IOC Cleanup

ioc_cleanup provides a reproducible, transparent, and traceable workflow for cleaning tide gauge data from IOC (Intergovernmental Oceanographic Commission) stations worldwide.

IOC Cleanup database

All stations with clean data between 1st of January 2020 and the 31st of december 2025.

Getting Started

Prerequisites

• Python 3.11+ (recommended)
• ~24 GB free disk space for raw IOC 2020-2025 data

Installation

git clone https://github.com/seareport/ioc_cleanup.git
pip install -r requirements.txt

Minimal example

import searvey
import ioc_cleanup as C

station = "abed"
df_raw = searvey.fetch_ioc_station(station, "2020-01-01", "2026-01-01")

trans = C.load_transformation_from_path(
    "../transformations/abed_bub.json"
)

df_clean = C.transform(df_raw, trans)

JSON transformations

Each cleaned station is described by a single JSON file that records every operation applied to the raw signal, so the dataset is fully reproducible from the raw IOC data plus the transformations.

• Zenodo - the complete dataset includes transformations.tar.gz. See Data Access.
• GitHub - also mirrored on the GitHub releases.
• Build your own - follow the JSON schema and load it with C.load_transformation_from_path(...).

Example for maya station:

From raw signal...

... using JSON transformation ...

... to clean signal