Quantitative evaluation of three-dimensional cloud radiative transfer effect using multi-satellite data

This study evaluates three-dimensional (3D) cloud radiative transfer effects (CRE) quantitatively using a 3D radiative transfer model, MCstar, and A-train collocated CloudSat/CPR and Aqua/MODIS cloud data. The 3D extinction coefficients are constructed by a newly devised Minimum cloud Information Deviation Profiling Method (MIDPM) (Okata et al., 2017) that extrapolates the scaled extinction coefficient profiles derived from CPR radar profiles (Marchand et al., 2008) and MODIS cloud optical thickness (Nakajima and King, 1990; Nakajima and Nakajima, 1995; Kawamoto et al., 2001) at nadir into off-nadir regions within MODIS swath. The method is applied to water clouds, for which the 3D radiative transfer ( RT) simulations are performed.

In this study, the radiative fluxes thus simulated are compared to those obtained from CERES as a way to validate the MIDPM-constructed clouds and our 3D radiative transfer simulations for 97 and 92 cloud cases on July 2, 2007 and July 1, 2009, respectively, in South California. The results show that the simulated solar and thermal radiative fluxes against CERES data values within 0.37~60Wm^-2 and 0.36~20Wm^-2 respectively.

One of the large biases arises from the 1D assumption for cloud property retrievals. While 3D-RT effects on the angular variability of cloud optical thickness are evident, it is concluded that a cloud effective radius bias is itself an important contributor to this variability ( Lang et al., 2015) . In this study, we analyze these cloud properties biases derived from cloud heterogeneity using multi satellite imager data like Second generation GLobal Imager ( SGLI) with a high spatial resolution ( 250 m) within MODIS pixel with spatial resolution (1km).

Furthermore, it is effective to analyze the characteristics which the cloud morphology is particularly sensitive to the solar insolation geometry as one of the 3D-RT effects to investigate these flux biases details (Okata et al., 2017). The 3D-RT effects are classified into three types which cannot be estimated by the 1D-RT corresponding to different simple three morphology types.

We performed these simulations for the satellite-constructed real cloud cases, and quantitatively classified 64 cases into the characteristic 3D cloud RT effects and morphologies. In addition to, the biases against CERES values are analyzed based on this quantitative classification method.