An Australian convective wind gust climatology using Bayesian hierarchical modelling.

Area of Science:

• Meteorology
• Climatology
• Atmospheric Science

Background:

• Quantifying severe convective wind gust risks requires a spatially complete climatology.
• Existing observational networks are often too sparse to capture all events.

Purpose of the Study:

• To develop a spatially complete convective wind gust climatology for Australia.
• To correct for observational biases in wind gust data.

Main Methods:

• Coupling observational and ERA-Interim reanalysis data (2005-2015).
• Developing Bayesian Hierarchical models using weather station data and severe weather indices (SWI).
• Estimating seasonal gust frequencies and correcting for observational biases.

Main Results:

• Different SWI combinations proved effective for different seasons (e.g., Lifted Index for autumn/winter; Microburst Index for spring/summer).
• Minimum event counts occurred in winter, primarily along southwestern Western Australia, Tasmania, and Victoria.
• Summer showed the highest event counts, with maxima in northern Western Australia, the Northern Territory, and northeast New South Wales.

Conclusions:

• Bayesian hierarchical models effectively create a spatially complete wind gust climatology.
• The models can be applied using only reanalysis data, demonstrating versatility.
• Understanding seasonal and regional variations in wind gust events is crucial for risk assessment.