Repository for detection of oil palm in Google Earth Engine.
In this study we develop an approach to map oil palm on a large scale for South East Asia including age of plantations. Our pixel-based classification approach uses a set of input features derived from Copernicus Sentinel-1 data as well as from additional auxiliary datasets. The high resolution and frequency of Sentinel-1 observations enables us to detect not only large oil palm plantations, but also smallholder farms under 1ha in size. Using a continuous flow of Sentinel-1 data we aim to make our approach operational and enable oil palm plantation expansion monitoring.
The map can be interactively navigated here.
The data can be downloaded at https://dare.iiasa.ac.at/85/
DOI: 10.22022/ESM/05-2020.85
