Guidelines for sharing datasets

[tnipen]

Thanks for sharing your Verif dataset with the community. Your dataset allows those interested in verification to play around with real-life datasets. All datasets can be interesting, but here are some tips for making them even more useful:

• Try to aim for datasets in the size range 1 - 100 MB. Other sizes are fine too, but this range is ideal for experimenting with.
• Try to provide the data in NetCDF. This makes it faster to work with. Use scripts/text2nc.py to convert data in text format to NetCDF.
• When posting a datasets, describe the dataset a bit to give some context. If the dataset contains multiple files (e.g. different modelling systems or post-processing schemes), then describe what the differences between them are.
• Describe how to download the data. You can host the data yourself, or consider uploading your dataset to the verif Zenodo community: https://zenodo.org/communities/verif/
• Describe the license for using the data