R-approach to analyze CMIP5 climate simulation data. Is designed as far as a functions set. Package-wrapping should be done in future.
Processing of monthly-aggregated data to assess multidecadal dynamics of the (area-aggregated) data.
- Automated search and check-up of nc-files inside the specified folder
- Accurate temporal interpolation for calculation of the month/seasonal means
- Re-grid to a user-defined grid (to make possible inter-model comparison)
- Prompt visualization of climatological fields
- Prepare CMIP5 data files. CMIP5 simulation data are availavle via data distribution nodes for non-commercial research and educational purposes. We are deeply grateful to the World Data Center for Climate in Hamburg (https://cera-www.dkrz.de) for a granted access to the CMIP5 simulation data with a perfect user interface
- Download the project files
- Set the data directory and the results directory in the local-config "0_ProcessCMIP5_Config_local.R". It is convenient to set that file to --assume-unchanged in git
- Extract the models meta data
- Check that continuous time-series are present for each requires time span
- Form list of spatial matrices for each model by requested time spans
- Calculate ensemble estimations
- Visualize!
We highly acknowledge participating modeling groups of the CMIP5 project and the World Data Center for Climate in Hamburg for making a huge simulation work and granting access to its results.