DSpace
 

Osaka Prefecture University Repository >
D.Former Osaka Prefecture University:(1955-2004) >
0200.College of Agriculture/Graduate School of Agriculture and Biological Sciences >
Scientific report of the Graduate School of Agri. and Bio. Sci. >
Vol.57 >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10466/3077

Title: Multiplicative Methods for Assessment of Variation in Rainwater Chemical Content with the Time of Rain Events
Authors: SUWARDI, Annas
KANAI, Takenori
KOYAMA, Shuhei
Keywords: AMMI
ANOVA
chemical content
PCI
rainwater
SOM
Issue Date: 31-Mar-2005
Publisher: Graduate School of Agriculture and Biological Sciences, Osaka Prefecture University
Citation: 大阪府立大学大学院農学生命科学研究科学術報告. 2005, 57, p.1-8
Abstract: Periodic measurement of the concentration of chemicals dissolved in rainwater to differing time of rain events was important in detecting rainwater contamination. Multiplicative methods such as the AMMI (additive main effects and multiplicative interaction) analysis and the SOM (self-organizing map) algorithm were useful for studying patterns of the variation of the rainwater chemical content relative to the month in which the time of rain occurred. The AMMI incorporated both additive main and multiplicative interaction effects into an ANOVA (analysis of variance) of AMMI in testing the significances of the analyzed factors. The multiplicative normal distribution in term of the confidence interval 95% on the AMMI was employed to assess the stability of the chemical content in rainwater during the rainy season. Overlapping of the two first PCI (principal component of interaction) into a biplot was usually used for interpreting the components of interaction. However, in the present study the AMMI biplot not eligible to be used for displaying the interaction patterns since the second PCI was not significant according to F-test (Fisher's test) at the level of α=0.05. We then used the SOM to overcome this deficiency of AMMI. This method therefore offered benefits in inspecting the possible correlations of the multi-parameter of the vector components in the input data. A plane map of SOM has effectively revealed the interaction patterns as well as, visualizing the relations between the concentrations of chemicals in rainwater to the times of rain events.
URI: http://hdl.handle.net/10466/3077
Appears in Collections:Vol.57

Files in This Item:

File Description SizeFormat
KJ00004360353.pdf1.04 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.