Multiple wind hazards in non-hurricane prone regions

The estimation of the wind speed values used in codes and standards is an integral part of the wind load prescription process. In a number of codes and standards wind speeds outside of hurricane prone regions are estimated using a single probability distribution developed from observed wind speed data, with no distinction made between the type of causal wind hazard (e.g., thunderstorm). Distinct non-hurricane wind hazards have been shown to possess different probability distributions and estimation of non-hurricane wind speeds based on a single probability distribution has been shown to be unconservative in some cases, especially at locations that receive extreme wind speeds from multiple wind hazards (storm types) at somewhat regular intervals. Current treatment regarding non-hurricane wind hazards in worldwide codes and standards is touched upon in this work. Meteorological data is available at a considerable number of United States (US) stations that have information on wind speed as well as the type of causal wind hazard. In this paper probability distributions are fit to distinct storm types and the results of these distributions are compared to fitting a single probability distribution as well as considering multiple storm types independently in a "mixed" distribution. Consideration in this paper is given to two extreme value distributions (e.g., Gumbel, generalized Pareto) when fitting distinct storm types. It is shown that how those distributions are used may affect wind speed estimation. US areas prone to multiple non-hurricane wind hazards are identified.