Research On Cloud Macro And Micro Physical Properties Based On CloudSat/CALIPSO Satellite Data And Its Application In FAMIL2

The representation of cloud and cloud-radiation interaction in current general circulation models(GCMs)remains a vast challenge,which contributes to one of the largest uncertainties in the simulation of cloud feedback process under global warming background.The problem stems from the fact that cloud radiation depends primarily on a large variety of cloud properties,which to some extent dominate the simulation performance of cloud radiative effect(CRE).To cope with this problem,it is highly necessary to extend our knowledge of cloud macro-and microphysical properties with observation.By using 2007-2010 four-year Cloud Sat/CALIPSO satellite data,the similarities and differences of simulated cloud fraction as well as the sources of bias in reanalyses are investigated.The CRE simulations of different cloud properties in climate models are also evaluated.Moreover,realistic distributions of cloud macro-and microphysical properties are explored based on satellite observations.Finally,with a fitting scheme of the observed critical relative humidity proposed,a RH-based cloud scheme has been improved and implemented in FAMIL2 climate model.Main conclusions are summarized as follows:(1)All three reanalyses can basically capture the horizontal pattern and vertical structure as in Cloud Sat/CALIPSO,they show considerable biases against satellite retrievals and differ from one another as well.ERA-Interim and JRA-55 perform better for low and mid clouds,but exhibit apparent underestimation for high clouds,whereas MERRA-2 succeeds in representing high clouds but dramatically underestimates low and mid clouds.Results also show that,the underestimation of cloud fraction is presumably due to a small specification or prediction of subgrid-scale variability of moisture in models,where clouds are less likely to form at the same humidity.(2)The net CREs are overestimated in almost all CMIP6 models,which is due to the underestimation of long-wave(LW)CREs as well as the overestimation of short-wave(SW)CREs.In addition,it is found GCMs commonly overestimate the internal SW CRE for clouds that are optically thick and morphologically large,whereas lead to an underestimation for the remaining cloud types.The simulations of frequency of occurrences of these clouds are on the opposite.The similar conclusion is reached for LW CRE as well.The models underestimate the RFO of thick ice clouds even more pronounced than those for warm clouds.The biases of cloud occurrence dominate over the internal CRE biases in a decomposition of CRE errors.(3)The derived critical relative humidity from Cloud Sat/CALIPSO exhibits distinctive vertical structures at different latitudes.The diagnosed critical relative humidity shows bimodal characteristics in the vertical direction,which first decreases with altitude and then increases upward.Moreover,the observed critical relative humidity in the middle and low latitudes is generally smaller than those in the high latitudes,and the smallest values are presented in the subtropical areas.Larger critical relative humidity means smaller subgrid-scale variability.Considering its distributions by altitude and latitude,the fitting scheme combining two exponential functions captures the observed distribution characteristics of critical relative humidity very well,which then improves a RH-based cloud scheme.(4)Cloud horizontal inhomogeneity parameter V exhibits obvious spatial distribution characteristics at different altitudes and latitudes,which is relatively insensitive to seasonal variability,land-sea distribution,and diurnal cycle.Further study shows that the horizontal inhomogeneity parameter V is in a good relation with different cloud types,and cloud inhomogeneity is found to be generally weaker in warm clouds than ice clouds,weaker in stratiform clouds(e.g.,Ns,Ci)than cumuliform clouds(e.g.,Cu).(5)Although cloud fraction shows clear seasonal variations,the cloud microphysical properties do not.However,they do show significant variations in the vertical direction.For liquid clouds,both cloud mass and number concentration gradually decrease with height,leading to the effective radius being nearly uniformly spread in the range of 8–14?m.For ice clouds,the cloud mass and effective radius decrease with height,whereas the number concentration is nearly uniform in the vertical.Besides,the cloud microphysical properties show remarkable differences among different cloud types.Cloud mass and number concentration are larger in cumuliform clouds,whereas smaller in cirrus clouds.By comparing cloud properties among the Tibetan Plateau,East China and the western North Pacific,results show the values are overall smaller for liquid clouds but larger for ice clouds over the Tibetan Plateau.(6)With the parameterization of critical relative humidity in the former analysis,a RH-based cloud scheme is improved and implemented in FAMIL2 climate model.Results of AMIP experiments show that the modified cloud scheme significantly improves the negative bias of cloud fraction in FAMIL2,which is largely underestimated or even missed with the original cloud scheme,especially for mid and high-level clouds over low latitudes and low-level clouds over the eastern coast of the ocean.In addition,results of aqua-planet experiments show that the global averaged net CRE change is positive in FAMIL2 for+4K SST experiments,which implies that clouds play a role in accelerating global warming.Meanwhile,the calculated climate sensitivity?of original and modified schemes is 0.632 and 0.605 K m² W^-1,respectively,which is closer to those obtained by other models.