Nowadays, for many different reasons, groundwater is contaminated, creating a serious health hazard in both man and animals ( Kazi et al., 2009 ). Major contaminants comprise nitrates ( Mahvi et al., 2005 ), heavy metals ( Taghinia Hejabi et al., 2011 ), and organic compounds ( Manecki & Gałuszka, 2012 ). Table 1 summarizes different groundwater composition ranges found across countries. As indicated in the literature, there is a large variation between groundwater in terms of values and concentrations of constituents.

Generally, groundwater has potential hydrogen (pH) values averaging between 6.6 and 8.3, indicating a neutral to slightly alkaline nature of groundwater. According to WHO, the pH of water does not have a direct effect on human health; however, it is related to other chemical constituents of water in general ( Mostafa et al., 2017 ; Saalidong et al., 2022 ).

Table 1. Variation of selected parameters in different groundwaters.

Electrical conductivity (EC) of groundwater is an indicator of the presence of metal ions and inorganic elements ( Egbueri , 2019 ; Lin & Chang, 2000 ). Gueddari et al. (2022) and Qureshi et al. (2011) have indicated that the electrical conductivity of groundwater is related to the level of mineralization and provides information on salinity levels. Additionally, a relationship between electrical conductivity and total dissolved solids has been evidenced by some researchers ( Thirumalini & Joseph, 2009 ), which are both indicators of salinity levels. Various extremes of mean electrical conductivity (EC) and total dissolved solids (TDS) values were found in the groundwater selected for this study, especially in Italy and Qatar with mean values of 3979 μS/cm and 5038.1 mg/l, respectively. According to Ismail et al. (2020) ; Kumari & Rai (2020) , the reason for the high salinity of groundwater in these areas of semi-arid climatic conditions is due to the evapotranspiration exceeding the precipitation, and the lack of drainage. Furthermore, it was found that agricultural activities involving the use of fertilizers and irrigation with highly saline water can result in a concentration of salts in the soil due to high evaporation, which can lead to the leaching of salts and nutrients into the aquifer ( Ahmad et al., 2020 ). Alfarrah & Walraevens (2018) and Barlow & Reichard (2010) studies have indicated that extensive groundwater extraction can result in increased EC due to saltwater intrusion in coastal areas. Temperature is an important parameter which controls micro-organism activity and chemical equations ( Schürch et al., 2018 ). In this review, the average groundwater temperature is between 14.2 and 29.1˚C. An increase in water temperature can, for example, lead to a decrease in the concentration of dissolved oxygen, favoring the presence of pathogenic bacteria and thus causing a decrease in the microorganisms that indicate the good quality of the resource ( Garnier, 2012 ). The temperature of groundwater is related to climate and hydrogeology. Groundwater loses or gains heat from the ground surface. According to Taylor and Stefan (2009) , the average annual ground surface temperature is controlled by climate and land use (surface cover). If climate and land use do not vary over time, the average annual groundwater temperature and the average annual ground surface temperature are theoretically identical.

The average chemical oxygen demand (COD) in groundwater for this study is from 1.1 to 11 mg/l. The presence of COD in water indicates organic matter content, which is related to a richer pathogenic microflora ( Dippong et al., 2019 ). These authors add that COD is an indirect chemical index of water pollution by inorganic oxidizable substances such as S²⁻, ferrous salts, and ${\text{NO}}_2^{-}$. Previous studies by Ahmed et al. (2004) and Halim et al. (2010) found that the presence of organic matter in sediments can play an important role in the creation of anaerobic environments in groundwater.

Turbidity in groundwater results from an extremely fine colloidal suspension in the form of finely divided matter (organic and inorganic), microorganisms such as plankton, and clay ( Majeed & Nashaat, 2022 ). The average groundwater turbidity in selected groundwater ranged from 0.4 to 7.8 NTU. This indicates that dissolved solids in the groundwater are low and that microorganisms have difficulty in multiplying or resisting disinfection ( Egbueri , 2019 ).

The total hardness (TH) varies widely between countries. Average values are between 53.3 and 2120.2 mg/l. Physically, hardness can depict the resistance of water to lathering soap ( Todd, 2008 ). According to Saana et al. (2016) , TH chemically indicates the total concentration of Mg²⁺ and Ca²⁺ in mg/l of CaCO₃ equivalent. Hard water consumption can cause white incrustations on boilers and cooking utensils ( Kaushik et al., 2002 ) and can scale water pipes and water heaters, and can require more soap for laundry ( Akram & Rehman, 2018 ). Hard water is also responsible for anencephaly, urolithiasis, some cardiovascular disorders, and some types of cancer ( Marghade , 2020 ).

Generally, about 95% of the ions encountered in groundwater are Mg²⁺, Ca²⁺, K⁺, Na⁺, ${\text{NO}}_3^{-}$, ${\text{HCO}}_3^{-}$, ${\text{SO}}_4^{2-}$, and Cl⁻ ( Sundaram et al., 2009 ). According to Abdel-Satar et al. (2017) , the main source of ions in groundwater is rock lithology, in contrast to anthropogenic sources.

As with total hardness (TH), the average concentration of calcium and magnesium ions in groundwater varies widely, from 7.6 to 570.2 mg/l and from 15 to 169.1 mg/l, respectively. The presence of calcium ions in groundwater can be naturally associated with the dissolution of silicate, phosphate, and sulfate minerals, and the dissolution of carbonate minerals ( Cobbina et al., 2012 ). The presence of magnesium in groundwater can be attributed to geological sources such as biotite, pyroxene, and dolomite ( Saana et al., 2016 ).

Average values of alkalinity are fairly low, with a peak average of 178.3 mg/L. Groundwater alkalinity is principally due to the presence of ions such as OH⁻, ${\text{CO}}_3^{2-}$, or ${\text{HCO}}_3^{-}$ ( Wolf-Gladrow et al., 2007 ). Among these ions, bicarbonate is the main form because it is formed in large quantities by the action of carbonates on the basic materials of the soil ( Kumari & Rai, 2020 ). Bicarbonate is naturally found in water, mainly through the CO₂ dissolution of carbonate-containing minerals or by the combination of rainwater and CO₂ ( Ismail et al., 2018 ).

Average nitrate ( ${\text{NO}}_3^{-}$) concentrations in groundwater range from 0.4 to 95 mg/l, while the average phosphate content is low and varies from 0.1 to 0.8 mg/l. Nitrates and phosphates are the major plant nutrients derived from fertilizers ( Eg bueri, 2019 ). ${\text{NO}}_3^{-}$ provides favorable growing conditions for algae and other aquatic plants, which use and deplete oxygen in the water, rendering it tasteless ( Dippong et al., 2019 ). High nitrate contents in drinking water result in substantial health risks for pregnant women and infants and contribute to stomach cancer incidence ( Adimalla , 2019 ; Ward et al., 2018 ). Further, the presence of high ${\text{NO}}_3^{-}$ in water may influence organisms by blocking hemoglobin and producing methemoglobin ( Dippong et al., 2019 ). Under natural conditions, the concentration of nitrates does not exceed 10 mg/l in water as reported by Cushing et al. (1973) . ${\text{NO}}_3^{-}$ pollution is principally driven by the decomposition of organic matter ( Bahir et al., 2020 ), animal and human wastes, urban domestic sewage, intense fertilization, septic tank effluents ( Zhang et al., 2014 ). ${\text{NO}}_3^{-}$ leakage to groundwater over time is related to the geological and hydrogeological structure of an area ( Racoviteanu , 2016 ). ${\text{NO}}_3^{-}$ is highly mobile once it reaches groundwater ( Eary et al., 1989 ), hence the load is very rapidly transferred.

Chloride (Cl⁻) mean concentrations in groundwater vary widely from 3.8 to 6289.5 mg/L. Chlorides are common constituents of natural water ( Ahmad et al., 2020 ). According to Adimalla and Venkatayogi (2018) , the salty taste and laxative effects of drinking water may be due to high concentrations of Cl⁻. Natural sources of Cl⁻ contained in groundwater include saline seeps, water-rock interactions, and rainwater, while anthropogenic sources include gypsum-based fertilizers ( Vengosh et al., 2002 ), municipal landfill leachate, wastewater pollutants, industrial facility effluents, and road salt ( Srinivasamoorthy et al., 2014 ).

Mean sulfate ( ${\text{SO}}_4^{2-}$) values in groundwater show a wide range, from 1.4 to 4977.2 mg/l. High concentrations of ${\text{SO}}_4^{2-}$ in groundwater result from natural sources such as the dissolution of anhydrite and gypsum rocks, and from anthropogenic sources such as fertilizers from agricultural activities ( Czerewko et al., 2003 ). Alternatively, according to Egbueri (2019) , low sulfate levels indicate that the investigation area is not an industrial site where higher concentrations are expected due to industrial processes and emissions. According to Ghalib (2017) , ${\text{SO}}_4^{2-}$ in water induces metal corrosion in the distribution system with low-alkalinity water.

Sodium (Na⁺) and potassium (K⁺) contents have mean values ranging from 4.9 to 1466 mg/l and 0.5 to 90.2 mg/l, respectively. Generally, Na⁺ and K⁺ ions in the GW are associated with each other, but the level of Na⁺ is higher than that of K⁺ ( Al Suhaimi et al., 2019 ). Na⁺ content may originate from the dissolution of sodium-bearing minerals, including sodium plagioclases such as albite ( Egbueri , 2019 ). Kumar et al. (2009) and Meybeck (1987) reported that a Na/Cl ratio greater than 1 indicates that sodium ions are derived from silicate weathering. K⁺ in groundwater: it can naturally result from feldspar weathering in igneous rocks and silicate and clay mineral weathering in sedimentary rocks ( Ahmad et al., 2020 ). In addition, chemicals from fertilizers and industries are essential sources of K⁺ in groundwater ( Mallick et al., 2018 ). Trace element concentrations in the groundwater are mostly less than one mg/L ( Ahmad et al., 2020 ; Ahmed et al., 2019 ; Dippong et al., 2019 ; Egbueri, 2019 ; Hernández-Mena et al., 2021 ; Loh et al., 2020 ; Talib et al., 2019 ).

Trace elements represent insoluble materials that are often suitable for metal precipitation in an alkaline environment ( El gawad et al., 2008 ). Common metals present in the groundwater include iron, manganese, zinc, arsenic, and fluoride. According to Dippong et al. (2019) , a relatively high amount of iron (Fe) in drinking water has no harmful effects on the human body, but can result in a metallic taste and an opalescent yellow color. High manganese (Mn) content forms brown to black deposits and gives a specific muddy odor and metallic taste, which is easily confused with hydrogen sulfide ( Dippong et al., 2019 ). The presence of zinc (Zn) is beneficial to both humans and other ecological entities, although at higher levels it can be harmful to human biological systems ( Adeyemi & Ojekunle, 2021 ). Arsenic (As) is one of the carcinogenic elements present in groundwater, resulting from both natural processes and human activities ( Massoudinejad et al., 2020 ; Paydar et al., 2020 ). Regarding fluoride, Ali et al. (2016) highlighted fluoride-related health problems such as skeletal and dental fluorosis in humans, which in turn have serious socioeconomic implications.

Water contamination and human health are interrelated ( Khan et al., 2009 ). Consumption of contaminated water may result in serious human health hazards. In 2002, a study in Bangladesh reported that tens of millions of people were poisoned to varying degrees by arsenic in well water ( Harvey et al., 2002 ). A study by Nyanganji et al. (2011) conducted on the groundwater quality of Dass town in Nigeria found high concentrations of manganese, calcium carbonate (total water hardness), and E. coli that exceeded the safe WHO and SON safe drinking standards. Accordingly, water-borne diseases were observed at the Dass General Hospital (about 110 cholera cases, 3345 cases of diarrhea, 1522 cases of dysentery, and 1527 cases of typhoid).

In Togo, in the district of Adakpamé located in the commune of Lomé, it has been reported that water consumption from wells and boreholes is responsible for various microbial diseases, such as cholera, typhoid fever, bacillary dysentery, diarrhea and gastroenteritis, hepatitis A and E, and amoebic dysentery ( Sokegbe et al., 2018 ).

Groundwater contaminants have different sources ( Figure 1 ). Some are of human origin; for example, saltwater intrusion along coastal areas ( Alfarrah & Walraevens, 2018 ; Mtoni et al., 2013 ; Xu et al., 2021 ) and nitrate pollution caused by agricultural activities and improper construction of septic tanks ( Almasri & Kaluarachchi, 2004 ; Corniello et al., 2007 ; Pastén-Zapata et al., 2014 ), while others are of natural origin, such as the solubilization of components like fluoride (F−) ( Adimalla et al., 2019 ; Kumar et al., 2018 ; Li et al., 2019 ).

Exponentially more solid waste is generated each year as a result of rapid population growth, industrial development, and urbanization. However, rainwater percolates through solid waste in landfills and generates a toxic liquid called landfill leachate with significant amounts of organic matter, inorganic salts, heavy metals, and ammonia nitrogen ( Costa et al., 2019 ). Leachate composition principally depends on the amount of precipitation, the source of the landfill waste, and the age of the landfill.

Nutrient contamination of groundwater, especially nitrate ( ${\text{NO}}_3^{-}$), is increasing in agricultural areas owing to excessive application of inorganic and organic nitrogen fertilizers ( Shishaye , 2015 ; Xue et al., 2009 ). As a highly soluble substance, nitrate diffuses easily through the vadose zone before reaching groundwater. The nitrate, upon frequent use in agricultural landscapes, can accumulate significantly and remain in groundwater for decades ( Sasakova et al., 2018 ).

Based on their widespread agricultural use and high mobility in the environment, some pesticides, such as atrazine and its derivatives, Atrazine-desisopropyl and desethyl-atrazine, are common contaminants of groundwater ( APVMA, 2008 ; Giddings, 2005 ). Once atrazine enters an aquifer system, it persists longer than in surface water and soil due to the lack of photodegradation and generally anaerobic conditions ( Schult, 2016 ). Within an aquifer, atrazine has a 2 - 18-month-long effective half-life ( Giddings, 2005 ). Further groundwater contamination concerns include persistent pesticides such as hexazinone and bromacil, both of which have high mobility in groundwater ( Shaw et al., 2012 ). Pesticides, once dissolved in groundwater, are also able to move through the hydrologic system and potentially affect the environment ( APVMA, 2008 ).

Industrial wastewater is one of the major concerns associated with the rapid industrialization in the world today. Industrial effluents are characterized by significant amounts of organic and inorganic chemicals and their derivatives ( Bougherira et al., 2014 ). Moreover, most industries are quite small and do not have sewer lines. To mitigate the negative effects of the hazardous components in industrial wastewater, adequate treatment of the effluent is mandatory prior to its release into water bodies or onto the land. However, Bougherira et al. (2014) report that most industries lack adequate treatment plants. Therefore, highly colored and toxic chemical effluents reach the river. According to Malik et al. (2019) , discharges from nickel and other metal-plating industries are seriously impairing the land. Common heavy toxic metals in wastewater are chromium (Cr), arsenic (As), cadmium (Cd), mercury (Hg), zinc (Zn), copper (Cu), cobalt (Co), and nickel (Ni) ( Gardea-Torresdey et al., 2005 ).

Urbanization is a pervasive global trend. About 50% of the world’s population is in urban areas, with a projection of 60% by 2030, according to Burns et al. (2005) . Residents around major municipalities and in rural areas with no access to public sewage systems for latrine and kitchen discharges are expected to eliminate wastewater on-site.

A septic tank is one of the most common methods of on-site wastewater management. Septic tanks, as a means of wastewater disposal in homes, however, may have a major impact on groundwater. This is an issue in many countries. Groundwater is considered a source of local and regional groundwater contamination in Nigeria ( Adetunji & Odetokun, 2011 ), in the United States ( Katz et al., 2011 ), in Ghana ( Takal & Quaye-Ballard, 2018 ), and in China ( Lu et al., 2008 ).

Septic tank wastewater often contains nitrogen, which is a major concern because of public health concerns and its potential adverse ecological effects ( Yang et al., 2017 ). Of the nitrogen species, nitrate (NO3) is the most prevalent. Generally, NO3 contamination of groundwater results from nitrogen movement in the unsaturated zone (vadose) and transformation processes in the unsaturated or saturated zones ( Tang et al., 2004 ).

In coastal areas, seawater intrusion has been reported in many areas as a global environmental concern, including the Mekong Delta in Vietnam ( Xiao et al., 2021 ), Ibeno in Southeastern Nigeria ( Inim et al., 2020 ), the northwest coast of Oman ( Ahmed & Askri, 2016 ), North Sinai in Egypt ( Gad & Khalaf, 2015 ), and Cyprus ( Kathijotes & Panayiotou, 2013 ). Research on the mechanisms and impacts of seawater intrusion has been conducted lately in various locations. Hydrochemical and isotopic parameters of groundwater have been applied by Galliari et al. (2021) in South American coastal wetlands for the purpose of examining the dominant processes of groundwater salinity. Based on a series of geostatistical methods, Sivakarun et al. (2020) investigated the dominant factors of hydrochemical processes in shallow coastal groundwater in India. In a thorough analysis of anthropogenic and natural processes governing groundwater salinization, Telahigue et al. (2020) evaluated the sources of different groundwater bodies based on the hydrogeochemical and isotopic characteristics of unconfined groundwater on Djerba Island in southeastern Tunisia.

Seawater intrusion is often closely linked to anthropogenic activities in densely populated areas. Salinization of superficial aquifers can result in the salinization of soils, with consequent decreases in crop yields, deterioration of vegetation, and gradual environmental decline, along with a decline in the quality of farmlands ( Shi & Jiao, 2014 ). Furthermore, seawater intrusion induces direct pollution of groundwater, which makes it unsuitable for industrial production, irrigation, and consumption. Global warming induces a rise in sea level that may lead to large-scale seawater intrusion. Thus, it is necessary to review the future trends of seawater intrusion, the mechanism of occurrence, and the current status to better identify the inevitable risks of seawater intrusion that will arise in different regions. Based on Kamal et al. (2020) , the significant increase in Cl and TDS levels in coastal groundwater towards the coastline is a good indicator of seawater intrusion. However, water-rock interaction is affecting the salinity of groundwater ( Nefzaoui et al., 2023 ; Vespasiano et al., 2021 ).

Groundwater pollution can affect socioeconomic development, environmental quality, and human health. Numerous studies reveal, in particular, a health risk to human populations from significant levels of persistent organic pollutants (metals, fluoride, and nitrate) ( Sunitha et al., 2022 ). Infants and children are particularly vulnerable to the effects of these contaminants ( Adimalla et al., 2020 ; Kadam et al., 2021 ; Mohammadpour et al., 2022 ; Shalyari et al., 2019 ; Su et al., 2021 ; Wu et al., 2019 ; Yin et al., 2020 ). High nitrate concentrations in drinking water for formula preparation are responsible for infant methemoglobinemia, also known as “blue baby syndrome.” In addition, irrigation using groundwater contaminated with wastewater containing persistent contaminants such as heavy metals is likely to cause vegetables and grains to accumulate toxic elements, thus posing risks to human health ( Kumar et al., 2022 ; Minhas et al., 2022 ; Pratap et al., 2021 ).

Groundwater pollution can have a negative impact on the quality of forests and land. In fact, groundwater contamination is likely to promote the degradation of soil quality. According to Wu et al. (2014) , the high salinity of groundwater is one of the main factors of soil salinization in many agricultural areas in arid regions. Besides, a nuisance to vegetation growth may be due to the accumulation of soluble salts and some contaminants, including toxic metals, in the root zone. Teng et al. (2018) reported that due to surface water and groundwater interactions, contaminants from groundwater can be transported, resulting in deterioration of surface water quality.

For sustainable economic development, balancing human demand with the renewal rate of natural resources is essential ( Awan, 2013 ). Fresh groundwater is probably the most valuable natural resource. Yet, under chronic groundwater contamination, the freshwater supply can be depleted, breaking the balance of demand and water availability to the degree of socio-economic stress and even war. Water shortages as a result of contamination are likely to become a factor in a future conflict between locals ( Ricart et al., 2021 ), ultimately delaying the socioeconomic development of the community. The issue of groundwater contamination is not just an environmental issue; it is also a social issue, requiring close cooperation among social scientists and environmental scientists.

The DRASTIC method by the National Water Well Association (NWWA) and the United States Environmental Protection Agency (USEPA) was developed in 1987 to assist administrators, planners, and managers ( Aller et al., 1987 ). The principle of vulnerability assessment by the DRASTIC method is based on three fundamental assumptions: the contaminant is considered to have the same mobility as water; the contaminant is carried from the soil surface to the aquifer by effective infiltration, i.e., by vertical flow; and the source of potential contamination is located at the soil surface.

The DRASTIC method assesses the intrinsic vertical vulnerability of aquifers to contamination by taking into account seven hydrogeological parameters. DRASTIC matches the first letters of the seven parameters: Depth to water (D); Net Recharge (R); Aquifer media (A); Soil media (S); Topography (T); Impact of the vadose zone (I); and Hydraulic Conductivity of the aquifer (C). First, the parameters are classified into ranges (for continuous variables) or classes (for thematic data) and then assigned a rating value (r) between 1 and 10, impacting the potential pollution. Second, weight multipliers (w), between 1 and 5, are assigned to each parameter to balance and reinforce their significance. The output is a vulnerability index (VI) that is a weighted sum of the rating value (r) multiplied by (w), the weight associated with each of the seven parameters: D, R, A, S, T, I, and C:

$VI=Dr\ast Dw+Rr\ast Rw+Ar\ast Aw+Sr\ast Sw+Tr\ast Tw+Ir\ast Iw+Cr\ast Cw$ (1)

(where D, R, A, S, T, I, and C are the seven parameters of the DRASTIC method, W being the weight of the parameter, and R the associated score or rate).

Two versions of the DRASTIC method exist: the standard DRASTIC version, applied in the case where the contaminants considered are inorganic pollutants, and the DRASTIC pesticide version, applied in the case where the contaminants considered are pesticides. The values of the ratings (r) and weights (W) of the parameters, in both versions of DRASTIC, are shown in Table 2 . Calculated DRASTIC index values reflect the hydrogeological vulnerability of the aquifer, with specific zones recognized as being more sensitive than others to groundwater contamination. DRASTIC index values range from 23 to 226 for the standard version and from 26 to 256 for the pesticide version. These values are grouped into five classes, each corresponding to a degree of vulnerability: very high, high, medium, low, and very low ( Aller et al., 1987 ) ( Table 3 ). On the other hand, Engel et al. (1996) propose classifying values into four further classes ( Table 4 ). According to Aller et al. (1987) , the contaminant imitates the mobility of groundwater. Therefore, contaminants released at the surface are likely to reach groundwater. This approach focuses on man-made contamination rather than pollutants released at the surface or at depth by processes such as injection wells, animal waste lagoons, or leaking underground storage tanks.

Table 2. Parameter ratings and weights in standard and pesticide versions of DRASTIC based on Aller et al. (1987) .

Table 3. Vulnerability assessment criteria for standard and pesticide.

DRASTIC model applications range from the local scale, with the study by Tomer et al. (2019) in the National Capital Territory of Delhi, India, to the continental scale, through a critical application of the DRASTIC method by Rama et al. (2022) to investigate South American groundwater.

An important benefit of the DRASTIC method resides in its flexibility to eliminate and integrate specific parameters into the model, depending on the availability of data and the conditions of the area under study ( Ahmed et al., 2022 ; Singh et al., 2015 ; Thirumalaivasan et al., 2003 ). Moreover, rate and weight scores can be adjusted according to field measurement data.

Nevertheless, it has been indicated by Hamza et al. (2015) that groundwater contamination is influenced in the same way by all parameters, insofar as each parameter indicates a significant impact regardless of the weighting assigned to the parameters. Decision-makers and researchers, therefore, need to go beyond the assumed weight of a given factor when assessing groundwater vulnerability and to perform an in-depth scientific analysis if they are to manage groundwater contamination effectively.

The DRASTIC approach is a useful tool for groundwater vulnerability assessment due to its simplicity, the availability of widely estimated or widely available data, and its low cost. The DRASTIC approach is a useful tool for groundwater vulnerability assessment due to its simplicity, the availability of widely estimated or widely available data, and its low cost. The application of the Geographic Information System (GIS) allows us to produce a map that is easy to both understand and integrate into the decision-making process ( Abunada et al., 2021 ).

As mentioned earlier, the DRASTIC method is one of the most widely used methods for assessing the vulnerability of groundwater resources, owing to its ease of application and performance. At present, there are two options for adapting the DRASTIC method to field conditions and improving the results obtained with the DRASTIC method:

The first option for improving DRASTIC’s efficiency, based on an in-depth scientific analysis of the data, is to optimize the weighting and rating of the model’s parameters. Some successful approaches have been experimented with in recent years. Single-parameter sensitivity analysis (SA-DRASTIC), fuzzy pattern recognition (F-DRASTIC), and entropy information (E-DRASTIC) were applied by Sahoo et al. (2016) to Kanpur City in India to more appropriately weight DRASTIC factors for vulnerability to groundwater contamination while comparing the performance of subjective (DRASTIC, SA-DRASTIC) and objective (F-DRASTIC, E-DRASTIC) models based on weighting. The results showed the effectiveness of objective methods in assessing the vulnerability of Kanpur City.

In the second option, DRASTIC’s performance can be enhanced by altering its original parameters, whether by adding or substituting with other parameters, such as irrigation type and land use. A study was carried out in the Tiruchirappalli district (India), an area with extensive agricultural practices. To ensure a better assessment of the area’s vulnerability, Jenifer and Jha (2018) used the original DRASTIC and DRASTIC-P models, to which two further parameters were added: lineament density (LD) and land use (LU). The two original models (DRASTIC and DRASTIC-P) were then compared with six modified versions of these models, namely DRASTIC-LU, DRASTIC-LD, DRASTIC-LULD, DRASTIC-P-LU, DRASTIC-P-LD, and DRASTIC-P-LULD. All eight vulnerability models were validated using a single water quality parameter (F⁻, Cl⁻, and ${\text{NO}}_3^{-}\text{-N}$). With 61% and 68% accuracy for nitrate and chloride concentrations, respectively, the most accurate model was DRASTIC-P-LDLU, and the second most accurate model was DRASTIC-LDLU, with 59% and 61% accuracy for the same concentrations.

The DRASTIC method for delineating vulnerable zones underestimates groundwater pollution potential ( Bai et al., 2012 ). In the literature, the application of the DRASTIC method works for all potential contaminants, regardless of anthropogenic activities. Therefore, the complexity of a specific vulnerability assessment may lie in the wide diversity of pollutants present in nature, whose behavior and fate in hydrogeological units are highly disparate. As a result, there is a need for careful use of a simple system that guarantees the accuracy of results ( Ferreira & Oliveira, 2004 ).

A further limitation of the DRASTIC method is that the data used to generate the various parameters applied by the approach influence the reliability of these parameters. In most cases, most information comes from interpolation ( Barbulescu, 2020 ), as in the case of parameters conditioned by the lithology of the environment, such as hydraulic conductivity, effective aquifer recharge, the impact of the vadose zone, and water table depth ( Kouz et al., 2020 ). This aspect leads to errors when generating parameter values, as it is only accurate within the intervals delimited by the punctual data. Consequently, the DRASTIC model can only be used as a relative assessment tool and not to provide an absolute assessment of groundwater vulnerability.

Moreover, another limitation of the DRASTIC approach is that it assumes that water and contaminants penetrate vertically from the ground surface to the water table. However, the method ignores the different situations in karst aquifers, where water and contaminants flow laterally through shallow holes ( Oke , 2017 ).

The Susceptibility Index (SI) is a model developed in Portugal by Ribeiro (2000) to assess specific vertical vulnerability to agricultural pollution (mainly nitrates but also pesticides). The model involves five parameters. One is land use (LU), and the other four are identical to four parameters already used in the DRASTIC method, namely depth to water (D), net recharge (R), aquifer media (A), and topography (T). The SI method omits three DRASTIC parameters: hydraulic conductivity, soil environment, and aquifer vadose zone. Indeed, the inclusion of the additional parameter (land use), according to Ribeiro (2000) , in the SI method compared to the DRASTIC method is based on the fact that land use remains a key and influential factor in the contamination of groundwater by pollution generated by anthropogenic activities. As per Brindha and Elango (2015) and Teixeira et al. (2015) , the integration of the land use factor in groundwater quality assessment is a key issue that should be taken into account when predicting the effect of anthropogenic activities on groundwater quality. In addition, Francés et al. (2001) consider the soil environment as a factor indirectly represented by land use, so its use remains a repetition. Further, the deletion of the vadose zone and hydraulic conductivity is justified by the fact that these two factors overlap ( Stigter et al., 2006 ), and, besides that, hydraulic conductivity reflects the aquifer media already included ( Engel et al., 1996 ).

The Susceptibility Index (SI) is computed by summing the products of the ratings and the corresponding parameter weights:

$SI=Dr\ast Dw+Rr\ast Rw+Ar\ast Aw+Tr\ast Tw+OSr\ast OSw2$ (2)

(where D, R, A, T, and OS are the five parameters of the SI method, W being the weight of the parameter, and R the associated score or rate).

With the SI method, the ranges for water depth, net recharge, topography, and aquifer media are the same as in DRASTIC, whereas the ranges used for land use are derived from the Portuguese scientists’ classification ( EUROPEAN COMMUNITY, 1993 ), as depicted in Table 4 .

Table 4. Main land use classes and corresponding rates.

The weights attributed to SI parameters, ranging from 0 to 1, are shown in Table 5 in accordance with the parameters’ importance in vulnerability.

Table 5. Weights allocated to SI parameters.

Based on the index values obtained from equation (2), Table 6 lists the four degrees of vulnerability of the SI method.

Table 6. Vulnerability categories for the SI method.

Geographic Information Systems (GIS) and remote sensing technologies are increasingly being used in groundwater contamination risk assessment. These technologies, combined with the SI method, allow us to develop an integrated approach with an emphasis on heterogeneous environments that take into account geochemical, hydrological and geological data in order to improve the vulnerability assessment accuracy ( Anane et al., 2013 ; Bartzas et al., 2015 ). Moreover, the SI approach, similar to the DRASTIC method, has been developed to assess the vulnerability of aquifers on a medium and large scale (e.g., 1: 200,000 - 1: 50,000).

Over the past few years, numerous aquifer vulnerability assessments have been carried out all over the world using the SI method. An application of the SI method by El Himer et al. (2013) to assess the vulnerability of the hydrogeological watershed of the Oualidia-Sidi Moussa wetland in Morocco reveals that the area is characterized by medium to high vulnerability to pollution. According to El Himer et al. (2013) , the high vulnerability to pollution reflects the nitrate levels measured in groundwater during sampling campaigns. The SI method’s land-use parameter has proved useful in the Caldas da Cavaca area in Central Portugal, highlighting urban areas where agricultural land and buildings are concentrated ( Teixeira et al., 2015 ). For Teixeira et al. (2015) , most of the Caldas da Cavaca area has moderate and low-to-moderate vulnerability values, due to the presence of high slopes, rocky outcrops, and less weathered granitic zones. Recently, Ghouili et al. (2021) applied the SI method in the northeast of Tunisia to assess the vulnerability of the Takelsa phreatic aquifer to contamination. In order to validate the generated vulnerability maps, high-salinity groundwater areas were compared to their respective vulnerability indices. From these validated vulnerability maps, Ghouili et al. (2021) reported that these high-vulnerability areas can be associated with low slopes, sandy soils, shallow water tables, and highly agricultural areas with very high recharge rates.

One of the weaknesses of the SI method is that it fails to identify the path that the contaminant will take through the hydrogeological system. Indeed, the SI method is limited to vertical movements, ignoring the lateral migration of elements. Therefore, the SI method deals with the water’s sources of contamination rather than the state of pollution itself. Other SI method limitations include the need to correlate the vulnerability index with the nitrate concentration measured in the field in order to validate the method. Indeed, the use of the SI method on the shallow aquifer of Nabeul-Hammamet in Tunisia ( Anane et al., 2013 ) and the southern aquifer of Teheran ( Noori et al., 2019 ) revealed an overestimation of vulnerability due to the impact of dilution, highlighting the difference between the most vulnerable zones and the most contaminated zones. Moreover, Anane et al. (2013) reported that groundwater recycling, which contributes to the accumulation of pollutants, leads to an underestimation of vulnerability as two factors are not taken into account, namely the unsaturated zone and the soil conditions.

The basic principle of the DRASTIC and SI methods is the use of parameters within a weight-class system. However, the weights assigned to the different parameters by the two approaches differ. Therefore, the resulting intrinsic vulnerability also varies according to their inputs. To achieve greater confidence in the vulnerability assessment approach, the best strategy is to conduct case studies in areas where contamination has occurred. These studies ultimately make it possible to compare the results of different vulnerability assessment methods in terms of their consistency and relevance.

In the Nabeul-Hammamet shallow aquifer in Tunisia, Anane et al. (2013) tested the effectiveness of the DRASTIC, DRASTIC-Pesticide, and Susceptibility Index methods for determining the degree of vulnerability to pollution. According to Anane et al. (2013) , there was a major difference between the vulnerability maps obtained using the DRASTIC and SI methods, as the DRASTIC and SI categories overlapped in only 27% of the total aquifer area, with SI reflecting field reality better than DRASTIC. The reason is that the SI method gives greater weight to the LU parameter than to the other four factors. Meanwhile, the Nabeul-Hammamet area is an agricultural zone and is, therefore, less likely to pollute groundwater. On the other hand, Anane et al. (2013) reported a good correlation between SI and Pesticide DRASTIC (64% of the total aquifer area). Such a match can be explained by the fact that both models assess the specific vulnerability to human activity, especially agriculture practiced in the Nabeul-Hammamet. Another factor is the high weight given to land use in the SI method and to soil media in the Pesticide DRASTIC method, as soil characteristics have a strong influence on land use. An assessment of groundwater vulnerability by Duarte et al. (2019) in Portugal’s Serra da Estrela Mountains shows that SI vulnerability index values tend to be lower when compared with DRASTIC vulnerability index values, which tend to be more intermediate. This makes DRASTIC the most balanced choice for a strict display of intrinsic groundwater vulnerability. Conversely, the influence of specific agricultural and urban land use classes is only represented on the SI map. An application of the DRASTIC and SI methods to study vulnerability to potential nitrate pollution in the Timahdite-Almis Guigou groundwater in Morocco Amrani et al. (2019) revealed that the coincidence rate of groundwater nitrate concentrations with the different vulnerability classes established was 61.54% and 76.92%, respectively, for the DRASTIC and SI methods. From this rate of coincidence, the SI method appears to be the most appropriate for assessing vulnerability to nitrate pollution. In southern Portugal, Stigter et al. (2006) applied the DRASTIC and SI methods to nitrate contamination and groundwater salinization in two agricultural regions (Campina de Faro and Campina da Luz). In Campina da Luz, characterized by karstified limestone aquifers, both DRASTIC and SI methods overestimated vulnerability due to neglect of the dilution aspect. Despite having significant control over contamination levels, the DRASTIC method ignores groundwater dilution, thereby leading to inaccurate results ( Aller et al., 1987 ). SI minimizes DRASTIC’s error by eliminating the impact of the vadose zone. In fact, SI incorporates land use, which provides valuable additional information. Meanwhile, using the SI method tends to overestimate vulnerability, which is preferable to underestimation, since it involves the security of uncertainty ( Elshall et al., 2020 ). A study by Brindha and Elango (2015) to identify the most appropriate method for assessing groundwater vulnerability in a weathered rock aquifer in southern India found that highly vulnerable areas to pollution in DRASTIC and Pesticide DRASTIC were similar to, but differed from, those identified by the SI method, which includes “land use” as one of the input parameters. The LU parameter, concludes Brindha and Elango (2015) , is a crucial input that needs to be taken into account in any hydrogeological context.