Abstract

The first-ever engineering-derived tornado wind speed maps have been produced for the contiguous United States. Using multi-variate statistical analysis of 11 tornado and physiographic variables, we developed 9 broad tornado climatology regions. We analyzed regional and national data to produce probabilistic models for: population bias; EF-Scale distribution; tornado path length, width, direction, and translational speed; radius of maximum winds; tornado path length intensity variation; variable path widths within a tornado; mean to maximum path width ratios; and maximum damage widths relative to local path width. We used a probabilistic load and resistance modeling framework to develop engineering-derived wind speeds from the EF-Scale system by analyzing the most common Damage Indicator (FR12, one- and two-family residences) used to rate EF3-5 tornadoes. The developed wind speed distributions, based on 44 3-D models of FR12 structures, are broad and encompass the original judgment-based EF-Scale wind speeds. We used these data to support a probabilistic tornado wind field model, simulate tornado wind speed swaths, and develop regional tornado hazard curves. The tornado hazard curves depend on the building or “target” size. Fifty-one tornado wind speed maps were developed for 8 target sizes ranging from geometric points to 4 million sq ft targets and for 8 return periods from 300 to 10 million years. Numerous judgments and assumptions were required in the model development. Many of the probabilistic models consider both epistemic and aleatory uncertainties. The derived mean wind speeds are “best-estimates” and are intended for use in engineering design and safety analysis.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call