Research Data Management Plan for OceanParcels

Summary statement

The most valuable type of data created in TOPIOS and OceanParcels is python and C code, stored on github. This code should be self-containing and documented enough to be run and understood by others without much effort.

Each sub-project (e.g. PhD chapter, MSc thesis, Bsc thesis) will require a new repository on the OceanParcels github page. Initially, this repository will be private, only accessible to Team Members within the OceanParcels team. Team members working on a project will make sure that the online repository is an up-to-date reflection of the project, and that it contains all the python and/or jupyter scripts used in the project.

Most importantly, the project repository will at the end contain the python or jupyter scripts to recreate all the Figures used in the manuscript/chapter/thesis, as well as any information on the data required as input for these scripts. This ensures full reproducibility and transparency of at a minimum the creation of the Figures.

Data Storage and Back-up

All of the scripts required to run, analyse and plot the trajectories used in papers and theses will need to be version controlled within the appropriate project repository on the OceanParcels github account.

Experiments with OceanParcels will almost always require input data in the form of hydrodynamic ocean model data in NetCDF format. These are at least the velocities, but can also include other variables such as temperature, salinity, plankton concentration etc. Each Parcels python script and jupyter notebook should clearly state the source of these NetCDF files (ideally as a DOI if known, or otherwise as a web archive). Even if the data is already stored locally, it is still better to give, as a comment, the original source of the NetCDF files.

Within TOPIOS, datasets of observational data as reported in the literature will also be collected. These should be converted to NetCDF format where possible, with clear attribution to the original study (including DOI) in the comments attribute. The resulting NetCDF data files will be stored on YODA, see this tutorial.

Data Access and Security

Every team member will have access to all the private repositories on the OceanParcels github account and to all the hydrodynamic ocean model data on the gemini.science.uu.nl server, with the goal to facilitate sharing of code between team members. Erik van Sebille has responsibility for granting and revoking of team member status.

Data Preservation and Archiving

Once a manuscript or thesis is published, the goal is for the repository associated with that paper to be made public. This is good practice in Open Science, and only in specific cases and with good reason can this be changed. Each repository containing code (scripts and notebooks) will then also be archived on zenodo.org, so that it obtains a permanent DOI.

Data Sharing and Reuse

Since the scripts used to generate all the Figures and other analysis of each published paper/thesis in principle is open to the public, data sharing is a matter of pointing interested parties to the appropriate github or zenodo page.

All code will in principle fall under the same MIT license as the Parcels license (see https://github.com/OceanParcels/parcels/blob/master/LICENSE.md). Exceptions to this rule can be made (and should be made when external code is sued that is published under a different license), in discussion with the project leader.

Summary Table

Name of group/project	Tracking Of Plastics In Our Seas (TOPIOS) and OceanParcels
Funding body(ies) and grant number	TOPIOS is funded by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 715386).
Partner organizations	Utrecht University, Imperial College London
Project duration	Start: 2017-04-01 End: 2022-03-31
Date first version	2017-08-17
Date last update	2020-04-09
Current Version	V1.1
Name of researcher with responsibilities for data management	The final responsibility for this DMP lies with Dr Erik van Sebille ([email protected])