R code for Modeling the Scaling of Short‐Duration Precipitation Extremes with Temperature

These files contain the R code for the Quantile Regression models for the scaling of short‐duration precipitation extremes with (dew point) temperature. The code is based on the main elements of the following Technical Note:

• Van de Vyver, H., Van Schaeybroeck, B., De Troch, R., Hamdi, R., & Termonia, P. (2019). Modeling the scaling of short‐duration precipitation extremes with temperature. Earth and Space Science, 6, 2031--2041.  https://doi.org/10.1029/2019EA000665

There are two different quantile regression models included: (i) CC model: linear quantile regression model for estimating the Clausius-Clapeyron (CC) rate, and (ii) CC+ model: piecewise linear quantile regression model for estimating the CC rate, the super-CC rate, and the (dew) point temperature where the transition from CC scaling to super-CC scaling takes place.

The output consists of the estimated scaling rates, the change-point, 95%-Confidence Intervals, BIC, and a Goodness-of-Fit which measures the relative strength of the predictor, i.e. (dew point) temperature, to the hourly precipitation extremes.

The files include:

• "PP1h.TEMP.DEW.UCCLE.RData": data input file with declustered hourly precipitation, and corresponding daily mean (dew point) temperature.
• "QR.functions.R": file with Least Absolute Deviation (LAD) Functions to be Minimized.
• "QR.R": R-script for estimating the CC and CC+ models
• "QR.BOOT.R": R-script for estimating the uncertainty with the Bootstrap.