Publication Date:
2023-08-29
Description:
We propose an analytical expression describing the probability of occurrence of tornadoes as a function of selected meteorological variables. Our study is based on the available lists of recorded tornadoes events in the USA and Europe matched with the meteorological conditions provided by the high resolution ERA5 reanalysis data. The statistical validation of the results has been carried out using Bootstrap techniques and Monte Carlo permutation procedures. The considered variables are the convective available potential energy, the vertical wind shear, the storm-relative helicity, and the lifting condensation level. Our analysis suggests that the most suitable representation for the probability of occurrence of tornadoes is a bivariate exponential-like expression as a function of the convective available potential energy and either the wind shear or the storm relative helicity. This expression allows to compute the probability of tornado occurrence using data extracted from operational weather prediction and available climate projections, thus compensating for the lack of resolution needed to explicitly model tornadoes in these simulations. The proposed methodological approach can be used to obtaining formula of increasing accuracy as larger datasets of tornado events will become progressively available in the future.
Language:
English
Type:
info:eu-repo/semantics/conferenceObject
