The 'ghost city' phenomenon in China: mapping and categorization at intra-urban scale.

The excessive growth of urban landscapes in China over the past two decades has been accompanied by a particular phenomenon, namely ‘ghost cities’. It refers to a new development that is running at severe under capacity, i.e. with drastically fewer people and businesses than there is available urban space for. This phenomenon mostly does not affect the whole city, but exists at district or block level. However, due to the limitations of data resolutions, this phenomenon is predominantly studied at the city or at maximum district level. In this study, we provide an overview of recent block-level research efforts. We recapitulate a classification approach for mapping this phenomenon at the block level (i.e. ‘ghost block’) and its area-wide application across entire China based on multi-source remote sensing data. Furthermore, we categorize the identified ‘ghost blocks’ based on frequency, intra-urban spatial locations, and construction periods. Results show that 1) 1048 ‘ghost blocks’ with a coverage area of 354 km ² are identified within the urban landscapes in China; 2) seven types allow to describe frequencies, spatial patterns and development periods of ‘ghost blocks’. Furthermore, relations between these types and urban development processes in China are discussed.