Sobol sensitivity approach for the appraisal of geomedicalhealth risks associated with oral intake and dermal pathways of groundwater fluoride in a semi-arid region of south India

The main objective of the present investigation is appraisal of human health hazard based on the intake and dermal contact of fluoride enriched potable groundwater used for rural water supply in a semi-arid region (Shanmuganadhi River basin) of south India. A variance decomposition based Sobol sensitivity method was used to assess the relative contribution as well as interaction of input variables for both oral and dermal models. Three different scores were evaluated: FOE (first order effect), SOE (second order effect) and TE (total effect) for different age groups of population including gender (kids, women and men). The spatio-temporal mapping indicates that about 26% of water supply wells exceeded the recommended limit (WHO) of fluoride (> 1.5 mgl(-1)) for safe intake. These wells spread over 104.03 km(2) area consisting 16 villages in the basin. To assess the human health risk related to fluoride enrichment in potable water, hazard index (HI) was calculated as per USEPA guidelines. The non-carcinogenic risk based on oral intake ranges from 0 to 1.81, from 0 to 1.59 and from 0 to 1.29 for kids, women and men respectively. Nearly 30%, 21% and 12% of well samples exceeded the upper permissible limit (HI > 1) for kids, women and men respectively. The Sobol sensitivity analysis reveals that, C-w, (concentration of F- in water) and IR (intake rate) combination plays a vital role in the HQ(oral) model for the appraisal of health hazard in kids. However, these two parameters have negligible effect on health hazard for adult population (men and women). Therefore, lower age group people especially kids have significant ill effect due to the consumption of fluoride enriched potable water. The model output suggests that body weight (BW) has least effect on health hazard in the lower age group of population. Interestingly 'exposure frequency (EF)' and 'body weight (BW)' have not much effect on health hazards related to dermal contact, and 'skin surface area (SA)' has minimal effect for men (TE = 0.92) and women (TE = 1.26). However, 'SA' has vital effect for kids (TE = 17.39). Because of this, older age group people have more dermal risk than the younger people. Therefore, the HQ dermal results indicate that 49%, 64% and 67% of samples possess non-carcinogenic risk to the kids, women and men respectively.