Inferring the Role of Ice, Liquid and Thermodynamic Phase Shifts in the Inter Annual Low-Cloud Optical Depth Feedback

Progress in narrowing the range of model-predicted climate sensitivity has been hampered by disparate responses of clouds to global warming. Recent developments have shown that the shortwave cloud optical depth feedback is likely too negative in the mid- and high latitudes across an ensemble of models. A recent emergent constraint on the low cloud optical depth in models has been identified, however, a critical piece of missing information is the dominant physical mechanism responsible for the changes in low cloud optical depth with global warming in the mid- and high latitudes. Here, we develop a novel method to separate the contributions of ice, liquid and thermodynamic phase shifts these clouds to the total mid-latitude low cloud optical depth as a function of cloud top temperature. The method is applied to the Moderate Resolution Imaging Spectroradiometer (MODIS) Aqua daily Level 3 dataset in the tropics, subtropics and mid-latitudes during different seasons and over different surface types. The results confirm those of previous studies that have used International Satellite Cloud Climatology Project (ISCCP) and models participating in the third phase of the Cloud Model Intercomparison Project (CMIP3) to examine cloud optical depth variations with temperature. Namely, cloud optical thickness decreases with temperature at warmer mixed-phase cloud temperatures, whereas optical thickness increases with temperature at colder mixed-phase cloud temperatures over land in the mid-latitudes. The novel decomposition method attributes the increases in cloud optical depths with temperature to thermodynamic phase shifts and the decreases in cloud optical depths with temperature primarily to changes in cloud liquid water path. The methodology reveals that the contribution of ice water path on cloud optical depth variations with temperature only exceeds that from liquid water path at temperatures below -25°C.