2 
00:00:06,613 --> 00:00:15,520 
Theory is all well and good. 
But if we're 

3 
00:00:15,520 --> 00:00:20,340 
going to invest money in trying to limit 
C02 emissions to avoid climate change, 

4 
00:00:20,340 --> 00:00:25,620 
really we want something more. 
We want to have evidence that everything 

5 
00:00:25,620 --> 00:00:31,540 
is working the way it is thought that it 
should theoretically. 

6 
00:00:31,540 --> 00:00:35,360 
The way to do that is to see if the 
Earth is in fact warming up. 

7 
00:00:36,710 --> 00:00:43,460 
There are three different major 
types of temperature measurements. 

8 
00:00:43,460 --> 00:00:47,690 
Those over land, those over sea, and those 
from satellites. 

9 
00:00:48,900 --> 00:00:51,460 
Land temperature records go back 

10 
00:00:54,190 --> 00:00:59,200 
to the invention of the thermometer by 
Gabriel Fahrenheit in 1714. 

11 
00:01:00,600 --> 00:01:04,320 
The thermometer is very easy to calibrate,

12 
00:01:04,320 --> 00:01:07,210 
because you can just stick it in boiling 
water and then stick 

13 
00:01:07,210 --> 00:01:12,290 
it in ice water and make a scale between 
those two things. 

14 
00:01:12,290 --> 00:01:17,600 
And since we know today what temperatures 
ice melts and boils at, 

15 
00:01:17,600 --> 00:01:19,750 
It's very easy to interpret 

16 
00:01:21,370 --> 00:01:25,840 
very old thermometer measurements in terms 
of what the temperature really was. 

17 
00:01:25,840 --> 00:01:33,020 
This is as opposed to measurements of 
CO2 concentration or something like that,

18 
00:01:33,020 --> 00:01:37,450 
that would have used methods that are hard 
to reconstruct today.

19 
00:01:37,450 --> 00:01:40,530 
Some kinds of old data are 
really not worth that much. 

20 
00:01:40,530 --> 00:01:43,201 
But old temperature data is pretty 
reliable. 

22 
00:01:45,925 --> 00:01:50,260 
One possible 

23 
00:01:50,260 --> 00:01:54,120 
problem with the land temperature 
measurements is called 

24 
00:01:54,120 --> 00:01:58,120 
the urban heat island effect. 
This is a very real effect. 

25 
00:01:58,120 --> 00:02:04,310 
The idea is that in a vegetated 
landscape, the incoming 

26 
00:02:04,310 --> 00:02:11,068 
sunlight energy can be carried back away 
by the latent heat in water vapor. 

27 
00:02:11,068 --> 00:02:16,230 
Because plants mine water from the ground 

28 
00:02:16,230 --> 00:02:19,040 
and they encourage it to 
evaporate. 

29 
00:02:19,040 --> 00:02:22,400 
Whereas in a urban landscape The 
same heat input

30 
00:02:22,400 --> 00:02:27,500 
can only be balanced by warming up 
the ground enough...

31 
00:02:27,500 --> 00:02:31,150 
that infra red energy leaving can 

32 
00:02:31,150 --> 00:02:34,260 
can balance the incoming 
light. 

33 
00:02:34,260 --> 00:02:36,670 
And so in 

34 
00:02:38,010 --> 00:02:43,075 
maps of temperatures or infra red images 
from space around cities,

35 
00:02:43,075 --> 00:02:45,640 
it's very clear that it is warmer in 

36 
00:02:45,640 --> 00:02:49,520 
urban environments than it is in rural 
environments. 

37 
00:02:49,520 --> 00:02:56,010 
The problem would come if a lot of the 
places where they measure temperatures...

38 
00:02:57,590 --> 00:03:03,150 
have transitioned over time from 
relatively green places like 

39 
00:03:03,150 --> 00:03:06,020 
this, to relatively urban places like 
that,

40 
00:03:06,020 --> 00:03:10,670 
and if you then took the warming, it would 
be due to the urban heat island effect,

41 
00:03:10,670 --> 00:03:14,760 
and you call that global warming due to 
rising C02 concentrations. 

42 
00:03:14,760 --> 00:03:21,000 
That would be a wrong attribution. 
But there have been many 

43 
00:03:22,600 --> 00:03:28,940 
studies that have taken the thermometer 
data and put them together into the global 

44 
00:03:28,940 --> 00:03:30,040 
average temperature change,

45 
00:03:30,040 --> 00:03:34,200 
and they take different methods of trying 
to filter out

46 
00:03:34,200 --> 00:03:37,720 
the urban heat island effect from 
their averages,

47 
00:03:37,720 --> 00:03:43,800 
by taking stations out if they seem like 
they're likely to be a problem. 

48 
00:03:43,800 --> 00:03:47,680 
And as you can see in this plot, 
basically everybody is 

49 
00:03:47,680 --> 00:03:51,710 
getting the same answer for the average 
temperature of the earth. 

50 
00:03:51,710 --> 00:03:54,180 
It turns out that the urban heat Island 

51 
00:03:54,180 --> 00:03:58,030 
effect, while it is a very real effect, Is 
not

52 
00:03:58,030 --> 00:04:02,600 
an important one for setting the 
global average temperature. 

53 
00:04:02,600 --> 00:04:05,655 
It's not confounding the 
global warming 

54 
00:04:05,655 --> 00:04:09,580 
signal that's being measured by temp, by 
thermometers on land.