Distributions of Uncertainty by Climate Impact

These summary tables describe the distributions of results produced by allowing only one of statistical, model, and weather uncertainty to vary, to determine the contributions of each.

These files are extracted from the general ACP Monte Carlo results using scripts from https://github.com/ClimateImpactLab/acp-impacts/tree/master/extract as defined below:

The contents of variation-split are produced by uncertain.py.  These results are organized into files with the filename template ----.csv.  VARIATION may be "impact", for the variation driven by statistical uncertainty; "model", for the uncertainty driven by GCM choice; "weather", for the variation driven by weather realization; or "total", allowing all of these to vary simultaneously.  IMPACT may be any of the estimated impacts, RCP can be RCP 2.6, RCP 4.5, RCP 6.0, or RCP 8.5, and YEAR is 2020, for an average from 2020 - 2039; 2040 for an average of 2040 - 2059; or 2080 for an average of 2080 - 2099.

The contents of baseline-testing are produced by uncertain-test.py.  These results are organized into files with the filename template ----.csv.  The filename parts are as described above, with the addition of the name of the GCM that is held fixed when the GCM is not allowed to change.

variation-split.tsv is produced by calcvar.py and is a summary of the results in variation-split/.

baseline-testing.tsv is produced by calcvar-test.py and is a summary of the results in baseline-testing/.

This data is provided for non-commercial research and educational purposes.