Keywords
hydrology, uncertainty, model calibration, model performance measurement
Start Date
1-7-2010 12:00 AM
Abstract
The effective rainfall-flow module of a "top-down" hydrological model haslinear, fixed dynamics, giving a two-exponential instantaneous unit hydrograph (IUH)often interpreted as summing slow- and quick-flow components. Such components sufferuncertainty due to ignorance of rainfall variation between samples and ignorance of delayin the IUH. If they are obtained from linear-in-parameters models, ill conditioning in theconversion amplifies the uncertainty. The extent of these uncertainties is analysed, withexamples. The paper also considers alternatives to performance assessment of such modelsby Nash-Sutcliffe efficiency. Rather than employing the sample mean as an outputpredictionbenchmark, they use almost equally simple predictors taking the flow correlationstructure into account. Again numerical examples are given.
Limitations in Interpretable Top-Down Effective Rainfall-Runoff Modelling
The effective rainfall-flow module of a "top-down" hydrological model haslinear, fixed dynamics, giving a two-exponential instantaneous unit hydrograph (IUH)often interpreted as summing slow- and quick-flow components. Such components sufferuncertainty due to ignorance of rainfall variation between samples and ignorance of delayin the IUH. If they are obtained from linear-in-parameters models, ill conditioning in theconversion amplifies the uncertainty. The extent of these uncertainties is analysed, withexamples. The paper also considers alternatives to performance assessment of such modelsby Nash-Sutcliffe efficiency. Rather than employing the sample mean as an outputpredictionbenchmark, they use almost equally simple predictors taking the flow correlationstructure into account. Again numerical examples are given.
