A Comparison of Cloud Classification Methodologies: Differences between Cloud and Dynamical Regimes

Classifications of cloud data into Cloud Regimes (CR) and compositing based on meteorological parameters, Dynamic Regimes (DR), are often used in the statistical analysis of cloud properties. We apply the Self-Organizing Map technique to International Satellite Cloud Climatology Project (ISCCP) D1 joint histograms of optical depth versus cloud top pressure to produce CR. We then apply the same methodology to ERA–Interim pressure vertical velocity output to produce a set of DR. The CR created improved the separation between high-level cloud regimes compared to previous work. Composites of IS2CCP joint histogram data using the DR produce coherent groupings similar to those in the CR scheme, highlighting the utility of the DR, particularly in regions of ascent. Both classifications display coherent geographical patterns and reproduce relationships between vertical velocity and cloud properties. However, the CR produces more coherent clusters with higher intra-cluster similarity and a greater range of independent cloud classes. Two independent tests of the classifications compare composites using ISCCP FD output and CloudSat observations. The regional variability of longwave cloud radiative effect for particular nodes is significantly higher in the DR than the CR scheme suggesting a poorer classification. Composite mean CloudSat reflectivity–altitude joint histograms represent all major cloud types in the CR scheme, while the DR grouping is less coherent and misses classes. Our analysis suggests that the CR scheme produces a more meaningful classification than the DR scheme. Contingency table analysis indicates a low association between classifications, suggesting classifications combining both schemes will have additional value.