Abstract

A water safety plan (WSP) is a preventive comprehensive risk assessment and management approach to ensuring the safety of a drinking water supply from source to tap for public health protection. The concept was introduced in the last decade in international guidance documents and has been applied widely across a varied range of water supply systems, particularly, the public water utilities and to a lesser extent towards small systems. Mainstreaming water safety intervention for small systems however, would ensure safe household water to a wider population, alleviate poverty and hunger through water for use in support of livelihood activities, and help towards achieving the sustainable development goals. Self-supply hand-dug wells in Abeokuta, Nigeria, were assessed using the step-by-step World Health Organization WSP model, mainly from the relevant system assessment to operational monitoring and management procedures. This paper reviewed the methodology of water safety planning and flagged the issue of ‘who’ conducts WSP for small systems. The paper also evaluated major control measures critical to self-supply and suggested an apt WS planning model for the systems. The WSP framework for self-supply systems incorporated an institutional aspect for WSP coordination.

INTRODUCTION

Small systems are the non-utility-managed water systems. Embedded in the general description of small water systems are self-supply systems (SS), such as hand-dug wells or rainwater harvesting, which have remained a coping strategy at the household level for millions of the population not served with tap or communal water (Sutton 2004; Foster 2008). Self-supplies are private initiatives by individuals or households to improve their own water provisions, without waiting for help from government or non-governmental organisations (Carter 2006). The systems are also generally unregulated (Oluwasanya et al. 2011).

Constant calls have made over the years to recognise and mainstream SS as one of the water supply strategies (Sutton 2004; Carter 2006; Oluwasanya et al. 2011). In recent times, upgraded self-supply sources have been promoted as formal service-delivery models for varied groups such as pro-poor or rural water supply interventions to help achieve universal access (Mekonta et al. 2015). It is also endorsed as group-based self-supply sources to reach improved service-level targets in informal settlements in urban and isolated areas (Mekonta et al. 2015; Gowing et al. 2016). Promoting self-supply as a formal service-delivery model necessitates the need to engage in water safety planning for such sources. Ensuring water safety planning for such small systems is also embedded in, and crucial to achieving, the current global sustainable development goals (Maurice 2013; United Nations 2014).

Water safety plans (WSP) as described by the World Health Organization (WHO 2004, 2011) are a preventive risk assessment and management approach to ensuring the safety of a drinking-water supply from source to tap for public health protection. The WHO WSP guidance (Davison et al. 2005; Bartram et al. 2009), which highlighted a stepwise process (currently 11 steps) for the development of WSP, is generally focused on applications to public and community water supplies, with distinct utility-managed features (catchment, treatment and distribution networks), and with an identifiable provider.

The WHO WSP framework does not however fit the realities of SS, as such sources are usually non-piped and non-utility managed. Similarly, the owners and the operators of the systems do not have the necessary skills to assess the potential risks of their water sources. Rather, household perception of the wholesomeness of water sources determines the handling (collection and storage) and usage of water from the sources. Generally, sources bearing water that is (suspended) sediment- or particle-free is perceived to be good. Hence the need for an appropriate water safety planning framework for SS.

To suggest an apt water safety planning model for SS, this paper reviewed the method of the WHO WSP and flagged the issue of ‘who’ conducts WSP for small systems. The paper also evaluated the major control measures that are critical to SS for inclusion in the recommended water safety framework. Control measures, in the context of WSP, are those steps in drinking-water supply that directly affect water quality and that collectively ensure that drinking water consistently meets health-based targets (Schmoll et al. 2006). The paper is part of wider research that assessed SS in Abeokuta, Ogun State, Nigeria, towards the development of a suitable WSP for the systems. Abeokuta, the capital of Ogun State, is located in southwest Nigeria.

MATERIALS AND METHODS

Towards the formation of a self-supply water safety framework

The owners and users of the wells selected for water quality investigation in the wider research were retained for semi-structured interviews. The main respondent inclusion criterion was being an owner or user of a self-supply well. One hundred and five (105) respondents were interviewed: nine key informants, such as senior officials of the government institutions that were considered relevant to water management and safety, and 96 interviewees (24 source owners, 68 resident users and four non-resident users). Questions were asked to obtain respondents' perceptions on the water source, operations, management, maintenance, household usage and handling, and on individual health and safety (Oluwasanya et al. 2011; Oluwasanya & Carter 2013). In part, the study identified 20 control measures (Table A1 in Appendix A, available with the online version of this paper) and spotted the need for a coordinating institution to oversee self-supply water safety planning.

Consequently, the WHO WSP was reviewed and the stepwise processes summarised into four main components as a proposed WSP model for SS (Figure 1). Three of the components were as contained in the WHO WSP outline. The fourth component was suggested. The components are systems assessments (Box A), operational monitoring (Box B), management and communication (Box C), and institutional framework (Box D) (Figure 1). The underlying assumption of the proposed framework was that the fourth component (Box D) must be functional for self-supply WSP to materialise (Figure 1). Box D represents the most critical component on which the other boxes depend. For instance, the question of ‘who’ resides in Box D – who initiates safety plans; who describes the system or carries out risk assessment; who coordinates source management; who is responsible for implementing control measures or organising the supporting programs, among further questions. Noting the general lack of relevant expertise of the self-supply owners/users.

Figure 1

Proposed water safety framework for SS (research-driven).

Figure 1

Proposed water safety framework for SS (research-driven).

To verify the proposed framework, a structured interview was designed for a cross-section of the 105 respondents. Fourteen respondents were selected by a simple random method to participate in the verification study. The small sample size was due to (additional) cost, time constraints, and willingness of respondents to participate in a repeat interview session. The objectives of the verification study were to assess the feasibility of the proposed framework (Figure 1), answer the ‘who’ questions, and validate the acceptability of the identified control measures to be included in the SS WSP. Verification was done by asking five basic questions (Figure 1) around each of the identified existing and recommended control measures in Table A1 (see Appendix A).

The unit of data analysis for the descriptive case study is respondents expressed in number of respondents or percentage of the number of respondents.

RESULTS AND DISCUSSION

Water safety planning framework for SS

The research-driven water safety planning framework for SS, as an outcome of the WHO WSP review, is shown in Figure 2. Contents of the three components (A, B and C) adapted from the WHO WSP and details of the fourth component (D) suggested for inclusion in the SS water safety planning model are: [A] System description: detailed understanding of the water source; [B] Risk assessments: identification and assessment of the hazards/hazardous events, causes and risks associated with the water source; [C] System management.

  • Source monitoring: involves the supervision of the following:

    • ○ Operations; Operators; Access to water sources

  • Source improvements (upgrade, cleaning and maintenance)

  • Source and household water handling. Handling refers to a user's activities around the source or household water from collection to storage e.g. (un)hygienic placement of operational tool (bucket and rope on the floor).

Figure 2

Research-driven WSP framework for SS Systems [NB: E1a: Ministry of Water Resources; E1b: Ministry of Health].

Figure 2

Research-driven WSP framework for SS Systems [NB: E1a: Ministry of Water Resources; E1b: Ministry of Health].

[D] Systems institutional framework: referred to as system managers (Oluwasanya & Carter 2013). System managers are made up of two main subjects; the source owners (S) and the external institutions or influencers. The possible external influencers (with expected roles in parenthesis) are as follows:

  • The government – E1

    • ○Ministry of Water Resources – E1a (validation/verification/supporting programs)

      • ▪ Department of Community Water Supply or Department of Self-supply Systems Management (DSSM). The DSSM is proposed; currently not in existence in Nigeria

    • ○Ministry of Health – E1b (validation and verification)

      • ▪ Department of Public or Environmental Health

  • Educational Institution; University – E2 (independent surveillance)

    • ○Academic Department of Water Resources Management

    • ○Academic Department of Microbiology or Environmental Health

  • Regulators – E3 (surveillance)

    • ○National Environmental Protection Agency and Standards Organisation.

Evaluation of control measures with specified implementing actors

Evaluation of the identified control measures and responses to the basic questions in Figure 1 are presented in Table 1. The number of responses for each control measure varied from nine to 14. In a few instances, related control measures were grouped. For example, views on control measures 13 to 16 associated with household water treatment and hygiene were given as being related. At least 12 responses were obtained for each control measure (Table 1). Respondents are either for (Y) or against (N) a measure, while a few claim acceptability based on specified pre-conditions (C). The evaluation of some of the critical measures are herewith highlighted.

Table 1

Source (self-supply wells) management and control measures: specified actors and level of acceptability

SNControl measuresFrequency
Actors
YesNoCGovernment
LLASORUOthers
Federal
StateLocal
NS
MOHMAWRWBLGCDPH 
Standardised well construction design – – – – 
Dedicated pump installation 13 – – – – – – – – – – 
Cleaning of well area 10 – – 
Minimum distance to source of waste – – – – – 
Usage of dedicated bucket – – – – – 
Minimum age limit – – – – 
Source hygiene management rules 13 – – – – 
Sanction of unruly behaviour 12 – – – – 
Display and compliance with access time 10 – – – – – 
10 Access time supervision 10 – – – – – 
11 Lock well when out of use – – – – – 
12 Standard bucket recovery system 10 – – – – – 
13 Household water treatment policy – – – – – – – 
14 Hand washing before storage usage – – – – – – – 
15 Usage of in-house storage cover – – – – – – – 
16 Usage of in-house storage bailer – – – – – – – 
17 Monitoring of water quality status – – – – – – – 
18 Inspection of well handling & hygiene – – – – – – 
19 Regulation of source management – – – – – – 
20 Compulsory supporting programs – – – – – – – 
SNControl measuresFrequency
Actors
YesNoCGovernment
LLASORUOthers
Federal
StateLocal
NS
MOHMAWRWBLGCDPH 
Standardised well construction design – – – – 
Dedicated pump installation 13 – – – – – – – – – – 
Cleaning of well area 10 – – 
Minimum distance to source of waste – – – – – 
Usage of dedicated bucket – – – – – 
Minimum age limit – – – – 
Source hygiene management rules 13 – – – – 
Sanction of unruly behaviour 12 – – – – 
Display and compliance with access time 10 – – – – – 
10 Access time supervision 10 – – – – – 
11 Lock well when out of use – – – – – 
12 Standard bucket recovery system 10 – – – – – 
13 Household water treatment policy – – – – – – – 
14 Hand washing before storage usage – – – – – – – 
15 Usage of in-house storage cover – – – – – – – 
16 Usage of in-house storage bailer – – – – – – – 
17 Monitoring of water quality status – – – – – – – 
18 Inspection of well handling & hygiene – – – – – – 
19 Regulation of source management – – – – – – 
20 Compulsory supporting programs – – – – – – – 

N = 14; MOH, Ministry of Health; MAWR, Ministry of Agriculture and Water Resources; WB, Water Board; LGC, Local Government Council; DPH, Department of Public Health; NS, Non-specified arm of government; LLA, Landlords Association; SO, Source owners; RU, Resident users; C, Conditional; Others, Head of household, any sensible adult, caretaker, NGO, Universities; –, Nil.

The first control measure is associated with hand-dug well construction. The action is the introduction of a standardised design for hand-dug wells. The well construction standard is to include construction design and quality of construction materials. Of the 13 respondents who remarked on the first control measure, three respondents are in favour of the introduction of standardised design for hand-dug wells. Another three respondents suggested a pre-condition for acceptability while eight respondents are against the measure. The respondents in favour of standard well construction design opined that the measure would bring about three positive impacts. Well water would be used for all household purposes, improve hygiene within the well area, and provide ‘better’ (in terms of quality) water. The users equally suggested the State, the local government council or the property owners association (Landlords Association, LLA) as the best possible implementers of the control action (Table 1). The LLA are expected to influence the usage of approved design among source owners (also property owners) within the associations.

Two pre-conditions were however mentioned by the users in the C category. One of the responders opined that the introduction of standardised well construction design is feasible ‘for those who can afford it’. Another claimed ‘only with new wells’. One of the source owners stated ‘I will not dig a new well’! Invariably, the users do not see the possibility of owners of existing wells complying with a regulation that suggests reconstruction or construction upgrade (new technologies are, however, emerging to ensure hand-dug well reconstruction or upgrade without the collapse of existing annular holes). Affordability is also seen as a limitation with either reconstruction or building of new wells in conformity with standard designs.

The number of respondents in the N category suggested that water users generally are not in support of standard well design. All the eight respondents declined based on affordability with comments like ‘high poverty level’ or ‘…too expensive’. A particular responder quoted the price of a pre-cast concrete ring lining as N3,500.00 (about 18 USD) per ring.

The second measure is usage of a dedicated pump (manual or motorised) for hand-dug well operation. Of the fourteen respondents that commented on pump installation, one respondent gave conditional acceptance, 13 respondents were against the action, and no respondents spoke in support of the measure (Table 1). The general concern was on ‘… the cost involved’. The price of a motorised pump is currently about N25,000.00 (126 USD). Self-supply well operation using a dedicated pump was previously identified as a best-practice control action for both access and hygiene management of such wells (Oluwasanya & Carter 2013). The best practice is however rather prohibitive. In the absence of dedicated pumps, enforcement of the usage of a dedicated bucket for self-supply well operation represented an alternative control action. The alternative action was debated by nine respondents. Five of the nine respondents agreed to the prospect. Two respondents each were in the C and the N categories.

The two respondents in the N category argued that ‘people will not wait for their turn’, and ‘the bucket can be spoilt’. Non-acceptability of the usage of a dedicated bucket based on the argument that users ‘will not wait their turn’ is flawed since water users generally ‘wait their turn’ at public water points with single tap stands. There should therefore be no justification for well-water users not to do the same if the action would ensure safe water. The notion that the hand-dug well operational bucket is a consumable that requires continual replacement was well reasoned. A responder figured that ‘We [the resident users] can be contributing the money for replacement or better rotate the replacement among ourselves’! The proposed solution is however dependent on ‘co-operation’ among the resident users.

One of the C category responders entertained the reservation of ‘if the resident users co-operate’. The other C category responder who believed that usage of a dedicated bucket is functional only in a ‘strictly private residence’ also implied the importance of some form of co-operation among resident users. A strictly private residence could mean a single household residence or residence with restricted access to non-resident users. Co-operation of resident users in self-supply management cannot be overemphasised. Resident users were identified as major players in self-supply well (source) management (Oluwasanya & Carter 2013). The five respondents in the Y category believed that disease (cholera) prevention is the main drive for the acceptance of usage of a dedicated bucket.

Source hygiene rules and sanctions include control actions Nos 7 and 8 in Table 1. Thirteen respondents commented on formalisation of hygiene rules and sanctions. All 13 respondents were unanimous in the acceptability of hygiene rules but one respondent believed that sanctions may be difficult to enforce (Table 1). The respondents were unanimous in the acceptability of hygiene rules and sanctions for three reasons: ‘the well areas will be neat’, ‘Minimise certain diseases’, and ‘improve water quality such that more people will have confidence to drink well water’. The responders however disagreed on the control measure implementers. One particularly believed that ‘the government should enlighten the people’. Two other responders supported the previous claim but suggested that the enlightenment should be via the radio and television. Another argued that the resident users, and not the government, are better suited to implement and enforce hygiene rules and sanctions. Nonetheless, the majority (6) believed that source owners are best placed to implement hygiene rules and sanctions (Table 1).

Formulation of regulatory measures was included in control measures 17 to 19 in Table 1. The regulatory measures included hand-dug well water-quality monitoring, inspection of well handling and source hygiene practices, and formal regulation of source management practices. Eleven respondents debated the referred regulatory measures. Seven respondents were in the C category, three in the N category, and only one respondent was in the Y category (Table 1). The Y category respondent believed that adoption of regulatory measures would bring about the provision of clean well water. The seven C category responders however argued that acceptability of regulatory measures would depend on whether the regulatory measures are ‘clear and well communicated’; ‘The government can do it [i.e. formulate the measures]’; the measures become a [government] ‘policy’; enforcement ‘does not involve payment by water users’; and if there is ‘adequate enlightenment’. The responders in summary recognised (1) the role of the government in policy or regulation formulation, (2) the need for adequate dissemination of regulatory measures, (3) the importance of enlightenment, and (4) the need for incentives. The highlighted observations invariably signify major factors that may influence the acceptance of regulatory measures by small systems' water users.

The role of the government in policy or regulation formulation is critical for the acceptability or adoption of control actions. Six of the 11 responders believed that source management regulation and formulation of any regulatory measure should be the responsibility of the government. One of the responders expected that government water regulatory plans should be disseminated by source owners to resident users. Another responder assigned the role of enlightenment to non-governmental organisations (NGOs) and to educational institutions like the universities. But one of the N category responders disapproved because of the financial implications. The woman queried ‘would the government give us [the source owners] money to do this [monitor water quality and inspect source handling]’?

Water-user acceptability of control measures: implication for water safety planning

To summarise the implication of the enumerated control measures to WSP, the degree of acceptability of each of the 20 control actions is highlighted in Table 2. A strong acceptance or rejection level is assigned to control actions with more than 70% of respondents in support of acceptance or rejection. Acceptance or rejection based on a simple majority is assigned to control measures with 50–70% respondent acceptance or rejection. Strong acceptance if specified conditions are met is allocated to control actions that had more than 70% respondent acceptance when the percentage of respondents in the Y and C categories are summed up. Consequently, five groups of control actions were derived (Table 3).

Table 2

Degree of acceptability of identified control measures

Control measures (Serial nos)Rejection (%)Acceptance (%)Acceptance (conditional) (%)
1a 57 43 
2b 93 
3c 77 15 
4d 10 60 30 
5d 22 56 22 
6c 17 75 
7c 100 
8c 92 
9c 71 22 
10c 71 22 
11d 14 64 22 
12c 71 22 
13d 30 40 30 
14e 40 30 30 
15e 40 30 30 
16e 40 30 30 
17d 27 64 
18d 27 64 
19d 27 64 
20d 55 36 
Control measures (Serial nos)Rejection (%)Acceptance (%)Acceptance (conditional) (%)
1a 57 43 
2b 93 
3c 77 15 
4d 10 60 30 
5d 22 56 22 
6c 17 75 
7c 100 
8c 92 
9c 71 22 
10c 71 22 
11d 14 64 22 
12c 71 22 
13d 30 40 30 
14e 40 30 30 
15e 40 30 30 
16e 40 30 30 
17d 27 64 
18d 27 64 
19d 27 64 
20d 55 36 

aStrong rejection (>70%).

bRejection based on simple majority (50–70%).

cStrong acceptance (>70%).

dStrong acceptance if conditions are met (>70%).

eAcceptance based on simple majority (50–70%).

Table 3

General acceptance level of control measures, N = 20

Control measure groupsNumber of measures
Strong acceptancea 
Strong rejectionb 
Acceptance based on simple majoritya 
Rejection based on simple majorityb 
Strong acceptance if specified conditions are meta 
Control measure groupsNumber of measures
Strong acceptancea 
Strong rejectionb 
Acceptance based on simple majoritya 
Rejection based on simple majorityb 
Strong acceptance if specified conditions are meta 

aGeneral acceptance.

bGeneral rejection.

Table 3 shows the control measures that were generally acceptable to hand-dug well users and, as such, recommended for inclusion in WSP for self-supply hand-dug wells. The rejected measures however may not be included in the plans. The control measures (serial number in parenthesis) that were strongly acceptable to water users include: Cleaning of well area (3); Minimum age limit (6); Source hygiene management rules (7); Sanction of unruly behaviour (8); Display and compliance with access time (9); Access time supervision (10); and Standard bucket recovery system (12). The control actions that possess strong acceptability if various specified conditions are met include: Minimum distance to sources of contamination (4); Usage of dedicated bucket (5); Locking of dug well when not in use (11); Household water treatment policy (13); Monitoring of water quality status (17); Inspection of well handling and hygiene (18); Regulation of source management (19); and Compulsory supporting programs (20).

The various major requirements for general acceptability of control measures include: (1) Provision of public burial grounds that are easily accessible (minimum bureaucratic procedure of procurement) for both private and public usage; (2) Public enlightenment focused on the need and expected benefits from stipulated control actions; (3) Co-operation especially among the resident users – the source managers; (4) Establishment of sanctions for non-compliance; (5) Appropriate training of relevant actors in the following areas: access control, operation management, hygiene management, bucket recovery tools and system development; (6) Provision of incentives in especially the following areas: subsidies to enhance public acceptance of approved well construction design with usage of recommended construction material, provision of water safety facilities like laboratories for water quality monitoring and surveillance, and provision of household water treatment solutions and subsidies to encourage the adoption; (7) Government involvement in the design and implementation of supporting programs and dissemination of water safety information to users.

The control measures with acceptability based on simple majority were: (1) Hand washing before water retrieval from storage (14); (2) Usage of in-house storage cover (15); and (3) Usage of in-house storage bailer (16). These actions were regarded as actions for enlightenment rather than control.

The control measure that was generally not acceptable to water users was the imposition of a dedicated pump. Water users would not accede to the action. Introduction of standardised well construction design, including specifications for the quality of construction materials, was rejected based on simple majority (57%; Table 2). A shift to general acceptance (from current 43% to more than or equal to 50%) may, however, be achieved with the provision of subsidies to encourage acceptance, and if a range of design and construction materials are stipulated.

Dealing with poverty in water safety planning for SS

The issue of affordability is recurrent in this study. Statements such as ‘…high poverty level’, ‘…too expensive’ or ‘…the cost involved’ are fundamental to affordability and cannot be overemphasised in the context of SS in the study area and by extension to many developing countries. However, the assumption beneath the ‘too expensive’ theory, which is particularly directed towards the non-acceptability of standard well design, should be queried. The assumption is that such designs would involve prescription of pre-cast concrete ring linings and that pre-cast ring linings are generally ‘too expensive’. It should be noted that standard design may not necessarily involve usage of expensive construction materials or techniques. It is also important to understand that the drive behind any control action should be safety; human, source and water safety, and not high cost. Nonetheless, the concern for affordability should be taken seriously as it may mar the implementation of WSP for household water sources in regions where poverty levels are generally elevated. Consequently, stipulated control measures that would involve incurring expenses should include a range of options (where possible) from best practice (top of the range) to safe practice (acceptable safe materials and designs). For instance, regarding standard well design, it could be a range of, for example, use of pre-cast concrete ring linings (best practice) to cemented block linings (safe practice). Water users are then free to choose within a range of flexible options according to affordability.

The role of incentives in the acceptability of control actions for self-supply owners/users

The expectation of possible incentives with the implementation of regulatory measures is also noted in this study. The idea of incentives, as suggested by some of the responders, signified that incentives may play a key role in acceptability or adoption of regulatory actions. Incentives may be provided in two ways; first, in the form of subsidies through, for instance, the establishment of micro-financing. Subsidies could be provided to source owners to improve or upgrade source construction quality and design to the recommended standard. The second means of creating incentives may be in the form of providing facilities that could enhance the adoption of regulatory measures. An example is laboratory facilities, which could be provided for source water quality monitoring. Such facilities could be made affordable and easily accessible to water users. With appropriate incentives in place, certified source construction documents or certification of safe water outlets, for instance, issued by the relevant authority could be provided on demand to surveillance or regulatory operatives. In this way, appropriate incentives could facilitate the adoption and implementation of regulatory measures and by extension enable the water safety of small systems to serve as a regulatory tool as intended.

The role of a coordinating agency for SS

The need for a coordinating institution to oversee and develop water safety guidance for SS was previously argued (Oluwasanya & Carter 2013). In this paper, the role of such a coordinating agency was conceptualised within the recommended water safety planning framework for SS (Figure 2). The implementing actors or relevant task owners were also specified. Water safety planning for SS may not be achievable until a coordinating institution is established. Also, such a coordinating agency can only be effective if SS managers (source owners or designated resident users) are fully incorporated (Figure 2). Similarly, enforcement of the regulatory tool for small systems within the study area and by extension, in developing country regions, is critical but only when water safety planning for such sources is achieved. By suggesting a framework and providing relevant control actions, this paper offers guidance for water safety planning for SS. The information from this study may also be useful in the facilitation of water safety development for small systems. To ensure safe water provision, achieve universal access and in part help to meet the global sustainable goals, this paper recommends that water safety planning for small systems should be promoted.

CONCLUSIONS

The WHO WSP step-by-step framework was considerably modified for adaptation to SS. While the systems assessment component of the WHO WSP model remained fundamentally relevant, the management component shifted to accommodate the source management realities of SS, which is hinged on the peculiar ownership characteristics. Another major modification is the inclusion of the institutional framework, which is pivotal to WSP development and implementation for SS. The paper identified that lack of a coordinating institution to claim responsibility for the broad management of individually owned water sources may generally hinder water safety management of SS. Finally, this paper provides a framework (Figure 2) that could guide local specific adaptation of water safety planning/processes for small water systems.

REFERENCES

REFERENCES
Bartram
,
J.
,
Corrales
,
L.
,
Davison
,
A.
,
Deere
,
D.
,
Drury
,
D.
,
Gordon
,
B.
,
Howard
,
G.
,
Rinehold
,
A.
&
Stevens
,
M.
2009
Water Safety Plan Manual: Step-by-Step Risk Management for Drinking-Water Suppliers
.
World Health Organization
,
Geneva, Switzerland
.
Carter
,
R. C.
2006
Investigating Options for Self-Help Water Supply: From Field Research to Pilot Interventions in Uganda
.
Rural Water Supply Series
,
WSP/RWSN
,
St Gallen
,
Switzerland
.
Davison
,
A.
,
Howard
,
G.
,
Stevens
,
M.
,
Callan
,
P.
,
Fewtrell
,
L.
,
Deere
,
D.
&
Bartram
,
J.
2005
Water Safety Plans: Managing Drinking-Water Quality from Catchment to Consumer
.
WHO
,
Geneva
,
Switzerland
.
Foster
,
S.
2008
Urban water supply security in Sub-Saharan Africa: making best use of groundwater
. In:
Session 5, International Conference of Groundwater and Climate in Africa
,
24–28 June 2008
,
Kampala, Uganda
,
available at
:
http://www.worldbank.org/gwmate (accessed 6 November 2009)
.
Gowing
,
J.
,
Parkin
,
G.
,
Forsythe
,
N.
,
Walker
,
D.
,
Haile
,
A. T.
&
Alamirew
,
D.
2016
Shallow groundwater in sub-Saharan Africa: neglected opportunity for sustainable intensification of small-scale agriculture
?
Journal of Hydrology and Earth Systems Sciences
10
,
519
549
.
EGU open access
,
DOI:10.5194/hess-2015-549
.
Mekonta
,
L.
,
Butterworth
,
J.
&
Holtslag
,
H.
2015
Self-supply as a Formal Service Delivery Model for Rural Water in Ethiopia: Ambition, Approaches, Challenges and Opportunities
.
IRC Briefing Note, February 2015
,
IRC, The Hague
,
The Netherlands
.
Oluwasanya
,
G.
&
Carter
,
R.
2013
Management of urban self-supply wells in Abeokuta, Nigeria: an assessment and redefinition of roles for water safety planning
.
Nigerian Journal of Hydrological Sciences (NJHS)
2
,
85
109
. .
Oluwasanya
,
G.
,
Smith
,
J.
&
Carter
,
R.
2011
Self supply systems: urban dug wells in Abeokuta, Nigeria
.
Journal of Water Science and Technology: Water Supply
11
(
2
),
172
178
.
Schmoll
,
O.
,
Howard
,
G.
,
Chilton
,
J.
&
Chorus
,
I.
2006
Protecting Groundwater for Health: Managing the Quality of Drinking-Water Sources
, 1st edn.
IWA
,
London
,
UK
.
Sutton
,
S.
2004
Self-supply: A Fresh Approach to Water for Rural Populations
.
Field Note, November 2004
,
WSP/RWSN/DFID
,
UK
.
United Nations
2014
Report of the Open Working Group of the General Assembly on Sustainable Development Goals, issued as document A/68/970, United Nations, available at: http://undocs.org/A/68/970 (accessed 2 April 2015)
.
WHO
2004
Guidelines for Drinking-Water Quality
, 3rd edn,
Vol. 1
.
WHO
,
Geneva
,
Switzerland
.
WHO
2011
Guidelines for Drinking-Water Quality
, 4th edn.
WHO
,
Geneva
,
Switzerland
.

Supplementary data