Next Article in Journal
Comprehensive Hydrological Analysis of the Buha River Watershed with High-Resolution SHUD Modeling
Next Article in Special Issue
Pathogen Contamination of Groundwater Affecting Drinking Water Quality with Potential Health Effects in Pavlodar Region, Kazakhstan
Previous Article in Journal
Limnological Characteristics and Relationships with Primary Productivity in Two High Andean Hydroelectric Reservoirs in Ecuador
Previous Article in Special Issue
On-Site Chlorine: A Promising Technology in Drinking Water Treatment in Santa Cruz, Bolivia
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 16
Issue 14
10.3390/w16142014
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Applying Microbial Source Tracking Techniques for Identification of Pathways of Faecal Pollution from Water Sources to Point of Use in Vhembe District, South Africa

by
Opelo Tlotlo Wryl Mochware
1,*,
Mathoto Lydia Thaoge-Zwane
2 and
Maggy Ndombo Benkete Momba
1,*
1
Department of Environmental, Water and Earth Sciences, Arcadia Campus, Tshwane University of Technology, P/B X 680, Pretoria 0001, South Africa
2
Department of Biotechnology and Food Technology, Arcadia Campus, Tshwane University of Technology, P/B X 680, Pretoria 0001, South Africa
*
Authors to whom correspondence should be addressed.
Water 2024, 16(14), 2014; https://doi.org/10.3390/w16142014
Submission received: 7 June 2024 / Revised: 11 July 2024 / Accepted: 12 July 2024 / Published: 16 July 2024
(This article belongs to the Special Issue Water Quality Monitoring and Public Health)
Browse Figures
Versions Notes

Abstract

:
A safe water supply is a necessity, but it remains one of the backlogs of services rendered in rural areas of developing countries. This leads to vulnerable communities using water from available sources that is unsafe as it is contaminated with faecal matter. Microbial source tracking (MST) methods are gold-standard techniques that detect the exact sources of faecal contamination. This study, therefore, tracked and identified the exact sources of faecal contamination from the catchment to the point of use in rural areas of Vhembe District Municipality. Collected water samples (n = 1048) were concentrated by membrane filtration for the enumeration and detection of E. coli, followed by DNA extraction. The extracted DNA was subjected to a quantitative polymerase chain reaction (qPCR) to track target host-specific Bacteroidales genetic markers from the water source to the point of use. Rivers and dams exhibited maximum E. coli counts of up to 90 CFU/100 mL during the wet season and up to 50 CFU/100 mL during the dry season. Due to the effective treatment of these water sources, no E. coli bacteria were detected in any of the sampled municipal drinking water treatment plants at the point of treatment, while this indicator bacterium was detected at the point of use (households), with a maximum of 4 CFU/100 mL recorded during both the wet and dry seasons. Overall, the most prevalent MST marker exhibited during the wet season was BacCan (dog-associated, 6.87%), followed by BacCow (cow-associated, 5.53%), while Pig-2-Bac (pig-associated, 2.48%) was the least prevalent. The most prevalent marker exhibited during the dry season was BacCan (5.34%), followed by BacCow, with Pig-2-Bac (1.72%) being the least prevalent. A positive correlation (r = 0.31, p = 0.001) was established between the presence of the MST markers and detected E. coli from water sources to the point of use. The knowledge of the faecal contamination attributes in both public and domestic domains will assist in developing prevention and control strategies.

1. Introduction

The past few decades have seen massive efforts to increase the provision of domestic water. Unfortunately, safe domestic water is still unavailable to many people, mostly those located in sub-Saharan Africa, South Asia, and East Asia [1]. Yet, safe drinking water is a necessity and a human right for all and can help curb the occurrence of waterborne diseases. Approximately 2 billion people have gained access to water around the world since 2000. The number of people who have gained access to safe drinking water have increased to 5.8 billion by 2020 [2]. Unsafe contaminated water and inadequate sanitation lead to the transmission of diseases such as diarrhoea, while poor water and sanitation management expose individuals to health risks. Approximately 84,2000 people are reported to die yearly from diarrhoea due to exposure to unsafe drinking water and sanitation. Safely managed water and sanitation services could prevent the 361,000 deaths of children under five years which occur yearly in Africa. While diarrhoea has been greatly associated with contaminated water [2], reports suggest that enteric pathogens are often harboured by different water bodies; these include rivers, dams, spring water and groundwater. In this regard, it is therefore crucial to determine the sources of contamination expeditiously for informed stances on corrective measures and water quality management [3].
South Africa is a water-scarce country and only a few of its main rivers are in good condition as 60% of them are being exploited [4,5]. The faecal contamination of water sources is still a problem, although efforts are being made to try and reduce it. This is due to the failure in identifying the actual source of faecal contamination. Animals and humans are both reservoirs of pathogenic microorganisms in water. Notable animal-to-human transmission sources include the inadequate treatment of wastewater, agricultural practices, septic tank leakages, domestic animals, and wildlife. The management of water bodies can be greatly improved if these sources of contamination are known [4] and strategic pollution prevention measures are developed and implemented.
It is important to note that monitoring water bodies for all enteric pathogens is not possible using culture-based methods. It is relatively costly and time-consuming when using these culture-based techniques to identify diverse pathogens, including protozoa, viruses, and bacteria, commonly found in aquatic environments. Equally, following an approach of detecting only one or a few microorganisms gives a wrong impression of the quality and safety of water for human usage as pathogens other than those tested for may be present [6]. Faecal indicator microorganisms have, therefore, been used to assess the quality of drinking water for more than a century. This approach is focused on measuring the concentrations of faecal indicator bacteria (FIB) such as faecal coliforms, Escherichia coli, and Enterococcus spp. However, there are some limitations to the traditional FIB methods [7]. These include that FIB are the inhabitants of the gastrointestinal tract of humans and other warm-blooded animals and they can survive and thrive outside the host. Studies have also shown a poor correlation between pathogens and FIB presence [8]. Even so, the key limitation is that FIB cannot discriminate between faecal sources, and therefore they do not identify the actual source of faecal contamination [9], while this aspect is pivotal for the delineation of public health risk and the implementation of mitigating procedures thereof. Researchers have, therefore, over many years developed a method which can unambiguously identify the actual source of faecal contamination in water bodies.
Microbial source tracking (MST) is a means of examining aquatic environments for the identification of sources of faecal contamination utilising some indicator bacteria [10]. Library-dependent MST (LD-MST) and library-independent MST (LI-MST) methods are available for the identification of the sources and origin of faecal contamination. However, the LD-MST method has shown limitations regarding the identification of the actual source, while the LI-MST techniques in previous studies have proven to extensively identify the source of contamination from environmental samples. These LI-MST methods are able to precisely quantify target sequences from the host-associated microorganisms using quantitative polymerase chain reaction (qPCR) assays [6]. The members of the order Bacteroidales which are anaerobic and found in the gastrointestinal tracts of both humans and animals are usually employed as the target because they are found in greater concentrations as compared to E. coli and they are host-specific. The core fundamental intention of MST is that some faecal pathogens are strongly related to a specific host; the MST methods employ genes belonging to the genus Bacteroides as markers. The recognised characteristics of these host-specific markers may be used as marker assays to identify the host. Host-specific 16S rRNA gene sequences are used in order to discriminate between different hosts, whether human or animal [11]. A specific gene marker is associated with a particular host and is expected to be present in all the host group associates [12]. Microbial source tracking assays have been designed with the use of quantitative polymerase chain reaction (qPCR) to identify bacteria obtained from a specific source [7].
Identifying the exact source of faecal contamination is imperative as human waste can enter water sources using various pathways, such as septic tank leakage, the discharge of untreated wastewater, surface runoff carrying human faeces due to open defaecation, dysfunctional sewer systems, and other sources. While enteric pathogens can be harboured by both animal and human faeces, studies have highlighted the significance of animal-to-human transmission, especially in developing countries, as animal husbandry is habitual in these countries [13]. This poses a major risk to humans as the end-users of the water as they are exposed to enteric pathogens [14]. The HF183 marker with its sequence found in the 16S rRNA gene of the cultured Bacteroides dorei was one of the first human-associated markers developed, and is now mostly found in the literature [4,6]. Amongst other animals, pigs are known to be sources of faecal contamination. To date, only one 16S rRNA gene marker that is pig-specific (Pig-2-Bac) is known to have been developed [15]. A study by Schriewer et al. [16] recommended BacCan as one of the two dog-associated assays. Microbial source tracking (MST) assays have been shown to be regionally specific and therefore, the performance of the MST markers must be assessed for any new geographic region. This is carried out by testing the specificity, sensitivity, and accuracy of the markers in the area of interest [17]. Boehm et al. [18] extensively evaluated MST assays targeting humans and animals; however, the significance of outcomes in other areas may potentially be limited. This is due to geographical differences, which may noticeably affect the sensitivity, accuracy, and specificity of the assays as the stool samples were sampled in a different area [17]. Studies using Bacteroidales for MST have been employed in some developing countries like Tanzania [19,20,21] and Kenya [22]. However, only limited information is available on the use of MST marker assays to identify the faecal pollution sources of water bodies in the Vhembe District, Limpopo Province, South Africa.
Pathways posing a threat to human health can be identified by assessing faecal pollution from numerous faecal–oral transmission pathways and determining the exact sources of faecal contamination, being either human or animal. These assessments will provide the evidence base for planning appropriate WASH interventions to improve conditions in specific areas. The vulnerability of communities to water contaminated with faecal matter is a matter of concern; exposure to faecal contamination occurs through the consumption of contaminated public drinking water sources and in households by contaminated hands and incorrectly storing water. The knowledge of the faecal contamination sources of both public and domestic domains will assist in developing effective prevention and control strategies [16].
The current study was, therefore, conducted to track the sources of faecal pollutants from different water sources to the point of use using host-specific Bacteroidales qPCR assays. To achieve the aim of the study, two objectives were set: firstly, to identify faecal pollution of water sources by detecting and enumerating E. coli as a model FIB; and secondly, to identify the pathways of faecal pollution between the water source and the point of use by microbial source tracking methods.

2. Materials and Methods

2.1. Study Design

Figure 1 illustrates the design of the study conducted in the selected rural areas of the Vhembe District Municipality. Prior to the commencement of the study, rural communities around the Vhembe District were visited. Different water sources were identified and various activities happening around the water sources were extensively observed. Only accessible communities with more than two different water sources available were considered as study areas.

2.2. Study Site and Population Description

Limpopo Province is one of the poverty-stricken areas in South Africa and consists mainly of rural communities in which the majority of people are subsistence farmers. Human and animal activities in this area contribute greatly to the contamination of the rivers [23]. The Vhembe District Municipality is located in the Limpopo Province and comprises four (4) local municipalities, namely Thulamela, Makhado, Musina, and Collins Chabane. According to Stats SA, in 2016 [24] the total population of this district comprised 1,393,949 people, with a land area of 27,969,148 km2. The district experiences a large number of water outages with a substantial number of communities relying on unsafe surface water [25].

2.3. Scientific Ethics Clearance and Informed Consent

An ethical clearance was granted by the Faculty Committee for Research Ethics (FCRE) of the Tshwane University of Technology before the commencement of the study. The study proposal was presented to Vhembe District Municipality representatives and permission was granted to conduct the study around the district in all four local municipalities, namely Musina, Thulamela, Collins Chabane, and Makhado. Permission to conduct the study in the respective communities was also obtained from chiefs and tribal authorities. Household members were also given a brief verbal explanation of the study and consent was given prior to conducting the study.

2.4. Collection of Water Samples from Different Sources

In total, 1048 water samples were collected on rainy days (wet season) between February and May 2021 and on days with no rain (dry season) between June and August 2021 in the communities of five villages in the Vhembe District, namely Tshakhuma, Tshidzini, Tshilapfene, Dididi, and Tshivhulani. The samples were collected from different water sources, which included communal taps, yard taps, springs, wells, boreholes, dams, and rivers used by the above-mentioned rural communities. The sampling points are outlined in Figure 2 and Figure 3. The water samples were collected aseptically according to the South African National Standard [26] guidelines for water sampling. A total volume of 1 L for each water source and point of use was aseptically collected from different water sources using sterile Duran® Schott glass bottles. For treated (chlorinated) water samples, glass bottles containing 120 mg of sodium thiosulphate (Na2S2O3) were used to neutralise any residual chlorine in the water samples [27]. Each sample was labelled with its unique description and transported on ice to the research laboratory at the Tshwane University of Technology for analysis. All the samples were analysed in triplicate within 24 h.

2.5. Microbial Water Quality Analysis

2.5.1. Detection and Enumeration of E. coli

Membrane filtration techniques were followed for the detection and enumeration of E. coli from the water samples as described in Standard Methods for the Examination of Water and Wastewater [28]. The water samples were filtered through a membrane filter with a pore size of 0.45 µM (Millipore™, Merck, Modderfontein, South Africa). The membrane filter was then placed onto the Chromocult® (Merck) agar plate, followed by incubation at 36 ± 1 °C for 24 h. The experiments were conducted in triplicate for all the samples. All the violet colonies were counted and recorded as colony-forming units (CFUs) per 100 mL.

2.5.2. Detection of Faecal Sources of Contamination in Water Sources Using Host-Specific Bacteroidales qPCR Genetic Markers

Sample Preparation and Genomic DNA Extraction

A water sample of 500 mL for final water or 100 mL for raw water was filtered through a membrane filter with a pore size of 0.22 µm (Millipore™, Merck, Modderfontein, South Africa) [29]. The filters were placed in test tubes containing 50 mL of sterile phosphate-buffered saline solution (PBS). The test tubes were vortexed to dislodge bacteria from the filter paper. The solution with the re-suspended cells was centrifuged at a speed of 2000 rpm for 15 min and at 4 °C. The supernatant was discarded and this procedure was repeated three times in order to concentrate the cells. The pellet containing the cells was subjected to DNA extraction as described elsewhere [29]. Bacterial genomic DNA was extracted from the pellets using the Quick-DNA Fecal/Soil Microbe Microprep Kit (Zymo Research, Inqaba Biotechnical Industries, Pretoria, South Africa. The extraction was performed following the manufacturer’s instructions and the recommendation by Okabe (2007) [22].

2.5.3. Tracking and Detection of Sources of Faecal Contamination by Host-Specific Bacteroidales Genetic Markers Using qPCR

Validation of Bacteroidales Genetic Marker Assays in the Study Area

The preliminary selection of MST markers was based on the animals observed within the study area. Previous assays (Table 1) were selected to corroborate their relevance in the study area.
The selection of the most appropriate assay in terms of performance was based on the sensitivity, specificity, and accuracy of the assays as described elsewhere [31]. The equations given below were used to calculate these parameters.
Sensitivity = TP/(TP + FN) × 100
where TP represents the number of samples that are true positives, while FN represents the number of false-negative samples.
Specificity = TN/(TN + FP) × 100
where TN represents the number of samples that are true negatives and FP represents the number of false-positive samples.
Accuracy is determined by dividing the total number of the predicted true positives and true negatives by the total number in the data set. The calculation was conducted using Equation (3) as given below.
Accuracy = (TP + TN)/(TP + FP + TN + FN) × 100
In total, 51 DNA samples extracted from the faecal samples from different hosts obtained in the Tshwane University of Technology Water Research Laboratory were used. One human-specific assay (HF183), ruminant-specific assay (BacCow), pig-specific assay (Pig-2-Bac), chicken-specific assay (Cytb), and dog-specific assay (BacCan) were tested in this study using the CFX96 Touch Real-Time PCR System (Bio-Rad, Hercules, CA, USA). Briefly, a reaction mixture of a total volume of 25 µL containing 7.5 µL of PCR grade water, 12.5 µL of GoTaq® Probe qPCR Master Mix, 1 µL of both forward and reverse primers, 0.5 µL of probe, and 3 µL of DNA template was used. The qPCR cycling conditions were as follows: GoTaq® DNA polymerase activation at 95 °C for 2 min, followed by 40 cycles at 95 °C for 15 s, and annealing and extension at 60 °C for 1 min. All the reaction assays were run in triplicate.

2.5.4. Standard Curves for Performance and Interpretation of qPCR Results

A quantitative polymerase chain reaction was employed to assess and verify the cross-reactivity of microbial source marker assays with target and non-target human and animal faecal samples. A standard curve was generated using the 10-fold serial dilutions of plasmid DNA with the sequence of interest, and the limit of detection (LOD) was determined for each assay. The lower limit of quantification (LLOQ) for each individual MST assay was determined as the average cycle threshold (Ct) value correlating to the minimum concentration within the linear spectrum of detection where at least 95% of the dilution repeats were detected. The LOD was determined by rounding the LLOQ to the closest integer number. Further standard curve parameters were percent efficiency-calculated using the slope of the curve and the y-intercept. The equations below were used to calculate the above-mentioned parameters.
Efficiency % = [10(−1/slope)] − 1
If the Ct value obtained was within the quantification range, the sample was regarded as positive. If the Ct value was recorded outside the quantification range or if the target of interest was not identified, the sample was deemed negative. The quantity of gene copy numbers was calculated from standard curves and recorded as gene copies per nanogram (ng).
Quantity = 10^(Cq − b)/m
where b is the y-intercept, m is the slope of the linear regression, and Cq is the PCR cycle number.
The results were normalised to Log10 gene copies/mL [8,16]. A regression analysis was conducted to eliminate outliers (by removing the Ct values with a residual value larger than +3 or smaller than −3) [16].

2.5.5. Data and Statistical Analysis

A descriptive data analysis of the detected E. coli and MST markers was performed using Microsoft Excel or IBM SPSS Statistics 22.0. The concentrations of faecal source markers were log-transformed and no detection samples were assigned as zero (0). Sensitivity, accuracy, and specificity were used to validate the internal matrices for each genetic marker. The abundance and distribution trends of the markers were analysed using the analysis of variance (ANOVA). The correlation between the detected genetic markers and E. coli was calculated using Pearson’s correlation coefficients (r). The t-test was used to compare the distribution of the MST markers between the dry and the wet season. The statistical significance of the abundance of the MST markers was determined using the t-test regression analysis. The Kruskal–Wallis test was employed to assess the differences in the E. coli counts of the various water sources between the different villages. The Mann–Whitney test was employed to compare the differences in the E. coli counts of the various water sources between the wet and dry seasons.

3. Results

3.1. Microbiological Quality Assessment of Water Samples

3.1.1. Detected E. coli Concentrations in Different Water Sources

In total, 1048 water samples were analysed for the enumeration and detection of E. coli using culture-based methods. All the river and dam water samples (n = 96) were contaminated with E. coli during the wet and dry season. The highest concentration of E. coli (90 CFU/100 mL) during the wet season and the dry season (50 CFU/100 mL) was recorded in the Mutshindudi River, as illustrated in Figure 4. The lowest concentration of E. coli was detected in the Ngwedi River upstream during both the wet season (8 CFU/100 mL) and the dry season (6 CFU/100 mL). The E. coli concentrations at the point of treatment were slightly lower at the Tshakhuma Water Treatment Plant (60 CFU/100 mL) as compared to the Tshakhuma Dam (86 CFU/100 mL), while there was a slight difference between the concentrations recorded at the Nandoni Dam (48 CFU/100 mL) and raw water prior to treatment at the Nandoni Water Treatment Plant (41 CFU/100 mL). After treatment at the point of treatment (water treatment plants), no E. coli counts (0 CFU/100 mL) were recorded in any of the municipal tap water samples, suggesting the effective water purification processes of raw water at all the treatment plants. In contrast, contamination was recorded at 15/50 (30%) of the households receiving municipal water from these treatment plants in the wet season. The data in Figure 4, Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9 suggest that the contamination of municipal drinking water occurred in the households, the end-users of municipal water.
The highest recorded concentration of E. coli in the municipal tap water in the households was 4 CFU/100 mL in both the wet and dry seasons. In most households, 0 CFU/100 mL was recorded in both the wet (35/50 (70%)) and dry (8/50 (16%)) seasons. All the spring water samples harboured E. coli in the wet season, while only two (2) did in the dry season. In contrast, the highest E. coli concentration was recorded in the dry season (21 CFU/100 mL) in the Dididi spring. Furthermore, the highest E. coli concentration (8 CFU/100 mL) was recorded in the spring water in storage containers in the Dididi and Tshilapfene households. In only one household (1/8) in Tshivhulani using spring water, no E. coli contamination was recorded. The lowest concentration of E. coli was recorded in borehole water stored in JoJo tanks in the households; only 3/18 (17%) harboured E. coli during both the wet and dry seasons. In most households (83% and 78%) using borehole water, 0 CFU/100 mL was recorded during both the wet and dry seasons, respectively, while E. coli was detected in all the household containers storing river water during the dry season. The highest concentration of E. coli was recorded in a household in Tshivhulani which uses Mutshindudi River water (90 CFU/100 mL). Figure 4, Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9 below show the concentrations of E. coli recorded in the different water sources in both the wet and dry seasons in the village study cohorts.
Table 2 below shows a comparative analysis between the E. coli concentrations in the water samples from the different water sources. In the households, the container-stored river water exhibited the highest concentrations of E. coli (22 ± 31.87 CFU/100 mL), followed by spring water (2 ± 2.47 CFU/100 mL) in the wet season. The lowest E. coli concentration was recorded in the dry season in the municipal tap water samples at the dwellings (0 CFU/100 mL). Mean differences in counts were significant among all the water sources (p ≤ 0.05), except for spring water (p ≥ 0.05).

3.1.2. Quality Assurance and Validation of MST Marker Assays

Perplexity may be experienced during the quantification of the markers using qPCR. To obtain reliable results, the assays must be optimised to achieve the utmost sensitivity, specificity, and accuracy [32]. The assay-tested amplification efficiencies ranged between 85% and 94%. The performance of the cow-associated assay (BacCow) was assessed with 15 cow faeces samples and the target marker was detected in 93% of the samples (14/15). High specificity was obtained for Pig-2-Bac and HF183 (100%) while the lowest specificity was recorded for BacCan (80%). The HF183 marker was the least accurate (89%) compared to animal markers as illustrated in Figure 10. Table 3. illustrates the validation standard curve performance of host-specific Bacteroides markers (BacCow, Cytb, HF183, Pig-2-Bac, BacCan) using standard serial dilutions of a plasmid DNA with known concentration.

3.2. Prevalence of Sources of Contamination Detected in Study Villages

Overall, both animals and humans were found to be the sources of the faecal contamination of water sources in the target villages, as can be seen in Table 4 and Table 5 and in Figure 5, although the prevalence of the associated host-specific Bacteroidales gene markers varied from one village to another and during the seasons. In general, Tshidzini exhibited the highest sources of contamination compared to the other villages during both the wet and the dry seasons, with an average prevalence of 32.69% and 23.08%, respectively (Table 4 and Table 5). Tshakhuma displayed the lowest sources of contamination in its water sources during the dry seasons with a total prevalence of 11.61% and Tshilapfene was the highest during the wet season (15.63%). The most prevalent faecal source was found to be dogs (total average prevalence of 6.87%), which predominated in all the study cohort communities during both seasons, with Tshidzini displaying the highest rate (9%). Overall, pig faecal source markers (2.48%) were the least prevalent, with the exception of Dididi Village during winter (5%). The variation in the presence of the different sources of faecal contamination in the cohort study areas was statistically significant (p < 0.05) in all five study areas during the wet season (Table 4).
The prevalence of the MST markers was relatively lower in the dry season as compared to the wet season. Overall, faecal sources were identified in 17% of the water samples collected in the five (5) target study areas. All the faecal source markers were present in all the study communities, although the chicken marker was most predominant in Tshivhulani (5.36%). The pig marker (average 1.72%) followed by the human marker (average 2.6%) were the least detected in all the study villages in the dry season, as shown in Table 5. The variation in the presence of different sources of faecal contamination at the cohort study areas was statistically significant (p < 0.05) in all five study villages in the dry season.

Prevalence of Detected Host-Specific Bacteroidales Gene Markers in Water Sources Used in the Target Study Villages

Overall, both the human and animal sources of faecal contamination were detected in water sources, at the catchment level, in the municipally treated water at the household level, in spring water, in hand-dug well water, and during the storage of spring water at the household level, although the prevalence of the markers varied from one water source to another, as shown in Table 6. In the catchments, the cow marker (26%) was found to be the most prevalent, followed by the dog marker (19%). The chicken marker displayed the lowest prevalence (3%). Only three of the target markers were detected in the raw water prior to the treatment at the treatment plants with the cow marker (18%) again being the most prevalent. No sources of contamination in terms of the presence of the markers were detected in any of the municipal water samples post-treatment at the treatment plants and neither in the communal tap water samples. At the household level, the municipal container-stored water exhibited the prevalence of the markers, with BacCan (4%) being the highest and HF183 (0.3%) the lowest. The dog marker exhibited high prevalence rates in water samples of the hand-dug wells (50%), spring water (21%), and spring water in containers in the households (7%). Hand-dug well water also displayed a relatively high BacCow prevalence (38%) compared to other water sources. Pearson’s correlation was employed to assess the variance in faecal source prevalence in the wet and dry seasons. A strong correlation was observed for spring water (r = 0.87), while a weak correlation was observed for spring water in containers in the households (r = 0.22). The variance in the recorded sources of faecal contamination at different water sources was statistically significant except at hand-dug wells (p = 0.17).

3.3. Relationship between E. coli Concentrations and Detected MST Markers

As can be seen in Table 7 and Table 8, correlations were established between the prevalence of E. coli and the faecal source markers during the wet and dry seasons. Overall, E. coli prevalence positively and strongly correlated with BacCan (r = 0.93), Cytb (r = 0.90), and HF183 (r = 0.70) in the wet season, while BacCan (r = 0.45) and Pig-2-Bac (r = 0.35) had a moderate and weak correlation, respectively. In the dry season, E. coli prevalence positively and strongly correlated with BacCan (r = 0.64), Cytb (r = 0.54), BacCan (r = 0.70), and HF183 (r = 0.63) while a moderate correlation was observed between E. coli and Pig-2-Bac (r = 0.41), as shown in Table 8.

3.4. Distribution and Transmission Pathways of Sources of Contamination between Water Sources and Point of Use per Village

Of the overall sources of faecal contamination identified in Tshakhuma, three (3) (human, cow, and dog) were detected in the Tshakhuma Dam, which is the main water supply source for the Tshakhuma WTP and the municipal taps. Figure 11 shows that the same faecal sources were also detected in the raw intake water of the Tshakhuma Water Treatment Plant (WTP) prior to treatment. No sources of faecal contamination were observed in the WTP final water, suggesting that the treatment processes were efficient at the plant. However, different faecal sources were identified at the municipal water point of use (households), of which 60% of the treated water samples showed pig faecal contamination, and 40% of the samples showed chicken faecal contamination; these findings suggest that contamination could have occurred at the –point of use as some of these households keep domestic animals in their yards. Of the three (3) sources of faecal contamination detected in Tshakhuma, the main source of the contamination of spring water in the households was found to be dogs, with 72% of the samples testing positive for the BacCan marker. However, no contamination source was found in the households using borehole water.
In Tshivhulani, as shown in Figure 12, two (2) sources of faecal contamination were identified in the Vondo Dam, which is the main municipal raw water supply source. With the exception of the Phiphidi WTP and the communal tap, cows, dogs, pigs, and chickens were found to be the main sources of the faecal contamination of water used by the community of Tshivhulani. Some households in Tshivhulani use water directly from the Mutshindudi River where cow, dog, and human faecal sources were identified. Two of the sources of faecal contamination identified in the Mutshindudi River, namely cow and human, were also detected in the end-user water samples (in the households) when using the river as the main water source. More sources of contamination were identified at the household level for those who also use spring water as their main water supply. Of all the different main water sources used by households, the largest number of the faecal sources of contamination were detected in spring water.
As can be seen from Figure 13, the residents of Tshilapfene also depend on surface water from the Vondo Dam and Phiphidi Water Treatment Plant, and spring and borehole water. Overall, these water sources were contaminated with animal faeces (cows, dogs, pigs, and chickens) and human faeces with the exception of the treated water. Faecal contamination sources from cows and dogs were identified at the main municipal raw water supply source, which is the Vondo Dam. Raw water from the Phiphidi WTP displayed three (3) sources of faecal contamination (cows, dogs, and humans) at the point of treatment; however, after treatment, none of these were detected, implying the effectiveness of the water treatment processes. In spite of its absence at the point of treatment, the BacCan marker (dog) reappeared in the dwellings at the point of use. Of the five (5) target faecal contamination sources, four (4) Bacteroidales gene markers (for dogs, pigs, humans, and chickens) were identified in the households of the communities using spring water. Spring water also exhibited the presence of dog and chicken markers.
All five (5) target sources of faecal contamination were detected in the Luvuvhu River, which is the raw water intake source of the municipal water treatment plant, as shown in Figure 14. After the treatment of this intake river water at the Xikundu WTP, no faecal contamination sources were detected in the final water. However, at the household level, all five sources of faecal contamination reappeared in the end-user water samples (at the dwellings). A similar observation was noted in the Ngwedi River water, where five target sources, also observed at the catchment level, reappeared in the end-user water samples (in the households) using the very same untreated Ngwedi River water. Fewer sources of faecal contamination were observed in the households using spring and borehole water, as only one faecal source was identified in the respective households (cows and dogs, respectively).
As can be seen in Figure 15, the Luvuvhu River also supplies intake water to the Nandoni WTP. After treatment, the final water is supplied to the residents of Dididi Village. Results revealed four (4) different faecal contamination sources (cows, pigs, humans, and chickens) from the Luvuvhu River to the municipal tap water in Dididi. In the Nandoni Dam, which also receives water from the Luvuvhu River, three main sources of faecal contamination were identified (cows, dogs, and humans). Of these three faecal sources detected in the Nandoni Dam, cow and human faecal contamination was detected in the Nandoni WTP intake raw water. Cow and human faecal sources were consistently present between the main water source and the water treatment plant prior to treatment. No faecal contamination source was detected in the Nandoni WTP final water, although the households receiving this municipal water exhibited three faecal contamination sources (cows, pigs, and dogs). The sources of faecal contamination detected in the households supplied by the municipal water treatment plant were not entirely the same as those in the main water source (Figure 15). Almost all four (4) target faecal sources were present in the piped spring water that did not undergo treatment; of all the faecal contamination sources, three (3) were detected in the households (cows, dogs, and pigs) in addition to the chicken faecal source that was not initially present in the spring water. While these faecal contamination sources were consistently present between the spring (sources) and the end-user water samples (in the households), none of them were identified in the borehole water samples of the households.

4. Discussion

Vigorous agricultural practices predominantly existing in the rural areas of developing countries may have a detrimental impact on the quality of both groundwater and surface water, and South Africa is not an exception [33]. The contamination of water sources as a result of agricultural activities due to the use of fertilisers has also been reported by previous investigators in the Vhembe District Municipality and this may have a negative impact on the quality of water supplies [34,35]. Although agriculture contributes to the income of rural communities for their survival, it has been observed in India, through microbial source tracking methods, that the faecal contamination of domestic water sources by animals is more prevalent than human contamination [36]. By using these methods in this study, it was possible to track the sources of faecal contamination from the various water sources (river, dam, spring, and hand-dug well) to treated water and/or the household level.
The identification of E. coli was employed in the present study as a model faecal indicator bacterium (FIB) aimed at assessing the faecal contamination of water sources in target areas. This bacterium serves as a better indicator in comparison with faecal coliforms. Fewer instances of false positives are reported to be experienced with this indicator [37]. Moreover, E. coli serves as an indicator of the microbial quality of the water and signals the presence of recent faecal contamination [38]. A substantial increase in E. coli concentrations during the rainy season has been illustrated in this current study (Table 2); this may be due to the surface runoff during storm events that can be considered as the classical source for the contamination of surface water bodies. Previous studies have also pointed out the relationship between E. coli concentrations and rainfall events. The assumption puts forward that the survival of E. coli in water bodies is influenced by water levels, moisture, and the availability of energy in these environmental habitats [39]. This clearly justifies higher concentrations of E. coli found in catchments (rivers and dams) during the wet season where the maximum recorded E. coli concentration was 90 CFU/100 mL, while a minimum E. coli concentration of 50 CFU/100 mL was observed during the dry season during the (Figure 4). Higher concentrations of E. coli were also recorded in the wet season ((85 CFU/100 mL) in the households utilising untreated river water, while a maximum concentration of only 34 CFU/100 mL was detected in the households utilising untreated river water in the dry season. Furthermore, the lowest concentrations (0 CFU/100 mL) were recorded in the municipal treated water in the households and river water in the households in the wet season (Figure 5). However, E. coli concentrations in the dry season may have been influenced by favourable temperatures during this season [40]. As observed in Figure 4, a higher maximum concentration was recorded in the dry season (21 CFU/100 mL) than in the wet season (10 CFU/100 mL) at the Dididi spring.
For the past century, faecal indicator bacteria (FIB) such as E. coli have been considered as a tool for assessing the microbial quality of drinking water. Unfortunately, FIB cannot distinguish between faecal contamination from humans and animals. Furthermore, FIB have been reported to be present in the environment without faecal contamination and also poorly correlate with the presence of enteric pathogens [41]. Consequently, the use of faecal markers associated with specific hosts for microbial source tracking (MST) has demonstrated their capability to differentiate between the human and non-human sources of faecal contamination in an aquatic environment. Variations, such as those caused by dietary patterns on gastrointestinal tract microbiota [42] or variations in geography [43], have the potential to have a substantial effect on the efficiency of MST markers, as evidenced through the various validation studies carried out previously [18,21,29,44]. As a result, prior to the application of the MST markers, it was critical to assess the performance of human and animal faecal markers in this geographical area. All the assays in this study successfully showed an efficiency above 85%. In this study, HF183 showed great sensitivity (80%), specificity (100%), and accuracy (89%) (Figure 10). Although the assay was highly specific, the human-associated markers have previously shown substantial cross-reactivity with chicken faecal samples in South Asian investigations [29,45]. In this study, the HF183 marker was shown to be suitable in the study area as its specificity was 87%. The BacCow marker was initially developed as a cattle faecal source marker [11], even though a California-based validation study demonstrated cross-reactivity with other faecal samples acquired from other ruminants and non-ruminants such as the horse [46]. In this study, BacCow was detected in 14/15 of the cow faecal samples analysed; however, there was some cross-reactivity with (2/10) the chicken faecal samples (Figure 10). This is consistent with the study outcomes of previous investigators [8,35]. The BacCow marker was therefore applicable as a cattle-associated marker in the Vhembe District Municipality. Although various pig microbial source tracking markers have been established, Pig-2-Bac has consistently performed satisfactorily in trials. In this study, Pig-2-Bac had a sensitivity of 88%, a specificity of 100%, and an accuracy of 93% (Figure 10). No cross-reactivity was experienced with faecal samples from other animals, even though previous studies conducted in China and Thailand showed higher sensitivity and lower specificity [47]. A study in indoor-household environments of the Peruvian Amazon [30] also showed higher sensitivity (100%) and lower specificity (88.5%) as compared to those of the current study. The BacCan dog-associated marker cross-reacted with the cow and chicken faecal samples, resulting in a much lower acceptable specificity (80%). In a comparable vein, this specific marker had a specificity of 45% in Nepal, although it was 97% in India and Singapore [29,37]. The Cytb assay marker for chicken showed a cross-reaction with one human faecal sample (Figure 10). This resulted in a sensitivity of 88%, which aligns with the results of a study conducted by Schiaffino et al. [30].
Determining the source(s) of faecal contamination in water sources is required for remediation measures and the evaluation of the possible health risks relating to faecal microorganisms [48]. These water sources may be exposed to faeces from different sources such as surface runoff containing human faeces due to open defaecation or animal faecal matter, and human sewage contamination due to the discharge of inadequately treated wastewater into water sources. During the rainy season, surface runoff that carries animal and human faecal matter can elevate the concentrations of faecal indicator bacteria [49]. Microbial source tracking was therefore employed in this study to assess the presence of faecal pollution in different water sources and analyse the pathways of faecal contamination between water sources and points of use. It was shown that all the water treatment plants successfully decontaminated their intake raw water as no markers were detected in their final water (Table 6). This could be the result of an effective chlorination process at the point of treatment. This suggests that contamination took place at the point of use as the presence of domestic animals in the households correlated well with the detected sources of faecal pollution markers in the households. Furthermore, the World Health Organization [2] highlights that animal faecal matter is one the major sources of faecal pollution, which results in 80% of the diseases in developing countries. Liang et al. [50] recommend that applying MST markers alongside conventional indicator organisms to evaluate water quality could provide enhanced methods for predicting pathogen occurrence in ambient waters. A study by Waso et al. [51] assessed the correlation between MST markers and faecal indicator bacteria. The study found that the human marker HF183 significantly and positively correlated with E. coli (p = 0.024). This corresponds with findings in the current study, where the presence of all the target faecal source markers (HF183, BacCow, Cytb, Pig-2-Bac, BacCan) correlated positively with the presence of E. coli (r = 0.70, r = 0.93, r = 0.90, r = 0.35, r = 0.43, respectively) in the water samples. The proportion of water sources with detected animal and human faeces contamination differed greatly between the water sources (Table 7). The inflow of stormwater runoff carrying faecal matter is likely a trigger, as evidenced by the reality that the detection frequency of faecal markers was the highest in surface water, followed by spring water and finally the lowest in private borehole water (Table 6). Broadly, this ranking of water sources in terms of microbiological quality was expected since these water sources are more isolated from the human environment, and hence less prone to faecal contamination than surface water. Surface freshwater sources are the most vulnerable to various zoogenic and human-caused hazards, with spatio-temporal factors influenced by climate change. Surface waters are more vulnerable than other water systems to the massive, direct, and immediate impacts of several of these pollutants [52]. This result is in line with the findings of this study as the highest frequency of faecal source contamination was recorded in catchments (Figure 4). Additionally, the Ngwedi River in Tshidzini harboured all the target sources of faecal contamination. Animal and human faecal contamination was not detected in municipal treated water at the point of treatment (water treatment plants) but was detected in municipal treated water samples in the dwellings in the current study. These findings point to remote hotspots where dwelling exposure to animal and human-derived faecal pathogens is spatial and or temporally aggregated. It is important to emphasise that the majority of human-caused faecal pollution is due to the impact of underground structures, such as defects in sewer infrastructure as well as seepage from faulty septic tanks affecting adjacent wells, as verified by MST using qPCR assays [20]. However, in this study, the prevalence of target faecal source markers was relatively low (Table 4 and Table 5). The human-associated (HF183) and cow-associated (BacCow) markers were found in 36% and 82% of the surveyed households, respectively, in a study of community water sources and home-stored water in a rural area of Odisha, India [29]. These findings suggested that drinking water supplies were contaminated by both animal and human waste, with cows being thought to be the primary source. Sub-Saharan African countries reported similar findings in drinking water [53]. Similar results were observed in this study, although the dog-associated markers were found to be the most prevalent (4%) in municipal tap drinking water samples in the dwellings.

5. Conclusions

The current study investigated and tracked the sources of faecal pollutants from different water sources to the point of use, using host-specific Bacteroidales qPCR assays (BacCow, BacCan. Cytb, Pig-2-Bac, and HF183) in five villages of the Vhembe District Municipality. The results revealed that disparities existed in the study villages as some households had no access to safe drinking water and relied on unsafe water sources such as rivers and hand-dug wells, while the findings also demonstrate that these sources harboured high concentrations of E. coli. Furthermore, despite the effective water treatment processes of municipal water at all the treatment plants, faecal source markers were detected in some of the samples of the municipal tap water in the dwellings. This suggests that contamination does take place in the households. The presence of host-associated faecal markers correlates with the presence of faecal indicator organisms and the presence of domestic animals in the dwellings. This study found that most of the communities have access to piped municipal water, but some areas experience frequent municipal water supply interruptions (e.g., Dididi Village), and therefore residents have to resort to any source of water in their vicinity. There is, thus, a dire need for the provision of potable water to all the people living in rural areas in order to prevent the spread of waterborne diseases. Moreover, in view of the finding that faecal contamination takes place in the homestead, there is still a critical need for awareness and education on proper sanitation and hygiene practices and the safeguarding of the water provided in rural areas. The findings derived here are useful for authorities who are responsible for the surveillance and control of drinking water quality to minimise the health risks caused by pathogens by identifying the sites where intervention is required.

Author Contributions

M.N.B.M. conceived the project; M.N.B.M. and O.T.W.M. designed the experiments; O.T.W.M., M.L.T.-Z. and M.N.B.M. performed the experiments and analysed data. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported by the South African Research Chairs Initiatives (SARChI) in Water Quality and Wastewater Management under the leader of MNB, funded by the Department of Science and Technology, as administrated by the National Research Foundation (UID87310). Additional funding was received from Tshwane University of Technology. Opinions expressed and conclusions arrived are those of the authors.

Data Availability Statement

All data are included in artcicle.

Acknowledgments

We thank Ingrid G. Buchan (Pr.Sci.Nat.) for language editing. Sample collection was carried out with the assistance of Arinao Murei, Barbara Mogane, Dikeledi Prudence Mothiba, Mulalo Mudau, Jeridah Matlhokha Sekgobela, and Ndamulelo Musumuvhi. The study site map was created by Phumlani Zwane.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Dungumaro, E.W. Socioeconomic differentials and availability of domestic water in South Africa. Phys. Chem. Earth (Parts A/B/C) 2007, 32, 1141–1147. [Google Scholar] [CrossRef]
  2. World Health Organization (WHO). Water, Sanitation, Hygiene and Health: A Primer for Health Professionals (No. WHO/CED/PHE/WSH/19.149); World Health Organization: Geneva, Switzerland, 2019. [Google Scholar]
  3. Shahra, E.Q.; Wu, W. Water contaminants detection using sensor placement approach in smart water networks. J. Ambient Intell. Humaniz. Comput. 2020, 14, 4971–4986. [Google Scholar] [CrossRef]
  4. Bernhard, A.E.; Field, K.G. Identification of nonpoint sources of fecal pollution in coastal waters by using host-specific 16S ribosomal DNA genetic markers from fecal anaerobes. Appl. Environ. Microbiol. 2000, 66, 1587–1594. [Google Scholar] [CrossRef] [PubMed]
  5. Crookes, C.; Hedden, S.; Donnenfeld, Z. A delicate balance: Water scarcity in South Africa. ISS S. Afr. Rep. 2018, 2018, 1–24. Available online: https://www.wrc.org.za/wp-content/uploads/mdocs/ISS_A%20delicate%20balance.pdf (accessed on 14 May 2023).
  6. Harwood, V.J.; Staley, C.; Badgley, B.D.; Borges, K.; Korajkic, A. Microbial source tracking markers for detection of fecal contamination in environmental waters: Relationships between pathogens and human health outcomes. FEMS Microbiol. Rev. 2014, 38, 1–40. [Google Scholar] [CrossRef] [PubMed]
  7. Hinojosa, J.; Green, J.; Estrada, F.; Herrera, J.; Mata, T.; Phan, D.; Pasha, A.T.; Matta, A.; Johnson, D.; Kapoor, V. Determining the primary sources of fecal pollution using microbial source tracking assays combined with land-use information in the Edwards Aquifer. Water Res. 2020, 184, 116211. [Google Scholar] [CrossRef]
  8. Ahmed, W.; Sritharan, T.; Palmer, A.; Sidhu, J.P.S.; Toze, S. Evaluation of bovine feces-associated microbial source tracking markers and their correlations with fecal indicators and zoonotic pathogens in a Brisbane, Australia, reservoir. Appl. Environ. Microbiol. 2013, 79, 2682–2691. [Google Scholar] [CrossRef]
  9. Field, K.G.; Samadpour, M. Fecal source tracking, the indicator paradigm, and managing water quality. Water Res. 2007, 41, 3517–3538. [Google Scholar] [CrossRef] [PubMed]
  10. Ko, H.Y.; Cho, K.; Park, S.; Kim, J.H.; Kang, J.H.; Jeong, Y.S.; Choi, J.D.; Sin, Y.; Lee, C.; Ko, G. Host-specific Bacteroides markers-based microbial source tracking in aquaculture areas. Microbes Environ. 2018, 33, 151–161. [Google Scholar] [CrossRef]
  11. Kildare, B.J.; Leutenegger, C.M.; McSwain, B.S.; Bambic, D.G.; Rajal, V.B.; Wuertz, S. 16S rRNA-based assays for quantitative detection of universal, human-, cow-, and dog-specific fecal Bacteroidales: A Bayesian approach. Water Res. 2007, 41, 3701–3715. [Google Scholar] [CrossRef]
  12. Seurinck, S.; Defoirdt, T.; Verstraete, W.; Siciliano, S.D. Detection and quantification of the human-specific HF183 Bacteroides 16S rRNA genetic marker with real-time PCR for assessment of human faecal pollution in freshwater. Environ. Microbiol. 2005, 7, 249–259. [Google Scholar] [CrossRef] [PubMed]
  13. Fuhrmeister, E.R.; Ercumen, A.; Pickering, A.J.; Jeanis, K.M.; Crider, Y.; Ahmed, M.; Brown, S.; Alam, M.; Sen, D.; Islam, S.; et al. Effect of sanitation improvements on pathogens and microbial source tracking markers in the rural Bangladeshi household environment. Environ. Sci. Technol. 2020, 54, 4316–4326. [Google Scholar] [CrossRef] [PubMed]
  14. Kapoor, V.; Pitkänen, T.; Ryu, H.; Elk, M.; Wendell, D.; Santo Domingo, J.W. Distribution of human-specific Bacteroidales and fecal indicator bacteria in an urban watershed impacted by sewage pollution, determined using RNA-and DNA-based quantitative PCR assays. Appl. Environ. Microbiol. 2015, 81, 91–99. [Google Scholar] [CrossRef] [PubMed]
  15. Mieszkin, S.; Furet, J.P.; Corthier, G.; Gourmelon, M. Estimation of pig fecal contamination in a river catchment by real-time PCR using two pig-specific Bacteroidales 16S rRNA genetic markers. Appl. Environ. Microbiol. 2009, 75, 3045–3054. [Google Scholar] [CrossRef]
  16. Schriewer, A.; Goodwin, K.D.; Sinigalliano, C.D.; Cox, A.M.; Wanless, D.; Bartkowiak, J.; Ebentier, D.L.; Hanley, K.T.; Ervin, J.; Deering, L.A.; et al. Performance evaluation of canine-associated Bacteroidales assays in a multi-laboratory comparison study. Water Res. 2013, 47, 6909–6920. [Google Scholar] [CrossRef] [PubMed]
  17. Gawler, A.H.; Beecher, J.E.; Brandão, J.; Carroll, N.M.; Falcão, L.; Gourmelon, M.; Masterson, B.; Nunes, B.; Porter, J.; Rincé, A.; et al. Validation of host-specific Bacteriodales 16S rRNA genes as markers to determine the origin of faecal pollution in Atlantic Rim countries of the European Union. Water Res. 2007, 41, 3780–3784. [Google Scholar] [CrossRef] [PubMed]
  18. Boehm, A.B.; Van De Werfhorst, L.C.; Griffith, J.F.; Holden, P.A.; Jay, J.A.; Shanks, O.C.; Wang, D.; Weisberg, S.B. Performance of forty-one microbial source tracking methods: A twenty-seven lab evaluation study. Water Res. 2013, 47, 6812–6828. [Google Scholar] [CrossRef] [PubMed]
  19. Pickering, A.J.; Julian, T.R.; Mamuya, S.; Boehm, A.B.; Davis, J. Bacterial hand contamination among Tanzanian mothers varies temporally and following household activities. Trop. Med. Int. Health 2011, 16, 233–239. [Google Scholar] [CrossRef] [PubMed]
  20. Mattioli, M.C.; Pickering, A.J.; Gilsdorf, R.J.; Davis, J.; Boehm, A.B. Hands and water as vectors of diarrhoeal pathogens in Bagamoyo, Tanzania. Environ. Sci. Technol. 2013, 47, 355–363. [Google Scholar] [CrossRef]
  21. Jenkins, M.W.; Tiwari, S.; Lorente, M.; Gichaba, C.M.; Wuertz, S. Identifying human and livestock sources of fecal contamination in Kenya with host-specific Bacteroidales assays. Water Res. 2009, 43, 4956–4966. [Google Scholar] [CrossRef]
  22. Pickering, A.J.; Julian, T.R.; Marks, S.J.; Mattioli, M.C.; Boehm, A.B.; Schwab, K.J.; Davis, J. Fecal contamination and diarrhoeal pathogens on surfaces and in soils among Tanzanian households with and without improved sanitation. Environ. Sci. Technol. 2012, 46, 5736–5743. [Google Scholar] [CrossRef] [PubMed]
  23. Vhembe District Municipality. Integrated Development Plan (IDP) of Vhembe District Municipality Limpopo South Africa; Vhembe District Municipality: Thohoyandou, South Africa, 2019.
  24. Stats SA (Statistics South Africa). The State of Basic Service Delivery in South Africa: In-Depth Analysis of the Community Survey 2016 Data; Statistics South Africa: Pretoria, South Africa, 2016.
  25. Sibiya, J.E.; Gumbo, J.R. Knowledge, attitude and practices (KAP) survey on water, sanitation and hygiene in selected schools in Vhembe District, Limpopo, South Africa. Int. J. Environ. Res. Public Health 2013, 10, 2282–2295. [Google Scholar] [CrossRef]
  26. SANS 5667-5:2006/ISO 5667-5:2006; South African National Standard (SANS). Water Quality–Sampling–Part 5: Guidance on Sampling of Drinking Water from Treatment Works and Piped Distribution Systems. SABS Standards Division: Pretoria, South Africa, 2006.
  27. Murray, A.L.; Kumpel, E.; Peletz, R.; Khush, R.S.; Lantagne, D.S. The effect of sodium thiosulfate dechlorination on fecal indicator bacteria enumeration: Laboratory and field data. J. Water Health 2018, 16, 70–77. [Google Scholar] [CrossRef]
  28. APHA; AWWA; WEF. Standard Methods for the Examination of Water and Wastewater; American Public Health Association (APHA), American Water Works Association (AWWA), Water Environment Federation (WEF): Washington, DC, USA, 2005; p. 21. [Google Scholar]
  29. Okabe, S.; Okayama, N.; Savichtcheva, O.; Ito, T. Quantification of host-specific Bacteroides–Prevotella 16S rRNA genetic markers for assessment of fecal pollution in freshwater. Appl. Microbiol. Biotechnol. 2005, 74, 890–901. [Google Scholar] [CrossRef]
  30. Schiaffino, F.; Pisanic, N.; Colston, J.M.; Rengifo, D.; Olortegui, M.P.; Shapiama, V.; Yori, P.P.; Heaney, C.D.; Davis, M.F.; Kosek, M.N. Validation of microbial source tracking markers for the attribution of fecal contamination in indoor-household environments of the Peruvian Amazon. Sci. Total Environ. 2020, 743, 140531. [Google Scholar] [CrossRef]
  31. Odagiri, M.; Schriewer, A.; Hanley, K.; Wuertz, S.; Misra, P.R.; Panigrahi, P.; Jenkins, M.W. Validation of Bacteroidales quantitative PCR assays targeting human and animal fecal contamination in the public and domestic domains in India. Sci. Total Environ. 2015, 502, 462–470. [Google Scholar] [CrossRef]
  32. Bustin, S.A.; Benes, V.; Garson, J.A.; Hellemans, J.; Huggett, J.; Kubista, M.; Mueller, R.; Nolan, T.; Pfaffl, M.W.; Shipley, G.L.; et al. The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Clin. Chem. 2009, 55, 611–622. [Google Scholar] [CrossRef]
  33. Shabalala, A.N.; Combrinck, L.; McCrindle, R. Effect of farming activities on seasonal variation of water quality of Bonsma Dam, KwaZulu-Natal. S. Afr. J. Sci. 2013, 109, 7. [Google Scholar] [CrossRef]
  34. Fouche, P.S.O.; Vlok, W.; Roos, J.C.; Luus-Powell, W.; Jooste, A. Establishing the Fishery Potential of Lake Nandoni in the Luvuvhu River, Limpopo Province; WRC Report No. 1925/1/12; Water Research Commission: Pretoria, South Africa, 2013; pp. 1–135. [Google Scholar]
  35. Traoré, A.N.; Mulaudzi, K.; Chari, G.J.; Foord, S.H.; Mudau, L.S.; Barnard, T.G.; Potgieter, N. The impact of human activities on microbial quality of rivers in the Vhembe District, South Africa. Int. J. Environ. Res. Public Health 2016, 13, 817. [Google Scholar] [CrossRef] [PubMed]
  36. Ercumen, A.; Pickering, A.J.; Kwong, L.H.; Arnold, B.F.; Parvez, S.M.; Alam, M.; Sen, D.; Islam, S.; Kullmann, C.; Chase, C.; et al. Animal feces contribute to domestic fecal contamination: Evidence from E. coli measured in water, hands, food, flies, and soil in Bangladesh. Environ. Sci. Technol. 2017, 51, 8725–8734. [Google Scholar] [CrossRef]
  37. Hachich, E.M.; Di Bari, M.; Christ, A.P.G.; Lamparelli, C.C.; Ramos, S.S.; Sato, M.I.Z. Comparison of thermotolerant coliforms and Escherichia coli densities in freshwater bodies. Braz. J. Microbiol. 2012, 43, 675–681. [Google Scholar] [CrossRef] [PubMed]
  38. Odonkor, S.T.; Mahami, T. Escherichia coli as a tool for disease risk assessment of drinking water sources. Int. J. Microbiol. 2020, 2020, 2534130. [Google Scholar] [CrossRef] [PubMed]
  39. Chen, H.J.; Chang, H. Response of discharge, TSS, and E. coli to rainfall events in urban, suburban, and rural watersheds. Environ. Sci. Process Impacts 2014, 16, 2313–2324. [Google Scholar] [CrossRef] [PubMed]
  40. Egilmez, H.I.; Morozov, A.Y.; Clokie, M.R.; Shan, J.; Letarov, A.; Galyov, E.E. Temperature-dependent virus lifecycle choices may reveal and predict facets of the biology of opportunistic pathogenic bacteria. Sci. Rep. 2018, 8, 9642. [Google Scholar] [CrossRef] [PubMed]
  41. Murei, A.; Kamika, I.; Samie, A.; Momba, M.N.B. Assessment of the water sources for potential channels of faecal contamination within Vhembe District Municipality using sanitary inspections and hydrogen sulphide test. Sci. Rep. 2023, 13, 6250. [Google Scholar] [CrossRef] [PubMed]
  42. Zhang, Y.; Wu, R.; Lin, K.; Wang, Y.; Lu, J. Performance of host-associated genetic markers for microbial source tracking in China. Water Res. 2020, 175, 115670. [Google Scholar] [CrossRef] [PubMed]
  43. Yatsunenko, T.; Rey, F.E.; Manary, M.J.; Trehan, I.; Dominguez-Bello, M.G.; Contreras, M.; Magris, M.; Hidalgo, G.; Baldassano, R.N.; Anokhin, A.P.; et al. Human gut microbiome viewed across age and geography. Nature 2012, 486, 222–227. [Google Scholar] [CrossRef]
  44. Reischer, G.H.; Ebdon, J.E.; Bauer, J.M.; Schuster, N.; Ahmed, W.; Åström, J.; Blanch, A.R.; Blöschl, G.; Byamukama, D.; Coakley, T.; et al. Performance characteristics of qPCR assays targeting human-and ruminant-associated Bacteroidetes for microbial source tracking across sixteen countries on six continents. Environ. Sci. Technol. 2013, 47, 8548–8556. [Google Scholar] [CrossRef] [PubMed]
  45. Nshimyimana, J.P.; Cruz, M.C.; Thompson, R.J.; Wuertz, S. Bacteroidales markers for microbial source tracking in Southeast Asia. Water Res. 2017, 118, 239–248. [Google Scholar] [CrossRef]
  46. Raith, M.R.; Kelty, C.A.; Griffith, J.F.; Schriewer, A.; Wuertz, S.; Mieszkin, S.; Gourmelon, M.; Reischer, G.H.; Farnleitner, A.H.; Ervin, J.S.; et al. Comparison of PCR and quantitative real-time PCR methods for the characterization of ruminant and cattle fecal pollution sources. Water Res. 2013, 47, 6921–6928. [Google Scholar] [CrossRef]
  47. Somnark, P.; Chyerochana, N.; Mongkolsuk, S.; Sirikanchana, K. Performance evaluation of Bacteroidales genetic markers for human and animal microbial source tracking in tropical agricultural watersheds. Environ. Pollut. 2018, 236, 100–110. [Google Scholar] [CrossRef] [PubMed]
  48. Unno, T.; Staley, C.; Brown, C.M.; Han, D.; Sadowsky, M.J.; Hur, H.G. Fecal pollution: New trends and challenges in microbial source tracking using next-generation sequencing. Environ. Microbiol. 2018, 20, 3132–3140. [Google Scholar] [CrossRef]
  49. Sidhu, J.P.; Hodgers, L.; Ahmed, W.; Chong, M.N.; Toze, S. Prevalence of human pathogens and indicators in stormwater runoff in Brisbane, Australia. Water Res. 2012, 46, 6652–6660. [Google Scholar] [CrossRef] [PubMed]
  50. Liang, L.; Goh, S.G.; Gin, K.Y.H. Decay kinetics of microbial source tracking (MST) markers and human adenovirus under the effects of sunlight and salinity. Sci. Total Environ. 2017, 574, 165–175. [Google Scholar]
  51. Waso, M.; Khan, S.; Khan, W. Microbial source tracking markers associated with domestic rainwater harvesting systems: Correlation to indicator organisms. Environ. Res. 2018, 161, 446–455. [Google Scholar] [CrossRef] [PubMed]
  52. Paruch, L.; Paruch, A.M. An overview of microbial source tracking using host-specific genetic markers to identify origins of fecal contamination in different water environments. Water 2022, 14, 1809. [Google Scholar] [CrossRef]
  53. Hamzah, L.; Boehm, A.B.; Davis, J.; Pickering, A.J.; Wolfe, M.; Mureithi, M.; Harris, A. Ruminant fecal contamination of drinking water introduced post-collection in rural Kenyan households. Int. J. Environ. Res. Public Health 2020, 17, 608. [Google Scholar] [CrossRef]
Water 16 02014 g001
Figure 1. Schematic diagram illustrating the study design.
Figure 1. Schematic diagram illustrating the study design.
Water 16 02014 g001
Water 16 02014 g002
Figure 2. Map of South Africa and Limpopo Province showing the Vhembe District, local municipalities, and the five selected villages and sampling points of the study site.
Figure 2. Map of South Africa and Limpopo Province showing the Vhembe District, local municipalities, and the five selected villages and sampling points of the study site.
Water 16 02014 g002
Water 16 02014 g003
Figure 3. Map showing the sampling points at water sources (dams and rivers) in the Vhembe District and local municipalities.
Figure 3. Map showing the sampling points at water sources (dams and rivers) in the Vhembe District and local municipalities.
Water 16 02014 g003
Water 16 02014 g004
Figure 4. Concentrations of E. coli in catchments (rivers and dams) and in the raw water at the treatment plants recorded in both the wet and dry seasons in the cohort study villages.
Figure 4. Concentrations of E. coli in catchments (rivers and dams) and in the raw water at the treatment plants recorded in both the wet and dry seasons in the cohort study villages.
Water 16 02014 g004
Water 16 02014 g005
Figure 5. Concentrations of E. coli in the final water at the treatment plants and in spring water recorded in both the wet and dry seasons in the cohort study villages.
Figure 5. Concentrations of E. coli in the final water at the treatment plants and in spring water recorded in both the wet and dry seasons in the cohort study villages.
Water 16 02014 g005
Water 16 02014 g006
Figure 6. Concentrations of E. coli in the municipal tap water in the households recorded in both the wet and dry seasons in the cohort study villages.
Figure 6. Concentrations of E. coli in the municipal tap water in the households recorded in both the wet and dry seasons in the cohort study villages.
Water 16 02014 g006
Water 16 02014 g007
Figure 7. Concentrations of E. coli in the spring water stored in containers in the households recorded in both the wet and dry seasons in the cohort study villages.
Figure 7. Concentrations of E. coli in the spring water stored in containers in the households recorded in both the wet and dry seasons in the cohort study villages.
Water 16 02014 g007
Water 16 02014 g008
Figure 8. Concentrations of E. coli in borehole water stored in JoJo tanks in the households recorded in both the wet and dry seasons in the cohort study villages.
Figure 8. Concentrations of E. coli in borehole water stored in JoJo tanks in the households recorded in both the wet and dry seasons in the cohort study villages.
Water 16 02014 g008
Water 16 02014 g009
Figure 9. Concentrations of E. coli in river water stored in containers in the households recorded in both the wet and dry seasons in the cohort study villages.
Figure 9. Concentrations of E. coli in river water stored in containers in the households recorded in both the wet and dry seasons in the cohort study villages.
Water 16 02014 g009
Water 16 02014 g010
Figure 10. Performance of animal- and human-associated Bacteroidales MST markers in human and animal stool samples.
Figure 10. Performance of animal- and human-associated Bacteroidales MST markers in human and animal stool samples.
Water 16 02014 g010
Water 16 02014 g011
Figure 11. Distribution trends of faecal contamination sources between water source and point of use in Tshakhuma (WTPR: water treatment plant raw water; WTPF: water treatment final water).
Figure 11. Distribution trends of faecal contamination sources between water source and point of use in Tshakhuma (WTPR: water treatment plant raw water; WTPF: water treatment final water).
Water 16 02014 g011
Water 16 02014 g012
Figure 12. Distribution trends of faecal contamination sources between water source and point of use in Tshivhulani.
Figure 12. Distribution trends of faecal contamination sources between water source and point of use in Tshivhulani.
Water 16 02014 g012
Water 16 02014 g013
Figure 13. Distribution trends of faecal contamination sources between water source and point of use in Tshilapfene.
Figure 13. Distribution trends of faecal contamination sources between water source and point of use in Tshilapfene.
Water 16 02014 g013
Water 16 02014 g014
Figure 14. Distribution trends of faecal contamination sources between water source and point of use in Tshidzini.
Figure 14. Distribution trends of faecal contamination sources between water source and point of use in Tshidzini.
Water 16 02014 g014
Water 16 02014 g015
Figure 15. Distribution trends of faecal contamination sources between water source and point of use in Dididi.
Figure 15. Distribution trends of faecal contamination sources between water source and point of use in Dididi.
Water 16 02014 g015
Table 1. qPCR primer and probe sequences of gene-specific molecular markers tested for in all water samples.
Table 1. qPCR primer and probe sequences of gene-specific molecular markers tested for in all water samples.
HostTarget NameSequence (5′-3′)DyeReference
HumanHF183–1ATCATGAGTTCACATGTCCGFAM-TAMRAKapoor et al., 2015 [14]
BthetR1CGTAGGAGTTTGGACCGTGT
BthetP1CTGAGAGGAAGGTCCCCCACATTGGA
CowBacCow-CF128CCAACYTTCCCGWTACTCFAM-TAMRAKildare et al., 2007 [11]
BacCow-305rGGACCGTGTCTCAGTTCCAGTG
BacCow-257pTAGGGGTTCTGAGAGGAAGGTCCCCC
PigPig-2-Bac41FGCATGAATTTAGCTTGCTAAATTTGATFAM-MGBMieszkin et al., 2009 [15]
Pig-2-Bac RACCTCATACGGTATTAATCCGC
Pig-2-Bac113TCCACGGGATAGCC
ChickenCytb FAAATCCCACCCCCTACTAAAAATAATFAM-MGBSchiaffino et al., 2020 [30]
Cytb RCAGATGAAGAAGAATGAGGCG
Cytb PACAACTCCCTAATCGACCT
DogBacCan1545fGGAGCGCAGACGGGTTTTFAM-MGBKildare et al., 2007 [11]
BacUni-690r1CAATCGGAGTTCTTCGTGATATCTA
BacUni-656pTGGTGTAGCGGTGAAA
Table 2. Comparison of E. coli concentrations (CFU/100 mL) in water samples from different water sources.
Table 2. Comparison of E. coli concentrations (CFU/100 mL) in water samples from different water sources.
Water SourceWet SeasonDry Seasonp-Value
Mean
SD		MinMaxdfMean
SD		MinMaxdf
Catchments (RD) (n = 96)45 ± 30.698907922 ± 15.05650790.01
RWPT (n = 40)35 ± 17.9112603119 ± 16.96848310.01
MTWPT (n = 40)00031000310.00
Communal tap
(n = 8)		000700070.00
MTWHHY
(n = 400)		0.64 ± 1.20043990.37 ± 0.90043990.02
SW (n = 24)8 ± 4.87110239 ± 8.58021230.08
SWCHH (n = 248)2 ± 2.47082391.08 ± 1.97092390.05
JHHBW (n = 144)0.17 ± 0.41021430.14 ± 0.28011430.05
RWCHH (n = 40)22 ± 31.870853910 ± 12.14134390.03
Note(s): HH: households; SD: standard deviation; Min: minimum; Max: maximum; df: degree of freedom; RD: river and dam; RWPT: raw water at point of treatment; MTWPT: municipal tap water at point of treatment; MTWHHY: municipal tap water in household inside yard; SW: spring water; SWCHH: spring water in containers in households; JJHHBW: JoJo tank in households containing borehole water; RWCHH: river water in containers in households.
Table 3. Standard curve values for five host-specific MST markers.
Table 3. Standard curve values for five host-specific MST markers.
Standard Curve Parameters for Host-Specific Markers
Target SpeciesSpecific MarkerSlopey-InterceptLinearity (R2) Efficiency
(%)		LLOQ (Ct) ValueGene Copy Number per µLLog10 Gene Copies per ng
CowBacCow-CF128−3.6740.510.99258726.632.85 × 1037 37.45
ChickenCytb−3.7539.620.99268526.511.18 × 1037 37.07
HumanHF183−3.7439.840.98908526.742.49 × 103737.40
PigPig-2-Bac−3.4839.020.97839426.821.08 × 103838.03
DogBacCan−3.6339.310.99318927.665.02 × 103838.70
Note(s): Ct: cycle threshold value; LLOQ: lower limit of quantification.
Table 4. Prevalence of the detected sources of contamination in all five study villages during the wet season.
Table 4. Prevalence of the detected sources of contamination in all five study villages during the wet season.
Sampled Study VillagesDetected Sources of Contamination in All Five Study Villages during the Wet Season
OverallBacCow (Cow)Cytb (Chicken)HF183 (Human)Pig-2-Bac (Pig)BacCan (Dog)p-Value
Tshakhuma
(n = 112)		17 (15%)5 (4%)3 (3%)2 (2%)1 (1%)8 (7%)0.022
Tshivhulani
(n = 112)		28 (25%)7 (6%)6 (5%)3 (3%)2 (2%)10 (9%)0.015
Tshilapfene
(n = 96)		15 (16%)4 (4%)2 (2%)2 (2%)2 (2%)5 (5%)0.030
Tshidzini
(n = 104)		34 (33%)8 (8%)10 (10%)4 (4%)3 (3%)9 (9%)0.019
Dididi
(n = 100)		24 (24%)6 (6%)5 (5%)4 (4%)5 (5%)4 (4%) 0.007
Table 5. Prevalence of the detected sources of contamination in all five study villages during the dry season.
Table 5. Prevalence of the detected sources of contamination in all five study villages during the dry season.
Sampled Study VillagesDetected Sources of Contamination in All Five Study Villages during the Dry Season
OverallBacCow (Cow)Cytb (Chicken)HF183 (Human)Pig-2-Bac (Pig)BacCan (Dog)p-Value
Tshakhuma
(n = 112)		13 (11%)3 (3%)2 (2%)2 (2%)1 (1%)5 (4%)0.055
Tshivhulani
(n = 112)		22 (20%)6 (5%)6 (5%)3 (3%)2 (2%)5 (4%)0.031
Tshilapfene
(n = 96)		13 (14%)2 (2%)3 (3%)2 (2%)2 (2%)4 (4%)0.084
Tshidzini
(n = 104)		24 (23%)6 (6%)4 (4%)3 (3%)2 (2%)9 (9%)0.011
Dididi
(n = 100)		21 (21%)6 (6%)5 (5%)3 (3%)2 (2%)5 (5%)0.097
Table 6. Prevalence of detected sources of faecal contamination in all five study villages during the wet and dry season.
Table 6. Prevalence of detected sources of faecal contamination in all five study villages during the wet and dry season.
Sampled Water SourcesDetected Source of Faecal Contamination from Different Water Sources in the Study Areas
BacCow (Cow)Cytb (Chicken)HF183 (Human)Pig-2-Bac (Pig)BacCan (Dog)p-Value Pearson Correlation (r) (Wet and Dry Season)
Catchments (R and D) (n = 96)25 (26%)3 (3%)7 (7%)6 (6%)9 (9%)0.0020.47
RWPT (n = 40)7 (18%)0 (0%)2 (5%)0 (0%)2 (5%)0.0190.56
MTWPT (n = 40)0 (0%)0 (0%)0 (0%)0 (0%)0 (0%)0.0580.33
Communal tap
(n = 8)		0 (0%)0 (0%)0 (0%)0 (0%)0 (0%)0.0550.29
MTWHHY
(n = 400)		4 (1%)12 (3%)1 (0.3%)8 (2%)16 (4%)0.0060.70
JHHBW (n = 144)2 (1%)2 (1%)0 (0%)2 (1%)5 (4%)0.0240.62
Hand-dug well (n = 8)3 (38%)2 (25%)1 (13%)0 (0%)4 (50%)0.1710.48
RWCHH (n = 40)5 (13%)0 (0%)5 (8%)0 (0%)0 (0%)0.0020.51
SW (n = 24)5 (21%)1 (4%)1 (4%)1 (4%)(21%)0.0370.87
SWCHH (n-248)2 (1%)14 (6%)4 (2%)7 (3%)17 (7%)0.0110.22
Note(s): HH: households; SD: standard deviation; Min: minimum; Max: maximum; RD: river and dam; RWPT: raw water at point of treatment; MTWPT: municipal tap water at point of treatment; MTWHHY: municipal tap water in household inside yard; SW: spring water; SWCHH: spring water in containers in households; JJHHBW: JoJo tank in households containing borehole water; RWCHH: river water in containers in households.
Table 7. Correlation between the prevalence of E. coli and the faecal source markers in the study village households in the wet season (p-value was <0.05).
Table 7. Correlation between the prevalence of E. coli and the faecal source markers in the study village households in the wet season (p-value was <0.05).
Correlation Coefficient (r)
WETE. coliBacCowE. coliCytbE. coliBacCanE. coliHF183E. coliPig-2-Bac
E. coli1
BacCow0.931
E. coli10.931
Cytb0.900.950.901
E. coli10.9310.901
BacCan0.430.400.430.520.431
E. coli10.9310.9010.431
HF1830.700.850.700.770.70−0.100.701
E. coli10.9310.9010.4310.701
Pig-2-Bac0.350.500.350.350.35−0.580.350.860.351
Table 8. Correlation between the prevalence of E. coli and the faecal source markers in the study village households in the dry season (p-value was <0.05).
Table 8. Correlation between the prevalence of E. coli and the faecal source markers in the study village households in the dry season (p-value was <0.05).
Correlation Coefficient (r)
DRYE. coliBacCowE. coliCytbE. coliBacCanE. coliHF183E. coliPig-2-Bac
E. coli1
BacCow0.641
E. coli10.641
Cytb0.540.790.541
E. coli10.641.000.541
BacCan0.700.530.700.090.701
E. coli10.6410.541.000.701
HF1830.630.950.630.860.630.520.631
E. coli1.000.6410.5410.7010.631
Pig-2-Bac0.410.390.410.670.410.220.410.650.411
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Mochware, O.T.W.; Thaoge-Zwane, M.L.; Momba, M.N.B. Applying Microbial Source Tracking Techniques for Identification of Pathways of Faecal Pollution from Water Sources to Point of Use in Vhembe District, South Africa. Water 2024, 16, 2014. https://doi.org/10.3390/w16142014

AMA Style

Mochware OTW, Thaoge-Zwane ML, Momba MNB. Applying Microbial Source Tracking Techniques for Identification of Pathways of Faecal Pollution from Water Sources to Point of Use in Vhembe District, South Africa. Water. 2024; 16(14):2014. https://doi.org/10.3390/w16142014

Chicago/Turabian Style

Mochware, Opelo Tlotlo Wryl, Mathoto Lydia Thaoge-Zwane, and Maggy Ndombo Benkete Momba. 2024. "Applying Microbial Source Tracking Techniques for Identification of Pathways of Faecal Pollution from Water Sources to Point of Use in Vhembe District, South Africa" Water 16, no. 14: 2014. https://doi.org/10.3390/w16142014

APA Style

Mochware, O. T. W., Thaoge-Zwane, M. L., & Momba, M. N. B. (2024). Applying Microbial Source Tracking Techniques for Identification of Pathways of Faecal Pollution from Water Sources to Point of Use in Vhembe District, South Africa. Water, 16(14), 2014. https://doi.org/10.3390/w16142014

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop