Page 196 - Fundamentals of Physical Volcanology

9780632054435_4_011.qxd 12/10/2007 12:33PM Page 173

VOLCANIC HAZARDS AND VOLCANO MONITORING 173

Table 11.1 Data on eruption frequency and probability for 92 eruptions of Mount Vesuvius.

Volcanic Explosivity Eruption type Number of eruptions Probability value
Index value
1 Lava flow 21 0.228
1.5 Strombolian crisis 15 0.391
2 Strong Strombolian activity 17 0.576
2.5 Strong Strombolian 10 0.685
activity + lava fountains
3 Small eruption column 15 0.848
4 Subplinian eruption 5 0.902
5 Plinian eruption 8 0.989
6 Ultra-Plinian eruption 1 1.000

Taken from section 7.4.1 in Dobran, F. (2001) Volcanic Processes – Mechanisms in Material Transport. Kluwer Academic/
Plenum Publishers, New York, 590 pp. With kind permission of Springer Science and Business Media.

1 P(intensity ≥ VEI) = 1 − exp[−6.634 ∆t
exp(−1.18 VEI)] (11.1)

0.8 where ∆t is the time interval between eruptions in
years. Some examples of the implications of this
are that, in any 10 year period, the probability
0.6
Probability of a Plinian eruption with VEI = 5 occurring is
only 0.16, or about 1 in 6, whereas the chance of
a Strombolian eruption with VEI = 2.5 is close to
0.4
0.97, i.e., extremely likely.
0.2
11.4 Monitoring volcanoes and short-term
eruption prediction
0
0 1 2 3 4 5 6
VEI Volcanoes that erupt frequently are commonly mon-
itored in various ways for two reasons: to learn
Fig. 11.11 Probability of a given Volcanic Explosivity
more about the internal structure of the volcano
Index (VEI). (Based on fig. 7.19 in Volcanic Processes:
and to look for signs of activity. The two processes
Mechanisms in Material Transport, by F. Dobran, Kluwer
are linked, because the more we understand about
Academic/Plenum Publishers, New York, 2001, 590 pp.
With kind permission of Springer Science and Business a particular volcano, the more sense we are likely
Media.) to make of indicators of impending eruptions.
Traditionally, monitoring volcanoes has implied
from Dobran (2001) treating a data base of 92 making measurements at ground level on the vol-
eruptions at Mount Vesuvius. The probability value, cano. Unfortunately, all continuous monitoring exer-
P, for each VEI in Table 11.1 is the cumulative frac- cises involving instrumentation on a volcano are
tion of eruptions for which the intensity is less than expensive and vulnerable. Usually at least several
or equal to that VEI. For example, the P value for instruments have to be left in place on the volcano
VEI ≤ 2.5 is (21 + 15 + 17 + 10)/92 = 0.685. Plotting and the measurements are either recorded on site
the P values against their corresponding VEI values, in some simple computer-controlled data storage
a graph such as Fig. 11.11 is found, which can be device or returned to a distant base via a radio or
fitted by the double exponential function satellite link. All parts of the system have to be