The sediment transport, involving the movement of the bedload and suspended sediment in the basins, is a critical environmental concern that worsens water scarcity and leads to degradation of land and its ecosystems. Machine learning (ML) algorithms have emerged as powerful tools for predicting sediment yield. However, their use by decision-makers can be attributed to concerns regarding their consistency with the involved physical processes. In light of this issue, this study aims to develop a physics-informed ML approach for predicting sediment yield. To achieve this objective, Gaussian, Center, Regular, and Direct Copulas were employed to generate virtual combinations of physical of the sub-basins and hydrological datasets. These datasets were then utilized to train deep neural network (DNN), conventional neural network (CNN), Extra Tree, and XGBoost (XGB) models. The performance of these models was compared with the modified universal soil loss equation (MUSLE), which serves as a process-based model. The results demonstrated that the ML models outperformed the MUSLE model, exhibiting improvements in Nash–Sutcliffe efficiency (NSE) of approximately 10%, 18%, 32%, and 41% for the DNN, CNN, Extra Tree, and XGB models, respectively. Furthermore, through Sobol sensitivity and Shapley additive explanation–based interpretability analyses, it was revealed that the Extra Tree model displayed greater consistency with the physical processes underlying sediment transport as modeled by MUSLE. The proposed framework provides new insights into enhancing the accuracy and applicability of ML models in forecasting sediment yield while maintaining consistency with natural processes. Consequently, it can prove valuable in simulating process-related strategies aimed at mitigating sediment transport at watershed scales, such as the implementation of best management practices.

The acid volatile sulphide (AVS) and simultaneously extracted metals (ΣSEM) method is increasingly used for risk assessment of toxic metals. In this study, we assessed spatial and temporal variations of AVS and ΣSEM in river sediments and floodplain soils, addressing influence of flow regime and flooding. Slow-flowing sites contained high organic matter and clay content, leading to anoxic conditions, and subsequent AVS formation and binding of metals. Seasonality affected these processes through temperature and oxygen concentration, leading to increased levels of AVS in summer at slow-flowing sites (max. 37 μmol g-1). In contrast, fast-flowing sites hardly contained AVS, so that seasonality had no influence on these sites. Floodplain soils showed an opposite AVS seasonality because of preferential inundation and concomitant AVS formation in winter (max. 3-30 μmol g-1). We conclude that in dynamic river systems, flow velocity is the key to understanding variability of AVS and ΣSEM. Flow velocity is the key to understanding variability of AVS and ΣSEM in river sediment.

Polycyclic Aromatic Hydrocarbons (PAHs) transported by surface runoff result in nonpoint source pollution and jeopardize aquatic ecosystems. The transport mechanism of PAHs during rainfall-runoff events has been rarely studied regarding pervious areas. An experimental system was setup to simulate the runoff pollution process on PAHs-contaminated soil. The enrichment behavior of soil-bound PAHs was investigated. The results show that soil organic matters (SOM), rather than clay particles, seem to be the main carrier of PAHs. The enrichment is highly conditioned on runoff and erosion processes, and its magnitude varies among PAH compounds. It is not feasible to build a simple and universal relationship between enrichment ratio and sediment discharge following the traditional enrichment theory. To estimate the flux of PAHs from pervious areas, soil erosion process has to be clearly understood, and both organic carbon content and composition of SOM should be factored into the calculation.

The Soil and Water Assessment Tool (SWAT) was used to assess the impact of climate change on sediment, nitrate, phosphorus and pesticide (diazinon and chlorpyrifos) runoff in the San Joaquin watershed in California. This study used modeling techniques that include variations of CO2, temperature, and precipitation to quantify these responses. Precipitation had a greater impact on agricultural runoff compared to changes in either CO2 concentration or temperature. Increase of precipitation by ±10% and ±20% generally changed agricultural runoff proportionally. Solely increasing CO2 concentration resulted in an increase in nitrate, phosphorus, and chlorpyrifos yield by 4.2, 7.8, and 6.4%, respectively, and a decrease in sediment and diazinon yield by 6.3 and 5.3%, respectively, in comparison to the present-day reference scenario. Only increasing temperature reduced yields of all agricultural runoff components. The results suggest that agricultural runoff in the San Joaquin watershed is sensitive to precipitation, temperature, and CO2 concentration changes. Agricultural runoff is significantly affected by changes in precipitation, temperature, and atmospheric CO2 concentration.

A dynamic flux model for lakes taking into account the interactions between atmospheric depositional and biogeochemical processes (BIODEP model) was used to assess the importance of internal cycling and biogeochemical processes with respect to accumulation of four different polychlorinated biphenyls (PCBs) (congeners 28, 52, 101 and 153) in Lake Redo, a high-altitude lake in the Spanish Pyrenees. To investigate the effect of temperature on sediment accumulation, the model was run at different temperatures and corresponding sediment inventories were plotted vs. reciprocal temperature. In line with measurements from a previous study looking at sediment inventories of 19 European high-altitude lakes (MOLAR study), we found a stronger temperature dependence of lake sediment concentrations for the less volatile/less soluble PCBs. Seasonal and annual mass balances were investigated and highlighted the importance of internal lake processes controlling the differences in sediment accumulation for the different PCB congeners. For example, the higher temperature dependence of sediment inventories for the high chlorinated congeners is due to the fast dynamics of water-to-sediment transport in comparison to atmospheric deposition processes. A dynamic flux model was used to assess the importance of internal lake processes in controlling sediment accumulation for different PCB congeners.

High resolution sediment records in the Yangtze Delta front were constructed to reveal recent environmental changes in response to river basin human activities. Increases in nutrient and organic C influxes that began in the 1950s, together with elevated primary productivity and increased chemical fertilizer application, suggested a shift toward anthropogenic-predominated environmental changes during this period. The depletion of total organic C (TOC), total N (TN), and biogenic Si (BSi), along with the decline in sedimentation rate and coarsening of sediment coincided with the development of hydrological engineering in the river basin from the 1980s. Reservoir Si retention substantially altered river mouth primary productivity community composition from diatoms to non-diatoms, thereby changing the BSi/TOC molar ratio in the sediment profile. Estimation of biogenic component burial fluxes was conducted to assess the variation and potential impacts. A recent dramatic decline in biogenic component burial in the delta area suggested a low nutrient removal efficiency in this region, due to the decrease in sediment discharge. Consequently, more nutrients have been further transported to the inner shelf and open waters instead of being buried in the delta sediment, thereby increasing the environmental pressure in the Yangtze Delta and adjoining coastal area.

In remote, tropical areas human influences increase, potentially threatening pristine seagrass systems. We aim (i) to provide a bench-mark for a near-pristine seagrass system in an archipelago in East Kalimantan, by quantifying a large spectrum of abiotic and biotic properties in seagrass meadows and (ii) to identify early warning indicators for river sediment and nutrient loading, by comparing the seagrass meadow properties over a gradient with varying river influence. Abiotic properties of water column, pore water and sediment were less suitable indicators for increased sediment and nutrient loading than seagrass properties. Seagrass meadows strongly responded to higher sediment and nutrient loads and proximity to the coast by decreasing seagrass cover, standing stock, number of seagrass species, changing species composition and shifts in tissue contents. Our study confirms that nutrient loads are more important than water nutrient concentrations. We identify seagrass system variables that are suitable indicators for sediment and nutrient loading, also in rapid survey scenarios with once-only measurements.

Understanding of the pre-development, baseline denudation rates that deliver sediment to the Great Barrier Reef (GBR) has been elusive. Cosmogenic ¹⁰Be in sediment is a useful integrator of denudation rates and sediment yields averaged over large spatial and temporal scales. This study presents ¹⁰Be data from 71 sites across 11 catchments draining to the GBR: representing 80% of the GBR catchment area and provide background sediment yields for the region. Modern, short-term, sediment yields derived from suspended load concentrations are compared to the ¹⁰Be data to calculate an Accelerated Erosion Factor (AEF) that highlights denudation “hot-spots” where sediment yields have increased over the long-term background values. The AEF results show that 58% basins have higher modern sediment yields than long-term yields. The AEF is considered a useful approach to help prioritise on-ground investments in remediation and the additional measured empirical data in this paper will help support future predictive models.

Increased sediment loads within algal turfs, can be highly detrimental to coral reef systems. However, significant knowledge gaps remain in relation to sediment dynamics, especially linking suspended sediments, sedimentation and turf-bound sediments. To examine these links, a series of different methods for quantifying suspended sediments, sedimentation and the accumulation of turf sediments were compared, simultaneously, on an inner-shelf reef. We revealed that the amount and composition of sediment quantified using different methods varied markedly, with commonly employed measures of sedimentation failing to accurately reflect patterns of sediment accumulation in turfs. Our results highlighted the propensity for turfs to trap and retain sediments, with turfs accumulating approximately 2.6 times more sediment than traps, and 6 times more sediment than SedPods, over a seven-day period. This study highlights the major, but often overlooked, role that algal turfs can play in sediment dynamics on coral reefs.