The study was conducted in the Copperbelt province in Kitwe district in Ipusukilo informal settlement in Zambia, located at latitude 12˚45' S and longitude 28˚20' E ( Figure 1 ). The area has a population of 31,405 distributed in three distinct zones stretching along the Kafue River basin. The region experiences a mean annual temperature of 23˚C and an average annual precipitation of 1226 mm ( Banda, 2023 ).

Administratively, Ipusukilo settlement is divided into three distinct sections ( Macwani et al., 2009 ). The areas of interest were the zones using shallow well water for their drinking use. Based on this criterion, Zones 1 (Southeast), 2 (Southwest), and 3 (Northwest) were purposively selected as strata. A total of 16 wells were then randomly selected from these zones for microbial water quality sampling. Water samples were collected in triplicate from each site in October (dry season) and December (wet season). Giving a total of 48 water source samples in the dry and wet season. GPS coordinates for the sampled shallow wells were collected and are presented in Figure 1 .

The pH and temperature of water were measured on-site using a multi-parameter. This was done by dipping the electrode of the calibrated instrument into the sample and allowing it to stabilize to give a reading. It was then rinsed with deionized water before being used to measure other samples.

The sample collection procedures were based on the American Public Health Association ( APHA, 2005 ) standards. Water samples for microbial analysis were collected using sterilized 500 ml high-density polyethylene (HDPE) bottles to avoid contamination. Sterile gloves were worn, and care was taken to avoid contact with the bottle’s interior. The bottles were rinsed with the sampled water and sealed after filling. Samples were stored in a cool box at 4˚C and transported to the Department of Environmental Engineering Laboratory at Copperbelt University for analysis within six hours.

The Membrane Filtration Technique was used to analyze water quality from shallow wells for total coliform and E. coli concentrations. The setup included a filter holder, a 0.45 µm sterile membrane filter, and a vacuum pump. A 100 mL water sample was drawn through the membrane filter, which collected coliform bacteria. This filter was then placed onto a selective growth medium using forceps ( APHA, 2005 ). For total coliform detection, M-Endo Agar was used and incubated at a temperature range of 35˚C to 37˚C for 24 hours. E. coli was identified using HiChrome and incubated at 35˚C for 24 hours. After incubation, the petri dishes were removed and examined for bacterial growth. The colonies were then counted using a colony counter. On M-Endo Agar, total coliform colonies appeared as red or pink with a metallic sheen, while E. coli colonies displayed a blue color on HiChrome.

The total bacterial count in every 100 mL was reported as Colony Forming Units (CFU) per 100 mL calculated using the formula:

$\frac{\text{CFU}}{100\text{mL}}=\frac{\text{Number of CFU}}{\text{Volume of water filtered}}\times 100\text{mL}$(1)

A sanitary survey was conducted at each sampling site during water sample collection process. The sanitary survey followed a standardized format that was adopted by Lloyd & Bartram (1991) . The sanitary scores from shallow wells were plotted against E. coli counts to evaluate their relationship.

Clinical health records from the health facility in the settlement were reviewed to identify the most prevalent water-related diseases a year before the study period. The data collected included the age and the specific disease recorded during the dry and wet seasons. The prevalence rates of water-related diseases were calculated using the equation below:

$PR=\frac{P}{N}\times 100$(2)

where PR denotes the prevalence rate, P denotes all persons with a specific condition at one point in time, and N means population in the target area.

Descriptive statistics, including the mean and standard deviation, were used to summarize the data on microbial water quality. The Shapiro-Wilk test was performed to assess data normality. Furthermore, Pearson correlation tests were conducted to explore the relationships between total coliform and E. coli levels and the physicochemical parameters of water during both dry and wet seasons. A paired t-test was used to determine if there were any significant seasonal variations in microbial water quality. All statistical analyses were conducted using R Studio software.

The study protocol was approved by the Ethics Committee of Egerton University EUISERC/APP/364/2024 and the Ethics Committee of the Tropical Diseases Research Centre TDREC237/10/24. Authorization to conduct the study was granted by the Zambia National Health Research Council NHRA-1612/05/10/2024 and the Zambia Provincial District Health Office.

Temperature varied slightly between seasons, measuring 27.2˚C ± 0.5˚C in the wet season and 27.5˚C ± 0.3˚C in the dry season Table 1 . Shallow wells exhibited higher temperatures in both seasons, likely due to their proximity to the surface and direct exposure to the sun ( Morris et al., 2003 ). Mean pH values remained within the WHO guideline range of 6.5 - 8.5. The pH increased from 6.6 ± 0.5 in the dry season to 7.2 ± 0.5 in the wet season. The high mean pH value recorded in the wet season is likely due to increased organic matter dissolution following rainfall events, as suggested by Chauque et al. (2021) .

Total coliform counts were 79 ± 24 CFU/100 mL in the dry season and 124 ± 40 CFU/100 mL in the wet season. Escherichia coli, a key indicator of microbial contamination, measured 26 ± 16 CFU/100 mL in the dry season and 55 ± 20 CFU/100 mL in the wet season. Both exceeded the WHO guideline of 0 CFU/100 mL, indicating bacterial contamination ( Figure 2 ).

The presence of E. coli in a water source suggests fecal contamination and the potential presence of harmful pathogens and other bacteria that can cause water-related diseases. The results revealed that contamination of drinking water with E. coli was 87.5% and 100% during the dry and wet seasons, respectively. Shallow wells, particularly in informal settlement areas, are highly susceptible to contamination due to their limited depth ( Nayebare et al., 2022 ). Several studies have reported fecal contamination of shallow wells. For instance, research conducted by Haramoto (2018) in Nepal found that 100% of the sampled shallow wells contained detectable levels of E. coli, a finding that aligns with the results of the current study. Similarly, Segut et al. (2024) reported that 80.6% of shallow wells in Kenya were contaminated with fecal coliforms, which is notably higher than the 60% contamination rate observed in shallow wells in Ethiopia ( Gizachew et al., 2020 ).

The World Health Organization (WHO) provides guidelines to help assess the safety of drinking water based on classified risk levels of E. coli contamination ( WHO, 2017 ). Water that falls into the high-risk category is considered unsafe and has a higher chance of causing waterborne diseases ( Saima et al., 2023 ; Yang et al., 2025 ). In this study, 75% of shallow wells during the dry season fell within the high-risk category (11 - 100 CFU/100 mL). Contamination levels worsened in the wet season, where all sampled shallow wells (100%) exceeded the WHO zero threshold for safe drinking water, placing them in the high-risk category ( Table 2 ).

A paired t-test revealed significant seasonal variation of bacteria counts in water sources. Total coliform counts were lower during the dry season than the wet season (p < 0.001) and E. coli counts were lower during the dry season than during the wet season (p < 0.001) ( Table 3 ).

**Significant at p < 0.001.

The seasonal fluctuation observed underscores the heightened vulnerability of drinking water sources to microbial contamination during the wet season, thereby posing significant public health risks ( Berihun et al., 2023 ). The relationship between rainfall patterns and water quality deterioration has been observed in several regions including, Ghana, Nigeria, Guatemala, and Sri Lanka. Studies have shown that the wet season significantly increases microbial contamination in water sources ( Dongzagla et al., 2021 ; Ayeni et al., 2023 ; Caballero et al., 2024 ; Thilakarathna et al., 2025 ). This trend suggests that increased rainfall not only facilitates the transport of pathogens but also creates favorable conditions for bacterial proliferation, further compromising water quality.

The role of hydrogeological factors in influencing water quality cannot be overlooked. During the wet season, rising water tables, increased infiltration, surface runoff, and rapid recharge of shallow aquifers facilitate the transport of pathogens into groundwater. These wells, which often lack adequate casing or protective barriers, become highly susceptible to contamination from surrounding environments, particularly from sanitation facilities such as pit latrines ( Machona et al., 2025 ). Rainwater infiltration can introduce fecal matter from human waste into groundwater supplies, significantly increasing bacterial loads. The absence of effective drainage systems in settlement areas further exacerbates the situation, allowing contaminated water to stagnate and seep into drinking water sources ( McGill et al., 2019 ; Awuah, 2024 ).

The complexities of groundwater quality dynamics have also been illustrated by Kanyerere et al. (2012) , who noted that in Malawi, variations in seasonal patterns of contamination can be attributed to localized environmental conditions and sanitation practices. However, the findings of this study contrast with results from Uganda ( Nayebare et al., 2022 ), where higher microbial counts were reported during the dry season. The high E. coli counts during the dry season were attributed to the greater accumulation of fecal matter at the land surface and less-diluted, fecally contaminated recharge. In some arid and semi-arid regions, reduced availability of water during the dry season may lead to water stagnation and higher concentrations of contaminants, explaining the observed differences ( Bouchaou et al., 2024 ; Water and Development, 2024 ).

The relationship between microbial and physicochemical parameters was assessed using the Pearson correlation coefficient to evaluate the strength of association between E. coli, total coliforms, temperature, and pH. The correlation coefficient ranges between −1 and +1. A correlation coefficient of ±0.1 to ±0.39 represents a weak linear relationship, while a coefficient between ±0.4 and ±0.69 indicates a moderate correlation. A strong correlation is observed within the ±0.7 to ±0.89 range and coefficients from ±0.9 to ±1.0 reflect a very strong linear association ( Schober & Schwarte, 2018 ). The results of the statistical testing for the relationship between the variables are presented in Figure 3 .

Temperature significantly influenced microbial proliferation more so than die-off, as evidenced by the positive correlation with coliform bacteria. Specifically, E. coli showed a significant positive correlation with temperature during the dry season (r = 0.72, p = 0.002; Figure 3(a) ) and the wet season (r = 0.86, p = 0.001). Additionally, pH was also positively correlated with E. coli during both the dry season (r = 0.54, p = 0.032) and the wet season (r = 0.66, p = 0.005; Figure 3(c) ). For total coliforms, a positive correlation was observed with temperature in the dry season (r = 0.55, p = 0.028; Figure 3(b) ) and wet season (r = 0.74, p = 0.001), and with pH in both the dry (r = 0.63, p = 0.009) and wet seasons (r = 0.52, p = 0.041; Figure 3(d) ). These findings are consistent with those of Shamsudin et al. (2017) in Malaysia, who reported a strong correlation between E. coli and both pH (r = 0.9) and temperature (r = 0.76), highlighting the role of these physicochemical parameters in supporting microbial growth when sufficient nutrients are present. Similarly, a mean temperature of 32˚C was associated with increased E. coli counts in Kenya ( Powers et al., 2023 ), Bangladesh and Nepal, but decreased E. coli counts in Tanzania ( Charles et al., 2022 ). The pH of water can influence the survival and growth of coliform bacteria in water ( Suehr et al., 2020 ). In this study, the pH levels recorded from shallow wells fell within the World Health Organization accepted range of 6.5 to 8.5 for drinking water. Exceeding this range can negatively impact the survival and proliferation of coliform bacteria ( Hong et al., 2010 ). The positive correlation identified in this study aligns with findings from Nigeria by Olalemi et al. (2021) who demonstrated a similar positive relationship between E. coli and pH levels in well water. Additionally, the results agree with the work of Youssef et al. (2014) in Morocco, who reported a comparable positive correlation.

A sanitary survey was conducted for each of the selected shallow wells using the sanitary survey questionnaire ( Lloyd & Bartram, 1991 ). The total risk scores obtained were compared to the established risk categories ( Etang, 2000 ; WHO, 2017 ) and are presented in Table 4 . These scores remained unchanged throughout the study period. The results indicated that 50% of the wells were classified as high risk, while 12.5% were categorized as very high risk, with scores ranging from 9 to 10. Additionally, 18.75% of the wells fell into the intermediate risk category, whereas the remaining 18.75% were classified as low risk.

*Sanitary risk score: 9 - 13 = very high; 6 - 9 = high; 3 - 5 = intermediate; 0 - 2 = low ( Lloyd & Bartram, 1991 ).

The sanitary risk scores from shallow wells were plotted against E. coli counts, revealing a positive non-significant correlation during the dry season (p = 0.106) and the wet season (p = 0.434) ( Figure 4 ). The lack of statistical significance supports the limitation of sanitary surveys in predicting water quality as reported by Snoad et al. (2017) and Misati et al. (2017) . The findings of this study indicate that while the overall sanitary risk score did not correlate with E. coli presence, specific risk factors such as proximity of the well to the pit latrine (p = 0.025), exposed rope and bucket (p = 0.025) and poorly sealed well walls (p = 0.05) showed a significant association with E. coli presence. These findings contrast with those of Luby et al. (2008) , where individual risk factors did not show a significant association with microbial water quality, but agree with the findings of Ercumen et al. (2017) in Bangladesh, where individual components of the sanitary risk were associated with the presence of E. coli. Similarly, Tadele & Tekile (2025) identified five risk factors associated with the presence of coliforms. This suggests that while the composite sanitary risk score may not correlate with E. coli presence, specific sanitary risk factors can significantly be associated with microbial water contamination.

The shallow wells in the study area are hand-dug, and water is drawn manually using a rope and a plastic bucket. The repeated use of the rope, which often comes into contact with the ground before being submerged back into the water, can introduce bacteria from contaminated surfaces, further compromising water quality ( Gnimadi et al., 2024 ). Additionally, most wells in the study area are of the open type and lack protective brick or stone linings along the sides, a feature similarly observed by Grönwall et al. (2010) .

Physical barriers, such as concrete covers and slabs for shallow wells, may provide protection against surface runoff contaminated with fecal matter. However, in areas with a high water table, the risk of well contamination is heightened by underground seepage, particularly when well walls are inadequately sealed. This risk is further exacerbated when wells are located near pit latrines, as the use of pit latrines has been recognized as a contributing factor to groundwater contamination ( Elisante & Muzuka, 2016 ; Rivett et al., 2022 ; Hinton et al., 2024 ; Odewade et al., 2025 ). It is recommended that wells should be located at least 30 meters or more, away from pit latrines ( Tillett, 2013 ; Ngasala et al., 2021 ; Nenninger et al., 2023 ). In Ipusukilo, where the water table is high the proximity of wells to pit latrines may likely increase the risk of contamination from fecal matter. A study by Ashuro et al. (2021) in Ethiopia, found that the presence of latrines uphill from water sources was associated with fecal coliform contamination, likely due to the downward movement of contaminants through surface runoff or groundwater seepage. This implies that the landscape and hydrological dynamics play a critical role in the extent of fecal contamination.

The prevalence rates of water-related diseases based on the season are presented in Figure 5 . The most prevalent water-related disease during the dry season was diarrhea, with 245 cases reported. The most prevalent diseases in the wet season were cholera and diarrhea. The seasonal fluctuation highlights the significant public health risks posed by waterborne diseases, particularly during the wet season when microbial contamination peaks ( Siamalube & Ehinmitan, 2024 ). Water-related diseases pose a major threat to public health in areas with limited access to safe drinking water, sanitation, and hygiene ( Battersby et al., 2019 ; WHO, 2024 ). The consumption of contaminated drinking water is a major transmission route for pathogens responsible for diseases such as diarrhea, cholera, typhoid fever, and dysentery ( Mapingure et al., 2024 ). E. coli, a fecal indicator bacterium, is strongly associated with gastrointestinal infections, which can lead to severe dehydration and in extreme cases, mortality, particularly among young children and older individuals ( Okesanya et al., 2024 ). The increased prevalence of E. coli contamination in the wet season observed in this study is consistent with higher incidences of waterborne disease outbreaks during periods of heavy rainfall. This highlights the importance of implementing effective water treatment, improving sanitation, and implementing public health interventions to mitigate the risks associated with microbial contamination and ensure access to safe drinking water.