Author/s:

Islam, A.
Bodrud-Doza, M.
Rahman, M.
Samiran, S.
Shen, S.

Publisher:

Springer Link

Year of Publication:

2017

Descriptive statistics, correlation, regression, and geostatistical modeling are applied to assess the trace elements of groundwater and their spatial distribution at the Rangpur district of Bangladesh. A total number of 47 water samples have been collected from wells at depth ranging from 10 to 53 m. The descriptive statistics results show that the mean concentrations of iron (Fe), manganese (Mn), and barium (Ba) have exceeded the permissible limits and those concentrations are alarming to human health and their surrounding environments. Furthermore, Mn, Zn, Al, and Ba concentrations reveal the highly positive skewed and are considered to be extreme. The statistical results demonstrate that groundwater trace element quality is mainly related to natural/geogenic sources followed by anthropogenic sources in the study area even though they show significant correlations among them. The multiple regression models are developed for prediction of each trace element of groundwater samples. The spatial analysis of groundwater trace elements is performed by geostatistical modeling. The cross validation results reveal that kriging models are produced to show the most accurate spatial distribution maps for all trace elements except Ba concentration. The semivariogram models have demonstrated that most of the elements have shown moderate to strong spatial dependence suggesting less agronomic/residential area influences. The findings of the multiple regression model and the correlation matrix are also consistent with the spatial analysis results. It is anticipated that outcomes of this study will provide insights for decision makers taking adaptive measures for groundwater trace element monitoring in Rangpur district, Bangladesh.

comments powered by Disqus