loading page

Process-based climate model development harnessing machine learning: II. model calibration from single column to global
  • +7
  • Frédéric Hourdin,
  • Danny Williamson,
  • Catherine Rio,
  • Fleur Couvreux,
  • Romain Roehrig,
  • Najda Villefranque,
  • Ionela Musat,
  • Fatoumata Bint Diallo,
  • Laurent Fairhead,
  • Victoria Volodina
Frédéric Hourdin

Corresponding Author:hourdin@lmd.jussieu.fr

Author Profile
Danny Williamson
University of Exeter
Author Profile
Catherine Rio
Centre national des recherches météorologiques (CNRM), Université de Toulouse, Météo-France, CNRS
Author Profile
Fleur Couvreux
Université Toulouse, CNRM, Meteo-France, CNRS
Author Profile
Romain Roehrig
CNRM, Université de Toulouse, Météo-France, CNRS
Author Profile
Najda Villefranque
Centre National de Recherches Météorologiques
Author Profile
Ionela Musat
Author Profile
Fatoumata Bint Diallo
Laboratoire de Météorologie Dynamique
Author Profile
Laurent Fairhead
Author Profile
Victoria Volodina
The Alan Turing Institute
Author Profile


We demonstrate a new approach for climate model tuning in a realistic situation. Our approach, described in detail in Part I, systematically uses a single-column configuration of a global atmospheric model on a series of test cases for which reference large-eddy-simulations are available. The space of free parameters is sampled running the single-column model from which metrics are estimated in the full parameter space using emulators. The parameter space is then reduced by retaining only the values that are consistent with the metrics computed on large eddy simulations within a given tolerance to error. The approach is applied to the recently designed 6A version of the LMDZ model, itself the result of a long investment in the development of physics parameterizations and by-hand tuning. The boundary layer is revisited by increasing the vertical resolution and varying parameters that were kept fixed so far. The approach allows us to automatically reach a tuning as good as that of the 6A version, after some improvements are done at process scale. This approach helps accelerate the introduction of new parameterizations, by avoiding a tedious manual tuning process and preventing some of the error compensations that could occur if calibration was carried out directly with the full atmospheric model. This way of using machine learning techniques allows us to maintain the physical foundations of the model and to ensure that the improvement of global metrics is obtained for a reasonable behavior at process level. That is, we get things right for the right reasons.
Jun 2021Published in Journal of Advances in Modeling Earth Systems volume 13 issue 6. 10.1029/2020MS002225