Search

Economic development, population growth, industrialization, and urbanization dramatically increase urban water quality deterioration, and thereby endanger human life and health. However, there are not many efficient methods and techniques to monitor urban black and odorous water (BOW) pollution. Our research aims at identifying primary indicators of urban BOW through their spectral characteristics and differentiation. This research combined ground in-situ water quality data with ground hyperspectral data collected from main urban BOWs in Guangzhou, China, and integrated factorial data mining and machine learning techniques to investigate how to monitor urban BOW. Eight key water quality parameters at 52 sample sites were used to retrieve three latent dimensions of urban BOW quality by factorial data mining. The synchronically measured hyperspectral bands along with the band combinations were examined by the machine learning technique, Lasso regression, to identify the most correlated bands and band combinations, over which three multiple regression models were fitted against three latent water quality indicators to determine which spectral bands were highly sensitive to three dimensions of urban BOW pollution. The findings revealed that the many sensitive bands were concentrated in higher hyperspectral band ranges, which supported the unique contribution of hyperspectral data for monitoring water quality. In addition, this integrated data mining and machine learning approach overcame the limitations of conventional band selection, which focus on a limited number of band ratios, band differences, and reflectance bands in the lower range of infrared region. The outcome also indicated that the integration of dimensionality reduction with feature selection shows good potential for monitoring urban BOW. This new analysis framework can be used in urban BOW monitoring and provides scientific data for policymakers to monitor it. Image 1 • Hyperspectral bands provide unique info concerning severity of black-odorous water. • Factorial data mining synthesizes water quality parameters into reduced dimensions. • Reduced dimensions establish correlation between water quality and spectral data. • Machine learning based feature extraction selects critical spectral bands. • Reduced dimensions and critical spectral bands reveal condition of water quality. [ABSTRACT FROM AUTHOR]

Long-term air pollution data are essential for formulating air quality management policies and assessing their corresponding impacts on public health. However, missing data are inevitably encountered during air pollution observations at different sites. This study proposed a machine learning approach that utilizes data from neighboring sites to reconstruct missing data. Hourly observation data from three neighboring sites in the Pearl River Delta (PRD) region in South China, were used for data retrieval, including the NC site (2006–2015), JXL site and PYZX site (2014–2022). The overlapped data (2014.05–2015.12) were used to train and evaluate the machine learning models. The performance of 11 algorithms (CatBoost, XGBoost, LightGBM, LightGBMXT, LightGBMLarge, RandomForestMSE, ExtraTreeMSE, NeuralNetTorch, NeuralNetFastAI, KNeighborsDist, and KNeighborsUnif) for the retrieval of major air pollutants, including O 3 , NO 2 , PM 2.5 , PM 10 and SO 2 was benchmarked by a set of evaluation metrics. CatBoost showed the best performance; thus, it was adopted for air pollutant data reconstruction in NC (2016–2022) and PYZX (2008–2014). Long-term data (2006–2022) at the NC were obtained by combining the observation and retrieval data. In the past 15 years, the O 3 concentration of NC has increased by 72% at a rate of 0.83 ppb yr−1 (3.2% yr−1). On the contrary, substantial reductions were observed for NO 2 (61%), PM 2.5 (51%) and PM 10 (42%) at the NC site, with the rates of −1.27 ppb yr−1 (−5.9% yr−1), −1.96 μg m−3 yr−1 (−5.8% yr−1) and −2.32 μg m−3 yr−1 (−5.2% yr−1), respectively. SO 2 exhibits the most pronounced reduction (79%) among all species, with two distinct rates of −4.10 ppb yr−1 (−27.4% yr−1) and −0.40 ppb yr−1 (−6.2% yr−1), for 2008–2012 and 2012–2022, respectively. This study demonstrates the feasibility of machine learning in filling the data gap of air pollution monitoring network and highlights the importance of continuous long-term air pollution data in reviewing air quality management policies. [Display omitted] • A machine learning approach is proposed to reconstruct missing air quality data. • 11 algorithms were compared and CatBoost exhibited the best results. • Missing data reconstruction was demonstrated using data in Guangzhou. [ABSTRACT FROM AUTHOR]

Abstract: Near-surface atmospheric aerosol samples were collected at the sampling frequency of 2–3 d per week for one year from August 2006 to August 2007 at a low latitude station in Tianhe District, Guangzhou, Guangdong Province of southern China. The samples were analyzed for cosmogenic nuclide 7Be and persistent organic pollutants, i.e. organochlorine pesticides (OCPs), polychlorinated biphenyls (PCBs), and polybrominated diphenyl ethers (PBDEs). The annual average 7Be concentration was 2.59 mBq m−3, with the maximum occurred in May (8.45 mBq m−3) and minimum in late August and early September (0.07 mBq m−3). Winter and spring were the seasons in which the 7Be concentrations were high while summer and autumn were the lower 7Be seasons. Spring peaks in 7Be in the near-surface atmospheric aerosols may have associated with the “spring leak maximum” episode. The annual average ∑OCPs concentration was 345.6 pg m−3, ∑33PCBs 317.6 pg m−3, and ∑31PBDEs 609.0 pg m−3. The variation trends in the time-series of 7Be, OCPs, PCBs, and PBDEs in near-surface atmospheric aerosol showed both common features and differences. Significant correlations (R 2 = 0.957 and 0.811. respectively, p = 0.01) were observed between the monthly average 7Be concentrations and those of ∑PCBs and ∑PBDEs in summer, autumn, and early winter. The difference between the seasonal variation features of OCPs and PCBs (and PBDEs) could be attributed to the different source functions and physical–chemical properties which could control the behaviors of these compounds in air–aerosol partitions as well as atmospheric transport. [Copyright &y& Elsevier]

Improving the comprehensive utilization level of energy has been a core goal for the optimal design of the integrated energy system (IES), which requires an accurate assessment of energy efficiency. The widely used criterion of conventional energy efficiency derived from the first law of thermodynamics has gradually been criticized for its shortcomings of neglecting the quality of energy. As a concept considering both the quantity and quality of energy, exergy efficiency has been recognized as a more reasonable index in recent decades. Nevertheless, it is a complicated problem to calculate the exergy efficiency among multiple energy forms in an electricity-gas-heating-cooling coupled IES, limiting its application in IES planning. This paper hence proposes the method of calculating the exergy efficiency of an IES by using a black-box model, after defining the concept of energy quality coefficient (EQC) that values the quality of various energy forms involved. And then, a multi-objective planning model considering both exergy efficiency and economy is thus proposed for the joint planning of energy generators, storages and networks, wherein a five-tier energy hub is modeled to deal with the multi-energy couplings. To reduce the difficulty of optimizing, several convex relaxation measures are designed to transform the non-convex original problem into a convex one. The proposed framework is verified in a real-world study case located in Guangzhou, China. • A criterion for accessing comprehensive energy utilization level is proposed. • A five-tier energy hub model for multi-energy couplings is established. • A multi-objective joint planning model and its solution method is proposed. [ABSTRACT FROM AUTHOR]