Explore the vast range of titles available.

Satellite estimates of suspended‐sediment concentration (SSC) are widely used to understand ecosystem functions and resilience of coastal environments. Yet, they rely largely on fair‐weather observations. Combining continuous in situ time series with multi‐decadal satellite records and reanalysis products, we quantify a systematic fair‐weather bias that misrepresents coastal SSC in both magnitude and sign. Negative biases dominate where cloudy conditions coincide with energetic sediment mobilization, causing satellites to miss high‐SSC events and underestimate long‐term means by up to ∼20%. In contrast, positive biases arise where clear‐sky conditions align with elevated wave energy. Bias magnitude and direction are driven by local couplings among cloud cover, wave energy and sediment mobilization, enabling prediction and correction. Climate‐model projections indicate only modest changes in cloudiness over this century, suggesting the bias will persist in the future. Ignoring fair‐weather limitations distorts sediment budgets and biogeochemical fluxes, underscoring the need for bias‐aware algorithms in global coastal assessments.

Accurately predicting sediment dynamics and understanding their intrinsic contributors are pivotal for sustainable environment and water management. While machine learning (ML) enables precise predictions, its “black‐box” nature hinders transparency and credibility, posing challenges in interpretability and uncertainty quantification (UQ). To achieve trustworthy ML for riverine sediment timeseries predictions, this study proposes an integrated ML framework, enhancing key steps: feature selection, UQ, and interpretation. Lagged hydro‐environmental variables are incorporated via rigorous feature selection. SHapley Additive exPlanations (SHAP) and conformal prediction are utilized to refine interpretability and UQ, respectively. Based on 41‐year multi‐source data and three ensemble learning algorithms (LightGBM, XGBoost, and random forest (RF)), this study models daily suspended sediment concentration (SSC) separately for seven subtropical watersheds and evaluates overall and local accuracy. Key findings include: (a) Discharge and precipitation dominate SSC variability (explaining ∼56.8% and ∼18.9% of the variability, respectively). Sampling‐day discharge and accumulative lagged precipitation should be prioritized as predictors. Precipitation‐discharge interaction effects on SSC exhibit simple threshold effects, whereas the interaction effects of hydrological (precipitation, discharge) and environmental (SPEI, land cover) factors involve complex, bidirectional threshold effects. (b) LightGBM and XGBoost excel in long‐term/general prediction, while RF outperform for short‐term/extreme value predictions. (c) Conformal prediction‐based UQ provides probabilistic information to quantify prediction reliability and efficiency, alongside uncertainty sources: discharge (∼38.9%) > precipitation (∼33.4%) > land cover (∼19.6%) > SPEI (∼8.1%). This framework advances trustworthy ML in riverine sediment modeling, while its algorithm‐agnostic design ensures potential scalability to support broader hydrological applications and informed environmental decision‐making. Plain Language Summary Sediment dynamics are complexly shaped by diverse factors, tied to many ecological impacts such as aquatic ecosystem degradation, habitat alteration, and reservoir silting. Trustworthy sediment prediction facilitates targeted measures establishment by policymakers. Machine learning offers promising and flexible avenues for such predictions; however, as a data‐driven method, its “black‐box” nature reduces user trust and hinders its applications, particularly in ecological fields where decisions carry far‐reaching consequences. Researchers have recognized that interpretable and uncertainty quantification are indispensable for guiding effective communication, scientific discovery, and actionable insights. Yet, the integration of such methods with ML remains inadequate in sediment modeling. Therefore, this study proposes an integrative framework covering pivotal modeling aspects: rigorous predictor selection, high‐efficiency simulation, overall and local performance evaluation, distribution‐free uncertainty quantification, and enhanced interpretation. The framework proposed can refine model accuracy, interpretability, and practical applicability, while being extensible with an algorithm‐agnostic design to other hydrological components and spatiotemporal scales. Key Points The framework integrates time‐lag embedded feature selection, SHapley Additive exPlanations interpretation and conformal prediction to boost model credibility Sampling‐day discharge and accumulative lagged precipitation dominate river suspended sediment variations and are top‐priority predictors Prediction reliability links to current discharge, lagged rain, and land use predictors, while efficiency links to hydroclimate predictors

Reliable information of hydrological processes within a river basin is essentially required for developing an appropriate strategy for achieving sustainable development goals. The present study assesses the streamflow of a pro-glacial stream and also intends to estimate the contribution of suspended sediments, erosion rate, and the headwater contribution of the Panchachuli glacier. A field study during ablation period was carried out to measure streamflow and suspended sediment concentration (SSC). Further, HBV model was used to estimate the snowmelt. The average seasonal streamflow and SSC during the gauging period (July to October) for the basin were measured to be 7.17 m3/s, and 1.52 g/l in 2018, and 6.84 m3/s, and 1.21 g/l in 2019, respectively. Snowmelt contribution in total streamflow was 54.75% in 2018 which is reduced to 49.16% in 2019. Similarly, glacier melt contributes to 32.62% of its total runoff share in 2018 which was reduced to 28.73% in 2019. The rainfall runoff in total runoff increased to 12.62% from 2018 to 2019. Rainfall-runoff in its total runoff contribution showed an increased share of 22.13% in 2019. The streamflow, SSC, and suspended sediment load (SSL) showed a strong positive correlation for both the years. The suspended sediment yield (SSY), SSL, and erosion rate of the basin were found as high as compared to the other Himalayan basins in Himachal Pradesh, Jammu and Kashmir, and Ladakh and non-Himalayan regions that was found low when compared to other glaciers in Uttarakhand.

Conventional methods of measuring water discharge and suspended sediment concentration (e.g., water sampling and moving acoustic Doppler current profiler [ADCP]) present challenges in large tidal rivers due to temporal and spatial constraints. This study introduces a novel approach to monitor water discharge and suspended sediment discharge (SSD) in large tidal rivers. Total water discharge and SSD exhibit notable variability in tidal rivers due to the river–tidal interactions; understanding this variability and its causes is essential for effective tidal river management. From June to November 2023, a field study was conducted at Nanjing (NJ) to continuously monitor water discharge, suspended sediment concentration (SSC), and SSD in the tidal reaches of the Yangtze River using coastal acoustic tomography (CAT). Total water discharge ranged from 8,765 to 43,356 m3/s, with a mean of 27,825 m3/s, while tidal discharge varied between −11,998 and 9,983 m3/s, with a mean of 69 m3/s. SSC ranged from 0.02 to 0.09 kg/m3, and SSD ranged from 110 to 3,823 kg/s. Tidal variations in SSC and SSD were within ±0.04 kg/m3 and −1,252 to 1,410 kg/s, respectively. Over short timescales, tides caused instantaneous fluctuations in velocity, water discharge, and SSD, with tides contributing −40% to instantaneous water discharge and SSD at NJ. Over seasonal timescales, no significant wet/dry variations were observed in water discharge, SSC, or SSD during a few months of 2023. Long‐term CAT application (e.g., decades) is required to reveal trends in tidal river dynamics. Plain Language Summary Due to temporal and spatial limitations, traditional methods for measuring suspended sediment concentration (SSC) and discharge, such as moving acoustic Doppler current profilers (ADCP), fail to directly measure transect variations in water discharge, SSC, and SSD in tidal reaches of the Yangtze River. This study developed a new method using coastal acoustic tomography (CAT). Two CAT systems were utilized to continuously measure water discharge, SSC, and SSD at the Nanjing Tidal Station. The CAT results were highly consistent with traditional methods, showing a correlation coefficient greater than 0.9. This study demonstrates the potential of CAT for continuous, real‐time monitoring of water discharge, SSC, and SSD in large tidal rivers. The results showed that mean water discharge, SSC, and SSD are primarily driven by river flow at Nanjing, while tides induce instantaneous variations in water discharge and sediment transport. Key Points Coastal acoustic tomography enabled water discharge and suspended sediment discharge (SSD) monitoring in Nanjing tidal reach of Yangtze River Total water discharge and SSD at Nanjing varied from 8,765 to 43,356 m3/s and 110–3,823 kg/s from June to November 2023, respectively Tides can directly trigger instantaneous variations in sediment discharge, while average sediment discharge is river‐dominated at Nanjing

Dams disrupt the natural flow of water and sediment along rivers. Reservoirs trap a significant amount of sediment, which substantially alters downstream hydrology, channel morphology, and sediment transport capacity. The longitudinal recovery of suspended sediment concentration (SSC) along rivers is potentially a new metric for estimating downstream responses to dams over space, rather than time, but is rarely quantified due to the lack of spatial SSC data. Satellites can estimate SSC along rivers where no field data exist and provide a high enough spatial resolution for assessing downstream recovery at the scale of tens to hundreds of kilometers. Here, we use a recently published database of spatially explicit SSC observations derived from Landsat to quantify if a river recovers or not, the SSC recovery percentage, and SSC recovery length downstream of large dams across the Contiguous United States (CONUS). Rivers recover SSC downstream of most dams (71%). The chance of a river recovering and the length of river required to recover SSC is associated primarily with the size of the reservoir (mean storage, km3) and the size of the river (mean discharge, m3/s). Rivers were more likely to recover SSC downstream of run‐of‐river, navigation dams compared to large storage or hydropower dams. Our results suggest that rivers typically recover suspended sediment downstream of dams, influenced by factors like dam storage, purpose, and river channel characteristics. Key Points Satellite‐derived suspended sediment concentration (SSC) shows rivers recovery to pre‐reservoir SSC downstream of 71% of dams in the US The length of river required to longitudinally recover SSC downstream of dams is most strongly associated with reservoir and river size Multidecadal satellite SSC observations provide a new metric for assessing the long‐term mean downstream response to dams

Machine learning (ML) has become a powerful tool for predicting suspended sediment concentration (SSC). Nonetheless, the ability to interpret the physical process is considered the main issue in applying most of ML approaches. In this regard, the current study presents a novel framework involving four standalone ML models (extra trees (ET), random forest (RF), categorical boosting (CatBoost), and extreme gradient boosting (XGBoost)) and their combination with genetic programming (GP). Three metrics (coefficient of correlation (r), root mean square error (RMSE), and Nash–Sutcliffe model-fit efficiency (NSE)) and a more advanced interpretation system SHapley Additive exPlanations (SHAP) are used to assess the performance of these models applied to hydro-climatic datasets for prediction of SSC. The calibration process was based on data from 2016 to 2020, and the validation was done for 2021 data. Further description and application of the framework are provided based on a case study of the Bouregreg watershed. The results revealed that all implemented models are efficient in SSC prediction with NSE, RMSE, and r varying from 0.53 to 0.86, 1.20–2.55 g/L, and 0.83–0.91 g/L respectively. Box plot diagrams confirm the enhanced performance of these combined models, and the best-performing ones for the four hydrological stations being the combined RF + GP model at the Aguibat Ziar station, the combined XGBoost + GP model at the Ain Loudah station, the CatBoost model at the Ras Fathia station, and the RF model at the Sidi Med Cherif station. The interpretability results showed that flow (Q) and seasonality (S) are the features most impacting SSC. These outcomes indicate that the applied models can extract accurate and detailed information from the interactions between the hydroclimatic factors and the generation of sediment by erosion (output). ML approaches illustrated the good reliability and transparency of the models developed for predicting SSC in a semi-arid setting, offered new perspectives for reducing ML models' “black box” character, and provided a useful source of information for assessing the consequences of SSC on water quality. The SHAP system and exploring other interpretable techniques are recommended to provide further information in future research. In addition, incorporating additional input data could enhance SSC predictions and deepen understanding of sediment transport dynamics.

Suspended sediment concentration prediction is critical for the design of reservoirs, dams, rivers ecosystems, various operations of aquatic resource structure, environmental safety, and water management. In this study, two different machine models, namely the cascade correlation neural network (CCNN) and feedforward neural network (FFNN) were applied to predict daily-suspended sediment concentration (SSC) at Simga and Jondhara stations in Sheonath basin, India. Daily-suspended sediment concentration and discharge data from 2010 to 2015 were collected and used to develop the model to predict suspended sediment concentration. The developed models were evaluated using statistical indices like Nash and Sutcliffe efficiency coefficient (N ES ), root mean square error (RMSE), Willmott’s index of agreement (WI), and Legates–McCabe’s index (LM), supplemented by a scatter plot, density plots, histograms and Taylor diagram for graphical representation. The developed model was evaluated and compared with CCNN and FFNN. Nine input combinations were explored using different lag-times for discharge (Q t-n ) and suspended sediment concentration (S t-n ) as input variables, with the current suspended sediment concentration as the desired output, to develop CCNN and FFNN models. The CCNN4 model with 4 lagged inputs (S t-1 , S t-2 , S t-3 , S t-4 ) outperformed the other developed models with the lowest RMSE = 95.02 mg/l and the highest N ES  = 0.0.662, WI = 0.890 and LM = 0.668 for the Jondhara Station while the same CCNN4 model secure as the best with the lowest RMSE = 53.71 mg/l and the highest N ES  = 0.785, WI = 0.936 and LM = 0.788 for the Simga Station. The result shows the CCNN model was better than the FFNN model for predicting daily-suspended sediment at both stations in the Sheonath basin, India. Overall, CCNN showed better forecasting potential for suspended sediment concentration compared to FFNN at both stations, demonstrating their applicability for hydrological forecasting with complex relationships.

In‐situ observations of hydrodynamics and suspended sediment concentrations (SSCs) were conducted on an abandoned lobe in the northern part of the modern Yellow River Delta, China. The SSC record at the site is found to be the superposition of a general trend (fast increase and slow decrease cycle) caused by storm waves (SubSSC1) and relatively smaller fluctuations caused by tidal currents (SubSSC2). Physically, this indicates that storm waves eroded the bottom sediments while tidal currents then re‐suspended and advected the suspended sediments in the study area. To further obtain the suspended sediment transport parameters, first, SubSSC1 is modeled with significant wave height which incorporates a “memory curve” to consider the remaining impacts of historical waves. It is detected that waves in the past 75 hr still influence the present SSC which is reasonable because 75 hr is roughly the typical duration of a normal storm. Second, SubSSC2 is modeled with tidal excursion and trigonometric functions with measured periodicities. Finally, some sediment transport parameters, for example, the background SSC, the horizontal SSC gradient, the tidal constituents that advect it, and their relative time lags are optimized from the best fits of the measured and modeled SSC time series. The proposed framework for model construction and parameter optimization can be extended to other sea areas for inferring sediment transport parameters from field SSC time series at a specific station. Plain Language Summary The evolution of our coastal zone fundamentally depends on the transport volume and direction of sediment, and understanding them is very beneficial for our coastal engineering construction and long‐term planning. Conducting in situ observations on‐site is one of the most reliable methods, but its observation cost is expensive. Therefore, we hope to extract as much information as possible from as few observation points as possible. This article successfully extracted the above information from the observation data of a station. First, we analyzed the data to preliminarily clarify that the suspended sediment in the area mainly comes from local erosion resuspension and advection transport from other regions. Furthermore, we conducted data modeling (i.e., constructed mathematical expressions of suspended sediment components from different sources, but with undetermined coefficients). Finally, we adjusted the model parameters to approximate the measured results, determined the undetermined coefficients, and explained the practical significance of the undetermined coefficients in physics. The analysis method we proposed can be extended to other sea areas. Key Points Storm wave‐induced suspended sediment concentration (SSC) variation is modeled with a “memory curve” of wave height Tidally‐induced SSC variation is modeled with tidal excursion and trigonometric functions Sediment transport parameters are estimated from the optimal matching of measured and modeled SSC time series

【Background and Objective】Estuarine sluices are widely used for flood control and water management, but their construction can significantly alter tidal dynamics and sediment transport, leading to downstream siltation, which affects navigation, ecology and flood safety. Understanding these processes is critical for sustainable management of tidal rivers. This paper investigates the impact of the Baxianyan sluice on tidal dynamics and downstream siltation in the Jiao (Ling) River estuary. 【Method】We used numerical simulation and physical modeling to systematically analyze hydrodynamic and sedimentary responses, as well as the spatial patterns of siltation after sluice construction The numerical model simulated the changes in tidal flux, flow velocities and sediment transport, while the physical model reproduced the morphological evolution under controlled laboratory conditions.【Result】In the near-field zone (within 10 km), the sluice reduced the high-and low-tide volumes by more than 30%, and the flow velocity by 75%-90%, accompanied by a synchronous attenuation of sediment concentration. In the mid-to-far zone (10-40 km), tidal currents gradually recovered, and sediment transport exhibited an ‘upstream migration from downstream’ pattern, reflecting a dynamic cycle of ‘tidal energy attenuation-sediment trapping- resuspension’ During dry years, sediment deposition evolved in four stages: ‘rapid initial accumulation-progressive balance-downstream migration-distal stabilization’ Five years after the sluice construction, total siltation asymptotically reached 23 million m3. The numerical model accurately captured the longitudinal gradients and lateral heterogeneity of the sediment deposition, while the physical model accurately reproduced the large-scale morphological changes, both showing that tidal energy attenuation, channel widening, and restructuring of estuarine hydrodynamics were the primary drivers of downstream siltation.【Conclusion】Our results demonstrate that the combination of numerical and physical models are robust for predicting and analyzing siltation processes in estuarine sluice projects. The identified ‘tidal energy attenuation-channel widening-hydrodynamic reconstruction’ mechanism can help improve sediment management in tidal estuarine engineering constructions.

This study examines the local, intratidal effects of suspended sediment concentrations (SSCs) on the hydrodynamics and vertical mixing in the Ems Estuary, located on the border between Germany and The Netherlands, during summer and winter seasons when the estuary turbidity maximum (ETM) is located upstream and adjacent to the study site, respectively. Measurements of density, SSCs, turbulent kinetic energy dissipation, and current velocity were collected and analyzed over a semi-diurnal tidal cycle in August of 2018 and January of 2019 as part of the collaborative Ems-Dollard Measurement (EDoM) campaign. During August, the estuary turbidity maximum was located 25 km upstream from the measurement site and local SSCs were low. Results revealed that under these conditions, suspended sediment minimally impacted vertical mixing by stabilizing density near-bottom during flood tide, while typical salinity-induced tidal straining patterns dominated. During January, the ETM was located only 5 km upstream of the measurement site leading to higher local sediment concentrations. Salinity-induced straining of the density occurred on early flood tide, creating stratification that suppressed vertical mixing. The suppression was enhanced by the contribution of vertical gradients in SSC to density, as signified by the gradient Richardson number. Suppression of vertical mixing by sediment-enhanced stratification was most significant within the hour following maximum flood currents when elevated velocity shear occurred. The variability observed between the local dynamics during August and January were attributed to greater sediment concentrations due to the ETM proximity in January. The intratidal asymmetry of vertical mixing observed under higher SSCs likely has implications for sediment transport.