 |
| Titel |
Comparing bottom-up and top-down parameterisations of a process-based runoff generation model tailored on floods |
| VerfasserIn |
Manuel Antonetti, Simon Scherrer, Michael Margreth, Massimiliano Zappa |
| Konferenz |
EGU General Assembly 2016
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 18 (2016) |
| Datensatznummer |
250123740
|
| Publikation (Nr.) |
 EGU/EGU2016-3044.pdf |
|
|
|
|
|
| Zusammenfassung |
| Information about the spatial distribution of dominant runoff processes (DRPs) can improve
flood predictions on ungauged basins, where conceptual rainfall-runoff models usually appear
to be limited due to the need for calibration. For example, hydrological classifications based
on DRPs can be used as regionalisation tools assuming that, once a model structure and its
parameters have been identified for each DRP, they can be transferred to other areas where
the same DRP occurs.
Here we present a process-based runoff generation model as an event-based spin-off of the
conceptual hydrological model PREVAH. The model is grid-based and consists
of a specific storage system for each DRP. To unbind the parameter values from
catchment-related characteristics, the runoff concentration and the flood routing
are uncoupled from the runoff generation routine and simulated separately. For
the model parameterisation, two contrasting approaches are applied. First, in a
bottom-up approach, the parameters of the runoff generation routine are determined a
priori based on the results of sprinkling experiments on 60-100 m2 hillslope plots at
several grassland locations in Switzerland. The model is, then, applied on a small
catchment (0.5 km2) on the Swiss Plateau, and the parameters linked to the runoff
concentration are calibrated on a single heavy rainfall-runoff event. The whole system
is finally verified on several nearby catchments of larger sizes (up to 430 km2)
affected by different heavy rainfall events. In a second attempt, following a top-down
approach, all the parameters are calibrated on the largest catchment under investigation
and successively verified on three sub-catchments. Simulation results from both
parameterisation techniques are finally compared with results obtained with the traditional
PREVAH. |
|
|
|
|
|
|