摘要
This study examines cloud radiative forcing （CRF） in the Asian monsoon region （0° 50°N, 60° 150°E） simulated by Intergovernmental Panel on Climate Change （IPCC） Fourth Assessment Report （AR4） AMIP models. During boreal winter, no model realistically reproduces the larger long-wave cloud radiative forcing （LWCF） over the Tibet Plateau （TP） and only a couple of models reasonably capture the larger short-wave CRF （SWCF） to the east of the TP. During boreal summer, there are larger biases for central location and intensity of simulated CRF in active convective regions. The CRF biases are closely related to the rainfall biases in the models. Quantitative analysis further indicates that the correlation between simulated CRF and observations are not high, and that the biases and diversity in SWCF are larger than that in LWCF. The annual cycle of simulated CRF over East Asia （0°-50°N, 100°-145°E） is also examined. Though many models capture the basic annual cycle in tropics, strong LWCF and SWCF to the east of the TP beginning in early spring are underestimated by most models. As a whole, GFDL-CM2.1, MPI-ECHAM5, UKMO-HadGAM1, and MIROC3.2 （medres） perform well for CRF simulation in the Asian monsoon region, and the multi-model ensemble （MME） has improved results over the individual simulations. It is suggested that strengthening the physical parameterizations involved over the TP, and improving cumulus convection processes and model experiment design are crucial to CRF simulation in the Asian monsoon region.
This study examines cloud radiative forcing （CRF） in the Asian monsoon region （0° 50°N, 60° 150°E） simulated by Intergovernmental Panel on Climate Change （IPCC） Fourth Assessment Report （AR4） AMIP models. During boreal winter, no model realistically reproduces the larger long-wave cloud radiative forcing （LWCF） over the Tibet Plateau （TP） and only a couple of models reasonably capture the larger short-wave CRF （SWCF） to the east of the TP. During boreal summer, there are larger biases for central location and intensity of simulated CRF in active convective regions. The CRF biases are closely related to the rainfall biases in the models. Quantitative analysis further indicates that the correlation between simulated CRF and observations are not high, and that the biases and diversity in SWCF are larger than that in LWCF. The annual cycle of simulated CRF over East Asia （0°-50°N, 100°-145°E） is also examined. Though many models capture the basic annual cycle in tropics, strong LWCF and SWCF to the east of the TP beginning in early spring are underestimated by most models. As a whole, GFDL-CM2.1, MPI-ECHAM5, UKMO-HadGAM1, and MIROC3.2 （medres） perform well for CRF simulation in the Asian monsoon region, and the multi-model ensemble （MME） has improved results over the individual simulations. It is suggested that strengthening the physical parameterizations involved over the TP, and improving cumulus convection processes and model experiment design are crucial to CRF simulation in the Asian monsoon region.
基金
supported by the CAS project under Grant No. KZCX2-YW-Q11-01
the Major State Basic Research Development Program of China under Grant No. 2006CB403607
the National Natural Science Foundation of China (Grant Nos.40523001, 40821092, 40875034)
