Interregional differences in MODIS-derived cloud regimes

Cloud regimes based on histogram clustering offer a potentially useful tool for observational analysis of clouds. The utility of the regimes depends on their ability to identify cloud structures that are associated with distinct meteorological conditions. In this study, active remote sensing observations from CloudSat and CALIPSO are binned according to the cloud regimes derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument. The occurrence of CloudSat radar reflectivity by altitude, as well as CloudSat-CALIPSO retrievals of cloud cover, precipitation occurrence, cloud radiative effect, and multilayered cloud structure are analyzed for each MODIS cloud regime and for different regions. While the picture of the regimes given by CloudSat and CALIPSO is generally consistent with that derived from MODIS, substantial region-to-region variability is found within the regimes. The regimes constrain the shortwave cloud radiative effect well, while the longwave effect and the precipitation occurrence exhibit more variability. The joint distributions of radar reflectivity and altitude also reveal differences in the structure of clouds within each regime. Thus, it appears that there is region-dependent variability within each regime, resulting from the different meteorological environments. Among the major differences in the cloud structure are cloud top height in convective clouds and the number of distinct cloud layers in boundary layer clouds. Thus, passive optical sensors appear limited in their ability to characterize clouds and assign them to distinct regimes. The differences can be used to estimate the cloud regime inherent variability for studies that use them as a proxy for the climate effects of clouds.