Urban stormwater outfall pollution is characterized by abrupt onset and drastic fluctuations, making it difficult for conventional monitoring methods to achieve rapid response. Taking the typical stormwater outfall in Xiamen as a case study, this research developed a machine learning-based rapid water quality estimation method. Two ensemble learning models, Random Forest (RF) and eXtreme Gradient Boosting (XGBoost), were compared, and the Harris Hawks Optimization (HHO) algorithm was introduced for automatic hyperparameter tuning. The results showed that both models performed poorly under default hyperparameters, while HHO optimization significantly improved model accuracy. Specifically, the HHO-XGBoost model achieved R2 values of 0.905 and 0.866 for NH3-N and TN, respectively. Feature importance and SHapley Additive exPlanations (SHAP) analyses revealed that HHO optimization enhances model interpretability by suppressing redundant feature interference and strengthening the weights of dominant variables. Furthermore, by comparing HHO with Random Search and Bayesian Optimization, this study confirmed that HHO possesses stronger global search capability in nonlinear hyperparameter spaces. The findings demonstrate that a machine learning framework driven by easily measurable and real-time monitoring indicators can serve as an effective supplement to existing monitoring systems, providing technical support for trend diagnosis and rapid screening of urban stormwater outfall pollution.
This study assesses how different types of anthropogenic impact affect the layer-by-layer distribution of soil organic carbon (SOC) stocks in soils of the Rostov agglomeration. The objects studied were natural Haplic Chernozems under herbaceous and woody phytocenoses, anthropogenically transformed urbic horizons within Urbic Technosols, and the natural buried horizons beneath them. In total, 80 soil pits were described; total, inorganic and organic carbon contents were determined by high-temperature catalytic combustion, and stocks were calculated from bulk density for successive 10 cm layers to a depth of 70 cm. Differences were evaluated using the Mann–Whitney test. Natural Haplic Chernozems under both herbaceous and woody vegetation showed a pronounced four-member vertical stratification of carbon stocks: a surface maximum, a transitional zone of declining organic matter input, an intermediate zone of relative statistical homogeneity, and a distinct lower horizon with minimum stocks; under woody phytocenoses, the boundaries of these zones were displaced to greater depths. Urbic Technosols displayed pseudo-homogenisation from 20 cm downwards with high individual variability, although their median SOC stocks were generally comparable to natural analogues, possibly due to reclamation practices. The most pronounced carbon-pool degradation occurred in Haplic Chernozems buried beneath technogenic deposits: natural stratification was preserved, but absolute stocks were significantly lower than in all other groups, consistent with cessation of fresh organic matter input and depletion of previously sequestered carbon, or with partial removal of topsoil before burial. These results show that anthropogenic impact can both reduce absolute carbon stocks and distort their natural vertical architecture, which must be considered when assessing the carbon sequestration potential of urban soils.
Endocrine-disrupting chemicals (EDCs) are increasingly detected in drinking water sources due to their widespread use and environmental persistence. Their potential to interfere with hormonal systems, even at trace levels, poses serious concerns for human health, including developmental, reproductive, and metabolic effects. This systematic review aimed to identify and compare the most effective treatment technologies to remove EDCs from drinking water, assessing their performance, analytical approaches, limitations, and relevance to public health and regulatory frameworks. Following PRISMA guidelines, a comprehensive literature search was conducted across 4 databases (PubMed, Scopus, Web of Science, Cochrane Library), resulting in the inclusion of 19 experimental studies published from the last 20 years. The studies were assessed for methodological quality and risk of bias using the Cochrane ROB-2 tool. A wide spectrum of EDCs was studied, including estrogens, per- and polyfluoroalkyl substances (PFAS), bisphenols, parabens, and drug residues, across diverse treatment configurations. Conventional methods (e.g., coagulation, chlorination) have shown limited efficacy; instead, advanced techniques such as ozonation, nanofiltration, activated carbon adsorption, and advanced oxidation processes (AOPs) have mainly achieved removal > 90
The study involved textural and geochemical analyses of 25 topsoil samples from the Lesser Zab River Basin, northeastern Iraq, including the source, distribution, and weathering behavior of trace elements (TEs) and rare earth elements (REEs) and their relationship with topsoil texture. Clay loam, silt loam, and silty clay loam are the predominant textures, with quartz, feldspar, calcite, and dolomite as dominant mineral phases and smectite and illite as the main clay minerals. Soils are moderately alkaline with low organic matter and comparatively high Fe2O3 content. While most TEs fell below Post-Archean Australian Shale (PAAS) reference values, Cr, Co, and Ni were markedly enriched-locally up to an order of magnitude above background. This enrichment reflects a mafic–ultramafic source distinct from the lithogenic REE factor, linked to ophiolitic contributions. Geo-accumulation index, enrichment factor (EF), and pollution index (PI) values indicated no pollution by Cu, Zn, As, Pb, and Cd; localized high Cr, Co, and Ni (Igeo up to 3.8, EF up to 59, PI up to 21) were linked to ophiolitic outcrops, indicating a geogenic origin of this anomaly. Total REEs averaged 55.92 mg/kg, significantly lower than the North American Shale Composite (NASC) and PAAS values, with LREE enrichment and HREE depletion (PAAS-normalized (La/Yb)n = 28.76–66.34).
Urban air pollution is a major environmental challenge in arid cities, where meteorological conditions strongly influence pollutant accumulation, transport, and dispersion. This study analyzed the relationship between meteorological variables and particulate matter (PM2.5 and PM10) concentrations in the San Juan Metropolitan Area, an arid urban environment in central western Argentina, over a short-term monitoring campaign during the cold season. Continuous air quality and meteorological monitoring was conducted across residential, industrial, and commercial zones. Generalized additive models (GAMs) were applied to evaluate the effects of air temperature, relative humidity, atmospheric pressure, wind speed, wind direction, and global radiation on PM concentrations. Significant differences in particulate matter concentrations were observed among urban land-use zones, with industrial and commercial zones showing the highest PM2.5 and PM10 levels, while residential zones presented lower concentrations. Pronounced diurnal variability was also detected, with concentration peaks occurring during morning and evening periods, particularly in commercial and industrial zones. Meteorological variables emerged as key drivers of PM variability, although their influence differed according to particle fraction and urban zone. Atmospheric pressure and wind speed were the most influential predictors across most models, emphasizing the role of atmospheric stability and ventilation processes in regulating PM accumulation and dispersion. PM2.5 was more sensitive to stable atmospheric conditions and low wind speeds, whereas PM10 was more strongly associated with wind-related resuspension processes. These findings provide new insights, based on a short-term monitoring campaign, into meteorological controls of particulate matter in arid urban environments and may support the development of effective air quality management strategies in regions with similar climatic and urban characteristics.
