Datasets for mloz: A Highly Efficient Machine Learning-Based Ozone Parameterization for Climate Sensitivity Simulations

Datasets for reproducing JAMES paper: mloz: A Highly Efficient Machine Learning-Based Ozone Parameterization for Climate Sensitivity Simulations NetCDF files started with fullchem*: A coarse-grained (60°×10°) version of the training datasets (from UKESM full chemistry simulations), downsampled due to storage constraints. NetCDF files started with coef* : full-resolution mloz coefficients NetCDF files started with Scaler* : full-resolution scalings for temperature or ozone used for normalization. NetCDF files started with mloz*: Coarse-grained (60°×10°) online ozone prediction from mloz in UKESM or ICON.