Out-of-cloud convective turbulence: Estimation method and impacts of model resolution

Convectively induced turbulence (CIT) poses both a serious threat to aviation operations and a challenge to forecasting applications. CIT generation and propagation processes occur on scales between 10 and 1000m and therefore are best treated with high-resolution cloud-resolving models. However, high-resolution model simulations are computationally expensive, limiting their operational use. In this study, summertime convection in the North Dakota region is simulated over a 1-week period using a variety of model setups that are similar to those utilized in operational and research applications. Eddy dissipation rate and Ellrod index, both popular turbulence metrics, are evaluated across various model resolutions and compared with pilot reports from aircraft. The Ellrod index was found to be extremely sensitive to model resolution and overestimated turbulence intensity. The variability of turbulence values with respect to model resolution and distance away from convection is also examined. Turbulence probability was found to be the greatest when farther than 20 mi (32.2 km) away from convective cores. Model resolution was found to influence the intensity of predicted turbulence, and the model setup with the highest horizontal and vertical resolution predicted the highest turbulence values. However, the influence on turbulence intensity of vertical resolution and convective properties, such as storm depth, was found to be minimal for 3-km horizontal grid spacing.