Source apportionment of organotin compound based on spatial variation using multiple linear regression (MLR) and hierarchical agglomerative cluster analysis (HACA) in Southern Coast of Peninsular Malaysia

High Organotin (OT) concentration including monobutyltin (MBT), dibutyltin(DBT), tributyltin (TBT), monophenyltin (MPT), diphenyltin (DPT), and triphenyltin (TPT) in Thais gradata has been reported in Southern Coast of Peninsular Malaysia. The OT concentration secondary data obtained from eight locations along the southern shores of Peninsular Malaysia shows high correlations between the OT compounds, especially among butyltins (BTs). Multiple Linear Regression (MLR) model was used for recovering the missing of OT compounds. Cross validation MLR by Leave-One-Out (LOO) was applied to eliminate independent variables that exhibit the lowest contribution in variance. Independent variables accounted for approximately 92.57% of the variance in OT concentrations are 14.16% (MBT), 26.15% (DBT), 41.22% (TBT), and 11.04% (TPT). TBT has slower biological degradation compared to MBT and it is dependent on environmental conditions and also physical state of TBT. Two clusters were formed based on Hierarchical Agglomerative Cluster Analysis (HACA). Four significant Organotin (sources) were identified which are BTs and TPT that are responsible for the variations of OT concentration dataset in the Southern Coast of Peninsular Malaysia. These results suggested that by using statistical analysis, it is capable to estimate the significant sources of OT. Furthermore, analysis of OT concentration should mainly focus on BTs in order to save cost and time for further research.

PDF ID 31453.pdf Restricted to Repository staff only Download (438kB)

Actions (login required)

View Item