Ice cloud properties in ice-over-water cloud systems using Tropical Rainfall Measuring Mission (TRMM) visible and infrared scanner and TRMM Microwave Imager data

A multilayered cloud retrieval system (MCRS) is updated and used to estimate ice water path in maritime ice-over-water clouds using Visible and Infrared Scanner (VIRS) and Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) measurements acquired over the Tropics between January and August 1998. Lookup tables of top-of-atmosphere 0.65-mm reflectance are developed for ice-over-water cloud systems using radiative transfer calculations for various combinations of ice-over-water cloud layers. The liquid and ice water paths, LWP and IWP, respectively, are determined with the MCRS using these lookup tables with a combination of microwave (MW), visible (VIS), and infrared (IR) data. LWP, determined directly from the TMI MW data, is used to define the lower-level cloud properties to select the proper lookup table. The properties of the upper-level ice clouds, such as optical depth and effective size, are then derived using the Visible–Infrared Solar-infrared Split-Window technique (VISST), which matches the VIRS IR, 3.9 mm, and VIS data to the multilayer cloud lookup table reflectances and a set of emittance parameterizations. Initial comparisons with surface-based radar retrievals suggest that this enhanced MCRS can significantly improve the accuracy and decrease the IWP in overlapped clouds by 42 and 13% compared to using the single-layer VISST and an earlier simplified MW–VIS–IR (MVI) differencing method, respectively, for ice-over-water cloud systems. The tropical distribution of ice-over-water clouds is the same as derived earlier from combined TMI and VIRS data, but the new values of IWP and optical depth are slightly larger than the older MVI values and exceed those of single-layered clouds by 7 and 11%, respectively. The mean IWP from the MCRS is 8–14% greater than that retrieved from radar retrievals of overlapped clouds over two surface sites, and the standard deviations of the differences are similar to those for single-layered clouds. Examples of a method for applying the MCRS over land without MW data yield similar differences with the surface retrievals. By combining the MCRS with other techniques that focus primarily on optically thin cirrus over low water clouds, it will be possible to more fully assess the IWP in all conditions over ocean except for precipitating systems.