Introduction - Extreme Events: A Physical Reconstruction and Risk Assessment - page 14

Imagine a community of intelligent bugs that live on the sides of a smooth
sided dish shaped pond. They are a complex community with many specialist
occupations. They have been living in the pond for about 100 years. The principal
natural hazard facing their society is variations in water levels in the pond. The
engineer bugs monitor the water levels with precision instruments and have
done so for nearly 100 years on a daily basis. They have a lot of data. The scientist
bugs are interested in the deposition of sediments around the sides of the pond
and they recognise that the position of the sediments relates to the level of water
in the pond at the time of deposition. The scientist bugs are also interested in
the chemical composition of the sediments and they use this information to
monitor pollution levels emitted by the bug community. The sediments around
thesides of the pond can be likened to a dirt ring around a bath tub. Few would
question that the height of the dirt ring is a reflection of the water level at the
time of sediment deposition, but the dirt ring does not record those water levels
as accurately as an instrument.
The scientist bugs observe another sequence of sediments at a much higher
elevation on the sides of the pond. They conclude that these sediments were
deposited by considerably higher water levels. The engineer bugs agree and tell
the scientist bugs that their model shows that this must have been the 1 in
500 year (0.2% AEP) event. The scientist bugs collect samples from the higher
level sediments for radiocarbon analyses which return ages of around 150 years.
The engineer bugs say that the sediments still represent the 1 in 500 year event
and that it is an outlier. The scientist bugs wish to test this hypothesis so they
examine a range of higher level sediments around the sides of numerous ponds
throughout the region. In the end they manage to date events spanning the last
1000 years. The average return interval for the higher level events is 90 years.
They present their data to the engineer bugs who dismiss it as an inaccurate
measure of past water levels. They still say that the higher water levels represent
the 500 year or even less frequent events.
Adilemma exists. Both groups see the behaviour of this hazard from dif-
ferent perspectives. The engineer bugs regard accuracy of measurements as a
guide to the veracity of their hypothesis. The scientist bugs wish to test this
hypothesis or model outcome and suggest that it is not necessary to have data
as precise as that derived from instruments to do this. They say that despite the
fact that they can only determine the water level to within 10--20% accuracy, the
presence of the higher level sediments together with the radiocarbon ages with
their uncertainty margin of a decade, demonstrate that the initial hypothesis is
likely flawed and needs to be reviewed. They can show that these events around
theregion definitely do not have a return interval of 500 years. They argue that
thesediments represent real events that actually occurred and the model only