A grid-level fixed-asset model developed for China from 1951 to 2020

Abstract. To better aid quick and accurate assessment of economic loss after the occurrence of future damaging earthquakes, we developed a grid-level fixed-asset model for China covering the period from 1951 to 2020. The modelling process can be divided into two stages: (1) the compilation of provincial-level fixed-asset data series using the perpetual inventory method (PIM) and fixed-asset-related statistics and (2) the disaggregation of provincial-level fixed assets into grid-level assets (1 km×1 km resolution) using different combinations of remote sensing ancillary data (i.e. nighttime light, built-up surface area, population) for different periods, considering their temporal availability. As of 2020, the total estimated value of fixed assets in China reached CNY 589.31 trillion (at 2020 price levels). Consistency checks have been performed by comparing our modelled fixed assets with those from other studies and data sources at different administrative levels, and good consistency has been achieved. An application to the direct economic loss estimation of the 2023 Ms 6.2 Jishishan Earthquake that occurred in Gansu Province, China, is also presented to demonstrate the potential of the developed fixed-asset data for future damaging earthquake loss estimation in China. In conclusion, the limitations of the developed fixed-asset model are discussed to shed light on directions for future improvement. The modelled grid-level fixed-asset maps from 1951 to 2020 can be conveniently extended to more recent years as new statistics on fixed assets become available.