API tools faq
paste
Advertisement
lepuspfelix

gholl

lepuspfelix
Dec 12th, 2011
129
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 35.08 KB | None | 0 0
raw download clone embed print report
  1. [16:38:59] <Gerrit> Does anyone object if I paste 425 lines here?
  2. [16:39:46] <+red-squirrel> go ahead
  3. [16:40:28] <smuckerz> err
  4. [16:40:29] <smuckerz> yeah
  5. [16:40:29] <smuckerz> i do
  6. [16:40:34] <smuckerz> Gerrit
  7. [16:40:36] <smuckerz> pastbin
  8. [16:40:40] <smuckerz> you fimilar with it?
  9. [16:40:51] Faqtotum [~danielg4@gimpelevich.san-francisco.ca.us] has joined ##belz
  10. [16:40:51] ChanServ [ChanServ@services.] has set mode +v Faqtotum
  11. [16:45:09] r0bby_ [~wakawaka@guifications/user/r0bby] has quit IRC: Read error: Connection reset by peer
  12. [16:45:42] r0bby_ [~wakawaka@guifications/user/r0bby] has joined ##belz
  13. [16:46:13] <Gerrit> Vanaf video:
  14. [16:46:13] <Gerrit> 1:28 start
  15. [16:46:13] <Gerrit> Thanks for the introduction, my name is Gerrit Holl and I will present my
  16. [16:46:13] <Gerrit> licentiate thesis in the next 30 minutes, welcome all those in the room
  17. [16:46:13] <Gerrit> welcome to tohse connected via internet, while I talk about my thesis, you
  18. [16:46:15] <Gerrit> have printed copies over there or in the e-mail announcement from before.
  19. [16:46:18] <Gerrit> 1:47 frame 2
  20. [16:46:20] <Gerrit> What is it I will talk about? I will start by motivating my thesis
  21. [16:46:23] <Gerrit> project.
  22. [16:46:25] <Gerrit> I will talk about cilmate change, which is the ultimate context of my
  23. [16:46:28] <Gerrit> thesis work.
  24. [16:46:30] <Gerrit> And, the motivation and why we care.
  25. [16:46:33] <Gerrit> And then I'm going to introduce the papers and while I introduce the
  26. [16:46:35] <Gerrit> papers I will also discuss the involved theory behind each paper and
  27. [16:46:38] <Gerrit> (resolve?) the context.
  28. [16:46:40] <Gerrit> So first I will talk about paper 1, paper 2, and finally paper 3, which is
  29. [16:46:43] <Gerrit> it.
  30. [16:46:45] <Gerrit> (...) from the first two papers
  31. [16:46:48] <Gerrit> But always
  32. [16:46:50] <Gerrit> 2:35 frame 3
  33. [16:46:53] <Gerrit> 2:38 frame 4
  34. [16:46:55] <Gerrit> But always the first question, and maybe the most important question
  35. [16:46:58] <Gerrit> why are we doing this, why on Earth, why do we care, for anything
  36. [16:47:00] <Gerrit> 2:48 frame 5
  37. [16:47:03] <Gerrit> For me, the prime motivation of this thesis work is...
  38. [16:47:05] <Gerrit> this is planet Earth, this is a photograph taken by the Apollo astronauts,
  39. [16:47:08] <Gerrit> the last Apollo mission on its way to the moon, but this is not a picture
  40. [16:47:10] <Gerrit> of the moon, but a picture of the Earth. Antarctica is at the top, because
  41. [16:47:13] <Gerrit> that happens to be how the spacercaft was oriented. We are used to see
  42. [16:47:15] <Gerrit> Antarctica at the bottom, but that is pretty random.
  43. [16:47:18] <Gerrit> We all live on Earth, we inhabit this planet, we have only one planet, we
  44. [16:47:20] <Gerrit> need to cherish it. Are we cherishing it?
  45. [16:47:23] <Gerrit> 3:27 frame 6
  46. [16:47:25] <Gerrit> The climate is changing. There is a lot of information in this figure, but
  47. [16:47:28] <Gerrit> teh fact that the climate is changing is hiddin in a small number at teh
  48. [16:47:30] <Gerrit> bottom that is net absorption, 0.9 W m-2, due to humans changing the
  49. [16:47:33] <Gerrit> composition of teh Earth atmosphere.
  50. [16:47:35] <Gerrit> One element of the climate that is difficult to understand, that has a lot
  51. [16:47:38] <Gerrit> of forcings and feedbacks in the climate system, are the clouds.
  52. [16:47:40] <Gerrit> Here, you see, clouds they have an important, complicated feedback,
  53. [16:47:43] <Gerrit> because they reflect incoming solar radiation, they absorb outgoing
  54. [16:47:45] <Gerrit> longwave radiation, they emit radiation, both upward and downward, and
  55. [16:47:48] <Gerrit> sometimes they have a cooling effect, sometimes they have a cooling
  56. [16:47:50] <Gerrit> effect.
  57. [16:47:53] <Gerrit> The net effect from the heating and cooling is a function of the latitude,
  58. [16:47:55] <Gerrit> a function of the altitude, they are important for hydrology, and clouds
  59. [16:47:58] <Gerrit> are actually pretty poorly understood.
  60. [16:48:00] <Gerrit> 4:45 frame 7
  61. [16:48:03] <Gerrit> To get a good understanding of clouds, to get a good quantification of
  62. [16:48:05] <Gerrit> clouds, we need satellite observations.
  63. [16:48:08] <Gerrit> And my thesis is about satellite observations of clouds, in particulary,
  64. [16:48:10] <Gerrit> about ice clouds, and one quantity that I'm focussing on a lot, is the Ice
  65. [16:48:13] <Gerrit> Water Path, which is the column mass density of ice in a cloud, so you
  66. [16:48:15] <Gerrit> might say it's, how much ice is there in a could, might sound like a
  67. [16:48:18] <Gerrit> pretty fundamental thing to know, but even a question like this, as this
  68. [16:48:20] <Gerrit> figure shows, even such a question the estimates vary by one order of
  69. [16:48:23] <Gerrit> magnitude, if you take the extremes, the estimates that are very high,
  70. [16:48:25] <Gerrit> the estimates that are very ligh, the differences are so huge, that we
  71. [16:48:28] <Gerrit> basically don't have a clue how much ice there is in a cloud.
  72. [16:48:30] <Gerrit> Which is a bit of a problem, because we just established that clouds are
  73. [16:48:33] <Gerrit> so important.
  74. [16:48:35] <Gerrit> Well, there are some fundamental reasons that the measurements vary so
  75. [16:48:38] <Gerrit> widely, because some are based on infrared, some are based on microwave,
  76. [16:48:40] <Gerrit> and the physics of the microwave obseravtions or an infrared observation
  77. [16:48:43] <Gerrit> are quite different.
  78. [16:48:45] <Gerrit> For example, microwave radiation, wavelength is large in relation to the
  79. [16:48:48] <Gerrit> ice particle, and therefore, and then there is not so much attenuation,
  80. [16:48:50] <Gerrit> so, ... cloud with small particles is completetly transparent to microwave
  81. [16:48:53] <Gerrit> obseraviotns.
  82. [16:48:55] <Gerrit> On the other hand, infrared observations, they are seeing even the thin
  83. [16:48:58] <Gerrit> clouds, and when a cloud is thick, it's completely opaque, it cannot see
  84. [16:49:00] <Gerrit> through the cloud, so when looking at the cloud, will also see part of the
  85. [16:49:03] <Gerrit> cloud
  86. [16:49:05] <Gerrit> But maybe if we have a combination of both, maybe we can do a better job.
  87. [16:49:08] <Gerrit> And that is what my thesis is about: can we do a better job?
  88. [16:49:10] <Gerrit> 7:03 frame 8
  89. [16:49:13] <Gerrit> So now I will talk about my first paper, whic his considered as a first
  90. [16:49:15] <Gerrit> step toward improving this.
  91. [16:49:18] <Gerrit> As I mentioned, the different components have different strengths, and
  92. [16:49:20] <Gerrit> it's not just me saying that.
  93. [16:49:23] <Gerrit> 7:17 frame 9
  94. [16:49:25] <Gerrit> In April, last year in April, I was at the international workshop of
  95. [16:49:28] <Gerrit> Space-based observations of frozen precipitations, IWSSM, where one
  96. [16:49:30] <Gerrit> recommendation from one of the working groups was that: mathups of
  97. [16:49:33] <Gerrit> spaceborne passive microwave observations with Cloudsat are critical for
  98. [16:49:33] mobile [~Gossamer@h33.55.188.173.dynamic.ip.windstream.net] has quit IRC: Ping timeout: 248 seconds
  99. [16:49:35] <Gerrit> improving the global snowfall detection capabilities of passive
  100. [16:49:38] <Gerrit> instruments.
  101. [16:49:40] r0bby_ [~wakawaka@guifications/user/r0bby] has quit IRC: Remote host closed the connection
  102. [16:49:40] <Gerrit> Say what you need to do is to combine CloudSat, which is an active,
  103. [16:49:43] <Gerrit> scientific radar, with passive, passive microwave, and other passive
  104. [16:49:45] <Gerrit> instruments, so that you can improve the retrievals from the passive ones.
  105. [16:49:48] <Gerrit> They said it in April 2011, already more than, almost a year before, I had
  106. [16:49:50] <Gerrit> a pubilaciotn in Atompsheri Measurement Techniques, on Collocating
  107. [16:49:53] <Gerrit> satellite based radar and radiometer measurements — methodoloy and usage
  108. [16:49:55] <Gerrit> examples.
  109. [16:49:58] <Gerrit> It's my first paper, which is pretty much, closely exactly what is
  110. [16:49:59] r0bby_ [~wakawaka@guifications/user/r0bby] has joined ##belz
  111. [16:50:00] <Gerrit> critically needed according to the workshop, so it's nice to get some independent
  112. [16:50:03] <Gerrit> confirmation that my work is relevant, maybe ...
  113. [16:50:05] <Gerrit> Matchups, I call them collocations, ..., I will come back to this figure
  114. [16:50:08] <Gerrit> later, but a collocation is just different insturments looking at the same
  115. [16:50:10] <Gerrit> place at the same time, might eb at completely different sattelites,
  116. [16:50:13] <Gerrit> different characteristics, but to combine them, synergy, can reach goals
  117. [16:50:15] <Gerrit> that you cannot reach with either satellite alone.
  118. [16:50:18] <Gerrit> 9:24 frame 10
  119. [16:50:20] <Gerrit> This satellite I'm collocating with, this CloudSat, that's a bit of a
  120. [16:50:23] <Gerrit> special satellite.
  121. [16:50:25] <Gerrit> CloudSat is part of a constellation of satellites known as the A-Train.
  122. [16:50:28] <Gerrit> But I'm focussing on CloudSat, I'm not using the other satellites so much
  123. [16:50:30] <Gerrit> at the current moment, but
  124. [16:50:33] <Gerrit> the A-Train, CloudSat carries a Cloud Profiling Radar, it's an active
  125. [16:50:35] <Gerrit> instrument, it looks only straight down, it sees only a very little bit
  126. [16:50:38] <Gerrit> it's dying, we have some five years of data, but unlike weather
  127. [16:50:40] <Gerrit> satellites, it's not going to get replaced and replaced all the time
  128. [16:50:43] <Gerrit> and because it's an active instrument, because it's designed to profile
  129. [16:50:45] <Gerrit> clouds, not for other things, we have good reasons to believe that the CloudSat ice water path
  130. [16:50:48] <Gerrit> product is the best ice water path product we have.
  131. [16:50:51] <Gerrit> CloudSat is an active instrumente, on the other hand, also close to the
  132. [16:50:53] <Gerrit> A-Train, are weather satellites, that contain passive instruments, but
  133. [16:50:56] <Gerrit> that are operational, that means if this satellite dies, they will
  134. [16:50:58] <Gerrit> replace it, because they are essential for the everyday weather forecast.
  135. [16:51:01] <Gerrit> Those satellites carry instruments, such as passive microwave and
  136. [16:51:03] <Gerrit> infrared, and those are the ones are collocating with the cloudsat
  137. [16:51:06] <Gerrit> But what are all those satellites actually measure?
  138. [16:51:08] <Gerrit> 10:58 frame 11
  139. [16:51:11] <Gerrit> measure, they don't measure ice, they are way above the clouds, below the
  140. [16:51:13] <Gerrit> clouds we also don't measure ice from the ground, so you measure by remote
  141. [16:51:16] <Gerrit> sensing, so the actual quantity that is being measured is electromagnetic
  142. [16:51:18] <Gerrit> radiation, which is emitted by any object at any temperature, for example
  143. [16:51:21] <Gerrit> by the Sun in the visible because it's so hot it's in the visible, the
  144. [16:51:23] <Gerrit> Earth which emits mostly in the infrared, here I'm also showing the
  145. [16:51:26] <Gerrit> microwave.
  146. [16:51:28] <Gerrit> Radiometers, such as the High-resolution Infarred Radiatien Sounder, or
  147. [16:51:31] <Gerrit> the Advanced Very High Resolution Radiomete,r they measuer radiation
  148. [16:51:33] <Gerrit> intensity, in teh infrared such as AVHRR, or in the microwave such as MHS, or
  149. [16:51:36] <Gerrit> actively such as the cloud profiling radar, the CPR.
  150. [16:51:38] <Gerrit> (NOTE: backgrounds not clear on projector)
  151. [16:51:41] <Gerrit> They measure actively such as the cloud profiling radar.
  152. [16:51:43] <Gerrit> And as I said, the passive instruments are carried on operational
  153. [16:51:46] <Gerrit> satellites, the cloud profiling radar is carried on cloudsat, and what I
  154. [16:51:46] <smuckerz> Gerrit
  155. [16:51:46] <smuckerz> Gerrit
  156. [16:51:47] <smuckerz> Gerrit
  157. [16:51:47] <smuckerz> Gerrit
  158. [16:51:48] <Gerrit> do is matching the two, the CloudSat is active, so I need to
  159. [16:51:51] <Gerrit> ONVERSTAANBAAR
  160. [16:51:53] <Gerrit> Microwave we don't see any curve here, because energy-wise the microwave
  161. [16:51:56] <Gerrit> is way way less, orders of magnitude less, than the infrared. But that
  162. [16:51:58] <Gerrit> doesn't matter as long as we can measure it, beacuse it still contains
  163. [16:52:01] <Gerrit> really
  164. [16:52:03] <smuckerz> Gerrit
  165. [16:52:03] <smuckerz> Gerrit
  166. [16:52:03] <Gerrit> valuable information in the context of atmospheric remote sensing, even if
  167. [16:52:04] <smuckerz> Gerrit
  168. [16:52:06] <Gerrit> it's so little radiation in absolute ways
  169. [16:52:07] <smuckerz> hey Gerrit
  170. [16:52:08] <Gerrit> 13:02 frame 12
  171. [16:52:09] <smuckerz> seriously
  172. [16:52:11] <Gerrit> back to those collocations.
  173. [16:52:13] <Gerrit> In this paper I present some statistics from (...) collocations, and
  174. [16:52:14] <smuckerz> why the hell
  175. [16:52:16] <Gerrit> collocation statistics. This was before NOAA-19 was launched. And then I
  176. [16:52:17] <smuckerz> don't you use pastebin?
  177. [16:52:18] <Gerrit> found only NOAA-18 with global collocatoins, which can be explained by the
  178. [16:52:21] <Gerrit> fact that NOAA-18 is close to the A-Train. The collocations, they occur
  179. [16:52:23] <Gerrit> globally, but not all combinations of all angles, due to the, the exact
  180. [16:52:26] <Gerrit> orbit of NOAA-18 and of the cloudsat, collocations occur in ... patterns
  181. [16:52:28] <Gerrit> this shows the latitude and the angle at which collocations occur, so it
  182. [16:52:31] <Gerrit> means that above the equator there are no collocations that occur at
  183. [16:52:33] <Gerrit> exactly the same time exactly then. This is how the orbits are, so there
  184. [16:52:36] <Gerrit> are some boundary conditions in what kind of collocations can we look at
  185. [16:52:38] <Gerrit> 13:58 frame 13
  186. [16:52:41] <Gerrit> The more important issue here is that the satellite have quite different
  187. [16:52:42] <+Faqtotum> smuckerz: he probably pasted that all at once
  188. [16:52:43] <Gerrit> footprints.
  189. [16:52:46] <Gerrit> I will zoom in a bit on part of this figure.
  190. [16:52:48] <Gerrit> Here zie je de green ellipse, which is the footprint from the MHS,
  191. [16:52:51] <Gerrit> microwave humidity sounder; HIRS which is the high-resolution infrared
  192. [16:52:53] <Gerrit> radiation sounder, as showed in an earlier plot, has lots of nice
  193. [16:52:56] <Gerrit> channels, but the footprints is such that it's difficult to use, because
  194. [16:52:58] <Gerrit> here, it doesn't even cololcate with the cloudsat at all.
  195. [16:53:01] <Gerrit> and then those bricks, that's the AVHRR, which has less nice channels than
  196. [16:53:03] <Gerrit> the HIRS, it's approximately the same frequency range, but it's a bit
  197. [16:53:06] <Gerrit> poorer resolution, maybe poorer calibartion as well, but it's my best bet
  198. [16:53:08] <Gerrit> because at least it collocates
  199. [16:53:11] <Gerrit> Now for the MHS collocations with the CloudSat, there are currently two
  200. [16:53:13] <Gerrit> options, either collocate each one pixel with the MHs pixel, each one
  201. [16:53:16] <Gerrit> footprint with the entire MHS footprint, or average them, even if I
  202. [16:53:18] <Gerrit> average the cloudsat, still over a much smaller area than the MHS,
  203. [16:53:21] <Gerrit> something always need to take into account when doing statistics with it.
  204. [16:53:23] <Gerrit> It's a bit easier with AVHRR, because there is only, can just take the
  205. [16:53:26] <Gerrit> closest and if it's too far, it's too far
  206. [16:53:28] <Gerrit> and with the HIRS, sometimes there are cocllocations, but most of the
  207. [16:53:31] <Gerrit> tmie, there actually are not, because those different HIRS-pixel are
  208. [16:53:32] <smuckerz> holy hell
  209. [16:53:33] <Gerrit> pretty far away.
  210. [16:53:36] <Gerrit> The CloudSat looks only at nadir and that's why we don't fill the entire
  211. [16:53:36] <smuckerz> are you fucking serious
  212. [16:53:38] <Gerrit> pixel, also with AVHRR, we can fill the entire MHS pixel, which is
  213. [16:53:39] <smuckerz> Gerrit
  214. [16:53:41] <smuckerz> Gerrit
  215. [16:53:41] <Gerrit> (OVERSTAANBAAR)
  216. [16:53:43] <smuckerz> what
  217. [16:53:43] <Gerrit> 15:58 frame 14
  218. [16:53:44] <smuckerz> the
  219. [16:53:45] <smuckerz> fuck
  220. [16:53:46] <Gerrit> Still, I used collocations between CPR and MHS, and I will present some
  221. [16:53:48] <Gerrit> results from my first paper, from applications from those collocations
  222. [16:53:51] <Gerrit> One application is to collocate CloudSat MHS and just compare the ice
  223. [16:53:53] <Gerrit> water path. In the plot in the introduction, the one from Salomon Eliasson
  224. [16:53:56] <+Faqtotum> smuckerz: frankly, i find this much easier to read than a pastebin
  225. [16:53:56] <Gerrit> and collaborators, I showed a comparison based on the latitudinal average, but
  226. [16:53:58] <Gerrit> here I really compare point by point and of course, yes, they cover
  227. [16:53:59] r0bby_ [~wakawaka@guifications/user/r0bby] has quit IRC: Remote host closed the connection
  228. [16:54:01] <Gerrit> different areas, but still, here we can confirm that the MHS ice water
  229. [16:54:01] <smuckerz> hey red-squirrel
  230. [16:54:02] <smuckerz> hey red-squirrel
  231. [16:54:02] <smuckerz> hey red-squirrel
  232. [16:54:03] <smuckerz> hey red-squirrel
  233. [16:54:03] <Gerrit> path, based on
  234. [16:54:04] <smuckerz> hey red-squirrel
  235. [16:54:04] <smuckerz> hey red-squirrel
  236. [16:54:05] <smuckerz> hey red-squirrel
  237. [16:54:05] <smuckerz> hey red-squirrel
  238. [16:54:06] <Gerrit> a product called MSPPS, has a much much lower ice water path than CloudSat, since we
  239. [16:54:06] <smuckerz> hey red-squirrel
  240. [16:54:07] <smuckerz> hey red-squirrel
  241. [16:54:09] <Gerrit> believe the CloudSat to be correct we arrive that the MSPPS too low, even
  242. [16:54:11] <Gerrit> for the thicker clouds here that it is supposed to be able to see.
  243. [16:54:14] <Gerrit> Another application I present in the first paper is just to do the statistics,
  244. [16:54:16] <Gerrit> to get the brightness tempearture which, from the MHS, which is a unit of
  245. [16:54:18] <smuckerz> hey red-squirrel
  246. [16:54:19] <Gerrit> radiation intensity and to compare this directly against the ice water
  247. [16:54:19] <smuckerz> hey red-squirrel
  248. [16:54:19] <smuckerz> hey red-squirrel
  249. [16:54:20] <smuckerz> hey red-squirrel
  250. [16:54:20] <smuckerz> hey red-squirrel
  251. [16:54:21] <smuckerz> hey red-squirrel
  252. [16:54:21] <smuckerz> hey red-squirrel
  253. [16:54:21] <Gerrit> path from cloudsat. This is a nice relationship, we can see here for
  254. [16:54:24] <Gerrit> example that the thin clouds below 100 grams per square meter are
  255. [16:54:27] <Gerrit> transparent for microwave radiation, but for thick clouds there is a nice,
  256. [16:54:28] r0bby_ [~wakawaka@guifications/user/r0bby] has joined ##belz
  257. [16:54:29] <Gerrit> also for those clouds, which is actually already above 100 grams per
  258. [16:54:31] <Gerrit> square meter, which is pretty close to the origin on the left figure,
  259. [16:54:34] <Gerrit> on the right hand we have ..., already on 100 grams per square meter it
  260. [16:54:36] <Gerrit> should be possible to measure, when I compare against the simulated
  261. [16:54:39] <Gerrit> datasetso I can confirm that the simulated dataset is mostly correct, there are some small
  262. [16:54:42] <Gerrit> differences, I discuss them in my first paper, but the main conclusion
  263. [16:54:44] <Gerrit> here is, collocations and simulations, they agree pretty well.
  264. [16:54:47] <Gerrit> But the main, what I find the nicest, the most promising application in my
  265. [16:54:49] <Gerrit> first paper, is to actually develop a new product. How do we do that? From
  266. [16:54:52] <Gerrit> the collocations we get a big table of MHS brightness temperatures,
  267. [16:54:54] <Gerrit> radiances, and CloudSat ice water path. And by getting this table, by
  268. [16:54:57] <Gerrit> getting this database, we can train a, in this case I call it an
  269. [16:54:59] <Gerrit> artificial neural network, which is a regression technique that learns to
  270. [16:55:01] <smuckerz> hey red-squirrel
  271. [16:55:01] <smuckerz> hey red-squirrel
  272. [16:55:02] <Gerrit> see patterns, so I show it a lot of combinations of brightness temperatures
  273. [16:55:02] <smuckerz> hey red-squirrel
  274. [16:55:03] <smuckerz> hey red-squirrel
  275. [16:55:03] <smuckerz> hey red-squirrel
  276. [16:55:04] <smuckerz> hey red-squirrel
  277. [16:55:04] <smuckerz> hey red-squirrel
  278. [16:55:04] <Gerrit> and ice water path, then after this I can input new brightness
  279. [16:55:07] <Gerrit> temperatures where I don't hvae collocated ice water path, and
  280. [16:55:09] <Gerrit> (onverstaanbaar)
  281. [16:55:12] <Gerrit> 19:12 frame 15
  282. [16:55:14] <Gerrit> Now this is also still in, still in progress, but this should be pretty promising,
  283. [16:55:17] <Gerrit> and I'm going to continue with that at a later academic point
  284. [16:55:20] <Gerrit> So, in my first paper I had collocations between passive, operational and
  285. [16:55:21] red-squirrel [~rs@unaffiliated/red-squirrel] is now known as rs-afk
  286. [16:55:22] <Gerrit> active, scientific satellites.
  287. [16:55:25] <Gerrit> It's nice.
  288. [16:55:27] <Gerrit> In my second paper, I continue with collocations, but I shift the focus.
  289. [16:55:30] <Gerrit> I shift the focus away from clouds, actually.
  290. [16:55:32] <Gerrit> And look directly at the brightness temperatures of operational
  291. [16:55:35] <Gerrit> satellites.
  292. [16:55:37] <Gerrit> 19:44 frame 16
  293. [16:55:40] <Gerrit> So, second paper.
  294. [16:55:42] <Gerrit> Actually, I'm co-author, it's a paper that was written by Viju John, but I
  295. [16:55:45] <Gerrit> did the collocation part, which is a very important part of the paper, an
  296. [16:55:46] <+Faqtotum> smuckerz: haha
  297. [16:55:47] <Gerrit> essential part.
  298. [16:55:50] <Gerrit> Here it's called, they're called simulateous nadir overpasses, so two,
  299. [16:55:52] <Gerrit> here, two polar orbiting satellites, such as those polar orbiting
  300. [16:55:55] <Gerrit> operational satellites are, there are about six of them, they pass over
  301. [16:55:57] <Gerrit> the pole at each orbit, so pretty frequently they have collocations at the
  302. [16:56:00] <Gerrit> orbit, at the poles sorry.
  303. [16:56:02] <Gerrit> Over the equator, they at a certain point over the equator they pass only once
  304. [16:56:05] <Gerrit> in a couple of weeks, but every orbit they pass ONVERSTAANBAAR
  305. [16:56:07] <Gerrit> what's frequently done is that such collocations at the pole are used to
  306. [16:56:10] <Gerrit> intercalibrate the sensors
  307. [16:56:12] <Gerrit> now, due to orbital drift we actually sometimes have
  308. [16:56:15] <Gerrit> 20:47 frame 17
  309. [16:56:17] <Gerrit> we actually have sometimes have collocations ONVERSTAANBAAR
  310. [16:56:20] <Gerrit> As the figure here on the left shows, that the local time ascending node
  311. [16:56:22] <Gerrit> of the different satellites drifts.
  312. [16:56:25] <Gerrit> The local time ascending node is the time that a sunsynchronous satellite
  313. [16:56:27] <Gerrit> crosses the equator.
  314. [16:56:30] <Gerrit> Those satellites normally are supposed to have a constant time, local time
  315. [16:56:32] <Gerrit> they cross the equator, but due to other gravitational influences this is
  316. [16:56:35] <Gerrit> not really constant, and there is some drift.
  317. [16:56:37] <Gerrit> Thanks to this drift, there are moments that different satellites have the
  318. [16:56:40] <Gerrit> same local time ascending node.
  319. [16:56:42] <Gerrit> (NOTE: looks different?)
  320. [16:56:45] <Gerrit> They have the same local time ascending node, and that means in this
  321. [16:56:47] <Gerrit> period, we actually have global collocations. That means we can compare
  322. [16:56:50] <Gerrit> bias at the equator with the bias at the high latitudes, and do a
  323. [16:56:52] <Gerrit> prediction whether it is right.
  324. [16:56:55] <Gerrit> So there are there pairs of satellites, ... one month, ... two months,
  325. [16:56:57] <Gerrit> where such collocations occur.
  326. [16:57:00] <Gerrit> Most occur between NOAA-19 and NOAA-18, and that is because their orbits
  327. [16:57:02] <Gerrit> were close to each other for a long time.
  328. [16:57:05] <Gerrit> NOAA-19 was just launched, NOAA-18 was just changing the drift, so we can
  329. [16:57:07] <Gerrit> look at some statistics.
  330. [16:57:10] <Gerrit> 22:47 frame 18
  331. [16:57:12] <Gerrit> I show parts of the plot paper by Viju John and me and collobarators,
  332. [16:57:15] <Gerrit> where we compare for different latitudes, in this case look at the bias in
  333. [16:57:17] <Gerrit> MHS, no in AMSU-B, which is also a microwave radiometer, in channel 5 with
  334. [16:57:20] <Gerrit> NOAA-15 and NOAA-16. It is pretty clear here that there is a strong
  335. [16:57:22] <Gerrit> dependence on latitude.
  336. [16:57:24] <smuckerz> hey red-squirrel
  337. [16:57:25] <smuckerz> hey red-squirrel
  338. [16:57:25] <Gerrit> At -90 degrees the bias is much less than at +90 degrees.
  339. [16:57:25] <smuckerz> hey red-squirrel
  340. [16:57:27] <Gerrit> Already, here, if you use only the polar observations, you could already
  341. [16:57:30] <Gerrit> tell that it's not the same.
  342. [16:57:32] <Gerrit> Howeve,r in the next example, MetOp-A/NOAA-17 channel 5, if you use only
  343. [16:57:35] <Gerrit> the polar collocations, you reach the conculsions that there is maybe between 1
  344. [16:57:37] <smuckerz> Faqtotum
  345. [16:57:37] <Gerrit> and 1.5 kelvin in bias
  346. [16:57:40] <Gerrit> If you then correct one of them, down, you would correct it globally,
  347. [16:57:42] <Gerrit> perhaps at the equator, whereas there is actually no bias at all between the two
  348. [16:57:44] <smuckerz> Faqtotum don't you find this shit rediculas
  349. [16:57:45] <Gerrit> because this bias, it depends on latitude, and why does it depend on
  350. [16:57:47] <Gerrit> latitude, because it depends on the scene temperature, and the scene
  351. [16:57:49] <smuckerz> Faqtotum don't you find this shit rediculas and very rude
  352. [16:57:50] <Gerrit> humidity.
  353. [16:57:52] <Gerrit> therefore, we can conclude that it is simply not correct to use only polar
  354. [16:57:55] <Gerrit> SNOs for intercalibration, it is simply too, too simplistic, that is the
  355. [16:57:57] <Gerrit> conclusion from this paper
  356. [16:58:00] <Gerrit> this has also been compared to other metheds
  357. [16:58:02] <Gerrit> 24:36 frame 19
  358. [16:58:03] <smuckerz> Faqtotum, i can't believe he is fucking pasting all this shit in here
  359. [16:58:04] <+Faqtotum> smuckerz: i find it very funny
  360. [16:58:05] <Gerrit> instead, I will proceed, I'm running out of time, but I will proceed
  361. [16:58:07] <Gerrit> my third paper where I shift the focus from collocations to radiative
  362. [16:58:10] <Gerrit> transfer
  363. [16:58:12] <Gerrit> 23:53 frame 20
  364. [16:58:15] <Gerrit> radiative transfer, the modeling, is calculating all those meauruemnts, in
  365. [16:58:16] <smuckerz> Faqtotum, i told his dumbass to use fucking pastebin
  366. [16:58:17] <Gerrit> our case the passive measurements, so we can calculate, we know what the
  367. [16:58:20] <Gerrit> atmosphere is, and then we model other radiation ...
  368. [16:58:22] <Gerrit> this is governed by the radiative transfer equation and I will go through
  369. [16:58:25] <Gerrit> step-by-step, starting with the clear-sky parts, which is absorption, and
  370. [16:58:27] <Gerrit> actually we can consider extinction, since it is just scattering out, we
  371. [16:58:30] <Gerrit> just need to know how much scattering there is, I will focus on the
  372. [16:58:32] <Gerrit> absorption and the emission of gases, so the arrow reperesnts the
  373. [16:58:35] <Gerrit> atmospheric radiation
  374. [16:58:37] <Gerrit> on the left hand of the equation where you have a change in radiation
  375. [16:58:40] <Gerrit> intenisty along the line of sight
  376. [16:58:42] <+Faqtotum> smuckerz: the drama is rather entertaining
  377. [16:58:42] <Gerrit> and then there is extinction, which is absorption plus scattering
  378. [16:58:45] <Gerrit> and there is emission, eh, and as I'll show extinction causes radiation
  379. [16:58:47] <Gerrit> along the line of sight to get less, but emission causes it to get more
  380. [16:58:50] <Gerrit> so later I will focus on the 3rd term, which is related to scattering, but
  381. [16:58:52] <Gerrit> can be considered more complicated
  382. [16:58:55] <Gerrit> but with those ingredients we can
  383. [16:58:57] <Gerrit> 26:08 frame 21
  384. [16:59:00] <Gerrit> do clear-sky simulations. what is shown on the figure in the top, then we can
  385. [16:59:02] <Gerrit> actually calculate it
  386. [16:59:05] <Gerrit> I'll show the parts, part of this figure, I will zoom in on the part I draw in
  387. [16:59:07] <Gerrit> red here
  388. [16:59:10] <Gerrit> show a zoom
  389. [16:59:12] <Gerrit> so I show between, the next plot shows just the different results from
  390. [16:59:15] <Gerrit> radiative transfer between 10 and 13 micrometer
  391. [16:59:17] <Gerrit> 26:40 frame 22
  392. [16:59:20] <Gerrit> ehm, for each atmospheric gas, ro for all atmospheric gases together, we
  393. [16:59:22] <Gerrit> can calculate the opacity or the optical depth, which is just a measure for how
  394. [16:59:25] <Gerrit> opaque it is
  395. [16:59:27] <Gerrit> the higher the opacity, the more radiation is blocked
  396. [16:59:30] <Gerrit> that's a function of wavelength and of species
  397. [16:59:32] <smuckerz> Faqtotum, i find it impressive that someone would do this
  398. [16:59:32] <Gerrit> so here we take some ingredients in the atompsehire
  399. [16:59:35] <Gerrit> we take water vapour, ozone, co2, n2o, etcetera
  400. [16:59:37] <Gerrit> that's the clear-sky part
  401. [16:59:40] <Gerrit> with this we can calculate clear-sky radiation
  402. [16:59:42] <Gerrit> which is by comparison to the cloudy simulation simple
  403. [16:59:43] <+Faqtotum> smuckerz: indeed
  404. [16:59:45] <Gerrit> but my thesis focusses on clouds, so I want to calculate clouds
  405. [16:59:47] <Gerrit> 27:25 frame 23
  406. [16:59:50] <Gerrit> that's why the 3rd term here becomes importan,t which is the scattering
  407. [16:59:52] <Gerrit> source term
  408. [16:59:55] <Gerrit> as radiation meets cloud particles, it can be scattered away, but it can
  409. [16:59:57] <Gerrit> also be scattered into the line of sight, and that's what's reperesented
  410. [17:00:00] <Gerrit> by the 3rd term, which starts with the phase function.
  411. [17:00:02] <Gerrit> 27:47 frame 24
  412. [17:00:05] <Gerrit> the phase function describes the photon
  413. [17:00:07] <Gerrit> the photon meets an ice particle and gets scattered, then it changes the
  414. [17:00:10] <Gerrit> direction of the radiation
  415. [17:00:12] <Gerrit> now in this figure, the photon gets scattered twice and in the end it's
  416. [17:00:15] <Gerrit> more or less, going it's in a very similar direction as when it started
  417. [17:00:17] <Gerrit> so it's important really, necessary to consider the scattering
  418. [17:00:20] <Gerrit> the scattering phase function is the probability density function in
  419. [17:00:22] <Gerrit> describing in what direction the radiation gets scattered
  420. [17:00:25] <Gerrit> unlike the photon in this animation, not every photon gets scattered in
  421. [17:00:27] <Gerrit> the same way
  422. [17:00:30] <Gerrit> one way to model this, is to really follow each photon step by step
  423. [17:00:32] <Gerrit> it's a slow method, slightly faster than this photon, but still a slow
  424. [17:00:35] <Gerrit> method that we can directly apply a scattering phase function such as the
  425. [17:00:37] <Gerrit> two examples in the figure
  426. [17:00:40] <Gerrit> and this scattering phase function, and also the scattering cross section,
  427. [17:00:42] <Gerrit> which determines how likely it is that a scatetring event occurs
  428. [17:00:45] <Gerrit> depends strongly on the size of the, on the wavelength in relation to the
  429. [17:00:47] <Gerrit> size of the particle
  430. [17:00:50] <Gerrit> in the figure there is an example, there's an infrared phase function,
  431. [17:00:52] <Gerrit> which nonlinear for a 50 micrometer particle, an infrared phase function
  432. [17:00:55] <Gerrit> wavelength is much smaller than the size of the particle, then most of the
  433. [17:00:57] <Gerrit> radiation goes, even if it gets scattered it goes straight ahead,
  434. [17:01:00] <Gerrit> whereas in the microwave, it's much more, in this case it's a similar
  435. [17:01:02] <Gerrit> size, size of particle is similar to the size of the wavelength,
  436. [17:01:05] <Gerrit> and it's more scattered in all directions, the phase function
  437. [17:01:07] <Gerrit> so knowing this, we can
  438. [17:01:10] <Gerrit> 29:50 frame 25
  439. [17:01:12] <Gerrit> solve now single frequency, which is, including the clear-sky part, and
  440. [17:01:15] <Gerrit> the cloudy part.
  441. [17:01:17] <Gerrit> But this is rather expensive, unfortunately.
  442. [17:01:20] <Gerrit> If we want, for example, to simulate an entire channel for AVHRR channel
  443. [17:01:22] <Gerrit> 5, you would need about 5000 calculations, which is too many to do
  444. [17:01:25] <Gerrit> practically, certainly if one wants to cloudy radiative transfer
  445. [17:01:27] <Gerrit> but this can be optimised, even in the cloudy case, and that is what my
  446. [17:01:30] <Gerrit> 3rd paper is about
  447. [17:01:32] <Gerrit> 30:26 frame 26
  448. [17:01:35] <Gerrit> Simulation cloudy thermal infrared radiances with an optimised frequency
  449. [17:01:37] <Gerrit> grid in the radiative transfer model ARTS
  450. [17:01:40] <Gerrit> It's a ... sentenec, but what it means that instead of 5000 frequencies,
  451. [17:01:42] <Gerrit> to simulate this channel, you can do with only five
  452. [17:01:44] r0bby_ [~wakawaka@guifications/user/r0bby] has quit IRC: Remote host closed the connection
  453. [17:01:45] <Gerrit> this is a continuation of work presented by
  454. [17:01:47] <Gerrit> 30:47 frame 27
  455. [17:01:49] mrmist [~mrmist@freenode/staff/pdpc.active.mrmist] has joined ##belz
  456. [17:01:50] <Gerrit> Stefan Buehler and collaborators in 2010 it was published, called
  457. [17:01:52] <Gerrit> simulated annealing, and it's a numerical method to give the same channel
  458. [17:01:55] <Gerrit> radiance, in this application, as the large number of frequencies
  459. [17:01:57] <Gerrit> now this paper was focussing directly on clear-sky radiation, but I
  460. [17:02:00] <Gerrit> and ... and derived for clear-sky, applying this method,
  461. [17:02:02] <Gerrit> 31:19 frame 28
  462. [17:02:05] <Gerrit> I get five channels, five frequencies with associated weights, all inside
  463. [17:02:07] <Gerrit> AVHRR channel five, and the daring part here, is that I'm actually
  464. [17:02:10] <Gerrit> applying it to cloudy simulations
  465. [17:02:12] <Gerrit> 31:32 frame 29
  466. [17:02:15] <Gerrit> does it work? how do we know it works.
  467. [17:02:17] <Gerrit> We can test it, and what do we see?
  468. [17:02:20] <Gerrit> ... it works
  469. [17:02:22] <Gerrit> we have the references brightness temperature here
  470. [17:02:25] <Gerrit> where we used only one photon per frequency for 5000, more than 5000
  471. [17:02:27] <Gerrit> frequencies, which took about 20 times longer to run than the fast setup,
  472. [17:02:30] <Gerrit> that is on the y-axis here, where I used 1000 photons times five
  473. [17:02:32] <Gerrit> frequencies
  474. [17:02:35] <Gerrit> the total number of photons is similar, but the reference is more than 20
  475. [17:02:37] <Gerrit> times faster
  476. [17:02:39] r0bby_ [~wakawaka@guifications/user/r0bby] has joined ##belz
  477. [17:02:40] <Gerrit> now I have a setup that I can use for whatever I'm going to do in the
  478. [17:02:42] <Gerrit> future
  479. [17:02:45] <Gerrit> 32:17 frame 30
  480. [17:02:47] <Gerrit> which is, among other things, a cloudy study of radiances through clouds,
  481. [17:02:48] <smuckerz> hello mrmist
  482. [17:02:49] <mrmist> Gerrit: you seem to be posting a novel here. I'd suggest using pastebin
  483. [17:03:01] <smuckerz> he was already told
  484. [17:03:04] <smuckerz> to use pastebin
  485. [17:03:24] rs-afk [~rs@unaffiliated/red-squirrel] is now known as red-squirrel
  486. [17:03:26] <smuckerz> 23 minutes ago
  487. [17:03:30] <smuckerz> to be precise
  488. [17:03:32] <+Faqtotum> it seems to have stopped
  489. [17:03:36] <+Faqtotum> idk why
  490. [17:03:45] <mrmist> it does.funny how that happens sometimes.
  491. [17:04:00] <smuckerz> hmm...
  492. [17:04:03] <+Faqtotum> maybe he crashed irssi
  493. [17:04:09] <smuckerz> what's the term...
  494. [17:04:09] <+Faqtotum> happens sometimes
  495. [17:04:10] <smuckerz> shun?
  496. [17:05:07] <+Faqtotum> afk
  497. [17:07:17] <smuckerz> mrmist, thank you
  498. [17:07:26] <mrmist> np
  499. [17:07:44] <smuckerz> no clue if you did anything
  500. [17:07:49] <smuckerz> but you did show up
  501. [17:07:49] <smuckerz> :)
  502. [17:07:50] <mrmist> I joined
  503. [17:08:07] robbyoconnor [~wakawaka@guifications/user/r0bby] has joined ##belz
  504. [17:08:14] <smuckerz> yup :)
  505. [17:08:19] <smuckerz> that was something :)
  506. [17:08:27] Bateau [~Ghost@unaffiliated/section9] has joined ##belz
  507.  
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!