基于PCA-SOM的北京市平谷区地下水污染溯源

为了解北京市平谷区地下水污染物来源,以平谷区2010—2018年监测数据为基础,使用PCA(主成分分析法)识别了地下水水质指标因子,使用自组织映射识别了污染物的空间分布.结果表明:通过监测指标间的Pearson检验发现, 平谷区地下水电导率与ρ(Ca2+)(p=0.936)、总碱度与ρ(HCO₃2-)(p=0.981)、ρ(Mg2+)与总硬度(p=0.944)指标之间显著相关.地下水化学类型主要以HCO₃-Ca型为主,其次为HCO₃-Mg型.NH₄+、SO₄2-、Cd、Fe(Ⅱ)、NO₂指标空间分布离散性和差异性较大,存在局部富集现象.通过因子分析法筛选出影响平谷区地下水水质的8个公因子,首要影响因子为溶滤-富集作用(贡献率为22.398%),次要影响因子为农业、养殖业和填埋场等人为活动作用(贡献率为16.533%),雨水下渗作用(贡献率为8.035%)、工业源人为活动(贡献率为7.466%)对地下水也有一定影响.通过比较各指标的SOM(Self-Organizing Map,自组织映射)特征图像和监测井映射特征图像,发现NH₄+受山前地带林业、种植业和平原地带农业、养殖业的双重影响,Na+、Mn受平原地带人为活动的影响;同时,NH₄+、NO₃-、NO₂三者之间及Fe(Ⅱ)与Fe(Ⅲ)之间来源不同,Cd、Al、氰化物三者具有同一来源.研究显示,PCA-SOM(PCA与SOM相结合)可以对地下水化学组分来源进行定性识别与定量分析.      

Evaluation of Water Quality in the Chillán River (Central Chile) Using Physicochemical Parameters and a Modified Water Quality Index

The Chillán River in Central Chile plays a fundamental role in local society, as a source of irrigation and drinking water, and as a sink for urban wastewater. In order to characterize the spatial and temporal variability of surface water quality in the watershed, a Water Quality Index (WQI) was calculated from nine physicochemical parameters, periodically measured at 18 sampling sites (January–November 2000). The results indicated a good water quality in the upper and middle parts of the watershed. Downstream of the City of Chillán, water quality conditions were critical during the dry season, mainly due to the effects of the urban wastewater discharge. On the basis of the results from a Principal Component Analysis (PCA), modifications were introduced into the original WQI to reduce the costs associated with its implementation. WQI_DIR2 and WQI_DIR, which are both based on a laboratory analysis (Chemical Oxygen Demand) and three (pH, temperature and conductivity), respectively, four field measurements (pH, temperature, conductivity and Dissolved Oxygen), adequately reproduce the most important spatial and temporal variations observed with the original index. They are proposed as useful tools for monitoring global water quality trends in this and other, similar agricultural watersheds in the Chilean Central Valley. Possibilities and limitations for the application of the used methodology to watersheds in other parts of the world are discussed.     

抑制斑点噪声的SAR与多光谱图像融合方法

通过将主成分变换与IHS变换相结合,避免了会导致SAR信息量损失的滤波过程,实现了一种可降低SAR斑点噪声影响的遥感影像融合算法。与主成分变换融合方法、IHS变换融合方法在主观视觉效果分析和客观统计参数两方面进行比较,新方法不仅很好地保留了多光谱影像的光谱信息,增强了结果影像的空间细节表现能力,而且有效地抑制了SAR斑点噪声,提高了结果影像的信息量与清晰度。     

The Distribution and Sources of Polycyclic Aromatic Hydrocarbons in Surface Sediments Along the Egyptian Mediterranean Coast

Coastal marine sediment samples were collected from 31 sampling stations along the Egyptian Mediterranean Sea coast. All sediment samples were analyzed to determine aliphatic and polycyclic aromatic hydrocarbons (PAHs) as well as total organic carbon (TOC) contents and grain size analysis. Total concentrations of 16 EPA-PAHs in the sediments were varied from 88 to 6338 ng g⁻¹ with an average value of 154 ng g⁻¹ (dry weight). However, the concentrations of total aliphatic were varied from 1.3 to 69.9 ng g⁻¹ with an average value of 15.6 ng g⁻¹ (dry weight). The highest contents of PAHs were found in the Eastern harbor (6338 ng g⁻¹), Manzala (5206 ng g⁻¹) and El-Jamil East (4895 ng g⁻¹) locations. Good correlations observed between a certain numbers of PAH concentrations allowed to identify its origin. The average total organic carbon (TOC) percent was varied from 0.91 to 4.54%. Higher concentration of total pyrolytic hydrocarbons (∑COMB) than total fossil hydrocarbons (∑PHE) declared that atmospheric fall-out is the significant source of PAHs to marine sediments of the Egyptian Mediterranean coast. The selected marked compounds, a principal component analysis (PCA) and special PAHs compound ratios (phenanthrene/anthracene vs fluoranthene/pyrene; ∑COMB/∑EPA-PAHs) suggest the pyrogenic origins, especially traffic exhausts, are the dominant sources of PAHs in most locations. Interferences of rather petrogenic and pyrolytic PAH contaminations were noticed in the harbors due to petroleum products deliveries and fuel combustion emissions from the ships staying alongside the quays.     

Comparative Metal Distribution in Hair of Pakistani and Libyan Population and Source Identification by Multivariate Analysis

Using nitric acid-perchloric acid wet digestion based FAAS method ten selected metals (Cd, Co, Cr, Fe, K, Mn, Na, Ni, Pb and Zn) were determined in the scalp hair of male Pakistani and Libyan donors (n = 62), between 3–54 years age and residing in typical urban areas. The study was taken up to identify sources of metal distribution in two diverse population segments exposed to different environments. Sodium emerged with the highest mean hair concentration of 511.0 μg/g and 496.9 μg/g for Libyan and Pakistani donors respectively, followed by K, at 266.2 μg/g and 242.4 μg/g for the two donor categories. Cadmium showed the lowest mean hair concentration for both Pakistani (0.380 μg/g) and Libyan (0.530 μg/g) donors. Levels of Na, K, Pb, Cr and Cd were higher in hair of Libyan donors compared with Pakistani counterparts which showed higher comparative levels of Zn, Fe, Co, Ni and Mn. Strong metal-to-metal correlations were found between Na–K (r = 0.638) and Cd–Co (r = 0.574) for Pakistani donors, while for Libyan donors, Na–K (r = 0.680) and Fe–Mn (r = 0.624) correlations emerged as strongly significant. Only K was found to have a strong positive correlation (r = 0.611) with age for Pakistani donors while this correlation was significantly negative (r = −0.500) for Libyan donors, probably arising from individual food habits. Other metals showed no viable relationship with age. Principal Component Analysis (PCA) and Cluster Analysis (CA) of the data evidenced food, industrial emissions and automobile emissions as possible sources of metal distribution. The results of the present study are compared with those reported for subjects from other regions of the world.     

Tracing the sources of refractory dissolved organic matter in a large artificial lake using multiple analytical tools

Structural and chemical characteristics of refractory dissolved organic matter (RDOM) from seven different sources (algae, leaf litter, reed, compost, field soil, paddy water, treated sewage) were examined using multiple analytical tools, and they were compared with those of RDOM in a large artificial lake (Lake Paldang, Korea). Treated sewage, paddy water, and field soil were distinguished from the other sources investigated by their relatively low specific UV absorbance (SUVA) values and more pronounced fulvic-like versus humic-like fluorescence of the RDOM samples. Microbial derived RDOM from algae and treated sewage showed relatively low apparent molecular weight and a higher fraction of hydrophilic bases relative to the total hydrophilic fraction. For the biopolymer types, the presence of polyhydroxy aromatics with the high abundance of proteins was observed only for vascular plant-based RDOM (i.e., leaf litter and reed). Molecular weight values exhibited positive correlations with the SUVA and the hydrophobic content among the different RDOM, suggesting that hydrophobic and condensed aromatic structures may be the main components of high molecular weight RDOM. Principal component analysis revealed that approximately 77% of the variance in the RDOM characteristics might be explained by the source difference (i.e., terrestrial and microbial derived) and a tendency of further microbial transformation. Combined results demonstrated that the properties of the lake RDOM were largely affected by the upstream sources of field soil, paddy water, and treated sewage, which are characterized by low molecular weight UV-absorbing and non-aromatic structures with relatively high resistance to further degradation.     

Concentrations of PCDD/PCDF in soil close to a secondary aluminum smelter

Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDDs/Fs) were analyzed in samples of the emissions from a secondary aluminum smelter (ALS) and soil samples around the plant. The purpose was to estimate the impact of the emissions on the surrounding environment.PCDD/F soil concentrations were higher in the proximity of the plant, exceeding the limit adopted in Italy in soils for green areas and residential uses and the upper limit of several reference concentrations. The most contaminated sites were less than 500 m from the plant and the dioxin concentration with the distance from the ALS.Principal component analysis (PCA) showed that emissions from the ALS were the source of PCDD/F contamination in the soils closest to the plant. Multivariate data analyses such as PCA are therefore useful to identify sources of emission causing contamination.     

Determination of direct and fugitive PM emissions in a Mediterranean harbour by means of classic and novel tracer methods

Remodelling works are frequent in harbour areas, given that they must adapt to rapidly changing market requirements. Depending on their nature (construction, demolition, etc.), these works may be carried out during long periods of time and thus exert a significant impact on the air quality at the harbours and their surroundings. The air quality at the Valencia harbour was studied during an enlargement project. Four sampling stations aimed to cover the major dust-emitting activities in the harbour. In addition, a sampling campaign was carried out to collect dust materials at their emission sources (e.g., loading and unloading of clinker and petroleum coke, as well as the enlargement works). Differences obtained between PM levels at the monitoring sites were correlated with the distance between sampling stations and enlargement works and/or harbour operations, as well as with the preferential wind direction. In all cases, the days with the highest PM_10 levels were linked to wind directions coinciding with the location of the enlargement works or the clinker and petroleum coke stocks. Classic source apportionment methodologies (PCA and CMB) were applied to the data, but their interpretation was complex due to the similar chemical signatures of PM originating from direct and fugitive emissions from stocked materials (e.g., clinker), and the enlargement works. To overcome this difficulty, a novel non-statistical approach was used to obtain quantitative estimations of the contributions from sources (petroleum coke and clinker), based on the correlation between specific PM components (e.g., carbonaceous species) and source tracers (e.g., V). Finally, a qualitative test using phenolphthalein was devised to identify the presence of clinker on the filters, which provided positive results. This novel tracer approach is thus recommended for harbour authorities.     

模式识别在火灾调查中的汽油分类问题的应用研究

在火灾调查中,检测汽油成分并对其进行正确分类尤为重要.运用GC-MS对90#和93#两种普通汽油的共50个样本进行检测,所得的GC-MS原始数据通过PCA方法进行处理,以提取有用信息,避免冗余变量进入后续计算.在此基础上应用KNN方法对这两种汽油助燃剂进行分类.结果表明, KNN方法对这两种汽油的分类准确率达到100%,且当初始数据未经标准化预处理时也能达到同样准确的分类效果.研究表明:将模式识别方法正确地运用到助燃剂鉴定和分类工作中有助于火灾调查.     

苏州市区灰霾现象形成的气象条件分析

随着社会和经济的快速发展,灰霾现象的发生有逐渐上升趋势,严重影响城市形象和群众的身心健康。苏州市的灰霾现象发生频率1—2月最高,7—8月最低。通过寻找形成灰霾现象的主要气象控制因子,有助于判别不同气象条件在灰霾天气形成中的作用。采用基于因子分析的主成分提取方法,将6个气象因子观测资料整合为3个主成分,并逐一揭示4个典型季节各主成分的支配因子在灰霾天气形成中的作用。结果表明,热量条件是影响1—2月、4—5月和7—8月霾日形成的主要气象条件。动力条件是10—11月霾日形成的主要气象条件,水分条件影响相对较弱,仅起辅助作用。