GSA Annual Meeting in Phoenix, Arizona, USA - 2019

Paper No. 251-4
Presentation Time: 8:50 AM

PARAMETRIC AND NON-PARAMETRIC ASSESSMENT OF ARSENIC VARIABILITY CONTROLS IN GROUNDWATER SYSTEMS


IJUMULANA, Julian1, MTALO, Felix W.2, BHATTACHARYA, Prosun3, IRUNDE, Regina2 and LIGATE, Fanuel J.4, (1)Department of Water Resources Engineering, College of Engineering and Technology, University of Dar es Salaam, Dar es Salaam, Tanzania, United Republic of; KTH-International Groundwater Arsenic Research Group, Dept. of Sustainable Development, Environmental Science and Engineering, KTH Royal Institute of Technology, Teknikringen 10B, Stockholm, SE-10044, Sweden, (2)Department of Water Resources Engineering, College of Engineering and Technology, University of Dar es Salaam, Dar es Salaam, Tanzania, United Republic of, (3)KTH-International Groundwater Arsenic Research Group, Department of Sustainable Development, Environmental Science and Engineering, KTH Royal Institute of Technology, Teknikringen 10B, Stockholm, SE-10044, Sweden, (4)KTH-International Groundwater Arsenic Research Group, Dept of Sustainable Development, Environmental Science and Engineering, KTH Royal Institute of Technology, Teknikringen 10B, Stockholm, SE-10044, Sweden

Arsenic contamination in drinking water systems based on groundwater resources is a global agenda. The occurrence, variability and distribution has been associated with local geology, hydrogeology, geochemistry and climate of an area. Recently, elevated concentrations of Arsenic have been reported in various places within the Lake zone in the North-west of Tanzania. In this study, multicriteria approach has been developed and used to study the effects of local lithology, topography, soils and physiochemical parameters on the occurrence and distribution of Arsenic in drinking water systems based on groundwater resources. Water samples were collected from boreholes, shallow wells and natural springs used as major sources of drinking water in the East and South-west of Lake Victoria. The samples were analyzed to determine arsenic concentrations and other physiochemical parameters such as pH, EC, TDS and salinity. The results of laboratory analysis were integrated with existing GIS datasets such as elevation, soil and geological data, and the effects of surface slope, soil type and local geology on arsenic concentrations variability were studied. Arsenic concentrations increased with an increase in elevation for the South-western samples contrally to the Eastern samples. The highest average arsenic concentrations (84.8 ±70.8 µg/l and 170±66.4 µg/l for S-W and E site respectively) were found among shallow wells located in the ferralic cambisols underlain by Neoarchaean granitic complex (Kavirondian-Nyanzian Supergroup) and eutric planosols underlain by Neoarchaean greenstone belts in the South-western and eastern site respectively. The analysis of variance (ANOVA) results indicated that arsenic concentrations in the sources did not differ significantly (F= 0.7656 , p-value=0.4673) for samples in the S-W site but differed significantly as evidenced by Kruskal-Wallis rank sum test (p-value = 0.02775). The proposed methodology may be useful in identification of safe aquifer where depth information is not available.

Key words: Arsenic variability, drinking water quality, parametric and non-parametric analysis, GIS