Articles | Volume 13, issue 3
Data description paper
24 Mar 2021
Data description paper |  | 24 Mar 2021

High-resolution global map of smallholder and industrial closed-canopy oil palm plantations

Adrià Descals, Serge Wich, Erik Meijaard, David L. A. Gaveau, Stephen Peedell, and Zoltan Szantoi

Data sets

High resolution global industrial and smallholder oil palm map for 2019 Adrià, Descals; Serge, Wich; Erik, Meijaard; David, Gaveau; Stephen, Peedell; Zoltan, Szantoi

Model code and software

oil_palm_global v1.0 (Version 1.0) A. Descals

Short summary
Decision-making for sustainable vegetable oil production requires accurate global oil crop maps. We used high-resolution satellite data to train a deep learning model that accurately classified industrial and smallholder oil palm, the main oil-producing crop. Our results outperformed previous studies and proved the suitability of deep learning for land use mapping. The global oil palm area was 21±0.42 Mha for 2019; however, young and sparse plantations were not included in this estimate.
Final-revised paper