Pre-processed climate simulations ensembles for metrics computation

The dataset contains pre-processed multi-annual mean 2D maps of surface temperature, rainfall and energy fluxes computed from ensembles of climate simulations. The ensembles consist in : 1) CMIP5 and CMIP6 multi model ensembles 2) perturbed physics ensembles run with the LMDZ global atmospheric model and IPSL coupled model during an automated tuning procedure. All the dataset are provided on the original model grid, and scripts are provided to interpolate the dataset on a 120x90 global regular grid. The interpolates dataset and scripts were used to compute evaluation metrics for a publication in "Science Advances" in 2023: "Towards machine-assisted tuning avoiding the underestimation of uncertainty in climate change projections" by Frédéric Hourdin, Brady Ferster, Julie Deshayes, Juliette Mignot, Ionela Musat, Daniel Williamson and in the associated Supplementary information.