Evaluating night-time light sources and correlation with socio-economic development using high-resolution multi-spectral Jilin-1 satellite imagery of Quito, Ecuador
C. Scott Watson,John R. Elliott,Marco Córdova,Jonathan Menoscal,Santiago Bonilla-Bedoya
DOI: https://doi.org/10.1080/01431161.2023.2205983
IF: 3.531
2023-05-06
International Journal of Remote Sensing
Abstract:Artificial light at night (ALAN) has positive and negative effects on social, economic, environmental, and ecological systems, and will increase with urban expansion. In this study, we used a multi-spectral 1.5 m resolution night-time acquisition from a Jilin-1 satellite over the city of Quito, Ecuador, to evaluate spatial lighting patterns in an expanding and topography complex-built environment. We demonstrated a requirement for robust georeferencing and orthorectification due to the complex topography, with errors on the order of 4–6 pixels (5.8–8.4 m CE95). We also quantified differences in observed brightness due to the image acquisition and local geometry. Street light type was distinguishable between high-pressure sodium (HPS) and light emitting diode (LED) sources (F1-score = 0.72–0.83) using a shark random forest decision tree approach. Additionally, street lights could be located within 10 m (F1-score = 0.71) with balanced omissions and commissions. Spatial trends revealed that the road network was the dominant source of illumination, accounting for 45% of illuminated pixels, whereas built-up areas accounted for 23%. Overall, 68% of all illuminated pixels were on or within 10 m of the road. Higher socio-economic development was associated with higher proportions of LED lighting, greater road network lighting and density of street lights, higher overall radiance for built-up areas and the road network, and greater coverage and illumination of designated green spaces. The broad impacts of ALAN mean that addressing the causes and consequences of lighting inequalities is a complex issue. Nonetheless, Jilin-1 night-time imagery offers a low-cost way to map and monitor light sources at high-resolution that will be beneficial to city-planners and progressing Sustainable Development Goals. Image acquisition geometry affects the observed radiance and would require consideration when evaluating spatial and temporal trends. Jilin-1 night-time satellite imagery offers a low-cost mechanism to map and monitor light sources at high-resolution that is sufficient to identify individual sources such as street lights, buildings, and recreational spaces. Radiance intensity was correlated with the degree of socio-economic development. In the higher socio-economic classes, we observed higher proportions of LED lighting, greater road network lighting and density of street lights, higher overall radiance for built-up areas and the road network, and greater coverage and illumination of designated green spaces.
imaging science & photographic technology,remote sensing