Highlights
•Data fusion allows representing land uses difficult to capture with land cover.
•Diverse land systems represented in one mapping product.
•Spatial assessment of shifting cultivation land systems for entire Lao territory.
•Agricultural frontiers signalled by forest-mosaic land systems.
Land cover data is widely used for the design and monitoring of land use policies despite the incapability of this type of data to represent multiple land uses and land management activities within the same landscape. In this study, we operationalized the concept of land systems for the case of the Lao PDR (Laos). Distinct land systems like shifting cultivation and plantations (land concessions) cannot be fully captured by land cover inventories alone, in spite of their relevance for land use policies. Using a decision tree and a matrix approach, we integrated several datasets for the period 2010/11, including land cover, an agricultural census and a land concession inventory. We selected thresholds for distinguishing land systems based on an expert survey. The resulting 17 land systems cover the whole territory of Laos and represent landscapes of 2 × 2 km pixel size. The largest area is occupied by smallholder agriculture land systems intertwined with forests. Only 27% of the territory are agriculturally undisturbed, dense forest systems. The assessment can serve as a basis to identify areas that could change shortly and locates the full range of land systems, from land concessions to smallholder systems, in one, integrated spatial assessment. The land system representation can help policy makers to link land systems to the diversity of different stakeholders and their backgrounds and support discussions about ecologic and socio-economic consequences of different land uses within a landscape.