Evaluation of cloud microphysical parameterizations in cloud-resolving model simulations using the ARM observations

Description

Clouds modulate the distribution of energy and water within the atmosphere and regulate the hydrological cycle. Cloud microphysical parameterizations are critical for the representation of cloud microphysical properties in both cloud-resolving and climate models. In this study, we analyze the capabilities of a cloud-resolving model (CRM) with advanced bulk microphysics schemes to simulate the microphysical properties and evolution of convective clouds and anvil cirrus over the Southern Great Plains (SGP) site in the midlatitudes and Kwajalein Atoll in the tropics. To evaluate the simulated cloud properties, we use observations from the Atmospheric Radiation Measurement (ARM) Climate Research Facility 1997 Summer Intensive Observation Period at the SGP site and the Kwajalein Experiment (KWAJEX) field campaign. The CRM simulations are evaluated using, in particular, precipitation records, radiative fluxes, and radar reflectivity values from the ARM millimeter-wavelength cloud radar (MMCR) and the Kwajalein precipitation radar.

Preliminary analysis of the ARM SGP case shows that precipitation events during this period are well captured by the model, but the outgoing longwave radiation (OLR) is considerably underestimated and the model generates too much high cloud when compared with the MMCR observations. We find that the excess cloud production is due in part to the microphysical parameterizations but that the periodic lateral boundary conditions of the CRM also play an important role. Compared to the cloud radar reflectivity histograms from the ARM MMCR observations, the simulated cloud radar reflectivity reproduces the shape and the mode of the reflectivity histograms at levels above 11 km. The agreement deteriorates at lower levels, presumably because more complicated microphysical processes are involved. In our study we focus especially on the causes of the overproduction of ice and high-level clouds in the CRM simulations. Improvements of the ice microphysics scheme and resulting impacts on the simulation are presented.