环境中大气颗粒物源解析方法对比研究
|
摘要
源解析技术的发展为环境管理提供了新思路,其研究结果对空气污染控制政策的制定起着重要作用。目前,大气颗粒物源解析研究多采用化学质量平衡法和因子分析法两种方法,而投影寻踪回归分析法在大气源解析分析中有其突出的特点,所以本文选择了这三种方法进行对比研究。本文首次将基于免疫进化算法的投影寻踪回归法用于源解析研究。大气颗粒物源解析系统软件可以VB6.0作为程序的开发语言,并结合Access数据库开发而成。本文基于CMB模型对四个城市进行源解析基础之上,采用因子分析(FA)和投影寻踪回归分析(PPR)方法,建立相应的源解析模型,编写相应的处理软件,对吉林省四个城市的采暖期及非采暖期源和受体样品数据进行源解析处理,计算最终各源的贡献率或贡献值,并对处理结果进行了对比分析。在实际工作中,选用适当的方法和模型,对源和受体分别分析,即多种方法结合,可使得到的结果更加准确。
Total suspended particulate (TSP) is a kind of complex contaminant in the ambient air. Currently,TSP is seriously overproof in China,while severe pollution of TSP is the main factor of the ambient air of the Jilin Province in many years.TSP is harmful to the health of inhabitants living and it has caused tremendous damage on the national economy.The development of Source Apportionment technology provides a new idea on environmental management, the research results play an important role of air pollution control policies.Therefore, enhanceing the sources analysis of TSP of pollutionand pollution control is particularly important to achieving the fundamental improvement of air quality of Jilin Province.
Source apportionment studies of TSP adopt mostly CMB and FA .To determining the source contributing to the receptor concentration and the contribution rate of quantitative sources ,CMB uses the measured physical and chemical properties of TSP of the polluted sources and receptors.CMB bases on mass conservation, it has a clear principle, can be easily accepted and is the used widely methods. The main objective of FA is concentrating information, reducing the dimension of indicators, simplifing the structure of indicators, avoiding multicollinearity problems between variables under the loss of principal information as little as possible.The analysis of indicator system is more simple and effective.On the study of TSP source apportionment, the public factor often represents the source of TSP, the number of the distinguishing factors of FA is the number of pollution sources, according to the relative size of the factor loading of FA in combination with the understanding of the sources pollution nature, the types of pollution sources represented by public factors is qualitatively descripted.
PPR has a prominent feature on the study of source apportionment. Without any request to the structure and characteristics of the data,linear or nonlinear, normal or non-normal data can be effectively dealt with by PPR.The source apportionment of TSP on PPR may not limited by the conditions,has the good objectivity and be more practical.And PPR takes full advantage of the spectrum distribution of the source and the data information of monitoring and analysis of atmospheric samples, the model has good stability and anti-jamming, high precision, strong reliability.
Visual Basic is a visual, object-oriented ,using event-driven ,high-level structured programming language,it can be used to developing all kinds applications under the Windows environments. It is easy to learning, efficient, and powerful.To make it better and more powerful, Visual Basic 6.0 increases many new features on the base of the original Visual Basic version.The source apportionment system software can use VB6.0 as a program development language, combining with Access database .The software users can easily query, modify and input the required monitoring data.
On the bases of the calculation of source apportionment.for four cities by the CMB model, This article uses FA and PPR methods, the builds up the corresponding model, compiles the calculation program, analyzes and calculates the data of source and receptor samples in the heating period and non-heating period,ultimately the contribution rate or contribution value of every source.Given full consideration to the characteristics of four typical cities, combined with the results of CMB analysis methods, the following results of comparative analysis of TSP source apportionment is available.
The results of TSP source apportionment in the heating period is that raised dust is emission sources of larger TSP contributions of Jilin, Siping andTonghua,soil dust is the major source of air pollution of Baicheng by the CMB model,while road dust and soil dust is emission sources of larger TSP contributions of Jilin and Tonghua, road dust and raised dust is the main source of air pollution of Siping and Baicheng by the PPR model.
The results of TSP source apportionment in the non-heating period is that the emission sources of larger TSP contribution of Jilin, Baicheng, Tonghua and Siping as follows: construction dust, soil dust wind, road dust and dust by the CMB model,while the proportions of the remaining various types dust sources are not low, in addition to coal dust, the distribution of every points is different by the PPR model.
The results of TSP source apportionment in the whole year is that, by the CMB model, the sources of 5 types of contribution on TSP of Jilin as follows: soil dust,raised dust, construction dust, road dust and coal dust, the results are consistent with the analytical results of the city's heating period and non-heating period, soil dust and coal dust is the sources of 5 types of large contribution on TSP of Baicheng, and others have smaller impact on air quality, road dust and coal dust is the main sources in Tonghua, raised dust is the main source in Siping, by the FA model,soil dust is the main source in Jilin, coal dust is the main source in Baicheng, construction dust and coal dust is the main sources in Tonghua, road dust is the main source in Siping.
CMB model is more effective for needing the contribution value and contribution rate. Under the known receptor and unknown sources,according to the correlation of the receptor data Combination with the knowledge of the characteristics element of the source e, FA model infer the source, the analysis on the source data is more accurate.FA requires a large amount of receptor data.the more data, more accurate results. Undre the case of various types of the known sources, PPR model compares the ratio of the effort of the same types of receptors on the sources,the is simiply, software is easily operated,understanding and using.This is of great help for qualitative analysis of source apportionment study. The models of three methods is different,the type of results is different, the results also vary. In practice work, selecting the appropriate methods and models On the analysis of source and receptor, respectively, that is combined with many ways could make the results more accurate.
Total suspended particulate (TSP) is a kind of complex contaminant in the ambient air. Currently,TSP is seriously overproof in China,while severe pollution of TSP is the main factor of the ambient air of the Jilin Province in many years.TSP is harmful to the health of inhabitants living and it has caused tremendous damage on the national economy.The development of Source Apportionment technology provides a new idea on environmental management, the research results play an important role of air pollution control policies.Therefore, enhanceing the sources analysis of TSP of pollutionand pollution control is particularly important to achieving the fundamental improvement of air quality of Jilin Province.
Source apportionment studies of TSP adopt mostly CMB and FA .To determining the source contributing to the receptor concentration and the contribution rate of quantitative sources ,CMB uses the measured physical and chemical properties of TSP of the polluted sources and receptors.CMB bases on mass conservation, it has a clear principle, can be easily accepted and is the used widely methods. The main objective of FA is concentrating information, reducing the dimension of indicators, simplifing the structure of indicators, avoiding multicollinearity problems between variables under the loss of principal information as little as possible.The analysis of indicator system is more simple and effective.On the study of TSP source apportionment, the public factor often represents the source of TSP, the number of the distinguishing factors of FA is the number of pollution sources, according to the relative size of the factor loading of FA in combination with the understanding of the sources pollution nature, the types of pollution sources represented by public factors is qualitatively descripted.
PPR has a prominent feature on the study of source apportionment. Without any request to the structure and characteristics of the data,linear or nonlinear, normal or non-normal data can be effectively dealt with by PPR.The source apportionment of TSP on PPR may not limited by the conditions,has the good objectivity and be more practical.And PPR takes full advantage of the spectrum distribution of the source and the data information of monitoring and analysis of atmospheric samples, the model has good stability and anti-jamming, high precision, strong reliability.
Visual Basic is a visual, object-oriented ,using event-driven ,high-level structured programming language,it can be used to developing all kinds applications under the Windows environments. It is easy to learning, efficient, and powerful.To make it better and more powerful, Visual Basic 6.0 increases many new features on the base of the original Visual Basic version.The source apportionment system software can use VB6.0 as a program development language, combining with Access database .The software users can easily query, modify and input the required monitoring data.
On the bases of the calculation of source apportionment.for four cities by the CMB model, This article uses FA and PPR methods, the builds up the corresponding model, compiles the calculation program, analyzes and calculates the data of source and receptor samples in the heating period and non-heating period,ultimately the contribution rate or contribution value of every source.Given full consideration to the characteristics of four typical cities, combined with the results of CMB analysis methods, the following results of comparative analysis of TSP source apportionment is available.
The results of TSP source apportionment in the heating period is that raised dust is emission sources of larger TSP contributions of Jilin, Siping andTonghua,soil dust is the major source of air pollution of Baicheng by the CMB model,while road dust and soil dust is emission sources of larger TSP contributions of Jilin and Tonghua, road dust and raised dust is the main source of air pollution of Siping and Baicheng by the PPR model.
The results of TSP source apportionment in the non-heating period is that the emission sources of larger TSP contribution of Jilin, Baicheng, Tonghua and Siping as follows: construction dust, soil dust wind, road dust and dust by the CMB model,while the proportions of the remaining various types dust sources are not low, in addition to coal dust, the distribution of every points is different by the PPR model.
The results of TSP source apportionment in the whole year is that, by the CMB model, the sources of 5 types of contribution on TSP of Jilin as follows: soil dust,raised dust, construction dust, road dust and coal dust, the results are consistent with the analytical results of the city's heating period and non-heating period, soil dust and coal dust is the sources of 5 types of large contribution on TSP of Baicheng, and others have smaller impact on air quality, road dust and coal dust is the main sources in Tonghua, raised dust is the main source in Siping, by the FA model,soil dust is the main source in Jilin, coal dust is the main source in Baicheng, construction dust and coal dust is the main sources in Tonghua, road dust is the main source in Siping.
CMB model is more effective for needing the contribution value and contribution rate. Under the known receptor and unknown sources,according to the correlation of the receptor data Combination with the knowledge of the characteristics element of the source e, FA model infer the source, the analysis on the source data is more accurate.FA requires a large amount of receptor data.the more data, more accurate results. Undre the case of various types of the known sources, PPR model compares the ratio of the effort of the same types of receptors on the sources,the is simiply, software is easily operated,understanding and using.This is of great help for qualitative analysis of source apportionment study. The models of three methods is different,the type of results is different, the results also vary. In practice work, selecting the appropriate methods and models On the analysis of source and receptor, respectively, that is combined with many ways could make the results more accurate.
引文
[1]齐文启,孙宗光,边归国.环境监测新技术[M].北京:化学工业出版社,2003.9
[2] Erome. O. Nriagu. Vanadium in the environment[M].Part1:Chemistry and Biochemistry. 1998,AWiley Interscience Publication, Canada.
[3] Murray M. Finkelsteina, Michael Jerrettb. Study of the relationships between Parkinson’s disease and markers of traffic-derived and environmental manganese air pollution in two Canadian cities[J]. Environmental Research.2007:1~13
[4] Shui-Ping Wu, Shu Tao, Zhi-Huan Zhang, Tian Lan, Qian Zuo. Characterization of TSP-bound n-alkanes and polycyclic aromatic hydrocarbons at rural and urban sites of Tianjin, China[J]. Environmental Pollution.2004(147):203~210
[5] D. W. Dockery, C. A. PopeⅢ., Annu. Rev. Public Health[J], 1994, 15:107~132.
[6] F. Sezer Tural?og?lu, Alper Nuhog?lu, Hanefi Bayraktar. Impacts of some meteorological parameters on SO2 and TSP concentrations in Erzurum, Turkey. Chemosphere[J].2005(59):1633~1642
[7] Guor-Cheng Fang, Yuh-Shen Wu, Jyh-Cherng Chen, Jui-Yeh Rau, Shih-Han Huang, Chi-Kwong Lin. Concentrations of ambient air particulates (TSP, PM2.5 and PM2.5–10) and ionic species at offshore areas near Taiwan Strait[J]. Journal of Hazardous Materials. 2006 (B132):269~276
[8] Ying Wang, Guoshun Zhuang, Aohan Tang, Wenjie Zhang, Yele Sun, Zifa Wang, Zhisheng An. The evolution of chemical components of aerosols at five monitoring sites of China during dust storms[J]. Atmospheric Environment.2007(41)1091~1106
[9] Ying Wang, Guoshun Zhuang, Xingying Zhang, Kan Huang, Chang Xu, Aohan Tang, Jianmin Chen, Zhisheng An. The ion chemistry, seasonal cycle, and sources of PM2.5 and TSP aerosol in Shanghai[J]. Atmospheric Environment. 2005(40):2935~2952
[10]潘贵仁,李春燕,邢海斌.唐山市大气颗粒物源解析研究[C] .昆明.中国环境科学学会大气环境分会.2000,129~135
[11]李先国,范莹,冯丽娟。化学质量平衡受体模型及其在大气颗粒物源解析中的应用[J].中国海洋大学学报,2006,36(2):225~228
[12]戴树桂,朱坦,白志鹏.受体模型在大气颗粒物源解析中的应用和进展[J].中国环境科学,1995,15(4):252~256
[13] Gorden G E.Receptor models[J].Environ Sci Technol,1980,14(7):792~800
[14]余建英,何旭宏.数据统计分析与SPSS应用[M].北京:人民邮电出版社,2003,4
[15] Blifford I.H,Mceker G.O..Afactor analysis model of large scale pollution[J],Atmospheric Environment,1967,1(1)
[16]张蓓,叶新,井鹏.城市大气颗粒物源解析技术的研究进展[J].能源与环境.2008,3
[17]杨钰,金永民,孙晓怡,陈娟.用因子分析法解析抚顺市大气颗粒物来源[J].辽宁城乡环境科技.2005,4
[18]唐孝炎,李金龙,栗欣等.大气环境化学[M].第1版.北京:高等教育出版社,1990
[19]孙文爽,陈兰祥.多元统计分析[M].第1版.北京:高等教育出版,1994
[20]张尧庭,方开泰.多元统计分析引论[M].第1版.北京:科学出版社,1982
[21] Kruscal J B.Toward a practical method which helps uncover the structure of a set ofmultivariate observations by finding the linear transformation which optimizes a new index of condensation[M].New York: In Statistcal Computer Academic,1969.
[22] Switzer P.Numerical Classification.In Geostatistics[M]. New York:Plenum Press,1970.
[23] Friedman J H,Tukey J W.A projection pursuit algorithm for exploratory data analysis[J].IEEE Trans On Computer,1974,23(9):881~890
[24] Glover David M,Hopke Philip K.Explotation of multivariate atmospheric particulate compositional data by projection pursuit[M] .Atoms Environ,Tarrytown:Pergamon Press Inc,1994,28(8):1411~1424
[25]付强,赵小勇.投影寻踪模型原理及其应用.北京:科学出版社,2006
[26] Huber P J.Projection pursuit(with discussion) [J].Ann. Statist,1985,13:435~525
[27] Pawlak Z.Rough Sets[J].International Journal of Computer and Information Sciences,1982,11:341~356
[28] Sun C.,Snape CE.,Mcrae C.,et.al..Resolving coal and petroleum-derived polycyclic aromatic hydrocarbons (PAHs)in some contaminated land samples using compound-specific stable carbon isotope ratio measurements in coinjunction with molecular fingerprint[J].fuel,2003(82):2017~2023
[29]李祚泳,丁恒康.BP网络应用于大气颗粒物的源解析[J].中国环境监测.2005,4:74~83
[30] Henry R.C..Dealing with near Collinearity in Chemical Mass Balance Receptor Models[J].Amtospheric Environment,1992,26A:933~938
[31]冯银厂,吴建会,朱坦.济南市环境空气中TSP和PM10来源解析研究[J].环境科学研究,2004,17(2):1~5
[32] Kowalczyk G.S., Gorden G.E., Rhelngrover S.W. Identification of Atmospheric Particulate Sources in Washington D.C. Using Chemical Element Balances[J]. Environ. Sci. Technol, 1982,16(2):79~90.
[33] Kasahara Mikio, ChoiKum-Chan, Takahashi Kanji. Source Contribution of Atmospheric Aerosols in Japan by Chemical Mass Balance Method[J]. Atmos. Environ., 1990,24A(3):457~466
[34]段永红,陶澍卜,王学军,李本纲,朱利中,骆永明.天津表层土壤中多环芳烃的主要来源.环境科学[J],2006,27(3):524~527
[35]刘文菁,黄世鸿,刘小红.南京市总悬浮颗粒物(TSP)及地面积法来源解析[J].气象科学,2001,21(1):87~94
[36]高运川,曹健,刘美华.上海市吴淞工业区大气环境中总悬浮颗粒物的源解析[J].上海环境科学,2003,22(2):51~53
[37]朱建新,周蓉.CMB法应用及对乌鲁木齐市大气TSP来源解析[J].新疆环境保护,1996,18(1):29~38
[38]杭维琦,黄世鸿.南京市区环境空气中总悬浮颗粒物的来源解析[J].环境监测管理与技术,2000,12(4):18~21
[39]谢骅,李联盟.我国若干地区总悬浮颗粒物和沉积尘来源解析[J].气象科学,1999,19(1):26~32
[40]谢骅.陕西气溶胶总悬浮颗粒物来源解析[J].气象科学,1998,24(7):25~29
[41]张晶,陈宗良.北京市大气小颗粒物的污染源解析[J].环境科学学报,1998,18(1):62~67
[42]高春香,宋进华,梁秀婷.呼市市区大气总悬浮颗粒物来源解析[J].内蒙古气象,2001, 2:46~48
[43]陈思龙,郑有斌.重庆城区大气颗粒物污染来源解析[J].重庆环境科学,1996,18(6):24~28
[44]王灿星,易林,林建忠.杭州市区大气中PM10的污染特征及其源解析[J].仪器仪表学报,2003,24(4):535~536
[45]朱坦,白志鹏.秦皇岛市大气颗粒物来源解析研究.环境科学研究[J],1995,8(5):49~55
[46]王起超,李东侠,方凤满。长春市空气中总悬浮微粒分布规律及来源的探讨[J].地理科学,200222(3):355~359
[47]朱振明,闫怀忠.济南市市区大气颗粒物污染来源研究[J].预防医学文献信息,2002,8(6):646～654
[48]宋宇,唐孝炎,方晨,张远航,胡敏,曾立民.北京市大气细粒子的来源分析[J].环境科学,2002,23(6):11~16
[49]胥晓瑜,傅立新,马永亮,周中平.化学质量平衡模型在香港污染源解析中的应用[J].四川环境,2000,19(4):12~15
[50]杨丽萍,陈发虎.兰州市大气降尘污染物来源研究[J].环境科学学报.2007.4(4):499~502
[51]刘莉莉,何宪铭,李屏雄,周洁.昆明市大气总悬浮颗粒物污染来源识别研究[J].云南环境科学.1994,12(1):21~24
[52]张雪敏.厦门市大气可吸入颗粒物源解析的研究[J].环境科学与技术.2007.11(11):51~54
[53]杨秋梅,找过涛,张成红,倪师军,施泽明.成都城区近地表大气尘源解析研究[J].广东微量元素科学.2007(9):32~38
[54]王帅杰,朱坦.大气颗粒物源解析技术研究进展[J].环境污染治理技术与设备,2002,3(8):8~12
[55]李祚泳.用投影寻踪回归进行大气颗粒物的污染源解析[J].中国环境科学,1999,19(3):270~272
[56]李祚泳,丁晶,张欣莉.成都市大气颗粒物源解析的PPR法[J].环境科学研究,2000,13(5):38~40
[57]邹本东,徐子优,华蕾等.因子分析法解析北京市大气颗粒物PM10的来源.中国环境监测,2007,23(2):79~85
[58]李建良,蒋勇,汪光先.计算机数值方法[M].第1版.南京:东南大学出版社,2000
[59]姜微,罗晓虹,姜林等.IEA-PPR用于降水的pH预测模型构建[J].化学研究与应用,2005,17(4):545~547
[60]张显库.VB实用编程技术——从基础到开发[M].大连:大连理工大学出版社,1997
[61]新智工作室.VB 6.0中文版教程[M].北京:电子工业出版社,2000
[62]王文义,周兵.VB与Internet应用[M].西安:西安交通大学出版社,1998,9
[63]张淮野,袁怡.数据库设计与编程实例详解——使用Access、SQL与VB[M].北京:电子工业出版社,2001,2
[64]铭政,刘瑞莲.VB编程技巧280例[M].上海:上海科学普及出版社,2002.
[65]庞一鸣.VB问答式教程[M].北京:清华大学出版社,2002,4.
[2] Erome. O. Nriagu. Vanadium in the environment[M].Part1:Chemistry and Biochemistry. 1998,AWiley Interscience Publication, Canada.
[3] Murray M. Finkelsteina, Michael Jerrettb. Study of the relationships between Parkinson’s disease and markers of traffic-derived and environmental manganese air pollution in two Canadian cities[J]. Environmental Research.2007:1~13
[4] Shui-Ping Wu, Shu Tao, Zhi-Huan Zhang, Tian Lan, Qian Zuo. Characterization of TSP-bound n-alkanes and polycyclic aromatic hydrocarbons at rural and urban sites of Tianjin, China[J]. Environmental Pollution.2004(147):203~210
[5] D. W. Dockery, C. A. PopeⅢ., Annu. Rev. Public Health[J], 1994, 15:107~132.
[6] F. Sezer Tural?og?lu, Alper Nuhog?lu, Hanefi Bayraktar. Impacts of some meteorological parameters on SO2 and TSP concentrations in Erzurum, Turkey. Chemosphere[J].2005(59):1633~1642
[7] Guor-Cheng Fang, Yuh-Shen Wu, Jyh-Cherng Chen, Jui-Yeh Rau, Shih-Han Huang, Chi-Kwong Lin. Concentrations of ambient air particulates (TSP, PM2.5 and PM2.5–10) and ionic species at offshore areas near Taiwan Strait[J]. Journal of Hazardous Materials. 2006 (B132):269~276
[8] Ying Wang, Guoshun Zhuang, Aohan Tang, Wenjie Zhang, Yele Sun, Zifa Wang, Zhisheng An. The evolution of chemical components of aerosols at five monitoring sites of China during dust storms[J]. Atmospheric Environment.2007(41)1091~1106
[9] Ying Wang, Guoshun Zhuang, Xingying Zhang, Kan Huang, Chang Xu, Aohan Tang, Jianmin Chen, Zhisheng An. The ion chemistry, seasonal cycle, and sources of PM2.5 and TSP aerosol in Shanghai[J]. Atmospheric Environment. 2005(40):2935~2952
[10]潘贵仁,李春燕,邢海斌.唐山市大气颗粒物源解析研究[C] .昆明.中国环境科学学会大气环境分会.2000,129~135
[11]李先国,范莹,冯丽娟。化学质量平衡受体模型及其在大气颗粒物源解析中的应用[J].中国海洋大学学报,2006,36(2):225~228
[12]戴树桂,朱坦,白志鹏.受体模型在大气颗粒物源解析中的应用和进展[J].中国环境科学,1995,15(4):252~256
[13] Gorden G E.Receptor models[J].Environ Sci Technol,1980,14(7):792~800
[14]余建英,何旭宏.数据统计分析与SPSS应用[M].北京:人民邮电出版社,2003,4
[15] Blifford I.H,Mceker G.O..Afactor analysis model of large scale pollution[J],Atmospheric Environment,1967,1(1)
[16]张蓓,叶新,井鹏.城市大气颗粒物源解析技术的研究进展[J].能源与环境.2008,3
[17]杨钰,金永民,孙晓怡,陈娟.用因子分析法解析抚顺市大气颗粒物来源[J].辽宁城乡环境科技.2005,4
[18]唐孝炎,李金龙,栗欣等.大气环境化学[M].第1版.北京:高等教育出版社,1990
[19]孙文爽,陈兰祥.多元统计分析[M].第1版.北京:高等教育出版,1994
[20]张尧庭,方开泰.多元统计分析引论[M].第1版.北京:科学出版社,1982
[21] Kruscal J B.Toward a practical method which helps uncover the structure of a set ofmultivariate observations by finding the linear transformation which optimizes a new index of condensation[M].New York: In Statistcal Computer Academic,1969.
[22] Switzer P.Numerical Classification.In Geostatistics[M]. New York:Plenum Press,1970.
[23] Friedman J H,Tukey J W.A projection pursuit algorithm for exploratory data analysis[J].IEEE Trans On Computer,1974,23(9):881~890
[24] Glover David M,Hopke Philip K.Explotation of multivariate atmospheric particulate compositional data by projection pursuit[M] .Atoms Environ,Tarrytown:Pergamon Press Inc,1994,28(8):1411~1424
[25]付强,赵小勇.投影寻踪模型原理及其应用.北京:科学出版社,2006
[26] Huber P J.Projection pursuit(with discussion) [J].Ann. Statist,1985,13:435~525
[27] Pawlak Z.Rough Sets[J].International Journal of Computer and Information Sciences,1982,11:341~356
[28] Sun C.,Snape CE.,Mcrae C.,et.al..Resolving coal and petroleum-derived polycyclic aromatic hydrocarbons (PAHs)in some contaminated land samples using compound-specific stable carbon isotope ratio measurements in coinjunction with molecular fingerprint[J].fuel,2003(82):2017~2023
[29]李祚泳,丁恒康.BP网络应用于大气颗粒物的源解析[J].中国环境监测.2005,4:74~83
[30] Henry R.C..Dealing with near Collinearity in Chemical Mass Balance Receptor Models[J].Amtospheric Environment,1992,26A:933~938
[31]冯银厂,吴建会,朱坦.济南市环境空气中TSP和PM10来源解析研究[J].环境科学研究,2004,17(2):1~5
[32] Kowalczyk G.S., Gorden G.E., Rhelngrover S.W. Identification of Atmospheric Particulate Sources in Washington D.C. Using Chemical Element Balances[J]. Environ. Sci. Technol, 1982,16(2):79~90.
[33] Kasahara Mikio, ChoiKum-Chan, Takahashi Kanji. Source Contribution of Atmospheric Aerosols in Japan by Chemical Mass Balance Method[J]. Atmos. Environ., 1990,24A(3):457~466
[34]段永红,陶澍卜,王学军,李本纲,朱利中,骆永明.天津表层土壤中多环芳烃的主要来源.环境科学[J],2006,27(3):524~527
[35]刘文菁,黄世鸿,刘小红.南京市总悬浮颗粒物(TSP)及地面积法来源解析[J].气象科学,2001,21(1):87~94
[36]高运川,曹健,刘美华.上海市吴淞工业区大气环境中总悬浮颗粒物的源解析[J].上海环境科学,2003,22(2):51~53
[37]朱建新,周蓉.CMB法应用及对乌鲁木齐市大气TSP来源解析[J].新疆环境保护,1996,18(1):29~38
[38]杭维琦,黄世鸿.南京市区环境空气中总悬浮颗粒物的来源解析[J].环境监测管理与技术,2000,12(4):18~21
[39]谢骅,李联盟.我国若干地区总悬浮颗粒物和沉积尘来源解析[J].气象科学,1999,19(1):26~32
[40]谢骅.陕西气溶胶总悬浮颗粒物来源解析[J].气象科学,1998,24(7):25~29
[41]张晶,陈宗良.北京市大气小颗粒物的污染源解析[J].环境科学学报,1998,18(1):62~67
[42]高春香,宋进华,梁秀婷.呼市市区大气总悬浮颗粒物来源解析[J].内蒙古气象,2001, 2:46~48
[43]陈思龙,郑有斌.重庆城区大气颗粒物污染来源解析[J].重庆环境科学,1996,18(6):24~28
[44]王灿星,易林,林建忠.杭州市区大气中PM10的污染特征及其源解析[J].仪器仪表学报,2003,24(4):535~536
[45]朱坦,白志鹏.秦皇岛市大气颗粒物来源解析研究.环境科学研究[J],1995,8(5):49~55
[46]王起超,李东侠,方凤满。长春市空气中总悬浮微粒分布规律及来源的探讨[J].地理科学,200222(3):355~359
[47]朱振明,闫怀忠.济南市市区大气颗粒物污染来源研究[J].预防医学文献信息,2002,8(6):646～654
[48]宋宇,唐孝炎,方晨,张远航,胡敏,曾立民.北京市大气细粒子的来源分析[J].环境科学,2002,23(6):11~16
[49]胥晓瑜,傅立新,马永亮,周中平.化学质量平衡模型在香港污染源解析中的应用[J].四川环境,2000,19(4):12~15
[50]杨丽萍,陈发虎.兰州市大气降尘污染物来源研究[J].环境科学学报.2007.4(4):499~502
[51]刘莉莉,何宪铭,李屏雄,周洁.昆明市大气总悬浮颗粒物污染来源识别研究[J].云南环境科学.1994,12(1):21~24
[52]张雪敏.厦门市大气可吸入颗粒物源解析的研究[J].环境科学与技术.2007.11(11):51~54
[53]杨秋梅,找过涛,张成红,倪师军,施泽明.成都城区近地表大气尘源解析研究[J].广东微量元素科学.2007(9):32~38
[54]王帅杰,朱坦.大气颗粒物源解析技术研究进展[J].环境污染治理技术与设备,2002,3(8):8~12
[55]李祚泳.用投影寻踪回归进行大气颗粒物的污染源解析[J].中国环境科学,1999,19(3):270~272
[56]李祚泳,丁晶,张欣莉.成都市大气颗粒物源解析的PPR法[J].环境科学研究,2000,13(5):38~40
[57]邹本东,徐子优,华蕾等.因子分析法解析北京市大气颗粒物PM10的来源.中国环境监测,2007,23(2):79~85
[58]李建良,蒋勇,汪光先.计算机数值方法[M].第1版.南京:东南大学出版社,2000
[59]姜微,罗晓虹,姜林等.IEA-PPR用于降水的pH预测模型构建[J].化学研究与应用,2005,17(4):545~547
[60]张显库.VB实用编程技术——从基础到开发[M].大连:大连理工大学出版社,1997
[61]新智工作室.VB 6.0中文版教程[M].北京:电子工业出版社,2000
[62]王文义,周兵.VB与Internet应用[M].西安:西安交通大学出版社,1998,9
[63]张淮野,袁怡.数据库设计与编程实例详解——使用Access、SQL与VB[M].北京:电子工业出版社,2001,2
[64]铭政,刘瑞莲.VB编程技巧280例[M].上海:上海科学普及出版社,2002.
[65]庞一鸣.VB问答式教程[M].北京:清华大学出版社,2002,4.