Assessing the Impact of Stochastic Cloud Microphysics in High-Resolution Models Using GOES-16 Infrared Brightness Temperatures

In this study, we use cloud-sensitive infrared brightness temperatures from the GOES-16 Advanced Baseline Imager (ABI) to explore the impact of adding stochastic perturbations to key microphysical process rates in the Thompson-Eidhammer microphysics scheme. In particular, the Stochastic Parameter Perturbation (SPP) method developed by Berner et al. was used to add stochastic perturbations with realistic spatial and temporal correlation patterns to various parameters in the Thompson-Eidhammer scheme during WRF model simulations performed using a domain configuration similar to that used by the HRRR model. Sensitivity experiments have been performed in which perturbations were added to the shape parameter of the gamma distribution describing the cloud droplet number spectra, to grid-resolved vertical velocities that control the activation of cloud and ice condensation nuclei, as well as to aerosols influencing the ice nuclei concentration. Preliminary results indicate that the stochastic perturbations can exert a large influence on the cloud field. Versions of the Thompson-Eidhammer scheme in which stochastic perturbations were applied to one or more parameters will be included in the Center for the Analysis and Prediction of Storms (CAPS) ensemble during the 2018 Hazardous Weather Testbed (HWT) Spring Experiment. This paper will discuss results from the HWT simulations as well as our preliminary sensitivity experiments. The accuracy of the forecast cloud field will be assessed through detailed comparisons of observed and simulated GOES-16 ABI brightness temperatures using a variety of statistical techniques. The development of a stochastic version of a cloud microphysics scheme is important because it promotes a more realistic representation of the underlying uncertainty in various microphysical parameters, many of which are given constant values even though observations indicate that these parameters may vary over several orders of magnitude depending upon the atmospheric conditions. This study represents an initial step toward development of cloud microphysics schemes containing realistic perturbations.