Detecting spatial pattern of inequality from remote sensing

Spatial inequalities across the globe are not easy to detect and satellite data have shown to be of use in this task. Earth Observation (EO) data combined with other information sources can provide complementary information to those derived from traditional methods. This research shows patterns of inequalities emerging by combining global night lights measured from Earth Observation, population density and built-up in 2015. The focus of the paper is to describe the spatial patterns that emerge by combing the three variables. This work focuses on processing EO data to derive information products, and in combining built-up- and population density with nighttime emission. The built-up surface was derived entirely from remote sensing archives using artificial intelligence and pattern recognition techniques. The built-up was combined with population census data to derive population density. Also the nighttime emission data were available from EO satellite sensors. The three layers are subsequently combined as three colour compositions based on the three primary colours (i.e. red green and blue) to display the “human settlement spatial pattern” maps. These GHSL nightlights provide insights in inequalities across the globe. Many patterns seem to be associated with countries income. Typically, high income countries are very well lit at night, low income countries are poorly lit at night. All larger cities of the world are lit at night, those in low-income countries are often less well lit than cites in high-income countries. There are also important differences in nightlights emission in conflict areas, or along borders of countries. This report provides a selected number of patterns that are described at the regional, national and local scale. However, in depth analysis would be required to assess more precisely that relation between wealth access to energy and countries GDP, for example. This work also addresses regional inequality in GHSL nightlights in Slovakia. The country was selected to address the deprivation of the Roma minority community. The work aims to relate the information from the GHSL nightlights with that collected from field survey and census information conducted at the national level. Socio-economic data available at subnational level was correlated with nightlight. The analysis shows that despite the potential of GHSL nightlights in identifying deprived areas, the measurement scale of satellite derived nightlights at 375 x 375 m and 750 x 750 m pixel size is too coarse to capture the inequalities of deprived communities that occur at finer scale. In addition, in the European context the gradient of inequality is not strong enough to produce strong evidence. Although there is a specific pattern of GHSL nightlights in settlements with high Roma presence, this cannot be used to identify such areas among the others. This work is part of the exploratory data analysis conducted within the GHSL team. The exploratory analysis will be followed by more quantitative assessments that will be available in future work.
JRC.E.1-Disaster Risk Management