 |
| Titel |
A framework for global river flood risk assessment |
| VerfasserIn |
H. C. Winsemius, L. P. H. Van Beek, A. Bouwman, P. J. Ward, B. Jongman |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250064119
|
|
|
|
|
|
| Zusammenfassung |
| There is an increasing need for strategic global assessments of flood risks. Such
assessments may be required by: (a) International Financing Institutes and Disaster
Management Agencies to evaluate where, when, and which investments in flood
risk mitigation are most required; (b) (re-)insurers, who need to determine their
required coverage capital; and (c) large companies to account for risks of regional
investments.
In this contribution, we propose a framework for global river flood risk assessment. The
framework combines coarse scale resolution hazard probability distributions, derived from
global hydrological model runs (typical scale about 0.5 degree resolution) with
high resolution estimates of exposure indicators. The high resolution is required
because floods typically occur at a much smaller scale than the typical resolution
of global hydrological models, and exposure indicators such as population, land
use and economic value generally are strongly variable in space and time. The
framework therefore estimates hazard at a high resolution (Â 1 km2) by using a) global
forcing data sets of the current (or in scenario mode, future) climate; b) a global
hydrological model; c) a global flood routing model, and d) importantly, a flood
spatial downscaling routine. This results in probability distributions of annual flood
extremes as an indicator of flood hazard, at the appropriate resolution. A second
component of the framework combines the hazard probability distribution with
classical flood impact models (e.g. damage, affected GDP, affected population) to
establish indicators for flood risk. The framework can be applied with a large number
of datasets and models and sensitivities of such choices can be evaluated by the
user.
The framework is applied using the global hydrological model PCR-GLOBWB,
combined with a global flood routing model. Downscaling of the hazard probability
distributions to 1 km2 resolution is performed with a new downscaling algorithm, applied on
a number of target regions. We demonstrate the use of impact models in these regions based
on global GDP, population, and land use maps. In this application, we show sensitivities of
the estimated risks with regard to the use of different climate input datasets, decisions made
in the downscaling algorithm, and different approaches to establish distributed estimates of
GDP and asset exposure to flooding. |
|
|
|
|
|
|