9.4 Silhs: a Monte Carlo Interface between Clouds and Microphysics

Clouds exhibit spatial variability on a range of scales, and this variability influences microphysical processes such as drizzle formation (Pincus and Klein 2000; Larson et al. 2001). For this and other reasons, large-scale models may benefit from including the effects of subgrid cloud variability on microphysical process rates. For instance, feeding the fraction of a grid box occupied by cloud into a microphysics parameterization may help better characterize the moist parts of a grid box in which drizzle forms. Accuracy may also be improved by including information on within-cloud variability. Unfortunately, embedding information about subgrid cloud variability in microphysics schemes is laborious. In a model such as the Weather Research and Forecasting (WRF) model, which contains many microphysics schemes, embedding subgrid information separately in each microphysics scheme would be especially time consuming. In this paper, we test an alternative Monte Carlo method of driving microphysics schemes with information about subgrid variability. The method assumes that the distribution of subgrid variability is available for each grid box and time step. It then draws samples from that distribution. The sample points are fed, one by one, into the microphysics scheme in use. Finally, the process rates from each call to the microphysics are averaged and fed back into the simulation. This Monte Carlo metholodogy is related to the McICA method which has been used to drive radiative parameterizations (e.g. Barker et al. 2002; Pincus et al. 2003; Räisänen et al. 2004; Räisänen and Barker 2004; Räisänen et al. 2005; Pincus et al. 2006). For reasons to be discussed below, we call this method the Subgrid Importance Latin Hypercube Sampler (SILHS). SILHS separates the calculation of subgrid variability and the calculation of local microphysical process rates. Because of this, SILHS can include effects of subgrid variability without requiring modifications to the microphysics code itself. We test SILHS by performing month-long, 3D simulations of marine low clouds off the coast of Chile. These clouds were observed during the VAMOS Ocean-Cloud-Atmosphere-Land Study – Regional Experiment (VOCALS-REx) (Wood et al. 2011). We find that the use of SILHS noticeably increases cloud cover.