2 
00:00:02,440 --> 00:00:16,470 
A different way of measuring 
the temperature is over the ocean. 

4 
00:00:16,470 --> 00:00:20,250 
And this is good because it 
doesn't have any possibility of 

5 
00:00:20,250 --> 00:00:24,000 
an urban heat island effect and so it's a 
separate check. 

6 
00:00:24,000 --> 00:00:28,450 
But there has been a change in the way 
that temperatures have been measured at 

7 
00:00:28,450 --> 00:00:35,300 
sea over time. 
In in old times they would cast 

8 
00:00:35,300 --> 00:00:39,680 
a bucket over the side on a rope and then 
pull a 

9 
00:00:39,680 --> 00:00:43,590 
bucket full of sea water on deck and stick 
a thermometer in it. 

10 
00:00:43,590 --> 00:00:49,140 
And this had the possibility of bias, 
because water 

11 
00:00:49,140 --> 00:00:53,870 
on the outside of the bucket could 
evaporate and could cool that bucket down 

12 
00:00:53,870 --> 00:01:00,200 
just a bit relative to what it was in the 
actual water that the boat is floating in. 

13 
00:01:00,200 --> 00:01:04,190 
In the mid '40s, there was a 
transition in the sea surface 

15 
00:01:04,190 --> 00:01:08,090 
temperature data toward measurements that 
were made

16 
00:01:08,090 --> 00:01:11,930 
on the cooling water intake to the 
engine. 

17 
00:01:11,930 --> 00:01:15,720 
They have to bring water in to cool 
down the engine as it's running, 

18 
00:01:15,720 --> 00:01:17,390 
you measure the temperature as the water 

19 
00:01:17,390 --> 00:01:20,000 
comes in, before you've dumped heat into 
it. 

20 
00:01:20,000 --> 00:01:22,330 
And that can tell you a sea surface 
temperature. 

21 
00:01:22,330 --> 00:01:26,000 
Of course this cooling intake might be 
coming in from a little 

22 
00:01:26,000 --> 00:01:29,640 
bit below the surface; it may not be 
exactly the same temperature as here. 

23 
00:01:31,640 --> 00:01:38,680 
To put together the land and the sea 
surface temperature records. 

24 
00:01:38,680 --> 00:01:40,380 
you can't just compare them directly 

25 
00:01:40,380 --> 00:01:42,740 
because they're different quantities.

26 
00:01:42,740 --> 00:01:46,440 
The land tends to heat up more than the 
ocean would 

27 
00:01:46,440 --> 00:01:49,430 
the same reason as this urban 
heat island effect. 

28 
00:01:49,430 --> 00:01:51,410 
The ocean would be like a super 

29 
00:01:51,410 --> 00:01:53,910 
vegetative place with lots and lots of 

30 
00:01:53,910 --> 00:01:56,780 
capability for water vapor to carry away 
heat. 

31 
00:01:56,780 --> 00:02:03,170 
And also, the ocean is a big heat sink 
itself, and so the land surface 

32 
00:02:03,170 --> 00:02:08,180 
is actually expected to warm more quickly 
than the ocean surface for those reasons. 

33 
00:02:08,180 --> 00:02:10,490 
The way to compare these 

34 
00:02:10,490 --> 00:02:14,380 
different temperatures is within a climate 
model, 

35 
00:02:14,380 --> 00:02:16,870 
where you impose the temperature 

36 
00:02:16,870 --> 00:02:21,350 
at the sea surface from the measured ship 
data,

37 
00:02:21,350 --> 00:02:23,250 
and then you use the model 

38 
00:02:23,250 --> 00:02:26,370 
to calculate, to predict the temperatures 


39 
00:02:26,370 --> 00:02:31,480 
in the atmosphere, both in the air just above the 

40 
00:02:31,480 --> 00:02:34,280 
water, but also, more importantly, over 
the land. 

41 
00:02:34,280 --> 00:02:36,740 
And then you can compare these model 
predicted land 

42 
00:02:37,820 --> 00:02:42,450 
temperatures with the actual temperatures 
measured at those weather stations. 

43 
00:02:42,450 --> 00:02:45,620 
And it turns out that the correction, 
there was 

44 
00:02:45,620 --> 00:02:48,980 
a significant correction that had to be 
made between this 

45 
00:02:48,980 --> 00:02:54,000 
technique of measuring sea surface 
temperature in buckets versus 

46 
00:02:54,000 --> 00:02:59,470 
measuring the sea surface temperatures 
from from the cooling intakes. 

47 
00:02:59,470 --> 00:03:03,950 
But aside from that, the temperature 
signals that come 

48 
00:03:03,950 --> 00:03:07,660 
out from this are consistent with those 
from the after that 

49 
00:03:07,660 --> 00:03:11,650 
correction is made, both the land and the 
sea surface temperatures 

50 
00:03:11,650 --> 00:03:15,620 
are telling us the same thing about the 
temperatures of the Earth.