This repository contains a 10-m global oil palm extent layer for 2021 and a 30-m oil palm planting year layer from 1990 to 2021. The oil palm extent layer was produced using a convolutional neural network that identified industrial and smallholder plantations in Sentinel-1 data. The oil palm planting year was developed using a methodology specifically designed to detect the early stages of oil palm development in the Landsat time series. The repository contains the following data: - Grid_OilPalm2016-2021.shp: shapefile that delineates the 609 grid cells of 100 x 100 km where oil palm was found. - GlobalOilPalm_OP-extent.zip: 609 raster tiles of 100x100 km in geotiff format. The raster files show the results of the deep learning classification at a spatial resolution of 10 meters. The classes are the following: [0] Other land covers that are not oil palm. [1] Industrial oil palm plantations [2] Smallholder oil palm plantations. - GlobalOilPalm_YoP.zip: 609 raster tiles of 100x100 km in geotiff format. The raster files depict the year of oil palm plantation. The raster files have a spatial resolution of 30 meters. - Validation_points_GlobalOP2016-2021.shp: shapefile that contains the 18,812 points used to validate the global oil palm extent 2016–2021 and the oil palm age layer. Each point includes the attribute ‘Class’, which is the class assigned by visual interpretation of sub-meter resolution images, and the attributes ‘OP2016-2021’ and ‘OP2019’, which show the mapped classes in the oil palm extent 2016–2021 (this dataset) and the global oil palm layer 2019 (Descals et al., 2021), respectively. These attributes contain the following class values: [0] Other land covers that are not oil palm. [1] Industrial oil palm plantations. [2] Smallholder oil palm plantations. The oil palm extent and the planting year can be visualized at: https://ee-globaloilpalm.projects.earthengine.app/view/global-oil-palm-planting-year-1990-2021. This web map allows for the inspection of Landsat time series and the visualization of historical satellite images for a given oil palm plantation. If you are interested in obtaining the GeoTIFF file for a specific region, please refer to this Google Earth Engine script, which shows the IDs of each tile:https://code.earthengine.google.com/284cf1dc974295c57fe55bccf5acd83e Changelog v1.2: The validation dataset includes 1,000 additional points generated using stratified random sampling: 300 points for the class ‘smallholder oil palm’ and 700 points for ‘industrial oil palm’. The shapefile contains a new attribute, 'Sampling_type', which specifies whether the points were generated with simple or stratified random sampling.