基于机器学习的大气PM2.5中金属浓度预测模型的研究

基于2013—2018年哈尔滨市气象数据、大气污染物数据和细颗粒物（PM_2.5）中金属成分数据,采用机器学习方法探索大气PM_2.5中金属浓度预测模型,并选择最优模型进行污染物浓度预测。结果表明,多元线性回归（MLR）、人工神经网络（BP-ANN）、支持向量机（SVM）和随机森林（RF）4种模型中,RF对大气PM_2.5中5种金属［锑（Sb）、砷（As）、铅（Pb）、镉（Cd）、铊（Tl）］的浓度预测效果最佳,在训练集和测试集中表现均较稳定,其中相关系数（r）均＞0.7, 平均绝对误差（MAE）和均方根误差（RMSE）数值较小。RF在大气PM_2.5中金属浓度预测上具有较好的表现,可在缺乏监测和实验数据的情况下,实现对大气颗粒物中金属浓度的快速预测,为全面了解颗粒物中金属污染特征提供数据基础。     

Trace metal contents of the pacific oyster (Crassostrea gigas) purchased from markets in Hong Kong

Arsenic, cadmium, chromium, and copper concentrations of the Pacific oyster,Crassostrea gigas, purchased from four different markets were determined in this project. In general, gill tissue had the highest proportion of metal contents (34%–67%) when compared with other tissue parts (mantle, viscera, and adductor muscle), except for arsenic, which showed the highest level in adductor muscle (44%). Smaller oysters (longitudinal length of soft body part less than 6 cm) had higher metal levels than larger ones (longitudinal length of soft body part more than 6 cm), except copper. None of the four metals examined showed an obvious seasonal trend, although cadmium levels seemed to be higher in autumn and winter months. Arsenic, cadmium, and copper levels in oysters purchased from different markets and different months obtained in the present study were higher when compared with past reports. Cadmium levels, as high as 10.98 mg/kg (dry weight basis) have been obtained. This approaches the safety limit that may be hazardous to human health. Continual monitoring of cadmium and other trace metals of toxicological significance to man in Hong Kong seafood is recommended.     

Carbon dioxide and methane emission from subsurface peatlands and its controlling factors北大核心CSCD

Peatland is an efficient carbon dioxide (CO2) sink on the continent and plays an important role in global carbon cycle. Climate change and human activities, two of the notable global environmental issues, have accelerated the degradation of peatlands during recent years. Global warming will increase the rate of aerobic decomposition in the surface of peatlands. Carbon stored in the subsurface of peatlands will be metabolized if the climatic conditions become favorable for decomposition. This study reviewed the carbon circle of subsurface peatland in natural environment and in environments disturbed by human activity or climate change. Furthermore, the major factors (environmental and human factors) that affect the carbon cycle were also discussed. According to a previous study, subsurface peatland will rapidly participate in the carbon cycle when the peatland is degraded. Water level, vegetation, and temperature were the main natural factors affecting the carbon cycle, whereas drainage, farming, and grazing were the main anthropogenic factors. Further studies should focus on different soil layer carbon dynamics, inorganic carbon content, and conservation and restoration of peatlands. The study methods should be a combination of macro with micro scale and focus on developing deep peat research techniques. Most of the previous studies focused on greenhouse gas emission and their response factors in short-term experiments. Thus, the mechanism and process of subsurface carbon are not clear and needs further study. © 2018 Science Press. All rights reserved.     

磁性铁基改性生物炭去除水中氨氮

氨氮的过度排放是水体富营养化的一个重要原因.然而,随着环境法规的日益严格,传统方法处理效果难以达到要求.吸附法因高效、安全等优点近年来开始应用于去除水中的氨氮.本研究中以共沉淀法将磁性铁基材料负载到市政污泥生物炭上,结果表明其对水中氨氮有良好的去除效果.80℃下合成的材料（MB₈₀）在293 K下对氨氮的饱和吸附量可达17.52 mg·g^-1.动力学与热力学结果表明,MB₈₀吸附氨氮的过程更符合伪二级动力学和Langmuir等温线.MB₈₀对氨氮的吸附机制可归纳为静电吸引、孔隙填充、离子交换和氢键结合.且5次循环后对氨氮的吸附量仍十分理想,可达3.18 mg·g^-1.本研究的结果可以为高效去除水中氨氮提供一种行之有效的方法,并为市政污泥的处理提供新的出路.     

Environmental modeling of uranium interstitial compositions of non-stoichiometric oxides: experimental and theoretical analysis

Study of uranium interstitial compositions of non-stoichiometric oxides UO_2+x (x ∈ 0.1–0.02) in gas and condense phases has been presented, using various soft-ionization mass spectrometric methods such as ESI-, APCI-, and MALDI-MS at a wide dynamic temperature gradient (∈ 25–300 °C). Linearly polarized vibrational spectroscopy has been utilized in order to assign unambiguously, the vibrational frequencies of uranium non-stoichiometric oxides. Experimental design has involved xUO_2.66·yUO_2.33, xUO_2.66·yUO_2.33/SiO₂, xUO_2.66·yUO_2.33/SiO₂ (NaOH) and SiO₂/x′NaOH·y′UO₂(NO₃)₂·6H₂O, multicomponent systems (x = 1, y ∈ 0.1–1.0 and x′ = 1, y′ ∈ 0.1–0.6) as well as phase transitions UO₂(NO₃)₂·6H₂O → {U₄O₉(UO_2.25)} → U₃O₇(UO_2.33) → U₃O₈(UO_2.66) → {UO₃}, thus ensuring a maximal representativeness to real environmental conditions, where diverse chemical, geochemical and biochemical reactions, including complexation and sorption onto minerals have occurred. Experimental factors such as UV-irradiation, pH, temperature, concentration levels, solvent types and ion strength have been taken into consideration, too. As far as uranium speciation represents a challenging analytical task in terms of chemical identification diverse coordination species, mechanistic aspects relating incorporation of oxygen into UO _2+x form the shown full methods validation significantly impacts the field of environmental radioanalytical chemistry. UO₂ is the most commonly used fuel in nuclear reactors around the globe; however, a large non-stoichiometric range ∈ UO_1.65–UO_2.25 has occurred due to radiolysis of water on UO₂ surface yielding to H₂O₂, OH^·, and more. Each of those compositions has different oxygen diffusion. And in this respect enormous effort has been concentrated to study the potential impact of hazardous radionuclide on the environment, encompassing from the reprocessing to the disposal stages of the fuel waste, including the waste itself, the processes in the waste containers, the clay around the containers, and geological processes. In a broader sense, thereby, this study contributes to field of environmental analysis highlighting the great ability of various soft-ionization MS methods, particularly, MALDI-MS one, for direct assay of complex multicomponent heterogeneous mixtures at fmol–attomol concentration ranges, along with it the great instrumental features allowing, not only meaningful quantitative, but also structural information of the analytes, thus making the method indispensable for environmental speciation of radionuclides, generally.     

Suitability for human consumption and agriculture purposes of Sminja aquifer groundwater in Zaghouan (north-east of Tunisia) using GIS and geochemistry techniques

In Tunisia, the water resources are limited, partially renewable and unequally distributed between the wet north and the dry south of the country. The Sminja aquifer in Zaghouan city is located in north-east of Tunisia, between latitudes 36°38′ and 36°47′ and longitudes 9°95′ and 10°12′. This aquifer is used to satisfy the population needs for their domestic purposes and agricultural activities. Water analyses results are expressed by many methods, among which are geochemical methods combined with the geographic information system (GIS) (all schematic presentations of the diagram software (Piper, Riverside, Wilcox…), which can be used to assess the suitability of the Sminja aquifer groundwater for human consumption and irrigation purposes. A total of 23 wells were sampled in January 2013, and the concentrations of major cations (Na⁺, Ca²⁺, Mg²⁺ and K⁺), major anions (Cl^?, SO₄ ^2? and HCO₃ ^?), electrical conductivity and total dissolved solids were analysed. In the Sminja groundwater, the order of the cations dominance was Na > Ca > Mg > K and that of the anions was Cl > HCO₃ > SO₄. All of the analysed samples of the study area exceed chemical values recommended by the World Health Organisation guidelines and Tunisian Standards (NT.09.14) for potability but with different percentages. The aquifer spatial distribution of saturation indices reveals that all groundwater samples are under-saturated with gypsum, halite and anhydrite and are over-saturated with respect to calcite and dolomite based on water quality evaluation parameters for irrigation purposes; here, 87 % of samples in Sminja aquifer groundwater are suitable, whereas 13 % are unsuitable for irrigation uses.     

Inhibition of pyrite oxidation by surface coating: a long-term field study

Pyrite and other iron sulfides are readily oxidized by dissolved oxygen in aqueous phase, producing acidity and Fe²⁺, which causes significant environmental problems. Applications of surface coating agents (Na₂SiO₃ and KH₂PO₄) were conducted at Boeun (Chungbuk, South Korea) outcrop site, and their efficiencies to inhibit the oxidation of sulfide minerals were monitored for a long-term period (449 days). The rock sample showed positive Net Acid Production Potential (NAPP = 20.23) and low Net Acid Generation pH (NAGpH = 2.42) values, suggesting that the rock sample was categorized in the potential acid-forming group. For the monitored time period (449 days), field study results showed that the application of Na₂SiO₃ effectively inhibited the pyrite oxidation as compared to KH₂PO₄. Na₂SiO₃ as a surface coating agent maintained pH 5–6 and reduced oxidation of pyrite surface up to 99.95 and 97.70 % indicated by Fe²⁺ and SO₄ ^2? release, respectively. The scanning electron microscope and energy-dispersive X-ray spectrometer analysis indicated that the morphology of rock surface was completely changed attributable to formation of iron silicate coating. The experimental results suggested that the treatment with Na₂SiO₃ was highly effective and it might be applicable on field for inhibition of iron sulfide oxidation.     

Geoenvironmental factors related to high incidence of human urinary calculi (kidney stones) in Central Highlands of Sri Lanka

An area with extremely high incidence of urinary calculi was investigated in the view of identifying the relationship between the disease prevalence and the drinking water geochemistry. The prevalence of the kidney stone disease in the selected Padiyapelella–Hanguranketa area in Central Highlands of Sri Lanka is significantly higher compared with neighboring regions. Drinking water samples were collected from water sources that used by clinically identified kidney stone patients and healthy people. A total of 83 samples were collected and analyzed for major anions and cations. The anions in the area varied in the order HCO₃ ^? > Cl^? > SO₄ ^2? > NO₃ ^? and cations varied in the order Ca²⁺ > Mg²⁺ > Na⁺ > K⁺ > Fe²⁺. The dissolved silica that occurs as silicic acid (H₄SiO₄) in natural waters varied from 8.8 to 84 mg/L in prevalence samples, while it was between 9.7 and 65 mg/L for samples from non-prevalence locations. Hydrogeochemical data obtained from the two groups were compared using the Wilcoxon rank-sum test. It showed that pH, total hardness, Na⁺, Ca²⁺ and Fe²⁺ had significant difference (p < 0.005) between water sources used by patients and non-patients. Elemental ratio plots, Gibbs’ plot and factor analysis indicated that the chemical composition of water sources in this area is strongly influenced by rock–water interactions, particularly the weathering of carbonate and silicate minerals. This study reveals a kind of association between stone formation and drinking water geochemistry as evident by the high hardness/calcium contents in spring water used by patients.