An investigation of microphysics and subgrid‐scale variability in warm‐rain clouds using the A‐Train observations and a multiscale modeling framework. Issue 14 (24th July 2017)

• Record Type: Journal Article
• Authors: Takahashi, Hanii; Lebsock, Matthew; Suzuki, Kentaroh; Stephens, Graeme; Wang, Minghuai
• Is Part Of: Journal of geophysical research. Volume 122:Issue 14(2017)
• Page Start: 7493
• Page End: 7504
• Publication Date: 2017-07-24
• Subjects: warm‐rain clouds -- subgrid‐scale variability -- multiscale modeling framework; Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1002/2016JD026404
• Languages: English
• ISSNs: 2169-897X

Abstract

Abstract: A common problem in climate models is that they are likely to produce rain at a faster rate than is observed and therefore produce too much light rain (e.g., drizzle). Interestingly, the Pacific Northwest National Laboratory (PNNL) multiscale modeling framework (MMF), whose warm‐rain formation process is more realistic than other global models, has the opposite problem: the rain formation process in PNNL‐MMF is less efficient than the real world. To better understand the microphysical processes in warm cloud, this study documents the model biases in PNNL‐MMF and evaluates warm cloud properties, subgrid variability, and microphysics, using A‐Train satellite observations to identify sources of model biases in PNNL‐MMF. Like other models PNNL‐MMF underpredicts the warm cloud fraction with compensating large optical depths. Associated with these compensating errors in cloudiness are compensating errors in the precipitation process. For a given liquid water path, clouds in the PNNL‐MMF are less likely to produce rain than are real‐world clouds. However, when the model does produce rain it is able to produce stronger precipitation than reality. As a result PNNL‐MMF produces about the correct mean rain rate with an incorrect distribution of rates. The subgrid variability in PNNL‐MMF is also tested, and results are fairly consistent with observations, suggesting that the possible sources of model biases are likely to be due to errors in its microphysics or dynamics rather than errors in the subgrid‐scale variability produced by the embedded cloud resolving model. Key Points: To compensate for the smaller net cooling effect, warm clouds in PNNL‐MMF are optically thicker and warmer than the observations The biases in simulated clouds result from a combination of microphysical and dynamical errors rather than subgrid scale errors The PNNL‐MMF produces about the correct mean rain rate to satisfy atmospheric energy balance with an incorrect distribution of rates