Figure: These panels show boxplots of mean squared errors (with respect to SAFRAN reanalysis data) of spatial correlations computed at each grid cell for (a) temperature and (b) precipitation.
Results are presented for IPSL, "low resolution" SAFRAN and the results from MBC-CycleGAN
(in green) and 3 other bias correction methods. The closer the boxplots are to zero, the more the spatial properties are in line with those from SAFRAN. See  for further details.
Climate simulations have spatial biases, i.e., spatial properties (e.g., rainfall variability between different sites) can differ from observations. Consequently, corrections of spatial properties are needed, e.g. for impact studies . As proof-of-concept, we proposed to adapt a computer vision technique used for Image-to-Image translation tasks (named CycleGAN ) for the adjustment of spatial properties of climate simulations .
The proposed algorithm, named MBC-CycleGAN, consists in learning to transfer simulated maps (seen as images) from climate simulations with inappropriate spatial properties to more realistic images with spatial properties similar to the observed ones. We tested MBC-CycleGAN by performing two bias correction exercises, both with respect to SAFRAN references  for temperature and precipitation data over the region of Paris:
We compared the results for the correction of the spatial properties with other state-of-the-art bias correction methods (Quantile-quantile , R2D2  and dOTC ). MBC-CycleGAN produced reasonable adjustments of spatial correlations for temperature and precipitation compared to the other alternatives. In particular, better results are obtained for the correction of paired images ("low-resolution" SAFRAN).
These results suggest that CycleGAN is very promising for multivariate bias correction. Of course, further research is needed, such as extending our method to correct not only spatial properties but also inter-variable and/or temporal ones.
 Maraun, D. et al. Precipitation downscaling under climate change: recent developments to bridge the gap between dynamical models and the end user. Rev Geophys (2010). https://doi.org/10.1029/2009RG000314
 Zhu, J.Y., Park, T., Isola, P., Efros, A.A. Unpaired image-to-image translation using cycle-consistent adversarial networks. (2017) arXiv: 1703.10593
 François, B., Thao, S. & Vrac, M. Adjusting spatial dependence of climate model outputs with cycle-consistent adversarial networks. Clim Dyn (2021). https://doi.org/10.1007/s00382-021-05869-8
 Vidal, J. P., Martin, E., Franchistéguy, L., Baillon, M., Soubeyroux, J.M. A 50-year high-resolution atmospheric reanalysis over France with the Safran system. Int J Climatol (2010). https://
 Déqué, M. Frequency of precipitation and temperature extremes over France in an anthropogenic scenario: model results and statistical correction according to observed values. Glob Planet Change (2007). https://doi.org/10.1016/j.gloplacha.2006.11.030
 Vrac, M., Thao, S. R2D2 v2.0: accounting for temporal dependences in multivariate bias correction via analogue ranks resampling. Geosci Model Dev (2020). https://doi.org/10.5194/gmd-2020-132
 Robin, Y., Vrac, M., Naveau, P., Yiou, P. Multivariate stochastic bias corrections with optimal transport. Hydrol Earth Syst Sci (2019). https://doi.org/10.5194/hess-23-773-2019