A Comparative Assessment of Global Distribution of Individual Cloud Types Retrieved by Space-borne Radar, Lidar and Passive Sensors

Clouds play a dominant role in modulating the Earth's energy balance and constitute a major component of the hydrological cycle. According to the IPCC AR5 report, clouds are considered to be contributing the largest uncertainty in future climate projections. Preparing an observation based unified cloud climatology is necessary to extensively evaluate cloud representation in climate models. Here, we use ten years (2007-2016) of active radar measurement (CloudSat), four years (2007-2010) of active radar-lidar combined measurement (CloudSat-Calipso) and twelve years (2000-2012) of passive remote sensing (ISCCP) data products to understand the global cloud distribution. Total cloud climatology inter-comparison was carried out by the GEWEX cloud group. Here, our objective is to compare climatology of individual cloud types available from different sensors so as to identify the similarities and differences among them. We find that global mean cloud fractions of cumulus, stratocumulus-stratus (combined), altocumulus, altostratus, nimbostratus, cirrus and deep convective clouds from ISCCP are 7.9%, 3.4%, 5.2%, 6.8%, 5.9%, 15.2%, 3.9% respectively. The corresponding numbers for Cloudsat and Cloudtsat-Calipso are 4.7% and 19.2%, 15.9% and 10.8%, 6.1% and 9.8%, 7.1% and 10.4%, 4.1% and 4.7%, 5.7% and 12.9%, 1.4% and 1.2% respectively. All three data sets are consistent in retrieving nimbostratus cloud distribution. In case of deep-convective clouds, ISCCP retrieval shows its presence in ITCZ as well as subtropical storm-tracks. However, deep convective cloud distribution is mostly confined in the ITCZ in CloudSat and CloudSat-Calipso dataset. Cirrus cloud distribution is similar in Cloudsat-Calipso and ISCCP. Global mean cloud fractions are much lower for cirrus and cumulus for CloudSat. Our results facilitate understanding the strengths and weaknesses of the sensors and provide a context to interpret these data more meaningfully.