$40,000 to collect 100 groundwater samples, and the mean cost per sample to
analyze metals would be around $260 (USEPA 1997). A detailed description of
possible challenges in the collection and analysis of water-quality data can be found
here (Crocker and Bartram 2014). Additionally, the Environmental Protection
Agency provides comprehensive guidance on the possible costs associated with
collection, monitoring, and testing of water samples. Costs may vary according to
the instruments and laboratory facilities used, but even so, they may continue to be
an obstacle to conducting such research in low-income regions where groundwater
contamination is a severe problem.
This chapter focuses only on arsenic, fluoride, and nitrate because of their serious
human health impacts and global presence in groundwater. Considering the obstacles mentioned in the previous section, the chapter is structured in three sections, the
first of which highlights the prevalence of these groundwater contaminants and their
health impacts. The second section covers the application of various AI techniques to
the prediction of arsenic, fluoride, and nitrate contamination of groundwater. The
final section entails a discussion of the advantages of certain techniques and concludes by recommending future directions of exploration and practical application.
4.2 Groundwater Contamination
4.2.1 Arsenic
Arsenic, a metalloid and the world’s most 53rd-most abundant element, occupies
just 0.0001% of the earth’s crust and occurs in unpolluted groundwater in concentrations below 10 μg/L (Braman 1975). Out of arsenic’s five valence states, its
inorganic form, arsenite, is the most toxic; in groundwater, however, the organic
form of arsenic, arsenate, predominates (Braman 1975; Oremland and Stolz 2003).
Because of its adverse health effects on human beings, the Environmental Protection
Agency has categorized arsenic as a “Group A” element. Currently, more than
100 countries are experiencing geogenic arsenic contamination of groundwater,
and more than 296 million individuals’ lives are at risk (Fig. 4.1) (Chakraborti
et al. 2017).
Approximately 66% of European, 59% of Asian, 37% of American, 24% of
African, and 1% of Oceanian countries are impacted by arsenic contamination of
groundwater (Singh 2017; Murcott 2012). It has been more than two centuries since
arsenic contamination was first testified to in Germany in 1885 and more than one
since arsenic-induced health effects were discovered in 1917 in the province of
Cordoba, Argentina (Chakraborti et al. 2017; Murcott 2012; Nordstrom 2002). Over
time, advancements in arsenic research confirmed that prolonged exposure to arsenic
in water, food, soil, or air could lead to a variety of health problems, including both
carcinogenic and non-carcinogenic diseases (Chakraborti et al. 2017; Mazumder
2008; Chakraborti et al. 2003; Bhattacharya et al. 2010). These health issues are
summarized in Table 4.1.
4 Application of Artificial Intelligence in Predicting Groundwater Contaminants
73
analyze metals would be around $260 (USEPA 1997). A detailed description of
possible challenges in the collection and analysis of water-quality data can be found
here (Crocker and Bartram 2014). Additionally, the Environmental Protection
Agency provides comprehensive guidance on the possible costs associated with
collection, monitoring, and testing of water samples. Costs may vary according to
the instruments and laboratory facilities used, but even so, they may continue to be
an obstacle to conducting such research in low-income regions where groundwater
contamination is a severe problem.
This chapter focuses only on arsenic, fluoride, and nitrate because of their serious
human health impacts and global presence in groundwater. Considering the obstacles mentioned in the previous section, the chapter is structured in three sections, the
first of which highlights the prevalence of these groundwater contaminants and their
health impacts. The second section covers the application of various AI techniques to
the prediction of arsenic, fluoride, and nitrate contamination of groundwater. The
final section entails a discussion of the advantages of certain techniques and concludes by recommending future directions of exploration and practical application.
4.2 Groundwater Contamination
4.2.1 Arsenic
Arsenic, a metalloid and the world’s most 53rd-most abundant element, occupies
just 0.0001% of the earth’s crust and occurs in unpolluted groundwater in concentrations below 10 μg/L (Braman 1975). Out of arsenic’s five valence states, its
inorganic form, arsenite, is the most toxic; in groundwater, however, the organic
form of arsenic, arsenate, predominates (Braman 1975; Oremland and Stolz 2003).
Because of its adverse health effects on human beings, the Environmental Protection
Agency has categorized arsenic as a “Group A” element. Currently, more than
100 countries are experiencing geogenic arsenic contamination of groundwater,
and more than 296 million individuals’ lives are at risk (Fig. 4.1) (Chakraborti
et al. 2017).
Approximately 66% of European, 59% of Asian, 37% of American, 24% of
African, and 1% of Oceanian countries are impacted by arsenic contamination of
groundwater (Singh 2017; Murcott 2012). It has been more than two centuries since
arsenic contamination was first testified to in Germany in 1885 and more than one
since arsenic-induced health effects were discovered in 1917 in the province of
Cordoba, Argentina (Chakraborti et al. 2017; Murcott 2012; Nordstrom 2002). Over
time, advancements in arsenic research confirmed that prolonged exposure to arsenic
in water, food, soil, or air could lead to a variety of health problems, including both
carcinogenic and non-carcinogenic diseases (Chakraborti et al. 2017; Mazumder
2008; Chakraborti et al. 2003; Bhattacharya et al. 2010). These health issues are
summarized in Table 4.1.
4 Application of Artificial Intelligence in Predicting Groundwater Contaminants
73