The Pearson correlation showed that there is a strong correlation between the global distribution of model bias in NetCRF and CF and it is significantly influenced by the LL and HL clouds.Ĭlouds play important role in the Earth’s climate and weather by controlling the radiative and hydrological balance. The geographical distributions of the model bias in the NetCRF, however, can be significant reaching up to ±40 Wm −2 depending on the location and atmospheric level. This inconsistent result may be attributed to the parameterization of the optical properties in the model. The AMIP-CanESM5 model performed reasonably well simulating the global mean cloud radiative forcing (CRF) with slight negative biases in the NetCRF at the TOA and surface that would be expected if the model has a positive bias in CF.
This is particularly prevalent over the low level (LL) marine stratocumulus/cumulus, convectively active tropical latitudes that are normally dominated by high level (HL) clouds and at the polar regions where all models showed negative, positive, and positive bias corresponding to these locations, respectively. Although the models captured well the climatological features of CF, they also exhibited a significant bias in CF reaching up to 40% over some geographical locations. There were no significant differences between AMIP and coupled versions of the model, but the COSP-based model-simulated data showed larger biases.
On average, all models have a negative bias in the total CF with global mean biases (MBs) of 2%, 2.4%, 3.9%, 6.4%, 5.6%, and 7.1% for the coupled-CanESM5, AMIP-CanESM5, COSP-AMIP-CanESM5, coupled-CanESM2, AMIP-CanESM2, and COSP-AMIP-CanESM2, respectively, indicating that the CanESM5 has a smaller MB.
The simulated shortwave, longwave, and net cloud radiative forcing using the AMIP version of the CanESM5 were also validated against satellite observations based on the recent CERES radiation satellite products. The CFs produced using the CALIPSO-COSP simulator based on the CanESMs data at their atmospheric (AMIP) configuration are also evaluated. In this study, the annual and seasonal climatology of cloud fraction (CF) and cloud type simulated by the Canadian Environmental System Models (CanESMs) version 5 (CanESM5) and version 2 (CanESM2) at their fully coupled and AMIP configurations were validated against the CALIPSO-GOCCP-based CF.
0 kommentar(er)
