Role of high-resolution phenology and productivity data from Sentinel-2 in prediction of grassland management type (data/code)
DATA.zip: gl22_public.shp - Skåne grasslands along with mean HR-VPP parameters and all vegetation indices for 2022, used for Fig.1, Fig.4 gl22_conf_pred_public.shp - the result of the grassland management type prediction based on the ANN model + conformal prediction. Used for Fig.7,8 model_accuracy_vs_data_size.xlsx - used for Fig. 9 feature_importance_ann.csv - used for Fig.10 ANN_Report.txt - used for Fig.6 RF_Report.txt, XGB_Report.txt - used in Appendix E(B) CODE.zip: North_field_scatterplots.R - reproduces Fig.3 Skane_province_boxplots_KWallis.R - reproduces Fig. 4. and Appendices F, G Skane_local_dots_KWallis.R - reproduces Fig.5 and Appendix H
Authors
- Anna Terskaia (ORCID: https://orcid.org/0000-0002-1708-8382)
Institutions
- Lund University (SE)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-05-21
- DOI
- https://doi.org/10.5281/zenodo.20329409
- Primary Topic
- Remote Sensing in Agriculture
- Type
- article
- Field-Weighted Citation Impact
- 0.00