An automated calibration framework and open source tools for 3D lake hydrodynamic models. (December 2020)

• Record Type: Journal Article
• Authors: Baracchini, Theo; Hummel, Stef; Verlaan, Martin; Cimatoribus, Andrea; Wüest, Alfred; Bouffard, Damien
• Is Part Of: Environmental modelling & software. Volume 134(2020)
• Publication Date: 2020-12
• Subjects: Auto-calibration -- OpenDA -- Delft3D-FLOW -- Parameter estimation -- Observational uncertainty -- Model performance evaluation; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2020.104787
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Understanding lake dynamics is crucial to provide scientifically credible information for ecosystem management. In this context, three-dimensional hydrodynamic models are a key information source to assess critical but often subtle changes in lake dynamics occurring at all spatio-temporal scales. However, those models require time-consuming calibrations, often carried out by trial-and-error. Through a new coupling of open source software, we present here a flexible and computationally inexpensive automated calibration framework. The method, tailored to the calibration data available to the user, aims at (i) reducing the time spent on calibration, and (ii) making three-dimensional lake modelling accessible to a broader range of users. It is demonstrated for two different lakes (Lake Geneva and Greifensee) with an extensive multi-variable observational dataset. Models mean absolute errors are reduced by up to ~50% over the baseline. Guidelines on heat and momentum transfer parameters are given with their dependence on the observational setup. Highlights: 3D hydrodynamic models require time-consuming calibrations. A new method reduces the time spent on calibration and need for expert knowledge. Mean absolute errors are reduced by up to 50% with limited computational demand.