The effect of aerosols and sea surface temperature on China's climate in the late twentieth century from ensembles of global climate simulations

Over the late twentieth century, China has seen a strong increase in aerosol emissions, whose quantitative role for observed changes in surface solar radiation (SSR), surface air temperature (SAT), and precipitation remains debated. We use ensembles of transient sensitivity experiments with the global climate model ECHAM5 from the Max Planck Institute for Meteorology, Hamburg, Germany, combined with the Hamburg Aerosol Module to examine the effect of aerosols and prescribed, observation‐based sea surface temperatures (SSTs) on the above variables. Observations and control experiments agree reasonably well in eastern China in terms of SSR dimming (−6 ± 2 W/m 2 /decade, 1960–2000; stronger than in models of the Coupled Model Intercomparison Project Phase 5, CMIP5), statistically nonsignificant summer SAT trend (1950–2005), and drying in summer from 1950 to 1990 (−2.5% to −3.5% per decade, essentially via reduction of convective precipitation). Other observed features are not reproduced by the model, e.g., precipitation increase in the 1990s in the Yangtze River valley or, from the 1960s onward, the strong winter warming in northern China and Mongolia and SSR dimming in western China. Aerosol effects are stronger for sulfur dioxide than for black and organic carbon and are more pronounced at lower model resolution. Transient SSTs are crucial for decadal‐scale SAT variability over land, especially the strong warming in the 1990s, and, via SST forced reduction of cloud cover, for the ceasing of SSR dimming around the year 2000. Unforced cloud variability leads to relevant scatter (up to ±2 W/m 2 /decade) of modeled SSR trends at individual observation sites.