The Impact of Extreme Climate on Agricultural Production Resilience in China: Evidence from a Dynamic Panel Threshold Model

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Dynamic panel threshold model for climate resilience offers methodological advancement for empirical climate impact studies.

🏢実務担当者:Insights on threshold-based climate adaptation can guide agricultural risk management and investment planning under extreme scenarios.

🏛政策担当者:Highlights need for 'threshold-based' climate adaptation governance and risk-financing instruments to address tail risks in agriculture.

Against the backdrop of accelerating climate change, extreme weather events have increasingly caused yield losses in agricultural crops. Meanwhile, they undermine the stability of production systems, posing an increasingly severe threat to agriculture. This study draws on the “diversity–stability” hypothesis to construct a country-level measure of agricultural production resilience in China (ARES). Using output time series for multiple agricultural products, we capture the co-movements of shocks and system resilience through output stability and volatility. By combining ARES with climate exposure measures, we assemble a panel dataset covering 1343 counties over the period 2000–2023 and employ a dynamic panel threshold model to jointly account for persistence in ARES and state-dependent nonlinearities in climate impacts. The results reveal significant path dependence in ARES and pronounced threshold effects across climate dimensions. In the full sample, extreme high-temperature days become significantly detrimental after crossing the threshold, whereas extreme low-temperature days become significantly beneficial in the high-exposure regime. Extreme rainfall days and extreme drought days generally exhibit positive effects that weaken markedly beyond their respective thresholds, indicating diminishing marginal gains in ARES under severe exposure. The comprehensive climate physical risk index significantly suppresses ARES when it is below the threshold value; however, after surpassing the threshold, its marginal effect becomes significantly weaker. Heterogeneity analyses across hilly, plain, and mountainous areas, as well as nationally designated key counties for poverty alleviation and development, further show that threshold locations and regime-specific effects differ substantially by terrain and development conditions. These findings highlight the need for “threshold-based” climate adaptation governance, emphasizing targeted investments and risk-financing instruments to prevent ARES collapse under tail-risk regimes.

• semanticscholar https://doi.org/10.3390/agriculture16080825first seen 2026-06-29 09:00:44