Document(s) 8 of 19

Abstract

The range of factors that determine the quality of wastewater used by different irrigators is described, drawing on case studies from Nairobi, Kenya and Kumasi, Ghana. Not all urban irrigation relies on raw wastewater and it is misleading to consider wastewater as a uniform commodity. Dilution and natural remediation mean that irrigators use a range of water qualities and the authors raise the question of when a dilute wastewater stream is no longer classed as wastewater. World Health Organization (WHO) guidelines for the design of wastewater treatment plants are widely used as standards to judge the quality of untreated irrigation water. However, because of the gap between standards that lead to ‘no measurable excess risk of infection’ and the actual situation pertaining in many cities, urban planners either condemn all urban irrigators as posing a major health risk to the community, or turn a blind eye. The authors argue that a standard leading to ‘no measurable excess risk’ to health is an unattainable and unhelpful medium-term goal under the conditions of indirect wastewater use seen in many cities. Instead, there is a need for explicit debate of the levels of risk that may be acceptable to individuals and communities, and the costs and benefits that they bring with them. Informed debate, that is enabled to assess the risks associated with different water qualities and irrigation practice, may lead to the development of local water quality norms and wastewater management that account for the physical and social environments in which wastewater irrigation is actually practised.

Introduction

Types of urban wastewater irrigation

The information presented here comes from a larger study of urban and peri-urban irrigation practices carried out in Nairobi, Kenya and Kumasi, Ghana, from 1998 to 2001. That research aimed to describe and quantify the nature, extent and importance of informal, irrigated agriculture in the urban and peri-urban zones of those cities (Cornish et al., 1999, 2001; Cornish and Aidoo, 2000; Hide and Kimani, 2000; Hide et al., 2001). The research focus was not confined to irrigation with wastewater or the hazards associated with its use. Rather, the intention was to understand the range of practices that exist with regard to water sources, water and crop management, crop marketing, and the contribution of informal urban and peri-urban irrigation to household income and expenditure. The research showed that in both cities a minority of irrigators use the urban potable water

supply; many use shallow groundwater that is polluted to varying degrees whilst others draw water from streams or rivers that are also polluted to varying degrees by untreated, industrial and municipal wastewater. In Nairobi, 34% of the irrigators sampled diverted untreated sewage from trunk sewers directly onto their land. In Kumasi there is no extensive piped sewerage network and urban wastewater is either collected in septic tanks that are periodically emptied by tanker, or it is discharged directly into the small streams and rivers that drain the urban area. Tankers that empty the septic tanks discharge their contents into derelict waste stabilisation ponds that overflow directly into a river. Thus, whilst there is no direct use of untreated wastewater in Kumasi, many irrigators who draw water from the rivers downstream of the city are using a diluted wastewater stream. Table 6.1 summarises the different water sources used by informal irrigators in the two cities.

Table 6.1. Percentage of urban and peri-urban irrigators sampled drawing water from different sources.

In introducing this chapter the following points are emphasised:

1. Treated wastewater is not being used for irrigation in either city to the best of the authors’ knowledge. Irrigation is informal and irrigators obtain water where they can. In many cases their water source is highly polluted and in Nairobi raw sewage is used. It seems reasonable to presume that informal use of dilute and undiluted, untreated wastewater is common in other urban areas in sub-Saharan Africa.
2. It is an over simplification to consider ‘urban wastewater irrigation’ as a single activity with uniform characteristics, amenable to a standard response from planners, policy makers or technologists. Rather, there is a range of different physical conditions under which urban wastewater irrigation occurs. These conditions influence both the levels of risk to health faced by growers and consumers, and possible interventions that may reduce those risks while maintaining the benefits to irrigators and possibly to the wider environment. Recognition of this variation in conditions is essential to any effective discussion of wastewater irrigation practice, or to the formulation of recommendations regarding its regulation.
3. The issue of mixing, and thus diluting wastewater, with water from a natural water body merits comment: at what point does urban wastewater become simply a polluted water body? Many will know of urban ‘rivers’ and other water bodies that are little more than open sewers or cesspits. Although some mixing and dilution of wastewater has occurred in these water bodies it seems misguided to exclude them from a consideration of wastewater irrigation as they are characterised by the presence of urban wastewater. There is a need to define a level of dilution at which wastewater becomes polluted ‘natural’ water, but proposing that definition lies beyond the scope of this chapter.

Figure 6.1 shows the range of factors that determine the nature of wastewater irrigation at any location. The only non-physical factor considered is whether the irrigation takes place in a formal (authorised) or informal (unauthorised) setting. The figure does not include the wider social, economic or institutional factors that influence any given practice, although these are recognised as having an important influence on irrigators’ behaviour.

Figure 6.1 may not constitute a formal typology of wastewater irrigation, but it emphasises the range of factors that influence both the physical and biochemical quality of wastewater used for irrigation. The elements of Fig. 6.1 are used to describe three different types of wastewater irrigation practice drawn from sites in Nairobi and Kumasi.

Fig. 6.1. Factors determining the nature of wastewater irrigation. Diluted = Effluent mixed with other water before use in irrigation. Undiluted=No significant dilution of the effluent in a river or other water body before use in irrigation. Formal use = Use of wastewater with a certain level of permission and potential control by state agencies. Informal use = Use of wastewater without permission and control by state agencies.

Physical factors include the source of the wastewater, the means by which it moves from the source to the field, and whether or not any treatment occurs. ‘Discharge’ describes whether or not the wastewater is discharged into an intermediate water body – surface or groundwater – where dilution occurs before an irrigator obtains it for use. The differentiation between formal and informal (authorised/unauthorised) irrigation – an institutional factor – is often determined by whether the wastewater is obtained from a small number of potentially controllable locations, or from numerous, unknown locations. The on-farm conditions identified are those considered to have the greatest influence on the level of risk to health for either the irrigators or those consuming the crops they produce.

Types of Wastewater Irrigation in Nairobi and Kumasi

Mau Mau Bridge, Nairobi

Mau Mau Bridge lies upstream of Nairobi’s city centre and its industrial zone (see Fig. 6.2). There are irrigated farm plots adjacent to the Nairobi River. Farmers have constructed small dams and weirs in the river to divert water through channels to the lower areas of their farm plots. Using buckets and watering cans, water is drawn from hand-dug ponds at the end of the channels, to irrigate crops at higher elevations in the farm plots. On-farm irrigation methods therefore include surface furrows or basins and overhead sprinkling from cans.

Although Mau Mau Bridge is situated upstream of the main city and industries, slums are located on the slopes above the Nairobi River. Waste and wastewater from the slums are dumped onto the streets and into natural drainage channels from where they find their way into the river. Thus, untreated municipal wastewater mixes with river water and it is this mixed water that the irrigators at Mau Mau Bridge use.

A typical plot size is 60 × 20 m and farmers grow a mixture of vegetables, including tomatoes, cabbage, spinach, maize and French beans. Some of these are eaten raw and others are cooked before consumption. Crops are mainly grown for the local market but small quantities are also consumed by the irrigators’ families. All members of the irrigators’ families carry out irrigation and other farm work.

Maili Saba, Nairobi

Maili Saba is 15 km east and downstream of Nairobi city (see Fig. 6.2). There are both similarities and contrasts with Mau Mau Bridge in the way wastewater is obtained and used. At both sites the practice is informal with no government permission or infrastructure provided to support irrigation. However, at Maili Saba farmers remove manhole covers and block the city’s main sewer, diverting raw sewage onto their land. Their plots, typically 20 × 40 m, are irrigated by surface irrigation from a hand-dug canal system. Buckets or watering cans are not used. Irrigators grow kale, sweet potato, arrowroot and some green maize – crops that are cooked before being eaten. Much of the production is for home consumption but some is sold at the local markets. Assuming that the produce is well cooked the health risks associated with the use of undiluted sewage are confined to the family members including men, women and children who carry out the irrigation.

Fig. 6.2. Location of water-sampling sites () with a 20 km radius of Nairobi city centre (Hide et al., 2001).

Asago, Kumasi

Asago is situated 9 km downstream of Kumasi at the confluence of the Sisa and Oda Rivers (see Fig. 6.3). The Sisa collects untreated and partially treated municipal wastewater and untreated industrial wastewater. The wastewater constitutes municipal and industrial effluent, conveyed to the river by both road tankers and natural drainage flows. Farmers at Asago draw irrigation water from the perennial River Oda either by bucket, or other container or using motorised pumps (hired or owned). Considering the factors identified in Fig. 6.1, wastewater irrigation at this site is informal use of diluted wastewater using river water that has been mixed with untreated or insufficiently treated wastewater from stabilisation ponds.

Fig. 6.3. Location of water-sampling sites relative to Kumasi city centre (Cornish et al., 1999).

All farmers use some form of overhead application to irrigate a mix of vegetable crops including tomato, African aubergine (Solanum integrifolium) okra and chilli, some of which are always cooked and others eaten raw. Water is applied with watering cans, buckets or perforated tins. Irrigators who use pumps use PVC pipes to convey the water from the pump to a position within their fields and connect a short length of 50-mm lay-flat hose to the final pipe length. A worker then stands and sprays water from the hose-end onto the crop.

The vegetables are mainly grown for the Kumasi market but small quantities are also consumed at home by all members of the family, who carry out irrigation and other farm work.

Variations in Water Quality Between Sites

Field measurements of faecal coliform numbers demonstrate the large variation in microbiological water quality between sampling sites and the danger of considering all urban wastewater irrigation as equal. In all but two of the sites the mean faecal coliform count exceeds the World Health Organization (WHO) Health Guidelines for Use of Wastewater in Agriculture and Acquaculture (WHO, 1989) but the degree of exceedence varies widely. The question of whether guidelines, developed for the design of wastewater treatment plants assuming a requirement for ‘no measurable excess risk’ are appropriate and adequate for making judgements of health risk in diverse field conditions such as these, lies at the heart of this chapter and is examined in more detail below.

Following the example of Westcot (1997) mean faecal coliform count was used as the sole indicator of biological water quality for health risks. It is recognised that helminth infections pose the greatest of the risks associated with wastewater irrigation and that the WHO guidelines specify threshold values for both faecal coliform and helminth egg numbers. However, whilst laboratories and technicians are readily able to measure faecal coliform numbers, procedures for the accurate detection of helminth eggs are more demanding and less widely known. For this pragmatic reason, helminth egg numbers were not measured or reported.

Figure 6.4 shows the mean numbers of faecal coliforms recorded at different locations in a. Nairobi and b. Kumasi. Five samples were collected at 10-day intervals over a 40-day period. The sampling sites included three river sites, one well and one sewerage outlet. Their location relative to Nairobi City centre is shown in Fig. 6.2. The sampling sites in Kumasi included seven river sites and two wells. Their locations relative to central Kumasi are shown schematically in Fig. 6.3. Five samples were collected at 6-day intervals over a period of 26 days. In both cities the sampling period coincided with the dry season, when irrigation is mainly practised.

Most of the data from Nairobi show very high levels of pollution. Numbers of faecal coliform in the Nairobi River at Kimathi and Njiru Bridge, both situated downstream of the city centre, are as high as those recorded in effluent drawn directly from sewerage mains at Maili Saba. This is 10,000 times greater than the limit for unrestricted irrigation recommended by the WHO design guidelines for treated wastewater. Water at Mau Mau Bridge contains faecal coliform numbers that are 10 times greater than the recommended value. Mau Mau Bridge is situated upstream of Nairobi’s city centre where the Nairobi River water has only been mixed with municipal wastewater collected and disposed into the river through natural drainage channels. Only water drawn from the shallow well at Thiboro, upstream of Nairobi, yields water that lies within the WHO guideline limits.

In Kumasi levels of pollution are generally lower, with water from the two sites upstream of the city centre lying on or near the WHO threshold value for unrestricted irrigation. Asago, the most highly polluted site, exceeds the guideline by only 2-log. At Asago farmers draw their water from the perennial River Oda. This water is mixed with municipal and industrial effluent conveyed to the river by road tankers and natural drainage flows.

There is clearly great variation in the quality of the water used at different locations. This must be recognised in evaluating the likely health risks. Single threshold values, intended as a guideline in the design of treatment plants, even when they account for different forms of irrigation and crop types say nothing about the different levels of risk posed at these various sites.

Positive Impacts of Urban Wastewater Irrigation

The extent of urban wastewater irrigation and its contribution to food security

The areal extent and the number of households relying on irrigation within the two study areas are shown in Table 6.2. It is important to note that not all of these irrigators are directly reliant on wastewater. The two city studies characterised urban and peri-urban irrigation irrespective of water type. Farmers using shallow groundwater in areas remote from rivers draining the urban centres are not using a wastewater source, although the shallow wells sampled in Fig. 6.4 still indicate relatively high levels of faecal coliform contamination in the water they use.

Table 6.2. Extent of informal irrigation in the study areas.

^a Estimates are based on sampling of 410 farmers in Kumasi and 158 farmers in Nairobi.

    Mean number of faecal coliform/100 ml

1. Nairobi
   

   Mean number of faecal coliform/100 ml

2. Kumasi
   

Fig. 6.4. Geometric mean faecal coliform (count/100 ml) and coefficient of variation (CV) at different sites in Nairobi (mean of at least 5 sampling dates (Hide et al., 2001).

The large area of informal irrigation within a 40-km radius of Kumasi contrasts with the 6,400 ha under formal irrigation reported in the Food and Agriculture Organization’s (FAO’s) statistics for the whole of Ghana (FAO, 1995). Kumasi alone supports an area of informal irrigation almost twice that of all formal irrigation in the country, and further substantial areas of informal irrigation exist around Accra and Takoradi.

The smaller area of informal irrigation identified around Nairobi was recorded over a much smaller study area. Irrigated crop production is, for many, a relatively new activity. It is quite possible that such wastewater irrigation will continue to expand in the coming years. Figures on irrigated areas in Kenya for 1998 reported by the Ministry of Agriculture and Rural Development (cited by Muchangi in HR Wallingford, 2001) identify only 1,500 ha of urban irrigation for the whole country. This study identified more than 2,200 ha of informal irrigated agriculture within 20 km of the centre of Nairobi. As in Ghana, it appears that the extent and importance of urban irrigation is under-reported in official statistics.

In Nairobi, the average annual revenue per ha from irrigated plots is US$1,770, indicating that from the urban irrigated sector vegetables worth as much as US$3.9 million are being used in Nairobi each year. The seasonal (November–May) average revenue per ha for production around Kumasi was US$544 indicating a total value of food production in excess of US$6 million.

Fig. 6.5. Seasonal (November-May) profits (US$/ha) from irrigated cropping for 21 Kumasi farmers in three villages (Cornish et al., 2001).

The studies did not determine what fraction of the total annual vegetable consumption of the two cities these production figures represent, and did not consider that all of the production does not pass through the main city markets; some is sold in smaller, local markets outside the urban centres. However, it is clear that informal urban irrigation, much of which relies on wastewater sources, contributes significantly to the supply of fresh vegetable produce in both Nairobi and Kumasi.

Contribution of urban wastewater irrigation to the livelihoods of irrigators

Income generation is the main objective for most irrigators in Nairobi and Kumasi. Only a small percentage of the farmers surveyed said that directly supplementing their food supply was their main goal. By generating cash income, urban irrigation is an important means of alleviating poverty and enhancing livelihoods.

The average profit recorded in three different peri-urban villages around Kumasi is remarkably similar, indicating that different sources of water, distance to market, or other factors determined by location do not have a major influence on profit. There is certainly no evidence that water quality influences levels of income in the Kumasi study.

Although average profit in each village is similar, Fig. 6.5 shows that there is a wide range of levels of profit recorded by individual farmers. Numbers on the x-axis identify individual farmers – farmers 11–17 from village 1 (Dedesua), 21–27 from village 2 (Baworo) and 31–37 from village 3 (Atia). Farmers from all three villages are distributed across the whole range of profit per ha. Although the average profit is around US$340/ha, four of the farmers recorded profits of between US$650–800/ha. As the actual plot sizes are much smaller than a hectare the actual profits of these four farmers were in the range of US$220–470.

The situation in Nairobi is quite different. On average, incomes and profits per hectare are higher than in Kumasi, but plot sizes are much smaller. Furthermore, there is a clear trend in the levels of expenditure, income, and profit according to location.

Farmers to the east of Nairobi, at Thiboro operate on a commercial basis, albeit on very small plots, investing heavily in paid labour and other production inputs. Levels of revenue and profit reflect this investment with an average actual profit (from 0.126 ha) of US$607 (US$4,816/ha). At Mau Mau Bridge there is high

investment in production inputs, but no use of hired labour. The average revenue during the period of study was low due to a pest attack on a crop of green peppers. This clearly illustrates the relatively high-risk nature of irrigated vegetable production. Average actual profit was just US$86 or US$1,036/ha. The agricultural practices at Maili Saba are more subsistence in nature. Few exotic market vegetables are grown and very few inputs are purchased. Actual average profits were about US$70 (US$1,404/ ha) during the study period June–September 2000.

Trade-offs of urban wastewater irrigation

The case studies show that urban wastewater irrigation has a positive effect on the financial capital of the urban irrigators. However, wastewater irrigation potentially bears risks that may weaken the human, natural, and social assets of the irrigators and their families, making them more vulnerable to external shocks. Apart from the direct risk to health, water polluted with industrial effluents may also pollute soil and groundwater, thereby undermining the long-term sustainability of the natural resource base. An analysis of the risks would help to understand the actual trade-offs on the sustainability of the livelihoods of urban irrigators and their families: Do the benefits outweigh the risks and negative impacts of wastewater irrigation, and over what time frame should such benefits and costs be assessed? The recent increases in the numbers of urban dwellers engaging in urban wastewater irrigation in Nairobi and Kumasi indicate that in irrigators’ and family members’ own assessment the benefits outweigh the risks, at least in the short term.

Whatever the benefits may be for the irrigators, policy makers must safeguard the wider public interest. Although irrigation with untreated wastewater contributes substantially to the availability of fresh vegetables, and under controlled circumstances may be environmentally acceptable and a beneficial means of waste disposal, uncontrolled wastewater irrigation can lead to both chronic ill-health and more serious outbreaks of disease amongst irrigators and consumers. Policy makers and others working in this field need clearer guidance on the levels of risk associated with use of different qualities of untreated wastewater if they are to assess the trade-offs that exist between the costs and benefits. Some types of wastewater irrigation documented in these studies are probably unsustainable and may be regarded as unacceptable by most communities, when given information. However, in the absence of guidelines aimed specifically at the management of untreated wastewater irrigation it is difficult to make informed judgements about the costs, benefits, and trade-offs, associated with different practices.

The Dilemma

At present there are no microbiological irrigation water quality standards that acknowledge the concept of an acceptable level of health risk for irrigators and the wider community, other than zero risk. In the absence of other norms, the WHO microbiological quality guidelines for the design of wastewater treatment plants, where the effluent is intended to be used for irrigation, are used extensively to evaluate the health risks arising from the use of polluted water sources for irrigation (WHO, 1989). These guidelines are designed to ensure ‘no measurable excess risk’ of infection attributable to the use of wastewater as evaluated from epidemiological studies and risk assessment models. The guidelines prescribe that for unrestricted irrigation the faecal coliform (FC) count may not exceed 1,000/100 ml and that the helminth egg count should be below 1/l. FAO promotes the use of these guidelines to monitor the quality of water used to irrigate vegetables and other high-risk crops in the absence of other microbiological irrigation water quality standards (Westcot, 1997).

In adopting these guidelines for controlling the quality of water used for irrigation two anomalies emerge. Firstly, water for irrigation must meet a higher standard than that set by the British Government’s Statutory Instrument 1991 No. 1597 for coastal and freshwater bodies used for bathing (HMSO, 1991), which sets a limit of 2,000 FC/100 ml. Secondly, and more significantly, a high percentage of the world’s freshwater resources do not meet WHO water quality guidelines for unrestricted use, while in practice these waters are diverted for unrestricted irrigation. Data published by WHO (1989) show that 45% of 110 rivers tested around the world have FC levels of above 1,000/100 ml, while 15% have levels over 10,000/100 ml. In China 27% of the river sections monitored have a coliform count of more than 10,000/100 ml. It may be expected that near urban centres the water quality will be poor. Rapid urbanisation is putting further pressure on sanitation and treatment infrastructure that is already inadequate. In developing countries, where the majority of the large cities are located, the costs of necessary investments in water supply, sanitation and treatment facilities are far beyond those countries’ present economic potential (Niemczynowicz, 1996). In the foreseeable future, surface water quality close to urban centres is likely to deteriorate further rather than to improve, and irrigators will continue to use it. To insist that only treated wastewater be used for irrigation seems an unrealistic goal. What planners and technocrats urgently need is guidance on the levels of risk associated with the use of water whose quality falls below the ideal, ‘no risk’ threshold set in the present WHO guidelines.

Conclusions

A large number of urban and peri-urban irrigation farmers around Nairobi and Kumasi are using various forms of untreated wastewater for irrigated cropping under unregulated and informal arrangements. In general, both the numbers of irrigators and the volumes of untreated wastewater seem certain to increase in the short to medium term, as urban populations grow and investment in wastewater treatment infrastructure is constrained.

In these two case-study cities the wastewater used for irrigation displays a wide range of microbiological quality depending on location, dilution, and the effects of natural remediation. It is misleading to consider ‘wastewater irrigation’ as a single activity with uniform characteristics. The various pathways of wastewater acquisition, from source to field, must be identified and differentiation made between them.

Some forms of wastewater irrigation not only offer important financial gain to the growers, they may also represent a low-cost and beneficial means of using and ‘treating’ wastewater within acceptable and controllable levels of disease risk. However, so long as the focus remains on the management of formally treated wastewater and a policy of ‘no measurable excess risk to health’, guidance on what might constitute an acceptable risk, the risks associated with different types of practice, and the tools needed to make informed, pragmatic judgements remain lacking.

By using the WHO guidelines to make judgements over the safety of the use of wastewater, without taking the various ‘types’ of urban wastewater irrigation into consideration, policy makers and technocrats are driven towards inappropriate conclusions. There is inevitably a huge gap between a standard leading to ‘no measurable excess risk of infection attributed to the reuse of wastewater’ and the situation on the ground. Faced with such a gap, reactions are either to condemn urban irrigators as posing a major health risk to the community or to turn a blind eye because action seems impossible and ignorance is the preferred course. Neither approach is helpful and both are driven by the lack of appropriate standards, inappropriate use of the WHO microbiological quality guidelines for treated wastewater use in irrigation, and a failure to differentiate between different qualities of wastewater flows. In Nairobi, for example, after the publication of studies on informal irrigation in the peri-urban zone, and the wider emergence of ’urban agriculture’ as a planning issue, city authorities are now motivated to ban the practice without taking account of the various types of urban wastewater irrigation, and the range of water qualities which largely define the actual risks involved.

As explained by Hespanhol and Prost (1994) guidelines produced by the WHO are intended to provide guidance for making risk-management decisions related to the protection of public health based on current scientific research and epidemiological findings. They provide a common background from which national and regional standards can be derived. However, for the development of national or regional standards the economic, technical, social, cultural and political contexts need to be taken into consideration. Such an approach inherently incorporates a risk-benefit analysis. Shuval et al. (1997) describe a risk-assessment model that estimates the risk of infection associated with eating vegetables irrigated with wastewater of varying microbiological quality. The first step in applying risk-assessment approaches is the definition of an ‘acceptable’ risk of infection. Therefore, there is a need for explicit debate on the levels of risk that may be acceptable to producers and consumers of wastewater-irrigated crops and the costs and benefits that they bring with them. Pragmatic water quality standards based on such an approach that are pertinent only to the use of untreated wastewater, can better inform policy makers and technocrats as they seek to manage the real situation on the ground.

Acknowledgements

The wider research project on which this chapter is based was funded by the Infrastructure and Urban Development Department of the British Government’s Department for International Development (DFID). It was conducted by HR Wallingford Ltd in collaboration with the Kwame Nkrumah University of Science and Technology in Kumasi, Ghana, and the Smallholder Irrigation Scheme Development Organization (SISDO), in Nairobi, Kenya.

References

Cornish, G., Mensah, E. and Ghesquire. P. (1999) Water Quality and Peri-urban Irrigation. An Assessment of Surface Water Quality for Irrigation and its Implications for Human Health in the Peri-urban Zone of Kumasi, Ghana. Report OD/TN 95. HR Wallingford Ltd, Wallingford, UK, 44 pp.

Cornish, G.A. and Aidoo, J.B. (2000) Informal Irrigation in the Peri-urban Zone of Kumasi, Ghana. Findings from an Initial Questionnaire Survey. Report OD/TN 97. HR Wallingford Ltd, Wallingford, UK, 41 pp.

Cornish, G., Aidoo, J. and Ayamba, I. (2001) Informal Irrigation in the Peri-urban Zone of Kumasi, Ghana. An Analysis of Farmer Activity and Productivity. Report OD/TN 103. HR Wallingford Ltd, Wallingford, UK, 39 pp.

FAO (Food and Agriculture Organization of the United Nations). (1995) Irrigation in Africa in figures. FAO Water Reports no.7. Food and Agriculture Organization of the United Nations (FAO), Rome, 336 pp.

Hespanhol, I. and Prost, A.M.E. (1994) WHO guidelines and national standards for reuse and water quality. Water Resources 28(1), 119–124.

Hide, J.M. and Kimani, J. (2000) Informal Irrigation in the Peri-urban Zone of Nairobi, Kenya. Findings from an Initial Questionnaire Survey. Report OD/TN 98. HR Wallingford Ltd, Wallingford, UK, 26 pp.

Hide, J., Hide, C. and Kimani, J. (2001) Informal Irrigation in the Peri-urban Zone of Nairobi, Kenya: An Assessment of Surface Water Quality for Irrigation. Report OD/TN. HR Wallingford Ltd, Wallingford, UK, 105 pp.

HMSO (Her Majesty’s Stationery Office). (1991) The bathing waters (classification) regulations 1991. Statutory Instrument 1991, no.1597. www.hmso.gov.uk/si/si1991/Uksi_19911597_en_4.htm/ (accessed October 2002).

HR Wallingford. (2001) Informal peri-urban irrigated agriculture: opportunities and constraints. In: Proceedings of a workshop held at Kwame Nkrumah University of Science and Technology, Kumasi, Ghana, 7–9 March. HR Wallingford Ltd, Wallingford, UK, 158 pp.

Niemczynowicz, J.M. (1996) Megacities from a water perspective. Water International 21, 198–205.

Shuval, H., Lampart, Y. and Fattal, B. (1997) Development of a risk assessment approach for evaluating wastewater reuse standards for agriculture. Water Science Technology 35(11–12), 15–20.

Westcot, D. W. (1997) Quality control of wastewater for irrigated crop production. FAO Water Report no. 10. Food and Agriculture Organization of the United Nations (FAO), Rome, Italy, 86 pp.

WHO (World Health Organization). (1989) Cited in: Westcot, D. W. (1997) Quality control of wastewater for irrigated crop production. FAO Water Report no. 10. Food and Agriculture Organization of the United Nations (FAO), Rome, Italy, 86 pp.

WHO (World Health Organization). (1989) Health guidelines for the use of wastewater in agriculture and aquaculture: Report of a WHO Scientific Group. WHO Technical Report Series 778. World Health Organization, Geneva, Switzerland, 74 pp.

Document(s) 8 of 19