Time-scale-dependent Sediment–Discharge Coupling across Fourteen Catchments Using Wavelet Analysis

Suspended sediment dynamics exhibit strong temporal variability and nonlinear behavior, making it challenging to characterize their relationship with streamflow using traditional statistical or machine-learning approaches. In this study, we addressed the following questions: how does the coupling between suspended sediment concentration (SSC) and discharge change across temporal scales, and which hydrological, morphological, climatic, and land-use factors control these changes? To investigate this, we examine the time-scale-dependent and non-stationary coupling between SSC and discharge across fourteen catchments (0.94-2846 km²) spanning diverse climatic and geomorphic settings. We applied wavelet coherence (WTC) and partial wavelet coherence (PWTC) analyses to quantify both the total and precipitation-independent SSC-discharge coupling across time scales ranging from 2 to 512 days. The analysis is performed continuously in time and interpreted within short (2-32 days), intermediate (32-128 days), and long (128-512 days) temporal bands. We used Spearman correlation to explore links between coherence and catchment characteristics, including physiography, morphology, climate, hydrology, and land use. Across all catchments, SSC-discharge generally exhibits strong coupling, although the strength of this coupling can be weak and fragmented at some time scales, indicating a non-stationary sediment response to discharge variations. After removing the influence of precipitation, much of this coherence weakens or becomes more fragmented across time scales, demonstrating that a substantial part of the SSC-discharge relationship reflects their shared hydrological forcing by precipitation. Nevertheless, a part of coherent patterns persists in all catchments, implying that catchment characteristics also sustain SSC-discharge coupling beyond direct precipitation effects. At short time scales, coupling is primarily controlled by slope and maximum length of the catchment; at intermediate scales, by moisture accumulation, land use, and aspect; and at long time scales, by moisture, slope aspect, and pasture cover. Using data from fourteen catchments, this study moves beyond single-catchment analyses and shows that wavelet-based approaches can disentangle precipitation-driven sediment dynamics from those controlled by catchment characteristics, providing new insight into how intrinsic catchment properties regulate SSC-discharge interactions across multiple temporal scales.

Key words: Catchment characteristics, Hydro-sediment dynamics, Partial wavelet coherence (PWTC), Suspended sediment concentration, Wavelet coherence (WTC)