Using historical disturbance identified with LandTrendr in Google Earth engine for land cover mapping of oil palm landscapes

In Malaysia, land under oil palm plantation has been steadily increasing. Meanwhile voluntary measures to improve sustainability of palm oil production have been introduced including regulation of land conversion to oil palm plantations. The objective of this study is to assess the utility of Google Earth Engine with the LandTrendr algorithm for classifying land cover, as a first step towards developing a tool for land cover change detection in Peninsular Malaysia to support Roundtable on Sustainable Oil Palm (RSPO) certification. Ground validation data on land cover and disturbance events from satellite imagery were used to calibrate LandTrendr to detect and map change from forest to oil palm, other vegetation or urban; other vegetation to oil palm; and oil palm to oil palm (replanting). The resulting disturbance rasters were then used with a 2019 multispectral Landsat mosaic in a random forests supervised classification. The classified maps of 2019 land cover showed a small improvement in accuracy with the addition of LandTrendr rasters over using only Landsat imagery. Our results suggest that disturbance history may provide useful ancillary information to support remote sensing mapping and LandTrendr could potentially become a useful tool for detecting land cover change in the tropics. In many cases the maps performed better with the addition of LandTrendr rasters; however, the resulting difference was small in overall accuracy. The method improved the accuracy of oil palm, rubber, forest and urban land covers, while it decreased the accuracy for other land cover classes. Vegetation classes such as oil palm, rubber and forest were often confused and remain challenging to map.