Assessment of Springtime Cloud Forecasts from Two Mesoscale Models

Abstract:

Boundary layer clouds can affect the fate and transportation of atmospheric constituents in the atmosphere. Accurate dispersion modeling should include scavenging and transformations of constituents from boundary layer clouds that could lead to washout or altered constituents upon droplet evaporation. These prescribed interactions require accurate modeling of clouds within the boundary layer. Here we use ceilometer measurements taken at the Savannah River Site (SRS) in western South Carolina to validate cloud base height (CBH) and cloud fraction (CF) daily springtime (April to June) forecasts derived from two operational mesoscale models (WRF and RAMS). Forecast assessments of CBH and CF were completed for both models at 15-minute intervals. RAMS performed better at forecasting CF while WRF better forecasted CBH. Monthly averages show that both models generally underpredicted nighttime CF while RAMS better predicted CF during the daytime. RAMS predicted nighttime CBH better than WRF but WRF predicted CBH much better than RAMS in the daytime convective cases in May and June. RAMS daytime CBH was usually higher than the observed CBH and WRF nighttime CBH was generally much lower than the observed CBH. CBH observations showed trends of falling CBH in the morning hours. Because WRF started with lower CBH it was better able to simulate the increase in CBH during the daytime. RAMS resolved a similar increase in CBH but failed to sufficiently develop the drop in CBH that occurred in the observations. However, the excessively low WRF CBH during the overnight period points to an increase in clouds within the boundary layer that would affect dispersion. A combination of both models seems be important for characterizing dispersion depending on time of day. Further research needs to be done using different model configurations to improve model accuracy of cloud macrophysical properties.