High-Resolution Simulation of Observed Downburst Events

Downbursts, generated by collapsing thunderstorm updrafts, allow rapidly accelerating air to fall to the earth's surface, creating horizontally divergent winds that can exceed 60m/s. During the JAWS and NIMROD field campaigns, conducted in the 1970's and 80's over the central United States (US), considerable amount of downburst-related observational datasets were acquired. In-depth analyses of these unique datasets largely contributed towards our current understanding of this elusive atmospheric phenomenon. However, there is a definite opportunity to expand the existing observational database and comprehension of this topic.

Downbursts are not only intriguing from a fundamental atmospheric science perspective, they also have tremendous practical significance. Each year, straight line winds associated with downbursts cause substantial structural damage world-wide. Recently, several wind turbines in Minnesota and Iowa were either damaged or destroyed due to downbursts. Given that the majority of the US wind farms are being developed in the tornado alley of the US, a well-known breeding ground of downbursts, similar devastating events might become more frequent in the foreseeable future. This mere speculation justifies increased research in the arena of downburst characterization and modeling.

In this study, simulations of two observed downburst events using the Weather Research and Forecasting (WRF) model are conducted by performing seamless coupled simulations from synoptic-scale (order of ~100 km) down to sub-downburst-scale (order of ~100 m) flows. Observational data from ASOS, NEXRAD, and radiosondes are used for model validation. In addition, a detailed sensitivity study has been performed to understand the impacts of various physical parameterizations (e.g., microphysics) and spatial resolutions on the model results.