The WorldCereal project started and built a community-based open harmonized reference data repository at global extent ready for model training or product validation. Data from 2017 onwards were collected from many different sources, harmonized, and annotated. Therefore a harmonization protocol was developed to structure, annotate, harmonize and evaluate data. By the end of 2022 the repository holds around 75 million harmonized observations with standardized metadata available to the public.

We recommend continuing and institutionalizing this reference data initiative e.g. through GEOGLAM, and encouraging the community to publish land cover and crop type data following the open science and open data principles