Risk assessment of hand-dug well water: a case study of Aflao in Ghana Open Access

High population growth, rapid economic development, and urbanization have exerted significant pressure, particularly in developing countries, on the inadequate conventional piped water supply systems (Amin et al. 2019). As a result, millions of people are left behind without access to safe water services, making them vulnerable to a range of preventable illnesses affecting the quality of life and also undermining the fundamental human rights to safe water for economic development (WHO 2022). Nearly 2 million deaths and 123 million disability-adjusted life years are linked to inadequate access to safe water, and 2 billion global population lack access to safe and reliable sources of water (WHO 2022). This pressing global issue of water accessibility is more pronounced in developing countries, especially in sub-Saharan Africa (Ogunbode et al. 2024).

The coping strategies with inadequate formal and/ or conventional piped water systems in communities in developing countries are reliance on groundwater sources through developing boreholes and hand-dug wells (HDWs) (Cid Escobar 2024). Boreholes are typically deeper and better protected and therefore are generally regarded as safer options (Rauf et al. 2021). Most HDWs are unprotected wells and are considered unsuitable for consumption due to the high risk of contamination from surface runoff, nearby sanitation facilities, and other environmental factors (Addo et al. 2023; Kupa et al. 2024). Despite these safety concerns, many areas continue to rely on HDWs as a primary water source because of their affordability, ease of construction, and lack of access to more secure alternatives like boreholes or treated piped water systems (Malinga & Hashe 2024). In many developing countries, HDWs serve as essential water sources, particularly in rural areas of sub-Saharan Africa and parts of Asia (Zhang et al. 2020).

According to the Ghana Statistical Service (2021), 12% of the Ghanaian population relies on unprotected HDWs for their water needs. The Ghana National Water Policy recognizes access to safe water as a basic human right including protection of unprotected sources like HDWs (Ministry of Sanitation & Water Resources 2024). The policy also mandates water safety measures such as integration into national planning, regularly prepare and review national and Integrated Water Resources Management (IWRM) plans, promote practices that protect critical natural resources and prevent irreversible ecological damage and the ‘polluter pays’ principle to address ground and surface water contamination, ensuring equitable access and inclusive management for all, especially marginalized groups.

Most Ghanaian HDWs are unprotected with significant risk of environmental contamination (Addo et al. 2023). For instance, the absence of centralized wastewater treatment systems in many towns and communities has led to the proliferation of private on-site sanitation facilities (Asumadu et al. 2023). These cluster of sanitation facilities pose a significant risk of contaminating groundwater through the infiltration of wastewater with contaminants such as pathogens, heavy metals (HMs) and others (Addo et al. 2023). Evaluating the microbial and physicochemical quality of HDWs is crucial because it provides insights into the risk associated with harmful contaminants that are a threat to public health (Kushwah & Singh 2024). HDWs could pose significant health risks mostly because they are highly susceptible to microbial and chemical contamination (Xue et al. 2020). Yet there are inadequate studies in Ghana on the risk associated with HDW water sources especially in larger towns especially where such wells serve as the main water supply sources.

One of the busy border towns in Ghana within the Ketu-South Municipality, Aflao's main water source is HDW (Amoako et al. 2023). There is limited information like in other towns on the risk associated with the HDW water sources. This study therefore assessed the risk of the HDW water source in Aflao to provide an understanding of sanitary risk factors using the World Health Organization (WHO) observation checklist.

Ketu-South Municipal (Figure 1) is a low-lying area that ranges in elevation from 66 m inland to less than 15 m close to the coast. The coast is smooth and dotted with sandbars. The municipality's drainage system is dominated by seasonal streams and flows southward to Aflao (Babanawo et al. 2022). The average annual precipitation is 850 mm at the coast and 1,000 mm inland, with a double rainfall peak from April to July and September to October. The dry season, from December to February, is marked by harmattan winds. Rainfall during the minor season is low and unpredictable, particularly between Agbozume and Aflao (Babanawo et al. 2022). The hydrogeology includes recent and tertiary formations of unconsolidated sands and clays, as well as partially consolidated red continental deposits of sandy clay and gravel. Beneath newer coastal sediments lies a thick layer of marine sands, clay, shale, limestone, sandstone, and some gravel (Amoako et al. 2023).

In addition to its unique terrain, Aflao was chosen for this study because it is one of the busy large towns in Ghana with majority of its population (70%) relying heavily on HDWs (GSS 2021) due to persistent challenges with public water supply.

The study used a cross-sectional survey approach. The WHO checklist was used for the sanitary risk assessment (SRA) of the HDWs, along with laboratory analyses of water samples from HDWs.

A sanitary risk survey was conducted to determine the sanitary risk factors of 400 HDWs in Aflao, using WHO guidelines which involved identifying potential factors and sources of contamination. The sanitary inspection technique was adopted from the WHO Guide for Drinking Water Quality standard having a systematic checklist of a small number of specific inquiries (WHO 2024). This checklist addressed the 10 most fundamental potential sources and factors of well water contamination.

Water samples were collected from 20 HDWs using 500 ml sterilized plastic bottles and transported to the Council for Scientific and Industrial Research (CSIR) - Water Research Institute laboratory in Accra within 24 h under a controlled temperature of 4 °C in an ice-chest box during transportation following the protocol in APHA (1989).

Atomic absorption spectroscopy (AAS) (Table 1) was employed to determine the concentration of metals (Lead [Pb], Cadmium [Cd], Chromium [Cr], Nickel [Ni], and Arsenic [As]) by using the protocol of APHA (1989). Sample solution was aspirated into a flame and atomized. Through the flame, a light beam was focused, passing through a monochromator and onto a detector that determined how much light the flame absorbed. Since each metal has its characteristic absorption wavelength, a source lamp composed of that metal was used. The amount of energy at the characteristic wavelength of 279.5 nm absorbed in the flame was proportional to the concentration of the element in the sample over a limited concentration range.

Microbial analysis: Membrane filtration was used, employing a sterile 0.45 μm Millipore filter, Erlenmeyer flask, and vacuum source. Samples were filtered and placed on selective media, including Xylose Lysine Deoxycholate (XLD) Agar for Shigella and Salmonella, Membrane Faecal Coliform Agar for faecal coliforms (FC), total coliforms (TC) and Hi-Chrome Agar for E. coli, all in separate Petri dishes. Incubation occurred at 37 ± 2 °C for 18–24 h for the analysis. Clamps and forceps were sterilized before each use. All these procedures and techniques followed the APHA Standard Methods for the Examination of Water and Wastewater (APHA 1989).

Following Viban et al. (2021), regression analysis was used to examine the link between detected organisms and observed risk factors/sources, specifically utilizing the beta Poisson model. In our approach, ‘No’ responses to sanitary risk factors were assigned ‘absence of a risk factor’ and taken as the reference categories in the statistical analysis, allowing for a comparison with the ‘Yes’ responses, which indicated the presence of a risk factor. This involved the use of odds ratios (OR) at a 0.05 significance level. As demonstrated by Viban et al. (2021), an OR below one suggested no significant difference between contamination and observed risk factors. Conversely, an OR exceeding one indicates an association with the specific risk factor under consideration, while an OR of one implies no discernible difference between contamination and observed factors. SRA was obtained through the adoption of the WHO SRA checklist for well water. Household scores 9–10 risk factors indicate a ‘very high risk,’ while 6–8 signifies ‘high risk,’ 3–5 denotes ‘moderate risk, ‘1–2 suggests ‘low risk,’ and 0 ‘no risk’. The statistical tool employed for the analysis was Statistical Package for Social Sciences version 27.

Health risk assessment for HMs involves the use of guideline limits, and models such as hazard quotient (HQ), hazard index (HI), and the lifetime cancer risk (LCR) model for carcinogenic risks. The HQ evaluates non-carcinogenic effects and HI which is the sum of all HQ values for different metals, provides a cumulative risk assessment with values below ‘1’ indicating no significant risk. If the HI value is less than 1, there is no significant risk of combined health effects (Balogun et al. 2023). The LCR model estimates the probability of developing cancer over a lifetime due to exposure to metals like As, Cd, Cr, and Ni.

The surveyed population, composed of 400 respondents, exhibited a diverse demographic profile with an average age of 46 and a gender distribution of 76% females and 24% males. Education levels varied, ranging from 38% with no formal education to 3% attaining tertiary education. Employment diversity was notable, with 31% not employed, including about 7% retirees, 61% self-employed predominantly traders, and 8% in government positions. The average household size was 3, and residents had, on average, lived in the community for 20 years, occupying houses with an average size of nine people. All the assessed wells were privately owned by households and fully depended on for their domestic and drinking purposes.

The results of the SRA conducted on 400 HDWs are presented in Table 2. The findings indicate that several sanitary risk factors were widespread and needed to be addressed. The absence of a concrete floor around wells was noted in 94.5% of cases, while shallow well depths of less than 30 m and the incorrect handling of ropes and buckets were present in all wells studied (100%). Also, the absence of covers for wells was documented in 98.25% of wells, and insufficient drainage systems were noted in 88.5% of wells. These factors, due to their high prevalence, represent critical areas requiring urgent intervention to ensure the safety of water from these wells. However, some sanitary risk factors, while less prevalent, still pose significant concerns and should not be overlooked. The presence of latrines within 10 m of the wells was identified in 37.3% of the wells, and 20.8% were located below latrines positioned on higher ground. Wells surrounded by animal excreta or rubbish within a 10-m radius accounted for 35.3% of the total, and 31.8% were located near or under trees, which may contribute to organic debris and animal droppings entering the water. Inadequate construction of well headwalls, observed in 7.8% of cases, represents a less common but concerning issue that may allow surface water to contaminate the wells. While all the factors assessed permit attention to reduce contamination risks, it is evident that issues such as the absence of concrete floors, shallow well depths, improperly handled ropes and buckets, absence of covers, and poor drainage systems should be prioritized due to their higher prevalence and immediate threat to water quality. Factors with lower prevalence, such as proximity to latrines, trees, or rubbish, and inadequate headwalls, though less urgent, remain essential to address for long-term water safety improvements.

The mean concentration of total coliform was 579.7.00 CFU/100 ml with a standard deviation (SD) of 294.9 CFU/100 ml, indicating considerable variability among the samples. Faecal coliform had a mean of 32.6 CFU/100 ml and an SD of 54.7 CFU/100 ml, while E. coli recorded a mean concentration of 14.7 CFU/100 ml with an SD of 21.7 CFU/100 ml. Salmonella and Shigella were absent in all the samples (Table 3). The proximity of latrines within 10 m of a well, areas prone to flooding, and public toilet areas, as well as wells located under trees, wells designated for public use, latrines situated uphill from wells, the practice of digging and burying faeces within households, and the absence of proper drainage and well covers were significantly associated (p < 0.05) with increased levels of microbial contamination (Table 4). However, the presence of animal droppings did not show a significant association with contaminant levels (p > 0.5), nor did the presence of public toilets correlate with TC levels. Similarly, the absence of well covers, flood-prone areas concerning E. coli, and the practice of digging and burying faeces of E. coli were not significantly associated with contamination.

Key risk factors with higher OR (OR > 1) indicating a strong association with contamination included the presence of latrines within 10 m of a well (average distance between eight households with septic tank toilets and observed well water was 6 m, with 10–11 m), which showed significantly increased odds for FC (OR = 5.620, CI = 4.114–7.678), TC (OR = 1.247, CI = 1.183–1.313), and E. coli (OR = 2.699, CI = 1.799–4.051) contamination. Wells located under trees were also strongly associated with contamination, with OR and CI values of 1.890 and 1.220–2.928 for FC and 2.806 and 1.524–5.168 for E. coli. Additionally, wells designated for public use showed 2.516 times more likely to contaminate the well water with FC and 3.591 for E. coli. Latrines situated uphill from wells exhibited increased odds of contamination with OR values of 3.034 for FC and 1.710 for E. coli. The absence of a fence, cover, apron, and shallow well with an average of depth 3.84 m and water levels at 3.35 exhibited constant outcomes, with all observations and measurements consistently yielding the same result (risk factor observed). This implies that observed factors did not exhibit variation or fluctuation within the dataset. The unidirectional skewness of these constant observations and measurements suggests a lack of diversity or meaningful variability in the specific attributes assessed within the studied population as indicated in dash (–) in Table 4.

Pointing out the specific trend illustrated in Table 5, the analysis of the E. coli, total, and faecal coliform variable discovered that there were no statistically significant (p > 0.05) differences observed between any of the zones.

The results of the HMs analysis from all 20 samples consistently showed concentrations below the minimum detection limits (MDL) (Table 6): Pb (<0.005 mg/L), Cd (<0.002 mg/L), Cr and Ni (<0.010 mg/L), and As (<0.001 mg/L).

Microbial and chemical contamination of groundwater sources is a documented issue globally, often linked to poor sanitary conditions around sources (Baia et al. 2022; Amoako et al. 2023). Consistent with our findings, Anang et al. (2023) in Ghana observed significant microbial contamination levels in wells situated near sanitation facilities in a community. Similar patterns have been reported in China, where proximity to sanitation facilities emerged as a significant risk factor for contamination (Xue et al. 2020). In contrast, studies in Southeast Asia (Cao et al. 2012) reported lower contamination levels, likely reflecting differences in well-protection measures and sanitation infrastructure. These variations highlight the influence of regional factors such as hydrogeology, sanitation practices, and community behaviour on groundwater contamination.

The high frequency of microbial contamination of HDWs in the current study can be attributed to various factors, including environmental conditions and infrastructure limitations. Key predictors include the proximity of wells to latrines, lack of concrete floors and well covers, insufficient drainage systems, and the placement of wells under trees (Xue et al. 2020). These factors significantly (p < 0.05, OR > 1) increase contamination risks, as evidenced by the high levels of total coliform (mean = 579.7), faecal coliform (mean = 32.6), and E. coli (14.7) observed in this study (Table 4). Similar findings were reported by Anang et al. (2023), who linked faecal contamination in peri-urban Ghana to inadequate separation between sanitation facilities and water sources.

Cultural and environmental sanitation practices also influence contamination risks. In peri-urban communities characterized by slums or informal settlements, maintaining regulatory distances between wells and sanitation facilities is challenging due to limited space, hydrogeological variability, and cultural norms (Bhallamudi et al. 2019; Oyeniyi 2020). These challenges are compounded by poor maintenance of sanitation systems, as argued by Jenifer & Jha (2022) in India, where design flaws and substandard maintenance exacerbated groundwater contamination, regardless of proximity (Othoo et al. 2023).

The findings of this study aligned with several local investigations on HDW contamination in Ghana and global. For instance, Yousuf et al. (2021) identified poor sanitary conditions as a key driver of microbial contamination in wells across peri-urban US, consistent with our finding of a high mean score of TC (579.7.00 CFU/100 ml), FC (32.6 CFU/100 ml), and E.coli (14.7 CFU/100 ml) (Table 3) contamination. Similar to our findings, Houéménou et al. (2020) reported significant contamination in wells located near latrines and septic systems, suggesting a shared challenge of inadequate separation between water sources and sanitation facilities. In agreement with this study, Lutterodt et al. (2018) found that microbial contamination in HDWs was often linked to effluent seepage from on-site sanitation systems, particularly in densely populated areas. This underscores the difficulty of maintaining regulatory distances in communities with limited land availability and poor spatial planning.

This study also discovered additional risk factors, such as the placement of wells under trees and the lack of well covers and concrete floors, shallow wells and high water tables which may influence contamination of HDWs. These structural deficiencies are often overlooked in local studies but are critical for addressing water safety concerns (Braimah et al. 2021). Furthermore, seasonal factors such as heavy rainfall worsen contamination risks. Floods during the rainy season easily contribute to contamination through the leaching of contaminants, as documented by Cao et al. (2012), and this effect was further supported by Nascimento Santos et al. (2023), who observed a significant (p < 0.05) link between rainfall intensity and microbial contamination in well water samples.

In Senegal, Pouye et al. (2023) discovered that frequent rain can transport contaminants over greater distances than typically expected, undermining recommended safety distances for well placement. This observation defines the findings of this study, where wells located a distance from sanitation facilities still exhibited significant contamination possibly due to structural deficiencies identified in Aflao, such as the lack of concrete floors, the absence of cover and wells under trees. Therefore, well design and maintenance are important in controlling contamination (Jenifer & Jha 2022). Furthermore, Olalemi et al. (2021) study in Nigeria found that the absence of Salmonella and Shigella in well water, despite the presence of E. coli, could be attributed to differences in contamination sources, microbial communities, or environmental conditions. This aligns with our study as indicated oganisms (Salmonella and Shigella) were absent in analysed well water samples (Table 3).

An observation pertained to the deployment of standard well lining and parapet walls at all sample points, functioning as an effective barrier mechanism that delineated a protective interface between HDW water and potential contaminants originating from surface runoff (Okoye et al. 2023). Furthermore, this observation holds significance in the context of environmental hygiene, particularly given that the majority of these wells were typically situated within household premises and most houses owned a well which might reduce the risk of contamination from neighbours (Othoo et al. 2023).

The hydrogeological context has a key impact on the below detection limits of HMs recorded in the HDWs (Hao et al. 2024). Hydrogeological environments including recent and tertiary formations characterized by unconsolidated sands and clays, marine sands, and partially consolidated red continental deposits affect the prevalence of HMs in well water. These have significant adsorption capabilities for HMs, allowing them to retain metals in the soil and diminish their concentrations in groundwater (Bai et al. 2024). Coastal and riverine deposits influence groundwater recharge rates (Chmielarski et al. 2024). High recharge rates in river valleys and coastal regions can contribute to the diffusion of contaminants, contributing to reduced concentrations of HMs in the wells (Birla et al. 2020).

In some cases, HM contamination tends to come from natural mineral deposits located in deeper geological layers (Liu et al. 2024). Metals like As, Pb, and Cd may come from mineral-rich bedrock, but since HDWs do not reach these deep layers, they are less likely to locate water that has been in contact with such minerals (Richard et al. 2024). Groundwater flow through loosely consolidated sands and clays tends to be slower, allowing for higher contact with the soil matrix and enhanced possibility for metal attenuation (Wang et al. 2021). Additionally, the passage through marine clays and sands may promote the natural dilution of contaminants, resulting in lower detectable levels (Rahman et al. 2013).

The presence of industrial or anthropogenic activities near the wells or other water sources adds to HMs contamination through storm runoff and also geological features are identified as contributors to low HMs levels, with regions naturally low in metal concentrations yielding water with minimal contamination (Kapoor & Singh 2021). Therefore, the absence of industrial activity could be a potential factor in the below detection limit of targeted HMs in the samples result of the recent study.

This study assessed the risk of HDW water in Aflao. The results discovered that all 20 wells tested were contaminated with total coliform (mean = 579.7.00 CFU/100 ml), and E.coli (14.7 CFU/100 ml), and 18 wells contained faecal coliform (mean = 32.6 CFU/100 ml) and E. coli (mean = 14.7 CFU/100 ml), making the water unsafe for consumption by WHO standards (0 CFU/100 ml). Although HMs (Pb, Cd, Cr, Ni, and As) were below MDL - minimum detection limit (0.001–0.01 mg/L), there was a strong association (p < 0.05) between sanitary risk factors such as proximity to latrines, lack of concrete floors and covers, poor drainage, trees near wells, and shallow depth and microbial contamination. These findings recommend the need for better sanitation, replacement of components of the wells, education and promotion of well water treatment and hygiene to reduce contamination risks and safeguard public health in Aflao.

The findings are limited to the few numbers of purposively selected hand-dug well water samples which could be only indicative of samples taken given the potential biases with purposive sampling although selection criteria were defined. Also, seasonal variations (wet/rainy and dry) were not considered in the study design.

The study received ethical approval from the Committee on Human Research, Publication, and Ethics at Kwame Nkrumah University of Science and Technology (CHRPE) with reference number ‘CHRPE/AP/569/23’. Informed consent was also obtained from the Ketu-South Municipal Assembly and all participants according to the directives of CHRPE.

We are deeply grateful to the staff of CSIR-Water Research Institute, Accra, for their timely analysis of the samples. We would also like to extend our appreciation to the Ketu-South Municipal Assembly for approving to conduct of this study and to the owners of the hand-dug wells for their cooperation and for allowing access to their wells.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.