【作者】 冯金飞；

【导师】 卞新民；

【作者基本信息】 南京农业大学 ， 生态学， 2010， 博士

【摘要】 公路交通是环境中重金属污染物的主要排放源之一。由于耕地资源有限,全球公路沿线的作物种植仍非常普遍。鉴于农产品质量安全的隐患,公路交通对沿线农田土壤和作物的重金属污染一直受到学者和公众的广泛关注。大量研究表明,公路沿线土壤和作物均受到了不同程度的Pb、Cd、Cr、Zn等重金属污染,但对其污染特征及规律的认识尚不非常清楚。公路沿线土壤中重金属含量相对较低,而沿线的水稻、小麦、蔬菜、水果等农产品中均存在Pb、Cd、Zn等重金属超标的现象；土壤中累积量较高的重金属元素,在作物中含量并不一定高,而土壤中累积量较低的元素,在作物中含量却较高。导致这些不确定性的主要原因可能是由于对公路沿线重金属污染途径、土壤中不同重金属的生物有效性及其影响因素等的认识不足。高速公路作为未来交通的主干网络,对沿线农田环境的影响尤其严重。因此,研究高速公路沿线农田土壤和作物中重金属分布特征、累积规律和影响因素,可为我国高等级公路沿线作物生产的产地环境保护和种植规划提供科学依据和技术支撑。本论文以江苏省交通流量最大的两条高速——沪宁高速和京沪高速为研究对象,于2007-2009年选择车流量不同的六个典型路段,采集公路两侧农田土壤和作物(水稻、小麦)样品,测定土壤中重金属(Pb、Cd、Cr、Zn和Cu)总量和有效态含量,以及作物籽粒中重金属含量；分析高速公路沿线土壤和作物中重金属含量的分布特征和影响因素。同时采用盆栽对比试验和稳定性Pb同位素示踪法,分析高速公路沿线水稻、小麦中重金属的主要来源(大气或土壤),以及不同器官中大气来源重金属和土壤来源重金属所占比例。在产地监测和作物试验的基础上,借助人工神经网络法对高速公路两侧农田土壤和作物中重金属含量进行模拟,建立高速公路沿线土壤和作物重金属含量的预测模型。本文获得的主要结论如下：1.沪宁和京沪高速六个路段两侧土壤、水稻和小麦均受到不同程度的重金属污染,污染边界最远已经达到了路两侧330m。土壤中Pb、Cd、Cr、Zn和Cu含量均高于对照区土壤,但没有超过国家土壤环境二级标准；水稻和小麦中上述五种重金属含量均高于对照样品,部分样品中Pb、Cd、Zn含量超出了国家食品安全限量标准,其中Pb和Cd的超标率较高。2.高速公路两侧土壤和作物中重金属含量的空间分布特征差异明显。公路两侧土壤中Cd、Cr、Zn和Cu含量随与公路距离的增加而不断降低,Pb含量随与公路距离的增加而先增加再不断降低。高速公路两侧水稻和小麦籽粒中五种重金属含量均随与公路距离的增加而先增加再不断降低。公路两侧,作物中重金属含量高值区的分布与土壤中重金属含量高值区的分布存在差异,土壤中重金属含量高的区域作物中重金属含量并不一定高。3.高速公路沿线水稻和小麦中的重金属污染来源于不同的途径,大气污染途径不可忽视。盆栽对比试验和作物中Pb稳定性同位素组成分析结果显示,高速公路沿线水稻中累积的Pb、Cd和Zn部分来源于叶片对大气中重金属的吸收,而Cr和Cu主要来源于根系对土壤中重金属的吸收。高速公路旁水稻叶片中,大约有20%的Pb、35%的Cd和60%的Zn来源于大气；在水稻籽粒中,约有46%的Pb和41%的Cd来自于叶片的吸收和转运；在水稻茎中,约有49%的Zn来自于叶片的吸收和转运。公路旁不同距离水稻叶、茎、籽粒中大气来源的Pb、Cd或Zn的比率随与公路距离增加而不断降低。高速公路沿线小麦中Cd和Zn部分来源于叶片对大气中重金属的吸收,而Pb、Cr和Cu主要来源于根系对土壤中重金属的吸收。在公路旁小麦叶片中,大约有22%的Cd和29%的Zn来自于叶片对大气中重金属的吸收,在小麦籽粒中,约有21%的Cd和20%的Zn来自于叶片的吸收和转运,在小麦茎中,Pb、Cd、Cr、Zn和Cu主要来自于根系的吸收和转运。公路旁不同距离小麦叶、籽粒中大气来源的Cd和Zn的比率随与公路距离增加而不断降低。5.车流量是高速公路沿线土壤中重金属累积最主要的影响因素,沿线作物中重金属的累积受到车流量、风向、土壤性质、重金属有效态含量等因素的综合影响。沪宁高速和京沪高速沿线六个路段土壤中Pb、Cd、Zn和Cu累积指数与车流量呈显著正相关,风向、土壤pH、有机质等因素的影响相对较小。沪宁高速和京沪高速沿线6个路段水稻中Pb和Cd累积指数、小麦籽粒中Cd和Zn累积指数与车流量呈显著正相关。沿线水稻、小麦籽粒中重金属的累积受到车流量、风向、土壤性质、重金属有效态含量等因素的综合影响,不同重金属元素累积的最主要的影响因素各有不同。6.BP神经网络具有很强的自学习、自组织与自适应功能,具有高度非线性函数映射功能,将其应用于高速公路沿线农田土壤和作物中重金属含量分布的预测与评价,拟合精度较高,泛化能力好。能够对高速公路两侧土壤中Pb、Cd、Zn和Cu含量、对两侧水稻籽粒中的Pb和Cd含量、小麦籽粒中的Pb、Cd和Zn含量进行较好的拟合和泛化。更多还原

【Abstract】 The highway traffic is the main source of heavy metal pollution. Due to limited cropland area, it is very common to plant crops along the highways. So, in view of agricultural products safety, the heavy metal pollution by the highway traffic to the soils and crops along the highways is widely concerned by scientist and public. Lots of evidence has demonstrated that the soils and crops along the highway were contaminated at various degrees by heavy metals such as Pb, Cd, Cr and Zn etc. However, the traits and laws of the pollution were unclearly documented. The soils along the highway have lower heavy metal content while the crops, such as rice, wheat, vegetables and fruits, contain Pb, Cd, Cr and Zn over the national guidance limits. The heavy metal which is higher in soils is not necessarily higher in crops and which is lower in soils is higher in crops. These uncertainties may be mainly contributed to the uncertainties about the polluting pathways of the heavy metals, biological availability and its influence factors of heavy metal in soils. The express highway will impact agro-environment more seriously due to its main part in future traffic. Therefore, to study the distribution traits, accumulative laws and influence factors of heavy metals in agricultural soils and crops can provide scientific evidence and theoretical basis for the environmental protection and cultivation planning along the express highway.This study took the Shanghai-Nanjing and Beijing-Shanghai express highways which have the heaviest traffic flows as research objects. In 2008-2009, six typical sections with different traffic flows were selected for collection of the bilateral soils and crops (rice and wheat) along the highway. The total and available contents of heavy metals (Pb, Cd, Cr, Zn and Cu) in soil and the contents of heavy metal in grains of rice and wheat were measured. The distribution traits and influence factors of the heavy metal were analyzed. At the same time, by pot experiment and stable Pb isotope tracing method, source origin (atmosphere or soil) of the heavy metal in rice and wheat, and the proportion of the heavy metal from atmosphere or soils in different organs were analyzed. Based on the field monitoring and simulation experiment, the content of heavy metal in roadside agricultural soils and crops were simulated by the artificial neural network method. From the simulation, the prediction model of heavy metal contents in soils and crops were built. The main conclusions in our study were as follows:Along the six sections, the bilateral soil, rice and wheat were contaminated by heavy metal with the furthest border reaching 330 m. The Pb, Cd, Cr, Zn and Cu contents in soils were higher than the control but no more than the maximum allowable conentrations. These heavy metals in rice and wheat were also higher than control. The Pb, Cd and Zn contents in some plant samples were higher than the national guidance limit with higher over limit ratio existed in Pb and Cd.There existed obvious spatial distribution differences in heavy metal contents of roadside soils and crops. Along with the increased distance from highways, the Cd, Cr, Zn and Cu contents in soils decreased while Pb increased first and then decreased. And the five heavy metals in grains of rice and wheat all increased first and then decreased along with, the increased distance from highway. There were differences between the high value area of heavy metal content in soils and in crops. The area where the heavy metal was higher in soils hasn’t had necessarily higher heavy metal in crops.The heavy metals in rice and wheat along the express highway came from different pathways among which the atmosphere pathway couldn’t be neglected. Results from pot experiment and stable Pb isotope showed that part of Pb, Cd and Zn accumulated in rice were derived from the atmosphere via foliar uptake, while Cr and Cu were mainly from the soil via root uptake. In rice leaves, about 20% Pb,35% Cd and 60% Zn were from atmosphere. In rice grains, about 46% Pb and 41% Cd were derived from the atmosphere via foliar uptake. In rice stem,49% Zn were also derived from the atmosphere via foliar. The ratios of atmosphere-originated Pb, Cd and Zn in rice leaves, stem and grains decreased along with the increased distance from the highway. As for wheat, part of Cd and Zn in wheat came from atmosphere via foliar uptake, while Pb, Cr and Cu mainly came from the soil via root uptake. In wheat leaves, about 22% Cd and 29% Zn were from atmosphere. In wheat grains, about 21% Cd and 20% Zn were from the atmosphere. In wheat stem, Pb, Cd, Cr, Zn and Cu mainly came from absorption and transportation by roots. The ratios of atmosphere-originated Cd and Zn in rice leaves and grains decreased along with the increased distance from the highway.The traffic density had significant effect on the accumulation of heavy metals in soils. The accumulation of heavy metals in rice and wheat grains were affected by the traffic desenty, wind direction, soil pH and organic matter contents, and the total and available heavy metals contents. Accumulation coefficients of Pb, Cd, Zn and Cu in soils along the six sections had significant positive correlation with the traffic density. The influence of wind direction, soil pH and soil organic matter were releativly lower. The traffic fluxes were significantly positively correlated with the accumulation coefficients of Pb and Cd in rice and the accumulation coefficients of Cd and Zn in wheat grains. The accumulation of heavy metals in roadside rice and wheat was affected by the traffic density, wind direction, soil properties, and available contents. The most important factors for each heavy metals were different.The BP neural network has a strong ability of self-learning, self-organizing and self-adapting and a high function of nonlinear function mapping. So it will have high fitting precision and good generalization ability to apply the Bp neural network to the prediction and evaluation of the heavy metals distribution and content in soils and crops along with the express highway. And the BP neural network can fit and generate on the Pb, Cd, Zn and Cu content in soils, Pb and Cd contents in rice grains and Pb, Cd and Zn contents in wheat grains.更多还原