5.1.1 Methods for estimating activity data
For estimating activity data, maps serve multiple purposes. First, subject to classification error, they depict the general spatial distribution of land attributes in general and forest resources in particular. Second, maps that depict forest-related classes or that can be aggregated or converted to forest-related classes can serve as the basis for stratification. Maps depicting forest classes and particularly forest change classes can be used to support construction of stratified sampling designs for purposes of estimating activity data. Third, maps of continuous variables such as percent forest canopy cover and even biomass can be used directly with model-assisted methods to estimate rates of forest change and can be aggregated to produce forest class maps.
Factors that influence a country’s decisions concerning which data and methods to use for assessing activity data include the nature of the forests in the country, forest management practices, availability of various kinds of satellite data, existing satellite image analysis capabilities, availability of ground-based data and general level of technological capacity. Spatial resolution, annual observations of forest disturbances, and attribution of land-cover changes by disturbance type all influence activity data uncertainty (Mascorro et al., 2015).