Welcome to IDF’s documentation!

IDF

Developer: Felix Fauer (felix.fauer@met.fu-berlin.de) ClimXtreme B2.5- IDF-AF


IDF is a freva plugin that creates Intensity-Duration-Frequency Curves for Extreme Precipitation Events based on the IDF R package developed at FUB.

It analyses daily precipitation station files. Specific events can be plotted into the IDF curves and their average return period is calculated.

Notes:

  • The meaning of the d-GEV parameters Duration Offset, Multiscaling and Intensity Offset is explained in detail in Fauer et al. (2021). To summarize it briefely, when using sub-hourly data, the parameter duration-offset should be enabled, as well as intensity-offset. Multiscaling should be enabled, when the focus lies on long durations and can be disabled otherwise.

  • “Bootstrap given” gives the possibility to load bootstrap results from a former run to save time. This makes sense only, when data and model (including choice of parameters) does not change.

  • Further publications that are based on the IDF package are Ulrich et al. (2020), Ulrich et al. (2022) and Fauer et al. (2023)

References:

Fauer, F. S., Ulrich, J., Jurado, O. E., and Rust, H. W.: Flexible and consistent quantile estimation for intensity–duration–frequency curves, Hydrol. Earth Syst. Sci., 25, 6479–6494, [https://doi.org/10.5194/hess-25-6479-2021], 2021

Ulrich J., Jurado O. E., Peter M., Scheibel M., Rust H. W. Estimating IDF Curves Consistently over Durations with Spatial Covariates. Water. 2020; 12(11):3119. [https://doi.org/10.3390/w12113119]

Ulrich, J., Fauer, F. S., and Rust, H. W.: Modeling seasonal variations of extreme rainfall on different timescales in Germany, Hydrol. Earth Syst. Sci., 25, 6133–6149, [https://doi.org/10.5194/hess-25-6133-2021], 2021

Fauer, F. S., Rust, H. W. Non-Stationary Large-Scale Statistics of Precipitation Extremes in Central Europe, 05 February 2023, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-2542862/v1]

Updates:

[2023.05.23]

  • bootstapping and uncertainty is added (can be disabled)

  • the bootstrapping results are stored and can be loaded in another run to save time (when the model stays the same)

  • IDF Figures are improved

  • a figure with the distribution density is added

  • one can select a specific event and add it to the plot to see its return period. By default, the maximum recorded event is used.

  • the most important data are stored in an “rds” file. That is an R-specific format. The Program’s output lists its path after “- store results in data fromat in: [Path]”

[2022.08.08]

First working plugin example:

  • creation of conda environment for dependencies and libraries of the plugin.

  • adaptation of wrapper file for new Freva version.

NOTE: at the moment it only works either with 1 explicitely entered file in InputDirectory or 1 station file from the Databrowser (product=station).

Installation

To install the conda environment with the necessary libraries to run the plugin:

  1. first you need to have load conda first, by for example loading a Freva module:

$ module load clint xces

  1. then git clone and create the conda environment:

$ git clone https://gitlab.dkrz.de/freva/plugins4freva/idf.git
$ cd idf
$ make conda

This it will create the conda environment with R==4.1.3 AND install the IDF package from CRAN.

The conda environment can be activated via:

$ source /path/to/idf/plugin_env/bin/activate

to deactivate it, run:

$ conda deactivate

Indices and tables