1 00:00:00,090 --> 00:00:03,810 Okay. Good afternoon, everybody. Welcome to the Physics Colloquium. 2 00:00:04,230 --> 00:00:08,250 So I'm delighted today to be able to introduce a friend of mine from, uh, 3 00:00:08,400 --> 00:00:12,690 from Garching in Munich, where the European Southern Observatory is Bob Fosbury. 4 00:00:13,020 --> 00:00:19,710 He's been there for the vast majority of his career where he spent most of the time working with Hubble Space Telescope. 5 00:00:20,700 --> 00:00:29,790 But, um, I have interest particularly on active galactic nuclei in star forming galaxies so far universe material. 6 00:00:30,450 --> 00:00:40,850 But he has recently extended his wide range of interests to in particular the interpretation of colour images. 7 00:00:40,860 --> 00:00:47,130 And so he has a, a very interesting website that you might want to look at for some interesting spectroscopy of various objects. 8 00:00:47,640 --> 00:00:51,540 And today he's going to tell us about colours from Earth. 9 00:00:52,090 --> 00:00:55,770 Oh, thank you, Roger. Yes. 10 00:00:55,770 --> 00:00:58,980 Roger and I have known one another for a very long time, but I won't go into that. 11 00:01:02,700 --> 00:01:05,640 It's very nice of you to invite me here. I'm very, very happy to be here. 12 00:01:06,270 --> 00:01:13,530 As Roger implied, I spent a long time working on what I called Edge of the Universe stuff. 13 00:01:14,550 --> 00:01:18,960 And when I retired from the European Space Agency and the Hubble job a few years ago, 14 00:01:19,560 --> 00:01:23,070 I asked myself the question Should I continue with some very interesting collaborations 15 00:01:23,070 --> 00:01:26,790 I had about the point or whether I should do something quite different? 16 00:01:27,390 --> 00:01:31,950 And I decided to do something quite different. So in a sense, this is a bit of an experiment. 17 00:01:32,910 --> 00:01:36,780 It's a subject towards a range of subjects that I'll talk about this afternoon 18 00:01:37,230 --> 00:01:42,900 in which I'm I'm very interested in a very general sense about exoplanets, 19 00:01:43,710 --> 00:01:53,040 the history of the Earth and so on, but with a very specific purpose of looking at the Earth as an example of an extra. 20 00:01:53,100 --> 00:02:01,170 So looking at the Earth in ways that are very similar to the ways in which we would try and look at excellence in the future. 21 00:02:01,620 --> 00:02:09,660 So it's the future is possibly the next generation of very large telescopes, both on the ground and in space. 22 00:02:10,170 --> 00:02:15,329 And I think very many people are busy on thinking about this kind of problem. 23 00:02:15,330 --> 00:02:24,810 But I think there are possibly ways in which we can hopefully make the ground based telescopes much more competitive with the space telescopes. 24 00:02:25,200 --> 00:02:31,230 And if I'm going to tell you anything that's perhaps of interest for any experts in this area this afternoon, 25 00:02:31,620 --> 00:02:39,449 it'll be an idea about using large ground based telescopes for looking at atmospheric properties of 26 00:02:39,450 --> 00:02:44,850 terrestrial planets and avoiding some of the problems that we normally have from the from the ground. 27 00:02:46,050 --> 00:02:54,780 So really on moving from distant universe to looking back at the Earth as an exoplanet. 28 00:02:54,780 --> 00:02:59,010 And the obvious place to start, of course, is with Carl Sagan's Pale Blue Dot, 29 00:03:01,530 --> 00:03:13,020 which I think was a masterstroke of of PR back in 1990 to point the Voyager cameras back towards the solar system and take a portrait of the planets. 30 00:03:13,410 --> 00:03:21,330 It was actually quite difficult because from a distance of 6.1 billion kilometres, the planets are pretty close to the sun. 31 00:03:21,330 --> 00:03:23,970 Certainly the terrestrial planets are very close to the sun. 32 00:03:24,300 --> 00:03:31,890 And so the famous streaks in this image are sunbeams within the voyager camera, which is very hard to get rid of. 33 00:03:33,030 --> 00:03:36,360 Nonetheless, one concealer, one can see a pale blue dot there. 34 00:03:37,050 --> 00:03:41,280 And one of the things that's struck me many, many years ago is pale blue. 35 00:03:41,280 --> 00:03:48,320 Why? Why is the earth pale blue? I find it very difficult to understand exactly why that was true. 36 00:03:48,330 --> 00:03:54,270 And I hope I can say something this afternoon about that. So that was a masterstroke of imaging. 37 00:03:54,270 --> 00:04:01,710 And then I'll just go through a series of these images of Earth from well away from the the Earth itself. 38 00:04:02,760 --> 00:04:10,170 I'm sure you're all very familiar with these, but I think this is a lovely one from behind Saturn, using Saturn as a as a solar occult. 39 00:04:10,770 --> 00:04:17,130 And to look at the pale blue dot, which if you enlarge a bit, you know, you can convince yourself you can see the moon as well. 40 00:04:18,180 --> 00:04:22,830 This was done with the Cassini wide angle camera quite a long time ago in 2006. 41 00:04:23,250 --> 00:04:26,970 And you may remember in 2013, I think it was July, 42 00:04:27,870 --> 00:04:35,520 Nassr decided to take another image with the Cassini spacecraft using the the high resolution camera, 43 00:04:35,880 --> 00:04:39,990 and they tried to get everyone in the earth to wave at the time that they took the picture. 44 00:04:42,690 --> 00:04:52,080 What they also did actually, was to as well as to image the earth with from Cassini with Saturn in the foreground. 45 00:04:52,350 --> 00:04:55,979 They also had a spacecraft, the Messenger spacecraft to Mercury. 46 00:04:55,980 --> 00:04:59,760 And they looked back from the inside of the solar system to look at the earth. 47 00:05:00,080 --> 00:05:08,270 At more or less the same time. So there are two images of the earth. So I took the Cassini data and lodged it to individual pixels. 48 00:05:08,630 --> 00:05:13,280 And indeed, you can see the clear difference in colour between the earth, 49 00:05:13,280 --> 00:05:18,620 which has a lot of blue pixels around it, and the moon, which is rather brown by comparison. 50 00:05:23,290 --> 00:05:32,010 A nice picture I found from the the Mars Global Surveyor spacecraft from about 140 million kilometres of the earth in the moon. 51 00:05:33,330 --> 00:05:42,120 And you can begin to resolve the surface structures on the earth from that distance and of course, the very famous Earthrise picture. 52 00:05:43,710 --> 00:05:51,360 So Nassr have taken the trouble to build a couple of images of the earth called the blue marble. 53 00:05:51,390 --> 00:05:54,870 I don't know if any of you have found this on the Web. I love looking at them. 54 00:05:55,680 --> 00:05:58,049 They are composites of many individual images, 55 00:05:58,050 --> 00:06:05,700 but they processed in such a way that they're supposed to represent a fairly true colour image of the of the earth. 56 00:06:06,090 --> 00:06:10,230 So you can see which things contribute, which colours. 57 00:06:14,700 --> 00:06:20,890 Well, I did. I took that image and I just separated the red and the blue channels in the in the image. 58 00:06:20,910 --> 00:06:24,120 And this is the blue channel and this is the Red Channel. 59 00:06:24,660 --> 00:06:28,170 And you can see something interesting about the ocean. 60 00:06:28,470 --> 00:06:39,870 If you look at the the Gulf of Mexico here and Florida, you can see that Florida is brighter than the the Gulf of Mexico Ocean in the Blue Channel. 61 00:06:40,950 --> 00:06:52,110 But in the harder to see. But you can see it can here it looks it looks it looks darker than the Gulf of Mexico in in in in the Blue Channel. 62 00:06:53,610 --> 00:07:00,300 And that brought me to a little bit of history. I'm I'm a great admirer of C.V. Raman as a as a physicist. 63 00:07:00,780 --> 00:07:05,550 And, in fact, I knew Roman Sun Radhakrishnan, the radio astronomer, quite well. 64 00:07:07,680 --> 00:07:15,390 And I've read a lot of the stories about Raman doing experiments of the kind that I like doing with his pocket spectroscope and so on. 65 00:07:15,930 --> 00:07:23,100 And there's a very nice paper in the Proceedings of the Royal Society, written in 1921, published in 1922. 66 00:07:24,180 --> 00:07:30,420 Roman visited. He was already feted as a as as a famous scientist in India. 67 00:07:30,720 --> 00:07:34,980 And he visited Britain for about a year, but less than a year. 68 00:07:34,980 --> 00:07:43,350 I think where he went around to various places became a fellow of the Royal Society and so on and on his journey back to India by boat. 69 00:07:43,950 --> 00:07:48,210 He became fascinated by the colour of the sea, the colour of the ocean. 70 00:07:48,570 --> 00:07:50,010 And he spent the whole time, I mean, 71 00:07:50,010 --> 00:07:58,200 I can imagine Roman running around with his nickel prism and his spectroscope peering at the period the ocean under all different conditions. 72 00:07:58,800 --> 00:08:05,280 And he wrote a very nice paper entitled On the Molecular Scattering of Light in Water and the Colour of the Sea, 73 00:08:05,580 --> 00:08:11,910 in which he actually estimated the colour of the ocean and tried to understand what was producing the colour. 74 00:08:14,550 --> 00:08:21,990 And he did this in terms his unit of brightness was actually the air equivalent 75 00:08:21,990 --> 00:08:25,830 number of kilometres you needed to really scatter that amount of light. 76 00:08:26,460 --> 00:08:37,080 And so he had a rather odd units and his spectrum of the sea looked like this using scattering data as it was known at the time. 77 00:08:38,010 --> 00:08:45,120 But I've corrected that for the radio, scattering from the sky and produced the green curve here, 78 00:08:45,360 --> 00:08:52,380 using his data which represents the colour of the ocean as well as he had estimated it at the time. 79 00:08:52,890 --> 00:09:01,340 And you can see on this blue image that the image is very bright around the edge, and that's because of the increased optical depth of the atmosphere. 80 00:09:01,350 --> 00:09:05,610 So a little really, really scattering, going on around the edge, much less going on in the middle. 81 00:09:07,200 --> 00:09:15,749 But the combination of the of the scattering and the colour of the sea gives you the rather peculiar colour of the of the earth, 82 00:09:15,750 --> 00:09:18,960 which is a very unsaturated greenish blue. 83 00:09:19,530 --> 00:09:27,210 This is due to vibrational band absorption, the vibrational overtones, absorbing the red light very effectively. 84 00:09:27,780 --> 00:09:30,680 But here you have what I call the green sky phenomenon. 85 00:09:30,690 --> 00:09:38,309 So you have a really scattering source function, lambda the minus four, but you also have right really scattering same function. 86 00:09:38,310 --> 00:09:41,430 You have the absorption as well. So you combine those two together. 87 00:09:42,120 --> 00:09:48,600 It's the same effect that gives you the the actual green twilight sky somewhere between the orange and the blue. 88 00:09:48,900 --> 00:09:56,430 You see, as you go upwards from the horizon, you go through red, orange, apple green and then eventually blue. 89 00:09:56,460 --> 00:10:02,370 The apple green is exactly this effect of the combination between the the radio 90 00:10:02,370 --> 00:10:06,720 scattering source function and the each of the minus one over land as the fourth. 91 00:10:07,070 --> 00:10:13,080 Also, each of the minus lambda is the fourth from the scattering anyway. 92 00:10:13,080 --> 00:10:17,190 So I think that gives us an understanding of the two contributors. 93 00:10:18,480 --> 00:10:22,080 So rather balanced contributors to the pale blue dot the rest. 94 00:10:23,020 --> 00:10:34,110 The pale bit coming from all the clouds. So this brings me to address the ways in which we will look at terrestrial exoplanets. 95 00:10:35,760 --> 00:10:43,440 It's a rather simple argument. I mean, we either look at reflected light coming from the parent star bouncing off the planets, 96 00:10:44,100 --> 00:10:48,509 and that's extremely difficult for very obvious reasons. 97 00:10:48,510 --> 00:10:54,990 The stars are extremely bright, 1 to 10 billion times for solar type stars and earth like planets. 98 00:10:55,740 --> 00:11:00,990 Somewhat better, but still daunting in the thermal infrared. So that's going to be a very difficult problem. 99 00:11:00,990 --> 00:11:01,410 And, you know, 100 00:11:01,410 --> 00:11:09,510 a lot of thought has been given to the design of the European extremely large telescope and its instruments to address that particular problem. 101 00:11:10,170 --> 00:11:12,120 But it is going to be extraordinarily difficult. 102 00:11:12,870 --> 00:11:20,090 The other way, of course, is transmission through the atmosphere as the planet transits transits in front of the parent star. 103 00:11:22,860 --> 00:11:26,129 We were talking just before Coffey about the transit of Venus and you have this 104 00:11:26,130 --> 00:11:30,930 image of a rather large sun and a rather tiny Venus transient transiting across it. 105 00:11:31,650 --> 00:11:38,460 So this is a problem as well because during the transit you're actually intercepting a rather small amount of light. 106 00:11:38,970 --> 00:11:44,040 And if you want to learn about the atmosphere, you have to look at their light that has actually transmitted this very, 107 00:11:44,040 --> 00:11:47,040 very thin skin of atmosphere around the earth. 108 00:11:47,040 --> 00:11:53,790 So that's that's a rather daunting problem as well. But, you know, we're getting big telescopes. 109 00:11:53,790 --> 00:11:57,449 We're making very stable, precise instruments. 110 00:11:57,450 --> 00:12:06,210 So sometime it'll happen. What do we expect to see from something like a terrestrial planet, given that it could be at any stage of its evolution? 111 00:12:07,140 --> 00:12:11,790 Well, we'll see reflection and transmission of the atmosphere, the effect of clouds and so on. 112 00:12:12,900 --> 00:12:18,360 We'll see oceans by reflection, although oceans are actually very dark, there's very little reflection from the oceans. 113 00:12:18,660 --> 00:12:30,000 There are lovely blue colour, but they don't reflect very much blue light. And the oceans of course have a profound effect on the atmosphere. 114 00:12:30,000 --> 00:12:35,460 They provide water for clouds and so on. We will see continents by reflection. 115 00:12:35,850 --> 00:12:51,060 And also, I'm sure we have the possibility of seeing the effects of volcanic activity on up on potential earth like planets life, of course, 116 00:12:51,300 --> 00:12:58,080 very lengthy discussions about biomarkers and what they are and how many of them you need to convince yourself that there's life on a planet. 117 00:12:58,080 --> 00:13:09,510 But certainly the kind of biomolecules that we can see, the 203 methane, nitrogen dioxide, etc., some of those we I think will be able to see. 118 00:13:10,800 --> 00:13:15,570 And O2 may be difficult, but I think ozone will be probably the easiest to see. 119 00:13:16,290 --> 00:13:18,240 Still very difficult, but the easiest to see. 120 00:13:19,650 --> 00:13:27,180 They'll be bioproducts, for instance, you know, the oxygen oxidisation, state of rocks and so on, but also chlorophyll. 121 00:13:27,180 --> 00:13:32,640 Chlorophyll is a very interesting biomarker because it has such a strong spectral signature in the red 122 00:13:33,060 --> 00:13:40,230 at about 700 nanometres where the reflectivity of of plant material chains changes very abruptly. 123 00:13:40,560 --> 00:13:50,370 So that is an interesting biomarker. And of course the extension of that is to look for non-equilibrium signatures in general, whatever they might be. 124 00:13:53,160 --> 00:13:59,730 Civilisation, I, I put that down, but I think, you know, 20, 30 years ago people thought, oh, well, 125 00:13:59,880 --> 00:14:08,310 the obvious way to look for life on planets throughout the galaxy is to look for civilisation or look for broadcast. 126 00:14:08,310 --> 00:14:13,709 But, you know, you look at the earth today. I mean, how much energy are we putting into broadcasting, into outer space? 127 00:14:13,710 --> 00:14:21,390 Not very much. And probably in a few years time, nothing at all. So I'm not sure that civilisation is going to be done in this kind of way. 128 00:14:25,810 --> 00:14:30,490 So I want to talk in the middle part of my talk. 129 00:14:30,490 --> 00:14:41,260 I want to talk about the looking at the earth as an exoplanet in those two ways, first of all, by reflection, and then secondly, by transmission. 130 00:14:42,370 --> 00:14:47,499 I think the more interesting part for me anyway, is the transmission, but so I'll allow a little bit more time for that. 131 00:14:47,500 --> 00:14:57,309 But of course, experiments like this have been done and the most valuable asset we have in performing these experiments is actually the moon, 132 00:14:57,310 --> 00:15:05,570 because it acts as a very good reflector nearby and we can do a lot of experiments by just bouncing light back off the moon. 133 00:15:06,250 --> 00:15:11,409 And that's been very valuable for us. You can also do it by flying spacecraft a long way away from the earth. 134 00:15:11,410 --> 00:15:15,580 And looking back at the earth, that has been done on quite a number of occasions. 135 00:15:16,420 --> 00:15:19,480 But they tend to be very short measurements. 136 00:15:20,040 --> 00:15:21,550 You can't do terribly much with them. 137 00:15:23,750 --> 00:15:32,350 So if you want to look at the integrated spectrum of the illuminated face of the earth, then Leonardo's your man. 138 00:15:32,350 --> 00:15:35,680 I mean, Leonardo da Vinci figured out how to do this by looking at earth shine. 139 00:15:37,300 --> 00:15:41,590 The light reflected off the dark part of the new moon. 140 00:15:42,430 --> 00:15:45,420 There's a beautiful example, actually, in Munich where I live. 141 00:15:45,430 --> 00:15:51,250 There was a wonderful example at the beginning of this currently nation where we saw a wonderful moon and a earthshine. 142 00:15:51,670 --> 00:15:59,950 It's also been done with other planets. So I found this Cassini image which showed Jupiter as being illuminated by Saturn by reflection. 143 00:16:00,700 --> 00:16:10,179 So this can be done. And this is a page from one of Leonardo's books which shows that he did indeed understand 144 00:16:10,180 --> 00:16:14,620 the geometry of the of the process with the sun and the and the earth and the moon. 145 00:16:16,840 --> 00:16:23,140 And this is an image, actually, I hate to admit it, but I think it's a synthetic image by Martin Kornmesser. 146 00:16:23,320 --> 00:16:30,040 But it's very realistic of the relative brightness of the, uh, the earthshine and the bright moon. 147 00:16:30,580 --> 00:16:31,059 And of course, 148 00:16:31,060 --> 00:16:40,600 the advantage of this kind of experiment is that you have the bright moon to calibrate the earth shine data so you can take the ratio between the two, 149 00:16:40,960 --> 00:16:48,280 get rid of all the solar nonsense, and just look at the effect of the Earth's atmosphere and reflectivity of the planet. 150 00:16:53,260 --> 00:16:56,290 This is an example of doing it from outer space. 151 00:16:56,320 --> 00:17:06,850 It's the epoxy spacecraft looking at the earth with three colour filters in the visible, and you can see the very various continents going past. 152 00:17:06,880 --> 00:17:11,860 You can see Saudi Arabia and you see the moon going past. 153 00:17:12,070 --> 00:17:15,040 See how dark the moon is. It's quite remarkable how dark the moon is. 154 00:17:16,030 --> 00:17:24,610 So that gives you a very direct visual impression of what the integrated earth looks like when you see it from a long way away. 155 00:17:24,850 --> 00:17:29,850 I'll repeat that movie in a little bit, replacing the red filter by a near-infrared filter. 156 00:17:29,890 --> 00:17:38,350 So you go along beyond the chlorophyll change and you can see the effect on the visibility of vegetation and forests. 157 00:17:40,270 --> 00:17:47,080 So this is a visible spectrum of Earthshine taken with the Palomar five metre, in fact, by some very good friends and colleagues of mine. 158 00:17:49,360 --> 00:17:53,820 It was done over two or 3 hours and it's an average of observations. 159 00:17:53,830 --> 00:17:59,940 The black line is an average of observations taken at that time. I mean, you can look at this and think, yeah, okay. 160 00:17:59,950 --> 00:18:04,329 It's not actually very exciting and it doesn't tell you very much. And the signatures are not tremendously strong. 161 00:18:04,330 --> 00:18:07,820 And I think that's that's the problem with looking at reflected light. 162 00:18:07,840 --> 00:18:13,480 You're looking down through the atmosphere. You're looking at light getting reflected back from the surface. 163 00:18:13,930 --> 00:18:19,150 You're also looking at a lot of light reflected from the upper parts of clouds. 164 00:18:19,660 --> 00:18:23,620 And you are and you look at the total path length through the atmosphere. 165 00:18:23,620 --> 00:18:30,400 And it's not very much I mean, it's a bit more than an earth mass on average, but an air mass on average, but not very much more. 166 00:18:30,850 --> 00:18:35,760 So all you see in the visible and a near infrared is a rise there. 167 00:18:35,770 --> 00:18:40,929 Doing due to really scattering is somewhat affected by the ozone. 168 00:18:40,930 --> 00:18:47,560 SHAPLEY Balance here you see rather weak oxygen, molecular oxygen absorptions and really very weak water. 169 00:18:47,570 --> 00:18:53,800 So, yeah, it's it's an interesting experiment to do, but it's hard to learn very much about the Earth from it. 170 00:18:56,600 --> 00:19:01,819 Well, I did do I had a room in when I was in Durham a year and a half ago. 171 00:19:01,820 --> 00:19:04,160 I had a room which overlooks the Botanical Gardens, 172 00:19:04,610 --> 00:19:13,819 and I used to get up and put my spectrometer out of the window and I took a spectrum of the botanical gardens and normalised that with the nearby sky. 173 00:19:13,820 --> 00:19:18,200 So this is a relative colour of the botanical gardens in Durham. 174 00:19:18,860 --> 00:19:24,889 Durham is well known to be fairly wet and you see this huge feature here. 175 00:19:24,890 --> 00:19:31,730 This is water absorption in the far right and this is the reflection of, of chlorophyll. 176 00:19:32,150 --> 00:19:41,220 Uh, so it has a very strong signature. And you may think it's surprising that while perhaps you do see this signature extremely weakly in the, 177 00:19:41,410 --> 00:19:44,660 the Palomar data of the Earthshine, it is extremely weak. 178 00:19:45,440 --> 00:19:49,730 And I think there's an interesting reason for that. It's because first tend to attract clouds. 179 00:19:50,060 --> 00:19:57,290 So you have a lot of nice little forests on the planet at any one time, but there's a high probability that those forests will be covered by cloud, 180 00:19:57,590 --> 00:20:02,060 and that's indeed a cloud that taken at about the same time from satellite data. 181 00:20:02,390 --> 00:20:12,140 And you can see the Amazon, uh, you can see the African equatorial belt, you can see a lot of the way that the northern regions are covered by cloud. 182 00:20:12,140 --> 00:20:15,950 So it's very little grassland or forest peeking through that. 183 00:20:15,950 --> 00:20:23,060 So that's a that's a tricky part of trying to observe chlorophyll during, uh, during daytime. 184 00:20:23,990 --> 00:20:27,080 This is going back to the, the epoxy image. 185 00:20:27,080 --> 00:20:31,280 This is the same set of data except for the exchange of filters. 186 00:20:31,790 --> 00:20:35,929 And I think you can see very clearly even Australia has a lot of vegetation at 187 00:20:35,930 --> 00:20:41,240 this particular time and you can see the northern uh the northern forests, 188 00:20:41,940 --> 00:20:50,269 you see the forests in Africa and of course the uh South America and North America shine up quite brightly. 189 00:20:50,270 --> 00:20:54,049 So the chlorophyll signal is, it is extremely strong. 190 00:20:54,050 --> 00:21:03,980 In fact, if you can see it, it's an extremely strong biomarker. So let's move on to the transmission efforts. 191 00:21:04,190 --> 00:21:08,240 And here we can really exploit the utility of of lunar eclipses. 192 00:21:08,600 --> 00:21:11,870 And we've done this. In fact, I so I'm an amateur in this. 193 00:21:11,870 --> 00:21:20,840 We only started this a couple of years ago, but we had the advantage of getting a young Chinese student on an exchange program with, 194 00:21:20,870 --> 00:21:25,670 uh, with Beijing, who spent two years with us and engaging. 195 00:21:26,120 --> 00:21:29,239 And I must say he was absolutely brilliant. 196 00:21:29,240 --> 00:21:33,620 And he did a number of lunar eclipse experiments, which I found extremely interesting. 197 00:21:33,620 --> 00:21:35,239 And I learned a tremendous amount from this. 198 00:21:35,240 --> 00:21:43,340 I mean, I including one thing we learned which may have a very profound effect on the use of large ground based telescopes. 199 00:21:43,730 --> 00:21:52,910 So let me go through this and I'll, you know, some of it will be, uh, some will be interesting in a technical sense what you have to look out for. 200 00:21:53,720 --> 00:21:59,510 But the last bit, I think, is, is more interesting in a in a philosophical and a utilitarian sense. 201 00:22:01,100 --> 00:22:05,209 So I'll just take you through this. I hope you'll learn something from this. 202 00:22:05,210 --> 00:22:08,630 I mean, there may be many experts in the audience, but I hope you'll learn something. 203 00:22:08,960 --> 00:22:17,720 I have to show one. So movie here of what transits look like gives us all a bit of a break, but, uh, actually, it does show two things. 204 00:22:17,790 --> 00:22:24,230 It shows, one, planets completely eclipsing the sun, which is what we're doing with lunar eclipses. 205 00:22:24,920 --> 00:22:30,440 And then it shows a more conventional view of a here we get the complete eclipse happening. 206 00:22:31,610 --> 00:22:36,049 And then a bit later we've got a small planet, planet going across the disk of the star, 207 00:22:36,050 --> 00:22:44,960 which is more typical of the, of the kind of transit, uh, we want to observe with spectrometers. 208 00:22:49,220 --> 00:22:55,040 It's still quite a big planet. I mean, it's much bigger than Venus in the Sun, but it gives you an idea of what they look like. 209 00:22:56,390 --> 00:22:59,590 So this is taken from the International Space Station. 210 00:22:59,790 --> 00:23:05,780 It's more or less what the moon sees in the umbra of a lunar eclipse. 211 00:23:06,470 --> 00:23:12,740 So when you have the earth cutting across the sun, it completely eclipses the sun, 212 00:23:13,190 --> 00:23:18,740 except that the Earth's atmosphere refracts a certain image into the Umbra. 213 00:23:19,100 --> 00:23:23,360 So if you're sitting in the umbra of the if you're sitting on the moon within the umbra, 214 00:23:23,750 --> 00:23:29,060 you actually see bright sunlight coming through the shell of the atmosphere at some point. 215 00:23:29,630 --> 00:23:33,680 And you can calculate how big that point is and so on, and what the effect of altitude is. 216 00:23:34,430 --> 00:23:40,910 But you also get the integral of the blue sky light from around the whole earth shining onto the umbra. 217 00:23:40,920 --> 00:23:45,140 So both of those sources affect the spectrum that you get. 218 00:23:49,490 --> 00:23:56,440 That's picture by so subtly. He is the guy who does in Baltimore who does very, very many of the Hubble images. 219 00:23:56,450 --> 00:24:01,010 He's the image maker. We also do it in Cushing, but salt is a very good friend of ours. 220 00:24:03,020 --> 00:24:12,410 This for this particular eclipse, this is the path of the moon in within the umbral within the umbral shadow. 221 00:24:12,920 --> 00:24:18,649 And what we do, actually, as we look at the Tycho crater, the Tycho is quite highly reflective. 222 00:24:18,650 --> 00:24:23,810 So it gives it just gives us more signal to noise than we would get by looking at a random part of the moon. 223 00:24:23,840 --> 00:24:31,700 So we just point the telescope at Tycho and let the eclipse proceed and then work out the geometry of the of the light path. 224 00:24:32,450 --> 00:24:34,910 And it turns out that for this particular eclipse, 225 00:24:35,210 --> 00:24:44,540 the turning point of the light from the sun to the moon crossed the Antarctic coast south of Australia. 226 00:24:45,860 --> 00:24:52,040 Interesting. I mean, what is the atmosphere like in and over the Antarctic coast? 227 00:24:52,550 --> 00:24:57,470 And Mount Travis is marked in there for a reason which will become apparent 228 00:24:57,470 --> 00:25:01,670 in a moment and it's raised an issue that we haven't completely solved yet. 229 00:25:01,670 --> 00:25:08,410 But this is a not to scale representation of the particular geometry. 230 00:25:08,420 --> 00:25:13,880 So you've got the sunlight coming in refracted in the Earth's atmosphere, which is very much exaggerated in depth here. 231 00:25:14,540 --> 00:25:22,790 Bouncing off the moon and going back to the telescope actually was not in Beijing itself, but not very far away from Beijing. 232 00:25:23,300 --> 00:25:25,400 It was actually a very nice house. 233 00:25:25,430 --> 00:25:33,020 The student from from China came to us with high resolution spectral data of this eclipse, and we were extremely impressed. 234 00:25:33,020 --> 00:25:38,480 It was far better than anything we'd seen before, and he was able to do very interesting things with it. 235 00:25:40,670 --> 00:25:49,640 You look at both the bright moon and the eclipsed moon through more or less the same vertical extinction. 236 00:25:50,060 --> 00:25:53,360 And so when you take the ratios that cancels out, in fact, 237 00:25:53,360 --> 00:25:59,900 you can often design the observation time so that the masses of the two observations are more or less identical. 238 00:26:00,500 --> 00:26:07,190 So in principle, even though Beijing is a great source of pollution, we shouldn't have any pollution showing due to Beijing. 239 00:26:07,220 --> 00:26:10,790 We should perhaps have pollution showing from Antarctica. 240 00:26:13,990 --> 00:26:23,740 This, again, is an exaggerated scale calculation of the image that is seen by the part of the moon we were looking at. 241 00:26:23,800 --> 00:26:31,750 So you get a distorted image of the sun, which varies slightly during the process of the eclipse, 242 00:26:32,200 --> 00:26:38,930 and it has an effect of altitude in the in the atmosphere around about 15. 243 00:26:38,950 --> 00:26:43,030 Actually, the mean the mean altitude is around 12 and a half kilometres. 244 00:26:43,690 --> 00:26:51,010 So it's it's almost like looking through a pinhole. So you get a clean path through the atmosphere, through a very small pinhole. 245 00:26:51,400 --> 00:26:54,970 And you can model on that particular basis. 246 00:26:56,290 --> 00:27:00,729 And this is what you see. We're very familiar with this picture now, but this is what you see. 247 00:27:00,730 --> 00:27:08,830 This is the ratio of the eclipsed the umbral eclipsed moon and the bright moon. 248 00:27:08,980 --> 00:27:18,490 So it's the ratio of the light that's passed through the Earth's atmosphere to the sunlight that's illuminating the same part of the moon. 249 00:27:19,720 --> 00:27:23,380 And, well, just a I mean, a crude description. 250 00:27:23,800 --> 00:27:30,220 You see the decreasing transmission as you go into the blue. 251 00:27:30,910 --> 00:27:35,920 And a lot of this is due to both really scattering, scattering and aerosol scattering. 252 00:27:37,000 --> 00:27:41,800 Interestingly, you see this really very broad feature here with these lumps and bumps. 253 00:27:42,100 --> 00:27:45,070 And this is due to ozone, the sharply bands of ozone, 254 00:27:45,490 --> 00:27:52,090 which are actually the strongest two lowering feature that we will see in the visible for an earth like planet, 255 00:27:52,330 --> 00:27:56,900 much stronger than the oxygen and even the water. The water is somewhat problematic. 256 00:27:56,920 --> 00:28:02,650 And it's interesting here you see some noise here where you should see water, but you don't see any water. 257 00:28:04,390 --> 00:28:08,790 This is a surprise because we had looked at other lunar eclipses and seen a fair amount of water, 258 00:28:08,810 --> 00:28:12,010 not as much as we see from the ground, but quite a lot. 259 00:28:12,010 --> 00:28:13,810 But in this case, we see very little water. 260 00:28:14,200 --> 00:28:21,610 So I showed this to the Alma guys at ESO and they said, yes, we know, don't tell us there's no water in Antarctica. 261 00:28:23,270 --> 00:28:26,620 So the water is presumably frozen out at that altitude. 262 00:28:26,620 --> 00:28:30,280 And we don't we really don't see any water at all or very little of it. 263 00:28:32,200 --> 00:28:43,200 The other features, I mean, the oxygen, molecular oxygen of yes, we also see these used to be called dinosaurs, the oxygen to oxygen to combination. 264 00:28:43,300 --> 00:28:47,620 So I think the modern terminology, they're called collision induced absorptions. 265 00:28:48,160 --> 00:28:54,230 They turn out to be quite important. They turn out to be quite strong under some circumstances and they're very useful because they depend. 266 00:28:54,550 --> 00:28:57,820 The strength of those bands depends quite drastically on the pressure. 267 00:28:58,180 --> 00:29:03,520 And so you can use them as indicators of whether you're actually seeing through the low atmosphere or not. 268 00:29:04,300 --> 00:29:11,740 Because many times when you look at a lunar eclipse, the low atmosphere will be obscured by aerosols and clouds and so on. 269 00:29:12,220 --> 00:29:16,330 And if you don't see any water, you may think, well, the water is just obscured by cloud. 270 00:29:16,330 --> 00:29:23,980 We don't see it at all. But if you see that, if you see these dimmer features, then you know, you're seeing a low into the atmosphere. 271 00:29:26,440 --> 00:29:28,510 So that was an interesting thing we learnt on the way. 272 00:29:31,720 --> 00:29:37,060 I did mention that you got a little bit of skylights as well as the pinhole image of the sun and in 273 00:29:37,060 --> 00:29:43,860 fact here the model without the skylight is this little faint line underneath here in the blue. 274 00:29:43,870 --> 00:29:50,500 So this, the skylight from around the earth contributes a small amount in the blue, but not very much. 275 00:29:55,680 --> 00:30:03,510 And these are the contributors. I mean, this is the this is the menu of molecules that you have to use to produce this kind of spectrum. 276 00:30:05,250 --> 00:30:11,670 Many of them are obvious, like the O2, the aerosol scattering, the molecular scattering, the Rayleigh scattering. 277 00:30:14,490 --> 00:30:17,500 The interesting things are, well, the very weak H2O. 278 00:30:19,410 --> 00:30:23,730 These are the dimer features you can see, which are reasonably strong, 279 00:30:24,600 --> 00:30:29,550 hard to see for an exo earth, but you know, they're interesting to know that they're there. 280 00:30:29,880 --> 00:30:33,990 The thing that surprised us was to have very strong nitrogen dioxide. 281 00:30:34,380 --> 00:30:38,790 We didn't expect this because nitrogen dioxide is normally considered to be a pollution product. 282 00:30:39,210 --> 00:30:42,600 You see lots of it over Los Angeles and I'm sure over Beijing as well. 283 00:30:43,350 --> 00:30:50,429 But we didn't really expect to see it over Antarctica. So I actually got in touch with Clive Oppenheimer, 284 00:30:50,430 --> 00:31:00,840 who's volcanologist at Cambridge and discussed this with him because he's actually the guy who runs the non service volcanic observatory. 285 00:31:01,260 --> 00:31:07,440 In fact, he spends a month of the year December camped at 3000 metres on the slopes of Mount Everest. 286 00:31:08,610 --> 00:31:13,230 I get the most amazing emails from him. At least I have the most amazing email origins. 287 00:31:13,650 --> 00:31:15,930 I got one from me the other day from Pyongyang. 288 00:31:16,110 --> 00:31:22,830 He was invited by the North Vietnamese to to look at a volcano on the Chinese border because they were very worried about it. 289 00:31:24,210 --> 00:31:26,100 Anyway, it turns out. 290 00:31:26,730 --> 00:31:35,280 Of course, I didn't know this, but it turns out that Mount Travis is the strongest point source of nitrogen dioxide in Antarctica. 291 00:31:35,850 --> 00:31:42,300 And it's not a volcanic gas. It actually gets fixed above the hot lava like there's a permanent lava lake in Mount Everest. 292 00:31:42,720 --> 00:31:50,670 And that the temperature, the high temperature above the lava lake fixes the nitrogen and produces a significant amount of nitrogen dioxide. 293 00:31:50,970 --> 00:31:55,500 So we thought for a while that nitrogen dioxide might have been coming from Mount Everest. 294 00:31:56,370 --> 00:31:58,950 We're a little sceptical about that now. 295 00:31:59,130 --> 00:32:08,610 It's, I think, more likely that the nitrogen dioxide is a is a decay product of nitrous oxide, which is produced by bacteria. 296 00:32:09,090 --> 00:32:11,220 And in the spring snow melt, 297 00:32:11,760 --> 00:32:19,890 the bacteria produce quite a lot of nitrous oxide and dump it into the atmosphere where it can get oxidised into nitrogen dioxide. 298 00:32:20,070 --> 00:32:28,260 But that's not really been resolved yet. I thought this plot end because it just shows you what you can do by looking at the moon with a spectrometer. 299 00:32:29,640 --> 00:32:40,560 This is the computed spectrum of nitrogen dioxide absorption, given the the basic cross-section data, and this is the residual of our observations. 300 00:32:40,560 --> 00:32:46,500 So I was surprised when I saw how how well we could do with such a complex, weak spectrum. 301 00:32:50,000 --> 00:32:56,480 Coming back to water for a moment. This is the top spectrum is a lunar is an earlier lunar eclipse observation. 302 00:32:56,780 --> 00:33:00,020 In fact, I think it was the first one that was published. It was published by. 303 00:33:00,260 --> 00:33:08,300 All right, Pele from the Canaries in Nature in 90 something 2009. 304 00:33:08,360 --> 00:33:15,200 Beg your pardon? And we look we looked at this, and you can see here, the water's quite weak, but there is some water. 305 00:33:15,470 --> 00:33:19,430 This is an eclipse above the Arctic rather than the Antarctic. 306 00:33:20,330 --> 00:33:26,780 But it actually happened when there was a volcanic eruption in in in the Arctic region. 307 00:33:27,080 --> 00:33:31,550 And so there was probably quite a lot of aerosol extinction. 308 00:33:32,420 --> 00:33:36,230 And this is what the spectrum of the sunset looks like. 309 00:33:37,490 --> 00:33:40,880 So your ground level, this is actually a sea level from Norfolk. 310 00:33:41,240 --> 00:33:43,760 And you see the water bands are extremely strong. 311 00:33:44,450 --> 00:33:50,360 You can also see these these oh four features, the diamond features are also quite strong in these data. 312 00:33:50,750 --> 00:33:54,409 This, incidentally, is taken with our one of our little portable spectroscope. 313 00:33:54,410 --> 00:34:01,729 So we have a number of portable electronic spectrometers, which we use for atmospheric and plant studies and so on, 314 00:34:01,730 --> 00:34:05,390 for generating reference spectra for a lot of these things. 315 00:34:06,020 --> 00:34:22,670 Very nice devices. I just wanted to put a little rest into the into the talk by talking a bit more about ozone because I realise, 316 00:34:23,270 --> 00:34:31,610 I realise that very few people realise what a profound effect ozone has on what we see around us, especially during Twilight. 317 00:34:32,960 --> 00:34:40,610 I came across this by accident quite a number of years ago when I was trying to understand the the colour of the zenith sky at twilight. 318 00:34:41,150 --> 00:34:50,270 When, when you, when you have a clear sky and the sun is setting and you look directly upwards, you see a remarkably deep, steely blue colour. 319 00:34:50,810 --> 00:34:54,170 And I try I measured this with a little three colour photographer, 320 00:34:54,170 --> 00:34:58,760 a light metre with three filters and tried to model it using Rayleigh scattering 321 00:34:58,760 --> 00:35:03,710 and I could get nowhere near modelling the colour blue from the zenith sky. 322 00:35:05,120 --> 00:35:09,769 And I eventually looking in Allen's astrophysical quantities, 323 00:35:09,770 --> 00:35:16,100 the little used to be in my day a green book where they have the coefficients of really scattering and aerosol scattering and so on. 324 00:35:16,310 --> 00:35:21,500 They also have a little table of the SHAPLEY band absorption coefficients in this spectral range. 325 00:35:21,500 --> 00:35:26,630 So I put that into my little model. And lo and behold, it's ozone that makes a blue colour. 326 00:35:26,930 --> 00:35:30,080 And I'll show you in a moment. If you take the ozone away, it's totally different. 327 00:35:31,340 --> 00:35:38,270 So this is a this is an observation, again, taken with one of our little portable spectrometers of about an hour before. 328 00:35:38,850 --> 00:35:43,670 Uh, yes. Well, it's a it's an hour and a quarter before sunset. 329 00:35:44,600 --> 00:35:54,259 And these are spectra normalised to the spectrum, taken an hour before sunset. 330 00:35:54,260 --> 00:36:00,230 So we just take the subsequent spectra and divide them by the, uh, the first spectrum. 331 00:36:00,470 --> 00:36:08,810 And you can see the development of this structure in the orange as the twilight proceeds and the, 332 00:36:09,200 --> 00:36:14,510 the blue lines are the ones before sunset and the red lines are the ones after sunset. 333 00:36:14,930 --> 00:36:18,530 And you can see this is an enormously strong absorption feature. 334 00:36:19,130 --> 00:36:24,950 This is a zero real zero level here. Um, and this has a very, very profound effect on the colour. 335 00:36:25,250 --> 00:36:33,440 So when you within about half an hour of, of sunset, the blue colour of the sky has almost nothing to do with right here with Rayleigh scattering. 336 00:36:33,440 --> 00:36:36,470 It's all due to ozone and it makes a tremendous effect. 337 00:36:36,860 --> 00:36:39,950 And you don't notice this so much because your eyes adapt to the blue. 338 00:36:40,400 --> 00:36:43,840 And it's only when you take a picture with a digital camera set. 339 00:36:44,060 --> 00:36:49,520 So sunlight, white balance that you realise you must have done something wrong because the pictures are so incredibly blue, 340 00:36:49,880 --> 00:36:55,070 but the colour temperature will show you and the colour temperature gets extremely high in twilight. 341 00:36:56,000 --> 00:37:01,520 So this not only tells you that you're dealing with something which has a strong spectroscopic signature, 342 00:37:01,790 --> 00:37:07,220 which is of interest for exoplanets, but it also tells you something about the earth we live on. 343 00:37:07,610 --> 00:37:12,230 And, uh, not many people realise that this is such an important effect. 344 00:37:12,620 --> 00:37:18,290 What I've done, I, I took what I think is the first observation of the SHAPLEY band absorption, 345 00:37:19,040 --> 00:37:24,740 which is reported in a little note in Astrophysical Journal in 1934 by Wolff, Moore and Melvin. 346 00:37:25,040 --> 00:37:29,060 And it's actually Absorbance they're measuring. So I turned it upside down. 347 00:37:29,510 --> 00:37:34,460 For the sake of argument. And you can see that this is done with a pocket spectroscope in a box camera. 348 00:37:34,920 --> 00:37:39,650 Uh, but it's a pretty good, pretty good. 349 00:37:42,460 --> 00:37:46,120 And this is what I was saying about the the colour temperature, the brightness. 350 00:37:46,120 --> 00:37:49,390 Of course, as you approach sunset goes down dramatically. 351 00:37:49,960 --> 00:37:53,830 This is the colour temperature measured with a a colour temperature metre. 352 00:37:54,010 --> 00:37:57,820 So it's just measuring in the visible spectrum. It doesn't go in far into the red. 353 00:37:58,810 --> 00:38:02,410 But you say it goes up to almost 15,000 kelvins. 354 00:38:03,400 --> 00:38:07,240 And this has profound effects on on animals, as I'll come to in a moment. 355 00:38:07,290 --> 00:38:13,469 It's it's quite remarkable. The eye level. 356 00:38:13,470 --> 00:38:18,420 This side, the blue hour this is known as the blue are by artists and poets. 357 00:38:19,020 --> 00:38:28,750 I don't know if any of you are that way inclined, but this is a phrase which is used quite a lot in artistic and poetic literature. 358 00:38:28,770 --> 00:38:35,310 The blue are they realise that this is a time when the the landscape around them was much slower than they might have expected. 359 00:38:38,140 --> 00:38:41,500 This is a neat little set of data we had. 360 00:38:41,830 --> 00:38:45,909 We were we were observing the moon from the CFT for another reason. 361 00:38:45,910 --> 00:38:54,490 But this gives us a picture of the spectrum as a function of the height of the tangent path through the Earth's atmosphere. 362 00:38:54,820 --> 00:39:00,430 So by looking at different times, at the reflected light from the moon, we could change the height. 363 00:39:00,430 --> 00:39:06,879 So this goes from about five kilometres altitude, up to about 100 kilometres, and you can see a low altitudes. 364 00:39:06,880 --> 00:39:08,770 You lose more or less everything in the visible. 365 00:39:09,670 --> 00:39:19,870 But there's a there's a region later on, just as you're coming out of Eclipse where you can again, you get this very blue spectrum due to the ozone. 366 00:39:20,620 --> 00:39:27,519 And I found there's a German lady photographer who is not actually a scientist, 367 00:39:27,520 --> 00:39:32,049 but she's very keen on photographing things of scientific interest and she's done a whole 368 00:39:32,050 --> 00:39:38,380 series of photographs of the Eclipse moon and this is one of them actually she gave to me. 369 00:39:39,150 --> 00:39:42,160 The intensity range here has been very much compressed, of course, 370 00:39:42,160 --> 00:39:48,040 but this shows you that as you get close to the edge of the umbral eclipse, you get a blue moon. 371 00:39:48,160 --> 00:39:56,560 And this is a blue moon for a reason, which is quite different from the normal accepted blue moon ni scattering explanation. 372 00:39:57,940 --> 00:40:02,560 You can't see this by eye if you look carefully at the right time, but it's I think it's quite remarkable. 373 00:40:04,150 --> 00:40:10,270 Another thing, which is I'm sure the observers amongst you have seen is the the Earth's Shadow or the dark segment. 374 00:40:10,600 --> 00:40:14,560 This is a picture taken of Paranal as the sun is setting. 375 00:40:15,370 --> 00:40:19,599 And you get this very dark band here, which has an interesting colour. 376 00:40:19,600 --> 00:40:22,780 It's, it's, it's a very steely grey blue. 377 00:40:23,110 --> 00:40:29,589 And in fact, this is, I think, the result of a double scattering. This is a set of data I got from a guy called Ray Lee. 378 00:40:29,590 --> 00:40:38,080 He's a very well known atmospheric physicist, works at the US Naval Academy and he had actually observed the spectrum of the dark segment, 379 00:40:38,330 --> 00:40:42,220 the to the orange and the green curves here in the visible. 380 00:40:42,550 --> 00:40:47,920 I think he had a trouble with background subtraction over here but and then this orange line 381 00:40:47,920 --> 00:40:54,550 is my model of the solar spectrum with two scatter rings and a and a deep ozone layer. 382 00:40:55,420 --> 00:40:58,659 So it's the earth's shadow is a shadow of the ozone layer. 383 00:40:58,660 --> 00:41:02,080 It's not actually a shadow of the solid earth. It's a shadow of the ozone layer. 384 00:41:06,400 --> 00:41:11,820 So this is a this is a sunset sky model. 385 00:41:11,830 --> 00:41:21,190 I took the spectral fist of the data and modelled the colour of the blue sky somewhere up in the air above you. 386 00:41:21,190 --> 00:41:26,049 Not necessarily the zenith, but to give you this I don't think the colour calibration is very good. 387 00:41:26,050 --> 00:41:30,910 But this is the the colour of the sunset sky, including ozone. 388 00:41:31,270 --> 00:41:34,990 And if you take the ozone away it becomes a sort of grey, strong yellow. 389 00:41:35,680 --> 00:41:39,310 So it has a very, very profound effect on the sky. 390 00:41:41,820 --> 00:41:50,820 So you learned something. Okay. And this picture taken by my son actually in Scotland, which I grabbed immediately. 391 00:41:50,970 --> 00:41:59,340 He's very careful, careful about his colour balance. This is a picture taken at dawn or just before slightly before dawn in Dumfries and Galloway. 392 00:42:00,390 --> 00:42:05,770 You can see the mountains on the snow is the snow on the mountains is fairly nice and white. 393 00:42:06,000 --> 00:42:11,730 But the rest of the picture is incredibly blue. And this is the ozone blue in the way it actually looks like. 394 00:42:12,120 --> 00:42:18,930 But if you were standing there looking at it, you would really want to see that because your eye would flat field it out or colour field it out. 395 00:42:19,680 --> 00:42:31,430 Now, this is the kind of subject that I now find very interesting, but this has some remarkable effects on the adaptation of animals behaviour. 396 00:42:32,600 --> 00:42:35,880 And there are many examples of this too. 397 00:42:35,910 --> 00:42:39,900 I'll point out to you one I heard about only fairly recently. 398 00:42:40,350 --> 00:42:54,899 Another one I worked on myself some some years ago. The people have looked at the eyes of reindeer in the in the Arctic regions and they 399 00:42:54,900 --> 00:43:00,870 find that during the winter the eyes adapt to see blue light very effectively. 400 00:43:01,500 --> 00:43:11,250 And they do this because the the top into the lining of the eye behind the retina is reflective so that they 401 00:43:11,250 --> 00:43:17,400 can maximise the the light coming in because there's a double pass processed through the through the retina. 402 00:43:17,940 --> 00:43:21,980 And in the summer, the temperature reflects orange light. 403 00:43:22,710 --> 00:43:28,290 And if you shine a torch into a reindeer eye in the summer, you see an orange reflection. 404 00:43:28,980 --> 00:43:33,540 But if you do the same thing in the winter, you see this very dark blue ultraviolet reflection. 405 00:43:33,930 --> 00:43:39,420 And so they adapt their eye to allow them to live in this incredibly blue environment. 406 00:43:39,990 --> 00:43:44,370 And that adaptation allows them to see food much more easily. 407 00:43:44,370 --> 00:43:47,729 They can see the lichen and they can see things on the ground. 408 00:43:47,730 --> 00:43:50,970 They can see that the predators more easily. 409 00:43:51,330 --> 00:44:00,389 And there's a thesis from the University of Tromso, which I found which shows two pictures of the top part of the orange, 410 00:44:00,390 --> 00:44:05,790 a turquoise one, a summer eye, and the the blue one have a winter eye. 411 00:44:06,750 --> 00:44:16,280 There's actually a lovely YouTube video and there's a group at UCL who have studied this, and they made a video about a year ago about reindeer eyes. 412 00:44:16,320 --> 00:44:22,650 If you look up reindeer eyes or something on YouTube, you'll find it very easily and it's a very nice description of what they did, 413 00:44:23,070 --> 00:44:26,190 but I don't think they realised it was something they were dealing with. 414 00:44:26,340 --> 00:44:29,610 I really must write to these guys. Mention Ozone anyway. 415 00:44:29,610 --> 00:44:35,100 So that's the reason. The other thing I did get involved in was this little guy. 416 00:44:35,610 --> 00:44:38,760 He's a madagascan lemur called The Eye. 417 00:44:38,790 --> 00:44:46,500 I'm sure you've heard of the eye. He's the lemur that has the extended finger that it's used for picking out bugs from the leaf litter and so on. 418 00:44:46,920 --> 00:44:56,650 They forage in the rain in the rainforest at twilight where they're shadowed, generally shadowed from direct sunlight and moonlight. 419 00:44:56,670 --> 00:45:06,510 So they're just looking at skylight and they have retained their their blue, obscene blue visual pigment to be very blue sensitive. 420 00:45:06,960 --> 00:45:12,050 And I by chance, via what I post on Flickr, 421 00:45:12,060 --> 00:45:18,810 I got in touch with a a primatologist in the US who was actually working on this problem and he was puzzled. 422 00:45:18,990 --> 00:45:24,090 He didn't know what the evolutionary driver was for this item and maintain this capability. 423 00:45:24,720 --> 00:45:32,740 And we were thinking that it was that the eye feeds on the not on the seats of the travellers tree itself. 424 00:45:32,740 --> 00:45:39,030 There's this big tree that grows in Madagascar called the traveller's tree, which is the same family as the banana. 425 00:45:39,900 --> 00:45:46,020 But it's a huge thing with big fan shaped leaves and it has big flower clusters about this big, 426 00:45:46,020 --> 00:45:51,420 hard things and it produces seed pods which when they open up into this little 427 00:45:52,080 --> 00:45:59,459 tripod of but it contains this artificially blue arrow fluff around the seed. 428 00:45:59,460 --> 00:46:06,210 So black seeds and I have this, this incredibly blue arrow, which I bought one of these seeds went somewhere, 429 00:46:06,210 --> 00:46:10,290 but I was convinced that this was dyed artificially to make the seeds more attractive. 430 00:46:10,590 --> 00:46:16,020 And it was only years later that I realised this was a real blue arrow. 431 00:46:16,080 --> 00:46:23,219 It's an extremely unusual blue in in animals or plants, and the protein has not yet been identified. 432 00:46:23,220 --> 00:46:30,150 As far as I know, people in University of Florida were looking at it. But the interesting thing was it was a by by-products of this investigation, 433 00:46:31,590 --> 00:46:40,590 this tree was known to produce a biological molecule, the molecule that you. 434 00:46:41,360 --> 00:46:46,660 I get with jaundice. I've forgotten the name, but it's something somebody remember. 435 00:46:47,900 --> 00:46:50,900 It's the. No, no. It doesn't matter. 436 00:46:51,620 --> 00:46:56,090 But this was apparently the first biological model molecule. 437 00:46:56,240 --> 00:47:00,980 It's the first animal protein that had been found directly in a in a plant. 438 00:47:02,000 --> 00:47:05,780 And there was a great deal of excitement about it. But anyway, I think that turned out to be a red herring. 439 00:47:06,020 --> 00:47:11,090 I think the the real reason for the adaptation, again, is the ozone colour, 440 00:47:11,120 --> 00:47:17,990 which gives you a very blue visual environment where they still can maintain some colour vision, even at very low light levels. 441 00:47:21,290 --> 00:47:33,860 Now. What have I got? I want to do another discussion of the excellent transit measurements. 442 00:47:34,250 --> 00:47:44,450 And this is the I mean, the normal way of doing transit measurements is you look at the you look at the photometry of the star being transited. 443 00:47:45,200 --> 00:47:50,780 And by looking at the depth of the eclipse signal as a function of wavelength, 444 00:47:51,020 --> 00:47:59,180 you can work out the, the, the absorption of the, uh, of the molecules in the planetary atmosphere. 445 00:47:59,450 --> 00:48:03,590 So it's a floating metrically based experiment. You have to measure it very precisely. 446 00:48:04,010 --> 00:48:06,440 This is why Hubble has been so successful in doing this. 447 00:48:06,590 --> 00:48:11,390 Not in earth like planets, but in somewhat bigger parts, because it can do such stable photometry. 448 00:48:11,720 --> 00:48:19,880 But it's extremely, extremely difficult to do this on the ground. And so we've been investigating two effects that you get during a transit, 449 00:48:21,530 --> 00:48:27,770 which are a result of the fact that the transiting planet is masking regions of the stellar surface. 450 00:48:28,370 --> 00:48:31,220 So the star is not just a uniformly amazing surface. 451 00:48:31,460 --> 00:48:41,120 It has variations and it has variations which are symmetric, like, uh, um, centre limb variations in brightness and line depth and so on. 452 00:48:41,450 --> 00:48:45,860 But it also has variations which are asymmetric, and that is the rotation of the, of the star. 453 00:48:46,370 --> 00:48:50,030 And the rotation of the star is called the Russell Mclauchlan Effect. 454 00:48:50,270 --> 00:48:55,430 And it's been used for very many years now in spectroscopic binaries and, and, 455 00:48:55,700 --> 00:49:03,020 and even in exoplanets in recent years to figure out whether the planets are rotating in the same direction as they as the star. 456 00:49:03,050 --> 00:49:13,190 So it's a well developed technique, but in fact it can be adapted to look at the effective radius of the planets because the part 457 00:49:13,190 --> 00:49:20,120 of the star that you actually mask or eclipse by the planet varies with the effect of radius, 458 00:49:20,630 --> 00:49:25,580 and that variation carries an intrinsic signal about the atmospheric properties. 459 00:49:26,420 --> 00:49:34,970 Now this is actually suggested back in 2004 by Chanel and in Leiden, but nobody had ever actually done it. 460 00:49:35,720 --> 00:49:40,490 Uh, so we decided to do it with a lunar eclipse and see if we could do it. 461 00:49:40,520 --> 00:49:44,299 It's a very easy case because you eclipse the entire sun. 462 00:49:44,300 --> 00:49:56,560 So it's a it's a strong signal in that sense. But to show you the way it works is that we have here the centres limb variation of the sun. 463 00:49:56,570 --> 00:50:06,979 This is the Kurt Sutter Atlas data. We just divide the the solar limb spectrum by the central spectrum and of course the 464 00:50:06,980 --> 00:50:11,110 lines don't completely go away because the lines are different deaths that the oh I mean, 465 00:50:11,120 --> 00:50:14,000 all the solar people know this very well. The lines are different deaths. 466 00:50:14,390 --> 00:50:20,240 So in the subtraction, you say this goes up to 20, 25%, especially in the blue. 467 00:50:21,020 --> 00:50:26,420 You get the effect of the central and variations of the line profiles and line deaths. 468 00:50:28,010 --> 00:50:36,770 Similarly, where, where you have a penumbral spectrum of the, uh, of the sun or penumbral spectrum of the, 469 00:50:37,130 --> 00:50:41,900 of the Earth's shadow, you're cutting off part of the sun by the, by the Earth. 470 00:50:42,320 --> 00:50:47,990 And so you're when you do your division, you're something different parts of the stellar surface. 471 00:50:47,990 --> 00:50:57,260 So this is a a set of data during a lunar eclipse where we've done the penumbra divided by the bright moon, we see a very similar pattern. 472 00:50:59,150 --> 00:51:05,960 Now, this is very important to understand in detail, because if you're looking for faint absorption lines due to the atmosphere of the planet, 473 00:51:06,320 --> 00:51:15,110 you have to understand what this central and variation effect can actually do to destroy your signal. 474 00:51:15,560 --> 00:51:18,920 And it's quite can be quite profound under some circumstances. 475 00:51:19,220 --> 00:51:23,630 That's just a simple diagram of a penumbral eclipse at different stages. 476 00:51:24,470 --> 00:51:30,620 And this is the peak solar atlas looking at a spectral line, one of the calcium triplet lines in the infrared. 477 00:51:30,860 --> 00:51:38,450 You can see the profiles are quite different. When you divide one by the other, you get this characteristic emission absorption feature. 478 00:51:38,960 --> 00:51:43,370 You can degrade the resolution a bit for the kind of observation you might make. 479 00:51:43,940 --> 00:51:49,040 And this is the example of this line with the CFT data. 480 00:51:49,220 --> 00:51:51,500 So you can see the kind of feature you get. 481 00:51:51,960 --> 00:52:02,480 And when you're trying to look for a in Earth's atmosphere absorption line in the ratio spectrum, you have to understand this. 482 00:52:03,350 --> 00:52:11,930 And in fact, it's quite daunting. I think this is the sodium one of the sodium d lines looking again at real data of the day. 483 00:52:12,020 --> 00:52:23,960 The day line has a rather weak centre limb variation, but this is the absorption line that we know exists because of the laser guide star work. 484 00:52:24,300 --> 00:52:32,790 We know the optical depth of the sodium in the Earth's atmosphere, and this is close to the maximum that we see in doing these experiments. 485 00:52:33,240 --> 00:52:40,319 And you can see the relative size of that compared to the the the variation effects. 486 00:52:40,320 --> 00:52:47,910 So that's and I don't think people realise that at that level that it's such a daunting problem. 487 00:52:51,960 --> 00:53:01,770 So if we can measure planetary atmospheric transmission from purely radial velocity data of the study taken during transit using this Rossiter 488 00:53:02,190 --> 00:53:15,209 in effect it since it doesn't require very high precision photometry it's a it's a real potential for ground based ELTs for earth like planets. 489 00:53:15,210 --> 00:53:20,330 And I, I suspect that this will become a serious scientific case for the E-ELT. 490 00:53:20,520 --> 00:53:30,989 We've already discussed it with the VLT people, and so we did this experiment using a lunar eclipse, realising that it was an easy experiment to do, 491 00:53:30,990 --> 00:53:36,240 but interested nonetheless in seeing what lessons we had to learn and how strong the signal actually was. 492 00:53:37,380 --> 00:53:45,350 So the amplitude of the signal, as I said, depends on the effect of radius, which is wavelength dependent because of the atmospheric transmission. 493 00:53:45,570 --> 00:53:50,520 And it's a kind of radial velocity tomography of the surface of the stellar surface that we do in this way. 494 00:53:52,140 --> 00:53:55,890 It will be dependent on surface activity, sunspots and so on. 495 00:53:55,890 --> 00:54:02,280 We're well aware of that. But we use the Harps telescope and I'm sure you all know Harps. 496 00:54:02,280 --> 00:54:07,979 It's the the great planet finder. It's a resolving power of over 100,000 beautiful data. 497 00:54:07,980 --> 00:54:15,150 It's wonderful to look at the data, I must say, as a spectroscopy. So I just look at the data, gosh, these are the shallow orders you can see here. 498 00:54:15,150 --> 00:54:21,600 That's a bit of a gap in the middle. There are 72 spectral orders going across that wavelength range. 499 00:54:21,600 --> 00:54:27,970 That's what a spectrum of the bright moon looks like. If we just look at a bit round the action k lines. 500 00:54:28,140 --> 00:54:29,310 The Beautiful, isn't it. I love that. 501 00:54:31,080 --> 00:54:42,150 So okay what we do is we measure for each harps spectral order and I think those 69 useful orders that we can use, 502 00:54:42,360 --> 00:54:49,740 we measure the radial velocity, okay? And we measure that at various points during the eclipse. 503 00:54:50,790 --> 00:55:00,870 And this is the, the basically the raw data before eclipse, the bright moon coming along, radial velocity, the first penumbra going down. 504 00:55:01,260 --> 00:55:04,410 And then there's the umbra. I won't talk about the Umbra. 505 00:55:04,500 --> 00:55:05,820 You can ask about it if you want. 506 00:55:06,390 --> 00:55:12,270 I think this is we're seeing a number of different effects here, but then we pick up the pin number again on the other side. 507 00:55:12,600 --> 00:55:15,750 So we got a good sample of the bright moon spectrum. 508 00:55:16,140 --> 00:55:27,540 Now, this is the computed radial velocity that we would expect after because of the the relative motions of the sun, the moon and the earth. 509 00:55:27,780 --> 00:55:32,700 And this is taken from the standard JPL horizon EPHEMERIS There's an offset here, 510 00:55:32,700 --> 00:55:37,500 which I think is due to the way the template is used in the in the HARPS pipeline. 511 00:55:38,110 --> 00:55:42,240 It doesn't matter because we, we just flatten everything by dividing by this curve. 512 00:55:42,750 --> 00:55:50,610 And so we get a zero based measurement of the velocity of the, of the reflected light from the moon. 513 00:55:56,860 --> 00:56:02,920 And obviously what you're doing your your first of all obscuring part of the sun that's redshifted. 514 00:56:03,430 --> 00:56:07,880 And then the earth moves across and then you're observing part of the sun, which is blue shifted. 515 00:56:08,230 --> 00:56:11,860 So it gives you the signature of the of the stellar rotation. 516 00:56:12,520 --> 00:56:21,420 And what we do here, we actually model the stellar rotation. We have a great of points, 100th of a solar radius grid of the. 517 00:56:21,430 --> 00:56:30,610 And so we assign velocities, the differential rotation and the actual rotational rotation to those points and then model the rotation. 518 00:56:31,090 --> 00:56:36,579 And this is just an extract of that signal before where you see the pre the first 519 00:56:36,580 --> 00:56:43,030 penumbra with this turn up because of the asymmetries and the second penumbra. 520 00:56:43,450 --> 00:56:51,340 So what we actually do, we take the maximum signal we have here of the of the velocity on one side and the other and 521 00:56:51,580 --> 00:57:01,389 plot those and those are the red and the green lines and the artificially normalise to zero. 522 00:57:01,390 --> 00:57:04,030 It doesn't matter. We don't, we don't worry about the normalisation. 523 00:57:04,930 --> 00:57:09,340 But I think the first thing you spot here, there's a correlated noise between those two signals. 524 00:57:09,730 --> 00:57:14,140 So we're not getting a uniform velocity as a function of wavelength and we can't use that directly. 525 00:57:14,440 --> 00:57:20,110 And what we think is happening is that the sun suffers from what's called a convective blue shift. 526 00:57:20,110 --> 00:57:23,890 The convective cells in the sun come up hot and go down cold. 527 00:57:24,310 --> 00:57:30,850 And so they're more light emitted from the upwelling than the than the down welling material in the sun. 528 00:57:31,300 --> 00:57:38,890 And it produces a blue shift that the surfaces understand this very well, but it depends on how deep the lines are and so on. 529 00:57:39,070 --> 00:57:40,870 So each individual harps. 530 00:57:40,870 --> 00:57:48,970 Order will contain a distribution of lines formed at the distribution of deaths, which will be different for each different order. 531 00:57:49,300 --> 00:57:56,200 And so the actual velocity we measure for each order will be, will have this correlated noise associated with it. 532 00:57:57,610 --> 00:58:05,950 And so, but the point is that the blue shift noise, the convective blue shift noise is symmetric around the centre of the sun, 533 00:58:06,310 --> 00:58:09,670 whereas the signal we're looking for is anti symmetric around the sun. 534 00:58:09,880 --> 00:58:18,430 So we just subtract the two eclipse measurements to get to the line underneath and then we just have to fiddle around a bit. 535 00:58:19,360 --> 00:58:26,470 We have to press the button. The. We have to correct, 536 00:58:26,480 --> 00:58:33,980 we have to take the continuum then darkening into effect because that has a a slight weighting effect on the on the velocity measurements. 537 00:58:34,730 --> 00:58:40,040 And then we we correct for that and we get a signal. 538 00:58:41,870 --> 00:58:48,350 So this is the variation of the radial velocity measured in metres per second against wavelength. 539 00:58:49,010 --> 00:58:49,399 Okay. 540 00:58:49,400 --> 00:59:03,290 And you've got this very characteristic signature and the red line here is a model of this derived from the components that I talked about in the, 541 00:59:03,650 --> 00:59:09,680 in the, in the eclipse spectra. So it's the basically the really scattering and the ozone absorption. 542 00:59:10,040 --> 00:59:14,540 So this bump here at 600 nanometres is the effect of the SHAPLEY ozone absorption. 543 00:59:14,960 --> 00:59:19,520 And the rise up here is due to the radius scattering. 544 00:59:19,520 --> 00:59:27,709 So we have simply from measurements of radial velocity done to a precision of well within the HARPS capabilities, 545 00:59:27,710 --> 00:59:32,390 it's a precision of a few metres per second. We actually make a measurement of the. 546 00:59:35,090 --> 00:59:40,249 Of the transmission of the atmosphere. So I think given that we're going to have extremely stable, 547 00:59:40,250 --> 00:59:47,090 extremely high resolution spectrometers on the very large telescopes, which are the last to get very high signal to noise. 548 00:59:47,480 --> 00:59:52,639 We're just looking at the light from the star. We're not looking at a planet at all, so we're just looking at the light from the star. 549 00:59:52,640 --> 00:59:59,690 So actually, it's a very I think it's a very cool method of trying to get information about the atmosphere. 550 01:00:00,020 --> 01:00:03,170 It's going to be hard. It's going to be in getting down to the centimetres per second. 551 01:00:03,410 --> 01:00:08,330 It's going to mean coping with the the intrinsic noise of the stellar surface and so on. 552 01:00:08,330 --> 01:00:12,920 But I think it's it's promising and we're trying to simulate some of these effects at the moment. 553 01:00:15,640 --> 01:00:22,450 Now. I should finish now. Um, I just very quickly want to show you some quite different things that we've been doing 554 01:00:22,450 --> 01:00:26,980 in trying to understand the earth and the effects it can have on spectrum and so on. 555 01:00:27,310 --> 01:00:31,180 I will run through all the quickly, but I think it just to give you a flavour. 556 01:00:31,640 --> 01:00:35,680 Um. We're interested in volcanoes. We're interested in volcanic gases. 557 01:00:35,920 --> 01:00:38,980 We're interested in the effect that it may have on the spectra. 558 01:00:39,340 --> 01:00:46,180 So we mounted an expedition to Mount Etna with the help of the Volcanic Observatory. 559 01:00:46,180 --> 01:00:50,170 Again, through this, Clive Oppenheimer in Cambridge helped us organise this. 560 01:00:50,860 --> 01:00:54,970 And we measured with them using our own spectrometers. 561 01:00:55,660 --> 01:01:00,879 We measured Mount Etna and they were very fortunate in having a fairly significant eruption at the time. 562 01:01:00,880 --> 01:01:08,770 So we were able to look at the lava as well as the volcanic plume and night time here of the of the eruption. 563 01:01:09,490 --> 01:01:17,049 And that's infrared 1 to 2 and a half micron spectrum of the lava seen at two distances, 564 01:01:17,050 --> 01:01:21,070 one in ten kilometres, and a rather crude estimate of the temperature of the lava. 565 01:01:21,310 --> 01:01:24,900 1260 Kelvins pretty close to what I expect for a basaltic lava. 566 01:01:24,910 --> 01:01:31,690 I think, um, the deviation of the, the red in there is due to the water. 567 01:01:31,690 --> 01:01:36,610 The water's a lot of water absorption and this is right down in the ultraviolet. 568 01:01:36,640 --> 01:01:39,790 We have a nice little ultraviolet spectrometer which actually was shown in that picture. 569 01:01:40,030 --> 01:01:49,730 It's only about this size. It's the one all the volcanologists use, I think an ocean optics spectrometer which starts measuring it to 90 nanometres. 570 01:01:49,750 --> 01:01:54,610 So we go from 290 to 510 nanometres and we can get the sulphur dioxide absorption. 571 01:01:54,940 --> 01:02:00,700 So that's a the Wiggles. There are a data and a model of the sulphur dioxide. 572 01:02:00,700 --> 01:02:09,160 We think we can measure column densities and so on. And I just want to finish with a rather nice experiment, which is, again, these are based. 573 01:02:09,460 --> 01:02:18,700 If you go to the north of Paranal in Chile, the very northern part of the Atacama Desert, it's probably the driest place on the planet. 574 01:02:19,210 --> 01:02:25,720 And there are big solar plains there which contain bacteria, which are cyanobacteria, 575 01:02:25,720 --> 01:02:29,170 probably evolved very little over the last two and a half billion years. 576 01:02:29,530 --> 01:02:34,209 And so we've been looking at the properties of this bacteria and testing the idea 577 01:02:34,210 --> 01:02:38,350 of remote sensing it in some way long time before we do that on the next to Earth. 578 01:02:38,350 --> 01:02:45,520 But we might be able to do it on Mars. So this is the region, the Big Salt Lake Uni. 579 01:02:45,940 --> 01:02:51,549 I've actually been there a fantastic place, 220 kilometres across. 580 01:02:51,550 --> 01:02:56,110 It's the it's the zero point for the chips model earth. 581 01:02:56,830 --> 01:03:00,400 So flat is within a centimetre of being completely flat. 582 01:03:00,760 --> 01:03:05,770 And then we go down here somewhere and it's within, it's not too close to the ocean. 583 01:03:05,770 --> 01:03:12,940 There's a mountain range, a low mountain range here. So it never gets the the view coming in from the fog coming in from the sea. 584 01:03:13,210 --> 01:03:17,920 So it's incredibly dry. I think the rainfall is about one millimetre every four years. 585 01:03:19,160 --> 01:03:25,360 Uh, you can see and these are the nodules of salt, pure salts covered by a bit of dust. 586 01:03:26,260 --> 01:03:30,610 And within the salt you get these bands of, uh, bacteria. 587 01:03:30,610 --> 01:03:34,480 If you can pronounce the word, your better than me croak out to the opposites. 588 01:03:34,920 --> 01:03:43,690 Um, it's a very well known bacterium. It's an extremophile which will survive almost anything in Antarctica and in the hot deserts. 589 01:03:44,440 --> 01:03:52,840 And some people even think this came from Mars, but maybe, uh, a lot of interest in this particular bacterium. 590 01:03:53,140 --> 01:03:59,470 But with our little spectrometers, we were looking at whether we could see either the transmission or the fluorescence of the chlorophyll. 591 01:03:59,680 --> 01:04:03,070 And in fact, the fluorescence of the chlorophyll is quite significant. 592 01:04:03,460 --> 01:04:10,420 And this red peak here is the effect of shining blue light on the on the rock and looking at the reflected light. 593 01:04:10,840 --> 01:04:17,230 So it's potentially this is a technique that's used by Earth resource satellites to look at the health of plants on the ground, 594 01:04:17,680 --> 01:04:20,950 and they measure the fluorescence by measuring the depths of Fraunhofer lines. 595 01:04:21,220 --> 01:04:29,650 Because if you've got a fluorescent signal underlying the reflected light, you reduce the depths of the front of the lines and that can be used. 596 01:04:29,650 --> 01:04:31,480 So I don't know whether it'll work. 597 01:04:32,080 --> 01:04:39,550 Um, this is my finishing slide, which I won't talk to, but it's one of the fun things one can do with a little spectrometer. 598 01:04:39,820 --> 01:04:43,930 It's the spectrum of a thundercloud, and it has all kinds of interesting features. 599 01:04:44,350 --> 01:04:46,150 Anyway, that's enough. Thank you very much.