 |
| Titel |
Evaluation of a geomorphology-based conceptual IUH in a mountain watershed |
| VerfasserIn |
J. J. López, F. N. Gimena, J. V. Giráldez, J. L. Ayuso, M. Goñi |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250026862
|
|
|
|
|
|
| Zusammenfassung |
| Hydrograph generation at a point in the drainage network, as a watershed response to a
rainfall event, is a complex process that depends on watershed and storm characteristics.
Among the available methods for hydrologic design, the unit hydrograph (UH) is one of those
most widely used. It is a conceptual model which assumes the linear systems theory and
incorporates the rainfall characteristics in the simulation process.
There is, obviously, a close relationship between the geomorphologic characteristics of
a watershed and its hydrologic response. During the past years hydrologists and
geomorphologists have worked together in a joint effort to characterize the global
average watershed response as a function of its geomorphologic properties. Since
Rodríguez-Iturbe and Valdés presented the Geomorphological Instantaneous Unit
Hydrograph (GIUH) there have been many attempts to propose an Instantaneous Unit
Hydrograph (IUH) that incorporates the geomorphological properties of the watershed.
Also, linear reservoir models were, and are still, very frequently used for simulating
rainfall-runoff processes and, more precisely, for determining the unit hydrograph of a
watershed.
The fundamental aim of this work is to present and evaluate an Instantaneous Unit
Hydrograph based on a cascade of linear Reservoirs obtained by taking the Geomorphology
of the watershed (IUHGR). The geomorphological association of reservoirs in this IUH is
characterized by means of the sub-basins into which the watershed is divided starting from
the drainage network. The formulation of this IUH, obtained from the one set up by López et
al.,(2005), has been carried out both for the case of the spatial uniformity of the sub-basins
and for that of considering their spatial variability. Also, this model’s evaluation was made
based on a detailed analysis of it and on a comparative study. For the latter, some IUHs
with some similar characteristics to the one proposed and which were sufficiently
vouched for in the literature were selected. These were: the Nash IUH, a linear
reservoir cascade model; the Chutha and Dooge IUH, an interpretation of the GIUH of
Rodríguez-Iturbe and Valdés describing the water pathways as cascades of unequal
linear reservoirs; and the Troutman and Karlinger IUH, with a geomorphologic
foundation, based on the assumption of linear flow through topologically random channel
networks.
The experimental watershed of 23.42 km2, is located in a forested area of the Ebro River
basin, at the southwestern end of the Pyrenees, in the northeastern part of the Navarre
province, (Spain). Several rainfall-runoff events have been selected for the analysis. The
effective rain hyetograph was estimated using the curve number method, adjusting the
value of the curve number and the initial abstraction coefficient for the different
events.
The validation of the model was performed by means of the model’s evaluation, in which
the factors influencing its formulation are analyzed, and a comparative study made with other
IUHs with similar characteristics to the one proposed and which are sufficiently vouched for
in the literature.
The model’s evaluation is performed, firstly, with a sensitivity analysis of the only
uncertain parameter on which the model depends, k, and, secondly, with the analysis of the
possible effect of the degree of division of the watershed into subwatersheds or order,
n.
The parameters of the different IUHs were obtained by fitting the computed hydrographs
resulting from the convolution with the respective formulations: IUHGR(u), IUHGR(v), Nash
(NIUH), Chutha and Dooge (CDIUH), and Troutman and Karlinger (TKIUH), and the
computed effective rain hyetographs, to the recorded hydrographs.
The fitting was carried out using the Rosenbrock algorithm, with the difference between
the unity and the Nash-Sutcliffe index, as the objective function. The goodness of the fit for
the validation was evaluated again with the Nash-Sutcliffe efficiency index, Ei,
equation.
The results indicate its satisfactory performance, with good fits to the observed
hydrographs and a lower number of identifiable parameters. |
|
|
|
|
|
|