On spatial conditional extremes for ocean storm severity

Area of Science:

• Environmental Science
• Statistics
• Oceanography

Background:

• Understanding spatial dependence in extreme environmental events is crucial for risk assessment.
• Existing models like max-stable processes have limitations in conceptual simplicity and flexibility for extremal dependence.
• Conditional extremes methodology provides a framework for analyzing spatial extremes.

Purpose of the Study:

• To develop and apply a novel model for conditional spatial extremes.
• To estimate the extremal dependence of ocean storm severity (significant wave height) in the North Sea.
• To investigate how this dependence changes with distance and geographical orientation.

Main Methods:

• Developed a new model for conditional spatial dependence, building on Heffernan and Tawn's work.
• Employed a Bayesian framework for parameter estimation.
• Analyzed significant wave height data along East-West and North-South transects in the Northern North Sea and Central North Sea.

Main Results:

• The conditional extremes "linear slope" parameter (α) decays approximately exponentially with distance for significant wave height on a Laplace scale.
• The decay of mean dependence is faster in the Central North Sea compared to the Northern North Sea.
• The East-West transect in the Northern North Sea exhibits the greatest persistence of mean dependence.

Conclusions:

• The proposed model offers a flexible and conceptually straightforward approach to spatial extreme value analysis.
• Spatial dependence of ocean storm severity decreases with distance, with regional variations.
• The alignment of transects with storm propagation directions influences the persistence of dependence.