Improving GCM representation of convective anvil cloud microphysics by using ARM Raman lidar and cloud radar observations

Principal Investigator(s):
Qiang Fu, University of Washington

Co-Investigator(s):
Xiaohong Liu, University of Wyoming

Collaborator(s):
Jennifer M. Comstock, Pacific Northwest National Laboratory
Laura D. Riihimaki, Pacific Northwest National Laboratory
Hsi-Yen Ma, Lawrence Livermore National Laboratory

Cloud detrainments from deep convective plumes are often called “anvil clouds”, which are largely composed of ice particles. The area covered by the anvil clouds is much larger than the area occupied by the convective cores. The anvil clouds are the primary type of high cloud for modulating the Earth’s radiative energy balance, which play an important role in determining the cloud feedback. However, large uncertainty exists in the representation of convective anvil clouds in global climate models because of a lack of knowledge, e.g., in the partitioning of convective condensates into those precipitating out and those detrained to form anvil clouds in both current and future climate.

In the current convective scheme of the Community Atmospheric Model Version 5 (CAM5) that is the atmospheric component of the Community Earth System Model Version 1 (CESM1) for the CMIP5, the conversion of cloud water to rainwater is determined through a tunable parameter. The rainwater is removed immediately from the updrafts either as surface precipitation or through evaporation in the atmosphere. All detrained hydrometeor is linearly partitioned into liquid and ice depending on temperature in the range -5C < T < -35C with corresponding hydrometeor numbers estimated from the detrained mass by assuming a mean volume radius of 8 and 25 μm for water droplets and ice particles. The cloud liquid and ice water contents and their number concentrations of the convective detrainment are then used as input to the stratiform two-moment bulk cloud microphysics scheme. A two-moment diagnostic convective microphysics scheme was recently developed to improve the representation of convective clouds in CAM5 but assumptions and treatments of a few key processes need to be tested and improved.

The overall objective of this proposed effort is to improve the understanding of convective cloud microphysics processes and their representation in CAM5 by comparing the model simulations with observations. In particular we will derive the cloud phase and cloud ice water content by synthesizing ARM Raman lidar (RL) and cloud radar observations at the ARM SGP and TWP Darwin sites. Previous studies showed that simulated macro- and micro-physical properties of anvil clouds are sensitive to the parameterization of convective microphysical processes, and observed ice water content provides a critical constraint on these parameterizations. In the proposed research, we will categorize hydrometeors into liquid, ice, and mixed-phase clouds, precipitating ice particles, and rain by synthesizing the Raman lidar and cloud radar observations, with a focus on the anvil clouds associated with the deep convection. We will derive ice cloud water content based on the cloud radar retrievals that will be constrained by the RL-derived extinction coefficients. The Raman lidar is unique because the extinction coefficient can be accurately derived, which is the most important factor determining the cloud radiative effect. We will run the standard CAM5 and CAM5 with the two-moment convective cloud microphysics scheme in the single column mode, weather-hindcast mode (DOE Cloud-Associated Parameterization Testbed), and in the free-running mode. The simulated convective anvil cloud properties (e.g., cloud occurrence and phase and cloud ice water content) as well as their sensitivity to various microphysical processes and parameters will be tested with the above Raman lidar and cloud radar data over the SGP and TWP Darwin sites. The representation of convective cloud microphysics in CAM5 will be constrained and improved with much increased understanding of related convective and microphysical processes. The global climate impacts on simulated temperature, cloud radiative forcing and condensate contents, precipitation (including the stratiform fraction of total precipitation), and climate variability (e.g., ENSO, MJO, and diurnal cycle of precipitation) will be elucidated with the improved CAM5.