Volcanic Hazards
The
multiplicity of events that can occur from a single volcano adds
to the complexity of hazard prediction and civil defence response
compared to straightforward earth-shaking events or a chemical spill.
The following section delves into the many types of hazards associated
with volcanic eruptions, following on from 1.5 in the previous section.
In addition to the following section, you should read pages 218-228
of Chapter 12 Volcanoes as a hazard of your textbook
by Bryant (1991).
Hazard versus risk
Volcanic
hazard is defined as the probability of an area being affected by
potentially destructive volcanic processes or products within a
given period of time. Volcanic risk is the possibility of a loss
(e.g. of life, property or production potential) within the area
subject to the hazards. Assessment of risk involves calculation
of the following relationship:
Risk = Value x Vulnerability x Hazard
where
value can include number of lives, property, civil works
and production capacity threatened, and vulnerability
is a measure of the proportion of the value (0-100%) likely to be
lost in a given hazardous event.
Volcanic disasters versus other types of disasters
On
a global basis, volcanic and related hazards occur less frequently,
affect fewer people, cause fewer casualties and smaller economic
losses than many other natural or man-made disasters (Fig. 5A, B).
Figure
5.
The
deadliest eruption in history was in 1815, Tambora, Indonesia which
killed 92 000 people. However, 500 000 people were killed in the
worst hurricane in the Ganges Delta, Bangladesh in 1970, and the
worst natural disaster in history was the Huahsien Earthquake in
China in 1556 which resulted in more than 820 000 fatalities.
In
the U.S.A. the annual economic loss due to volcanic activity is
an order of magnitude smaller than that from earthquakes, and almost
two orders of magnitude less than the loss from floods and ground
failures. It has been suggested that in the U.S. the average person
is more likely to die from a heart attack shovelling snow or from
a lightning strike than from either an earthquake or volcanic eruption.
However, for more densely populated countries such as Japan, Indonesia
and the Philippines, volcanic hazards have a far greater potential
for causing enormous economic and social losses.
Historical volcanic disasters
Table
5. Selected volcanic disasters and their respective human fatalities
in the period from 1000-1985 A.D. Adapted from Tilling (1989).
|
Primary
cause of death |
Volcano
|
Location
|
Year
|
Pyroclastic
flow
|
Lahars
|
Lava
flow
|
Tsunami
|
Starvation
|
Merapi
|
Indonesia
|
1006
|
1000 |
|
|
|
|
Kelut
|
"
|
1586
|
|
10000 |
|
|
|
Vesuvio
|
Italy
|
1631
|
|
|
c.
18000 |
|
|
Etna
|
"
|
1669
|
|
|
c.
10000 |
|
|
Merapi
|
Indonesia
|
1672
|
300 |
|
|
|
|
Awu
|
"
|
1711
|
|
3200 |
|
|
|
Oshima
|
Japan
|
1741
|
|
|
|
1480 |
|
Cotopaxi
|
Ecuador
|
1741
|
|
1000 |
|
|
|
Papadajan
|
Indonesia
|
1772
|
2960 |
|
|
|
|
Lakagigar
|
Iceland
|
1783
|
|
|
|
|
9340 |
Asama
|
Japan
|
1783
|
1150 |
|
|
|
|
Unzen
|
"
|
1792
|
|
|
|
15190 |
|
Mayon
|
Philippines
|
1814
|
1200 |
|
|
|
|
Tambora
|
Indonesia
|
1815
|
12000 |
|
|
|
80000 |
Galunggung
|
"
|
1822
|
|
4000 |
|
|
|
Nevado
del Ruíz
|
Colombia
|
1845
|
|
1000 |
|
|
|
Awu
|
Indonesia
|
1856
|
|
3000 |
|
|
|
Cotopaxi
|
Ecuador
|
1877
|
|
1000 |
|
|
|
Krakatau
|
Indonesia
|
1883
|
|
|
|
36420 |
|
Awu
|
Indonesia
|
1892
|
|
1530 |
|
|
|
Soufrière
|
St.
Vincent
|
1902
|
1560 |
|
|
|
|
Mont
Pelée
|
Martinique
|
1902
|
29000 |
|
|
|
|
Santa
Maria
|
Guatemala
|
1902
|
6000 |
|
|
|
|
Taal
|
Philippines
|
1911
|
1330 |
|
|
|
|
Kelut
|
Indonesia
|
1919
|
|
5110 |
|
|
|
Merapi
|
"
|
1951
|
1300 |
|
|
|
|
Lamington
|
P.N.G.
|
1951
|
2940 |
|
|
|
|
Hibok-Hibok
|
Philippines
|
1951
|
500 |
|
|
|
|
Agung
|
Indonesia
|
1963
|
1900 |
|
|
|
|
El
Chichón
|
Mexico
|
1982
|
>
2000 |
|
|
|
|
Nevado
del Ruíz
|
Colombia
|
1985
|
|
25000 |
|
|
|
|
|
Totals
|
|
65080 |
56900 |
28000 |
53090 |
89340 |
The loss of human lives is probably the best basis by which to compare
historic eruptions, other effects such as economic losses are not
completely known or difficult to compare for different times and
places. The historical record, presented in Table 5, is dominated
by a few particularly destructive events. Between 1000 A.D. and
1985, more than 300 000 people have been killed directly or indirectly
by volcanic activity. Most of these lives have been lost in the
circum-Pacific area, with over 65% of the losses in only three countries
- Colombia, Indonesia and Japan.
Table
6. Proportions of historical human fatalities from differing
volcanic hazards, adapted from Tilling (1989).
Primary cause of death
|
1600 - 1899
|
1900 - 1986
|
Pyroclastic flows/debris
avalanches |
18200 |
9.8 % |
36000 |
48.4 % |
Lahars and floods |
8300 |
4.5 % |
28400 |
37.4 % |
Tephra falls and ballistics
|
8000 |
4.3 % |
3000 |
4.0 % |
Tsunami |
43600 |
23.4 % |
400 |
0.5 % |
Starvation, disease etc.
|
92100 |
49.4 % |
3200 |
4.2 % |
Lava flows |
900 |
0.5 % |
100 |
0.1 % |
Gases and acid rains |
- |
- |
1900 |
2.5 % |
Other or unknown |
15100 |
8.1 % |
2200 |
2.5 % |
|
Totals |
186200 |
|
76000 |
|
Fatalities per year
|
620 |
|
880 |
|
The average number of fatalities per year in the 20th Century is
higher than that of the preceding three centuries (Table 6), despite
developments made in the scientific study of volcanic activity over
this time. However, in the 20th Century the incidence of fatalities
caused by indirect hazards (including starvation and tsunamis) has
decreased (Table 6), probably due to improved communications and
relief systems. These conclusions are however biased by the short
interval of time (compared to geologic time) considered, and that
there were no large tsunami-producing eruptions in the 20th Century.
The large increase in the number of deaths associated with pyroclastic
flows, debris avalanches and debris flows in the 20th century is
biased by the two disastrous events at Mont Pelée in 1902, and Nevado
del Ruíz in 1985.
Short and long term volcanic hazards
There are two broad levels of volcanic hazards:
(1) those that occur frequently (i.e. more than once per century)
and which most people living near a volcano are likely to experience
at least once during their lifetimes - here termed short or immediate
term hazards;
(2) those that occur less frequently (much less than once per century)
and which people are unlikely to experience in their lifetimes -
termed long term or potential hazards. An example of short term
hazards are lahars on the slopes of Ruapehu, affecting ski slopes
in 1969 and 1995. An eruption of Taupo volcano can be considered
a long term hazard, its average eruption recurrence interval is
of the order of 2000 years and no activity is known for the last
c. 2000 years.
These
two hazard levels need to be planned for in different ways. Where
short term hazards are clearly present, the volcano should be kept
under intensive surveillance. Areas at risk need to be zoned and
the activities within them restricted. At volcanoes where only long
term hazards are likely, some instrumental monitoring is desirable
and although it is difficult to restrict land uses, in some areas
developmental planning will need to take the potential hazards into
account.
|