Reproducible, transparent and traceable cleaning of IOC Tide Gauge Data

Accurate tide gauge records are essential for coastal monitoring, sea level analysis, and the calibration and validation of numerical models. However, global sea level data providers such as the Intergovernmental Oceanographic Commission (IOC)¹ often contain inconsistencies related to vertical datums, step changes, sensor noise, and undocumented interventions, which limit their direct applicability for modelling and validation purposes.

We present ioc_cleanup (github.com/oceanmodeling/ioc_cleanup) , an open-source Python repository designed to clean tide gauge time series using a reproducible and transparent workflow defined in structured JSON files. All transformations are traceable, version-controlled using Git, allowing for consistent quality control, peer-review and community-driven improvements. The framework explicitly addresses common data quality issues, including spikes, sensor noise, sensor replacement or substitution, and step changes, as well as the challenge of distinguishing bad data from genuine physical events such as storm-driven sea level extremes or tsunamis.

The cleaned datasets have been used for the calibration and validation of a global barotropic model, revealing systematic data quality patterns across stations and regions. While the framework is applied here to sea level data, the methodology is provider-agnostic and applicable to other geophysical time series.

By formalising expert-driven flagging and corrections in a transparent manner, ioc_cleanup provides a foundation for future developments, including the potential use of machine learning techniques to assist data flagging, reduce operator subjectivity, and extend spatial and temporal coverage. The framework offers a scalable contribution to other datasets (such as GESLA4²) and supports reproducible coastal data curation.

Citations:
[1] Flanders Marine Institute (VLIZ); Intergovernmental Oceanographic Commission (IOC) (2025): Sea level station monitoring facility. Accessed at https://www.ioc-sealevelmonitoring.org/ on 2025-12-15 at VLIZ. DOI: 10.14284/482

[2] Haigh, I.D., Marcos, M., Talke, S.A., Woodworth, P.L., Hunter, J.R. & Hague, B.S. et al. (2023) GESLA Version 3: A major update to the global higher-frequency sea-level dataset. Geoscience Data Journal, 10, 293–314. Available from: https://doi.org/10.1002/gdj3.174