Relative Humidity and Cloud Phase Distributions at 0°C to -40°C over the Southern Ocean: Scale Aware Comparisons between in-Situ Observations and CAM5 Simulations

Mixed-phase clouds have large impacts on Earth's climate system through cloud radiative effects. A key step to evaluate the representations of mixed-phase clouds in global climate models (GCMs) is to conduct comparisons between simulations and observations on comparable spatial scales. However, the large gap between the spatial resolutions of aircraft-based in-situ observations (e.g., ~100 m horizontal scale) and the horizontal grid box scales of GCMs (e.g., ~25 km to 100 km) produces a major obstacle for conducting direct comparisons between these two types of datasets.

In this work, we will present scale-aware comparisons between airborne in-situ observations over the Southern Ocean and simulations from the NCAR Community Atmosphere Model Version 5 (CAM5). The coexistence of ice particles and supercooled liquid water often imposes significant challenges to observational techniques due to the overlapping size ranges of various cloud hydrometeors, and thus it is difficult to rely on a single type of instrument to detect thermodynamic phases. To resolve this problem, we developed a comprehensive method that separates three cloud phases, ice, liquid and mixed-phase, based on a suite of instrumentation onboard the NSF Gulfstream-V research aircraft. For model-observation comparisons, in-situ observations were spatially averaged to various scales (i.e., 2, 5, 10, 20, and 40 km), which were compared with CAM5 simulations (i.e., 0.5 degree horizontal grid spacing) spatially interpolated onto the research aircraft flight track. Our scale-aware comparisons examined the sensitivity of comparison results to the horizontal scales of the two datasets. The comparison results to be shown include occurrence frequencies of three cloud phases, relative humidity distribution, ice and liquid supersaturation, ice and liquid water content, etc. Suggestions on cloud micro- and macrophysics parameterizations for ice and mixed-phase clouds in CAM5 model will be given as a result of our comparisons.