Environmental changes affect vegetation dynamics and analyses of both fossil and modern pollen assemblages can be an important source of evidence. Taxonomic identification of pollen grains, however, can be a daunting task, as existing reference collections are kept private and inaccessible to researchers.
This paper presents a new online, open and peer-reviewed database of pollen data from across the world, allowing submission and identification of unknown grains through crowdsourcing, as well as a repository for digitization and sharing of existing reference collections. Connected to the GBIF.org database, The Global Pollen Project (GPP) tool provides direct and easy linkage from pollen grains to occurrences for more than 1,500 plant species.
Generating new opportunities for research, the GPP tool can be a valuable resource for automatic pollen identification through machine learning. As an alternative to physical slides that degrade over time, the tool has already proven useful in university teaching by being incorporated in lectures and lab exercises.