Multifractal characteristics and extremes of high-resolution rainfall, application to climate change detection

The quality of rainfall statistics, especially the Intensity‑Duration‑Frequency curves, closely depends on the reliability of available data. However, it has been shown that most of the time series obtained with tipping bucket rain gauges have a lower measuring frequency than is normally assumed. This question is particularly important for urban hydrology, where it is important to take into account high frequency fluctuations of rainfall. Preliminary studies showed that the estimated number of floods was lower when low time resolution data were used, compared to number of floods obtained with the help of higher time resolution data. The deficit of high frequency data can lead to apparent breaks in the scaling laws, which unnecessarily and notoriously complicate rainfall modelling. It is therefore essential to quantify the quality of data before using them. We present a SERQUAL procedure that enables us to answer this question and we use this procedure to select sub-series having the qualities required for high-resolution analysis. A multifractal approach is then applied to the selected data to characterize the temporal structure and the extreme behaviour of rainfall. In the present paper we present a reliable estimate of the multifractal parameters of the five‑minute high resolution rainfall data for the four departments in France. These parameters can be used to calibrate or validate statistical and stochastic models. On the other hand, the evolution of the multifractal characteristics can also be used to evaluate the hydrological consequences of climate change. The obtained results show that the influence of climate change on precipitation is not perceptible for the studied periods in Ile-de-France. © 2014, Institut National de la Research Scientifique. All rights reserved.