Estimation of high-resolution emissions using functionally-filtered nighttime lights in monocentric and polycentric cities: Fusion of remotely sensed human activities and social media data for carbon neutral settlements

Urban areas play a fundamental role in local and large-scale greenhouse gas emissions reduction efforts since they contribute to >70 % of the global budget for anthropogenic carbon dioxide. With current rapid urbanization showing a unique trend compared to the past few centuries globally, it is essential to estimate high-resolution emissions from settlements to recognize the consequences of landscape conversion in the built environment. Here, we develop and test a brand-new methodological framework to estimate the high-resolution emissions using functionally-filtered nighttime lights (FNL) in monocentric and polycentric cities, via a fusion of remotely sensed human activities and social media data. Field surveys verified that FNL are well consistent with the real spatial distribution of the settlements compared with original mixed nighttime lights (OMNL). The new emission mapping showed that the pattern in the monocentric city is more spatially concentrated, in comparison with geographically dispersed pattern in the polycentric city. Further analysis revealed that hotspot areas of the new emissions maps in the monocentric and polycentric cities are only a sixth and an eighth of those from OMNL, and the number of extreme value points drops from 48 to 12 and from 67 to 15, respectively. Therefore, high-resolution emissions using FNL in monocentric and polycentric cities improve the monitoring and understanding of the urban emissions dynamics and allow for careful examination and revision of urban mitigation policies and strategies aimed at offsetting the impacts of rapidly expanding urban environments beyond the single city.