Submitted as: model description paper
12 Nov 2021
Submitted as: model description paper | 12 Nov 2021
Status: this preprint is currently under review for the journal GMD.

A Bayesian data assimilation framework for lake 3D hydrodynamic models with a physics-preserving particle filtering method using SPUX-MITgcm v1

Artur Safin1, Damien Bouffard1, Firat Ozdemir2, Cintia L. Ramón3,1, James Runnalls1, Fotis Georgatos2, Camille Minaudo4, and Jonas Šukys1 Artur Safin et al.
  • 1Eawag: Swiss Federal Institute for Aquatic Science and Technology, Switzerland
  • 2Swiss Data Science Center, Switzerland
  • 3Water Research Institute and Department of Civil Engineering, University of Granada, Spain
  • 4École Polytechnique Fédérale de Lausanne, Switzerland

Abstract. We present a Bayesian inference for a three-dimensional hydrodynamic model of Lake Geneva with stochastic weather forcing and high-frequency observational datasets. This is achieved by coupling a Bayesian inference package, SPUX, with a hydrodynamics package, MITgcm, into a single framework, SPUX-MITgcm. To mitigate uncertainty in the atmospheric forcing, we use a smoothed particle Markov chain Monte Carlo method, where the intermediate model state posteriors are resampled in accordance with their respective observational likelihoods. To improve the assimilation of remotely sensed temperature, we develop a bi-directional Long Short-Term Memory (Bi-LSTM) neural network to estimate lake skin temperature from a history of hydrodynamic bulk temperature predictions and atmospheric data. This study analyzes the benefit and costs of such state of the art computationally expensive calibration and assimilation method for lakes.

Artur Safin et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2021-305', Anonymous Referee #1, 22 Feb 2022
  • RC2: 'Comment on gmd-2021-305', Anonymous Referee #2, 24 Mar 2022

Artur Safin et al.

Data sets

Hydrodynamic model predictions (2019) Artur Safin, James Runnalls

Model code and software

MITgcm compilation and running scripts Artur Safin, Jonas Sukys

MITgcm source code modified to run with SPUX MITgcm developers (Jean-Michel Campin, Patrick Heimbach, Martin Losch, Gael Forget, Ed Hill, etc), Artur Safin

SPUX data assimilation package source code Artur Safin, Jonas Sukys, Firat Ozdemir, Fotis Georgatos

Artur Safin et al.


Total article views: 703 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
534 152 17 703 30 6 6
  • HTML: 534
  • PDF: 152
  • XML: 17
  • Total: 703
  • Supplement: 30
  • BibTeX: 6
  • EndNote: 6
Views and downloads (calculated since 12 Nov 2021)
Cumulative views and downloads (calculated since 12 Nov 2021)

Viewed (geographical distribution)

Total article views: 702 (including HTML, PDF, and XML) Thereof 702 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 28 May 2022
Short summary
Reconciling the differences between numerical model predictions and observational data is always a challenge. In this paper, we investigate the viability of a novel approach to the calibration of a three-dimensional hydrodynamic model of Lake Geneva, where the target parameters are inferred in terms of distributions. We employ a filtering technique that generates physically consistent model trajectories, and implement a neural network to enable bulk-to-skin temperature conversion.