![Hier klicken, um den Treffer aus der Auswahl zu entfernen](images/unchecked.gif) |
Titel |
Quantifying and attributing uncertainty in flood damage estimates |
VerfasserIn |
H. de Moel, J. C. J. H. Aerts |
Konferenz |
EGU General Assembly 2009
|
Medientyp |
Artikel
|
Sprache |
Englisch
|
Digitales Dokument |
PDF |
Erschienen |
In: GRA - Volume 11 (2009) |
Datensatznummer |
250025762
|
|
|
|
Zusammenfassung |
It is clear that the flood risk of river floods is growing because of continuous developments in
flood-prone areas, anthropogenic interference in the drainage system and other changing
environmental conditions. Economic developments and climate change are likely to
continue to increase flood risks in the future. A transition from protective flood
management to a more integrated risk management approach can be observed in many
European countries, in order to cope with these projected changes and its associated
uncertainties
Flood risk assessments are of key importance in flood risk management in order to make
well considered decisions. Direct monetary damage that could potentially result from a flood
is a key figure in flood risk assessments and can be used in cost-benefit analyses to facilitate
decision making regarding risk-reducing measures. However, many uncertainties surround
such flood damage estimates as uncertainties in the data underlying flood damage
calculations propagate through the calculation and accumulate in the final damage estimate.
Uncertainty exists in all components of flood damage models because of generalizations or
aggregation of information. This is acknowledged by various studies but mainly
uncertainty from a single source (e.g. water depth) is analyzed. A comprehensive study
into how uncertainty from different sources compare to one another is currently
lacking.
This research addresses this caveat and makes an effort to assess the variation
in final flood damage estimates resulting from assumptions and uncertainties in
land use data, water depth data, and stage-damage curves (and their associated
maximum damage). The input data for the flood damage assessment from these three
sources was varied in order to assess the sensitivity of the damage estimate for
uncertainty in each component. The data was varied by taking different sources
of land use information, introducing water depth errors manually, and by using
different sets of stage-damage curves. All possible combinations of input data were
used to calculate flood damages in order to present the full range of possible flood
damage estimates. The approach is applied to a case study area in the south of the
Netherlands on the south bank of the Meuse river: Land van Heusden/Maaskant (dikering
36).
It is found that variation in land use data has comparatively the smallest effect (~factor
1.2) on the resulting flood damage estimate, having a similar effect of a water depth error of
about 25 cm. The flood damage estimate is most sensitive to the shape of the stage-damage
curve and the maximum damage assumed for the different land use classes. Both can easily
result in a factor 2 change in flood damage estimate, similar to an error in water
depth of around 1.1 m. In total, absolute damage estimates can easily vary up to a
factor 5 or 6 as a result from different assumptions and errors in the underlying
data. Relative changes in flood damage (as a percentage of a reference situation),
on the other hand, are surrounded by a much lower uncertainty, around a factor
1.3. Given these results, it is clear that choosing stage-damage functions and their
respective maximum damages should receive much attention and that the use of absolute
numbers of (avoided) damage, in for instance cost-benefit analyses, can be misleading. |
|
|
|
|
|