Basic Earth Science and Society - Basic Research Opportunities in Earth Science - page 12

Applied research on different types of natural hazards has a common set of
practical objectives:
• education of the general populace about the threat of natural hazards and
the steps that can be taken to reduce risks (public preparation),
• quantification of what will happen during a particular event scenario
(deterministic hazard analysis) or over an ensemble of possible events
(probabilistic hazard analysis)—both are used by engineers to reduce
human casualties (safety engineering) and economic losses
(performance-based engineering).
• specification of individual disasters in terms of where, when, and how
large (event prediction)—in some cases, such as earthquakes and
tsunamis, an accurate prediction of occurrence times may not be possible,
in which case the objective of “when” may be weakened to “how
frequently” (long-term forecasting), and
• accurate and timely information about the occurrence and circumstances
of disastrous events for public notification (including early warning, when
possible) and for use by government officials and emergency
management personnel (rapid response).
Experience shows that these practical objectives are difficult to attain
without precise observations and good understanding of the phenomena involved
in natural hazards. Because the processes of magma motion through the upper
crust can be detected and analyzed, the most dangerous volcanic eruptions can be
predicted in a useful way. The 1991 eruption of Mt Pinatubo in the Philippines
was forecast by the U.S. Geological Survey (USGS) early enough to allow the
region to be evacuated and aircraft and other equipment to be removed from
Clark Air Force Base. The forecasts are estimated to have saved 5000 to 20,000
lives and economic losses of at least $200 million. 11 On the other hand, the most
dangerous earthquakes cannot be predicted deterministically, because no reliable,
precursory indicators of their timing, location, and magnitude have been
discovered. Is this because such precursors do not exist or because the fault-
rupture process is too poorly understood to know which kinds of behavior are
most diagnostic of a large, impending event? These questions remain the subject
of vigorous basic research, supported by EAR as part of NSF's participation in
the National Earthquake Hazard Reduction Program.
11 USGS Fact Sheet 115-97.