Dataset for "Microburst Detection with the WRF Model: Effective Resolution and Forecasting Indices"
Abstract: Microbursts are meteorological phenomena in the lower troposphere which can produce damaging surface winds and pose a severe risk to aircraft flying close to the ground. As these events usually span less than 4 km and 15 min, the spatiotemporal resolution is a challenge for numerical simulations. Although research of microburst using operative mesoscale models is scarce, the Weather Research and Forecasting model has been used in the diagnosis of this phenomenon. In this paper, such model is used to simulate several microburst conducive days using two different boundary conditions. The energy spectra of the simulations are computed to evaluate the effective resolution of the model. The results are in line with previous studies and produce no notable differences among the boundary conditions. Nonetheless, the energy spectra show an over-energetic troposphere at microscale resolutions, rendering the effective resolution inadequate for microburst forecasting using the simulated physics variables. Thus, mesoscale indices are analysed as a prognostic tool. The Wind Index, the Wet Microburst Severity Index and the Microburst Windspeed Potential Index do not show high forecasting performances, even though improving the results of climatology. Also, notable differences among the boundary conditions can be seen. The most consistent results are achieved by the Wet Microburst Severity Index. Note: Due to the high resolution used, the 40 simulations produced are over 7 TB is size. Only the WRF configuration files are provided here, to reproduce the simulations if necessary. The NCL scripts can be used to plot the Energy Spectra and several variables used in the evaluation. The R scripts can be used to plot ROC curves. See each file description for further detail.