From Cumulative CO₂ Emissions to Compound Climate Extremes: Is Eight-Dimensional Coherence a Mediating Mechanism? ——An Observational-Constrained Empirical Test of Global Compound Hot–Dry and Hot–Wet Events
Authors/Creators
Description
Compound hot–dry and hot–wet events are critical manifestations of climate change risk. Recent research has shown that compound extreme events respond nonlinearly to cumulative CO₂ emissions, with rare and severe events escalating faster than conventional climate model estimates suggest. This paper proposes a testable question: whether cumulative CO₂ emissions first alter the multidimensional coherence state among climate, land-surface, and ecological systems, and whether this coherence change subsequently modulates the probability of compound extreme events. We construct a joint global observational–CMIP6 database, forming an eight-dimensional state space with variables including temperature, precipitation, soil moisture, radiation, vegetation, carbon cycling, atmospheric circulation, and human exposure. We calculate phase differences, circular coherence, and phase dispersion, and establish a dynamic mediation model of CO₂ → coherence → compound extremes. By comparing with traditional TCoRE, joint probability, and marginal distribution models, we test whether incorporating eight-dimensional coherence explains additional risk variation, and examine robustness through causal lag analysis, counterfactual simulations, model ensembles, and observational constraints. We find that: (1) the eight-dimensional global coherence C_8 exhibits a systematic decline with increasing cumulative CO₂ emissions, with each 0.1 decrease in C_8 corresponding to a 35–55% increase in compound extreme event probability; (2) mediation analysis shows that C_8 changes explain approximately 38–52% of the total effect of CO₂ emissions on compound extreme events, indicating that coherence decline is a significant mediating pathway; (3) after observational constraint, the Coherence-Adjusted Compound Extreme Response (CACER) is 42–68% higher than the CMIP6 multi-model ensemble mean, highly consistent with the TCoRE findings of Li et al. (2026); (4) the energy (D1), moisture (D2), and circulation (D4) dimensions contribute most significantly and stably to the mediation effect. Our aim is not to replace cumulative emission response theory, but to test whether coherence state can serve as an intermediate dynamical variable connecting external forcing and compound risk. If stable results are obtained, this will provide empirical foundations for upgrading climate risk theory from “emissions–mean temperature” to “emissions–system coherence–compound risk.”
Files
Files
(170.3 kB)
| Name | Size | |
|---|---|---|
|
md5:db8563f7f2553a13575be6ff2b4da233
|
170.3 kB | Download |