Source apportionment can be an effective tool in mitigating soil pollution but its efficacy is often limited by a lack of information on the factors that influence the accumulation of pollutants at a site. In response to this limitation and focusing on a suite of heavy metals identified as priorities for pollution control, the study established a comprehensive pollution control framework using factor identification coupled with spatial agglomeration for agricultural soils in an industrialized part of Zhejiang Province, China. In addition to elucidating the key role of industrial and traffic activities on heavy metal accumulation through implementing a receptor model, specific influencing factors were identified using a random forest model. The distance from the soil sample location to the nearest likely industrial source was the most important factor in determining cadmium and copper concentrations, while distance to the nearest road was more important for lead and zinc pollution. Soil parent materials, pH, organic matter, and clay particle size were the key factors influencing accumulation of arsenic, chromium, and nickel. Spatial auto-correlation between levels of soil metal pollution and industrial agglomeration can enable a more targeted approach to pollution control measures. Overall, the approach and results provide a basis for improved accuracy in source apportionment, and thus improved soil pollution control, at the regional scale.

Nitrogen dioxide (NO₂) is an important air pollutant and highly related to air quality, short- and long-term health effects, and even climate. A national model was developed using the extreme gradient boosting algorithm with high-resolution tropospheric vertical column NO₂ densities from the Sentinel-5 Precursor/Tropospheric Monitoring Instrument and general meteorological variables as input to generate daily mean surface NO₂ concentrations across mainland China. Model-derived daily NO₂ estimates were high accuracy with sample-based cross-validation coefficient of determination of 0.83, a root-mean-square error of 7.58 μg/m³, a mean prediction error of 5.56 μg/m³, and a mean relative prediction error of 18.08%. It has good performance in NO₂ estimations at both regional and individual site scale. The model also performed well in terms of estimating monthly, seasonal, and annual mean NO₂ concentrations across China. The model performance appears to better than or comparable to most previous related studies. The seasonal and annual spatial distributions of surface NO₂ across China and several regional NO₂ hotspots in 2019 were derived from the model and analyzed. Also evaluated were the population exposure levels of NO₂ for cities in and provinces of China. At the national scale, about 12% of the population experienced annual mean NO₂ concentrations exceeding the Chinese national air quality standard. The nationwide model with conventional predictors developed here can derive high-resolution surface NO₂ concentrations across China routinely, benefitting air epidemiological and environmental related studies.

The construction of noise maps is of great significance for the development of urban sustainability and the protection of residents’ physical and mental health. The traditional noise map construction method is difficult to be widely used because of its low update frequency and high drawing cost. Based on the crowd-sensing technology and Latent Factor Model (LFM), this paper proposes a new noise map completion method called Spatial-Temporally Related LFM (STR-LFM) for solving the problem of data sparseness. First, the geographic information features including Point of Interest (POI), road network and building outline are fully excavated, and then combine the correlation of the samples in the time dimension to construct the similarity matrixes. After that, use the k-nearest neighbor algorithm to find out the similar samples of missing positions, and finally regard their weighted fusion as the predicted values. Experimental results show that the recovery error is lower than other commonly used methods, and the proposed method has better stability when faced with data sparseness problems at different levels.

Poor air quality is an emerging problem in Australia primarily due to ozone pollution events and lengthening and more severe wildfire seasons. A significant deterioration in air quality was experienced in Australia’s most populous cities, Melbourne and Sydney, as a result of fires during the so-called Black Summer which ran from November 2019 through to February 2020. Following this period, social, mobility and economic restrictions to curb the spread of the COVID-19 pandemic were implemented in Australia. We quantify the air quality impact of these contrasting periods in the south-eastern states of Victoria and New South Wales (NSW) using a meteorological normalisation approach. A Random Forest (RF) machine learning algorithm was used to compute baseline time series’ of nitrogen dioxide (NO₂), ozone (O₃), carbon monoxide CO and particulate matter with diameter < 2.5 μm (PM₂.₅), based on a 19 year, detrended training dataset. Across Victorian sites, large increases in CO (188%), PM₂.₅ (322%) and ozone (22%) were observed over the RF prediction in January 2020. In NSW, smaller pollutant increases above the RF prediction were seen (CO 58%, PM₂.₅ 80%, ozone 19%). This can be partly explained by the RF predictions being high compared to the mean of previous months, due to high temperatures and strong wind speeds, highlighting the importance of meteorological normalisation in attributing pollution changes to specific events. From the daily observation-RF prediction differences we estimated 249.8 (95% CI: 156.6–343.) excess deaths and 3490.0 (95% CI 1325.9–5653.5) additional hospitalisations were likely as a result of PM₂.₅ and O₃ exposure in Victoria and NSW. During April 2019, when COVID-19 restrictions were in place, on average NO₂ decreased by 21.5 and 8% in Victoria and NSW respectively. O₃ and PM₂.₅ remained effectively unchanged in Victoria on average but increased by 20 and 24% in NSW respectively, supporting the suggestion that community mobility reduced more in Victoria than NSW. Overall the air quality change during the COVID-19 lockdown had a negligible impact on the calculated health outcomes.

Ambient fine particulate matter (PM₂.₅) has been ranked as the sixth leading risk factor globally for death and disability. Modelling methods based on having access to a limited number of monitor stations are required for capturing PM₂.₅ spatial and temporal continuous variations with a sufficient resolution. This study utilized a land use regression (LUR) model with machine learning to assess the spatial-temporal variability of PM₂.₅. Daily average PM₂.₅ data was collected from 73 fixed air quality monitoring stations that belonged to the Taiwan EPA on the main island of Taiwan. Nearly 280,000 observations from 2006 to 2016 were used for the analysis. Several datasets were collected to determine spatial predictor variables, including the EPA environmental resources dataset, a meteorological dataset, a land-use inventory, a landmark dataset, a digital road network map, a digital terrain model, MODIS Normalized Difference Vegetation Index (NDVI) database, and a power plant distribution dataset. First, conventional LUR and Hybrid Kriging-LUR were utilized to identify the important predictor variables. Then, deep neural network, random forest, and XGBoost algorithms were used to fit the prediction model based on the variables selected by the LUR models. Data splitting, 10-fold cross validation, external data verification, and seasonal-based and county-based validation methods were used to verify the robustness of the developed models. The results demonstrated that the proposed conventional LUR and Hybrid Kriging-LUR models captured 58% and 89% of PM₂.₅ variations, respectively. When XGBoost algorithm was incorporated, the explanatory power of the models increased to 73% and 94%, respectively. The Hybrid Kriging-LUR with XGBoost algorithm outperformed the other integrated methods. This study demonstrates the value of combining Hybrid Kriging-LUR model and an XGBoost algorithm for estimating the spatial-temporal variability of PM₂.₅ exposures.

Biorefineries provide a platform for different industries to produce multiple bio-products enhancing the economic value of the system. The production of these biorefineries has led to an increase in the generation of biowaste. To minimize the risk of environmental pollution, numerous studies have focused on a variety of strategies to mitigate these concerns reflected in the vast amount of literature written on this topic. This paper aims to systematically analyze and review the enormous body of scientific literature in the biowaste and biorefinery field for establishing an understanding and providing a direction for future works. A bibliometric analysis is first performed using the CorTexT Manager platform on a corpus of 1488 articles written on the topic of biowaste. Popular and emerging topics are determined using a terms extraction algorithm. A contingency matrix is then created to study the correlation of scientific journals and key topics from this field. Then, the connection and evolution of these terms were analyzed using network mapping, to determine relationships among key terms and analyze notable trends in this research field. Finally, a critical review of articles was presented across three main categories of biowaste management such as mitigation, sustainable utilization, and cleaner disposal from the perspective of the biorefinery concept. Operational and technological challenges are identified for the integration of anaerobic digestion in biorefineries, especially in developing nations. Moreover, logistical challenges in the biorefinery supply-chain are established based on the economics and collection aspect of handling biowaste.

Quantitative estimation of soil organic carbon (SOC) is essential for the study of the C cycle and global C storage. Soil spectroscopic technology provides a cost-effective and time-efficient method for SOC quantification and has been successfully used to determine SOC storage. However, the SOC estimation accuracy remains limited by other soil properties, particularly soil water. In this study, we proposed a new deep learning algorithm named the Water Absorption Trough Dewatering Machine (WATDM) to improve estimations of SOC from soil reflectance spectra and reduce the effect of soil water. Soil water and reflectance spectral data of soil samples were measured using spectrometry. Based on the soil water contents derived from the water absorption troughs around 1900 nm, the optimal WATDM model was obtained and treated as the final model of the WATDM method, which performed better than a multiple linear regression model based on moist soil samples. The findings of this study indicate that the WATDM method can improve the estimation accuracy of SOC content by reducing the effect of soil water and can be used as a valuable new methodology within the spectroscopic estimation of soil properties.

During March 2020, most European countries implemented lockdowns to restrict the transmission of SARS-CoV-2, the virus which causes COVID-19 through their populations. These restrictions had positive impacts for air quality due to a dramatic reduction of economic activity and atmospheric emissions. In this work, a machine learning approach was designed and implemented to analyze local air quality improvements during the COVID-19 lockdown in Graz, Austria. The machine learning approach was used as a robust alternative to simple, historical measurement comparisons for various individual pollutants. Concentrations of NO₂ (nitrogen dioxide), PM₁₀ (particulate matter), O₃ (ozone) and Oₓ (total oxidant) were selected from five measurement sites in Graz and were set as target variables for random forest regression models to predict their expected values during the city’s lockdown period. The true vs. expected difference is presented here as an indicator of true pollution during the lockdown. The machine learning models showed a high level of generalization for predicting the concentrations. Therefore, the approach was suitable for analyzing reductions in pollution concentrations. The analysis indicated that the city’s average concentration reductions for the lockdown period were: -36.9 to −41.6%, and −6.6 to −14.2% for NO₂ and PM₁₀, respectively. However, an increase of 11.6–33.8% for O₃ was estimated. The reduction in pollutant concentration, especially NO₂ can be explained by significant drops in traffic-flows during the lockdown period (−51.6 to −43.9%). The results presented give a real-world example of what pollutant concentration reductions can be achieved by reducing traffic-flows and other economic activities.

In the present work, we propose a novel algorithm to determine the scattering coefficient of OA by evaluating the relationships of the MSEs for primary organic aerosol (POA) and secondary organic aerosol (SOA) with their mass concentrations at three distinct sites, i.e. an urban site, a rural site, and a background site in China. Our results showed that the MSEs for POA and SOA increased rapidly as a function of mass concentration in low mass loading. While the increasing rate declined after a threshold of mass loading of 50 μg/m³ for POA, and 15 μg/m³ for SOA, respectively. The dry scattering coefficients of submicron particles (PM₁) were reconstructed based on the algorithm for POA and SOA scattering coefficient and further verified by using multi-site data. The calculated dry scattering coefficients using our reconstructing algorithm have good consistency with the measured ones, with the high correlation and small deviation in Shanghai (R² = 0.98; deviations: 2.9%) and Dezhou (R² = 0.90; deviations: 4.7%), indicating that our algorithms for OA and PM₁ are applicable to predict the scattering coefficient of OA and Submicron particle (PM₁) in China.

Ambient ozone (O₃) concentrations have shown an upward trend in China and its health hazards have also been recognized in recent years. High-resolution exposure data based on statistical models are needed. Our study aimed to build high-performance random forest (RF) models based on training data from 2013 to 2017 in the Beijing-Tianjin-Hebei (BTH) region in China at a 0.01 ° × 0.01 ° resolution, and estimated daily maximum 8h average O₃ (O₃-8hmax) concentration, daily average O₃ (O₃-mean) concentration, and daily maximum 1h O₃ (O₃-1hmax) concentration from 2010 to 2017. Model features included meteorological variables, chemical transport model output variables, geographic variables, and population data. The test-R² of sample-based O₃-8hmax, O₃-mean and O₃-1hmax models were all greater than 0.80, while the R² of site-based and date-based model were 0.68–0.87. From 2010 to 2017, O₃-8hmax, O₃-mean, and O₃-1hmax concentrations in the BTH region increased by 4.18 μg/m³, 0.11 μg/m³, and 4.71 μg/m³, especially in more developed regions. Due to the influence of weather conditions, which showed high contribution to the model, the long-term spatial distribution of O₃ concentrations indicated a similar pattern as altitude, where high concentration levels were distributed in regions with higher altitude.