Long-term temporal structure of catchment sediment response to precipitation in a humid mountain badland area

Intraseasonal‐to‐Interannual Analysis of Discharge and Suspended Sediment Concentration Time‐Series of the Upper Changjiang (Yangtze River)

“…To achieve the above goal we use the method of wavelet transformation because of its ability to capture processes variability at multiple time‐scales (Torrence & Compo, 1998). The wavelet transform has been successfully applied in the past to a hydro‐climatic time‐series (Segele et al., 2009; Webster & Hoyos, 2004) and to discharge‐precipitation time‐series (Carey et al., 2013; Juez & Nadal‐Romero, 2020a,2020b) to determine periodicities and trends. We use the wavelet transform instead of other temporal analysis techniques such as Fourier spectral analysis (processes are identified by periodicity, e.g., L. C. Smith et al., 1998) or multi‐scale entropy analysis (processes are identified by their degree of complexity or randomness, e.g., Y. Wang et al., 2018) because of their favorable strengths: (a) wavelet transform can identify processes in both time and periodicity in the non‐stationary time‐series (contrary to Fourier spectral analysis), and (b) wavelet transform allows to localize dominant time‐scales within the complete temporal spectrum, ultimately identifying the variability of each time‐scale throughout the study period (contrary to Fourier spectral analysis and entropy analysis).…”

“…To achieve the above goal we use the method of wavelet transformation because of its ability to capture processes variability at multiple time-scales (Torrence & Compo, 1998). The wavelet transform has been successfully applied in the past to a hydro-climatic time-series (Segele et al, 2009;Webster & Hoyos, 2004) and to discharge-precipitation time-series (Carey et al, 2013;Juez & Nadal-Romero, 2020a,2020b to determine periodicities and trends. We use the wavelet transform instead of other temporal analysis techniques such as Fourier spectral analysis (processes are identified by periodicity, e.g., L. C. Smith et al, 1998) or multi-scale entropy analysis (processes are identified by their degree of complexity or randomness, e.g., Y.…”

Time Series Features for Supporting Hydrometeorological Explorations and Predictions in Ungauged Locations Using Large Datasets

“…The various open research pathways also include the simultaneous consideration of the spatial proximity of the stations and time series features for obtaining predictions, as well as the utilization of additional time series features from the stochastic (statistical) hydrology and data science fields, and even the utilization of the concept of massive feature extraction [24,25,27,29] from the latter of the aforementioned fields. In fact, although this work already covers a larger variety and a larger number of time series features than usual in stochastic hydrology, many more time series features are available (see, e.g., the ones investigated by Papacharalampous et al [34,35]; Hamed [50]; Montanari [51]; Ledvinka [52]; Ledvinka and Lamacova [53]; Juez and Nadal-Romero [54,55]) and could be useful in streamflow regionalization settings, given also the generally acknowledged significance of finding new informative predictors for obtaining improved predictions.…”

Six Decades of Hindsight Into Yesa Reservoir (Central Spanish Pyrenees): River Flow Dwindles as Vegetation Cover Increases and Mediterranean Atmospheric Dynamics Take Control

“…It thus provides a complete time‐scale representation of localized and transient phenomena occurring at different time‐scales (Torrence & Compo, 1998). In this research, we make use of the Morlet wavelet, which was successfully used in the past to analyze precipitation and discharge time‐series (Carey et al., 2013; Juez & Nadal‐Romero, 2021; Juez et al., 2021; Pérez‐Ciria et al., 2019). This wavelet type is characterized as: $$${\varphi }_{0}(\eta )={\pi }^{-1/4}{e}^{i{\omega }_{0}\eta }{e}^{-{\eta }^{2}/2}$$$ where $${\varphi }_{0}(\eta )$$ is the wavelet function, $$\eta $$, is a dimensionless time parameter, i is the imaginary unit and $${\omega }_{0}$$ is the dimensionless angular frequency taken as 6 as it provides a good match between time and frequency localization.…”

“…The Yesa reservoir was built for irrigation purposes, and the draining catchment underwent a process of re‐vegetation for the last decades (García‐Ruiz et al., 2015 ). We make use of robust statistical tool such as the wavelet transform method, which is able to distinguish different time‐scales of variability and localize changes in the modes of variability within time‐series (Carey et al., 2013 ; Juez & Nadal‐Romero, 2020 , 2021 ; Juez et al., 2021 ; Labat et al., 2005 ; Pérez‐Ciria et al., 2019 ; Restrepo et al., 2014 ). This is an important step forward in identifying changing patterns at different and non‐similar time‐scales, enabling explaining the causes of observed changes and understanding their implications (Zhang et al., 2011 ).…”