data for “Projected impacts of extreme heat stress on China’s economy...”
Description
**Code for the paper ‘Projected impacts of extreme heat stress on China’s economy: A high-resolution coupling of biometeorological modeling and multi-regional input-output analysis’** **Abstract:** Our study develops a multi‑model integrated assessment framework, based on five high‑resolution CMIP6 climate models, the Liljegren physical model, a localized exposure‑response function and the AMRIO multi‑regional input‑output model, to systematically investigate the spatiotemporal patterns of future heat stress risk in China, the mechanisms of labor loss, and the resulting cascading economic effects along industrial chains under the SSP585 high‑emission scenario. The multi‑model ensemble simulation of WBGT during the historical period shows excellent overall performance (R>0.95), with the 50‑km‑resolution HadGEM3‑GC31‑HM model exhibiting the optimal performance. Future extreme heat stress risk in China, as represented by the 95th percentile of WBGT, is projected to increase significantly, with outdoor WBGT daily maximum values in South China exceeding 34 °C. Heat stress intensification follows a distinct NW-SE gradient, with larger increases in the northwest and northeast regions (1.8–2.0 °C) and smaller increases along the southeast coast (1.0–1.2 °C). This spatial heterogeneity is physically driven by the nonlinear humidity response to warming, modulated by regional topography. Future labor loss risk is the product of spatial coupling between climate stress and socioeconomic vulnerability, and two types of highly vulnerable regions are identified: “development‑lagged” and “transition‑stressed”. Under the no‑intervention scenario, national total industrial value-added loss is projected to increase from 2.7% of GDP in 2017 to 4.7% of GDP in 2050, with a growth rate exceeding that of GDP. The transport and warehousing sector, as a critical “loss transmission node”, exhibits extremely high loss rates (reaching 12.62% in Northeast China), significantly amplifying systemic risk through a positive feedback loop. The uncertainty ranges of all simulations exhibit pronounced asymmetry, with the upper bound (107–145%) dominated by parameters of socioeconomic system resilience, the importance of which is on a par with climate forcing itself. The code can be run in the following sequence once the prerequisite data files are prepared. This is provided for reference only: - **R1–6** Read, clean, and save meteorological, population and economic data - **CW1–6** Calculate historical and future, indoor and outdoor WBGT data, and generate ensemble - **V1–2** Historical validation of WBGT data - **CA** Calculate air‑conditioning penetration based on future meteorological, population and economic data - **CL1–5** Simulate labor productivity loss with air‑conditioning penetration incorporated - **CE1–2** Organize the MRIO table and simulate economic losses using the AMRIO model - **P1–9** Produce figures
Files
Steps to reproduce
The code can be run in the following sequence once the prerequisite data files are prepared. This is provided for reference only: - **R1–6** Read, clean, and save meteorological, population and economic data - **CW1–6** Calculate historical and future, indoor and outdoor WBGT data, and generate ensemble - **V1–2** Historical validation of WBGT data - **CA** Calculate air‑conditioning penetration based on future meteorological, population and economic data - **CL1–5** Simulate labor productivity loss with air‑conditioning penetration incorporated - **CE1–2** Organize the MRIO table and simulate economic losses using the AMRIO model - **P1–9** Produce figures
Institutions
- Sun Yat-sen UniversityGuangdong, Guangzhou