Critical evaluation of the ISCCP simulator using ground-based remote sensing data

Description

Given the known shortcomings in representing clouds in Global Climate Models (GCM), comparisons with observations are critical. The International Satellite Cloud Climatology Project (ISCCP) diagnostic products provide global descriptions of cloud-top pressure and column optical depth that extend over multiple decades. The necessary limitations of the ISCCP retrieval algorithm require that before comparisons can be made between model output and ISCCP results, the model output must be modified to simulate what ISCCP would diagnose under the simulated circumstances. We evaluate one component of the so-called ISCCP simulator in this study by comparing ISCCP and a similar algorithm with various long-term statistics derived from ground-based sensors at the Atmospheric Radiation Measurement (ARM) Climate Research Facility’s Southern Great Plains (SGP) site. We find that were a model to simulate the cloud radiative profile with the same accuracy as can be derived from the ARM data, then the likelihood of that occurrence being placed in the same cloud-top pressure and optical depth bin as ISCCP of the nine bins that have become standard is on the order of 50%. While the ISCCP simulator clarified interpretation of the comparisons, we find minor discrepancies due to the parameterization of cloud-top pressure in the ISCCP simulator. The primary source of error seems to be related to discrepancies in visible optical depth that are not accounted for in the ISCCP simulator. We show that the optical depth discrepancies are largest when the assumptions necessary for plane parallel radiative transfer optical depths retrievals are violated.