Anais do XIX Simpósio Brasileiro de Sensoriamento Remoto

The goal of this work is to explore variables that are recurrently pointed out as factors influencing the night-time levels around the world, in the Amazonian context. In order to do that, we generalized variables related with physical, anthropogenic and socioeconomic contexts at the city level and analyzed their correlation with the night-time radiance data from the Day-Night Band (DNB) sensor, a detector from VIIRS-NPP. We found that the GDP per capita, road network density and the total number of households have the greatest correlation coefficients and Spearman's coefficients. Our approach focused on the residual values of the variables, and so we found that cities with the greatest residual values were also cities with a similar historical context of development, such as mining and populational nuclei. Our results suggest that, when associated with other indicators, night-time data is useful for identifying patterns of socioeconomic dynamics.