Model intercomparison reveals improved extreme precipitation simulation in China, indicating resolution's role in predictions.
As a pioneering component of the Coupled Model Intercomparison Project Phase 6 (CMIP6), the High‐Resolution Model Intercomparison Project (HighResMIP) addresses critical resolution limitations in global climate models. This study evaluates three HighResMIP models (25–100 km resolution) in simulating precipitation over China (1961–2014) and projects changes in extremes during 2015–2050 relative to 1985–2014. We quantify impacts of horizontal resolution and ocean–atmosphere coupling, revealing that: high‐resolution configurations improve extreme precipitation correlations by 0.1–0.25 (notably for consecutive dry days), yet yield < 5% improvement in mean precipitation accuracy. Projections indicate heavy precipitation days (R95p) will increase 15%–30% faster than total annual precipitation, while maximum dry spells (CDD) show high uncertainty (±22%–35% regional variability). Resolution effects amplify precipitation changes by 3%–8% in complex terrain. Ocean coupling moderates extremes by 10%–15% in continental interiors but intensifies southeastern droughts by 6–12 days/year. These findings demonstrate that while enhanced resolution improves regional precipitation forecasts, its effectiveness is spatially heterogeneous and requires complementary advances in model physics.
No takes yet. Share an insight, caveat, or question.
Jin-ge et al. (2025) studied this question.