Threshold structures in conceptual rainfall-runoff models : potential problems with calibration
Rain and rainfall -- Mathematical models.
Runoff -- Mathematical models.
Hydrology -- Mathematical models.
Committee ChairSorooshian, S. S.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractOne of the most frequently reported problems in the automatic calibration of conceptual rainfall-runoff (CRR) models is the inability to identify a unique and conceptually realistic parameter set for a model in use on a particular watershed. Sorooshian and Gupta [1983,1985] and Gupta and Sorooshian  have shown that model structure can be largely responsible for this problem. This thesis extends this work by further investigating calibration problems resulting from model structure, particularly when the CRR model contains one or more threshold elements. Two aspects of this problem were addressed. In the first, the practice of modeling nonlinear processes with linear models is analyzed. In the second, reparameterization techniques to improve parameter identifiability (Gupta and Sorooshian ; Sorooshian and Gupta ) are demonstrated on a four-parameter simplified CRR model. The results of these studies show that the parameter identification procedure can in fact be hindered by model structure and that the interdependence of model structure and parameter estimation methodologies must be considered by CRR model users and developers if automatic calibration is to be successful.
Degree ProgramHydrology and Water Resources