Estimation of entity‐level land use and its application in urban sectoral land use footprint: A bottom‐up model with emerging geospatial data

Abstract Land is an essential resource tomaintain the functioning of the socio‐economic system. Due to sectoral land data limitations, previous studies were primarily restricted to a coarse sectoral level or focused mainly on the global and national scales. However,fine‐scale land use data are required to provide tailored implications for municipal sustainable development. With emerging geographic data and novel methods, including point of interest data, road network data, and natural language processing, a bottom‐up model is developed to estimate the entity‐level artificial impervious land use. Then, we conducted a case study in Shanghai to investigate the spatial features, footprints, and intensities of sectoral land use. Our results indicated that 42 sectors in Shanghai had diverse spatial patterns. The transportation sector had the highest level of agglomeration among all sectors, and the manufacturing industry's adjacent land patches had higher sectoral heterogeneities than the service sector. The transportation sector had the largest direct and embodied land use footprint. The residential‐related sectors had higher land use intensities, while the high value‐added service sectors showed lower land use intensities. Our study indicates that this model offers a novel way of extracting entity‐level spatial land use information and is applicable for socio‐economic metabolism research. Future studies could incorporate remote sensing images and multiple databases to achieve higher resolution.