ABSTRACT
This chapter presents an operational remote sensing approach for monitoring forest cover at continental and global levels, based on a statistical sampling design and on satellite imagery from optical sensors of moderate spatial resolution (30 m × 30 m resolution). Object-based approaches use pixel clustering algorithms to create spectrally homogenous pixel groupings, which are thereafter treated as individual units for analysis. For the Global Forest Resources Assessment 2010, theFood and Agriculture Organization of the UN has extended its global and continental monitoring of forest cover changes to include analysis of remotely sensed land cover and land use as a complement to standard national reporting. The survey applies object-based image analysis methods to a globally distributed, systematic sample of moderate-resolution satellite imagery to estimate forest land cover and land use change for the periods 1990–2000 and 2000–2005. The chapter presents the scientific and technical methods that have been developed for monitoring forest cover changes in the framework of this global survey.
