Streamflow Data Analysis for Flood Detection using Persistent Homology

Abstract

Flooding is an environmental hazard that occurs almost everywhere around the world. Analysis of streamflow data can give us important climatic information for flooding events. Persistent homology (PH), a new analysis tool in topological data analysis (TDA) offers a new way to look at the information in a data set using qualitative approach. PH uses topology to extract topological features such as connected components and cycles that exist in the data set. In this paper, we present a new approach for streamflow data analysis for flood detection by using PH. An analysis was conducted at Sungai Kelantan, Malaysia. The result shows that PH gives different pattern of topological features for dry and wet periods. In particular, there are more persistent topological features in the form of connected components and cycles in the wet periods compared to the dry periods. We observed that the time series of the distance measure corresponding to the evolution of the components is consistent with the time series of the streamflow data. As a conclusion, this study suggests that the time series of the distance measure corresponding to the evolution of the components can be used for flood detection at Sungai Kelantan, Malaysia.