A-Train active views of global MODIS cloud regimes

In-depth understanding of the role of clouds on the water and energy budget requires meaningful classification of the various cloud mixtures encountered in different parts of the globe throughout the year. Cloud observations are performed from many space-based observational platforms with a variety of remote sensing algorithms applied to both passive and active sensors. Passive observations have been available longer, generally have much better spatial coverage, and have been recently used in a novel way to classify and group mesoscale cloud mixtures into “cloud regimes”, or “weather states”. In our presentation, we will show that when applying clustering analysis on 10 years of coincident MODIS cloud top pressure and cloud optical thickness retrievals we find that 10 regimes provide a good representation of the distinct ways cloud mixtures organize around the globe. A multitude of other datasets is employed to shine light on the nature of these cloud regimes beyond what can be gleaned from the passive observations alone. We find especially illuminating datasets coming from A-Train active observations, namely those from CALIOP/CALIPSO and CPR/CloudSat which can be spatiotemporally collocated with MODIS-Aqua cloud regime occurrences. We explore a variety of products from LARC's CCCM dataset, which samples and aggregates in a systematic fashion a variety of CloudSat and CALIPSO retrievals, to gain deeper insight on the MODIS cloud regimes. We examine various aspects of cloud vertical structure such as mean vertical location of their highest tops and lowest bases, vertical distribution of cloud fraction and water content, and number of simultaneously occuring cloud layers. When we associate our MODIS-Aqua cloud regimes with traditional cloud type classifications made possible by active observations, we find their makeup to be quite diverse, but largely consistent with expectations based on interpretation of the passive observations, namely the “mean” joint histogram representing each regime and the geographical distribution of their occurrences. Overall, the employment of active observations corroborates the robustness of the MODIS regime derivation technique, reaffirms the regimes' fundamental nature as appropriate building blocks for cloud system classification, clarifies their association with standard cloud types, and constitutes a critical intermediate step toward unveiling their distinct radiative and hydrological signatures.