this brilliant analogy to explain the structural blind spot. He 358 00:16:43.639 --> 00:16:45.879 calls it the dirty bedroom problem. 359 00:16:45.919 --> 00:16:48.039 The dirty bedroom. I like that. Explain that one. 360 00:16:48.159 --> 00:16:50.960 Okay, Imagine you own a beautiful house. Yeah, and your 361 00:16:51.000 --> 00:16:55.200 singular obsession is keeping the master bedroom flawlessly. 362 00:16:54.639 --> 00:16:57.799 Clean, Okay, pristine white carpet exactly. 363 00:16:58.120 --> 00:17:00.320 You do not want a single speck of dust on 364 00:17:00.360 --> 00:17:03.360 that white carpet. So you install a highly advanced security 365 00:17:03.360 --> 00:17:04.720 camera in the hallway. 366 00:17:04.440 --> 00:17:06.319 Leading up to the bedroom to monitor the path. 367 00:17:06.720 --> 00:17:09.839 Right, You monitor every person who walks down that hallway. 368 00:17:10.079 --> 00:17:12.119 You check their shoes, you make sure they wipe their feet. 369 00:17:12.559 --> 00:17:16.440 You are rigorously monitoring the approach, but you don't have 370 00:17:16.519 --> 00:17:18.839 a camera inside the bedroom itself. 371 00:17:18.960 --> 00:17:22.400 Ah, So you watch them walk down the hall perfectly clean, 372 00:17:22.799 --> 00:17:26.160 but right at the doorway they step in a microscopic 373 00:17:26.559 --> 00:17:27.680 puddle of mud. 374 00:17:27.640 --> 00:17:30.720 Or dust falls off their jacket as they cross the threshold. 375 00:17:30.799 --> 00:17:32.559 Right, And because you're only looking at the hallway, you 376 00:17:32.559 --> 00:17:35.920 have absolutely no idea they're tracking dirt all over your 377 00:17:35.920 --> 00:17:39.319 pristine white carpet until it's too late. The final destination 378 00:17:39.559 --> 00:17:40.599 is unmonitored. 379 00:17:40.799 --> 00:17:44.359 The unmonitored bedroom is the non common path. The wavefront 380 00:17:44.400 --> 00:17:47.000 sensor checked the light early on, but the light picked 381 00:17:47.039 --> 00:17:49.759 up optical dirt aberrations in the final few inches of 382 00:17:49.839 --> 00:17:52.359 glass and vacuum right before hitting the camera. 383 00:17:52.480 --> 00:17:55.599 Well, and because those aberrations go uncorrected, they just ruin 384 00:17:55.640 --> 00:17:56.559 the final image. 385 00:17:56.799 --> 00:17:59.839 The speckles bloom, the contrast is destroyed, and the ex 386 00:18:00.200 --> 00:18:01.160 planet vanishes. 387 00:18:01.440 --> 00:18:05.759 So the obvious, purely logical solution here is just put 388 00:18:05.759 --> 00:18:06.720 a camera in the bedroom. Right. 389 00:18:06.799 --> 00:18:07.880 It sounds so simple. 390 00:18:08.359 --> 00:18:10.160 If the problem is that we aren't looking at the 391 00:18:10.160 --> 00:18:13.880 final destination, why don't we just monitor the light at 392 00:18:13.880 --> 00:18:17.359 the exact spot where the final picture is taken. Why 393 00:18:17.400 --> 00:18:18.279 is that so difficult? 394 00:18:18.559 --> 00:18:22.400 It is intensely difficult because of the fundamental rules of 395 00:18:22.480 --> 00:18:25.279 quantum mechanics and something called photon starvation. 396 00:18:25.400 --> 00:18:27.839 Photon starvation meaning there's just not enough light. 397 00:18:28.200 --> 00:18:32.119 Right when you are trying to photograph an exoplanet, you're 398 00:18:32.160 --> 00:18:34.599 not dealing with a flood of light. You might be 399 00:18:34.640 --> 00:18:37.960 collecting only a few individual photons from that planet every minute. 400 00:18:38.039 --> 00:18:39.160 Oh wow, just a trickle. 401 00:18:39.440 --> 00:18:43.680