Causes of intermittent water supply in Lusaka City, Zambia

Water distribution systems are typically designed with the intention that they operate continuously under positive pressure. However, water supply systems in many cities around the world actually operate intermittently. In such systems, water supply duration varies greatly. In some cases, water is supplied less than 7 days a week for between 4 and 19 hours or more on supply days (Sashikumar et al. 2004). In others, water is supplied daily but for durations ranging from 1 to 2 hours (Ingeduld et al. 2008) to more than 23 hours (Danilenko et al. 2014). These variations can occur between district metered areas (DMAs) or townships on the same distribution system (Shiliya 2013).

Apart from supply scheduling, intermittent systems are complex, with large numbers of factors, both external and internal to the intermittence, affecting them. External factors include poor governance, demographic and economic dynamics, hydrologic regime change, poor system management and operation, unplanned system extensions, limited skilled manpower, poor electricity supply and lack of customer awareness. Internal factors are part of the self-reinforcing mechanism in the vicious downward spiral of intermittence (Ingeduld et al. 2008). They include excessive leakage, apparent water losses and poor compliance with paying water bills. The complex interactions between these factors sustain the intermittence and have major consequences for system operation and management. Handling such problems requires understanding and addressing their causal factors. The Lusaka Water and Sewerage Company (LWSC), Zambia, is used as a case study in this paper, which shows how the various external factors contribute to the water supply's intermittence in Lusaka. The work is based on a desk study and interviews with key stakeholders.

The government of the Republic of Zambia started water sector reforms seriously after 1991 and completed them with the National Water Policy (NWP) of 1994 (Chitonge 2011). The reforms were a donor-driven policy option, not led by the Zambian government (Op cit 2011). The reforms were meant to address problems faced by the sector, including poor legal and institutional frameworks, deteriorating water and sanitation services, inadequate human resource capacity, low stakeholder and community participation, limited and ever-decreasing capital investments, and the need to adapt to emerging international trends in water management (Ministry of Energy and Water Development 1994; Republic of Zambia 2010; Chitonge 2011).

Implementation of the NWP was to be guided by seven sector principles:

• (1)

  Separation of water resource functions from water supply and sanitation;

• (2)

  Separation of regulatory and executive functions within the water supply and sanitation sector;

• (3)

  Devolution of authority to local authorities and private enterprises;

• (4)

  Achievement of full cost recovery, in the long run, for water supply and sanitation services through user charges;

• (5)

  Human resource development leading to more effective institutions;

• (6)

  Technology appropriate to local conditions; and,

• (7)

  Increased Government of the Republic of Zambia (GRZ) spending priority and budget spending in the sector.

The key documents comprising water sector policy and the related legal framework are set out in Table 1 (NWASCO 2014b).

The major institutions involved in the Zambian water sector are the Ministry of Local Government and Housing, responsible for resources mobilization, the Ministry of Water Development, Sanitation and Environmental Protection, whose responsibilities are reflected in its name, and the National Water Supply and Sanitation Council (NWASCO), which regulates the water supply and sanitation sector, and advises the two ministries on water supply and sanitation issues. Even after new policies/laws and a comprehensive institutional framework, the water supply sector faces many challenges – see Table 2.

Water supply and sanitation infrastructure and services were deteriorating under the management of Lusaka City Council. To reverse this, commercialization (rather than privatization) of the water supply and sanitation services was seen as a means of increasing efficiency and reducing the fiscal burden on the state (Chitonge 2011). Lusaka's water and sewerage services were commercialized in 1988.

The infrastructure in Lusaka was taken over by LWSC, which started functioning as a limited company in 1990. While it currently supplies water and sanitation services to the whole Lusaka province (Figure 1), this paper focuses only on Lusaka city's water supply system (Figure 2).

Lusaka's water is supplied from both ground- and surface- water sources. More than 116 boreholes within and around the city provide 57% of the total supply. The remaining 43% is surface water abstracted from the Kafue River and treated at Iolanda water treatment plant (WTP), about 50 km from the city (Ndongwe 2013). While groundwater yield is limited, the river has unlimited yield for domestic purposes. The limiting factor is the WTP's capacity. Renovation of the existing plant and construction of another nearby, by the Millennium Challenge Corporation (MCC), are underway (Millennium Challenge Account – Zambia Limited 2013). The amounts of water that have been and are expected to be supplied by boreholes and WTPs up to 2035, and the corresponding water demand, including non-revenue water (NRW), are shown in Figure 3, which takes existing and planned WTP development into account.

From Iolanda WTP, water is pumped to Chilanga booster station and lifted to Stuart Park terminal reservoirs (Figure 4). Stuart Park is about 325 m higher than Iolanda. A pipe before Stuart Park takes some water from Chilanga to Lusaka water works where it is pumped to reservoirs at Lumumba Road, Woodlands and High Court.

The management of the network and water distribution is done through five zones or branches – Central, Chelstone, Kabulonga, Kabwata, and Lumumba including Matero (Millennium Challenge Account – Zambia Limited 2013) – which serve unplanned settlements (peri-urban areas) within their jurisdictions (Republic of Zambia 2011).

Water is fed to the branches' distribution systems from the reservoirs at Stuart Park, High Court, Woodlands and Lumumba Road, and various boreholes. The reservoirs and boreholes are managed by the production division, which supplies the branches intermittently. Each branch rations the water between DMAs to try and balance supply durations between DMAs connected to different sources.

Much of the water distribution system was built in the 1960s and 70 s (Government of the Republic of Zambia 2010). In 2013, some 21% of the network's distribution and transmission mains were considered to be of unknown age (Ndongwe 2013). The total network length was 1,308 km comprising 109 km transmission mains and 1,199 km of distribution mains. Many new extensions and network upgrades are now under way, which may change these statistics soon.

In developed countries, temporal events like droughts, pollution incidents, earthquakes, maintenance, and pipe or pump failure may result in water supply intermittence (Chandapillai et al. 2012). This may be called partial intermittence because, in most cases, only part of the system is affected and the supply normalizes when the cause is removed. In developing countries, intermittence could be said to be full-time, because water rationing and reduced supply duration are norms across the system, and it is not obvious how to achieve continuous supply. Temporary causes also arise, but seem insignificant and are obscured by major external causes discussed below.

Poor governance exists in the absence of common governance characteristics like participatory decision-making, adherence to the rule of law, transparency in dealings and accountability (Rogers & Hall 2003).

In Zambia, water infrastructure deterioration can be attributed to poor governance because of lack of resources due to the poor institutional and legal framework. These lead to a bloated civil service, lack of accountability, plundering of resources and corruption. The existence of these problems and the inertia in solving them could have been the concern of the UK's Department for International Development (DFID), which was working in Zambia in 2012 to support the Zambian government to improve its systems and skills, use its resources to deliver better services, beat corruption, and increase transparency and accountability to its citizens (DFID 2012).

Population growth and urbanization in Lusaka cause many service provision problems. These are amplified by poor governance aspects of land allocation that promote the mushrooming of peri-urban areas. More than 60% of Lusaka's population lives in such areas (Government of the Republic of Zambia 2010), which have been developed on plots obtained by residents illegally because of the failure of the formal land delivery practices (Lusaka City Council and Environmental Council of Zambia 2008).

Peri-urban areas in Zambia, like in many developing countries, lack water supply and sanitary facilities (de Waele & Follesa 2003; Cohen 2006; Republic of Zambia 2017) because utilities or governments are unwilling to provide services in them (Chitonge 2011). Independent supply systems, set up by NGOs and donors, face serious capacity and management problems (Dagdeviren 2008). Moreover, because there is no order in the alignment of plots and structures, service provision is difficult (Lusaka City Council and Environmental Council of Zambia 2008). Due to lack of waste disposal facilities, peri-urban areas also contribute heavily to both ground- and surface- water resource contamination (Republic of Zambia 2011; International Water Stewardship Programme 2016).

It has long been believed in Zambia that the country has an abundance of water resources. It is said that there is enough water to meet national demand even in a drought year (Government of the Republic of Zambia 2010). However, as shown in Table 2, water resource depletion has affected water supply since 2013. This, coupled with the low water levels in key hydropower production reservoirs in 2015, has changed the perception. The need to manage water resources has become clear but many factors make it difficult. It is known, for instance, that land use practices contribute significantly to hydrologic regime changes, land covered with forests helping to sustain stream base flows (Mishra et al. 2007). In Zambia, deforestation for settlement, agriculture and energy (charcoal) is rampart, and contributes to the reduction of both surface- and ground- water quantity and quality. Due to illegal land allocations, many water supply facilities have also been encroached (Ndongwe 2013), and the Zambian government seems unconcerned about protecting Lusaka's water sources. This is demonstrated, for instance, by location of the economic facility zone in south-eastern Lusaka (local forest No 26), an area known to be a recharge zone for the Lusaka aquifer (Bäumle & Nkhoma 2008; International Water Stewardship Programme 2016).

Poor system management and operation comprises many issues. One critical aspect is database management, which, when poor, makes effective analysis of the state and performance of water supply system infrastructure difficult (Klingel 2012), and revenue collection inefficient (Biswas & Tortajada 2010).

Because of this issue, LWSC took over the pipe network without proper records, which made it difficult to locate some network pipes and junctions. This has continued to be a problem. Network connectivity and sources are not clear, and DMAs are developed based on township boundaries. These characteristics make controlling water theft and leakage difficult. In fact, LWSC has challenges even fixing reported, visible leakages. This is attributed to the poor maintenance and operation budget, due to LWSC's poor financial position and laborious procurement procedures leading to lack of maintenance materials (Republic of Zambia 2011). Leakage contributes greatly to the utility's failure to meet the targeted NRW levels (Figure 5).

Tariffs, as a management tool, are too low. Instead of full cost recovery, current tariffs are limited to operation and maintenance costs (NWASCO 2016). For metered consumers, a two-part increasing block tariff structure is used. The lower block, up to 6 m³/month for a family of 6, is the social tariff, and is at or below the cost of service provision (NWASCO 2014a).

For unmetered consumers, a fixed charge tariff is used. This is estimated by metering a representative sample of consumers in the area in question and computing an average consumption for 12 months. Where there is a similar category of customers already metered, the metered customers' average consumption is applied (NWASCO 2014a).

Water distribution system extensions are important in developing countries because many residential areas do not have access to safe drinking water. However, in Lusaka, the problem is that such extensions are mostly unplanned because the utility company supplies water to locations after they are developed. This can be attributed to poor city planning. Due to lack of infrastructure investment (Republic of Zambia 2017), limited new water sources have been developed, although there are many new extensions. This has reduced supply duration and water volumes for areas that were supplied originally. Water resource addition has been principally from boreholes but this has been slow as it depends heavily on donor support.

In developing countries, there is typically a lack of the technical and managerial skills needed to improve water and sanitation service delivery to match rapid population growth and urbanization (Blair 2005; Cohen 2006). In Zambia, lack of skilled manpower in the water supply sector is a problem, although this was not apparently recognized in the sixth national development plan where the need to develop skills was seen mainly as an issue in water resources management (Republic of Zambia 2014). Whilst the concept of capacity building is admirable, the plan is short on detail as to what skills or capacity actually needs to be developed.

Electricity supply is very important to water utilities for pumping. As shown in Table 2, frequent power outages have contributed to the poor water supply since 2008. The power supply problem, as experienced by LWSC, is the absence of a shut off schedule from the electricity utility company, ZESCO limited. It appears that LWSC has no solution to this because many parts of the distribution system receive water directly from booster pumping stations. Because the supply boreholes are in residential areas, power cut off to a given area also affects boreholes there, and the use of elevated tanks to cushion the effects of power cut offs is minimal.

Lack of awareness has effects on water consumers as well as utilities. In Lusaka, consumers are affected because they have no basis for demanding improved supplies as they have experienced nothing better than the levels currently received. Utilities are affected by consumer's unwillingness to pay water bills, which contributes to the self-reinforcing intermittence mechanism. In this case, government fails to provide leadership as government entities are among the key payment defaulters (Table 2). Because of this, collection efficiency from the GRZ is a utility performance indicator (Millennium Challenge Account – Zambia Limited 2013). Comprehensive and lengthy customer awareness campaigns will have to be undertaken to influence water consumption trends or improve cooperation with new measures by utilities.

Most of the problems identified prior to the 1994 NWP still haunt the water sector in Zambia. Of the seven sector principles, only separation of regulatory and executive functions within the water supply and sanitation sector, and devolution of authority to local authorities and private enterprises were achieved by 2010 (Republic of Zambia 2010). This has contributed greatly to sustaining water supply intermittence in Lusaka.

Zambia's dependence on donors for the water supply development planning period 2010–2030 is evident in the National Urban Water Supply and Sanitation Programme (Government of the Republic of Zambia 2010). Almost all major infrastructure developments in the country are financed by donors. For example, the ongoing Lusaka Water Supply, Sanitation and Drainage infrastructure development project is an American Aid project through the MCC (Millennium Challenge Account – Zambia Limited 2013). Thus, it can be said that improving Lusaka's water supply system depends on donors improving the supply system's capacity rather than the Zambian government.

Water tariffs in Lusaka are generally lower than elsewhere in the sub-Saharan Africa but high poverty levels mean that many people spend more than 5% of their income on water (Dagdeviren 2008). Poverty levels are a governance issue (DFID 2012). Good governance would lead to increased investment in network extension and infrastructure renewal. This can reduce water losses and costs, thereby making water affordable for many and making it easier to achieve full cost recovery (Dagdeviren 2008). The cost of water supply is greatly affected by two factors. The first is the 75% increase in electricity tariffs in 2017, which is critical because of the extensive pumping from boreholes and booster stations. The other is that the cost of supplying water from Iolanda WTP is about 5 times more than that from boreholes (Republic of Zambia 2011). Thus, the increasing use of water from Iolanda necessarily implies supplying more costly water. In view of high poverty levels, the implications of raising tariffs can be dire. Water price segregation could offer an alternative to the low/high tariffs paradox in Zambia, because the variation in incomes is great, and basing tariffs on those with lowest incomes and poorly supplied with water, may be misleading.

A number of factors contribute to the high proportion of NRW. Leakage is high even when the distribution system's pressure ranges between 5 and 10 m in the dry season – the maximum is 15 m in the rainy season. System metering is poor. Much of the borehole water entering the system is unmetered and many meters malfunction. With poor system knowledge, there is concern that much of the increased supply from the WTP will be lost through leakage. Moreover, with no means of resolving capacity issues or improving maintenance habits, the infrastructure will become dilapidated in the long run, resulting in further poor water supply services.

The intermittent water supply mode in Lusaka and its continuation are issues of governance, and natural and technical problems. Thus, interdisciplinary approaches to exploring the interplays between these factors are recommended when developing solutions to supply intermittence in Zambia. The emphasis should be on identifying specific aspects that require human capacity development. This is critical as ignorance of the real problems could be a tragedy. As Mulambya said: ‘Tragedy is when the ignorant are ignorant that they are ignorant.’

The Commonwealth Scholarship Commission (CSC) is gratefully acknowledged for sponsoring PhD studies for the first author (ZMCA-2015-141). The authors further acknowledge the support of the UK Engineering and Physical Sciences Research Council (EPSRC) (EP/K006924/1).