Basic Study on Estimation of Dioxin Souces using Statistical Methods

Bibliographic Information

Other Title
  • 統計学的手法によるダイオキシン類の発生源推定のための基礎検討
  • トウケイガクテキ シュホウ ニ ヨル ダイオキシンルイ ノ ハッセイ ゲン スイテイ ノ タメ ノ キソ ケントウ

Search this article

Abstract

Many applied statistical methods have been used for identification and estimation of dioxin sources. In most case, however, it has merely discussed about basic parts which are kind and the number of data, parameters and/or calculation methods in spite of their importance.<BR>In this study, principal component analysis (PCA) was used for categorization of sample because it had been expected that isomer profiles of PCDD/Fs and/or planer PCBs account characteristics of samples. Input data were collected from published sources.<BR>Pretreatment of data was required before carrying out of PCA. A statistical standardization which is calculated by a formula, i.e. (value-mean) /Std. deviation, was determined as the best method among trial. The number of variable, which means a homologue or a compound, was sensitively affected to the result of PCA. From the results of examinations, the condition which uses 2, 3, 7, 8-position chlorine substituted compounds of PCDD/Fs and 1, 2, 3, 4, 6, 7, 9-HpCDD for input data was most suitable for source identification of dioxins.<BR>The most serious problem in the present study was the lack of data on dioxin isomers. In most case, available data contained only a few dioxin homologues, e.g. only 2, 3, 7, 8-substituted ones and congeners. Therefore sufficient comparative study was not conducted on the number of variable. It would be expected to get better result if more variables were used to the analysis.<BR>To collect and provide a lot of data including many isomers and homologues of dioxin in environment and sources must be most important work in these studies.

Journal

Citations (4)*help

See more

References(28)*help

See more

Details 詳細情報について

Report a problem

Back to top