Particulate matter (PM) emission is one of the leading environmental pollution issues associated with the coal mining industry. Before any control techniques can be employed, however, an accurate prediction of PM concentration is desired. Towards this end, this work aimed to provide an accurate estimation of PM concentration using a hybrid machine-learning technique. The proposed predictive model was based on the hybridazation of random forest (RF) model particle swarm optimization (PSO) for estimating PM concentration. The main objective of hybridazing the PSO was to tune the hyper-parameters of the RF model. The hybrid method was applied to PM data collected from an open-cut coal mine in northern China, the Haerwusu Coal Mine. The inputs selected were wind direction, wind speed, temperature, humidity, noise level and PM concentration at 5 min before. The outputs selected were the current concentration of PM₂.₅ (particles with an aerodynamic diameter smaller than 2.5 μm), PM₁₀ (particles with an aerodynamic diameter smaller than 10 μm) and total suspended particulate (TSP). A detailed procedure for the implementation of the RF_PSO was presented and the predictive performance was analyzed. The results show that the RF_PSO could estimate PM concentration with a high degree of accuracy. The Pearson correlation coefficients among the average estimated and measured PM data were 0.91, 0.84 and 0.86 for the PM₂.₅, PM₁₀ and TSP datasets, respectively. The relative importance analysis shows that the current PM concentration was mainly influenced by PM concentration at 5 min before, followed by humidity > temperature ≈ noise level > wind speed > wind direction. This study presents an efficient and accurate way to estimate PM concentration, which is fundamental to the assessment of the atmospheric quality risks emanating from open-cut mining and the design of dust removal techniques.

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.

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.

Excess exposure to fluoride causes substantive health burden in humans and livestock globally. However, few studies have assessed the distribution and controls of variability of ambient background concentrations of fluoride in soil. Ambient background concentrations of fluoride in soil were collated for Greater Melbourne, Greater Geelong, Ballarat and Mitchell in Victoria, Australia (n = 1005). Correlation analysis and machine learning techniques were used to identify environmental and anthropogenic influences of fluoride variability in soil. Sub-soils (>0.3 m deep), in some areas overlying siltstone and sandstone, and to a lesser extent, overlying basalt, were naturally enriched with fluoride at concentrations above ecological thresholds for grazing animals. Soil fluoride enrichment was predominantly influenced by parent material (mineralogy), precipitation (illuviation), leaching during palaeoclimates and marine inputs. Industrial air pollution did not significantly influence ambient background concentrations of fluoride at a regional scale. However, agricultural practices (potentially the use of phosphate fertilisers) were indicated to have resulted in added fluoride to surface soils overlying sediments. Geospatial variables alone were not sufficient to accurately model ambient background soil fluoride concentrations. A multiple regression model based on soil chemistry and parent material was shown to accurately predict ambient background fluoride concentrations in soils and support assessment of fluoride enrichment in the environment.

Cryptosporidium and Giardia are important parasites due to their zoonotic potential and impact on human health, often causing waterborne outbreaks of disease. Detection of (oo)cysts in water matrices is challenging and few countries have legislated water monitoring for their presence. The aim of this study was to investigate the presence and origin of these parasites in different water sources in Northern Greece and identify interactions between biotic/abiotic factors in order to develop risk-assessment models. During a 2-year period, using a longitudinal, repeated sampling approach, 12 locations in 4 rivers, irrigation canals, and a water production company, were monitored for Cryptosporidium and Giardia, using standard methods. Furthermore, 254 faecal samples from animals were collected from 15 cattle and 12 sheep farms located near the water sampling points and screened for both parasites, in order to estimate their potential contribution to water contamination. River water samples were frequently contaminated with Cryptosporidium (47.1%) and Giardia (66.2%), with higher contamination rates during winter and spring. During a 5-month period, (oo)cysts were detected in drinking-water (<1/litre). Animals on all farms were infected by both parasites, with 16.7% of calves and 17.2% of lambs excreting Cryptosporidium oocysts and 41.3% of calves and 43.1% of lambs excreting Giardia cysts. The most prevalent species identified in both water and animal samples were C. parvum and G. duodenalis assemblage AII. The presence of G. duodenalis assemblage AII in drinking water and C. parvum IIaA15G2R1 in surface water highlights the potential risk of waterborne infection. No correlation was found between (oo)cyst counts and faecal-indicator bacteria. Machine-learning models that can predict contamination intensity with Cryptosporidium (75% accuracy) and Giardia (69% accuracy), combining biological, physicochemical and meteorological factors, were developed. Although these prediction accuracies may be insufficient for public health purposes, they could be useful for augmenting and informing risk-based sampling plans.

Little information exists on interaction effects between air pollution and influenza vaccination on allergic respiratory diseases. We conducted a large population-based study to evaluate the interaction effects between influenza vaccination and long-term exposure to ambient air pollution on allergic respiratory diseases in children and adolescents.A cross-sectional study was investigated during 2012–2013 in 94 schools from Seven Northeastern Cities (SNEC) in China. Questionnaires surveys were obtained from 56 137 children and adolescents aged 2–17 years. Influenza vaccination was defined as receipt of the influenza vaccine. We estimated air pollutants exposure [nitrogen dioxide (NO2) and particulate matter with aerodynamic diameters ≤1 μm (PM1), ≤2.5 μm (PM2.5) and ≤10 μm (PM10)] using machine learning methods. We employed two-level generalized linear mix effects model to examine interactive effects between influenza vaccination and air pollution exposure on allergic respiratory diseases (asthma, asthma-related symptoms and allergic rhinitis), after controlling for important covariates.We found statistically significant interactions between influenza vaccination and air pollutants on allergic respiratory diseases and related symptoms (doctor-diagnosed asthma, current wheeze, wheeze, persistent phlegm and allergic rhinitis). The adjusted ORs for doctor-diagnosed asthma, current wheeze and allergic rhinitis among the unvaccinated group per interquartile range (IQR) increase in PM1 and PM2.5 were significantly higher than the corresponding ORs among the vaccinated group [For PM1, doctor-diagnosed asthma: OR: 1.89 (95%CI: 1.57–2.27) vs 1.65 (95%CI: 1.36–2.00); current wheeze: OR: 1.50 (95%CI: 1.22–1.85) vs 1.10 (95%CI: 0.89–1.37); allergic rhinitis: OR: 1.38 (95%CI: 1.15–1.66) vs 1.21 (95%CI: 1.00–1.46). For PM2.5, doctor-diagnosed asthma: OR: 1.81 (95%CI: 1.52–2.14) vs 1.57 (95%CI: 1.32–1.88); current wheeze: OR: 1.46 (95%CI: 1.21–1.76) vs 1.11 (95%CI: 0.91–1.35); allergic rhinitis: OR: 1.35 (95%CI: 1.14–1.60) vs 1.19 (95%CI: 1.00–1.42)]. The similar patterns were observed for wheeze and persistent phlegm. The corresponding p values for interactions were less than 0.05, respectively. We assessed the risks of PM1-related and PM2.5-related current wheeze were decreased by 26.67% (95%CI: 1.04%–45.66%) and 23.97% (95%CI: 0.21%–42.08%) respectively, which was attributable to influenza vaccination (both p for efficiency <0.05).Influenza vaccination may play an important role in mitigating the detrimental effects of long-term exposure to ambient air pollution on childhood allergic respiratory diseases. Policy targeted at increasing influenza vaccination may yield co-benefits in terms of reduced allergic respiratory diseases.

With the increase in population and industrialization, air pollution has become one of the global problems nowadays. Therefore, air pollutant parameters should be measured at regular intervals, and the necessary measures should be taken by evaluating the results of measurements. In order to prevent air pollution, pollutant parameters must be evaluated within the framework of a model. Recently, in order to obtain objective and more sensitive results with regard to air pollution nowadays, studies, which use machine learning algorithms in artificial intelligence technologies, have been carried out. In this study, PM₁₀ concentrations, which are obtained from 7 stations in Ankara province in Turkey, were trained with machine learning algorithms (LASSO, SVR, RF, kNN, xGBoost, ANN). The PM₁₀ concentrations of the years 2009–2017 of 6 stations in Ankara were given as input, and the PM₁₀ concentrations of the seventh station for the year 2018 were predicted. The model development stage was repeated for each station, and the performance and error rates of the algorithms were determined by comparing the results produced by the algorithms with the actual results. The best results were provided with ANN (R² = 0.58, RMSE = 20.8, MAE = 14.4). The spatial distribution of the estimated concentration results was provided through Geographic Information System (GIS), and spatial strategies for improving air pollution over land use were established.

Tropospheric ozone in the surface air has become the primary atmospheric pollutant in Hangzhou, China, in recent years. Previous analysis is not enough to decode it for better regulation. Therefore, we use the traditional atmospheric model, Weather Research and Forecasting coupled with Community Multi-scale Air Quality (WRF-CMAQ), and machine learning models, Extreme Learning Machine (ELM), Multi-layer Perceptron (MLP), Random Forest (RF) and Recurrent Neural Network (RNN) to analyze and predict the ozone in the surface air in Hangzhou, China, using meteorology and air pollutants as input. We firstly quantitatively demonstrate that the dew-point deficit, instead of temperature and relative humidity, is the predominant meteorological factor in shaping tropospheric ozone. Urban heat island, daily direct solar radiation time, wind speed and wind direction play trivial role in impacting tropospheric ozone. NO₂ is the primary influential factors both for hourly ozone and daily O₃-8 h due to the titration effect. The most environmental-friendly way to mitigate the ozone pollution is to lower the volatile organic compounds (VOCs) with the highest ozone formation potentials. We deduce that the tropospheric ozone formation process tends to be not only non-linear but also non-smooth. Compared with the traditional atmospheric models, machine learning, whose characteristics are rapid convergence, short calculating time, adaptation of forecasting episodes, small program memory, higher accuracy and less cost, is able to predict tropospheric ozone more accurately.