Severe Wind Hazard Assessment using Monte Carlo Simulation

Abstract

A Monte Carlo-based model to assess severe wind hazard is presented. Synthetic wind datasets for hazard analysis have been generated using Monte Carlo simulation of the physics of severe wind gust generation, to overcome the limitations of data-based statistical models. These statistical models consider extreme wind gust speeds to calculate the average probability of exceedance of a given wind speed in a single year (return period), and hence the return period is calculated using extreme value distributions. Monte Carlo wind hazard results are shown to be comparable to those produced by data-based statistical methods. They are similar to the results reported by Holmes (Australian Journal of Structural Engineering 4(1): [29][30][31][32][33][34][35][36][37][38][39][40] 2002) and also to the prescribed wind gust speeds of the Australian/NZ standards for wind loading of structures in Region A (non-cyclonic regions) of Australia (AS/NZS 1170.2 2002).

A review of severe thunderstorms in Australia

“…More recently, approaches using reanalysis environment data downscaled to 14km resolution have been explored to spatially extend this approach over Tasmania and approximate the 500-year return levels for wind magnitude from thunderstorms from extreme value theory (Sanabria and Cechet, 2010;Cechet et al, 2012). These results suggest that the overall hazard from these events was less than from synoptic winds, but still returned values of 133.2kmh -1 for the north, and 122.4kmh -1 for the south.…”

“…These results suggest that the overall hazard from these events was less than from synoptic winds, but still returned values of 133.2kmh -1 for the north, and 122.4kmh -1 for the south. These methods provide a way to assess risk for damaging convective winds, though do not provide information regarding its occurrence, and are subject to limitations arising from the problems of the observational record and the assumptions of extreme value theory (Sanabria and Cechet, 2010).…”

Recent changes in precipitation extremes in the Heihe River basin, Northwest China

“…The selection of an appropriate threshold is a crucial step for the application of the peak over threshold approach with a Generalized Pareto Distribution (GPD) (Saidi et al, 2013). The GPD fit is highly sensitive to threshold selection (Sanabria and Cechet, 2010). Therefore, the base period used to calculate the threshold can have a great effect on methods more sensitive to thresholds in the modeling of extreme climate events.…”

A severe wind hazard model for non-cyclonic regions of Australia using simulated climate data

“…The MC method works by simulating the physics of wind generation for synoptic wind conditions. The process consists of the numerical convolution of the mean wind speed and an empirical gust factor (GF) to produce gust wind speeds (Sanabria and Cechet, 2010). The gust factor is defined as the ratio of maximum wind speed (gust) and mean wind speed for the same time period (generally ten minutes duration).…”