Spatiotemporal resolved models were developed predicting daily fine particulate matter (PM₂.₅) concentrations across Switzerland from 2003 to 2013. Relatively sparse PM₂.₅ monitoring data was supplemented by imputing PM₂.₅ concentrations at PM₁₀ sites, using PM₂.₅/PM₁₀ ratios at co-located sites. Daily PM₂.₅ concentrations were first estimated at a 1 × 1km resolution across Switzerland, using Multiangle Implementation of Atmospheric Correction (MAIAC) spectral aerosol optical depth (AOD) data in combination with spatiotemporal predictor data in a four stage approach. Mixed effect models (1) were used to predict PM₂.₅ in cells with AOD but without PM₂.₅ measurements (2). A generalized additive mixed model with spatial smoothing was applied to generate grid cell predictions for those grid cells where AOD was missing (3). Finally, local PM₂.₅ predictions were estimated at each monitoring site by regressing the residuals from the 1 × 1km estimate against local spatial and temporal variables using machine learning techniques (4) and adding them to the stage 3 global estimates. The global (1 km) and local (100 m) models explained on average 73% of the total,71% of the spatial and 75% of the temporal variation (all cross validated) globally and on average 89% (total) 95% (spatial) and 88% (temporal) of the variation locally in measured PM₂.₅ concentrations.

Ambient concentrations of fine particulate matter (PM₂.₅) concentration is often used as an exposure surrogate to estimate PM₂.₅ health effects in epidemiological studies. Ignoring the potential variations in the amount of outdoor PM₂.₅ infiltrating into indoor environments will cause exposure misclassification, especially when people spend most of their time indoors. As it is not feasible to measure the PM₂.₅ infiltration factor (Fᵢₙf) for each individual residence, we aimed to build models for residential PM₂.₅Fᵢₙf prediction and to evaluate seasonal Fᵢₙf variations among residences. We repeated collected paired indoor and outdoor PM₂.₅ filter samples for 7 continuous days in each of the three seasons (hot, cold and transitional seasons) from 48 typical homes of Shanghai, China. PM₂.₅-bound sulfur on the filters was measured by X-ray fluorescence for PM₂.₅Fᵢₙf calculation. We then used stepwise-multiple linear regression to construct season-specific models with climatic variables and questionnaire-based predictors. All models were evaluated by the coefficient of determination (R²) and root mean square error (RMSE) from a leave-one-out-cross-validation (LOOCV). The 7-day mean (±SD) of PM₂.₅Fᵢₙf across all observations was 0.83 (±0.18). Fᵢₙf was found higher and more varied in transitional season (12–25 °C) than hot (>25 °C) and cold (<12 °C) seasons. Air conditioning use and meteorological factors were the most important predictors during hot and cold seasons; Floor of residence and building age were the best transitional season predictors. The models predicted 60.0%–68.4% of the variance in 7-day averages of Fᵢₙf, The LOOCV analysis showed an R² of 0.52 and an RMSE of 0.11. Our finding of large variation in residential PM₂.₅Fᵢₙf between seasons and across residences within season indicated the important source of outdoor-generated PM₂.₅ exposure heterogeneity in epidemiologic studies. Our models based on readily available data may potentially improve the accuracy of estimates of the health effects of PM₂.₅ exposure.

PM₁ might be more hazardous than PM₂.₅ (particulate matter with an aerodynamic diameter ≤ 1 μm and ≤2.5 μm, respectively). However, studies on PM₁ concentrations and its health effects are limited due to a lack of PM₁ monitoring data.To estimate spatial and temporal variations of PM₁ concentrations in China during 2005–2014 using satellite remote sensing, meteorology, and land use information.Two types of Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 6 aerosol optical depth (AOD) data, Dark Target (DT) and Deep Blue (DB), were combined. Generalised additive model (GAM) was developed to link ground-monitored PM₁ data with AOD data and other spatial and temporal predictors (e.g., urban cover, forest cover and calendar month). A 10-fold cross-validation was performed to assess the predictive ability.The results of 10-fold cross-validation showed R² and Root Mean Squared Error (RMSE) for monthly prediction were 71% and 13.0 μg/m³, respectively. For seasonal prediction, the R² and RMSE were 77% and 11.4 μg/m³, respectively. The predicted annual mean concentration of PM₁ across China was 26.9 μg/m³. The PM₁ level was highest in winter while lowest in summer. Generally, the PM₁ levels in entire China did not substantially change during the past decade. Regarding local heavy polluted regions, PM₁ levels increased substantially in the South-Western Hebei and Beijing-Tianjin region.GAM with satellite-retrieved AOD, meteorology, and land use information has high predictive ability to estimate ground-level PM₁. Ambient PM₁ reached high levels in China during the past decade. The estimated results can be applied to evaluate the health effects of PM₁.

A full-factorial test design was applied to systematically investigate the contribution and significance of water chemistry parameters (pH, divalent cations and dissolved organic carbon (DOC) concentration) and their interactions on the behavior and fate of copper nanoparticles (CuNPs). The total amount of Cu remaining in the water column after 48 h of incubation was mostly influenced by divalent cation content, DOC concentration and the interaction of divalent cations and DOC. DOC concentration was the predominant factor influencing the dissolution of CuNPs, which was far more important than the effect of pH in the range from 6 to 9 on the dissolution of the CuNPs. The addition of DOC at concentrations ranging from 5 to 50 mg C/L resulted in a 3–5 fold reduction of dissolution of CuNPs after 48 h of incubation, as compared to the case without addition of DOC. Divalent cation content was found to be the most influential factor regarding aggregation behavior of the particles, followed by DOC concentration and the interaction of divalent cations and DOC. In addition, the aggregation behavior of CuNPs rather than particulate dissolution explained most of the variance in the sedimentation profiles of CuNPs. These results are meaningful for improved understanding and prediction of the behavior and fate of metallic NPs in aqueous environments.

The main purpose of this study is to develop accurate equation for predicting methane emission into the environment during natural gas (NG) purging process. The process is carried out regularly in NG pressure reducing stations. For this purpose, a numerical investigation has been carried out to simulate NG exit flow from a purging valve during opening time. The simulation has been carried out using Ansys-Fluent code. To make the solution and results more similar to actual scenario, the valve is continuously opened in a transient turbulent flow. Initial condition, is assumed steady flow in the pipeline. Three-dimensional modeling is used to simulate the valve and connected pipe, and all of the effective parameters including, inlet pressure, pipeline diameter, valve diameter and purging process time (including the time which valve needs to get completely opened and also purging time) are investigated. For simplicity and also as the main component of NG is methane, methane is considered as working fluid (a real compressible gas). The numerical results show that discharging gas velocity is reached to a supersonic velocity at outlet section of valve. As the highest expected exit velocity is sonic velocity, the supersonic velocity is a surprised result. Looking at the streamlines show that this is due to a convergent-divergent nozzle occurrence (due to re-circulation zone near junction) in discharging pipe. Also results show that discharged mass flow rate has liner relation with pipeline pressure, second degree relation with valve diameter and has fourth-degree relation with valve to pipeline diameter ratio. To make the results more applicable for NG industry, two correlations have been developed for calculating the amount of released gas in steady state and unsteady state condition. Unsteady state correlation is valid for valve opening time and steady state correlation could be used while the valve is completely opened.

Understorey communities can dominate forest plant diversity and strongly affect forest ecosystem structure and function. Understoreys often respond sensitively but inconsistently to drivers of ecological change, including nitrogen (N) deposition. Nitrogen deposition effects, reflected in the concept of critical loads, vary greatly not only among species and guilds, but also among forest types. Here, we characterize such context dependency as driven by differences in the amounts and forms of deposited N, cumulative deposition, the filtering of N by overstoreys, and available plant species pools. Nitrogen effects on understorey trajectories can also vary due to differences in surrounding landscape conditions; ambient browsing pressure; soils and geology; other environmental factors controlling plant growth; and, historical and current disturbance/management regimes. The number of these factors and their potentially complex interactions complicate our efforts to make simple predictions about how N deposition affects forest understoreys. We review the literature to examine evidence for context dependency in N deposition effects on forest understoreys. We also use data from 1814 European temperate forest plots to test the ability of multi-level models to characterize context-dependent understorey responses across sites that differ in levels of N deposition, community composition, local conditions and management history. This analysis demonstrated that historical management, and plot location on light and pH-fertility gradients, significantly affect how understorey communities respond to N deposition. We conclude that species' and communities' responses to N deposition, and thus the determination of critical loads, vary greatly depending on environmental contexts. This complicates our efforts to predict how N deposition will affect forest understoreys and thus how best to conserve and restore understorey biodiversity. To reduce uncertainty and incorporate context dependency in critical load setting, we should assemble data on underlying environmental conditions, conduct globally distributed field experiments, and analyse a wider range of habitat types.

Biodegradation of 1,4-dioxane was examined in packed quartz and soil column flow-through systems. The inhibitory effects of co-contaminants, specifically trichloroethene (TCE), 1,1-dichloroethene (1,1-DCE), and copper (Cu²⁺) ions, were investigated in the columns either with or without bioaugmentation with a 1,4-dioxane degrading bacterium Pseudonocardia dioxanivorans CB1190. Results indicate that CB1190 cells readily grew and colonized in the columns, leading to significant degradation of 1,4-dioxane under oxic conditions. Degradation of 1,4-dioxane was also observed in the native soil (without bioaugmentation), which had been previously subjected to enhanced reductive dechlorination treatment for co-contaminants TCE and 1,1-DCE. Bioaugmentation of the soil with CB1190 resulted in nearly complete degradation at influent concentrations of 3–10 mg L⁻¹ 1,4-dioxane and a residence reaction time of 40–80 h, but the presence of co-contaminants, 1,1-DCE and Cu²⁺ ions (up to 10 mg L⁻¹), partially inhibited 1,4-dioxane degradation in the untreated and bioaugmented soil columns. However, the inhibitory effects were much less severe in the column flow-through systems than those previously observed in planktonic cultures, which showed near complete inhibition at the same co-contaminant concentrations. These observations demonstrate a low susceptibility of soil microbes to the toxicity of 1,1-DCE and Cu²⁺ in packed soil flow-through systems, and thus have important implications for predicting biodegradation potential and developing sustainable, cost-effective technologies for in situ remediation of 1,4-dioxane contaminated soils and groundwater.

Urban water pollution poses risks of waterborne infectious diseases. Therefore, in order to improve urban liveability, effective pollution mitigation strategies are required underpinned by predictions generated using water quality models. However, the lack of reliability in current modelling practices detrimentally impacts planning and management decision making. This research study adopted a novel approach in the form of Bayesian Networks to model urban water quality to better investigate the factors that influence risks to human health. The application of Bayesian Networks was found to enhance the integration of quantitative and qualitative spatially distributed data for analysing the influence of environmental and anthropogenic factors using three surrogate indicators of human health risk, namely, turbidity, total nitrogen and fats/oils. Expert knowledge was found to be of critical importance in assessing the interdependent relationships between health risk indicators and influential factors. The spatial variability maps of health risk indicators developed enabled the initial identification of high risk areas in which flooding was found to be the most significant influential factor in relation to human health risk. Surprisingly, population density was found to be less significant in influencing health risk indicators. These high risk areas in turn can be subjected to more in-depth investigations instead of the entire region, saving time and resources. It was evident that decision making in relation to the design of pollution mitigation strategies needs to account for the impact of landscape characteristics on water quality, which can be related to risk to human health.

Poor indoor air quality can have adverse effects on human health, especially in susceptible populations; however, few studies have measured multiple pollutants in facilities for susceptible populations at a national scale in South Korea. Therefore, we measured the concentrations of indoor pollutants (fine particulate matter (PM₁₀), CO₂, airborne bacteria (AB), total volatile organic compounds (TVOCs), and formaldehyde) to determine their possible relation to other indoor environmental factors and characteristics of facilities with susceptible populations, such as hospitals, geriatric hospitals, elderly care facilities, and postnatal care centers throughout South Korea. Indoor pollutants were sampled at 82 indoor facilities, including 62 facilities for susceptible populations. Spearman's correlation, Kruskal–Wallis, and Mann–Whitney analyses were used to examine the relationship among and differences between pollutants at indoor facilities and indoor/outdoor differences in PM₁₀ concentration. There were significant correlations between indoor temperature and AB concentration (r = 0.37, p < 0.01), TVOCs, and formaldehyde (r = 0.264, p < 0.01). Indoor PM₁₀ concentrations were higher than outdoor concentrations at all facilities for susceptible populations (p < 0.01). CO₂ might be a good indicator for predicting indoor pollutants when categorized into two levels (≤750 ppm and >750 ppm). The hazard quotient of formaldehyde was higher than the acceptable level of 1 for children under the age of eight in postnatal care centers, indicative of unsafe levels. Therefore, more depth study for exposure characteristics of formaldehyde and indoor air quality (IAQ) in postnatal care facilities as a national scale is needed for finding the children exposure levels.

Fine particulate matter (PM₂.₅) has been recognized as a key air pollutant that can influence population health risk, especially during extreme cases such as wildfires. Previous studies have applied geospatial techniques such as land use regression to map the ground-level PM₂.₅, while some recent studies have found that Aerosol Optical Depth (AOD) derived from satellite images and machine learning techniques may be two elements that can improve spatiotemporal prediction. However, there has been a lack of studies evaluating use of different machine learning techniques with AOD datasets for mapping PM₂.₅, especially in areas with high spatiotemporal variability of PM₂.₅.In this study, we compared the performance of eight predictive algorithms with the use of multiple remote sensing datasets, including satellite-derived AOD data, for the prediction of ground-level PM2.5 concentration. Based on the results, Cubist, random forest and eXtreme Gradient Boosting were the algorithms with better performance, while Cubist was the best (CV-RMSE = 2.64 μg/m3, CV-R² = 0.48). Variable importance analysis indicated that the predictors with the highest contributions in modelling were monthly AOD and elevation.In conclusion, appropriate selection of machine learning algorithms can improve ground-level PM2.5 estimation, especially for areas with nonlinear relationships between PM2.5 and predictors caused by complex terrain. Satellite-derived data such as AOD and land surface temperature (LST) can also be substitutes for traditional datasets retrieved from weather stations, especially for areas with sparse and uneven distribution of stations.