Skip to content

Latest commit

 

History

History
 
 

datasets-interactive-atlas

README.md

The Interactive Atlas Dataset

The Interactive Atlas Dataset comprises gridded information from a set of 25 variables and impact-relevant indices used in the Interactive Atlas, computed from CMIP5/6 and CORDEX historical and scenario simulations (see data-sources), with monthly temporal resolution and interpolated to common reference grids (including postprocessing –e.g. bias adjustment– when needed).

The resulting dataset underpins the Atlas chapter and the Interactive Atlas products. This final dataset is preserved in the long term and is openly accessible from the IPCC Data Distribution Center (IPCC-DDC catalog entry) in the DIGITAL.CSIC repository. A replica of the data is accessible through the Climate Data Store (CDS) of the CopernicusClimate Change Service (C3S), allowing interactive downloads using the CDS web interface or programmatic access via the CDS API service: “Gridded monthly climate projection dataset underpinning the IPCC AR6 Interactive Atlas”.

The metadata and scripts for reproducibily are included in this repository.

A subset with information spatially aggregated over the IPCC reference regions is avaialable in the section datasets-aggregated-regionally of this repository and is used to reproduce some of the figures of the Atlas chapter (see notebooks section).

Code for reproducibility of the Interactive Atlas Dataset

This repository contains the code needed to generate the intermediate dataset (Interactive Atlas Dataset) used to generate the different products displayed in the Interactive Atlas. The code takes as input the CMIP5/6 and CORDEX information described in the data sources section. For simplicity, the full workflow is split up in four main scripts which undertake independent tasks. In a nutshell:

  • 01_index_calculation.R computes a number of climate indices (see the table included in the own script) from CMIP5 (and/or CMIP6) model outputs. Due to the particularities of some of the indices available, two more companion scripts are also delivered: 01_index_calculation_SPI.R and 01_index_calculation_bias_corrected.R. Whereas the former is thought for SPI-6, the latter is used for those indices which are based on absolute thresholds (Frost Days, TX35 and TX40).
  • 02_interpolation.R interpolates the outputs (NetCDF files) produced by 01_index_calculation.R using Conservative Remapping (the approach used in EURO-CORDEX for CORDEX data). The folder called bash-interpolation-scripts contains some bash scripts which are required by 02_interpolation.R.
  • 03_ensemble_building.R creates a multi-model ensemble building on the interpolated information generated by 02_interpolation.R, returning NcML files.
  • 04_map_figures.R generates the final figures (maps) from the NcMLs returned by 03_ensemble_building.R.

Al these scripts build on the climate4R framework (Iturbide et al. 2019).

Script list