Dataset for "Future freshwater input for Greenland's fjords" submitted to JGR Earth Surface v1.1

# Datasets for Future Freshwater for Greenland's Fjords v1.1 This dataset contains the data used to study freshwater entering Greenland's fjords: runoff from the ice sheet, from peripheral glaciers and ice caps, and from ice-free tundra; precipitation falling directly on the fjord surface; and solid ice discharge. It covers 1950–2100 and includes both historical simulations and future climate projections (SSP1-2.6, SSP2-4.5, SSP5-8.5). Freshwater fluxes are from RACMO2.3p2 and MAR v3.12 forced by CESM2 and NorESM2; solid ice discharge is from CISM. The data are grouped by source of freshwater, model, and spatial or temporal resolution. The regional timeseries are extensively discussed in the manuscript, and are the preferred product by the authors. The fjord timeseries are added later. The code used to make the figures is available separately on GitHub: https://github.com/AnnekeV/FutureFresh This dataset supports the paper: Vries, A. et al., 2026. Future freshwater for Greenland's fjords. JGR Earth Surface. ## Explore it first, without downloading An interactive viewer for the ~830 individual fjord systems is at https://annekev.github.io/FutureFresh/ — click a fjord on the map to see its freshwater time series and seasonal cycle, and to find its id. ## What's in the dataset The record is in two parts, so that you do not have to download everything to get one time series. **Ready-to-use time series** — no code needed, gigatonnes throughout, regions named, units on every variable, NetCDF and CSV, monthly and annual: * `timeseries_per_region.zip` (4 MB) — the seven Greenland drainage basins plus a Greenland total, for five freshwater sources, four scenarios and up to four model combinations * `timeseries_per_fjord.zip` (60 MB) — the same fluxes summed per individual fjord system, for two fjord mask variants * `fjord_metadata_default.csv`, `fjord_metadata_incl_islands.csv` — position, drainage basin and size of every fjord id * `MANIFEST.csv` — every file with its size and SHA-256 checksum **Full preprocessed dataset** — for reproducing the paper's figures. Download all the `reproducibility_*.zip` archives and unzip them into one folder: * Topography and ice masks at 1 km resolution * Runoff from the Greenland Ice Sheet, glaciers and tundra * Precipitation data * MAR and RACMO climate model outputs * Region masks and mapping dictionaries * Gridded 1 km runoff climatologies * Per-fjord time series in their native units * Preprocessed and aggregated datasets ready for analysis ## Folder structure After unzipping the `reproducibility_*.zip` archives into a single folder: ├─ Icemask_Topo_Iceclasses_lon_lat_average_1km.nc # Topography and ice mask ├─ dict_calving_sensitivity.pkl # Calving sensitivity data ├─ mask_regions_RACMO_1km_optimized_output.nc # Optimized RACMO region mask ├─ masks1k.nc # Additional masks ├─ data/ │ ├─ Hist/ # Historical RACMO data │ ├─ MAR/ # MAR model outputs & masks │ │ └─ per_region_per_source/ # MAR outputs by region and source │ ├─ MOA/ # Melt over accumulation & climate indices │ ├─ climatologies/ # Gridded 1 km runoff climatologies │ ├─ fjord_timeseries/ # Per-fjord series in native units │ ├─ map/ # Mapping & plotting dictionaries │ ├─ precip/ # Precipitation datasets │ ├─ runoff/ # Runoff datasets (ice & tundra) │ └─ temp/ # Preprocessed/aggregated datasets That layout is unchanged from the previous version, so existing code that points at it keeps working. Set `FUTUREFRESH_DATA` to the folder you unzipped into, and the notebooks in the GitHub repository will find their data. The `timeseries_*` archives are separate from this tree and need none of it. ## What is new since the February 2026 version * Per-fjord time series for ~830 individual fjord "catchments", and a metadata table giving each fjord a position, a drainage basin and a size * Gridded 1 km runoff climatologies * Annual as well as monthly resolution throughout * Units, variable names, the generating script and a creation date recorded inside every file * Per-file SHA-256 checksums * `data/fjord_timeseries` and `data/climatologies` were missing from the previous upload; both are now included ## Licence Creative Commons Attribution 4.0 International (CC BY 4.0). The accompanying code is released under the MIT licence.