Estimating global economic well-being with unlit settlements

Abstract

It is well established that nighttime radiance, measured from satellites, correlates with economic prosperity across the globe. In developing countries, areas with low levels of detected radiance generally indicate limited development - with unlit areas typically being disregarded. Here we combine satellite nighttime lights and the world settlement footprint for the year 2015 to show that 19% of the total settlement footprint of the planet had no detectable artificial radiance associated with it. The majority of unlit settlement footprints are found in Africa (39%), rising to 65% if we consider only rural settlement areas, along with numerous countries in the Middle East and Asia. Significant areas of unlit settlements are also located in some developed countries. For 49 countries spread across Africa, Asia and the Americas we are able to predict and map the wealth class obtained from ~2,400,000 geo-located households based upon the percent of unlit settlements, with an overall accuracy of 87%.

Fig. 1. Global country-level unlit settlement percentages.

a map of countries classified according to their percentage of settlements (building footprints) with no associated satellite-derived nighttime radiance for urban and rural regions combined. b African and Asian countries with population exceeding 50 million ranked according to percentage of urban unlit settlements. c African and Asian countries with population exceeding 50 million ranked according to percentage of rural unlit settlements.

Fig. 2. Relationships between the Demographic and Health Surveys (DHS) wealth index categories (Poorer, Average Richer) and the percentage of unlit settlements for 31 African countries.

The boxplots show the mean percentage area of unlit settlements within a 2 km buffer of a DHS urban household cluster and a 5 km buffer of a DHS rural household cluster against the mode of the wealth indices of all households assigned to the household cluster. The midline represents the median with the lower and upper limits of the box being the 1st and 3rd quartiles. The lower and upper whiskers represent minima/maxima no further than 1.5 times the interquartile range from the hinge. Outliers appear as circles.

Fig. 3. Country-level maps based on out-of-sample predictions showing the estimated relative wealth class (Poorer, Average or Richer) within a 2.5 km pixel.

a Bangladesh. b Cambodia, c Nigeria. d; Uganda. Capital cities are shown. Map coordinates in kilometers north and east.

Fig. 4. A spatial comparison of this study’s results over Nigeria.

a Wealth index produced via deep learning. b The Subnational Human Development Index (SHDI). c The share of unlit settlements predicting the SHDI Income Index. The lower and upper boxplot bounds represent 25th and 75th percentiles, respectively. The lower and upper whiskers represent minima/maxima no further than 1.5 times the interquartile range from the hinge. The notched boxplot represents the 95% confidence interval. The gray background indicates 95% confidence interval.