 |
| Titel |
Relationships between rainfall data and water-related damage data for a case study in the Netherlands |
| VerfasserIn |
M. H. Spekkers, J. A. E. ten Veldhuis, M. Kok, F. H. L. R. Clemens |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250063660
|
|
|
|
|
|
| Zusammenfassung |
| Intense rainfall may generate overland flows and pooling in urban areas, causing damage to
buildings, infrastructure and inconvenience to people. Due to a lack of data on the
consequences of flooding, assessment of damages related to pluvial flooding remains a
challenge and so is the identification of explanatory factors for damage variations. Such
methods should take into account the relative short time scales at which urban drainage
processes respond to rainfall and local characteristics, such as the functioning of urban
drainage systems, topography and building properties.
The objective of this study is to establish relationships between rainfall extremes and
damage data from Dutch insurance industry for a case study in the Netherlands, Rotterdam.
For this study, an insurance database containing a series of 20 years of information on
water-related damages to properties and content in the Netherlands has recently been made
available for research by the Dutch Association of Insurers. A data-driven statistical method
is applied to analyse the damage database for the subset of the data related to the case study
Rotterdam. Rotterdam is relatively flat and is therefore typical for cities in lowland areas. The
subset of the database covers around 10000 of accepted damage claims recorded from 1998
until 2009 in Rotterdam. The rainfall data is based on a network of rain gauges (10-minute
temporal resolution) and an 11-year weather radar database (2.4-km horizontal resolution and
5-minute temporal resolution) composed by the Royal Netherlands Meteorological
Institute.
Data mining and statistical computing are carried out within R2.13.2 software
environment. The derived rainfall characteristics, such as peak rainfall intensity, rainfall
volume and duration are stored in a single dataset, as well as descriptive statistics
of the damage claims. The dataset is aggregated by day and postal district. The
strength of the relationships between rainfall characteristics and flood damage are
determined by means of multiple regression analysis. During the presentation the
results of the analysis will be discussed as well as implications for future research. |
|
|
|
|
|
|