README.md

This directory contains notebooks to generate LocalizeSL-based local SLR projections using a variety of temperature scenarios and ice sheet models, largely based around the IPCC AR5.

The final output of steps 1-4 is a Zarr store containing 10,000 Monte Carlo draws for each of the RCP scenarios and years (decadal), at each site ID (defined by LocalizeSL), for each corefile. Step 5 simply collapses these MC draws into quantiles.

The steps to produce this output are as follows:

1. convert-mat-version: Necessary to convert the MatLab version to work with Octave
2. generate-projected-lsl: Dask workers running Octave. For any corefile, call the LocalizeSL LocalizeStoredProjections function, followed by WriteTableMC, to get outputs as TSVs.
3. retrieve-num-gcms: Calculate number of GCMs for each site-year-scenario, for later use in clipping some sites due to data quality issues.
4. process-localizesl-output: combine all TSVs into a single Zarr store. Clip some sites based on data quality criteria.
5. create-slr-quantile: Take quantiles of the outputs of step (4), for use in the runs executed as part of Depsky et al. 2023.