Using nighttime light data to identify the structure of polycentric cities and evaluate urban centers.

Studying the structure of polycentric cities promotes a better understanding of the process of urban development and contributes to urban planning and management. However, existing studies cannot compare the level differences between urban centers and quantify the overall urban center development level within polycentric cities. Therefore, we combined nighttime light (NTL) data and the natural city (NC) to conduct this study from multiple perspectives. First, NCs were determined from Luojia-1A NTL data with high spatial resolution. Second, urban centers were selected from NCs. Third, urban center level (UCL) was proposed to compare the level differences of urban centers. Fourth, urban center development index (UCDI) was proposed to quantify the overall urban center development level within a polycentric city. A quantitative verification method was used to confirm that the detection accuracy of urban center identification is high. Furthermore, the applicability of the proposed method in different NTL datasets was verified by the identification of urban centers and the calculation of UCDIs. Unlike traditional methods, the shape and scope of the urban center identified using the proposed method are not restricted by administrative boundaries. Moreover, the level differences of urban centers and the overall urban center development level within a polycentric city, can be expressed by quantitative indicators, which helps in comparing the differences between urban centers.