浙江长兴煤山盆地土壤重金属来源解析及结果可靠性分析研究

发布时间：2018-02-16 06:42

本文关键词： 土壤重金属源识别源解析受体模型正定矩阵因子分解模型(PMF) 铅稳定同位素　出处：《浙江大学》2017年博士论文　论文类型：学位论文

【摘要】：土壤是环境的重要组成部分,也是人类赖以生存的自然环境和农业生产的重要资源。然而,在过去很长一段时间内,由于不合理的社会、经济发展,使许多有毒有害物质(如重金属)进入到土壤系统,对农作物生长和品质安全均产生了一定的负面影响,进而通过食物链的富集作用,对人体健康和安全构成了非常严重的潜在威胁。定性识别源类并量化各源类对于土壤重金属的贡献,是土壤重金属污染防治的前提。本文以浙江省长兴县西北部的煤山盆地农田土壤为主要研究对象,通过采集表层土壤样品,全面摸清当地土壤中的重金属污染特征和分布,并通过主成分分析、空间分析等手段,定性识别出区域内的主要污染源类。在此基础上,通过对潜在源类周围土壤剖面的对比分析,证实潜在污染源类对于周围土壤的污染;并使用铅稳定同位素技术对部分铅来源较为复杂的样点进行铅来源解析。最后,尝试使用正定矩阵因子分解(POsitive Matrix Factorization,PMF)模型确定因子的贡献及其成分谱,并结合剖面分析等对PMF结果的可靠性进行验证。主要结果如下:(1)煤山盆地表层(0-20cm)土壤镉(Cd)污染突出,平均浓度为0.43 mg kg-1,与国家土壤环境质量二级标准相比,点位超标率达30.71%。使用有限混合分布模型对表层土壤Cd浓度数据进行分析,确定当地的Cd污染阈值为0.231 mgkg-1。铅(Pb)、铜(Cu)、锌(Zn)在研究区域内污染并不严重,平均浓度分别为39.93 mg kg-1、17.25 mg kg-1和63.42 mg kg-1,均未超过国家土壤环境质量二级标准。(2)空间分析结果表明,Cd在千井湾地区和盆地东北部出现2处高值区,Pb在煤山盆地东北部的煤山工业区出现高值区。Cu和Zn的空间分布规律类似,两者的高值区与盆地内煤矿矿井的位置基本吻合。土壤pH在盆地中部有一条东西向的高值带,其空间分布规律与土壤Ca的空间分布特征相类似,并且,两者的高值区在水泥厂、采石场附近。使用主成分分析(Principal Component Analysis,PCA)共提取4个主成分,分别解译为铅蓄电池厂、建筑材料行业相关活动、煤矿开采相关活动这3类潜在人为活动源以及土壤母质这一自然源。分布在以上3类潜在人为源周围的大部分土壤样品均被有限混合分布模型分为受到Cd污染类。(3)在潜在污染源类附近采集的土壤剖面中,表层土壤(0-20cm)的Cd、Pb、Cu、Zn和Ca的浓度相对较高。其中Cd和Pb的最大浓度出现在铅蓄电池厂附近、Ca的最大浓度出现在灰岩矿区。剖面富集因子进一步证实了元素的表聚现象。各元素在不同土壤类型的自然剖面中,以及不同土壤类型的40cm以下土层中,含量相差较大。通过比较剖面表层中各元素的人为来源浓度,本研究确认研究区域内铅蓄电池厂的主要污染元素为Cd和Pb,水泥厂是Ca和Cd,煤矿开采活动则是Cd、Cu、Pb和Zn,石灰石开采仅为Ca,结果与PCA解析结果基本吻合。(4)在使用铅稳定同位素进行源解析时,建议使用丰度值来表示铅稳定同位素的组成。由于地壳样品中的204Pb含量很低,且变异程度小,所有地壳样品的铅稳定同位素组成近似落在一个平面上,可以在二维平面上展示样品的铅稳定同位素。使用丰度来表示铅稳定同位素的组成,消除了原来使用同位素比值表示组成时的不一致性问题,使得解析结果更加准确。通过对煤山盆地内部分土壤样品的铅稳定同位素分析表明,土壤背景值是土壤样品中铅的主要来源之一。(5)由于受到数据离群值值等影响,PMF模型无法对完整数据集得到合理的结果。而剔除部分数据点之后,PMF模型解析出的剖面表层土壤样品的各元素的土壤母质源的贡献大多小于剖面40 cm以下土层中的浓度,即PMF模型低估了土壤母质源的贡献。其主要原因可能是由于研究区域内的各种源类的成分谱在空间上并不稳定,呈现明显的异质性。使用剖面40 cm以下土层中的元素含量近似表征当地土壤母质的元素含量,各元素的自然含量变异较大,无法使用单一因子进行表征,需要至少4个因子才能较好地拟合剖面40 cm以下土壤元素浓度。水泥厂等人为活动源的成分谱在研究区域内不同空间位置也存在明显差异,无法使用单一因子来表征。
[Abstract]:Soil is an important part of the environment, but also an important resource for human survival of the natural environment and agricultural production. However, in the past for a long period of time, because of the unreasonable social and economic development, many toxic and harmful substances (such as heavy metals) into the soil system, which has a certain negative impact on the growth of crop quality and safety, and then through the enrichment of the food chain, constitutes a very serious potential threat on human health and safety. The qualitative identification of the source of all sources and quantify the contribution to soil heavy metals, is the premise of soil heavy metal pollution prevention. In Zhejiang province in northwestern Changxing County Meishan basin for farmland soil the main object of study, the soil samples were collected, in order to find out the local characteristics of heavy metal pollution in soil and distribution, and through principal component analysis, spatial analysis methods, qualitative Identify the main sources of pollution in the region. On this basis, through the comparison of potential sources around the soil profile analysis, confirmed the potential source of contamination to the surrounding soil pollution; sampling points and use of stable lead isotope technology on the part of the lead more complex sources lead to source apportionment. Finally, try to use the positive definite matrix factorization (POsitive Matrix Factorization, PMF) model to determine the contribution and component spectrum, combined with the profile analysis are adopted to verify the reliability of PMF results. The main results are as follows: (1) Meishan basin surface (0-20cm) soil cadmium (Cd) pollution, the average concentration of 0.43 mg kg-1, compared with the two level standard the national soil environment quality, point to exceed the standard rate of 30.71%. using the finite mixture distribution model of surface soil Cd concentration data were analyzed to determine the threshold of Cd pollution in the local area is 0.231 mgkg-1. Pb (Pb), copper (Cu), zinc ( Zn) in the study area are not seriously polluted, the average concentration was 39.93 mg and 63.42 kg-1,17.25 mg kg-1 mg kg-1, no more than two national soil environmental quality standards. (2) the results of spatial analysis showed that Cd in the northeast of 1000 wells Bay area and basin 2 high value area, high Pb value.Cu and Zn are similar in the spatial distribution of northeast Meishan basin mountain industrial zone, the high value area and Basin coal mine position basically. Soil pH has an east-west high value zone in the middle of the basin, the spatial distribution of soil Ca and the spatial distribution characteristics are similar. Also, the high value area in the cement factory of the two quarries nearby. Using principal component analysis (Principal Component, Analysis, PCA) were extracted from 4 principal components, respectively interpreted as a lead battery plant, building materials industry, coal mining activities related to these 3 types of potential Anthropogenic and natural sources of the soil. Most soil samples around the distribution in the above 3 types of potential anthropogenic sources were divided into finite mixture distribution model polluted by Cd. (3) a potential source of contamination in the soil profile type collected near surface soil (0-20cm), Cd, Pb, Cu. The concentration of Zn and Ca is relatively high. The maximum concentration of Cd and Pb in the vicinity of the lead storage battery factory, the maximum concentrations of Ca in limestone mining area. The enrichment factor profile further confirmed the element accumulation in surface elements in different soil types. The natural profile, and different types of soil below 40cm soil layer the content, the difference is great. The concentration of various elements from anthropogenic sources in the surface profile comparison, this study confirmed that the main pollution elements in the study area lead battery factory for Cd and Pb, Ca and Cd cement plant, coal mining activities are Cd, Cu, Pb and Zn, Limestone mining is only Ca and the results agree with the analytical results. (4) PCA source apportionment in the use of lead isotope abundance, suggest the use of value to indicate the composition of lead isotope. The content of 204Pb in the sample of the crust is very low, and the degree of variation is small, stable lead isotope composition of all samples fall near the crust in a plane, can show stable lead isotope samples in the two-dimensional plane. Using the composition of stable lead isotope abundance said, eliminating the original use of isotope ratios indicated the inconsistency of the composition, make the analytical results more accurate. The lead isotope of Meishan basin in the soil sample analysis showed that the background value of soil is one of the main sources of lead in soil samples. (5) due to the data outlier value, PMF model is unable to complete the data set and get reasonable results. After deleting some data points, the concentration of soil parent material source elements profile analysis of PMF model the surface of the soil sample's contribution is mostly less than 40 cm soil profile in the PMF model underestimates the contribution of soil source. The main cause is that within the study area of various source component spectrum is not stable in space, shows significant heterogeneity. The element contents in soil profile below 40 cm using the approximate representation of local soil, the content of elements of natural variation, unable to use the single factor characterization, need at least 4 factors can better fit the profile of 40 cm below the soil element concentration in cement plant. Human activities such as source component spectrum of different spatial locations in the study area there are also obvious differences, can not be used to characterize the single factor.

【学位授予单位】：浙江大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：X53;S153

【相似文献】