Premium
A multisignal automatic calibration methodology for hydrochemical models: A case study of the Birkenes Model
Author(s) -
Grosbois Ed,
Hooper Richard P.,
Christophersen Nils
Publication year - 1988
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/wr024i008p01299
Subject(s) - hydrograph , calibration , identifiability , least squares function approximation , range (aeronautics) , signal (programming language) , tracer , estimation theory , computer science , algorithm , mathematics , statistics , engineering , machine learning , geography , physics , archaeology , estimator , aerospace engineering , nuclear physics , programming language , flood myth
Most hydrochemical models employ immeasurable or “artificial” parameters that need to be estimated from calibration data. A methodology is presented here for the automatic calibration of artificial parameters, which improves parameter identifiability through increasing the information available for determining the parameter values. Instead of calibrating the model using only one signal (e.g., the hydrograph), multiple signals (e.g., chemical signals as well as the hydrograph) are considered simultaneously. Either a simple least squares or a weighted least squares objective function may be used. The methodology is applied to the hydrologic module of the Birkenes hydrochemical model using artificial data. From a wide range of starting points, a gradient search optimization technique is able to consistently locate the correct parameter values uniformly better when using two signals (the hydrograph and a conservative tracer) than when only one signal (the hydrograph) is used.