Groundwater dependence among poor urban people: out of sight is out of mind?

Abstract

Hundreds of millions of poor urban dwellers especially in Africa and Asia rely on groundwater from local, shallow wells or boreholes for domestic purposes; users may obtain such water through piped systems (taps), tankers, private vendors or directly from source. Many wells are not accounted for in the official statistics and the role of groundwater is frequently underestimated due to lack of knowledge, baseline data and assessments. One problem lies in the classification of unprotected dug wells as ‘unimproved’ water sources, another in the unawareness of how poor people often need more than one domestic water source. Easy access to well water, even if contaminated, is often beneficial to health because it facilitates good hygiene behaviour. The invisibility of groundwater undermines the potential for informed debate and policymaking regarding its increasing importance in urban areas.

Keywords:

• groundwater
• water access
• urban self-supply
• MDGs
• developing countries
• Bangalore
• Lusaka

1. Introduction

Improving easy access to water for drinking and other domestic purposes constitutes a significant step in the fight against poverty and ill health, and towards socio-economic development and well-being. Yet to hundreds of millions of urban dwellers the task of sourcing sufficient quantities of reasonably clean water is a daily stressor. Scarce water availability, poor water quality, inadequate sanitation facilities and improper hygiene behaviour are tightly linked factors that need to be tackled in an integrated manner to prevent disease (Eisenberg et al. 2007; Schmidt and Cairncross 2009). This is particularly true in the urban and peri-urban environment in developing countries, where infrastructure deficits are common and poor people are generally living under conditions of high density and a lack of services for water supply, sanitation and waste disposal.

Simultaneously, hundreds of millions of people rely on groundwater taken from local wells for drinking and other domestic purposes – at least a third of the population in cities in Sub-Saharan Africa and Asia according to the Demographic and Health Surveys (DHS) (Grönwall et al. 2010). Far from being a small and declining residual to the piped water supply system, groundwater from wells – one's own, someone else's or a shared well – is therefore vital especially for households in slums and low-income urban settlements. Well water can also be distributed by utilities that make conjunctive use of groundwater and surface water. For both individuals and utilities, groundwater may play a seasonal, yet strategic supplementary role, depending on geographic and other conditions. More than half of the world's megacities depend on groundwater in the sense that it constitutes at least a quarter of these cities' water supply (cf. Morris et al. 2003).

In both cases it seems that as a resource, groundwater is far from given the attention it deserves. This may be partly due to a lack of monitoring (baseline data) of hydrogeological and climatic conditions and the aquifer systems (locally and in relation to regional river basins) that impedes the ability to assess the situation. In part, this may also be because the number and kinds of wells (both private and public) in the city environment are seldom known or accurately reflected in statistics and official reports. The extent to which poor people especially self-supply with water from wells is therefore generally not accounted for in full. Furthermore, utilities do not always state publicly that groundwater is distributed to its customers, contributing to making groundwater dependence largely absent from official statistics.

At another level, part of the problem lies in the UN Millennium Development Goal (MDG) system's chosen assessment definitions and the methodology of the progress monitoring. The world is reportedly on track to achieve MDG target 10 (sometimes referred to as target 7c) of halving the proportion of the population without sustainable access to safe drinking water by 2015. From a worldwide average of over a billion people without such access in 2005, the number had gone down to 884 million people in 2010 (WHO and UNICEF 2000). In the MDG vocabulary, this means that a steadily rising number of people supposedly use an ‘improved’ source of water for drinking (and other domestic purposes such as hygiene): the water is distributed through pipes to the house or yard; fetched from public standpipes/taps; or taken from tube wells, boreholes or ‘protected’ dug wells. Several things can, however, be called into question in this regard, including whether the progress is actually as rapid as reported, and whether it is sound to make a distinction between water sources in the way it is done today. It is imperative to do so against the increasing insights into the importance of groundwater, not the least to poor people, in times of rapid urban growth.

Groundwater is often referred to as a ‘hidden’ resource. This article reviews in short the importance of groundwater and of different kinds of wells especially for the poorest section of society in urban areas of low- and middle-income (developing) countries. It argues that the role and value of groundwater, particularly from local wells, is often underestimated and needs to be accorded greater importance in city-level water resource management and strategic planning. The discussion is substantiated by examples from Bangalore, the so-called Silicon Valley of India, and Lusaka, capital of Zambia, as well as by aggregate statistics produced for the IIED Working Paper (Grönwall et al. 2010) that this article draws from.

2. Urban groundwater dependence

Urban poor people are, by and large, especially vulnerable to unreliable water access from both public utilities and private and informal sources. Although governments (at national and state level, where applicable) have the final responsibility to see to that everyone's human right to water is met (Grönwall 2008), urban poor households are generally not served directly by the water utilities. This group tends to depend more on three water sources: public standpipes (sometimes in the form of communal taps); small-scale private water providers (vendors, using tanker trucks or other means to distribute water); and/or private or communal/shared wells. In addition, especially urban dwellers in developing countries tend to make use of more than one source because their main choice may be unreliable for different reasons (Montgomery et al. 2003), this including consumers connected to the piped supply. The volumes of water available are regularly insignificant; for instance, no municipal water utility in either India or Pakistan provides a continuous water supply to all of its customers – many towns in India supply water for 2 hours or less per day, with some areas receiving water on alternate days or more seldom (Tayler 2008). This is particularly troublesome for poor households without proper storage facilities.

Strategies for sourcing water from elsewhere are therefore often necessary for both rich and poor. Self-supply is the choice for many and groundwater is commonly relied on. Apart from when taken from the local well (whether it belongs to the user or someone else, or is a shared well), public standpipes/taps may deliver water from a stand-alone well, and vendors in some parts of the world, such as India, sell groundwater taken from own or others' wells. In these latter cases, just as when the groundwater is distributed through pipes as part of the provider's reticulated water supply system, the users' dependence on – and influence over – the source is more indirect. In many cases the water users probably consume groundwater unknowingly.

This stands in contrast to the situation where water from local stand-alone wells is the choice by necessity. Lack of service provision in most low-income settlements and in very many peri-urban city areas means that groundwater is often used and depended upon more directly. Pumping water from different kinds of wells is a small-scale method that can often be applied locally with the help of low-cost technology. The wells may be dug or drilled (boreholes, tube wells); open or closed/covered; hand-drawn or equipped with handpumps or a motor – mainly electric submersible pumps – to lift the water. The wells may also be fitted with taps and pipes for distribution to more distant households (in which case the water may to some users appear and be defined as ‘piped’ water or simply ‘tap’ water).

Dug wells in the urban environment are predominantly shallow or extremely shallow, and constructed in areas where the water table is high so that water can be readily found at a depth of around 3–15 m. Such wells can normally be dug by hand in weathered regolith layers with clay, sand, gravel or mixed types of soils and only small boulders present. This means that self-supply is feasible for many poor people where the hydrogeological conditions are favourable and land-rights and space permit digging of wells, for instance in the owners' yards. This is the case in Lusaka's many peri-urban, low-income settlements.

Where people use their own dug, shallow wells because they are forced to fend for themselves, having a well of one's own provides freedom and great savings of time and energy and often also money. Shallow, dug wells are generally simple structures that the users – at least in principle – can construct and maintain on their own, meaning that their ‘functional sustainability’ is high; they can be kept in a good state of repair also in the longer run (Carter and Bevan 2008).

However, where the water table is too deep for digging of shallow wells and it is too costly for users to drill boreholes, equip them with pumping devices and so on, and to maintain the wells, the investments needed to reach the groundwater may prove prohibitive for individual households and smaller communities – such as in Bangalore, and parts of Lusaka. Poor urban dwellers may instead depend on groundwater being provided by other actors. Groundwater can be accessed through the utility's public standpipes or community-based taps, through vendors or through an NGO or foreign donor that arranges for water supply through drilling of medium-depth boreholes fitted with handpumps or electrically powered ones. A shared, community well may be feasible but the construction and consequent maintenance may depend on accessibility of (micro-)loans or other fundings, and the level of knowledge and expertise available.

Community boreholes but also deeper, dug wells are sometimes constructed by the authority in charge of water supply (such as the public utility in Bangalore) or by an NGO or church organisation active in the area (such as in Lusaka). The latter can especially be seen in villages that have consequently been merged with a growing adjacent city to become peri-urban areas. Water from such wells is more seldom distributed for free; users may have had to contribute economically for its construction and later also for the water (service) itself. In most Indian cities, though, water from public standpipes is still free.

Although many of those relying on groundwater take it for free from their own or shared wells, others thus have to purchase it. Water vendors – private small-scale suppliers of water in bulk (wholesale) or per pot/can – may sell water taken from wells. This is, for instance, common in India where landowners are legally entitled to draw unlimited amounts of groundwater from wells on their land.

3. Is urban groundwater development sustainable?

Although increased use of groundwater at the urban level is predicted as well as promoted by several experts in the field today, there are enormous problems connected to sinking water tables. ‘Over-extraction’ from aquifers, resulting in a falling water table, is reported from many parts of the world, for instance Lahore in Pakistan. This is pronounced where abstraction rates increase rapidly and changing precipitation patterns – some of which are linked to climate change – affect the recharge possibilities.

How much water can be withdrawn from a given aquifer or aquifer system under or by a city before the abstraction – recharge rate should be deemed unsustainable? The answer depends on a multitude of factors including the scale and technique of the abstractions, which is in turn connected to the purpose of the water use. But the question is complex if by sustainable we mean ecologic, economic as well as social aspects. For instance, the extensive use of groundwater from the confined sand/chalk aquifer of central London from the nineteenth century and onwards had long-term negative consequences with a lowered water table and springs and streams drying up, but it was economically beneficial for the development of the city as a major centre of population and manufacturing (Price 2002). Similar reasoning can be made in relation to many cities in developing and newly industrialised countries. Domestic use of groundwater tends to seem unproblematic when compared with the volumes of water required for irrigation and industrial needs. However, when individual wells dry up, this may have significant consequences for users depending on them.

Typically, crystalline (hard) bedrock means low-yielding and notoriously unpredictable aquifers and wells that are less sustainable. The reliability of the aquifers to yield water can be expressed as success or failure at the point and time of construction. The sustainability of the well once constructed is a function of increasing and changing types of demand, variability and intensity of precipitation, current and changing recharge potential, landuse changes and so on. The water availability may decrease due to competitive construction and deepening of nearby wells and during dry periods (seasons, years) when recharge is insufficient.

The groundwater availability within a city can vary hugely because the aquifer conditions are often not uniform. For instance, Bangalore's groundwater is taken from weathered crystalline bedrock. Although characteristically low-yielding, the conditions are highly variable and hence unpredictable. There may be differences between two nearby wells that can be explained by highly localised fissures and fractures. Some of the variability may also depend on the fact that urban aquifers usually receive additional recharge because of water being imported to the city through pipes and mains that leak substantially. Furthermore, the subsurface in urban environments regularly has secondary porosities and perhaps a permeability distribution comparable to shallow karst settings. Over-irrigation of lawns, golf courses and so on contributes to a situation where the recharge is as high as or higher than in equivalent rural areas despite hardened surfaces (Lerner 2002; Garcia-Fresca and Sharp 2005).

Whether there are risks of sinking water tables and aquifer depletion is also a question of what alternative water sources are available to different groups, in both socio-economic and technical respects, and what quantities are being pumped elsewhere from the same aquifer system, especially for utilities' water supply and by large commercial abstractors but sometimes also for irrigation. From the perspective of the generally water-deprived and under-served group of urban poor, temporary over-exploitation of aquifers may prove beneficial for development, health and well-being, provided that measures are taken to promote artificial recharge, rainwater harvesting, reuse, recycling and in the long run perhaps also reduced dependence on self-supply through wells.

Alarmist reports on (ground)water scarcity and looming water crises seldom take all the above into account, and are, in many cases, not accompanied by consistent or easily assessable observations of the actual development. This does not serve a situation where growing urban populations request more water. Increased use of groundwater necessitates increased attention to the resource, based on a holistic view of availability, supply, demand and needs and how the urban environment influences the dynamics of the water cycle.

4. Groundwater, climate change and growth

Climate change researchers suggest that as the world's reliable surface water supplies are likely to decrease due to increased temporal variations of river flow (caused by increased precipitation variability and decreased snow/ice storage), it might be beneficial to take advantage of the storage capacity of aquifers and increase groundwater withdrawals for different purposes, including urban water supplies. However, as pointed out this option is only sustainable where groundwater withdrawals remain well below recharge, and is not viable where groundwater recharge is projected to decrease (Kundzewicz and Döll 2009).

UN's Intergovernmental Panel on Climate Change has pointed to worrying gaps in knowledge and observational data about water-related impacts of climate change that are especially inadequate with respect to groundwater. This is much due to that the knowledge of current aquifer recharge is poor in both developed and developing countries, and because of considerable uncertainties with regard to (spatial distribution of) the projected precipitation changes (Kundzewicz et al. 2007; Bates et al. 2008). These insecurities are critical as the climate system and groundwater storage are fundamental and interacting parts of the hydrological cycle. Today's lack of data has implications both on further research and policymaking.

However, apart from there being difficulties ‘clouding the prediction’ of regional effects of future climate change on water resources, studies of these effects coupled with population growth models at regional and global scale suggest that population growth is likely to exert as great or greater an impact on the world's water resources as global warming might (Vörösmarty et al. 2000; Loaiciga 2009, p. 10). The demands for water rise also because of new and higher requirements due to altered lifestyle choices, especially in urban environments that, as such, are growing more rapidly than ever.

5. MDG target 10 and groundwater

5.1. Defining ‘improved’ water sources

The WHO and UNICEF's Joint Monitoring Programme (JMP), established in 1990, is the official mechanism of the UN system, mandated to monitor global progress towards the MDG target on water supply and sanitation by generating relevant data. In the year 2000, when indicative targets for water supply and sanitation coverage were endorsed as part of the Millennium Declaration, the methodology for assessment was partly changed in response to criticism: instead of reporting only coverage data from water service providers, user (consumer) based data were also collected as far as possible through household surveys and national censuses (WHO and UNICEF 2000). The means of water provision were thus emphasised in the definition of coverage whereas in past assessments, the coverage figures referred to ‘safe’ water supply (and ‘adequate’ sanitation) only. However, as a result of the lack of specific information on the quality of the drinking water and ditto for sanitation, the assessment ‘assumed that certain types of technology are safer or more adequate than others and that some of them could not be considered as “coverage”. The terms “safe” and “adequate” were replaced with “improved” to accommodate these limitations’ (WHO and UNICEF 2000, p. 4). Hence, the system builds on approving of certain water sources whereas others are regarded as unimproved from a wider health perspective.

The proxy indicators – based on technology type – for distinguishing between the improved and the unimproved sources were not employed without discussion. It was acknowledged that they can allow ‘only an approximate description of water and sanitation coverage’; that information about the quality of the water provided or about its use is not given; and that factors such as intermittence or disinfection could not be taken into account (WHO and UNICEF 2000). For instance, individual unprotected household wells may provide a better supply of water, both in terms of quantity and quality of water, than a household connection which may be subject to intermittence and poor water quality (cf. Shar et al. 2010). Likewise, water supplies from vendors or cart/tanker trucks may be adequate – but ‘experience suggests that such technologies are typically inferior to “improved” services’ (WHO and UNICEF 2000, p. 5). Furthermore, the definitions used in various countries' household surveys are not entirely standardised, meaning that cross-country comparisons are not always accurate.

The questionnaires used to collect information at household level ask respondents to identify their main source of water for drinking, defined as including also other domestic needs (although many countries survey water used for ‘other purposes’ separately). The respondent chooses from a list of technologies that includes household connection (water piped to the house or yard), public tap or standpipe, tube wells and boreholes, ‘protected’ dug wells and springs and rainwater collection (compare Table 1 below). This list is, however, not only based on safety criteria. It also involves the term ‘reasonable access’, broadly defined as the availability of at least 20 L per person per day, from a source within 1 km of the dwelling (WHO and UNICEF 2000).

Table 1. Sources of drinking water in the DHSs for Zambia 1992, 1996–1997, 2001–2002, 2007

DHS 1992; 1996–1997	DHS 2001–2002	DHS 2007
Piped into residence	Piped into dwelling	Piped into dwelling
	Piped into yard/plot	Piped into yard/plot
Public tap	Communal tap	Communal tap
	Piped to neighbour
Well in residence	Open well in yard/plot	Open well in yard/plot
Public shallow well	Open public well	Open public well/borehole
	Open well at neighbour
Public traditional well	Protected well in yard/plot	Protected well/borehole in yard/plot
	Protected public well	Protected public well
Spring	Spring	Spring

Hence, in developing the methodology of the JMP assessment considerable discussions were held regarding the acceptability of sources such as bottled water and vendor-provided supplies (including tanker truck supplies). As a result, water from private vendors is not considered ‘improved’ because the volumes secured from such sources may be severely limited by cost. Similarly, the expenditure of bottled water alone was estimated to prohibit adequate volumes of water for domestic uses other than drinking water for such would therefore have to be secured from other sources (WHO and UNICEF 2000).

The reasoning shows efforts to balance several angles of water access. It is paramount to take water availability (quantity) into account besides the issue of quality because only when drinking water is the main transmission route of water-, sanitation- and hygiene-related diseases will its quality be more important than the quantities available. Increased, easy access to water has the potential to lead to improved hygiene behaviour (Bostoen et al. 2007; Eisenberg et al. 2007; Schmidt and Cairncross 2009). A key in this regard is the possibility to wash hands – preferably under running water and with soap. It can be noted, though, that in the latest JMP update an improved source is defined in terms of quality, as ‘one that by the nature of its construction adequately protects the source from outside contamination, in particular with faecal matter’ (WHO and UNICEF 2010, p. 34).

Admittedly, being a global system for assessment of countries' progress towards the MDGs and based on the limited information collected, the JMP definitions must necessarily be simplifications of a difficult reality. However, many criticise the system. For instance, because the complexity of the issue has proven that the proxy indicators for target 10 are not well chosen: they oversimplify assessment matters, paint a misleading picture of the situation on the ground, and risk stifling real progress (Schäfer and Dölle 2007).

Further, water being provided, or accessible, from an improved source does not necessarily mean that it is adequate for the users' well-being and health (UN-HABITAT 2003; Manda 2009). This is a concern especially for poor people who depend on public standpipes and/or shared (albeit ‘improved’) wells where the distance to such, the limited time these may be open, the queues and the competition over the available water may prohibit the users from upkeeping a healthy standard of living. For instance, in Lusaka's large low-income settlements there are examples of one communal tap per 220 households and poor quality of the water purchased at those taps. Besides the competition and efforts to fetch the water from such taps, it is too expensive for many. At least 50% of the residents therefore depend on having dug their own, shallow wells.

5.2. Wells as improved versus unimproved water sources

Groundwater is considered as an improved source of water when distributed by way of pipes and taps, including public standpipes (standposts), and the population with access to such is counted towards MDG target 10. Similarly, tube wells, deep boreholes and ‘protected’ dug wells and springs are approved of in terms of safety. ‘Protected’ means that the well is protected from runoff water by a well lining or casing that is raised above ground level and a platform that diverts spilled water away from the well, and that it is covered to prevent bird dropping and animals from falling into it. In contrast, all unlined dug wells that are not ‘protected’ from runoff surface water (by being raised above the ground and equipped with a platform, or covered to protect it from bird droppings and animals) are defined as unimproved from a quality point of view.

Nevertheless, by distinguishing between wells in this manner a flawed definition of ‘safety’ is promoted. Although wellhead protection is important to prevent contamination, even protected wells can yield substandard, non-potable water. Urban groundwater resources are highly vulnerable to pollution from anthropogenic activities and where the groundwater is contaminated by arsenic, seawater, chemicals or ordinary sewage, the physical structure of an ‘improved’ well may provide little real protection. Similarly, several latrine types, including some of the improved sanitation categories, allow for faecal matters and wastewater from latrines to pollute aquifers by seepage through the ground, especially where the water table is shallow and the soil highly permeable. It can be noted that the 2006 Stockholm Water Prize Laureate, Asit K. Biswas, is harsh in his view in this regard: ‘If somebody has a well in a town or village in the developing world and we put concrete around the well – nothing else – it becomes an “improved source of water”; the quality is the same but you have “improved” the physical structure, which has no impact … [The official figures] are not only underestimating the problem, they are giving the impression the problem is being solved. What I'm trying to say is: that's a bunch of baloney’ (Jowit 2010).

When tested, the drinking water obtained from many improved sources has consequently not met the microbiological standards set by WHO and this problem is now increasingly acknowledged (United Nations Department of Economic and Social Affairs 2009). The approval of only such sources that are supposedly rendering safe and potable water has led to a challenge of measuring the water quality, such as through water safety indicators at the household level, that is still ‘beset by technical and logistical difficulties and by high cost’ (WHO and UNICEF 2010, p. 31). In the 2010 JMP update, quality-related questions such as ‘what definitions would be meaningful and assist decision-makers in the process of improving the drinking water situation in the world?’ were formulated, to be addressed by a new JMP task force (WHO and UNICEF 2010, p. 31). Most nations lack data on this parameter today, but water quality will have to be part of a revised MDG target beyond 2015. It will be increasingly important, then, to distinguish between the (relatively small volumes of) water vital for drinking and cooking purposes, and water necessary for other domestic uses. The quality of the latter can, at least in theory, be comparatively inferior and be taken from other sources, and/or undergo less treatment.

5.3. Groundwater and wells in the statistics

The JMP uses data from the USAID-sponsored DHS, conducted in 75 low- and middle-income countries (not including China), together with other national surveys such as the Multiple Indicator Cluster Survey. These surveys cover several thousand households in each country and the cluster samples are stratified to ensure that they are representative of urban and rural areas of each country. The findings based on these surveys are fundamental in assessing how far the development has come in attaining the MDGs, how well-invested money from donors and others are, whether the right policy recommendations have been made, in what regional areas more efforts seem to be needed and so on.

Limitations to the very design of the surveys affect the official statistics relating to water supply, not the least when groundwater is used by various means of distribution. The survey questionnaires only allow for the household's one main source of drinking water to be ticked (and in some countries and some surveys also the main source of water for other household purposes), although very many households use more than one source by necessity. From the formulations and the list used to define various ‘sources’ in the DHS (see www.measuredhs.com), it can be assumed that many respondents will have difficulties classifying what constitutes their main source, unless the interviewer explains the alternatives with examples from the local context. For instance, in the DHS 2007 the respondents in Zambia's capital Lusaka did not have the opportunity to answer that they purchase water from a ‘community-based system’, and therefore probably ticked ‘communal tap’ for this alternative (see Table 1). The former is an option for about 30% of Lusaka's urban population living in low-income areas and distributes water from local, protected boreholes, whereas the actual ‘communal taps’ provide a mix of surface water and groundwater (NWASCO 2009). The statistical error this gives rise to is not large but it contributes to misrepresenting the direct dependence on wells and may, potentially, affect decisions on further investments in this kind of supply scheme.

The terminology used in the questionnaires is also not consistent from year to year, which can be seen in the following example (Table 1) of ‘source of drinking water’ from the last four DHS surveys carried out in Zambia. (The respondents could also choose different surface water alternatives as well as water from tanker truck or cart, bottled water and ‘other’.)

According to the DHS 2007 results from Zambia, 76.6% of the urban water users in the country responded that they (mainly) take water from a piped source, whereas supposedly only 18.2% use groundwater (from stand-alone open or protected wells). However, an in-depth study of the city of Lusaka suggests that ca. 55% of the ‘piped’ water distributed by the public utility is drawn from boreholes, and that the (reticulated) water in ‘communal taps’ comes from the same mix of sources (Grönwall et al. 2010). It was also found during field work in Lusaka's settlements that people with dug wells may not want to admit that they use them, because such wells are banned by the authorities (Grönwall et al. 2010). Such a respondent may therefore state that the main source of water is any of the other options rather than ticking ‘open well in yard/plot’. In other words, from the Zambia DHS – and, consequently, the JMP reporting – it may seem as if the total dependence on groundwater is smaller than in reality.

It can also be noted that in Bangalore, many households – especially at the outskirts where the public utility has no or little infrastructure – purchase water in bulk from private groundwater vendors. The majority of their wells are, as such, of the ‘improved’ kind, and the vendors, their wells and trucks provide an important lifeline. For poor households that rely on purchasing water in this way, a price of Rs. 1–3 (USD 0.02–0.06) per 13 L container is of course often prohibitively high (though storage constraints, competition and distance/effort may be equally limiting). In both cases, though, this practice is not officially endorsed: where access to water is mainly through tanker trucks, it is not counted as progress towards the MDG target 10. Again, in the JMP reporting the total dependence on groundwater will seem smaller than in reality.

6. Aspects of a hidden asset: examples from Bangalore and Lusaka

It is often pointed out by experts within the field that groundwater as a resource is a ‘hidden asset’ and therefore under-researched. Clearly, groundwater and its importance is less acknowledged, and also less well understood, due to its occurrence out of sight below ground. It is likely that the invisible character of groundwater may result in local authorities and decision-makers having insufficient and inaccurate information about both its occurrence in and around the city environment, and the extent to which it is used by city dwellers. There is almost inevitably a lack of detailed information on the prevailing hydrogeological conditions, but also incomplete or missing records on the existing number of dug wells and boreholes, many of which are informal constructions. The characteristics of these wells, including water table, abstraction rate, potential quality problems and other data are pertinent for planning, development, protection and conservation of the groundwater resources in the short as well as long term. Groundwater modelling is complex but without knowledge of this kind, it is virtually impossible to track the impact of pollution and (over-)exploitation of groundwater spatially and over time to develop and implement evidence-based sustainability policies.

Examples of these challenges are visible from case studies of Bangalore and Lusaka. Table 2 summarises the relevant conditions for groundwater dependence in the two cities (full details are given in Grönwall et al. 2010). As can be seen, the data available are in many cases based on fairly coarse estimations only, or no data at all are available.

Table 2. Comparison of groundwater dependence in Bangalore and Lusaka

	Bangalore	Lusaka
City size	741 km^2	375 km^2
GDP (estimated)	USD 69 billion (2008)	N/A
Number of inhabitants (estimated)	7–9 M	ca. 1.7 M
Number of official slums/informal areas	473	35–40
Proportion of slum dwellers (estimated)	20–35%	65–70%
Proportion connected to water utility	<30%	<35%
Number of house connections	360,000 (2009)	N/A
Volume distributed by utility (all uses)	ca. 900,000 m^3/day	ca. 210,000 m^3/day
Demand (estimated by utility)	>1,300,000 m^3/day	ca. 400,000 m^3/day
Geological conditions, main rock	Granite	Limestone
Number of utility wells	ca. 10,000	ca. 72
Proportion of water from utility wells	N/A	>50%
Volume pumped from utility wells	N/A	130,000 m^3/day
Main type of private wells (depth)	Boreholes (60–300 m)	Dug wells (2–5 m)
Number of private boreholes on record	>110,000	>1900
Volume from private boreholes/day	N/A	80,000–350,000 m^3
Volume from other private wells	N/A (negligible)	N/A (substantial)
Poor people's direct dependence on groundwater	Via public standpipes; utility tankers; vendors; shared wells	Via own dug wells and community-based wells; vendors
Poor people's indirect dependence (conjunctive use)	–	Piped water supply (house connections); communal taps; kiosks

In critical respects, the prevailing conditions are fundamentally different in these two cities although residents in both rely heavily on groundwater: in Bangalore, where the utility's main water source is River Cauvery, groundwater is predominantly pumped from deep boreholes in the low-yielding crystalline bedrock. The wells are not only drilled by middle- and upper-class landowners but also by the public utility (and occasionally by a politician for the benefit of the poor). Unreliable electricity supplies create problems, as do poor maintenance and slow repair of pump motors that frequently burn out because of voltage fluctuations and have to be rewound. Furthermore, many experience sinking water tables and wells that dry up, or that fail already at the attempt to drill them, whereas others claim to have noted no significantly lowered groundwater levels. The concept of a ‘sustainable’ groundwater development deserves a wide interpretation under the prevailing conditions, because especially people residing in the peri-urban areas of Bangalore have few alternative water sources except for purchasing it from someone with a good yielding well. Slum dwellers rely mainly on irregular but essentially free water supplies from public standposts, some of which are connected to stand-alone wells. A slowly increasing number of slum dwellers are connected to the water supply network but no one from this group really have access to water from an own well; drilling and pumping equipment is too expensive.

Bangalore is in several ways a typical city of a middle-income country that experiences fast population and economic growth, and simultaneously faces a situation where the water made available from a surface water source at some distance falls short of the rapidly increasing demand. The complexities involved in water management are further compounded when groundwater dependence constitutes a significant component – as the sole source or one of several used. However, as groundwater is mainly accessed through deep boreholes belonging to more or less wealthy landowners, data consist of very vague estimates only. With regard to the wells belonging to the public utility, data are also scarce because concerted efforts to collect such are still to develop (see BWSSB 2010). Steps to induce aquifer recharge are most pertinent to the situation of Bangalore where whatever is feasible should be done to halt a lowering of the water table in affected areas. Measures to store water from precipitation through, for example, rooftop and underground structures are taken, although not yet on full scale. The public water supply utility has made rainwater harvesting mandatory – but only for its customers and certain buildings – whereas policymakers are so far reluctant to introduce legislation for groundwater abstractions from private wells. However, the political will to do so is likely to be of importance for the entire city's potential to access water in the future.

In Lusaka, the karst terrain makes for circumstances where the groundwater table is often extremely shallow, but the small depth to the system of underground channels and cavities reduces and/or eliminates completely the natural attenuation of pollutants that would otherwise occur through natural filtration. Due to this, the groundwater especially in Lusaka's low-income settlements is essentially as easily polluted as surface water in a stream. The water quality is therefore a major problem particularly during the wet season, and cholera outbreaks linked to oral-faecal transmission are common. Very many among the poorest access their water from shallow, unprotected wells dug in their own yards – the kind that are most at risk because of inferior solid waste disposal and on-site latrines – and can seldom afford to or bother to treat the water. Substandard sanitation and hygiene conditions are obstacles to improved well-being, and together with a low level of hygiene awareness this leads to ill health with endemic diarrhoeal disease and regular cholera outbreaks. The authorities are concerned by the dug wells and the official policy is to close them down. Doing this, the responsible authorities send out a dual health message: they seek to prevent people from using their wells for fear of disease but do not provide alternatives in the form of easy and/or affordable access to water, or improved infrastructure for sanitation, drainage, solid waste disposal and so on.

Whereas planners, the water supply authority, regulators and policymakers and so on in Lusaka are thus fully aware of the importance and widespread use of water from various kinds of wells in the city, more precise information about the extent of the groundwater dependence directly from wells and indirectly through the reticulated water distribution system is not available. Too little is also, generally, known about the causal relationship between water sources and disease transmission routes. Assumptions in this regard to the effect that certain sources are unapproved of have little impact, though. It is futile to think that the poorest in a city such as Lusaka, if self-supplying from wells for lack of affordable or otherwise accessible alternatives, will discontinue their use of available (ground)water sources because the authorities refer to potential health risks. According to a tap attendant in one settlement, people resort to quite innovative measures to cover their wells and avoid them being closed down (Veronica Katulushi, pers. comm.).¹

7. Concluding remarks

Increased water availability is key to improved health and quality of life. It can be assumed that urban poor people's dependence on water from wells of different kinds remains as large as ever, but that this is not fully appreciated in planning and decision-making at strategic city level or in a regional, Integrated Water Resources Management perspective. Being hidden and out of sight underground contributes to that in general, both monitoring of and statistics over urban groundwater use is very poor, and estimates are often dated and usually based on many assumptions (Foster 2009). To this comes that the poor as a group generally lack a voice to express their actual alternatives and strategies for water access, for instance regarding the multitude of sources that may have to be used, and reasons behind attitudes towards point-of-use treatment. Together these flaws result in an under-estimation of groundwater as a domestic water source today. This may, among other things, impede the follow-up of the intentions behind MDG target 10 – and of further assistance in serving people in low-income areas. It may furthermore undermine the potential for informed research and debate on water access and poverty alleviation.

The improved–unimproved dichotomy of MDG target 10 is of little value in relation to groundwater use and urban self-supply. Most poor people would benefit from access to larger amounts of water than are currently available to them, irrespective of the water quality (given that there is, for instance in Bangalore, often a high awareness of different water sources' potability, meaning that some are used only or mainly for non-drinking purposes). This objective cannot be reached if they are discouraged from using water from unimproved sources such as certain kinds of wells.

It is also reasonable to believe that the costs of efforts to improve unprotected wells would ultimately have to be met by the end users themselves. Instead, if the quality of the water from particular wells is actually found to be substandard and non-potable, this problem can be dealt with in a number of ways. Deterioration of the water quality should be prevented at all levels and by all means, but education about suitable point-of-use treatment, hand-washing and other hygiene-promoting activities are also vital to raise awareness about the various transmission routes for diarrhoea and the importance of safe water storage in the home. This is often more critical than the quality of the water source as not all water used must be of potable standard.

The increasing dependence on groundwater and wells needs to be acknowledged for the resource to be accurately taken into account by city planners and decision-makers. Measures need to be taken to safeguard this resource from contamination, both by users themselves and by the community and authorities. Groundwater that is already contaminated must be publicly recognised as such, though. A sufficient level of protection would entail a holistic awareness of aquifer systems, groundwater and well structures of different kinds, for instance through a renewed approach to the concept of Integrated Water Resources Management, one that takes the urban administrative scale into and not only the more traditional regional river basin perspective. For instance, the city of Bangalore is divided by a ridge and is drained by two different river basins. The water supplied by the public utility comes from only one of the rivers (the Cauvery), but a majority of Bangalore's residents self-supply with groundwater, either as a complement or as the sole water source. The quality of the groundwater is affected mainly by sewage, and the water table is sinking in parts of the city (Central Ground Water Board 2008). To deal with these problems, different institutions need to be better integrated.

Measures taken in relation to groundwater use and preservation always need to be contextualised; for instance, steps relating to rainwater harvesting and other aquifer recharge-inducing steps may be pertinent in most city environments where large amounts of water are drawn from wells. However, in a city like Lusaka such measures are largely irrelevant, considering how shallow its groundwater table is for most parts of the year. Rather, improved drainage possibilities to avoid floods during the rainy season would be important, together with treatment of water for drinking and cooking, and improvements of sanitation conditions.

This article and the IIED Working Paper it draws from show that even with insights into the local context it is not possible to draw many far-reaching, generalised conclusions on groundwater conditions and dependence from the data available. Follow-up studies of urban self-supply are necessary to increase understanding and awareness of local situations, including hydrogeology and well-monitoring circumstances as well as government policies, institutional capacity and potential interventions by way of foreign aid programmes and NGOs to complement the self-supply measures taken at household and community level.

Acknowledgements

This article is drawn from a longer IIED Working Paper by Jenny T. Grönwall, Martin Mulenga and Gordon McGranahan, ‘Groundwater, self-supply and poor urban dwellers: A review with case studies of Bangalore and Lusaka’ (2010). The author wishes to thank two anonymous reviewers for valuable comments.

Additional information

Notes on contributors

Jenny Grönwall

Jenny T. Grönwall is a freelance researcher trained as an interdisciplinary lawyer with subsequent focus on urban water and sanitation questions and special interest in the Bangalore situation. With a PhD from Linköping University, she has written on Indian water law and ecosystem services, equity and urban mapping at the Stockholm Resilience Centre as well as on groundwater and self-supply for the International Institute for Environment and Development (IIED).

Notes

The author was formerly with the Stockholm Resilience Centre.

1. Veronica Katulushi, tap attendant, George compound, Lusaka, February 2010 (as interviewed by Martin Mulenga, IIED).