有机化学品不同温度下(过冷)液体蒸气压预测模型的建立与评价

(过冷)液体蒸气压(PL)是评价化学品在环境中分配、迁移和归趋行为的重要参数。PL具有较强的温度依附性。发展一种能够精确预测不同环境温度下化学品PL的方法,有助于填补化学品生态风险评估的大量数据缺失。本研究收集整理了661种有机化合物在不同温度下(200~830 K)共计10 478个log PL值。在此基础上,采用偏最小二乘(PLS)回归和支持向量机(SVM)方法,构建了PL的线性和非线性预测模型。结果表明:2种模型均具有良好的拟合度、稳健性及预测能力,SVM模型的预测性能略高于PLS模型(PLS:R2adj.tra=0.912,RMSEtra=0.477,Q2ext=0.910;SVM:R2adj.tra=0.997,RMSEtra=0.092,Q2ext=0.967)。机理分析表明,温度是影响PL的主要因素,温度越高,蒸气压越大;其次,X1sol也影响PL大小,X1sol用来描述分子间的色散作用,分子间色散力越小,蒸气压越大;此外,化合物的氢键个数、极性和分子构型等因素也影响PL大小。采用Wiliams plot方法表征了PLS模型应用域。所建立的模型可用来预测烷烃、烯烃、醇、酮、羧酸、苯、酚、联苯、卤代芳香烃、含N化合物及含S化合物在不同温度下的PL数据。     

The Effect of Grain Size and Element Concentration in Identifying Contaminant Sources

The distribution of metal contaminants between different size fractions of marine sediments is well known. However, the use of size normalization techniques may alter the ability or usefulness in identifying potential sources in complex environments. In a reassessment of metal data from the shelf area of Sydney, Australia, the mud and sand fractions were investigated separately by PCA and PLS methodologies. The analyses were able to produce clear distinctions between industrial/urban sources when based on a suite of metals rather than individual (single-element) concentrations. Signature analysis by PLS with copper, lead, zinc, manganese, chromium, cobalt, nickel, and cadmium demonstrated the dispersion of the fine-grained contaminated material to the south in the East Australian Current. However, due to the commonality between many of the metals, a subset of four metals was used to define the signature. This significantly improved separation, showing clear plumes extending ～30 km from the source rivers.     

偏最小二乘(PLS)回归方法在中国东部植被变化归因研究中的应用

植被变化往往受到不同气候变量的综合作用,人类活动影响又使得植被对气候响应变得更为复杂,如何准确判别各种影响因素的相对重要性是植被变化归因研究中的一个关键点.研究基于偏最小二乘(PLS)回归方法,使用1982-2006 年的归一化植被指数(NDVI)数据,分析了降水、气温、日照、相对湿度和风等气候变量对中国东部植被变化的相对影响,并选取了NDVI变化较为典型的区域,量化了农业活动对该地区植被变化的相对贡献.PLS回归方法兼具了主成分分析和多元回归的优点,克服了众多自变量之间存在强烈交互相关导致的多元共线性问题.研究结果表明:① 1982-2006 年间,中国东部逐月NDVI的年际变化呈现出明显的南北差异.在12、1-5 月,NDVI以显著上升为主,上升区域主要位于淮河以北.在6-10月,NDVI以显著下降为主,下降区域主要为淮河以南的部分区域,特别是6 月江苏一带NDVI的大范围下降尤为明显.不过与NDVI发生显著变化的区域相比,更多区域的NDVI在各月并没有出现显著变化.② 在NDVI显著上升的站点,对NDVI变化最具解释意义的气候变量为气温,特别是冬末春初(2-3 月)的升温对黄淮海区域NDVI的显著上升具有主导控制作用.而对于NDVI 显著下降的站点,多数都不能从气候角度解释这些区域的NDVI 变化.③ 江苏省NDVI在6 月出现的大范围显著下降,与农业种植结构的调整,主要是棉花种植面积的减少以及油菜面积的增加具有显著关系.     

Particulate and semivolatile associated aliphatic hydrocarbon exhaust emission from heavy duty diesel vehicles

The exhaust emissions from two heavy duty diesel vehicles running on eight different fuel compositions were investigated regarding their content of high molecular weight (≥ C₁₂) aliphatic/ olefinic hydrocarbons. It was concluded that the emitted amount of semi‐volatile associated aliphatic hydrocarbons (range C₁₂‐C₂₂) depend on the fuel used in the engines and that these emissions mainly consisted of uncombusted fuel components. It was also found that uncombusted engine lubrication oil was the main constituent of the emitted particulate associated aliphatic hydrocarbons (C₁₇‐C₄₀). These constituted between 58% and 95% of the total emissions of the high molecular weight aliphatic compounds. Emission factors for the total of high molecular aliphatic hydrocarbons (C₁₂‐C₄₀) were demonstrated to be in the range of 15–100 mg/km. Some individual aliphatic hydrocarbons with cocarcinogenic effects were identified and quantified in both particulate and semi‐volatile phases of the exhaust. Multivariate data analysis was used to investigate the relationship between fuel parameters and emission of semi‐volatile aliphatic emission.     

基于地统计模型的上海大气污染物多建模方法的比较

地统计模型被广泛应用于环境空气污染物暴露模拟,但不同建模方法及其模拟结果之间的对比研究较少.基于上海2016~2019年55个环境空气监测点位的NO₂和PM_2.5观测数据,以及交通路网、排放源兴趣点和卫星数据等地统计变量,应用偏最小二乘回归（PLS）、监督学习线性回归（SLR）和机器学习随机森林（RF）这3种建模方法创建年暴露模型,并进一步应用普通克里金插值（OK）法分析模型残差,构建复合模型.应用交叉验证对模型的模拟效果进行检验,选取每一种建模方法的最优模型结构（是否应用OK）作为最终模型.结果表明,NO₂模型中表现最好的是RF-OK （R_mse²为0.70~0.82）和PLS-OK模型（R_mse²为0.78~0.84）;PM_2.5模型中PLS模型（R_mse²为0.62~0.71）优于SLR-OK （R_mse²为0.40~0.79）和RF-OK （R_mse²：0.31~0.56）模型.应用3种建模方法对上海1 km网格开展年暴露模拟和对比,NO₂模型间模拟结果的相关性（r为0.82~0.91）高于PM_2.5模拟结果的相关性（r为0.66~0.96）.基于3种模型2019年的模拟结果,评估了上海NO₂和PM_2.5的人群暴露水平.     

Does Soil Acidification Affect Spruce Needle Chemical Composition and Tree Growth?

In 1994, a large survey of soil chemistry was undertaken in thecounty of Värmland in central Sweden (Lundström et al., 1998).The southern part of the county was affected by soilacidification whereas there were no such indications in thenorthern part. To investigate the influence of soil chemistryon the trees at the specific sites, the survey was continued byan analysis of needle chemistry (Norway spruce) which wasundertaken at 150 of the 180 sites, and of tree growth at 65 ofthe 180 sites. Growth was expressed as a ratio between expectedgrowth, estimated with a national, empirical growth model, andthe growth observed in the field. In statistical analyses,using rank correlation, PCA and PLS, there were only weakindications of an influence of soil chemistry on needlechemistry and on tree growth. A moderate correlation betweennitrogen and sulphur in needles was found, which wasinterpreted as an effect of deposition and of processes in thetree canopy. No obvious regional pattern of the growth ratiowas found, in contrast to the clear pattern of soilacidification. The statistical analysis could not with anycertainty point out any of the soil chemistry variables asespecially important for the tree growth ratio.     

厌氧反应器废水资源化过程在线监测系统

为克服现行厌氧反应器状态离线分析的滞后性,充分利用资源,改进废水排放状态,研究了基于荧光光谱的废水资源化处理过程在线监测系统。系统以厌氧反应器中的关键产物色氨酸、核黄素及辅酶为主要检测组分,自动定位滤波轮并控制脉冲氙灯的闪烁次数,使用光电倍增管采集荧光信号。针对反应产物的混合体系,提出了将三维数据二维化的准三维荧光谱概念,优化获得了的3×3荧光强度矩阵,采用偏最小二乘法并优化其模型参数,实现了反应器中3种关键过程产物的实时在线检测、统计与分析,使其回收率范围达到85%至112%。     

Characterisation and determination of profiles of polycyclic aromatic hydrocarbons in a traffic tunnel in Gothenburg, Sweden

The concentrations of semi-volatile polycyclic aromatic hydrocarbons (PAHs), hydrocarbons (HCs), particulate matter (PM 1, 2.5 and 10 μm) and total suspended particles (TSPs) were measured in a traffic tunnel in Gothenburg, Sweden. Emission factors (EFs) were also calculated. These variables are assumed to provide good estimates of average vehicle emissions, since all types of vehicle, using all types of fuel, pass through this tunnel. It was shown that the majority of particle-associated PAHs were found on particles with an aerodynamic diameter of <1 μm. The concentrations of PAHs were one order higher in magnitude in air samples from the tunnel than in air samples at two urban locations. However, the PAH profiles of air samples from the tunnel and the urban sites were similar. This was demonstrated using principal component analysis (PCA). Finally, and notably, there was no significant change in the total emissions when the proportion of heavy-duty vehicles (HDVs) increased from 8% to 24%. Previously, diesel vehicles had been found to release larger quantities of PAHs and related substances. Advances in fuel quality, and HDV motor and exhaust system design during the last decade may have contributed to this promising result. However, it was shown, using partial least squares regression to latent structures (PLS), that some of the parameters measured displayed correlations with the proportions of HDVs and light-duty vehicles (LDVs). Concentrations of total HCs, TSPs, dibenzothiopene, phenantrene, anthracene and monomethyl-derivatives of phenantrene and anthracene were all correlated to the proportion of HDVs. The concentrations of naphthalene, some mono- and dimethylnaphthalenes and most large PAHs (with 5–7 fused rings) were correlated to the proportion of LDVs.     

市域能源碳排放影响因素分析及减碳机制研究——以福建省为例

选取福建省市域作为研究对象,应用扩展型的STIRPAT-PLS模型对2010~2016年福建省市域碳排放影响因子进行实证分析,探讨驱动福建省市域碳排放量增长的主要影响因素和各因素的影响程度,明确碳排放控制的主要领域.结果表明：总人口、城镇化率、人均GDP、第二产业比重和能源强度对碳排放量增加有正向驱动作用,而第三产业比重对碳排放量增加有负向驱动作用.总人口、城镇化率以及第二产业比重对碳排放量增长的贡献最大.据此,建议福建当前及未来时期减碳重点应是采取优化能源结构和促进产业结构升级,加速发展清洁、再生能源与提高能源效率相并重等策略.     

Molecular modeling in environmental science∗

Quantum mechanic computer models are often used in chemistry to predict properties of molecules and to simulate reaction pathways. Such models calculate bond lengths, bond angles, dihedral angles, the energy of the molecular‐orbitals (MO's), the dipole moment, the ionisation potential, and the heat of formation. These results lead to detailed information on electronic structures like bond orders, electronic charges and levels of frontier orbitals (HOMO, LUMO).

It is possible to calculate properties of ground state molecules as well as ionic and radicalic structures, or reaction intermediates and other compounds with very short lifetimes.

From the exact knowledge of electronic structure, frontier orbitals, and reaction intermediates it is possible to predict stability of chemicals in the environment but also of fictive chemicals, which are not yet synthesized. Some authors have shown that there are correlations between electronic structure and toxicity.

The most interesting models for environmental chemistry are semiempirical models, such as MINDO, MNDO or MOPAC. Theses programs are able to handle molecules with 30 and more heavy atoms (all elements without hydrogen), and it is possible to install them on main frame computers (CPU‐time several minutes to hours) and on personal computers, with coprocessors (CPU‐time several hours to a few days).

Normally the molecule is read in from a Z‐matrix (a matrix of polar coordinates of atoms connected in the molecule) and with symmetry data. Furthermore, programs are available which create a Z‐matrix from the molecular structure plotted on the screen with a mouse.