Global Forest Change dataset
Global Forest Change dataset, also known as Hansen Global Forest Change or simply Global Forest Change, is a global remote sensing dataset of tree cover extent and change produced by researchers at the University of Maryland and collaborators. Derived from time-series analysis of the Landsat program, it provides near-global coverage at roughly 30 metres per pixel. The maintained release covers 2000-2024 and includes layers for tree canopy cover in 2000, annual tree-cover loss and year of loss, cumulative tree-cover gain for 2000–2012, reference image composites from the first and last available years, and a data mask.
First described in a 2013 paper in Science, the dataset was created to provide a globally consistent annual record of tree-cover change using one methodology. The initial study reported 2.3 million square kilometres of tree-cover loss and 0.8 million square kilometres of tree-cover gain between 2000 and 2012 and found increasing loss in the tropics despite declines elsewhere. Its development depended on the opening of the Landsat archive and cloud-based computation, allowing the product to evolve from a one-time publication into a maintained data series.
The dataset became widely used in forest monitoring, conservation, environmental accountability and national reporting. It broadened access to high-resolution forest-change data beyond specialist mapping agencies and became an important input to downstream platforms such as Global Forest Watch.
Its interpretation has also been contested. Because it measures biophysical tree-cover change rather than land-use change, mapped loss may reflect harvesting, plantation cycles, fire, storms, pests or permanent conversion to another land use. Critics argued soon after publication that it could blur natural forests and plantations, while later validation studies found that default canopy thresholds often require calibration before the data can be used reliably in national statistics or as a proxy for deforestation.