Groundwater contamination in sub-Saharan Africa: Implications for groundwater protection in developing countries

Abstract

Rural and peri-urban communities in the developing world typically rely on boreholes and wells for drinking water, since freshwater on the land surface is heavily polluted and groundwater is perceived to be safe. Nonetheless, in these areas wastewater is managed onsite, often close to the groundwater abstractions points. Unbeknownst to most members of these local communities, this poses an unprecedented threat to their health, since chemicals and pathogens are leaking from onsite systems to groundwater. Polluted groundwater could also act as an environmental reservoir for bacteria and viruses, including new and emerging infectious diseases. In this study, groundwater from rural and peri-urban South Africa was assessed in terms of drinking water quality. Indicators of faecal pollution were identified across the examined boreholes, with E. coli and nitrates concentrations as high as 195 cfu 100mL-1 and 104 mg L-1, respectively, suggesting that onsite wastewater systems have grossly impacted groundwater quality. Elevated concentrations of fluoride and chloride (as high as 8.6 and 392 mg L-1, respectively) were also identified, suggesting that local geological setting also affects groundwater quality. Overall, groundwater was found in need of treatment before consumption, which is currently not the case. To inform local communities and the water industry about the problem, a simple classification model was developed. This could help assess water quality, identify the nature of pollution, suggest the type of treatment required, and prioritise the need for water and wastewater infrastructure investment in the developing world. It can also inform local communities about the impact of their current wastewater management practices on water resources and the pressing need to upgrade them.

Full text

Journal Pre-proof
Groundwater contamination in sub-Saharan Africa: implication to groundwater
protection in developing countries
Vhahangwele Masindi, Spyros Foteinis
PII: S2666-7908(20)30038-0
DOI: https://doi.org/10.1016/j.clet.2020.100038
Reference: CLET 100038
To appear in: Cleaner Engineering and Technology
Received Date: 28 June 2020
Revised Date: 16 December 2020
Accepted Date: 16 December 2020
Please cite this article as: Masindi, V., Foteinis, S., Groundwater contamination in sub-Saharan Africa:
implication to groundwater protection in developing countries, Cleaner Engineering and Technology,
https://doi.org/10.1016/j.clet.2020.100038.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

Journal Pre-proof

1
Groundwater contamination in sub-Saharan Africa: implication to groundwater
1
protection in developing countries
2
Vhahangwele Masindi
1&2
, Spyros Foteinis
3*
3
1
Magalies Water (MW), Scientific Services (SS), Research & Development (R&D) Division, Erf 3475,
4
Stoffberg street, Brits, 0250, South Africa, tel: 0123816602
5
2
University of South Africa (UNISA), College of Agriculture and Environmental Sciences (CAES),
6
Department of Environmental Sciences, P.O. Box 392, Florida, 1710, South Africa
7
3
School of Engineering, Institute for Infrastructure and Environment, University of Edinburgh, Edinburgh
8
EH9 3JL, United Kingdom
9
10
Abstract 11
Rural and peri-urban communities in the developing world typically rely on boreholes and wells 12
for drinking water, since freshwater on the land surface is heavily polluted and groundwater is 13
perceived to be safe. Nonetheless, in these areas wastewater is managed onsite, often close to 14
the groundwater abstractions points. Unbeknownst to most members of these local 15
communities, this poses an unprecedented threat to their health, since chemicals and 16
pathogens are leaking from onsite systems to groundwater. Polluted groundwater could also act 17
as an environmental reservoir for bacteria and viruses, including new and emerging infectious 18
diseases. In this study, groundwater from rural and peri-urban South Africa was assessed in 19
terms of drinking water quality. Indicators of faecal pollution were identified across the examined 20
boreholes, with E. coli and nitrates concentrations as high as 195 cfu 100mL
-1
and 104 mg L
-1
, 21
respectively, suggesting that onsite wastewater systems have grossly impacted groundwater 22
quality. Elevated concentrations of fluoride and chloride (as high as 8.6 and 392 mg L
-1
, 23
respectively) were also identified, suggesting that local geological setting also affects 24
groundwater quality. Overall, groundwater was found in need of treatment before consumption, 25
Journal Pre-proof

2
which is currently not the case. To inform local communities and the water industry about the 26
problem, a simple classification model was developed. This could help assess water quality, 27
identify the nature of pollution, suggest the type of treatment required, and prioritise the need for 28
water and wastewater infrastructure investment in the developing world. It can also inform local 29
communities about the impact of their current wastewater management practices on water 30
resources and the pressing need to upgrade them. 31
Keywords: Groundwater quality monitoring; unlined or semi-lined pit latrines and septic tanks; 32
water and wastewater management; pathogens; virions; bacteria. 33
* Corresponding author: sfoteini@exseed.ed.ac.uk and sfoteinis@gmail.com 34
1 Introduction 35
Nowadays, the human right to safe drinking water and access to sanitation is well 36
recognised through various initiatives, including the United Nations (UN) Resolution 64/292 (UN 37
General Assembly, 2010) and the UN’s Sustainable development goals (SDGs) (The United 38
Nations., 2018). Even though large strides have been made in this regard, currently around 1.5 39
10
9
people, or 19% of the global population, can only access basic drinking water services, 40
while a staggering 785 10
6
people having no access to water services whatsoever (The United 41
Nations., 2018). This situation is typical for rural and peri-urban communities in the Republic of 42
South Africa (RSA), where water distribution networks are often not in place and groundwater is 43
the sole source of drinking water since the freshwater on the land surface is heavily polluted 44
(Potgieter et al., 2020). As a result, in these communities drinking water is provided by 45
communal and private boreholes and wells, which is also the case elsewhere, e.g. in rural 46
Zambia (Kapulu et al., 2013). However, even though groundwater has been traditionally 47
perceived as a safe and dependable drinking water source this is often not the case, since it is 48
Journal Pre-proof

3
prone to pollution from both anthropogenic (Lapworth et al., 2012) and natural processes 49
(Rezaei et al., 2019). 50
Specifically, groundwater can be contaminated by a wide array of different pollutants, 51
including viruses, coliform bacteria, metals, oxyanions, pharmaceuticals, pesticides and 52
fertilizers (Manga et al., 2020). As a result, groundwater can act as an environmental reservoir 53
for bacteria and viruses, including new and emerging infectious diseases. Furthermore, natural 54
processes, such as weathering, coupled with the local geology can also affect groundwater 55
quality through leaching of cationic and anionic chemicals, e.g. fluorides, sulphates, heavy 56
metals and metalloids (Li et al., 2018). Particularly, fluoride in groundwater comprise a global 57
issue, with ~200 10
6
people being at health risk of endemic fluorosis (skeletal and dental) and 58
neurotoxicological implications in children, among others (Kaur et al., 2020). The problem 59
persist in RSA (Ncube and Schutte, 2005), the United Republic of Tanzania (WHO, 2004) and 60
further afield such as in China (He et al., 2020). In rural and peri-urban communities in RSA 61
groundwater pollution is grossly attributed to animal farming (Enitan-Folami et al., 2019) and 62
onsite sanitation systems ( Taonameso et al., 2018). The latter includes pit latrines and to a 63
smaller degree septic tanks, which are both used to manage human waste onsite. The “bush 64
method” can also cause groundwater contamination, but people are unaware of this problem as 65
was highlighted in a case study in Namibia (Claasen and Lewis, 2017). 66
The problem, per se, does not lie in the use of onsite sanitation systems, but to: i) their 67
proximity to groundwater abstraction points (Palamuleni and Akoth, 2015), and more importantly 68
ii) the fact are typically unlined or semi-lined, hence increasing the probability of groundwater 69
pollution through leakages to the aquifer (Shivendra and Ramaraju, 2015). Specifically, unlined 70
and semi-lined systems, which are grossly used in the developing world, are not watertight. 71
Thus, they allow the wastewater to leach to the ground and travel large distances within the 72
groundwater matrix. To make things worse, in many instances pit latrines are located in close 73
Journal Pre-proof

4
proximity (even a few m) from the boreholes or water wells (Enitan-Folami et al., 2019), which 74
makes groundwater unsuitable for direct consumption. However, strong misconceptions still 75
exists on groundwater’s suitability for drinking purposes, e.g. in a case study in India only 3 to 76
4% of the examined population was aware of the groundwater pollution problem (Das et al., 77
2019). The extent of the problem is highlighted by the fact that it has been estimated that 78
around 1.77 10
9
people use pit latrines as their primary means of sanitation (Graham and 79
Polizzotto, 2013). Watertight onsite sanitation systems could address, at least partly, this 80
problem, but these are more costly and are not typically used in RSA and the developing world. 81
Therefore, unbeknownst to most end users in the developing world, raw groundwater 82
consumption can affect their health and wellbeing. For this reason, detailed information about 83
the quality of the groundwater that serves local communities in RSA and further afield, should 84
exist, along with simple and easy to use tools to communicate the problem. This is addressed in 85
this work by providing actual measurements of the groundwater quality in rural and peri-urban 86
RSA and by proposing a simple and easy to use water quality classification model, tailored for 87
RSA and the developing world. To this end, groundwater’s current state in rural and peri-urban 88
RSA is first examined and assessed in terms of water quality. The presented results record the 89
quality of groundwater in a typical sub-Saharan region and also provide a point of reference for 90
future studies, which is grossly missing from the literature. More importantly, the proposed 91
borehole classification model can also be used to: i) provide insight and tools regarding the type 92
and extent of pollution and the required treatment, ii) help the water industry to prioritise 93
infrastructure investing, and iii) effectively disseminate the huge environmental and health 94
impact of onsite sanitation systems to local communities in the developing world and the need 95
to safeguard groundwater resources. The latter is of major importance for rural and peri-urban 96
communities in RSA and further afield, particularly when considering the adverse impacts of 97
climate change in groundwater resources (Dennis and Dennis 2012) and the fact that despite 98
Journal Pre-proof

5
being a water scarce country, its daily per capita water consumption (233 L) along with North 99
West province’s consumption (204 L) is well above the mean global consumption (180 L) 100
(DWS, 2017). Overall, this work progress the body of knowledge on groundwater quality in 101
developing countries, where the current sanitation practices affect, by and large, its quality and 102
is in line with UN’s Sustainable Development Goals (SDG no6: Clean Water and Sanitation) 103
(The United Nations, 2018). 104
105
2 Materials and methods 106
2.1 Area of Study 107
In this work, the northern part of the North West province, RSA (near the border with 108
Limpopo and Gauteng provinces) was examined (Figure 1 - generated using the open source 109
software Generic Mapping Tools (GMT) (Wessel et al., 2019)). RSA is geographically located in 110
the arid to semi-arid sub-Saharan region, experiencing relatively low rainfall (average annual 111
rainfall ~500 mm) with large and unpredictable seasonal precipitation patterns (Dennis and 112
Dennis 2012). A large part of the area under study falls within the Limpopo river catchment and 113
basin. In this region, the average annual precipitation is ~350 mm and temperature varies from 114
very cold (freezing) in winter to extremely hot weather (35 - 40ºC) during the warm seasons 115
(spring, summer, and autumn). Regarding the local geology, a huge belt of dolomite stretches 116
from Limpopo, to Mpumalanga, until North West province, which recharges low and deep 117
aquifers. Furthermore, the underlying geology of the North West province is mainly covered with 118
granites of the Bushveld Igneous Complex, shales of the Ecca Group, and dolerite of the Karoo 119
basin, while silty to clay soils mostly cover these rocks (Cobbing et al., 2008). 120
This area is also rich in valuable minerals, such as platinum, iron ore, and chromite, and 121
therefore, a strong mining industry exists. However, geological weathering and mining activities 122
Journal Pre-proof

6
might contaminate groundwater due to geological destabilisation. Worryingly, surface water in 123
this area has drastically deteriorated, due to discharges of effluents rich in nutrients, pesticides, 124
and contaminants of emerging concern. As a result surface water resources, i.e. dams, and 125
their feeding rivers and tributaries are currently infested with alien and native aquatic plants and 126
microbial contaminants (Masindi, 2020) threating fresh water security in South Africa (Musingafi, 127
2014). Therefore, in this area groundwater serves as the augmenting source of water supply for 128
domestic and agricultural use (Tessema et al., 2011). 129
However, in regions that are decentralised or lack water distribution networks, such as 130
rural and peri-urban areas in the North West province, South Africa, or further afield, 131
groundwater is typically the sole water source for domestic use. Nonetheless, this might not be 132
fit for human consumption. For this reason, in this work, 42 boreholes from the northern part of 133
the North West province, near the border with Limpopo and Gauteng provinces, were examined 134
in terms of water quality. This area is sparsely populated and therefore focus was placed on 135
collecting water samples that are representative for the examined rural and peri-urban 136
communities under study (Figure 1). 137
Journal Pre-proof

7
138
Figure 1: The area of study along with the location of the examined boreholes (generated using 139
GMT). 140
In this area, as well as in the developing world in general, groundwater pollution can be grossly 141
traced back to onsite sanitations systems, since wastewater infrastructure is practically non-142
existence. In the examined communities unlined or semi-lined pit latrines are typically used 143
(Figure 2), which release biological, organic, and inorganic contaminants through wastewater 144
leaching to the soil and by extension to the aquifer (Palamuleni and Akoth, 2015). This situation 145
is particularly problematic in aquifers with a high groundwater table or in fractured rock and 146
other high permeability soils, where the risk of contamination is considerable (Taonameso et al., 147
2018). Wastewater pollution can also combine with physicochemical contamination attributed to 148
Journal Pre-proof

8
natural processes, such as minerals weathering and leaching to groundwater, and exacerbate 149
the problem. 150
151
152
Figure 2: a) A semi-lined pit latrine and b) unlined pit latrine in rural communities of North West 153
province South Africa. 154
155
2.2 Sample collection 156
Water samples from 42 boreholes (Figure 1 and Table 1) were collected and assessed 157
in terms of drinking water quality. The boreholes are located in North West province, where 158
Magalies Water, a water board in RSA that supported this work, centrally operates. Specifically, 159
apart from providing bulk water supply, Magalies Water also assists local municipalities to 160
design, develop, construct and maintain basic water services. As such, it maintains strong links 161
Journal Pre-proof

9
with rural and peri-urban communities in RSA, which were used to ensure that a large number 162
of boreholes, across local communities, were examined. The reason why these locations were 163
chosen in twofold: i) they spatially cover the area of operation for Magalies Water, thus 164
highlighting a potential avenue for groundwater harvesting, treatment, and selling to local 165
communities and ii) they represent local communities in RSA that rely on boreholes for potable 166
water. To account for the fact that in RSA both private/communal and municipality/water-boards 167
owned boreholes are in operation, samples were collected from private, communal, and 168
municipal owned boreholes (Table 1). 169
Specifically, private and communal boreholes are used and operated by individuals, 170
which typically do not have the means and funds to treat the pumped water. On the other hand, 171
boreholes that belong to local municipalities in RSA are often operated by water boards, which 172
provide bulk water supply infrastructure and serve local municipalities. As such, a potential 173
source of revenue is the pumping and distribution of groundwater to local communities, typically 174
using basic water services that do not include water treatment. However, as the water quality 175
deteriorates, the need to invest in water treatment infrastructure increases, with water boards or 176
local municipalities footing the bill. This stresses their economic and financial viability. 177
Table 1 shows the coordinates of the boreholes under study, with their number 178
corresponding to the number shown in Figure 1. The name of the place/village that each 179
borehole is located, along with the sampling date and other relevant information are also listed 180
in Table 1. Specifically, boreholes no 1 to no 19 are owned, operated by, and serve private 181
users. In this sense, community members are free to pump and extract water from 5 out of 182
those 19 boreholes (shared or communal) while water from the remaining 14 boreholes is only 183
used by individuals (private). Furthermore, the boreholes no 20 to no 42 are owned by local 184
municipalities, but their operation and management has been outsourced to a water board 185
Journal Pre-proof

10
(Magalies Water). The water from these boreholes (municipality) is pumped into a central 186
reservoir for distribution. 187
Table 1: The borehole locations, along with the coordinates, the name of the place/village that 188
they are located, and the date of sampling, and other relevant information. Each borehole 189
number corresponds to the number shown in Figure 1. 190
Borehole
number
Name of
village/place
Coordinates
Ownership
Sampling
Date
Modus
operandi
Distribution
network
1
Bojating
House.no 227 -25.194454 27.380218
Private
14/11/2019
Individual
No
2
Kopman
House 372 -25.376777 27.260706
Private
14/11/2019
Individual
No
3
Ledig House
No.80034 -25.376735 27.063223
Private
14/11/2019
Individual
No
4
Lefaragathla
village -25.624681 27.197715
Communal
14/11/2019
Individual
No
5
Mamerotswe
village Masisi -25.417420 27.343129
Communal
27/11/2019
Individual
No
6
Bethanie
Mantabole sec -25.557147 27.614760
Private
27/11/2019
Individual
No
7
Ga-Mogajane -25.490027 27.362387
Communal
27/11/2019
Individual
No
8
Magono Luka -25.485397 27.181334
Communal
29/01/2020
Individual
No
9
Mogwase
Mayors house
Unit 4 -25.274689 27.208044
Private
29/01/2020
Individual
No
10
Mononono
no.10041 -25.048137 27.185608
Private
29/01/2020
Individual
No
11
Moruleng
House no.
50993 -25.177816 27.180697
Private
29/01/2020
Individual
No
12
Noka ya Lerato -25.432248 27.498308
Private
30/01/2020
Individual
No
13
Phatsima
House no.43 -25.391503 27.007310
Private
30/01/2020
Individual
No
14
Phokeng
House. 2755 -25.126563 27.377872
Private
30/01/2020
Individual
No
15
Ramokoka P.
School -25.152700 27.424657
Private
30/01/2020
Individual
No
16
Ramokoka
House
no.60150 -25.156430 27.420488
Private
30/01/2020
Individual
No
17
Rampa H/N
E9529 -25.126823 27.185427
Private
30/01/2020
Individual
No
18
Suncity
Bungalows -25.356957 27.107942
Communal
30/01/2020
Individual
No
19
Tswiri sec H/N
0637 -25.527175 27.530354
Private
30/01/2020
Individual
No
20
Mogogelo
-
25.352639
28.132789
Municipality
20/02/2020
Water board
Yes
21
Ratjiespan
-
25.328672
28.078022
Municipality
20/02/2020
Water board
Yes
Journal Pre-proof

11
22
Ga-Motla
-
25.361994
28.080164
Municipality
20/02/2020
Water board
Yes
23
Moeka
-
25.330206
28.054142
Municipality
20/02/2020
Water board
Yes
24
Jonathan
-
25.206758
27.869131
Municipality
20/02/2020
Water board
Yes
25
Wallmans-BH
-
25.091586
28.231306
Municipality
20/02/2020
Water board
Yes
26
Rabosula
-
25.280342
27.975525
Municipality
20/02/2020
Water board
Yes
27
Tlounane BH
-
25.117422
28.189069
Municipality
20/02/2020
Water board
Yes
28
Olverton BH
-
25.067772
28.197817
Municipality
20/02/2020
Water board
Yes
29
Vontini BH
-
25.054256
28.192347
Municipality
20/02/2020
Water board
Yes
30
Swartboom BH
-
25.043328
28.134539
Municipality
20/02/2020
Water board
Yes
31
Swartboom
Steel tank
-
25.039617
28.130050
Municipality
20/02/2020
Water board
Yes
32
Dipetlwane
BH
clinic
-
25.078317
28.121008
Municipality
20/02/2020
Water board
Yes
33
Dipetlelwane
-
25.081939
28.125753
Municipality
20/02/2020
Water board
Yes
34
Tladistad b/h 1
-
25.202489
28.044419
Municipality
20/02/2020
Water board
Yes
35
Tladistad b/h2
-
25.208247
28.042483
Municipality
20/02/2020
Water board
Yes
36
Tladistad b/h3
-
25.206897
28.037414
Municipality
20/02/2020
Water board
Yes
37
Mmatlhwaele
-
25.229386
28.013039
Municipality
20/02/2020
Water board
Yes
38
Mocheku b/h
-
25.216889
27.976783
Municipality
20/02/2020
Water board
Yes
39
Moema b/h
-
25.210172
27.974539
Municipality
20/02/2020
Water board
Yes
40
Dikebu b/h
-
25.194675
27.970108
Municipality
20/02/2020
Water board
Yes
41
Phedile b/h
-
25.053008
27.926653
Municipality
20/02/2020
Water board
Yes
42
Ruigtersloot
-
25.027769
27.852767
Municipality
20/02/2020
Water board
Yes
191
2.3 Sample analyses 192
After collection, the samples were immediately transported, following the standard 193
procedures for sample handling that are described elsewhere (Greenberg et al., 2010), to a 194
laboratory that is accredited by South African National Accreditation System (SANAS). The 195
samples were analyzed within 24 h from time of their collection, both in terms of 196
physicochemical characteristics and microbial indicators associated with bacteria of faecal 197
origin. Below, a short description of the equipment used and the analytical techniques applied 198
for sample assessment is given. Analytical quality control (AQA) and quality assurance (QA) of 199
the laboratory measurements were followed, using the US National Institute of Standards and 200
Technology (NIST) standards and the procedures that are in place on the SANAS accredited 201
laboratory, respectively. 202
Journal Pre-proof

12
Finally, standard and accredited methods and procedures were followed for samples 203
collection and analysis (in-situ and ex-situ), as described in the American Public Health 204
Association (APHA) guidelines (Greenberg et al., 2010). 205
2.3.1 Analysis of chemical and physical parameters 206
The analytical instruments that were used to determine the physical and chemical 207
parameters of the examined groundwater samples are listed in Table 2. It should be noted that 208
the total dissolved solids (TDS) were measured at 180 °C, while the electrical conductivity (EC) 209
and pH were measured at 25 °C. 210
Table 2: The analytical instruments that were used to determine the physical and chemical 211
parameters of the borehole water samples. 212
Parameters
Analytical e
quipment
Arsenic, Chromium,
Manganese, iron,
Calcium, and
Magnesium
Inductively coupled plasma mass spectrometry (ICP-MS), XSeries 2,
ICP-MS, supplied by Thermo scientific, from Hanna-Kunath-Str. 11
28199 Bremen, Germany. The ICP-MS was coupled to ASX-520 Auto
sampler.
Arsenic, Chromium,
Manganese, iron,
Calcium, and
Magnesium
Inductively coupled plasma - optical emission spectrometry (ICP-OES),
5110 ICP-OES vertical dual view, Agilent technologies Australia, Made
in Malaysia. The ICP-OES was coupled with Agilent SPS 4 Auto
sampler.
Nitrate, sulphate,
fluoride, chloride,
Magnesium, Calcium
and Hardness
Gallery plus photo spectrometer, Automated chemistry analyzer,
Supplied by Thermo Fisher scientific, Made in Vantaa, Finland.
pH, EC, and TDS HANNA Multi-parameter probe. HI-9828 Multi-Parameter Water Quality
Portable Meter.
213
2.3.2 Microbiology of the water 214
2.3.2.1 Total Plate Count 215
The total plate count (TPC) was determined following the APHA guidelines. Specifically, 216
a defined volume of the sample was transferred into a 90 mm size Petri dish, along with molten 217
Journal Pre-proof

13
agar, and then mixed. After solidification, the agar plates were inverted and incubated at 36˚C 218
for 48 h (Greenberg et al., 2010). Then, the colonies, which varied from single cells, pairs, 219
chains, to clusters, were measured to count the colony forming units. 220
2.3.2.2 Total coliforms and E.coli 221
Similar to TPC, the total coliform and E. coli concentrations were estimated following the 222
APHA guidelines. Specifically, measurements were carried out using the membrane filtration 223
technique, where the sample is pass through a cellulose filter (0.45 µm pore size). After 224
filtration, the samples were cultured on agar mediums and incubated at 36˚C for 24 h. Then, the 225
bacteria concentrations were measured (Greenberg et al., 2010). 226
2.4 Treatability index and borehole classification model 227
In the context of this work the physicochemical characteristics and microbial indicators of 228
each examined borehole were comprehensively examined. To effectively communicate 229
groundwater’s quality and suitability for drinking purposes, as well as provide the water industry 230
with information regarding its treatment requirements, a treatability index (TI) was developed 231
and used. The TI is an easy to use indicator that is often used to assess water quality (Enitan-232
Folami et al., 2019). Here, the TI is defined as the ratio of the measured value (MV) of the 233
contaminant under study to its maximum allowed limit (MAL) for drinking water, as shown in 234
Equation (1): 235
 

 !"!
(1) 236
Therefore, Equation (1) suggests that: 237
• When the treatability index is >1, the water will require treatment to comply with the 238
prescribed limits, since it is not suitable for human consumption, i.e. MV>MAL. 239
Journal Pre-proof

14
• When the treatability index is =1, the water will require limited to no treatment to comply 240
with the prescribed limit. This water is marginally suitable for human consumption, i.e. 241
MV=MAL. 242
• When the treatability index is <1, the water will require zero treatment to comply with the 243
required limit and its suitable for human consumption. MV<MAL. 244
Equation (1) also implies that a wide range of both physicochemical (e.g. TDS, pH, 245
hardness, F, and Fe) and biological (E. coli and Coliform bacteria) parameters should be 246
examined to assess water quality and its suitability for drinking purposes. The quality of the 247
water will be then determine by the parameter that exhibits the highest TI value, i.e. if one 248
parameter has a TI higher than unity then water will be unsuitable for human consumption, 249
regardless of the values of the other parameters. 250
Therefore, treatment will be required to lower all TI values below unity before water 251
consumption. The type (physicochemical or biological) and extent of pollution (TI value) will 252
largely determine the most appropriate technology and the associate cost of water treatment. 253
This information is particularly important for the water industry, since in cases where local 254
communities are served by water boards these will typically foot the bill for water treatment. 255
Therefore, when the TI value is much higher than unity, the water boards should either treat the 256
water or not used it at all. Nonetheless, TI values are difficult to communicate to decision- and 257
policy-makers and local communities. For this reason, here we propose a simple to use water 258
quality classification model, tailored for the developing world. 259
The suitability of borehole water for drinking purposes was assessed and quantified by 260
using a simple, yet effective, classification model tailored for the needs of the developing world. 261
Specifically, here we propose the use of a classification index, which was developed in the 262
context of this work and by consulting Magalies Water, to classify water quality in South Africa 263
and further afield. Many indexes have been proposed for water quality classification (Kempster 264
Journal Pre-proof

15
et al., 1997). In this work, the South African National Standard (SANS) 241-2:2015 for drinking 265
water (SABS, 2015) was used to assess groundwater’s quality. These specifications are based 266
on the guidelines of the World Health Organization (WHO) for drinking water (WHO, 2017) and 267
therefore are in line with those proposed further afield, such as in China (AQSIQ, 2017). Finally, 268
the water quality was divided into four different classes, as is proposed elsewhere (Enitan-269
Folami et al., 2019). These classes are based on the TI results (Table 3) and were developed 270
bearing in mind the current situation of rural and peri-urban communities in RSA. As such, they 271
are easy to be communicated to those communities and across the developing world, while the 272
use of colour in each class further highlight the current status of the borehole water. Specifically, 273
with blue and green colour the excellent and good quality water are shown, respectively. On the 274
other hand, with yellow colour (warning) the acceptable quality and with red (danger) the 275
unacceptable quality are presented (Table 3). We found, after long conversations with members 276
of these local communities, that the use of colour and this simple classification can effectively 277
communicate the water quality results to the public. 278
Table 3: The proposed water quality classification index for domestic purposes, modified from 279
Enitan-Folami et al. (2019)). 280
Class 1 T<0.5 Excellent quality
Class 2 TI<0.75 but >0.5 Good quality
Class 3 TI<1 but >0.75 Acceptable quality
Class 4 TI>1 Unacceptable quality
281
Using the SANS 241-2:2015 limits for drinking water the concentration of each examined 282
pollutant was categorized into the proposed four classes, using the corresponding TI values, as 283
shown in Table 4. Specifically, in cases where the contaminant concentration is up to 50% lower 284
than the specified limit (TI value 0.5 or higher), then the water quality is classified as class one 285
Journal Pre-proof

16
(1). If the contaminant concentration is between 50 – 25% less than the specified limit (TI value 286
0.5 to 0.75), the water is classified as two (2), while if it is 25 – 0% less than the specified limit 287
(TI value from 0.75 to 1) then it is classified as three (3). Finally, in cases when the contaminant 288
concertation is higher than the specified limit (TI value > 1) then the water quality is classified as 289
four (4). 290
As mentioned above, for each contaminant the MAL value was taken from SANS 241-291
2:2015. The water quality classification, based on the TI values and for each examined 292
contaminant/pollutant, is shown in Table 4. To accommodate an easier analysis, the examined 293
contaminants/pollutants were divided into to two different categories. The first comprise the 294
physicochemical characteristics and the latter the biological contamination, typically referring to 295
faecal indicator bacteria which are present as a result of human or animal waste. For each 296
borehole the TI value of each contaminant/pollutant was then estimated and water quality was 297
classified into the four different classes (Table 4). The final quality of the water is then defined 298
by the contaminant/pollutant with the highest TI value, i.e. even if all examined TI values are 299
below 1 and only one value is above 1, then water quality will be classified as unacceptable 300
(class 4). 301
302
Table 4: The proposed water quality classification model. 303
Pollution
source
Determinants
Limits
Class 1
Class 2
Class 3
Class 4
Physicochemical
characteristics
TDS
1
,
200
<600
600
-
900
900
-
1
,
200
>1
,
200
pH
5.5
-
9.7
7
6 or 8
5 or 9
<5 or >10
EC
*
170
85
127.5
170
>170
Hardness*
300
150
225
300
>300
Journal Pre-proof

17
Magnesium (Mg)*
100
50
75
100
>100
Calcium (Ca)
*
300
150
225
300
>
300
Iron
(Fe)
2
,
000
1
,
000
1
,
500
2
,
000
>
2
,
000
Manganese
(Mn)
400
200
300
400
>
400
Chloride
(Cl)
300
150
225
300
>
300
Fluoride
(F)
1.5
0.75
1.125
1.5
>
1.5
Sulphate
(SO²
#
$
)
500
250
375 500 >500
Arsenic (As)
10
5
7.5
10
>10
Chromium (Cr)
50
25
37.5
50
>50
Nitrate
s (NO
3
)
11
5.5
8.25
11
>
11
Biological
contamination
E.
coli
cfu
/100 mL
0
0
0
>
0
Coliform
cfu
/100 mL
5
7.5
10
>
10
TPC
cfu
/1mL
500
750
1
,
000
>
1
,
000
* Organoleptic and aesthetic limit, the water can be consumed but it will be visually, aromatically 304
or palatably unacceptable. 305
306
307
3 Results and discussions 308
First, the results for the examined physicochemical are presented and then the results 309
for the faecal indicator bacteria (biological contamination), which, as a preamble, were found to 310
grossly impacting the borehole water quality, are discussed. 311
3.1 Physicochemical parameters affecting water quality 312
Journal Pre-proof

18
Regarding the physicochemical parameters affecting water quality, here pH, EC, Ca, 313
Mg, hardness, SO
42-
, Mn, Fe, As, Cr, F, and Cl were considered. Among them, the limits for EC, 314
Ca, Mg, and hardness are indicative, since they refer to water’s organoleptic / aesthetic quality, 315
i.e. the water can be consumed but it will be visually, aromatically or palatably unacceptable. 316
Therefore, they are examined separately. Overall, it was found that the organoleptic / aesthetic 317
quality of the borehole water is relatively poor, with many samples being above the SANS 241-318
2:2015 prescribed limits for drinking water (Table 5). Particularly, the main problem lies on the 319
very high hardness levels, giving TI values at the threshold (yellow colour) or above (red colour) 320
of the aesthetic limit (Figure 3). 321
322
Figure 3: The groundwater quality, using the four class, of the organoleptic / aesthetic 323
parameters under study (generated using GMT). 324
Journal Pre-proof

19
Furthermore, Figure 4 shows the relationship between Ca, Mg and hardness for all 325
examined borehole samples. As was expected, the levels of Ca and Mg are proportional to the 326
hardness level of the borehole water. Even though high Mg, Ca, and hardness levels can affect 327
groundwater’s organoleptic / aesthetic quality they do have also some advantages, since 328
naturally soft water can be corrosive to metal water pipes and tanks, thus negatively affecting 329
human health (particularly if lead pipes are present). Finally, the Ca and Mg levels are directly 330
related to the EC levels, which is reflected in the overall high EC values shown in Table 5. 331
332
Figure 4: Relationship between hardness, Ca, and Mg levels in the examined borehole 333
samples. 334
Regarding the remaining physicochemical parameters, which if above the prescribed 335
limits imply that water requires treatment before being consumed, it was identified that iron (Fe) 336
and manganese (Mn) concentrations were below the SANS prescribed SANS 241-2:2015 337
prescribed limits, as was the case for H, and SO
42-
(only in sample no 8 their concentrations 338
were above the limits). In many samples As concentrations were high, but below the limit. High 339
0
200
400
600
800
1000
1200
1400
0
50
100
150
200
250
300
350
Limits
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Hardness level
Ca and Mg levels
Borehole number
Total Calcium Total Magnesium Hardness Total
Journal Pre-proof

20
As levels in groundwater have been reported in South Africa, owing to its release from highly 340
mineralised rocks, such as arsenopyrite (FeAsS), arsenical oxide, and scorodite 341
(FeAsO4·2H2O) (Abiye and Bhattacharya, 2019). Furthermore, Cl and particularly F 342
concentrations were, in many cases, above the limits for drinking water (Figure 5). F can lead to 343
dental or skeletal fluorosis (WHO, 2004) and is typically traced back to the existence of fluoride-344
containing minerals. Its release to groundwater is promoted when the pH and HCO
3
is high and 345
Ca and Na concentrations are low (Su et al., 2019). In general F concentrations vary, depending 346
on the type of rock the groundwater pass through, but usually do not exceed the 10 mg L
-1
347
reference (WHO, 2004), as was also the case here. However, in some African countries, such 348
as the United Republic of Tanzania, very high concentrations, of the order of 8 mg L
-1
have 349
been reported (WHO, 2004), which was also the case in sample no 33 (F = 8.6 mg L
-1
). For 350
context, in Panipat, Haryana, India F concentrations in groundwater ranged from 0.2 to 6.9 mg 351
L
-1
(mean value 1.4 mg L
-1
) (Kaur et al., 2020). In our case study they ranged from as low as 0 352
to as high as 8.6 mg L
-1
, with a mean value of 1.1 mg L
-1
. The problem of high F concentrations 353
in groundwater from North West Province, South Africa has also been highlighted elsewhere 354
(Monyatsi, 2012). Similarly, chloride was identified to be above the SANS limits in few 355
boreholes, suggesting that the excess levels should be removed before consumption. Finally, 356
the nitrate (NO%#) levels were above the limit in 22 out of the 42 examined samples, with sample 357
no 8 having an order of magnitude higher concentration than the SANS drinking water limit. 358
Overall, in terms the examined physiochemical, the water quality was poor in many of the 359
examined samples (Figure 5 and Table 5). 360
Journal Pre-proof

21
361
Figure 5: The groundwater quality, using the four class, of the physicochemical parameters 362
under study (generated using GMT). 363
Even though NO%# can naturally occur in groundwater, in most cases elevated 364
concentrations (> 3 mg L
-1
) are attributed to human activities, such as from ammonium fertilizers 365
and manure, industrial wastes, food processing wastes, and wastewater (Nebraska Extension, 366
2015). In the examined area industrial activities and intensive agriculture or animal farming does 367
not take place, apart from some agricultural activity in the boarder of North West with Limpopo 368
province. Therefore, the high nitrate concentrations, which in in 22 out of the 42 borehole 369
samples were well above the drinking water limit (Table 5) are attributed, by and large, to onsite 370
sanitation systems, i.e. to wastewater releases / leachates which pollute the aquifer. This is 371
further supported by the high concentrations of the examined faecal indicator bacteria, which 372
are discussed below. Furthermore, NO%# concentrations were found to range from 0.16 to 104 373
Journal Pre-proof

22
mg L
-1
, with the mean value being 15.2 mg L
-1
, while, for context, in Panipat, Haryana, India 374
NO%# concentrations in groundwater ranged from 0.5 to 69 mg L
-1
, with the mean value being 375
12.8 mg L
-1
(Kaur et al., 2020). 376
Table 5: The concentrations of physicochemical parameters under study. 377
Aesthetic limit*
class
Drinking water limit
class
Param
eter
EC
Ca
Mg
Hard
-
ness
pH
SO
42-
Mn
Fe
As
Cr
F
Cl
NO
3
&
&&
&
Units
mS m
-1
mg L
-1
mg L
-1
mg L
-1
-
mg L
-1
µ
g L
-
1
µ
g L
-1
µ
g L
-1
µ
g
L
-1
mg L
-
1
mg L
-1
mg L
-
1
Limit
170
300
100
≤300
≥5 to
≤9.7
500
400
2
,
000
10
50
1.5
300
11
No
1
37.0
15.0
7.5
69.0
1
6.5
4.3
4.3
4.4
0.5
0.8
0.1
43.0
21
4
2
85.0
40.0
35.0
244.0
3
8.3
37.0
2.3
3.5
4.7
0.8
0.6
91.0
12
4
3
78.0
70.0
28.0
291.0
3
7.4
38.0
2.3
54.0
1.0
0.8
1.7
58.0
5
4
4
49.0
24.0
40.0
226.0
3
7.4
35.0
2.3
24.0
1.2
0.8
0.1
52.0
12
4
5
30.0
16.0
5.8
63.0
1
7.4
6.6
9.2
119.0
1.0
0.8
0.1
39.0
14
4
6
58.0
51.0
15.0
190.0
2
7.1
13.0
2.7
56.0
0.9
0.8
0.3
58.0
30
4
7
201.0
66.0
27.0
277.0
3
6.7
318.0
6.8
422.0
3.6
0.8
0.2
326.0
31
4
8
262.0
176.0
179.0
1
,
176.0
4
8.0
515.0
2.3
237.0
7.0
0.8
0.0
285.0
104
4
9
83.0
37.0
26.0
199.0
2
7.6
85.0
12.0
3.5
6.4
0.8
0.2
98.0
0.16
2
10
77.0
75.0
37.0
339.0
4
7.7
48.0
2.3
4.4
1.0
0.8
0.1
62.0
11
4
11
80.0
41.0
27.0
214.0
2
7.8
82.0
9.1
4.4
5.9
0.8
0.1
110.0
0.22
2
12
77.0
41.0
21.0
191.0
2
6.5
71.0
49.0
84.0
2.2
0.8
0.1
121.0
21
4
13
117.0
45.0
88.0
477.0
4
8.0
47.0
4.3
3.5
3.1
0.8
0.2
77.0
31
4
14
56.0
34.0
43.0
261.0
3
8.5
41.0
2.3
40.0
1.4
0.8
0.1
63.0
3.9
1
15
58.0
30.0
14.0
130.0
1
6.8
63.0
2.3
4.4
0.2
0.8
0.1
69.0
4.7
1
16
47.0
41.0
20.0
185.0
2
8.2
5.4
2.3
4.4
3.4
0.8
0.3
23.0
8.8
3
17
35.0
29.0
10.0
113.0
1
7.9
3.0
2.3
29.0
0.7
0.8
0.4
24.0
8.7
3
18
117.0
84.0
13.0
262.0
2
6.9
66.0
2.3
3.5
2.8
0.8
0.9
186.0
0.77
2
19
90.0
86.0
28.0
330.0
4
7.4
61.0
3.4
120.0
1.8
0.8
0.8
88.0
14
4
20
131.0
95.0
30.0
359.0
4
7.2
39.0
2.3
27.1
1.6
0.8
2.3
123.0
17
4
21
238.0
89.0
65.0
490.0
4
7.5
24.0
2.3
3.5
5.0
1.8
1.1
381.0
58
4
22
83.0
60.0
23.0
243.0
3
7.4
8.8
2.3
3.5
1.2
0.9
1.9
47.0
8
4
23
197.0
122.0
63.0
562.0
4
7.3
23.0
2.3
3.5
3.5
2.0
0.9
284.0
14
4
24
152.0
92.0
41.0
398.0
4
7.1
69.0
2.3
3.5
3.1
1.9
0.9
134.0
12
4
25
125.0
83.0
26.0
314.0
4
7.7
76.0
10.0
16.3
3.4
0.8
2.9
159.0
3.6
4
26
229.0
106.0
47.0
458.0
4
7.2
122.0
72.0
12.5
3.2
0.8
1.0
343.0
57
4
27
117.0
82.0
30.0
328.0
4
7.6
73.0
2.3
34.8
2.2
1.1
1.4
84.0
16
4
28
64.0
55.0
12.0
187.0
2
6.9
8.1
2.3
3.5
1.4
0.8
2.9
64.0
14
4
29
94.0
32.0
11.0
126.0
1
8.0
61.0
9.8
440.0
1.2
0.8
5.1
97.0
0.16
4
30
65.0
41.0
16.0
170.0
2
6.6
7.8
9.9
734.0
1.5
1.0
1.4
79.0
17
4
Journal Pre-proof

23
31
68.0
61.0
15.0
215.0
2
7.5
8.9
2.3
3.5
1.0
0.8
3.1
38.0
3.8
4
32
25.0
8.4
6.8
49.0
1
6.5
2.4
2.3
3.5
0.7
0.8
0.9
21.0
5
2
33
80.0
22.0
6.0
79.0
1
7.8
17.0
38.0
4.6
2.3
0.8
8.6
124.0
0.97
4
34
20.0
8.8
4.0
38.0
1
6.7
1.6
6.5
67.4
1.1
1.1
1.9
13.0
4
4
35
104.0
96.0
72.0
535.0
4
7.7
70.0
2.3
25.8
1.6
1.3
0.3
19.0
22
4
36
232.0
219.0
87.0
905.0
4
6.9
27.0
2.3
85.7
8.1
4.3
1.4
392.0
18
4
37
23.0
9.9
6.6
52.0
1
6.3
1.6
2.3
3.5
0.0
0.8
0.1
23.0
18
4
38
24.0
11.0
7.7
59.0
1
6.7
1.6
2.3
3.5
0.1
0.8
0.0
26.0
2.7
1
39
15.0
4.6
4.3
29.0
1
6.2
1.6
15.0
3.5
0.1
0.8
0.1
13.0
3.6
1
40
20.0
6.7
5.2
38.0
1
6.6
1.6
2.3
3.5
0.6
0.8
0.1
24.0
5.2
1
41
15.0
7.1
5.2
39.0
1
6.4
1.6
2.3
3.5
0.2
0.8
0.1
8.4
3.1
1
42
18.0
10.0
5.7
49.0
1
6.7
1.6
16.0
65.3
0.6
0.8
0.1
11.0
1.6
1
Min
15
4.6
4
29
-
6.2
1.6
2.3
3.5
0
0.8
0
8.4
0.16
-
Max
262
219
179
1176
-
8.5
515
72
734
8.1
4.3
8.6
392
104
-
Mean
89.9
55.3
29.9
260.9
-
7.3
52.1
8.0
66.1
2.2
1.0
1.1
104.3
15.2
-
* water can be consumed but it will be visually, aromatically, or palatably unacceptable.
378
3.2 Biological contamination 379
The microbiological contaminants examined in the context of this work are the faecal 380
indicator bacteria E. coli, along with two indicators of faecal contamination, i.e. total coliform and 381
total plate count. As a preamble, the indicators for faecal contamination were found to be well 382
above the SANS limits across the examined samples (Figure 6). Specifically, in 20 out of the 42 383
boreholes E. coli bacteria, which are commonly found in human and animal intestines, and by 384
extension in their faeces, were identified. Alarmingly, E. coli concentrations as high as 195 per 385
100 mL were measured, which highlights the poor state of groundwater and the extent of 386
waterborne pathogen contamination. 387
Journal Pre-proof

24
388
Figure 6: The groundwater quality, using the four class, of the physicochemical parameters 389
under study (generated using GMT). 390
The same pattern, in terms of concentration compared to the prescribed drinking water 391
limit, was identified and in the other two examined indicators of faecal contamination (Table 6). 392
The boreholes that were identified to suffer from E. coli contamination, did also exhibit high total 393
coliform and total plate count concentrations, as was expected. However, the remaining 22 out 394
of the 42 boreholes samples, where E. coli contamination was not identified, did suffer from high 395
total coliform and total plate count concentrations, well above SANS limits for drinking water 396
limits. As a result, in all examined boreholes faecal contamination was identified, which, apart 397
from borehole no 37 (Class 3), render groundwater unsuitable for human consumption (Class 398
4). As such, it can be inferred that the groundwater that is used by rural and semi-urban 399
communities in RSA is grossly polluted by human actives, with the pollution possibly tracing 400
Journal Pre-proof

25
back to those communities (on-site sanitation systems), which is also supported by the high 401
NO
3-
. 402
Table 6: The microbiological contaminants under study, i.e. E. coli, Total Coliform, and Total 403
Plate Count. 404
Parameter
E. coli
Total Coliform
Total plate count
Class
Units
Count per 100mL
Count per 100mL
Count
per mL
Limit
0
10
1
,
000
Number
1
0
180
3
,
600
4
2
0
0
1
,
170
4
3
1
340
2
,
360
4
4
1
56
1
,
350
4
5
51
93
23
,
100
4
6
12
283
20
,
100
4
7
195
410
15
,
300
4
8
2
18
2
,
150
4
9
0
0
1
,
010
4
10
0
14
830
4
11
0
48
3
,
700
4
12
4
305
1
,
510
4
13
1
430
9
,
800
4
14
3
3
890
4
15
0
34
120
4
16
0
50
11
,
300
4
17
0
50
15
,
600
4
18
0
58
1
,
860
4
19
79
2
,
500
25
,
400
4
20
26
460
15
,
500
4
21
0
1
14
,
100
4
22
14
52
42
,
300
4
23
0
165
1510
4
24
0
38
1840
4
25
0
130
11
,
500
4
26
0
65
41
,
100
4
27
14
210
20
,
100
4
28
0
115
10
,
300
4
29
35
140
1
,
100
4
30
150
225
13
,
700
4
31
2
41
1
,
230
4
32
0
28
19
,
200
4
Journal Pre-proof

26
33
36
135
5
,
600
4
34
6
76
14
,
800
4
35
80
100
74
,
100
4
36
0
98
28
,
200
4
37
0
5
950
3
38
0
1
14
,
000
4
39
0
12
680
4
40
0
51
1
,
850
4
41
1
120
9
,
700
4
42
0
89
6
,
500
4
Min
0
0
120
Max
195
2
,
500
74
,
100
Mean
17.0
172.1
11
,
690.7
405
3.3 Groundwater quality 406
Large variations in the levels of the examined pollutants were observed across the 407
examined boreholes, with microbiological contaminants being the main concern. As a result all 408
examined samples were not fit for human consumption without treatment. To eliminate 409
biological contamination, a simple chlorination in the central water reservoir, where the 410
groundwater from the communal boreholes is gathered, might suffice. Furthermore, in South 411
Africa solar light is abundant and as such pathogen inactivation can also be achieved by means 412
of solar&driven oxidation, such as solar-Fenton oxidation (Ioannou-Ttofa et al., 2017), solar 413
photocatalytic treatment (Monteagudo et al., 2018), or even by means of a simple solar 414
disinfection (SODIS) (Porley et al., 2020). 415
Specifically, photocatalysis has arisen as a promising treatment technique that provides 416
safe and clean water. Its two main advantages are: i) its ability to destroy contaminants to 417
mineral end-products, without transferring contamination from one phase to another, and ii) that 418
its main “reagent” is irradiation itself in the presence of a photocatalyst (typically TiO2) (Foteinis 419
and Chatzisymeon, 2020). These, along with its simplicity, versatility, and robustness make it a 420
perfect candidate process for developing countries with high sunshine duration. More 421
Journal Pre-proof

27
specifically, the photocatalytic process finds optimal application under controlled conditions, i.e. 422
when using an artificial irradiation source (Davididou et al., 2018), however, this might impose 423
on the treatment system simplicity and versatility, with natural solar irradiance being more 424
suitable for developing countries. To enable local communities to access safe potable water in 425
these rural and peri-urban areas, a simple SODIS technique, which has showed successful 426
removal of a range of contaminants in similar settings, such as in rural India (Porley et al., 427
2020), could be employed. Larger systems, operating at village- or even at semi-industrial level, 428
such using a solar compound parabolic collector pilot plant for photo-Fenton treatment 429
(Davididou et al., 2019), could be used. The added benefit of such systems is that they can be 430
designed to operate at standalone mode, i.e. using PV panels and battery units instead of 431
electricity from the grid (Foteinis et al., 2018). This can be of outmost importance in remote and 432
insular areas, where electricity from the grid is not available. 433
However, given the physicochemical pollution identified in the majority of the samples 434
more robust treatment is required to remove the identified contaminants. For communal 435
boreholes, where water is pump into a central reservoir, water dilution might be a feasible 436
solution to reach the desired levels. For example, in communal boreholes F and Cl might be 437
diluted in the large water volume of the central reservoir where water is pumped from all 438
communal boreholes, thus avoiding the need for their removal. For the case of F this might be 439
beneficial, since F levels in the range of 0.5 to 1.0 mg L
-1
can prevent dental caries (tooth 440
decay) (WHO, 2004). Nonetheless, this is not the case for boreholes which are used by 441
individuals, where treatment to remove these contaminants is a perquisite before being used as 442
a source of potable water. Ion exchange resins or gravity membrane filters can be used to 443
remove physicochemical pollutants, such as NO%#, however, for the developing world simple 444
and low cost solutions, such as carbon (Porley et al., 2020) or biobased cellulose fibers (Tursi A, 445
2018) for contaminants filtering and adsorption, might be more preferable. In future works of our 446
Journal Pre-proof

28
group the feasibility of introducing the abovementioned treatment technologies, among others, 447
and the willingness of local communities to pay for improving the quality of drinking water will be 448
comprehensively examined. 449
Overall, the proposed TI, along with the borehole classification model, highlighted that 450
groundwater is unfit for human consumption, since pathogens (Table 6) along with other 451
pollutants (Table 5) were present. Since local communities rely on the existing boreholes for 452
potable water, the importance of assessing groundwater’s suitability with these simple tools is 453
highlighted. Furthermore, the water industry also requires information regarding the quality of 454
each borehole, to identify if and what type of treatment is required, either at each borehole or 455
directly at the main water reservoir, before distribution. Therefore, the TI and the borehole 456
classification model can be used to: i) identify the type and extent of the pollution, ii) suggest the 457
type of treatment before consumption, and iii) used to prioritise investment in water and 458
wastewater infrastructure. 459
Table 7: The water quality of each examined borehole. 460
number
Physico
-
chemical
Biological
Classification
number
Physico
-
chemical
Biological
Classification
1
4
4
4
22
4
4
4
2
4
4
4
23
4
4
4
3
4
4
4
24
4
4
4
4
4
4
4
25
4
4
4
5
4
4
4
26
4
4
4
6
4
4
4
27
4
4
4
7
4
4
4
28
4
4
4
8
4
4
4
29
4
4
4
9
2
4
4
30
4
4
4
10
4
4
4
31
4
4
4
11
2
4
4
32
2
4
4
12
4
4
4
33
4
4
4
13
4
4
4
34
4
4
4
14
1
4
4
35
4
4
4
15
1
4
4
36
4
4
4
16
3
4
4
37
4
3
4
17
3
4
4
38
1
4
4
Journal Pre-proof

29
18
2
4
4
39
1
4
4
19
4
4
4
40
1
4
4
20
4
4
4
41
1
4
4
21
4
4
4
42
1
4
4
461
4 Conclusions and recommendations 462
Groundwater’s quality in rural and peri-urban RSA was comprehensively examined. In 463
total 42 boreholes, which provide drinking water to local communities in North West Province 464
were sampled. It was identified that groundwater’s organoleptic / aesthetic quality was, in many 465
instances, above the SANS 241-2:2015 suggested limits, with hardness and EC primarily 466
affected. Even though this does not imply that water is unfit for human consumption, it suggests, 467
however, that its smell, taste, color, and turbidity might be affected. However, in many instances 468
the local geology was found to grossly affect groundwater’s physicochemical characteristics, 469
with Cl and particularly F having concentrations well above the drinking water limits, suggesting 470
the need for their removal before water consumption. Due to South Africa’s underlying geology, 471
which contains highly mineralised rocks, elevated As concentrations were identified, but below 472
the drinking water limit. 473
The main problem, by and large, was biological pollution, with the examined indicators of 474
faecal contamination being well above the prescribed drinking water limits in all but one 475
boreholes. Very high NO%# concentrations, up to 104 mg L
-1
, were identified in the majority of 476
the examined samples, with 22 boreholes exhibiting concentrations well above the drinking 477
water limit. This, along with the identified indicators of faecal contamination suggest that onsite 478
sanitation systems are grossly pollute the aquifer. Nonetheless, this problem has not been 479
acknowledged by the local communities, which erroneously perceive groundwater as safe to 480
consume without any prior treatment, which is typical for the developing world. To raise 481
awareness among the local communities and provide the water industry with tools to effectively 482
Journal Pre-proof

30
and sustainably identify and address the problem a treatability index along with a borehole 483
classification model were developed. 484
From the result of the model, it is highly recommended that different treatment techniques 485
should be employed to purify groundwater before consumption, focused on the microbial 486
contaminants. To this end, solar oxidation might be promising, given the high annual sunshine 487
duration and solar-radiation levels in RSA. Ion exchange and adsorption could be explored for 488
the removal of nitrates and fluoride, respectively, or for the removal of other contaminants. 489
Therefore, establishing a two-staged approach, i.e. the removal of organic and inorganic 490
contaminants and then disinfection, should be considered. Nonetheless, to effectively address 491
the problem local communities should move away from unlined or semi-lined onsite sanitation 492
systems, which could be achieved through capacity building and investment in wastewater 493
infrastructure. Finally, future works should also focus on the groundwater quality in relation to 494
depth and geology. 495
496
5 Acknowledgements 497
The authors would like to thank Magalies Water and the University of South Africa (UNISA) for 498
providing the necessary resources to successfully complete this extensive project. 499
Journal Pre-proof

31
References
Abiye, T.A., Bhattacharya, P., (2019). Arsenic concentration in groundwater: Archetypal study
from South Africa. Groundwater for Sustainable Development 9, 100246.
https://doi.org/10.1016/j.gsd.2019.100246
Claasen, T., Lewis, E., (2017). Nitrate and bacteriological assessment of groundwater in
Omaheke region, Namibia. International Science and Technology Journal of Namibia 10, 56-67.
http://hdl.handle.net/11070/2194
Cobbing, J.E., Hobbs, P.J., Meyer, R., Davies, J., (2008). A critical overview of transboundary
aquifers shared by South Africa. Hydrogeology Journal 16, 1207-1214.
https://doi.org/10.1007/s10040-008-0285-2
Das, R., Laishram, B., Jawed, M., (2019). Perception of groundwater quality and health effects
on willingness to procure: The case of upcoming water supply scheme in Guwahati, India.
Journal of Cleaner Production 226, 615-627. https://doi.org/10.1016/j.jclepro.2019.04.097
Davididou, K., Chatzisymeon, E., Perez-Estrada, L., Oller, I., Malato, S., (2019). Photo-Fenton
treatment of saccharin in a solar pilot compound parabolic collector: Use of olive mill wastewater
as iron chelating agent, preliminary results. Journal of Hazardous Materials 372, 137-144.
https://doi.org/10.1016/j.jhazmat.2018.03.016
Davididou, K., Nelson, R., Monteagudo, J.M., Durán, A., Expósito, A.J., Chatzisymeon, E.,
(2018). Photocatalytic degradation of bisphenol-A under UV-LED, blacklight and solar
irradiation. Journal of Cleaner Production 203, 13-21.
https://doi.org/10.1016/j.jclepro.2018.08.247
Dennis, I., Dennis R., (2012). Climate change vulnerability index for South African aquifers.
Water SA 38, 3, 417 – 426. DOI: 10.4314/wsa.v38i3.7
Journal Pre-proof

32
DWS, Department of Water and Sanitation (2017). Benchmark of Water Losses, Water Use
Efficiency and Non-Revenue Water in South African Municipalities (2004/05 − 2015/16).
Pretoria, South Africa, p. 57.
Enitan-Folami, A.M., Mutileni, N., Odiyo, J.O., Swalaha, F.M., Edokpayi, J.N., (2019).
Hydrochemical, bacteriological assessment, and classification of groundwater quality in
Thulamela Municipality, South Africa: potential health risk. Human and Ecological Risk
Assessment: An International Journal 1-15. https://doi.org/10.1080/10807039.2019.1644153
Foteinis, S., Chatzisymeon, E., (2020). Heterogeneous photocatalysis for water purification, in
Micro and Nano Technologies, Nanostructured Photocatalysts, 75-97, ISBN 9780128178362.
Foteinis, S., Monteagudo, J.M., Durán, A., Chatzisymeon, E., 2018. Environmental sustainability
of the solar photo-Fenton process for wastewater treatment and pharmaceuticals mineralization
at semi-industrial scale. Science of The Total Environment 612, 605-612.
https://doi.org/10.1016/j.scitotenv.2017.08.277
Graham, J.P., Polizzotto, M.L., (2013). Pit latrines and their impacts on groundwater quality: a
systematic review. Environ Health Perspect 121, 521-530. https://doi.org/10.1289/ehp.1206028
Greenberg, A.E., Clesceri, L.S., Eaton, A.D., (2010). Standard Methods for the Examination of
Water and Wastewater. American Public Health Association, Washington DC.
Ioannou-Ttofa, L., Foteinis, S., Chatzisymeon, E., Michael-Kordatou, I., Fatta-Kassinos, D.,
2017. Life cycle assessment of solar-driven oxidation as a polishing step of secondary-treated
urban effluents. Journal of Chemical Technology and Biotechnology 92, 1315-1327.
https://doi.org/10.1002/jctb.5126
Journal Pre-proof

33
He, X., Li, P., Ji, Y., Wang, Y., Su, Z., Elumalai, V., (2020). Groundwater Arsenic and Fluoride
and Associated Arsenicosis and Fluorosis in China: Occurrence, Distribution and Management.
Exposure and Health 12, 355-368. https://doi.org/10.1007/s12403-020-00347-8
Kapulu, M., Lusambo, E., Haankuku, C., (2013). Iron removal in borehole water: A case study of
Luapula Province. 116 pp. ISBN-10:3659400580
Kaur, L., Rishi, M.S., Siddiqui, A.U., (2020). Deterministic and probabilistic health risk
assessment techniques to evaluate non-carcinogenic human health risk (NHHR) due to fluoride
and nitrate in groundwater of Panipat, Haryana, India. Environmental Pollution 259, 113711.
https://doi.org/10.1016/j.envpol.2019.113711
Kempster, P.L., Van Vliet, H.R., Kühn, A., (1997). The need for guidelines to bridge the gap
between ideal drinking-water quality and that quality which is practically achievable and
acceptable. Water SA 23, 163-167.
Lapworth, D.J., Baran, N., Stuart, M.E., Ward, R.S., (2012). Emerging organic contaminants in
groundwater: A review of sources, fate and occurrence. Environmental Pollution 163, 287-303.
https://doi.org/10.1016/j.envpol.2011.12.034
Li, P., Wu, J., Tian, R., He, S., He, X., Xue, C., Zhang, K., (2018). Geochemistry, Hydraulic
Connectivity and Quality Appraisal of Multilayered Groundwater in the Hongdunzi Coal Mine,
Northwest China. Mine Water and the Environment 37, 222-237.
https://doi.org/10.1007/s10230-017-0507-8
Manga, M., Bartram, J., Evans, B.E., (2020). Economic cost analysis of low-cost sanitation
technology options in informal settlement areas (case study: Soweto, Johannesburg).
International Journal of Hygiene and Environmental Health 223, 289-298.
https://doi.org/10.1016/j.ijheh.2019.06.012
Journal Pre-proof

34
Masindi, V., (2020). Dataset on physicochemical and microbial properties of raw water in four
drinking water treatment plants based in South Africa. Data in Brief, 105822.
https://doi.org/10.1016/j.dib.2020.105822
Monteagudo, J.M., Durán, A., Chatzisymeon, E., San Martín, I., Naranjo, S., (2018). Solar
activation of TiO2 intensified with graphene for degradation of Bisphenol-A in water. Solar
Energy 174, 1035-1043. https://doi.org/10.1016/j.solener.2018.09.084
Monyatsi, L., (2012). Assessment of Groundwater Quality in the Rural Areas of the North West
Province, South Africa. Scientific Research and Essays 7(8), 903-914
https://doi.org/10.5897/SRE11.906
Musingafi, M., (2014). Fresh water sources pollution: A human related threat to fresh water
security in South Africa. Journal of Public Policy and Governance 1(2), 72-81.
Nebraska Extension, 2015. Irrigation and Nitrogen Management, in: Kranz, W.L. (Ed.). Institute
of Agriculture and Natural Resources, Lincoln, Nebraska, USA, p. 117.
Ncube, E.J., Schutte, C.F., (2005). The occurrence of fluoride in South African groundwater: A
water quality and health problem.Water SA 31(1) 35-40. https://doi.org/10.4314/wsa.v31i1.5118.
Palamuleni, L., Akoth, M., (2015). Physico-Chemical and Microbial Analysis of Selected
Borehole Water in Mahikeng, South Africa. International Journal of Environmental Research and
Public Health 12, 8619-8630. https://doi.org/10.3390/ijerph120808619
Porley, V., Chatzisymeon, E., Meikap, B.C., Ghosal, S., Robertson, N., 2020. Field testing of
low-cost titania-based photocatalysts for enhanced solar disinfection (SODIS) in rural India.
Environmental Science: Water Research and Technology 6, 809-816.
https://doi.org/10.1039/c9ew01023h
Journal Pre-proof

35
Potgieter, N., Karambwe, S., Mudau, L.S., Barnard, T., Traore, A., 2020. Human Enteric
Pathogens in Eight Rivers Used as Rural Household Drinking Water Sources in the Northern
Region of South Africa. International Journal of Environmental Research and Public Health 17.
https://doi.org/10.3390/ijerph17062079
Rezaei, A., Hassani, H., Hassani, S., Jabbari, N., Fard Mousavi, S.B., Rezaei, S., (2019).
Evaluation of groundwater quality and heavy metal pollution indices in Bazman basin,
southeastern Iran. Groundwater for Sustainable Development 9, 100245.
https://doi.org/10.1016/j.gsd.2019.100245
SABS Standards Division, 2015. South African National Standard (SANS), Drinking water - Part
2: Application of SANS 241-1. ISBN 978-0-626-29841-8.
AQSIQ, General Administration of Quality Supervision, Inspection and Quarantine of the
People’s Republic of China. Quality Standard for Groundwater (GB/T 14848-2017) AQSIQ;
Beijing, China: 2017. (In Chinese)
Shivendra, B.T. and Ramaraju, H.K., (2015). Impact of Onsite Sanitation System on
Groundwater in Different Geological Settings of Peri Urban Areas. Aquatic Procedia 4, 1162-
1172. https://doi.org/10.1016/j.aqpro.2015.02.148
Su, H., Wang, J., Liu, J., (2019). Geochemical factors controlling the occurrence of high-fluoride
groundwater in the western region of the Ordos basin, northwestern China. Environmental
Pollution 252, 1154-1162. https://doi.org/10.1016/j.envpol.2019.06.046
Taonameso, S., Mudau, L.S., Traoré, A.N., Potgieter, N., (2018). Borehole water: a potential
health risk to rural communities in South Africa. Water Supply 19, 128-136.
https://doi.org/10.2166/ws.2018.030
Journal Pre-proof

36
Tessema, A., Mengistu, H., Chirenje, E., Abiye, T., Demlie, M., (2011). The relationship
between lineaments and borehole yield in North West Province, South Africa: Results from
geophysical studies. Hydrogeology Journal 20. https://doi.org/10.1007/s10040-011-0803-5
The United Nations ( 2018). The Sustainable Development Goals Report, in: The United
Nations, Department of Economic and Social Affairs. New York, p. 40.
Tursi, C.E., Chidichimo, F., Beneduci, A., Chidichimo, G., (2018). Removal of Endocrine
Disrupting Chemicals from Water: Adsorption of Bisphenol-A by Biobased Hydrophobic
Functionalized Cellulose. International Journal of Environmental Research and Public Health
15, 2419. https://doi.org/10.3390/ijerph15112419
UN General Assembly, 2010. The human right to water and sanitation: resolution / adopted by
the General Assembly, 3 August 2010, A/RES/64/292.
Wessel, P., Luis, J.F., Uieda, L., Scharroo, R., Wobbe, F., Smith, W.H.F. and Tian, D., (2019).
The Generic Mapping Tools Version 6. Geochemistry, Geophysics, Geosystems 20, 5556-5564.
WHO, World Health Organization, 2004. Fluoride in Drinking-water, WHO/SDE/WSH/03.04/96.
United Kingdom, p. 17.
WHO, World Health Organization, 2017. Guidelines for drinking-water quality, Fourth edition
incorporating the first addendum, 4th ed + 1st add. Geneva Switzerland. ISBN 9789241549950.
Journal Pre-proof

- The water quality of a large number of boreholes was examined in South Africa
- Groundwater was heavily polluted and was in need of treatment before consumption
- Result suggested that wastewater from onsite systems leak to the groundwater
- Groundwater can be an environmental reservoir, for bacteria and virions transmission
Journal Pre-proof

Declaration of interest: none
Journal Pre-proof