Larger Sensitivity of Precipitation Extremes to Aerosol Than Greenhouse Gas Forcing in CMIP5 Models Academic Article uri icon

abstract

  • ©2018. American Geophysical Union. All Rights Reserved. The sensitivity of precipitation extremes (PEs; i.e., the change in PE per degree of change in global mean surface temperature) to aerosol and greenhouse gas (GHG) forcings is examined using the twentieth century historical multimodel ensemble simulations from the Coupled Model Intercomparison Program phase 5 (CMIP5). We find a robustly larger sensitivity of PE to aerosols than GHGs across all available models. The aerosol/GHG-induced sensitivity ratios for globe-averaged monthly maximum consecutive 5-day precipitation (RX5day) and maximum 1-day precipitation (RX1day) in the multimodel ensemble are 1.6 and 1.4, respectively. Over land, the corresponding ratios for RX5day and RX1day are 2.3 and 1.8, respectively. In particular, the aerosol forcing leads to several times greater sensitivity than GHG forcing in West Africa, eastern China, South and Southeast Asia, northwestern South America, and Eastern Europe. The atmospheric energy balance, dynamical adjustment, and vertical structure of forcing, all contribute to the difference in the PE sensitivity to the two forcings. It is shown that the fast response primarily contributes to the greater-than-one aerosol-to-GHG ratios of the PE sensitivities, as for the mean precipitation. This is because of a stronger rainfall suppression effect induced by the GHG atmospheric forcing. We also find that the aerosol-to-GHG ratios of the PE sensitivities depend on the defined extreme precipitation indices. The aerosol-to-GHG sensitivity ratio is larger for more loosely defined PE, and it gradually converges to one for more severely defined PE. Our results further highlight the importance of considering the anthropogenic aerosol reduction in projecting the change in PE.

altmetric score

  • 12.5

author list (cited authors)

  • Lin, L., Wang, Z., Xu, Y., Fu, Q., & Dong, W.

citation count

  • 12

publication date

  • August 2018