UDTR Global Rights of Nature Radar

Africa masks the global freshwater loss without ensuring water security

Identification

Public canonical key

radar:r-375fa446395fa2ca94cf6433fd81eb2f

Authors

Ashraf, Rateb

Reported year

2026

Reported language

und

Source type

Journal

First detected

2026-11-13T08:38:57.300734+00:00

Last metadata observation

2026-11-13T08:38:57.300734+00:00

Last HTTP verification

Not checked

Documentary abstract — original language

# African water-storage gains mask the global loss of land water: code and source data Deposit accompanying: Rateb, A., "African water-storage gains mask the global loss of landwater" (submitted to Earth’s Future, 2026). ## Contents - `source_data/`: one CSV per figure panel of the five main display items (fig1a through fig5c),  plus `data_inventory.csv` and `data_inventory.xlsx`, the full dataset inventory behind  Supplementary Table S7 (version, provider, resolution, coverage, role, citation key, path),  with `README.csv` describing each file's columns, units, and generating script. These CSVs  reproduce every plotted value, including the posterior intervals.- `scripts/`: the analysis and figure scripts sufficient to reproduce the paper's numbers and  display items (26 files; the selection covers the claim chain, not the full project history).- `Project.toml`: the Julia environment. ## Script map (what produces what) Shared infrastructure:- `_plots.jl`: figure conventions (Nature preset, colormaps, panel helpers, save paths).- `_mc_pipeline.jl`: the per-region Monte Carlo pipeline (heteroscedastic Matern-5/2 Gaussian  process, DerSimonian-Laird fusion with a Birge floor, 10,000 draws per region).- `_emergence_bayes.jl`: the Bayesian time-of-emergence estimator (structured systematics, full  GP covariance, surrogate-calibrated threshold, right-censoring).- `_nile_group.jl`: the Upper Nile aggregate (ten upstream reaches).- `D01_data_inventory.jl`: the dataset inventory; re-reads every source file and regenerates  the inventory CSV, the workbook, and the supplementary table. Posteriors and budgets:- `G25_all_regions_montecarlo.jl`, `B57_basin_montecarlo.jl`: continental and per-basin MC  posteriors (every velocity, reversal, and trend statement).- `B10_gci_uq.jl`: the compensation index posterior (200,000 joint draws).- `B50_sea_level_budget.jl`, `B53_cumulative_budget.jl`: sea-level expression and closed budget.- `B13_emergence.jl`: emergence runs; `B62_mask_failure_curve.jl`: the dated failure risk. The store (compartments):- `B45_compartment_budget.jl`: compartment variability (SMAP window).- `B63_where_water_lives.jl`: the per-basin trend account.- `B70_surface_water_rebuild.jl`: the complete 452-target surface-water inventory and the  large-waterbody leverage posterior; `B70c_budget_figs.jl`: the account display items.- `B68_epoch_budget.jl`: the account by epoch; `B69_cci_soil.jl`: the ESA CCI soil check;  `B18_concentration.jl`: basin concentration.- `B66_compartment_conditioning.jl`: the joint three-store conditioning model (observation  equations, identifiability gate, adaptive MH). Verification and figures:- `B47_number_audit.jl`: recomputes every headline number from the result artifacts and prints  computed against paper values.- `SF1_budget.jl`, `SF2_mask.jl`, `B16_emergence_delay.jl`, `SF4_unstable.jl`,  `SF5_security.jl`: the five main figures. `B08_trend_map.jl`: the per-mascon trend map  (Extended Data). ## Requirements Julia 1.12. Registered dependencies are in `Project.toml`. The scripts also use a set ofworkspace packages by the same author (GraceTools, GeoKit, TrendAnalysis, FigStd, Causality,ExtremesSuite), available from the author's GitHub (github.com/arateb) and archived with thisdeposit's DOI chain. ## Input data (public archives; not redistributed here) - GRACE/GRACE-FO mascons: JPL RL06.3, CSR RL06.3, GSFC RL06v2.0 (from PO.DAAC, CSR, and GSFC).- SMAP L4 root-zone soil moisture (NSIDC).- ESA CCI / C3S soil moisture (COMBINED, monthly).- DAHITI water-level and volume series (DGFI-TUM).- HydroLAKES; GLWD v2 wetlands; GDW dams; WorldPop; AQUASTAT withdrawal. Scripts read these from a local data store; paths are set at the top of each script. ## License Code: MIT. Source-data CSVs: CC BY 4.0.

Identifiers

DOI

10.5281/zenodo.23844461

Documentary context

Documentary family

unknown

Source-reported types

Journal

Publishers

Zenodo

Entity mentions are not evidence of legal recognition. Jurisdiction studied is separate from author affiliation country.

Sources

DataCite · Record 1

• Document · https://zenodo.org/doi/10.5281/zenodo.23844461

Documentary status

Relevance pending

Library

This identity is not yet included in the Legal Library.