Summary

Exceedance frequency curves. The central function method is combine_exceedance_curves in exceedance_frequency_curves.py. This function can be used to combine two exceedance impact or exceedance intensity curves with several parameters, such as

• coincidence_fraction: The fraction of the time unit according to which we say that two events (impacts or intensities) occur at the same time
• correlation_factor: A correlation factor to control if the different distributions should be sampled independently or with some correlation.
• aggregation_method: How to combine the values of the different curves, e.g. sumfor impact or max for intensities

These options are exemplified in example_combine_two_curves.ipynb.

Return Period Maps. One application of the above functionality is to change the resolution of return period maps in return_period_maps.py. E.g., if one wants to halven the resultion, four exceedance curves must be combined into a single one, which can be done using the above functionality. This is exemplified in change_resolution_RPMap.ipynb.

Requirements

All provided functionality and demnostrator notebooks can be run using a python environment with the climada package installed. If CLIMADA-related functionality is not required, a python environment with geopandas, matplotlib, numpy, scipy, and shapely is sufficient.

Demonstator notebooks

example_combine_two_curves.ipynb. Presenting different ways to combine two exceedance curves.

change_resolution_RPMap.ipynb. How to coarsen return period maps using the exceedance curve combination functionality.

estimate_AAI.ipynb. Some thoughts on how to compute average annual impacts with exceedance impact curves.