2 00:00:11,261 --> 00:00:15,491 Tree ring widths are one of the best a ways of piecing 3 00:00:15,491 --> 00:00:20,620 together climate changes of the last thousand years or so. 4 00:00:20,620 --> 00:00:23,964 A tree has a seasonal cycle of growth, If it's 5 00:00:23,964 --> 00:00:27,780 growing in a place that has a seasonal cycle in the weather. 6 00:00:27,780 --> 00:00:29,984 This method doesn't work in the tropics, 7 00:00:29,984 --> 00:00:31,900 where it's warm all the time. 8 00:00:31,900 --> 00:00:35,420 It works in the higher latitudes, where you have a distinct summer and winter. 9 00:00:35,420 --> 00:00:36,559 Most of the growth 10 00:00:36,559 --> 00:00:38,180 happens in the summertime. 11 00:00:38,180 --> 00:00:42,180 And wintertime, you get a layer that has a different color. 12 00:00:42,180 --> 00:00:46,410 You can count the rings in a stump, 13 00:00:46,410 --> 00:00:50,880 and each one of those rings corresponds to a year. 14 00:00:51,980 --> 00:00:55,200 How wide the ring is, 15 00:00:55,200 --> 00:00:58,290 in a particular year, depends on a whole bunch of things. 16 00:00:58,290 --> 00:01:02,820 In addition to temperature, it can depend on the availability of water. 18 00:01:02,820 --> 00:01:07,810 It can depend on sunlight, the age of the tree, things like that. 19 00:01:07,810 --> 00:01:12,550 There's a lot of noise in this method that has to be averaged 20 00:01:12,550 --> 00:01:16,960 out, or gotten rid of in one way or another. 21 00:01:16,960 --> 00:01:21,940 You get the primary calibration of tree ring width 22 00:01:21,940 --> 00:01:26,480 to temperature by comparing it with the instrumental temperature record, 23 00:01:26,480 --> 00:01:30,200 the time span over which there are measurements from thermometers. 25 00:01:31,390 --> 00:01:33,480 That can give you a calibration 26 00:01:33,480 --> 00:01:36,330 you can project further back into the past 27 00:01:36,330 --> 00:01:39,150 to say If the width was here, then using 28 00:01:39,150 --> 00:01:42,980 that same calibration the temperature must have been there. 29 00:01:44,270 --> 00:01:47,980 And the way noise is further reduced, 30 00:01:47,980 --> 00:01:52,348 is by looking at spatial patterns of these records from all 31 00:01:52,348 --> 00:01:53,060 around the world. 32 00:01:53,060 --> 00:01:56,250 This is a terribly drawn map, but you can use your imagination 33 00:01:56,250 --> 00:01:59,990 and figure this is South America and North America. 34 00:01:59,990 --> 00:02:03,260 If you have a distribution of tree ring records that 35 00:02:03,260 --> 00:02:06,870 cover the whole surface of the earth, or a large fraction 36 00:02:06,870 --> 00:02:11,570 of it, you can compare the variations that you see in 37 00:02:11,570 --> 00:02:16,760 the tree rings with patterns of variation like the El Nino cycle. 38 00:02:16,760 --> 00:02:21,900 From the instrumental record, we have some idea that when the Earth's 40 00:02:21,900 --> 00:02:26,090 climate changes, it changes with particular spatial patterns. 41 00:02:26,090 --> 00:02:31,110 The way that noise is eliminated from these tree ring records .. 42 00:02:32,780 --> 00:02:37,800 is by accepting the temperature changes from records that are coherent, in 43 00:02:37,800 --> 00:02:43,270 some way that's recognizable from the way the climate works today. 44 00:02:43,270 --> 00:02:46,130 And calling that real information. 45 00:02:46,130 --> 00:02:49,890 Then if there's some weird wacko record here that does something totally 46 00:02:49,890 --> 00:02:54,340 different than everybody else it will get excluded, by this method. 47 00:02:54,340 --> 00:03:00,810 It's not a secret, but there is a problem with the tree ring data. 48 00:03:00,810 --> 00:03:03,830 And that is that, in the most recent few 49 00:03:03,830 --> 00:03:13,780 decades of the instrumental records, starting from about 1970s, to today. 52 00:03:13,780 --> 00:03:16,880 The temperature, as measured by thermometers, has gone up, but the 53 00:03:16,880 --> 00:03:22,170 tree rings don't show that, and it's not clear why that is. 54 00:03:22,170 --> 00:03:26,190 It could be because rising CO2 concentrations 55 00:03:26,190 --> 00:03:28,770 actually fertilize the trees in some way. 56 00:03:28,770 --> 00:03:34,810 It could be because one of the components of acid rain is nitric acid. 57 00:03:34,810 --> 00:03:37,540 Which is nitrate, a fertilizer. 59 00:03:37,540 --> 00:03:41,500 Maybe fertilizer deposition has impacted the tree 60 00:03:41,500 --> 00:03:46,080 ring, the fidelity of the tree ring method. 61 00:03:46,080 --> 00:03:49,200 Does this destroy our confidence in the 62 00:03:49,200 --> 00:03:52,820 tree ring method to put together ancient climates? 63 00:03:52,820 --> 00:03:55,630 No. it doesn't destroy it. It undermines it a little bit. 64 00:03:55,630 --> 00:04:00,250 Like I said before, these proxy measurements are what they are. 65 00:04:00,250 --> 00:04:03,940 They're as good as they are and we have to use them with some caution. 66 00:04:03,940 --> 00:04:06,800 If we could actually have thermometers from a thousand years 67 00:04:06,800 --> 00:04:09,270 ago, that would be better, but this is what we have. 68 00:04:09,270 --> 00:04:16,750 It's possible that there's a cap on the the sensitivity of the method. 69 00:04:16,750 --> 00:04:18,420 We have to live with that possibility.