Data processing

We use pyglider to process our data.

We use the IOOS QARTOD toolbox to quality control our data. This toolbox applies flags on a per-variable basis. Flags are stored in a separate variable with the suffix _qc. Flags are not applied by default. Each flag variable includes metadata describing the parameters used for quality control, contributing variables and the version of the QARTOD toolbox used to generate the flags.

Data download

All our data are freely available from our ERDDAP server

To receive updates on our data products, including changes to processing, new variables, you can visit the Updates page or sign up to the mailling list

Please acknowledge data used in any publication with the following statement:

This study used data collected and made freely available by Voice of the Ocean Foundation ( accessed from

For data users

We have several jupyter notebooks with examples of how to query ERDDAP, perform server-side subsetting, and download data directly into Python available in a GitHub repo.

These notebooks were first presented as part of our 2023 ERDDAP introduction workshop in Gothenburg. These notebooks explain more of how ERDDAP works, and include examples of data analysis from other ERDDAP servers.

For data managers

We set up our ERDDAP with assistance from SMHI, EMODnet and ETT solutions. We use a docker image based on the one produced by axiom.

To manage our datasets, we have written a Python script and some shell helper scripts to add new datasets and update existing datasets. These scripts are available in a GitHub repo.

All datasets are subject to automated and manual quality control checks, however they may contain errors. Known errors/issues are detailed on the data updates page. Please direct all queries about data or the ERDDAP server to Data Scientist Callum Rollo and Head of Science Louise Biddle