1
00:00:07,525 --> 00:00:09,815
In 132 CE,

2
00:00:09,815 --> 00:00:12,145
Chinese polymath Zhang Heng

3
00:00:12,145 --> 00:00:15,585
presented the Han court with 
his latest invention.

4
00:00:15,585 --> 00:00:17,675
This large vase, he claimed,

5
00:00:17,675 --> 00:00:21,325
could tell them whenever an earthquake 
occurred in their kingdom–

6
00:00:21,325 --> 00:00:24,595
including the direction 
they should send aid.

7
00:00:24,595 --> 00:00:26,505
The court was somewhat skeptical,

8
00:00:26,505 --> 00:00:31,005
especially when the device triggered 
on a seemingly quiet afternoon.

9
00:00:31,005 --> 00:00:34,054
But when messengers came 
for help days later,

10
00:00:34,054 --> 00:00:36,614
their doubts turned to gratitude.

11
00:00:36,614 --> 00:00:41,114
Today, we no longer rely on pots to 
identify seismic events,

12
00:00:41,114 --> 00:00:46,150
but earthquakes still offer a unique
challenge to those trying to track them.

13
00:00:46,150 --> 00:00:49,139
So why are earthquakes so 
hard to anticipate,

14
00:00:49,139 --> 00:00:52,039
and how could we get better 
at predicting them?

15
00:00:52,039 --> 00:00:53,099
To answer that,

16
00:00:53,099 --> 00:00:57,539
we need to understand some theories 
behind how earthquakes occur.

17
00:00:57,539 --> 00:01:01,794
Earth’s crust is made from several vast, 
jagged slabs of rock

18
00:01:01,794 --> 00:01:03,749
called tectonic plates,

19
00:01:03,749 --> 00:01:08,479
each riding on a hot, partially molten
layer of Earth’s mantle.

20
00:01:08,479 --> 00:01:11,289
This causes the plates to 
spread very slowly,

21
00:01:11,289 --> 00:01:14,839
at anywhere from 1 to 20 
centimeters per year.

22
00:01:14,839 --> 00:01:17,289
But these tiny movements are powerful 
enough

23
00:01:17,289 --> 00:01:20,799
to cause deep cracks in the
interacting plates.

24
00:01:20,799 --> 00:01:22,399
And in unstable zones,

25
00:01:22,399 --> 00:01:27,219
the intensifying pressure may 
ultimately trigger an earthquake.

26
00:01:27,219 --> 00:01:30,249
It’s hard enough to monitor these 
miniscule movements,

27
00:01:30,249 --> 00:01:35,589
but the factors that turn shifts into 
seismic events are far more varied.

28
00:01:35,589 --> 00:01:38,399
Different fault lines juxtapose 
different rocks–

29
00:01:38,399 --> 00:01:42,269
some of which are stronger–or weaker–
under pressure.

30
00:01:42,269 --> 00:01:46,979
Diverse rocks also react differently to
friction and high temperatures.

31
00:01:46,979 --> 00:01:50,431
Some partially melt, and can release
lubricating fluids

32
00:01:50,431 --> 00:01:52,230
made of superheated minerals

33
00:01:52,230 --> 00:01:54,420
that reduce fault line friction.

34
00:01:54,420 --> 00:01:56,370
But some are left dry,

35
00:01:56,370 --> 00:01:59,140
prone to dangerous build-ups of pressure.

36
00:01:59,140 --> 00:02:03,680
And all these faults are subject to 
varying gravitational forces,

37
00:02:03,680 --> 00:02:08,531
as well as the currents of hot rocks
moving throughout Earth’s mantle.

38
00:02:08,531 --> 00:02:11,755
So which of these hidden variables 
should we be analyzing,

39
00:02:11,755 --> 00:02:15,955
and how do they fit into our 
growing prediction toolkit?

40
00:02:15,955 --> 00:02:19,829
Because some of these forces occur 
at largely constant rates,

41
00:02:19,829 --> 00:02:23,159
the behavior of the plates 
is somewhat cyclical.

42
00:02:23,159 --> 00:02:27,609
Today, many of our most reliable clues 
come from long-term forecasting,

43
00:02:27,609 --> 00:02:31,893
related to when and where earthquakes 
have previously occurred.

44
00:02:31,893 --> 00:02:33,573
At the scale of millennia,

45
00:02:33,573 --> 00:02:37,533
this allows us to make predictions 
about when highly active faults,

46
00:02:37,533 --> 00:02:38,873
like the San Andreas,

47
00:02:38,873 --> 00:02:41,943
are overdue for a massive earthquake.

48
00:02:41,943 --> 00:02:44,193
But due to the many variables involved,

49
00:02:44,193 --> 00:02:47,783
this method can only predict 
very loose timeframes.

50
00:02:47,783 --> 00:02:49,693
To predict more imminent events,

51
00:02:49,693 --> 00:02:55,303
researchers have investigated the 
vibrations Earth elicits before a quake.

52
00:02:55,303 --> 00:02:57,990
Geologists have long used seismometers

53
00:02:57,990 --> 00:03:01,920
to track and map these tiny shifts
in the earth’s crust.

54
00:03:01,920 --> 00:03:04,910
And today, most smartphones are 
also capable

55
00:03:04,910 --> 00:03:07,670
of recording primary seismic waves.

56
00:03:07,670 --> 00:03:09,740
With a network of phones around the globe,

57
00:03:09,740 --> 00:03:12,680
scientists could potentially 
crowd source a rich,

58
00:03:12,680 --> 00:03:16,425
detailed warning system that alerts 
people to incoming quakes.

59
00:03:16,955 --> 00:03:21,325
Unfortunately, phones might not be able
to provide the advance notice needed

60
00:03:21,325 --> 00:03:23,355
to enact safety protocols.

61
00:03:23,355 --> 00:03:26,075
But such detailed readings would still be 
useful

62
00:03:26,075 --> 00:03:29,425
for prediction tools like NASA’s 
Quakesim software,

63
00:03:29,425 --> 00:03:32,255
which can use a rigorous blend of 
geological data

64
00:03:32,255 --> 00:03:34,765
to identify regions at risk.

65
00:03:34,765 --> 00:03:36,765
However, recent studies indicate

66
00:03:36,765 --> 00:03:41,635
the most telling signs of a quake might be
invisible to all these sensors.

67
00:03:41,635 --> 00:03:43,005
In 2011,

68
00:03:43,005 --> 00:03:46,265
just before an earthquake struck 
the east coast of Japan,

69
00:03:46,265 --> 00:03:50,065
nearby researchers recorded surprisingly 
high concentrations

70
00:03:50,065 --> 00:03:54,425
of the radioactive isotope pair 
radon and thoron.

71
00:03:54,425 --> 00:03:58,167
As stress builds up in the crust right 
before an earthquake,

72
00:03:58,167 --> 00:04:02,407
microfractures allow these gases 
to escape to the surface.

73
00:04:02,407 --> 00:04:07,042
These scientists think that if we built 
a vast network of radon-thoron detectors

74
00:04:07,042 --> 00:04:08,992
in earthquake-prone areas,

75
00:04:08,992 --> 00:04:11,522
it could become a promising 
warning system–

76
00:04:11,522 --> 00:04:14,532
potentially predicting quakes 
a week in advance.

77
00:04:14,532 --> 00:04:15,232
Of course,

78
00:04:15,232 --> 00:04:17,432
none of these technologies 
would be as helpful

79
00:04:17,432 --> 00:04:20,572
as simply looking deep inside 
the earth itself.

80
00:04:20,572 --> 00:04:22,242
With a deeper view we might be able

81
00:04:22,242 --> 00:04:26,852
to track and predict large-scale 
geological changes in real time,

82
00:04:26,852 --> 00:04:29,734
possibly saving tens of thousands 
of lives a year.

83
00:04:30,034 --> 00:04:30,874
But for now,

84
00:04:30,874 --> 00:04:35,404
these technologies can help us prepare 
and respond quickly to areas in need–

85
00:04:35,404 --> 00:04:39,462
without waiting for directions 
from a vase.