 |
| Titel |
Analysis of extreme rainfall events using attributes control charts in temporal rainfall processes |
| VerfasserIn |
Maria Villeta, Jose Luis Valencia, Antonio Saá-Requejo, Ana María Tarquis |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250107926
|
| Publikation (Nr.) |
 EGU/EGU2015-7647.pdf |
|
|
|
|
|
| Zusammenfassung |
| The impacts of most intense rainfall events on agriculture and insurance industry can be very
severe. This research focuses in the analysis of extreme rainfall events throughout the use of
attributes control charts, which constitutes a usual tool in Statistical Process Control (SPC)
but unusual in climate studios. Here, series of daily precipitations for the years 1931-2009
within a Spanish region are analyzed, based on a new type of attributes control chart that
takes into account the autocorrelation between the extreme rainfall events. The aim is to
conclude if there exist or not evidence of a change in the extreme rainfall model of the
considered series.
After adjusting seasonally the precipitation series and considering the data of the first 30
years, a frequency-based criterion allowed fixing specification limits in order to discriminate
between extreme observed rainfall days and normal observed rainfall days. The
autocorrelation amongst maximum precipitation is taken into account by a New Binomial
Markov Extended Process obtained for each rainfall series. These modelling of the
extreme rainfall processes provide a way to generate the attributes control charts for
the annual fraction of rainfall extreme days. The extreme rainfall processes along
the rest of the years under study can then be monitored by such attributes control
charts.
The results of the application of this methodology show evidence of change in the model
of extreme rainfall events in some of the analyzed precipitation series. This suggests that the
attributes control charts proposed for the analysis of the most intense precipitation
events will be of practical interest to agriculture and insurance sectors in next future. |
|
|
|
|
|
|