Assessment of mountain river streamflow patterns and flood events using information and complexity measures

Abstract The availability of distinctly interpretable assessments to characterize and describe river discharge for mountainous rivers for different climatic events can improve our understanding of the various dynamics related to hydrological processes. Furthermore, information and complexity metrics can reveal invaluable information about the unseen processes that occur within a system. In this study, hourly streamflow records obtained from five gauging stations of a mountainous river were analyzed to quantify different patterns and characterize system states at both low and high frequencies using increasing aggregation lengths. In addition, we propose a new extension for the information and complexity theory, allowing it to be customized for flood assessment. Moreover, we clarify how a pattern (i.e., a word length) can be suitably defined by means of information and complexity metrics. Regarding low-frequency analyses, our results related to information and complexity metrics indicate two scaling regimes for river discharge, one of which may describe river memory characteristics. Regarding high-frequency analyses, our findings indicate the presence of an additional scaling regime that occurs along an hourly scale, captured by streamflow data, and is obtained using a novel hydroacoustic system. Additionally, power spectral density results confirmed our findings. A further significant result from our study is the clear correlation between complexity and fractal fluctuations, which should be addressed in future studies. In summary, this research focuses on new aspects of information and complexity metrics to be customized for the detection and understanding of temporal structures of streamflow patterns during both standard and extreme events.