A Variance Decomposition Approach for Risk Assessment of Groundwater Quality

Abstract

This research focuses on the assessment of fluoride doses in groundwater adopting the mathematical model employed by the USEPA. A total of 456 groundwater samples were tested to assess the spatial distribution of fluoride contamination in the study areas. Three age groups (children, teens and adults) were selected for two-way pathway exposure (potential dose and dermal dose) assessment. For uncertainty and sensitivity of inputs variables, a new emerging Sobol sensitivity analysis (SSA) technique was used to determine the relative importance of inputs using Monte Carlo simulation. Three types of effects, first-order effect (FOE), second-order effect (SOE) and total effect (TE) were calculated. The results showed that 96% of the samples analysed were within the standard acceptable level (1.5 mg l−1) of WHO guidelines. The spatial distribution depicts that the eastern and south-eastern parts of the study area have the higher concentrations with the few spots of elevated concentration in the middle of the north and the south-west areas. The mean value of Hazard Index for children in the study region is less than 1, whereas the 95th percentile exceeded the value of 1 for both children and teens. The FOE shows the concentration of fluoride (Cw) is highly sensitive followed by exposure frequency (EF), intake rate (IRw) and body weight (BW). The SOE scores revealed that IRw–BW are the most important input parameters for the assessment of oral health risk. For the dermal model, the highest value of Sobol score was recorded for interactions Cw–SA for adults followed by teens and children. Further, the results show that the older-age groups have more dermal risk than the younger-age groups. The research explores the feasibility of SSA technique to investigate the effects of individual input parameters for health risk model and whether it can be applied to another contaminant.