Statistical assessment of extreme weather phenomena under climate change - advanced study programme
Training in use of extreme value analysis to assess how frequency, duration, and intensity of extreme weather events could shift as part of global climate change.
Topics:
1. Analyzing how extreme events and their impacts evolve over time
2. Extracting extreme events from coarse resolution models that cannot explicitly resolve extremes using techniques such as statistical downscaling
3. Use of extreme value distributions (justified through statistical theory of extreme values)
4. Special characteristics of meteorological processes
a. Estimating return levels under non-stationarity
b. Model resolution vs. scales where extreme events occur
c. Weather spells such as heat waves or droughts
d. Temporal and spatial clustering of extremes
Activities:
- Two tutorials on extreme value analysis, one on basic statistical theory and another on its implementation using open source statistical language R
- Invited lecturers will discuss extreme value analysis from differing perspectives of atmospheric science, statistics, risk analysis, hydrology and other areas
- Students will work in groups on well-defined, but open-ended research projects