Evaluation of different machine learning approaches and aerosol optical depth in PM2.5 prediction.

Atmospheric Aerosol Optical Depth (AOD), derived from polar-orbiting satellites, has shown potential in PM2.5 predictions. However, this important source of data suffers from low temporal resolution. Recently, geostationary satellites provide AOD data in high temporal and spatial resolution. However, the feasibility of these data in PM2.5 prediction needs further study. In this paper, we analyzed the impact of AOD derived from Himawari-8 in PM2.5 predictions. Moreover, by combining wavelet, machine learning techniques, and minimum redundancy maximum relevance (mRMR), a novel hybrid model was proposed. The results showed that AOD missing rate over Yangtze River Delta region is the highest in Nanjing, Hefei, and Maanshan. In addition, missing rates are the lowest in winter and summer (∼80%). Moreover, we found that considering AOD, as an auxiliary variable in the model, could not improve the accuracy of PM2.5 predictions, and in some cases decreased it slightly. In comparison with other models, our proposed hybrid model showed higher prediction accuracy, R2 is improved by 11.64% on average, and root mean square error, mean absolute error, and mean absolute percentage error is reduced by 26.82%, 27.24%, and 29.88% respectively. This research provides a general overview of the availability of Himawari-8 AOD data and its feasibility in PM2.5 predictions. In addition, it evaluates different machine learning approaches in PM2.5 predictions. Our proposed framework can be used in other regions to predict different air pollutants concentrations and can be used as an aid for air pollution controlling programs.

Concentration-dependent effects of lithium on Daphnia magna: Life-history profiles and integrated biomarker response implementation.

The growing use of lithium (Li) in industrial and energy applications and increasing demand worldwide has inevitably resulted in its wide dispersal, representing a significant threat to aquatic systems. Unfortunately, as a ubiquitous emerging contaminant, the comprehensive toxicological information regarding Li at multifarious levels is limited. To diminish this gap, this work was focused to explore Li-induced cascading effects on Daphnia magna as a key species in freshwater ecosystems. Specifically, the organisms were chronically exposed to gradient Li concentrations with emphasis on characterizing life-history traits from individual to population scale, primarily as observed by a markedly concentration-dependent decrease along exposure gradients. In parallel, a robust set of biomarkers relating to energy reserves, antioxidant and biotransformation enzymes, cellular damage, ionoregulation and neurotoxicity were assayed for further understanding potential underlying mechanisms. As a result, biomarker alterations were characterized by significant decreases in energy storage and enzymatic profiles of antioxidant and biotransformation systems, not only triggering an imbalance between reactive oxygen species (ROS) generation and elimination under Li exposure, but compromising the fecundity fitness of phenotypical costs. In contrast, malondialdehyde (MDA) levels were remarkably enhanced as a consequence of inefficient antioxidant and biotransformation capacity leading to lipid peroxidation (LPO). Additionally, Li exerted a dose-dependent biphasic effect on the activities of superoxide dismutase (SOD), Na+,K+-ATPase and acetylcholinesterase (AChE) by interfering with inherent balance. In terms of responsive patterns and dose-effect trends, the integrated biomarker response indices (IBRv2) and star plots were consistent with the differences in biomarker profiles, not only presenting comprehensively biological effects in a visualized form, but signaling the importance of progressive induced changes in an integrative way. Overall, these findings highlighted the need for elucidating Li-produced impacts from a comprehensive perspective, providing valuable insights into better understanding the toxicity of Li in relation to aquatic ecosystem functioning and ecological relevance.

Mapping the social stock and spatiotemporal distribution of high-tech minerals from wasted mobile phones in China: 2001-2019.

In the twenty-first century, mobile phones have become one of the most indispensable electronic products in the international community. The pollution of wasted mobile phones has become an urgent problem worldwide and needs special attention. In this paper, we applied the consumption and usage method to calculate the high-tech mineral elements in China from 2001 to 2019. To analyze the spatial distribution of per capita high-tech minerals in China, we proposed a model (3D GHM) through which a 3D grid of high-tech minerals in wasted mobile phones can be obtained in 1 km resolution. The results showed that the total amount of wasted mobile phones in China from 2001 to 2019 was 8.6 billion, with a growth rate of 1026.7% in 2019 compared with 2001. Moreover, the spatiotemporal distribution of wasted mobile phones is characterized by more in the east and less in the west. The total amount of cobalt, palladium, antimony, beryllium, neodymium, praseodymium, and platinum in wasted mobile phones from 2001 to 2019 reached 42,422.4 tons. Based on our results, we proposed a system for efficient collecting and recycling of wasted mobile phones in China.

A novel framework to predict chlorophyll-a concentrations in water bodies through multi-source big data and machine learning algorithms.

Eutrophication happens when water bodies are enriched by minerals and nutrients. Dense blooms of noxious are the most obvious effect of eutrophication that harms water quality, and by increasing toxic substances damage the water ecosystem. Therefore, it is critical to monitor and investigate the development process of eutrophication. The concentration of chlorophyll-a (chl-a) in water bodies is an essential indicator of eutrophication in them. Previous studies in predicting chlorophyll-a concentrations suffered from low spatial resolution and discrepancies between predicted and observed values. In this paper, we used various remote sensing and ground observation data and proposed a novel machine learning-based framework, a random forest inversion model, to provide the spatial distribution of chl-a in 2 m spatial resolution. The results showed our model outperformed other base models, and the goodness of fit improved by over 36.6% while MSE and MAE decreased by over 15.17% and over 21.26% respectively. Moreover, we compared the feasibility of GF-1 and Sentinel-2 remote sensing data in chl-a concentration prediction. We found that better prediction results can be obtained by using GF-1 data, with the goodness of fit reaching 93.1% and MSE only 3.589. The proposed method and findings of this study can be used in future water management studies and as an aid for decision-makers in this field.

Geochemical behavior and fractionation characteristics of rare earth elements (REEs) in riverine water profiles and sentinel Clam (Corbicula fluminea) across watershed scales: Insights for REEs monitoring.

The increasing global demand for rare earth elements (REEs) has led to their recognition as emerging contaminants; however, the effect that biota have on the cycling of these elements at the watershed scale is not currently well understood. In this study, water samples and field freshwater clams Corbicula fluminea were concurrently collected along watershed gradients, and concentration profiles of 14 naturally occurring REEs were measured in operationally defined water fractions and soft tissues of the freshwater clams. Moreover, Post Archean Australian Shale (PAAS) normalized REE patterns, fractionation indices, and anomalous values were determined to further extract characteristic features. As a result, both the water and biological samples had variable REE compositions, with higher concentrations of light REEs (LREEs) than middle REEs (MREEs) and heavy REEs (HREEs), while decreasing concentrations were generally observed as filter pore size decreased, implying that large colloidal and particulate fractions were important carriers of REEs. The spatial distribution patterns of REEs revealed a clear site effect among profiles, with variability more pronounced among watersheds and with peaks in sites from a small watershed near the hotspots of the mining area, and then exhibited a decreasing trend with distance from there. Meanwhile, significant bioaccumulation of REEs was observed potentially reflecting different degrees of contamination gradients among the watersheds. The PAAS-normalized distribution patterns tended to be slightly enriched in MREEs, producing a peculiar "roof-shaped" feature and characteristic fractionation. Remarkably, bio-concentration factors (BCFs) highlighted the importance of large colloidal and particulate phases in assessing biologically available REEs for filter-feeding species. Collectively, our study strongly favored that accumulation patterns and fractionation characteristics of REEs in C. fluminea can serve as a reliable indicator of geochemical behavior, providing a promising biomonitoring tool to quantitatively denote different degrees of REE contamination and assess possible impacts in mining watersheds.

Landscape ecological risk assessment and driving factor analysis in Dongjiang river watershed.

The ecological and environmental quality of Dongjiang river watershed has great influence on Guangdong, Hong Kong and Macao. The landscape ecological risk assessment model could effectively monitor and assess environmental quality. In this study, spatial autocorrelation and geographic detector methods were used to explore the spatial characteristics of landscape ecological risk and their driving factors in the Dongjiang river watershed for four decades. The results showed that the ecological risks of Dongjiang River Source Watershed are mainly classified as low and intermediate, which are distributed in the hilly regions and the marginal mountainous regions at the junction of the Xunwu and Dingnan counties. From 1980 to 2018, the area of regions with the low ecological risk increased by 587.01 km 2. The size of regions with moderate, high and severe ecological risk decreased by 165.6 km 2, 258.82 km2 and 162.58 km2, respectively. Moreover, landscape ecological risk values exhibited an apparent spatial dependency, and high-risk areas cluster together. Among influencing factors, population density has the most significant impact on the change of landscape ecological risk in the Dongjiang river watershed, followed by elevation (DEM), human interface, vegetation index (NDVI), and urbanization level. However, the interaction of driving factors has a greater impact on the ecological risk of the Dongjiang river watershed than a single driving factor. The research provides good knowledge for environmental quality management, and the proposed methods can be used for other regions.