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.

International audience | In Guadeloupe, the use between 1972 and 1993 of chlordecone, an organochlorine insecticide, has permanently contaminated the island’s soil, thus contaminating the food chain at its very beginning. There is today a strong societal requirement for an improved mapping of the contaminated zones. Given the extent of the areas to be covered, carrying out soil tests on each plot of the territory would be a long and expensive process. In this article, we explore a method of demarcating polluted areas. The approach adopted consists in carrying out, using surface water analyses, a hydrological delimitation that makes it possible to distinguish contaminated watersheds from uncontaminated ones. The selection of sampling points was based on the spatial analysis of the actual and potential contamination data existing at the beginning of the study. The approach was validated by soil analyses, after having compared the contamination data of the watersheds with the soil contamination data of the plots within them. The study thus made it possible to highlight new contaminated areas and also those at risk of contamination and to identify the plots to be targeted as a priority during future analysis campaigns by State services.

In this paper, we are consider processing of soils erosion and his way on protection expoitation of water for population water supply, melioration's systems for irrigation and for strew of manger of the Lim river with the tug and suspend bank (Montenegro, Yugoslavia). We are given suggestions for protective utilization of agricultural and forest lands and the possibilities of water exploitation from Bistrica river (Montenegro, Yugoslavia). That is useful for fishing, irrigation, energetic, tourism and water supply. In this paper, you can meet some suggestions about exploitations of minerals and organic fertilizars that are not toxics and they not any influence on pollution of land and water.

Hydrological, psamological and pluviographical measurements in experimental watershed area in the watershed of River Topciderska (Serbia, Yugoslavia) was analyzed. Interrelations between suspended sediment concentration in streams on the one hand and runoff, rainfall, rainfall intensity and index of previous rainfall on the other hand was came from this analyze. Effects of these parameters on suspended sediment concentration was considered for a single rainfall epizode, during the period of torrential waves and as annual value. Specific coefficients (coefficient of annual rainfall structure and coefficient of annual runoff structure) were used in this paper, with help of them effects of torrential waves and intensity rainfall on average annual suspended sediment concentration were considered.

In this paper are presented autochthonous and all other alochthonous famous fishes in Lake Ohrid and its catchement area (The Former Yugoslav Republic of Macedonia). Introduction of new species in lakes and rivers in the Former Yugoslav Republic of Macedonia is prohibited, especially in Lake Ohrid and catchement area. The most probable period and possible way of introduction are presented and the reason of that introduction, as well. It is necessary, in the future, more attention to be paid on Lake Ohrid protection from introduction of new species, which may cause big changes in Lake Ohrid dynamics.

Problems encountered in evaluating surface water quality arise from the requirements of relevant regulations that these evaluations should be made in the last favourable hydrologic conditions during design flow rate (required minimum flow). However, the major part of measurements of water quality parameters is not performed under such conditions and the result must be corrected for the conditions under required minimum flow. Two possible approaches to solving this problem are presented in this paper and explained using an example of the Juzna Morava river basin up to the Station Korvingrad, Serbia (Yugoslavia). This basin is unhomogeneous regarding its hydrologic characteristics and mass flow of pollutants. The results of analysis are presented in the cartography form using the GIS technology.

Salt marsh estuaries serve as sources and sinks for nutrients and elements to and from estuarine water, which enhances and alleviates watershed fluxes to the coastal ocean. We assessed sources and sinks of mercury in the intertidal Plum Island Sound estuary in Massachusetts, the largest salt marsh estuary of New England, using 25-km spatial water sampling transects. Across all seasons, dissolved (FHg) and total (THg) mercury concentrations in estuarine water were highest and strongly enhanced in upper marshes (1.31 ± 0.20 ng L⁻¹ and 6.56 ± 3.70 ng L⁻¹, respectively), compared to riverine Hg concentrations (0.86 ± 0.17 ng L⁻¹ and 0.88 ± 0.34 ng L⁻¹, respectively). Mercury concentrations declined from upper to lower marshes and were lowest in ocean water (0.38 ± 0.10 ng L⁻¹ and 0.56 ± 0.25 ng L⁻¹, respectively). Conservative mixing models using river and ocean water as endmembers indicated that internal estuarine Hg sources strongly enhanced estuarine water Hg concentrations. For FHg, internal estuarine Hg contributions were estimated at 26 g yr⁻¹ which enhanced Hg loads from riverine sources to the ocean by 44%. For THg, internal sources amounted to 251 g yr⁻¹ and exceeded riverine sources six-fold. Proposed sources for internal estuarine mercury contributions include atmospheric deposition to the large estuarine surface area and sediment re-mobilization, although sediment Hg concentrations were low (average 23 ± 2 μg kg⁻¹) typical of uncontaminated sediments. Soil mercury concentrations under vegetation, however, were ten times higher (average 200 ± 225 μg kg⁻¹) than in intertidal sediments suggesting that high soil Hg accumulation might drive lateral export of Hg to the ocean. Spatial transects of methylated Hg (MeHg) showed no concentration enhancements in estuarine water and no indication of internal MeHg sources or formation. Initial mass balance considerations suggest that atmospheric deposition may either be in similar magnitude, or possibly exceed lateral tidal export which would be consistent with strong Hg accumulation observed in salt marsh soils sequestering Hg from current and past atmospheric deposition.

Ponds, depressional submerged landscapes that can store and process nitrogen (N)-enriched runoff from surrounding uplands, are recognized as biogeochemical hotspots for N removal. Despite their strong potential for N removal, information is limited concerning the specifics of their changing nature. Here, we investigated the dynamics of N removal rate in a typical agricultural pond from a hilly catchment, by unraveling the monthly and diel patterns of N₂ concentrations and fluxes. Our observations showed that the N pollution in the pond was severe. Its averaged total N level reached 3.6 mg L⁻¹, of which ∼72% consisted of NO₃–N. Meanwhile, the water samples were supersaturated with N₂, demonstrating N removal occurring in the pond. Further estimates of net N₂ fluxes indicated that N removal rates exhibited obvious day-and-night and monthly differences. On the diel scale, N removal rates exhibited a distinct diurnal cycle, with nocturnal rates around 20% higher than during the day. Such a diel pattern can be mainly explained by the fluctuation in DO levels, showing that at nighttime when photosynthesis is absent, low DO environments are conducive to N removal. On a monthly scale, the monthly rates ranged from 0.02 to 0.49 mmol N₂ m⁻² h⁻¹ (mean: 0.23 mmol N₂ m⁻² h⁻¹), with generally higher removal rates in the warmer and concurrently rainy months (June–September). N levels in the pond were the corresponding primary explanatory variables. Assembled data from both monthly and hourly scales provided a more complete picture of the changing nature of N removal in ponds. Future work should carefully consider the effects of altered environmental conditions triggered by hydrological events to better reveal the control mechanisms behind the time-immediate N removal from lowland ponds.