Code supporting "Human-caused sea level rise drives 21st-century worldwide water level extremes" by Gilford et al. (2026, Science Advances)

This repository is the Python code base for "Human-caused sea level rise drives 21st-century worldwide water level extremes" by Gilford et al. (2026), published in Science Advances.

Project abstract:

The rate and impacts of sea level rise vary considerably around the world, but the contribution of human-caused climate change to increases in local and regional flood risks has not yet been systematically explored. Because such information is critical to local decision making, legal proceedings, and loss and damage determinations, we quantify human-caused climate change’s contributions to sea level rise at worldwide locations using budget-based and semiempirical model methods. Results show that human-caused sea level rise is quantifiable at 97% of 519 tide gauge sites and is responsible for 58% (44 to 65%) of the observed daily extreme water level exceedances over 2000–2018. On average, human-caused sea level rise has caused a near-tripling in the number of days with attributable exceedances since the 1970s.

If you have any questions, comments, or feedback on this work or code, please contact Daniel Gilford or open an Issue in the repository.

Citation

If you use any part of this work, please cite this repository, Gilford et al. (2026), and include a link.

Gilford, D. M., Lin, Y., Dahl, K., Pershing, A., Kopp, R. E., & Strauss, B. (2026). Human-caused sea level rise drives 21st-century worldwide water level extremes. Science Advances, 12, eadz3595. https://doi.org/10.1126/sciadv.adz3595

The latest software release is available at:

Data Availability

• Observed Modern Global Mean Temperatures---HadCRUT5 (version HadCRUT5.5.0.2.0)
• Observed Common Era Global Mean Temperatures---PAGES2k (available here)
• Simulated Global Mean Temperatures---Google Cloud CMIP6 data (38 paired+unpaired model simulations)
• Gridded Sea Level Budget---Frederikse et al. (2020; F20)
• Literature-derived Attributable Fractions based in part on:
  • Antarctic Ice Sheet (AIS)---SROCC Report, sections 3.3.1.6 and 4.2.2.5
  • Greenland Ice Sheet (GrIS)---Trusel et al. (2018)
  • Mountain Glaciers---Roe et al. (2021)
  • Global Mean Thermosteric/Thermal Expansion---Marcos et al. (2014), Slangen et al. (2014), and Tokarska et al. (2019)
• Tide Gauge Water Levels---Global Extreme Sea Level Analysis (GESLA; version 3)
• Vertical Land Motion---Hammond et al. (2021)

Analysis output data files are available at , and from the author upon reasonable request.

Getting Started

Key Analysis Code Dependencies

• cartopy
• haversine
• intake
• joblib
• matplotlib
• netCDF4
• numba
• numpy
• pandas
• PyEMD
• scipy
• seaborn
• shapely
• tqdm
• xarray

The notebooks also import the gesla reader module (GeslaDataset/gesla.py), which ships with the GESLA version 3 download (see Data Availability) rather than as an installable package.

File Descriptions

Notebooks

• 00_cmip6_gmat.ipynb — Loads and analyzes CMIP6 global mean temperature (GMT) time series used as forcing and context for the study
• 01_SEmodel_create_temperature_forcing.ipynb — Generates GMT-forcing time series and related inputs required by the semi-empirical model
• 02_SEmodel_run_analysis.ipynb — Runs the SE model across scenarios/ensembles and performs core analysis on outputs
• 03_reorganize_F20_data.ipynb — Reorganizes and standardizes the F20 dataset into the repository’s expected format for downstream processing
• 04_GESLAstations_qc_and_matchvlm.ipynb — Performs quality control on GESLA station records and matches station records with vertical land motion (VLM) data
• 05_literature_attributable_fractions.ipynb — Gathers, computes, and outputs literature-based attributable fractions for sea-level changes and compares them to model results
• 06a_regrid_ASLR_to_GESLAstations.ipynb — Regrids ASLR model outputs onto GESLA tide gauge station locations for direct comparison
• 06b_regrid_F20_to_GESLAstations.ipynb — Regrids the F20 dataset onto GESLA tide gauge station locations for direct comparison
• 07a_regional_steric_analysis.ipynb — Analyzes regional steric sea-level changes and their spatial patterns
• 07b_regional_steric_attribution.ipynb — Attributes regional steric changes and outputs results
• 08_GESLA_trends_and_budget_closure.ipynb — Computes trends from GESLA data and assesses sea-level budget closure across tide gauge stations
• 09_SEmodel_apply_spatial_fingerprint.ipynb — Applies spatial fingerprints from the SE model to station data to attribute the pattern of local changes
• 10_define_ewl_thresholds.ipynb — Defines extreme water level (EWL) thresholds used to identify extreme events in the study
• 11_calculate_ewl_exceedances.ipynb — Calculates exceedances of EWL thresholds at stations
• 12_aggregate_exceedances.ipynb — Computes attributable differences between exceedance counts with and without ASLR at budget-closed stations, and aggregates them into decadal bins
• 13_plot_study_results.ipynb — Produces the study’s figures and summary visualizations for results and publication

Function Modules

• SE_model - Core Python implementation of the sea-level semi-empirical (SE) model used by the notebooks to simulate sea-level responses to GMT
• utilities - Collection of helper functions for data I/O, processing, regridding, and plotting used across the notebooks

Figures:

Main Text Figures:

• Fig. 1A+B. Worldwide maps showing (A) central-scenario budget-based attributable sea level rise since 1900 and (B) the fraction of total budgeted rise that is attributable at each gridded/tide-gauge location.
• Fig. 2A+B+C+D+E. Global maps showing the percentage contribution of each component (thermosteric, mountain glaciers, Greenland, Antarctic ice sheets, and regional steric residual) to total budget-based attributable sea level rise since 1900.
• Fig. 3A+B+C+D. Bar charts aggregating extreme water-level exceedances across tide gauges by year/decade, splitting counts and percentages into attributable versus non-attributable exceedances under budget-based and semi-empirical counterfactuals.
• Fig. 4A+B+C. Maps summarizing where human-caused sea level rise added EWL exceedance days in 2000–2018, what share of observed exceedances was attributable, and how attributable exceedances increased relative to 1970–1989.

Supplementary Figures:

• Fig. S1. Time series of global mean temperature anomalies showing the observed/reconstructed forcings and the Stable and CMIP6 natural-forcing counterfactuals used to drive semi-empirical sea-level simulations.
• Fig. S2. Semi-empirical model outputs of historical and counterfactual global mean sea level rise and their differences, quantifying attributable sea level rise distributions for Stable and CMIP6 scenarios.
• Fig. S3A+B. Global time series comparing budget-based and semi-empirical global mean sea level and ASLR across scenarios and show the component breakdown for the high budget-based ASLR case.
• Fig. S4A+B+C+D+E+F. At Charleston, SC, panels combine local ASLR components and totals with observed and attributable EWL exceedances (annual and decadal) under budget-based and semi-empirical counterfactuals.
• Fig. S5A+B+C+D+E+F. As in Fig. S4, but for Cape Town
• Fig. S6A+B+C+D+E+F. As in Fig. S4, but for Taranaki, New Zealand
• Fig. S7A+B+C+D+E+F. As in Fig. S4, but for Hong Kong
• Fig. S8A+B+C+D+E+F. As in Fig. S4, but for Sitka, Alaska
• Fig. S9. Probability density functions across GESLA stations comparing observed sea-level trends with budget trends and budget trends corrected for vertical land motion, over the satellite era.
• Fig. S10. Thermosteric attributable sea level rise since 1900 from low/central/high budget counterfactuals compared against the Liu et al. 2024 thermosteric attribution estimate.
• Fig. S11. A global map showing the total change in the regional steric (ocean residual) sea-level term over 1957–2018.

Authors

Code Development and Lead Author

• Daniel M. Gilford, PhD - GitHub, Linktree

Study CRediT (Contributor Roles Taxonomy)

Daniel M. Gilford (DMG), Yucheng Lin (YL), Kristina Dahl (KD), Andrew Pershing (AP), Robert E. Kopp (REK), Benjamin Strauss (BS)

• Conceptualization: BS, DMG, KD, REK, YL
• Methodology: AP, BS, DMG, KD, REK, YL
• Software: DMG, YL
• Validation: DMG, YL
• Formal analysis: DMG, YL
• Investigation: DMG, YL
• Data curation: DMG, YL
• Writing – original draft: DMG
• Writing – review and editing: AP, BS, DMG, KD, REK, YL
• Visualization: DMG, YL
• Supervision: AP, BS, KD
• Project administration: BS, KD
• Funding acquisition: AP, BS, REK

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

Funding from the Bezos Earth Fund, NSF Grant ICER-1663807, and NASA Grant 80NSSC17K0698.

AI Transparency

Code and documentation updates made after release v1.0.1 (2025-12-22) were developed with assistance from Claude Code. The analysis supporting the published article corresponds to v1.0.1 and earlier, and was written without AI coding assistance. All AI-assisted changes were reviewed by the authors, who remain responsible for the contents of this repository.