Large-scale and rapid perception of regional economic resilience from data-driven insights

Abstract

Developing general resilience measures that take into account spatio-temporal dynamics to withstand the adverse effects of shocks on the economy is urgent during the COVID-19 pandemic. However, rapid perception of city economic resilience at large scales is currently a challenge during disasters. Using machine learning to massively simulate hourly anthropogenic NO2 emissions from 2016 to 2020, a resilience quantification framework based on an undesired output perspective is proposed to assess the resilience of Chinese cities’ economic operations during the COVID-19 pandemic. The results show that NO2 can characterize economic activity except for the primary industry. Spatially, the economic resilience of Chinese cities at different stages of the pandemic showed a binary pattern of Huanyong Hu line divergence and north-south divergence, respectively. Temporally, economic resilience had a hysteresis effect. Moreover, cities with larger economies recovered more quickly, despite being hit harder. Measurement of economic resilience based on undesired output required integration of information on fluctuations and trends in emissions. Our study provides a new tool for perceiving resilience during disasters from an undesired output perspective to provide support and insight into city management and planning in the post-pandemic era.

Keywords

• COVID-19
• undesired output
• NO2
• economic resilience
• perceiving