FLOW RESISTANCE AND SEDIMENT TRANSPORT IN HIGH-GRADIENT STREAMS

Abstract

Soil erosion and sediment transport pose significant environmental challenges, leading to land degradation, reservoir sedimentation, and ecological imbalances. In mountainous regions, gravel-bed channels play a crucial role in transporting water and sediments downstream, impacting the sediment dynamics in alluvial plains. Traditionally, the focus has been on suspended sediments, with limited attention to the mobility of coarser bed materials. Monitoring transport of coarser grains in such energetic mountain streams can be challenging due to extreme flow events and limited accessibility. As such the spatial and temporal variable hydraulic forcing and stream bed dynamics makes the conventional bedload equations unreliable, emphasizing the need for a comprehensive understanding of hydraulic forcing and flow resistance. The non-uniform sediment on river bed particularly dominated by coarser grains in the irregular channels increases the resistance in the flow path causing an overall reduction in the coarse sediment transport. Further, the non-uniform sediments are likely to promote armouring and sediment clustering, causing increased resistance in the flow path. As such the meta-analysis on a large dataset (2805 measurements) compiled from the literature reveals that the velocity estimates made using the conventional flow resistance equations deviate particularly in the case of low relative submergence (y/D84 ≤ 1) and high geometric standard deviation of the bed material (σg ≥ 7.5). This discrepancy was assumed to probably occur because of additional resistance occurring due to distortion of free surfaces, spill losses caused by no-uniform bed grains in addition to drag and form resistance. Therefore, the geometric standard deviation of the bed material was considered as an additional parameter in the conventional flow resistance equations to include the additional resistance. The modified equations with consideration of sediment non-uniformity perform well in estimating the flow velocity both during low relative submergence and highly non-uniform bed material. Further, the spatial variability in flow parameters due to irregular cross-section, and sediment nonuniformity in gravel-bed rivers necessitates a nuanced approach for estimating bed material transport. Further, the Lagrangian-based approach for monitoring the mobility of painted tracers, cross-sectional elevations, and grain size distribution was applied in two Himalayan Rivers (Aglar and Paligad Rivers) in India. The hydraulic forcing (dimensionless shear stress, 𝜏*) averaged over wetted cross-section, together with its spatial distribution (𝜏* of each 0.5 m wide sub-sections), was applied to study the spatiotemporal variability in transport rates. While using local parameters, the total annual bed material transport bed material transport was estimated to be 67100 t (±20400 t) and 18400 t (±6000 t) in the Aglar and Paligad Rivers, respectively. Of this, nearly 60 % of transport occurred during the monsoon and the overall contribution of partial transport (PT) remained low (< 6 %). However, based on cross-section average parameters, total transport was estimated to be 42300 t (±15800 t) and 12200 t (±4700 t), in Aglar and Paligad, respectively, with nearly 79 % and 68 % occurring during the monsoon. Moreover, the contribution of PT increased to nearly 18 % and 29 % for the Aglar and Paligad Rivers, respectively. Additionally, the dependence of partial transport on Y and full transport on ds results in an abrupt shift in transport rates at the transition from partial to full transport, causing discontinuity (break) in transport curves. Therefore, a unified function was proposed to represent the extent of transport for both partial and full transport, yielding continuous transport curves. Further, the consideration of reduced slope to account for the additional flow resistance using modified flow resistance equations in the Wilcock and Crowe model yields a value of 201000 t and 111000 t compared to 682100 t and 207100 t using the total slope in Aglar and Paligad respectively, highlighting the benefits of accounting for flow resistance. This further necessitates the need for flow resistance partitioning and simple mathematical forms to directly account for flow resistance in the conventional bedload equations. For individual flood events, stochastic transport models, such as the Einstein-Hubbell-Sayre (EHS) model have been shown to adequately represent the longitudinal dispersion of tracers. We utilized field observations on gravel dispersion over nine years (2007-2015) in the snowmelt-driven regime of Halfmoon Creek, Colorado, USA to study gravel dispersion over a series of multiple events. The observations of flow, entrainment, and dispersion were used to develop a simulation model utilizing the EHS compound Poisson process. The observed mean virtual velocity of the tracer population was observed to slowdown with expended energy (cumulative stream power) after the 2010 large event. The base model deviates from the observations in representing tails, overpredicts the mean displacements, and shows a narrower spatial distribution. The heavy-tailed resting times indicate a prolonged immobilization period, suggesting the preferential movement of most mobile grains. As such, 34 % of most mobile grains were observed to constitute 50% of the total entrainments. The consideration of preferential movement explains the longitudinal spread but still overpredicts the displacement after the 2010 event. However, the preferential movement together with the underlying mechanisms causing deviations, such as reduction in virtual velocity, entrainment probability, and morphological trapping of meander bends, helps to adequately recreate the observed dispersive behavior. The historical flow records used for simulating dispersive behavior over an extended timeframe reveal the exhumation of immobile grains back in transport for the flow magnitudes larger than the previous occurrence. Further, large rivers originating from the mountains are efficient sediment conveyors, influencing river characteristics downstream, and any changes in sediment production due to natural and anthropogenic factors in uphills impact river morphology downstream. The Yamuna River (mainstem of Aglar and Paligad Rivers) has been regulated by the construction of the Lakhwar-Vyasi Dam which together with increased sediment mining, raised concerns about potential morphological implications in the downstream reaches. We explored morphological adjustments along a 46 km segment of the Yamuna River in the Himalayan foothills (Dakpathar to Hathni Kund Barrage), using a combination of remote sensing approaches. The cloud computing platform Google Earth Engine (GEE) was used to extract active river channels (including water and exposed sediment) using an MNDWI and NDVI thresholding approach from annually resolved temporal composite images from 1989 to 2021. Declassified CORONA, and Google Earth imagery were used to mark a check on the thresholding-based approach and to provide longer-term insights into channel change. RivMAP toolbox provides quantitative information on the rates of bank erosion and accretion, bank line shifts, and changes in active channel width, which shows substantial narrowing after 2013, coinciding with the construction of the Lakhwar-Vyasi Dam. We calculated a 67% reduction in mean active channel width, narrowing from ~800 m in 1989 to ~250 m in 2021. During the same period, evidence of sand mining in the active channel indicates a substantial increase in mining. The concept of stasis was explored under different flow regulation scenarios to suggest that the river has undergone a state of inactivity, where the river is doing nothing for much longer durations. This study highlights the complexity of sediment dynamics in Himalayan gravel-bed rivers, emphasizing the need for a multidimensional approach that considers spatiotemporal variability of critical parameters, particle-scale interactions, and the impacts of human interventions on river morphology. Scientific attention is crucial to assess the feedback between flow resistance, sediment transport and the potential consequences of flow regulation and sediment mining on these ecologically sensitive river systems.