Atmospheric mercury (Hg) poses human health and ecological risks once deposited and bio-accumulated through food chains. Source contribution analysis of Hg deposition is essential to formulating emission control strategies to alleviate the adverse impact of Hg release from anthropogenic sources. In this study, a Hg version of California Puff Dispersion Modeling (denoted as CALPUFF-Hg) system with added Hg environmental processes was implemented to simulate the Hg concentration and deposition in the central region of the Pearl River Delta (cPRD) at 1 km × 1 km resolution. The contributions of eight source sectors to Hg deposition were evaluated. Model results indicated that the emission from cement production was the largest contributor to Hg deposition, accounting for 13.0%, followed by coal-fired power plants (6.5%), non-ferrous metal smelting (5.4%), iron and steel production (3.5%), and municipal solid waste incineration (3.4%). The point sources that released a higher fraction of gaseous oxidized mercury, such as cement production and municipal solid waste incineration, were the most significant contributors to local deposition. In this intensive industrialized region, large point sources contributed 67–94% of total Hg deposition of 6 receptors which were the nearest grid-cells from top five Hg emitters of the domain and the largest municipal solid waste incinerator in Guangzhou. Based on the source apportionment results, cement production and the rapidly growing municipal solid waste incineration are identified as priority sectors for Hg emission control in the cPRD region.

Urban wastewater effluents bring large amounts of nutrients, organic matter and organic microcontaminants into freshwater ecosystems. The effects of this complex mixture of pollutants on freshwater invertebrates have been studied mainly in temperate rivers and streams with high dilution capacities. In contrast, Mediterranean streams and rivers have lower dilution capacities especially during the seasonal drought, and are therefore exposed to high concentrations of pollutants. Here, we assess the effects of urban wastewater pollution on invertebrate communities from Mediterranean streams under different hydrological conditions. Specifically, we assessed the invertebrates taxonomic composition and functional biological traits in 12 streams, differing in stream and substrate size (sand or cobbles), under low (2 surveys) and baseflow (1 survey) conditions. In each stream, we selected reaches both upstream and downstream of the wastewater discharge point. Our results indicate that urban wastewater pollution favours the most tolerant invertebrate taxa and homogenises functional trait composition over time. Changes in functional traits were more evident during the seasonal drought, when the low flow conditions at the upstream and downstream sites were more severe and, pollutant concentrations downstream were at their highest. However, the effects of urban wastewater pollution were not uniform in the downstream sites; as local invertebrate communities differed in according to the river substrate and stream size (i.e., width and discharge). Overall, urban pollution caused by wastewater enhanced both, taxonomic and functional differences between the invertebrate communities. Such an absence of homogenisation among wastewater pollution impacted sites was probably related to the relevant role of stream substrate-size as well flow conditions in the rivers receiving the impact. These are attributes that need to be considered when setting the pollutant discharge limits in rivers and streams receiving effluents.

In recent decades, most big cities in China have experienced severe air pollution accompanied by rapid economic and social development. Analysis of measurements of air pollutants form a fundamental basis for understanding the characteristics of air pollution and are important references for policy-making. In this study, five-year measurements of air pollutants at 6 sites in Lhasa, a typical high altitude big city in southwestern China, were analyzed from January 2013 to December 2017. Air pollutants at all the 6 sites in Lhasa generally displayed similar patterns of both diurnal and monthly variations, indicating the mixed atmospheric environment and the overall effect of the meteorological conditions in the city. Spatially, the air pollutant concentrations at the 6 sites were generally characterized by high concentrations of SO₂, NO₂, CO, PM₁₀ and PM₂.₅ at urban sites and high O₃ concentrations at suburban sites. In comparison with other provincial capital cities in China, Lhasa has low concentrations of air pollutants, except for O₃, and thus, better air quality. Although Lhasa has experienced rapid urbanization and economic development, air pollution conditions have remained rather stable and even decreased slightly in term of particular air pollutants. We suggested that the relatively isolated location, low air pollutant emissions associated with its industrial structure and renewable energy consumption, and effective air pollution control measures, collectively contributed to the synchronous improvement of the economy and air quality in Lhasa. Such “Lhasa pattern” may serve as a positive example for other regional hub cities in China and beyond that experience socioeconomic development and simultaneously seek to improve air quality.

Changzhou, an industrial city in the Yangtze River Delta, has been experiencing serious haze pollution, particularly in winter. However, studies pertaining to the haze in Changzhou are very limited, which makes it difficult to understand the characteristics and formation of winter haze in this area, and develop effective control measures. In this study, we carried out continuous online observation of particulate matter, chemical components, and meteorology in Changzhou in February 2017. Our results showed that haze pollution occurred frequently in Changzhou winter and exhibited two patterns: dry haze with low relative humidity (RH) and wet haze with high RH. Water-soluble inorganic ions (SO₄²⁻, NO₃⁻, and NH₄⁺) accounted for ∼52.2% of the PM₂.₅ mass, of which sulfate was dominant in wet haze periods while nitrate was dominant in other periods. With the deterioration of haze pollution, the proportion of nitrate in PM₂.₅ increased, while sulfate proportion increased under wet haze and decreased under dry haze. Dry haze and wet haze appeared under slow north wind and south wind, respectively, and strong north wind or sea breeze scavenged pollution. We found that formation of nitrate occurred rapidly in daytime with high concentrations of odd oxygen (Oₓ = O₃ + NO₂), whereas formation of sulfate occurred rapidly during nighttime with high RH, indicating that photochemistry and heterogeneous reaction were the major formation mechanisms for nitrate and sulfate, respectively. Through the cluster analysis of 36-h backward trajectories, five sources of air masses from three directions were identified. High PM₂.₅ concentrations (84.1 μg m⁻³ on average) usually occurred under the influence of two clusters (46%) from the northwest, indicating that regional transport from northern China aggravated the winter haze pollution in Changzhou. Emission reduction, particularly the mobile sources, and regional joint prevention and control can help to mitigate the winter haze in Changzhou.

Epidemiologists use prediction models to downscale (i.e., interpolate) air pollution exposure where monitoring data is insufficient. This study compares machine learning prediction models for ground-level ozone during wildfires, evaluating the predictive accuracy of ten algorithms on the daily 8-hour maximum average ozone during a 2008 wildfire event in northern California. Models were evaluated using a leave-one-location-out cross-validation (LOLO CV) procedure to account for the spatial and temporal dependence of the data and produce more realistic estimates of prediction error. LOLO CV avoids both the well-known overly optimistic bias of k-fold cross-validation on dependent data and the conservative bias of evaluating prediction error over a coarser spatial resolution via leave-k-locations-out CV. Gradient boosting was the most accurate of the ten machine learning algorithms with the lowest LOLO CV estimated root mean square error (0.228) and the highest LOLO CV Rˆ2 (0.677). Random forest was the second best performing algorithm with an LOLO CV Rˆ2 of 0.661. The LOLO CV estimates of predictive accuracy were less optimistic than 10-fold CV estimates for all ten models. The difference in estimated accuracy between the 10-fold CV and LOLO CV was greater for more flexible models like gradient boosting and random forest. The order of estimated model accuracy depended on the choice of evaluation metric, indicating that 10-fold CV and LOLO CV may select different models or sets of covariates as optimal, which calls into question the reliability of 10-fold CV for model (or variable) selection. These prediction models are designed for interpolating ozone exposure, and are not suited to inferring the effect of wildfires on ozone or extrapolating to predict ozone in other spatial or temporal domains. This is demonstrated by the inability of the best performing models to accurately predict ozone during 2007 southern California wildfires.

In the past few decades, extensive epidemiological studies have focused on exploring the adverse effects of PM₂.₅ (particulate matters with aerodynamic diameters less than 2.5 μm) on public health. However, most of them failed to consider the dynamic changes of population distribution adequately and were limited by the accuracy of PM₂.₅ estimations. Therefore, in this study, location-based service (LBS) data from social media and satellite-derived high-quality PM₂.₅ concentrations were collected to perform highly spatiotemporal exposure assessments for thirteen cities in the Beijing-Tianjin-Hebei (BTH) region, China. The city-scale exposure levels and the corresponding health outcomes were first estimated. Then the uncertainties in exposure risk assessments were quantified based on in-situ PM₂.₅ observations and static population data. The results showed that approximately half of the population living in the BTH region were exposed to monthly mean PM₂.₅ concentration greater than 80 μg/m³ in 2015, and the highest risk was observed in December. In terms of all-cause, cardiovascular, and respiratory disease, the premature deaths attributed to PM₂.₅ were estimated to be 138,150, 80,945, and 18,752, respectively. A comparative analysis between five different exposure models further illustrated that the dynamic population distribution and accurate PM₂.₅ estimations showed great influence on environmental exposure and health assessments and need be carefully considered. Otherwise, the results would be considerably over- or under-estimated.

In this study, carbon nanotube-based adsorbents, oxidized multi-walled carbon nanotube (OMWCNT) with non-magnetic property and OMWCNT-Fe₃O₄ and OMWCNT-κ-carrageenan-Fe₃O₄ nanocomposites with magnetic property, having different structural and surface properties were prepared and their adsorptive properties for the removal of toxic diquat dibromide (DQ) herbicide from water by adsorption were determined in detail. For each adsorption system, the effects of initial DQ concentration, contact time and temperature on the adsorption processes were determined. Equilibrium time was found to be 300 min for DQ solutions. OMWCNT showed faster adsorption and higher maximum adsorption capacity value than magnetic adsorbents. With increasing initial herbicide concentration from 5.43 mg.L⁻¹ to 16.3 mg.L⁻¹, the values of initial sorption rate exhibited a decrease from 29.1 mg.g⁻¹.min⁻¹ to 4.28 mg.g⁻¹.min⁻¹ for OMWCNT-DQ system, from 1.21 mg.g⁻¹.min⁻¹ to 0.823 mg.g⁻¹.min⁻¹ for OMWCNT-Fe₃O₄-DQ system and from 0.674 mg.g⁻¹.min⁻¹ to 0.612 mg.g⁻¹.min⁻¹ OMWCNT-κ-carrageenan-Fe₃O₄ system. Maximum adsorption capacity value of OMWCNT was approximately 2.8-fold higher than magnetic OMWCNT-Fe₃O₄ and 5.4-fold higher than magnetic OMWCNT-κ-carrageenan-Fe₃O₄ at 25 °C. Adsorption kinetic and isotherm data obtained for all adsorption systems were well-fitted by pseudo second-order and Langmuir models, respectively. Thermodynamic parameters indicated that the adsorption of DQ onto carbon nanotube-based adsorbents was spontaneous and endothermic process. Furthermore, OMWCNT having the highest herbicide adsorption capacity could be regenerated and reused at least five times. This study showed that carbon nanotube-based adsorbents with magnetic and non-magnetic property were of high adsorption performance for the removal of DQ from water and could be promising adsorbent materials for the efficient removal of herbicides from wastewaters.

Reduction of particulate matter (PM) has emerged as one of the most significant challenges in public health and environment protection worldwide. To address PM-related problems and effectively remove fine particulate matter (PM2.5), environmentalists proposed tree planting and afforestation as eco-friendly strategies. However, the PM removal effect of plants and its primary mechanism remains uncertain. In this study, we experimentally investigated the PM removal performance of five plant species in a closed chamber and the effects of relative humidity (RH) caused by plant evapotranspiration, as a governing parameter. On the basis of the PM removal test for various plant species, we selected Epipremnum aureum (Scindapsus) as a representative plant to identify the PM removal efficiency depending on evapotranspiration and particle type. Results showed that Scindapsus yielded a high PM removal efficiency for smoke type PM2.5 under active transpiration. We examined the correlation of PM removal and relative humidity (RH) and evaluated the increased effect of RH on PM2.5 removal by using a plant-inspired in vitro model. Based on the present results, the increase of RH due to evapotranspiration is crucial to the reduction of PM2.5 using plants.

Concerns about possible environmental implications of nano- and micro-plastics are continuously raising. Hence, comprehensive understanding of their behaviour, bioaccumulation and toxicity potential is required. Nevertheless, systematic studies on their fate and possible effects in freshwaters, as well as the influence of particle-specific and environmental factors on their behaviour and impacts are still missing. The aims of the present study are thus two-fold: (i) to examine the role of the surface charge on nanoplastic stability and acute effects to freshwater zooplankton; (ii) to decipher the influence of the refractory natural organic matter (NOM) on the nanoplastic fate and effects. Amidine and carboxyl-stabilized polystyrene (PS) spheres of 200 nm diameter characterized by opposite primary surface charges and neutral buoyancy were selected as model nanoplastics. The results demonstrated that the surface functionalization of the polystyrene nanoplastics controls their aggregation behaviour. Alginate or Suwannee River humic acid (SRHA) modified significantly the surface charge of positively-charged amidine PS nanoplastic and the aggregation state, while had no significant influence on the negatively-charged carboxyl PS nanoplastic. Both amidine and carboxyl PS nanoplastics were ingested by the zooplankton and concentrated mainly in the gut of water flea Daphnia magna and larvae Thamnocephalus platyurus, and the stomach of rotifer Brachionus calyciflorus. Amidine PS nanoplastic was more toxic than carboxyl one. The toxicity decreased in the order D. magna (48 h -immobilization) > B. calyciflorus (24 h - lethality) > T. platyurus (24 h - lethality). Alginate or SRHA reduced significantly the toxicity of both amidine and carboxyl PS nanoplastics to the studied zooplankton representatives. The implications of this laboratory study findings to natural environment were discussed.

Tris(1,3-dichloro-2-propyl) phosphate (TDCPP) has been frequently detected in environmental media and biological samples. However, knowledge of its adverse health consequences is limited. In the current study, Caenorhabditis elegans (C. elegans, L1 larvae) were exposed to TDCPP at environmentally relevant concentrations (control, 0.1, 1, 100 and 1000 μg L⁻¹) for 72 h to explore any association between TDCPP and the aging process. Some of the degenerative age-related indicators were observed, including locomotion behaviors and lifespan. As crucial biomarkers of aging, the accumulation of lipofuscin, and lipid peroxidation (LPO) products exemplified by 4-hydroxynon-2-enal (4-HNE) were detected. This product forms as a result of oxidative stress, as confirmed by an N-acetyl-L-cysteine (NAC) pharmacological assay. Moreover, a significant increase in reactive oxide species (ROS) production in a dose-dependent manner using a fluorescent probe was observed. For the underlying molecular mechanism of the above aging phenotypes, significantly upregulated transcription of genes related to antioxidant systems, especially a subset of glutathione S-transferase (gst-5, gst-6, gst-9, gst-10, gst-19, gst-24, gst-26, gst-29, gst-33, and gst-38), was found by RNA-Seq and further confirmed by RT-qPCR. The elevated glutathione S-transferase (GST) was attributed to the significant increase in 4-HNE because mutations in gst-5 and gst-24 inhibited the conjugation of GSTs with 4-HNE. Therefore, GST play an indispensable role in the detoxification process of TDCPP exposure and further confirmed LPO accumulation at the molecular mechanism level. In conclusion, TDCPP accelerated the aging process induced by the LPO products, 4-HNE, response to reactive oxidative species in C. elegans.