Measuring three-dimension urban expansion using multi-source data and change detection algorithm: A case study of Shanghai

Abstract

Irrational urban expansion debilitates the sustainability of the natural environment and human settlement environment. However, urban expansion research often prioritizes horizontal expansion over vertical growth. In this study, we proposed a method to measure three-dimensional urban expansion. Firstly, the random forest classifier was applied to extract construction land in the initial year and the final year to detect incremental construction land. Then, the Continuous Change Detection and Classification algorithm assisted in pinpointing the conversion time. The building height data was obtained through the map crawler and assigned to the construction land. Finally, the 3D urban expansion was measured by combining construction time and building height. Shanghai was selected as the study area. The results revealed that the construction land area increased by 1015.39 km² since 2000, with an average annual growth rate of 2.58 %; the building volume increased by 1.5 billion m³, with an average annual growth rate of 1.31 %. Though urban expansion continued to spread from the city center to the periphery, the rate of horizontal and vertical expansion has slowed down, and the tendency of horizontal expansion has enhanced. The method can support informed decision-making in urban land use, energy consumption, and human settlement environment.