Country-specific data sources and variable names used for population density estimation used for dasymetric weights.

* The variable names are used in Random Forest model output and throughout the text as reference to the specific data they were derived from. The first three letters are derived from the data type (e.g. “lan” indicates land cover) and the last three letters, if present, indicates what type of data each variable represents (e.g. “_cls” is a binary classification and “_dst” is a calculated Euclidean distance-to variable.† The default data for populated places is merged from several VMAP0 data sources. There are three VMAP0 data sets used: The point data pop/builtupp and pop/mispopp are buffered to 100 m and merged with the pop/builtupa polygons creating avector-based built layer. This layer is then converted to binary class and distance-to rasters for use in modeling.Country-specific data sources and variable names used for population density estimation used for dasymetric weights.