On the Automation of Flood Event Separation From Continuous Time Series

Abstract

Can machine learning effectively lower the effort necessary to extract important information from raw data for hydrological research questions? On the example of a typical water-management task, the extraction of direct runoff flood events from continuous hydrographs, we demonstrate how machine learning can be used to automate the application of expert knowledge to big data sets and extract the relevant information. In particular, we tested seven different algorithms to detect event beginning and end solely from a given excerpt from the continuous hydrograph. First, the number of required data points within the excerpts as well as the amount of training data has been determined. In a local application, we were able to show that all applied Machine learning algorithms were capable to reproduce manually defined event boundaries. Automatically delineated events were afflicted with a relative duration error of 20 and 5% event volume. Moreover, we could show that hydrograph separation patterns could easily be learned by the algorithms and are regionally and trans-regionally transferable without significant performance loss. Hence, the training data sets can be very small and trained algorithms can be applied to new catchments lacking training data. The results showed the great potential of machine learning to extract relevant information efficiently and, hence, lower the effort for data preprocessing for water management studies. Moreover, the transferability of trained algorithms to other catchments is a clear advantage to common methods.

Parsimonious statistical learning models for low-flow estimation

“…Regression methods cover a wide spectrum of models, and especially in the last decade there was increasing interest in statistical learning models in hydrology (Abrahart et al, 2012;Dawson and Wilby, 2001;Nearing et al, 2021;Solomatine and Ostfeld, 2008), with the terms "statistical learning" and "machine learning" being used synonymously. The applications include rainfall-runoff modeling by neural networks (e.g., Kratzert et al, 2019a, b), using support vector machines (SVM) for prediction of karst tracers (Mewes et al, 2020) or reference evapotranspiration (Tabari et al, 2012) and random forest for flood event classification (Oppel and Mewes, 2020). Nevertheless, the implementation of statistical learning methods for predicting low flow is still rare.…”

An Objective Time‐Series‐Analysis Method for Rainfall‐Runoff Event Identification

“…This hampers direct comparison of findings from independent research initiatives. Depending on the perspective the hydrologist is taking, focusing first on separating the rainfall into different events (e.g., Koskelo et al., 2012; Seibert et al., 2016) or first on separating the streamflow into different events (e.g., Fischer et al., 2021; Graeff et al., 2012; Merz et al., 2006; Oppel & Mewes, 2020), the chosen routine and corresponding assumptions can be quite different.…”

A methodology for the estimation of control flood wave hydrographs for the Horné Orešany reservoir

“…Thiessen et al ( 2019), who developed a method for identifying rainfall-runoff events in discharge time series, states that even though this process might be straightforward for a trained hydrologist, it is complicated to formulate rigid criteria that would enable the reliable identification of flood events. Currently, there is no single accepted method for automating this process, even though it has been the subject of substantial scientific efforts (Oppel and Mewes, 2020). As the beginning and end of a flood event are often associated with the intersection of baseflow and direct runoff curves, most methods try to separate the baseflow from the discharge time series.…”