How Does Our Definition of Anvil Cloud Fraction Affect Its Trend with Warming in Cloud-Resolving Simulations?

The tropical upper troposphere is home to extensive cirrus clouds detrained from thunderstorms. Due to the powerful shortwave and longwave radiative effects associated with anvil clouds, any change in anvil cloud amount with climate warming will cause a cloud-amount climate feedback. Therefore, it is necessary to better understand how anvil cloud amount changes with surface warming. In this study, we use data from idealized cloud-resolving simulations to address one aspect of that problem: How does one’s definition of anvil cloud fraction affect its trend with climate warming?

Peak cloud fraction measures the horizontal extent of clouds at the vertical level of maximum cloudiness. Although it is the traditional measure of anvil clouds, peak cloud fraction fails to account for the fact that clouds do not fully overlap with one another in the vertical direction. On the other hand, total cloud fraction measures what fraction of the upper troposphere contains a cloud anywhere within its vertical column. Since we wish to understand what area of the upper troposphere is subject to a cloud radiative effect, total cloud fraction is more appropriate for investigating changes in anvil-cloud area.

Using data from idealized cloud-resolving simulations, we quantify both definitions of anvil cloud fraction and assess their trends with warming. Consistent with earlier work, we find that peak cloud fraction declines with warming due to increasing upper-tropospheric static stability. However, the trend in total cloud fraction may be positive, neutral, or negative depending on surface temperature and model settings. This suggests that the physical mechanisms controlling the two types of cloud fraction are distinct and that the anvil cloud area climate feedback may not be readily constrained by process-resolving models.