Practical rare event sampling for extreme mesoscale weather

Area of Science:

• Atmospheric Science
• Computational Science
• Climate Modeling

Background:

• Extreme mesoscale weather events like tropical cyclones pose significant simulation challenges.
• Accurate simulation is crucial for understanding and mitigating damage from severe weather.
• Existing simulation strategies often lack the efficiency needed for extreme event analysis.

Purpose of the Study:

• Introduce a novel rare event sampling algorithm, quantile diffusion Monte Carlo (quantile DMC).
• Demonstrate the effectiveness of quantile DMC for simulating extreme tail behavior in various processes.
• Assess the feasibility of quantile DMC for modeling extreme mesoscale weather phenomena.

Main Methods:

• Developed and implemented the quantile diffusion Monte Carlo (quantile DMC) algorithm.
• Applied quantile DMC to sample extreme realizations of historical tropical cyclones (Hurricane Earl and Hurricane Joaquin).
• Compared the performance and efficiency of quantile DMC against other existing sampling methods.

Main Results:

• Quantile DMC successfully sampled extremely intense realizations of tropical cyclones.
• The algorithm demonstrated potential for providing low-variance extreme weather statistics.
• Initial results indicate advantages over other sampling techniques, though further efficiency improvements are noted.

Conclusions:

• Quantile DMC offers a promising approach for simulating extreme weather events.
• The algorithm has the potential to enhance the accuracy and efficiency of mesoscale weather simulations.
• Further research is needed to optimize quantile DMC for greater efficiency in practical applications.