The weighting layer here below represents the estimated population density in each 3 arc seconds (approximately 100 m at the equator) obtained using the Random Forest (RF)-based dasymetric mapping approach developed by Stevens et al. (2015)*. This weighting layer is then used to dasymetrically disaggregate population counts from administrative units into grid cells to obtain the final population distribution datasets (both remaining unadjusted and being adjusted to match the most recent UNPD estimates# available at the time of production). A brief description of the RF model and all its inputs, including the source and spatial distribution of each covariate, is also provided in this metadata report.

*Stevens, F. R., Gaughan, A. E., Linard, C. & Tatem, A. J. Disaggregating Census Data for Population Mapping Using Random Forests with Remotely-Sensed and Ancillary Data. PLoS ONE 10, e0107042 (2015).
https://esa.un.org/unpd/wpp/Download/Standard/Population/




Call:
 randomForest(x = x_data, y = y_data, ntree = popfit$ntree, mtry = popfit$mtry,      nodesize = length(y_data)/1000, importance = TRUE, proximity = rfg.proximity,      do.trace = F) 
               Type of random forest: regression
                     Number of trees: 500
No. of variables tried at each split: 7

          Mean of squared residuals: 0.436013
                    % Var explained: 93.46