10 - Extreme events over time
Published online by Cambridge University Press: 18 December 2009
Summary
The physical processes operating during extreme events (or natural hazards when they affect people) are now relatively well understood. There are also a broad array of approaches and methodologies available for unravelling the long-term history of extreme events from natural records. As a corollary, sufficient data has now been collected on the long-term behaviour of these phenomena to allow more realistic assessments of their frequency and magnitude compared to that achievable from short historical records alone. To date, however, few risk assessments, especially those dealing with atmospheric hazards utilise this long-term data.
One of the most interesting aspects of the long-term records is that they often show that short historical records do not accurately reflect the behaviour of a hazard over longer periods of time. Hazard behaviour appears to differ between the two temporal records in terms of variability and stationarity or non-stationarity. Short historical records often do not display the full variability of a hazard and they also often give the impression that the hazard displays stationarity. Of course a short historical record may reflect the true behaviour of a hazard, but a risk assessor can never be sure of this unless comparisons are made with actual records of the longer term behaviour. Many of the long-term records discussed in earlier chapters show, however, that short historical records poorly reflect the true nature of the hazard.
- Type
- Chapter
- Information
- Extreme EventsA Physical Reconstruction and Risk Assessment, pp. 251 - 267Publisher: Cambridge University PressPrint publication year: 2006