关键词: 职业热应激, 供应链网络, 级联风险, 间接损失

Abstract: Utilizing high-resolution meteorological and population grid data, combined with multi-regional input-output tables of 313 regions in China, this study assesses the systemic macroeconomic impact of occupational heat stress induced by extreme heat exposure. By employing the comprehensive heat stress indicator Wet-Bulb Globe Temperature (WBGT) and exposure-response functions, the direct losses across sectors with varying labor intensities are quantified. These direct losses were input into a self-developed dynamic complex supply chain network model based on adaptive agents to simulate daily indirect, or cascading, economic losses during the baseline period (2021–2022) and under the SSP1-2.6 and SSP3-7.0 scenarios for 2030 and 2060. The results demonstrate that indirect losses propagated through the supply chain network play a dominant role. During the baseline simulation period, the total national economic loss is approximately 2.966 trillion CNY, with indirect losses accounting for 59.2%, and the evolution of its peak significantly lags behind that of direct losses. The loss distribution exhibits a pronounced structural mismatch: high labor-intensity sectors, such as agriculture, forestry, and construction, serve as the physical sources bearing the direct losses of occupational heat stress, whereas cities located at the core nodes of the supply chain network (e.g., Beijing and Shanghai) and moderate-to-low intensity sectors (e.g., manufacturing and services) emerge as the epicenters of indirect losses. Under the four climate change scenarios, national total losses increase substantially relative to the baseline. The total loss under the 2060 SSP3-7.0 scenario is 3.19 times that of the baseline, and the share of indirect losses further increases to 66.1%. These findings suggest that climate risk management should shift from physical protection at the micro-individual level toward resilience building across the entire macro-level supply chain.

Key words: Occupational heat stress, Supply chain network, Cascading risk, Indirect loss