Data to: Ensuring Spatiotemporal Consistency in Multivariate Bias Correction for Climate Projections using Nested Vine Copulas

We provide raw and processed data as well as the evaluation results for the case study presented in the paper “Ensuring Spatiotemporal Consistency in Multivariate Bias Correction for Climate Projections using Nested Vine Copulas.” The dataset covers 22 grid points in the canton of Vaud, Switzerland, over the time period 1980-2022. The variables included are temperature, precipitation, relative humidity, wind speed and surface pressure. The files are structured as follows: nc_files.zip: Contains raw NetCDF files for ERA5-Land data (both on initial spatial resolution and remapped to CORDEX resolution) and gridded data products from MeteoSwiss, as well as interim pre-processed data for all CORDEX, ERA5-Land and MeteoSwiss datasets. Due to their size, the raw CORDEX simulations are not included here but can be downloaded from the Climate Data Store. The required API requests are provided in the supplementary material of the paper. processed.zip: Contains all the processed data that is actually used to run all the algorithms compared in the application. bc_data comprises all the bootstrap samples and original datasets, while corrected provide the bias-corrected data for all algorithms and bootstrap and original samples. The runtimes of each algorithm are also provided. results.zip: Contains evaluation metrics quantifying inter-variable, spatial, and temporal consistency, corresponding to the results presented in the manuscript figures and tables.