Cold-Air Damming Erosion: Physical Mechanisms and Numerical Prediction

Forecasters in the southeastern U.S. must contend with the challenge of Appalachian cold-air damming (CAD), which can affect several important weather parameters, including wintry precipitation, cloud and sky conditions, and the location of convection at the periphery of the CAD region. Despite improvements in numerical weather prediction (NWP), forecasters have noted a tendency for the cold dome in model forecasts to erode too quickly. Building on earlier work, we here investigate this phenomenon with the Weather Research and Forecasting (WRF) model. Our goals are to (i) identify the most important physical processes responsible for CAD erosion, (ii) examine the representation of these processes by numerical models, and (iii) identify model configurations that optimize forecast accuracy in these situations.

In earlier, unpublished research, we investigated the physical mechanisms of CAD erosion within the context of the bulk Richardson number, evaluated in the inversion layer near the top of the CAD cold dome. Processes that increase shear or reduce the strength of the inversion can potentially promote mixing and CAD erosion. Examination of the erosion of 90 observed CAD events indicates that cold advection aloft is the most effective erosion process. This often happens in conjunction with a cold-frontal passage, or cold advection in the wake of a coastal cyclone. Near-surface warming associated with solar heating is an effective erosion mechanism in cases of broken or scattered cloud cover. Lower-tropospheric divergence within the cold dome reduces the depth of the cold air, and can be associated with subsidence and drying within the dome.

Several aspects of model representation of CAD erosion processes are investigated using WRF model simulations of a case study from October 2002. The sensitivity to cloud-radiation interaction, the planetary boundary layer scheme, and the shallow mixing component of the convective scheme are evaluated. These results indicate that while all of these model components can affect predictions of CAD erosion, the shallow mixing component in the convective scheme may be most responsible for overly aggressive CAD erosion in operational model forecasts.