Skip to content

Suite of Jupyter Notebooks for calculation of regional and global metrics for evaluation of CMIP6 ESMs

Notifications You must be signed in to change notification settings

NCAR/ESM_regional_evaluation

Repository files navigation

ESM_regional_evaluation

Suite of Jupyter Notebooks for calculation of regional and global metrics for evaluation of CMIP6 ESMs

concat_cmip6, standardize_PNW_obs, and standardize_global_obs must be run on preexisting observational and CMIP6 data in order to create the appropriately formatted filed expected by all the other notebooks

cmip6_global_metrics and cmip6_regional_metrics compute all the metrics used in this evaluation over their respective regions. cmip6_regional_metrics can be easily adapted to any lat/lon box over the globe. Future functionality will add the ability to read shapefiles for region definition.

Once these metrics are computed, cmip6_global_regional_metrics can then be run to compute the Relative Error Scores and various other things like the effect on the future projections.

cmip6_metrics-splitSample isolates 1901-1950 as the evaluation period and verifies against 1951-2014

cmip6_perfect_model_eval evaluates all models against each "perfect" model and creates ranking distributions for all models, allowing some evaluation of model similarity

About

Suite of Jupyter Notebooks for calculation of regional and global metrics for evaluation of CMIP6 ESMs

Resources

Stars

Watchers

Forks

Packages

No packages published