Crowdsourced Vegetation Monitoring: Global Plant Trait Mapping
Source data and 1 km-resolution plant trait maps, integrating citizen science (GBIF) with professional data (sPlot) to map plant functional traits globally.
- Size
- 8.4 GB
- Format
- ZIP
- License
- CC-BY-4.0
- Access
- Open Access
- Source
- Zenodo (record 18108765)
Your edits to the notebook only affect your own copy in the Hub. Changed it and want the original back? Restore it here.
Challenge
Where on Earth do citizen-science (GBIF) and professional-survey (sPlot) plant-trait maps agree, and where do they systematically diverge? For which traits does crowdsourcing hold up, and which stay the exclusive domain of expert plots?
- Where to start
- Read SourceData.xlsx straight from inside SourceData.zip: use trait_id_mapping to decode TRY codes (X3106 = Plant height, X3117 = SLA, X26 = Seed mass), splot_gbif_correlation for how well the two maps agree per trait at 1 km, and all_results for the cross-validated R2 of the CIT, SCI and COMB models. Then pick one trait and stream just its rows from cv_obs_vs_pred.parquet (columns x, y, pred, trait_set_abbr; filter trait_id and trait_set_abbr in CIT/SCI) and pivot to compare the two predictions cell by cell.
- What to share
- A per-trait agreement ranking plus a global CIT-minus-SCI difference map for one trait (x, y are already an equal-area projection), and a short comment on where — geographically and for which trait types — the crowd matches the professionals and where it does not.
💬 Discuss this dataset, ask questions and share your results in its discussion thread. Challenges are open-ended — there's no single right answer.