Bangladesh Population Map Metadata Report

Prediction Weighting Layer Used in Population Redistribution

The data presented below represent the predicted number of people per ~100 m pixel as estimated using the random forest (RF) model as described in Stevens, et al. (2015). The following pages contain a description of the RF model and its covariates, their sources and any metadata collected for each covariate. The prediction weighting layer is used to dasymetrically redistribute the census counts and project counts to match estimated populations based on UN estimates for the final population maps provided by WorldPop.

Stevens, F. R., Gaughan, A. E., Linard, C., & Tatem, A. J. (2015). Disaggregating Census Data for Population Mapping Using Random Forests with Remotely-Sensed and Ancillary Data. PLOS ONE, 10(2), e0107042. doi:10.1371/journal.pone.0107042

plot of chunk predict_density

Bangladesh Census Data and Observed Population Density

These data are the population density values used to estimate the RF model used to create the prediction weighting layer you see above. Values represent population density as measured by people per hectare and calculated from population counts within each census unit. These values are used as the dependent variable during model estimation.

Bangladesh Census Data, 2011, Admin-level 5

Folder: Census
File Name: census_mauza.shp
Source: Bureau of Statistcs, Bangladesh, acquired by Gaughan, et al. for use in AsiaPop data products.
Description: These high spatial resolution census block data were attained through the Bangladesh Bureau of Statistics for 2011.
Class: polygon
Derived Covariates:
area, buff, zones,

class       : SpatialPolygonsDataFrame 
features    : 61835 
extent      : 298441, 772025, 2278578, 2946943  (xmin, xmax, ymin, ymax)
coord. ref. : NA 
variables   : 26

plot of chunk census_data


Random Forest Model and Diagnostics

These output and figures outline the estimated RF model that is used to predict the population density weighting layer. The model is fitted to the population density values for the preceding census data using covariates aggregated from the ancillary data sources summarized following the model diagnostics.


Call:
 randomForest(x = x_data, y = y_data, ntree = popfit$ntree, mtry = popfit$mtry,      nodesize = length(y_data)/1000, importance = TRUE, proximity = F) 
               Type of random forest: regression
                     Number of trees: 500
No. of variables tried at each split: 30

          Mean of squared residuals: 0.47
                    % Var explained: 66

plot of chunk random_forestplot of chunk random_forestplot of chunk random_forest

Covariate Metadata

Remotely-sensed, Classified Landcover

Folder: Landcover
File Name: bgd_ESA.tif
Source: MDA GeoCover Landcover Product, 30m
Description: Landcover from the Landsat-derived MDA GeoCover product, reclassified to match AfriPop coding and eventually broken down into binary classifications by aggregated land cover type (see Linard, et al., 2010 and Gaughan, et al. 2017 for category information).
Class: raster
Derived Covariates:
cls011, dst011, cls040, dst040, cls130, dst130, cls140, dst140, cls150, dst150, cls160, dst160, cls190, dst190, cls200, dst200, cls210, dst210, cls230, dst230, cls240, dst240, cls250, dst250, clsBLT, dstBLT,

class       : RasterBrick 
dimensions  : 6874, 5170, 35538580, 1  (nrow, ncol, ncell, nlayers)
resolution  : 100, 100  (x, y)
extent      : 279270, 796270, 2272009, 2959409  (xmin, xmax, ymin, ymax)
coord. ref. : +proj=tmerc +lat_0=0 +lon_0=90 +k=0.9996 +x_0=500000 +y_0=0 +a=6377276.345 +b=6356075.41314024 +towgs84=283.7,735.9,261.1,0,0,0,0 +units=m +no_defs 
data source : D:\Working_RF\data\BGD\Landcover\Derived\landcover.tif 
names       : landcover 
min values  :        11 
max values  :       210 

plot of chunk covariate_reports


Suomi NPP VIIRS-Derived 2012 Lights at Night, 15 arc-second

Folder: Lights
File Name: DEFAULT: VIIRS 2012
Source: http://ngdc.noaa.gov/eog/viirs/download_viirs_ntl.html
Description: These 'Lights at Night' data were derived from imagery collected by the Suomi National Polar-orbiting Partnership (NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) sensor. Data were collected in 2012 on moonless nights and though background noise associated with fires, gas-flares, volcanoes or aurora have not been removed it represents the best-available data for night-time light production.
Class: raster
Derived Covariates:
,

class       : RasterBrick 
dimensions  : 6874, 5170, 35538580, 1  (nrow, ncol, ncell, nlayers)
resolution  : 100, 100  (x, y)
extent      : 279270, 796270, 2272009, 2959409  (xmin, xmax, ymin, ymax)
coord. ref. : +proj=tmerc +lat_0=0 +lon_0=90 +k=0.9996 +x_0=500000 +y_0=0 +a=6377276.345 +b=6356075.41314024 +towgs84=283.7,735.9,261.1,0,0,0,0 +units=m +no_defs 
data source : D:\Working_RF\data\BGD\Lights\Derived\lights.tif 
names       :  lights 
min values  : -0.0023 
max values  :     316 

plot of chunk covariate_reports


WorldClim/BioClim Mean Annual Temperature 1950-2000, 30 arc-second

Folder: Temp
File Name: DEFAULT: BIO1
Source: http://www.worldclim.org/current
Description: WorldClim/BioClim 1950-2000 mean annual precipitation (BIO12) and mean annual temperature (BIO1) estimates (Hijmans et al., 2005) were downloaded, mosaicked and subset to match the extent of our land cover data for the mapping of this region.
Class: raster
Derived Covariates:
,

class       : RasterBrick 
dimensions  : 6915, 5181, 35826615, 1  (nrow, ncol, ncell, nlayers)
resolution  : 100, 100  (x, y)
extent      : 278670, 796770, 2269209, 2960709  (xmin, xmax, ymin, ymax)
coord. ref. : +proj=tmerc +lat_0=0 +lon_0=90 +k=0.9996 +x_0=500000 +y_0=0 +a=6377276.345 +b=6356075.41314024 +towgs84=283.7,735.9,261.1,0,0,0,0 +units=m +no_defs 
data source : D:\Working_RF\data\BGD\Temp\Derived\temp.tif 
names       : temp 
min values  :  149 
max values  :  267