Identifying and evaluating suburbs in China from 2012 to 2020 based on SNPP–VIIRS nighttime light remotely sensed data

Abstract

• China’s suburbs can be effectively identified based on SNPP-VIIRS data. • We provide a more lone-time series and fine-resolution suburb dataset in China. • China’s suburbs present a fluctuation-growth trend. • Population density, GDP, and roads are major driver factors of suburban development. Suburbs, as bridges between urban areas and rural hinterlands, are areas with the most intense urban–rural conflicts and drastic land use changes in the urbanization process. Accurate identification and evaluation of suburbs are important to effectively break the urban–rural dichotomy, improve the utilization and management of land resources, and promote urban–rural integration and coordinated development. Previous suburb identification studies have suffered from low identification efficiency owing to the influence of subjective factors, small scales, short time-series, and single data characteristics. Thus, we took China as experimental object, and attempted to identify suburbs from the Suomi National Polar-orbit Partnership’s Visible Infrared Imaging Radiometer Suite (SNPP–VIIRS) nighttime light remotely sensed data using the K-means algorithm and subsequent series of post-processing approaches. Thereafter, our study further evaluated the spatiotemporal dynamics and driving factors of suburb development. Accuracy verification results show that suburb identification based on SNPP–VIIRS data can identify more details than the existing urban area data and Defense Meteorological Satellite Program’s Operational Linescan System data. Compared with traditional mutation detection methods, the proposed method has the advantages of being fast, efficient, and less subjective. Furthermore, we found that China’s suburbs present a fluctuation-growth trend, with the proportions increasing from 0.6% to 1.3% in the period 2012–2020. China’s suburb development was mainly driven by the development of population density, GDP, and road network. Our study provides an innovative way to conduct a rapid, efficient, and large-scale and accurate suburb identification over a long time series, thereby facilitating the study of socio-environmental issues in the urbanization process. The annual series (2012–2020) of suburbs in China are available free of charge at https://doi.org/10.7910/DVN/M5EED5 .