Geography Reference
In-Depth Information
needed by urban places. For the sake of completeness some brief digressions are
made into the hazards of the last two categories, namely biological issues and from
human activity—although not personal behaviours, wars or terrorist attacks—since
they may produce similar disastrous effects to life and property. However it is worth
noting that many of these potential events inflicting disasters upon cities can be
reduced by many of the same procedures being adopted to combat natural hazards.
9.2.2
Natural Hazard Trends
Despite extensive reporting of the effect of natural hazards there is still a dearth of
reliable, global information about the exact numbers and losses, in deaths, disrup-
tion and property, even from recent events (Bouwer et al.
2007
). Hence it is not sur-
prising to find that it has been even more difficult to estimate impacts from historical
environmental hazards, although Smith and Petley (
2009
, p. 23) estimated that there
have been 35 major disasters with over 100 thousand killed since 1,000 BCE, with
seven events having over a million dead. In the last half century far more sources of
data have become available, often compilations by national governments, such as
The Canadian National Data Base (CNDB), but between government sources there
are many variations in accuracy and in types of compilation. The two most reliable
world sources at the present are the EMDAT (Emergency Events Data Base) records
complied in the University of Louvain's Centre for Research on the Epidemiology
of Disasters, both natural and human (CRED), and the annual estimates compiled
by Munich Re, a major insurance company. For inclusion in CRED a disaster must
have killed more than ten persons, and affected a hundred people, with a thresh-
old of 2000 for drought and famine cases, although those with government ap-
peals for national and international assistance are also included. Each source has
its own bias, with those stressing fatalities being more typical of reports from the
less developed world where more poor people die, whereas companies that focus
on insurance losses have greater reliability in the developed world, as well as much
better statistics on economic losses in the past 30 years. Although these different
sources show variations in the actual numbers, there is general agreement that there
has been a trend towards both increasing numbers and impacts of these hazards in
recent years. For example Fig.
9.1
shows the changes by type of hazard since 1980.
The rising general trend line is unmistakeable, although the variations by year must
be stressed. It also shows the estimated proportion of insured losses compared to
total losses at 5 year intervals, which reveals that in 2011 only 28 % of losses were
covered by insurance, a proportion often lower in previous years, since few of the
losses in developing countries are covered by insurance.
Figure
9.1
also shows that the increase is primarily due to meteorological and hy-
drological events, with geophysical causes much smaller and relatively stable, even
though these often have huge impacts and attract the most attention. The bigger
problem is trying to explain the trends. Jennings (
2011
) has argued that it is unlikely
that better reporting accounts for the increase, or that population increases alone is
a sufficient cause, but cautiously states that there is insufficient evidence to exclude
