Warm, tropical cumulus clouds represent one of the biggest sources of uncertainty in climate sensitivity to anthropogenic forcing. Information useful for gaining a better understanding about such clouds is relatively plentiful from satellites. But warm tropical clouds, such as those found in the trades, are often subpixel and due to the 1-D nature of the radiative transfer calculations used in retrieving cloud drop effective radius and cloud optical depth may introduce 3-D effects into the calculations, leading to systematic errors in satellite retrievals. To date, all validation for these satellite retrievals of cloud properties have been for stratiform clouds, not for cumulus. Here, we use in-situ aircraft measurements from RICO (“Rain In shallow Cumulus over the Ocean”) for these cumuli, and investigate how they compare to satellite retrievals from MODIS (“Moderate Resolution Imaging Spectroradiometer”). In doing so, the vertical structure of the cloud drop effective radius and liquid water content of these trade-wind cumuli are also investigated from the FSSP-100 and 2D-C cloud probes aboard the C-130 aircraft flown during RICO.
The constructed profiles show a general increase in cloud drop effective radius with increasing height up to around 2000 m altitude, and then start to show a decrease with height beyond that point, though with less samples. The liquid water content profile shows similar behavior. When using the aircraft data to compare to MODIS 2.1 μm band retrievals, aircraft data shows a much smaller mean effective radius than the MODIS retrievals. The MODIS 3.7 μm band shows an effective radius distribution closer to that of the aircraft. The spatial heterogeneity metric, Hσ, is also investigated for the MODIS retrievals. Analysis of Hσ shows that MODIS retrievals for more homogeneous clouds have a lower mean effective radius compared to heterogeneous retrievals, while cloud optical depth retrievals for homogeneous clouds have a higher mean optical depth than heterogeneous clouds. This overestimation of effective radius and underestimation of optical depth is in line with what is expected when 3-D radiative transfer effects play a major role, and shows that analysts should use caution when using such satellite retrievals.