A review of cloud top height and optical depth histograms from MISR, ISCCP, and MODIS

There are notable differences in the joint histograms of cloud top height and optical depth being produced from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Multiangle Imaging Spectro‐Radiometer (MISR) and by the International Satellite Cloud Climatology Project (ISCCP). These differences have their roots in the different retrieval approaches used by the three projects and are driven largely by responses of the retrievals to (1) stratocumulus (or more broadly low‐level clouds under temperature inversions), (2) small (subpixel) or broken low‐level clouds, and (3) multilayer clouds. Because each data set has different strengths and weakness, the combination tells us more about the observed cloud fields than any of the three by itself. In particular, the MISR stereo height retrieval provides a calibration insensitive approach to determining cloud height that is especially valuable in combination with ISCCP or MODIS because the combination provides a means to estimate the amount of multilayer cloud, where the upper cloud is optically thin. In this article we present a review of the three data sets using case studies and comparisons of annually averaged joint histograms on global and regional scales. Recommendations for using these data in climate model evaluations are provided.