Soil and Water Lab  
Biological and Environmental Engineering  

Back to Brett's Page

Back to Mali Research Page


A Thesis Presented to the Faculty of the Graduate School Of Cornell University In Partial Fulfillment of the Requirements for the Degree of Master of Science

By Brett Gleitsmann
May 2005

© 2005 Brett Gleitsmann


Permanent access to safe and sustainable water sources is a major concern for much of rural and peri-urban sub-Saharan West Africa.  In response to this problem, many international, regional and local water supply development organizations are currently working to improve the level of access that the local populations have to safe and sustainable water sources.  The West Africa Water Initiative (WAWI) was launched in Ghana, Mali and Niger in 2002 to promote broader partnerships between the various organizations working in the water supply development sector in the region.  As part of this collaborative effort, the Cornell International Institute for Food, Agriculture and Development (CIIFAD), one of the founding partners of WAWI, has supported and funded the present study.  The ultimate goal of this study is to generate a useful knowledge base that can be accessed by WAWI partners and other actors in the Malian water supply development sector to improve the sustainability of rural water supply projects in the region.  To this effect, domestic water-use patterns, choice-of-technology preferences, sustainability perceptions, regional pump conditions and general hygiene practices were observed and recorded during a ten-month study in the Koro district of the Mopti region in Mali.

This research has contributed insight into the complex nature of the rural water supply situation in the Koro region of Mali.  Choice-of-technology preferences vary according to several factors including local perception of water scarcity, individual water use requirements, and previous experience with various technologies.  Sustainability of various types of water supply infrastructure is dependent upon the degree to which the technology corresponds to the needs of the local community and the community’s ability to maintain and repair it over time.  Considering the poor state of the manual pumps observed in the district of Koro, it is apparent that efforts need to be made to ameliorate the situation.  Learning from previous development projects, the latest approaches address the problems of the limited availability of spare parts, the absence of trained technicians at the local level and the limited role of women in the pump management scheme.  Dedicating more time and resources to the maintenance and management aspects of rural water supply development is a positive action and should help to improve sustainability of newly installed water supply infrastructure.  However, the continued lack of community involvement in the decision-making phase of water supply development projects, deeply entrenched patriarchal structures that continue to constrain leadership and decision making participation of women and minorities, principal stakeholders in water use, and the potentially global inappropriateness of manual pumps as a sustainable long-term option remain to be addressed.

link to page top


I would like to wholeheartedly thank the chiefs of the three villages of study, notably, Alaye Bamadio of Yadianga, Aly Djimdé of Ogodouroukoro and Ali Niangaly of Benebourou, for their willingness to host and take care of a stranger from America.  The research itself would not have been possible without the help of my designated hosts in the three villages: Bakary Bamadio of Yadianga, Baye Djimdé and Antoine Sagara of Ogodouroukoro and Atémélou Niangaly of Benebourou.  And as for the day to day meals, water, clean clothes and fresh milk, I would like to thank the mothers, sisters and wives of my host families for all of their hospitality, hard work and support.  For their constant presence in Koro and their wonderful hospitality every time that I came through, I would like to thank the Kassambara family of Koro II (Gogo, Adema, Fanta, Aicha, Oumou and Ma).  For their logistical support and guidance, I would like to thank the staff of World Vision Mali, in particular Jean-Baptiste Kamaté, Josué Sogoba, Pathé Ongoiba, and Samuel Diarra for all of their support and assistance during my stay in Mali.  I would like to equally recognize and thank the staff of Winrock International, in particular Niels Hanssens and Bara Kassambara, for their total commitment to my project and for all of their guidance and assistance during my stay.  For their support from the very beginning of this research I would like to thank my professors, Dr. Tammo Steenhuis and Dr. Margaret Kroma.  Lastly, but most importantly, the research itself would not have been possible without the funding of the Cornell International Institute for Food, Agriculture and Development (CIIFAD) and the direction and vision of Dr. Norman to page top





link to page top


The World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF), estimate that nearly 1.1 billion people lack access to improved water supplies and that about 2.4 billion people lack access to improved sanitation facilities, with the vast majority of these people living in the developing countries. To achieve the international development target of halving the percentage of people without access to improved water supply or sanitation by the year 2015, an additional 1.6 billion people will require access to water supply and about 2.2 billion will require access to sanitation facilities by 2015, given the projected population increases (Brikké and Bredero 2003).  The first concerted global effort to meet this target was the International Drinking Water Supply and Sanitation Decade (1981-90).  Evaluations of water supply development projects during the UN Water Decade showed that non-sustainability of water supply projects and facilities left the majority of the target populations without access to adequate water supplies or sanitation facilities (Rotival 1991, Livingstone and McPherson 1993, Diamant 1992).        

One global response to the non-sustainability of the UN Water Decade was to promote the evolution of participatory approaches that began to consider the local populations as ‘participants’ rather than ‘beneficiaries’ as was previously the case (Carter et al. 1993).  These participatory programs typically focused on transferring ownership, responsibility and management to the local level by creating village water committees, requiring the community to financially contribute to the project, involving women in the management scheme, training local technicians, ensuring local availability of spare parts, etc.  This step towards a more participatory, locally-managed approach has been effectively integrated into most current water supply development projects, yet sustainability rates of these projects continue to remain low.      

Evaluations of these programs typically attribute the continued non-sustainability to continued external control of development projects due to institutional rigidity and a reluctance to allow community participation in the initial planning and design phases (Mangin 1991, MacRae Jr. and Whittington 1988).  Studies have shown that sustainability of water supply projects improves when communities are allowed to take a central role during all stages of the project, including design and planning (Mangin 1991, Williams 1998, Bah 1992).  As long as choice-of-technology decisions are made by an outside agency, community demands cannot be met, even if such demands have been duly assessed (Narayan 1995).  Ideally, the role of implementing agencies during the planning and decision making phase should be to provide interested communities with all relevant information such as the short- and long-term costs/benefits/tradeoffs of several available alternatives.  The technology to be offered must show benefits in terms of the community values, not just water quality improvements that the donors typically stress, but also convenience, time savings, improved access, etc. (Carter et al. 1993, Kendie 1992, Mu et al. 1990).  An informed choice-of-technology and level of service decision can then be made by the community, consolidating their role as the primary partner in a project.  This local choice can then be supported by the implementing agency by creating the needed supporting infrastructure (e.g., hygiene education, pump repair training, well maintenance) for the community-chosen type of technology.  This type of substantial participation between the two involved partners has a higher potential of leading to a more sustainable water supply project (Narayan 1995, Anonymous 1993).      

Beyond improving the structural sustainability of the water supply infrastructure, is the more important issue of multiplying the impact that these improved facilities have on the general health of the local populations.  Research and experience continually confirm that a safe water supply is not sufficient and that adequate sanitation facilities and hygiene practice are essential to improving the health of the local population (Brikké and Bredero 2003, Gasana et al. 2002, VanDerslice and Briscoe 1995, Briscoe et al. 1986, Feachem et al. 1978).  Water quality improvements are additionally minimized by a deterioration of water quality from the source to the point of use (Trevett et al. 2004, Wright et al. 2004, Genthe et al. 1997).  Studies have also shown that ready access to water and the resulting increases in the quantity used for hygiene can have a greater impact on health than water quality improvements (Nyong and Kanaroglou 2001, Cairncross 2003).  In brief, water supply development projects need to extend their scope beyond simply the provision of sustainable water supply infrastructure.  To have the greatest beneficial on the health of the local population, it will require an integrated multidisciplinary approach that works in close collaboration with the local population.      

The research for the current study was undertaken with the aim of evaluating the applicability and relevance of the above concepts as applied to rural water development projects operating in the water-scarce Koro district of Mali.  More specifically, domestic water-use patterns, choice-of-technology preferences, sustainability perceptions, and general hygiene practice were observed and recorded during a ten-month study throughout various villages in the Koro district as part of the Cornell International Institute for Food, Agriculture and Development (CIIFAD) contribution to the West Africa Water Initiative (WAWI).  The principal implementing partner of WAWI-Mali is World Vision and their initial objective within the WAWI project zones of Mali is to drill 225 successful (wet) boreholes and to equip these with new India Mark 2 hand pumps.  To complement these infrastructural improvements, CIIFAD operates within the WAWI partnership as a knowledge-generating and capacity-building institution.  CIIFAD plans to (1) ensure ecologically sustainable management of water and the broader natural resource base of which water is an integral component; (2) investigate the sustainability of micro-irrigation at the village-level in relation to the efficiency and effectiveness of water use through micro-irrigation innovation in Mali; and (3) investigate and assess local institutional mechanisms, gender relationships and patterns of decision-making in the management of water resources in WAWI areas in Mali (CIIFAD, 2003).  To this effect, the current research examines the relevance and importance of adopting an integrated, multi-disciplinary approach to rural water development projects in the Koro district of Mali.  Funding for this research was provided by CIIFAD and the National Science Foundations.

link to page top


Mali is a landlocked nation covering an area of 1,240,000 square kilometers, the northern three-quarters of which lie within the Saharan and Sahelian zones.  The country is home to approximately 11 million residents, 72% of whom live in the rural sector (Toulmin et al. 2000).  Rain-fed subsistence farming of cereals (millet and sorghum), animal husbandry, and irrigated paddy rice cultivation along the major rivers remain the principal agricultural activities despite poor soil fertility and highly variable rainfall. The specific zone of interest for this study is the district of Koro (Cercle de Koro) which corresponds with the first intervention zone of the WAWI project (Figure 1).  This district is classified as semiarid with an annual precipitation of 500 – 600 mm per year, and daytime temperatures typically exceed 33° C throughout the year.  There are three main seasons: the rainy season (mid-June – September), the cooler Harmattan period (October – January), and the hot season (February – mid-June).          Koro’s water resources are severely limited as there are no permanent rivers or lakes in the district and groundwater is typically deep and difficult to locate.  During the rainy season, small surface water reservoirs fill with water and become the primary water resource for the rural villages of the district.  At the onset of the dry season, these surface reservoirs begin to dry up and groundwater becomes the only available source of water for the remaining seven to eight months of the year.  This groundwater is typically located in discontinuous, slowly recharging aquifers at 50 – 100 meters below the sandy surface layers of Arenosol type soils, and confined under a thick impermeable layer of rock (30 – 60 meters).  The aquifers are highly fragmented and subsurface conditions may vary drastically across relatively short horizontal distances.

Figure 1: Region of Study
(click here for a full-size image)      

The three villages of this study, Yadianga, Ogodouroukoro and Benebourou for example, are all located within 12 kilometers of Koro and yet they demonstrate very different groundwater availability characteristics.  The village of Yadianga (pop. 2,473 est. 2001) typically has yearly access to a superficial unconfined aquifer at ~28 meters, but the wells accessing this aquifer begin to dry up during the dry season.  The village of Ogodouroukoro (pop. 612 est. 2001) has no viable amount of water stored within the shallow superficial aquifer and their only groundwater source is the deeper confined aquifer (~75 meters).  As for the village of Benebourou (pop. 1,570 est. 2001), they benefit from a plentiful shallow aquifer (~5 meters).

link to page top


To determine the effect of groundwater depth on water-use behaviors, a comparative study of three villages having different depths to the groundwater (5, 25 and 75 meters) was conducted in the district of Koro.  The villages of study were chosen in close collaboration with World Vision-Koro and their field staff.  The main selection criteria for the three villages were, 1) the existence of at least one manual pump and at least one large-diameter well, 2)  differing groundwater depths (5, 25 and 75 meters), and 3) a village population of approximately 2,000 residents.  Problem identification was carried out through interactive sessions with the village councils, village women’s groups, village hygiene groups, local mayors, NGO staff, and the village residents.  Formally structured in-depth water resource surveys were conducted with the head woman of 10 households per village.  The interviews were prepared with the help of Winrock and World Vision field agents, as well as the village hosts. The households were chosen with the help of the village hosts to ensure a representative cross-section of the population.  Structured interviews were undertaken with the regional representatives of World Vision, CARE, the Direction Nationale de l’Hydraulique et de l’Energie du Mali (DNHE), the mayors of Koro and Bondo, and the 4 pump managers from Yadianga.  Informal, unstructured interviews of 10 – 20 men per village were conducted throughout the study.  Average groundwater extraction rates were recorded at a major well (and pump, if applicable) for a period of 24 hours.  Water-use behavior, pump management schemes, water storage techniques and hygiene practice were obtained through direct observation of activities at the water sources as well as the in-depth individual surveys with local water users and pump managers.  A regional pump inventory and assessment was conducted with the help of the local mayor’s offices and several village representatives in each village having a manual pump in the counties of Bondo and Koro during the month of June.  Borehole data for 16 of the 44 inventoried pumps was provided by the DNHE.  General demographic information and county water resource planning documents were provided by the local mayor’s offices of the counties of Bondo and Koro.  Water quality analyses were conducted under the supervision of the DNHE in Bamako.

link to page top


Observations of the use and management of water resources in the three villages of study were recorded during the dry season of 2004.  The following section will address the type of access to groundwater in the villages of study, the way in which the groundwater was drawn to the surface, water-use at the household level and for livestock and gardening purposes, local preferences for various water supply infrastructure, manual pump sustainability on a regional level, and hygiene and sanitation conditions in the villages of study.

Groundwater Access       

The villages of study were found to have two major types of access to groundwater: large-diameter wells and manual pumps.  Large-diameter wells were classified into three main categories: reinforced, modern or traditional.  Wells having both concrete linings to the depth of the confining rock stratum and also a concrete margin at the opening of the well were classified as ‘reinforced.’  Wells having only the concrete margin and no concrete lining were classified as ‘modern.’  Within these two classifications, most of the wells were equipped with a structurally solid crossbar, either metallic or wooden, where the men and women attach their pulleys to more easily hoist the water.  Wells that did not meet these conditions were classified as traditional.        

The manual pumps found in the villages of study included the India-Mali and UPM pumps.  Both pumps are positive displacement pumps (piston pumps) that are operated by pressing down on a single metal lever that is connected to a series of tubes that mechanically lift the water.  India-Mali pumps were reported to be easier to use than the UPM and are approximately 1 meter high, while UPM pumps, known throughout the region as the ‘giant green pumps,’ are about 3 meters tall and require the user to pull down a 2.5-meter lever from above their head to the ground.  Across the district, foot pumps (Vergnet) and various solar-, diesel-, and wind-powered pumps were found as well.

Drawing Water      

For domestic water use, the responsibility for supplying the household with a daily supply of water rests firmly upon the women in the three villages of study, but the labor is typically divided up amongst various parties.  In the villages with shallower groundwater (Yadianga and Benebourou), school-aged children (7-17 years of age) were found to supply much of the actual labor for drawing and transporting water.  In these villages, water from the large-diameter wells was always drawn by hand.  Although women still pulled water by hand in the village with deeper groundwater (Ogodouroukoro), the young men (20-30 years of age) and draught animals played a much more prominent role in drawing water.  Often men drawing water for their animals at the well would help the women by lending them their donkey, or by pulling the cord by hand and then leaving the transportation of the water to the women.  Women would remain the responsible party, but the men were much more willing to help the women than in the other villages of this study, possibly owing to the more laborious nature of the activity as groundwater level is deepest in this village compared to the other two villages of study. Water for livestock is a primary responsibility for men (10-40 years of age) and they would use either draught animals or their hands.

Water Use      

Severely water-stressed environments, like the district of Koro, obligate the local population to follow a fundamental order of water-use priority.  People will always be required to satisfy the first element of the list (human consumption) before continuing to the second and third levels.[1]  These priorities can be ranked as follows:

  1. Human Consumption
  2. Livestock Consumption
  3. Other Activities (gardening, hygiene, brick building, etc.)      

Adhering to this philosophy, it seems appropriate to begin this analysis from the top of the list. 

Domestic Use      

Groundwater use at the home was more or less constant for the three villages of study (28 L/c/d ± 6 L/c/d).  The unexpected finding was that average per capita domestic water use was found to be highest in Ogodouroukoro where the water is deepest.  During the rainy season, groundwater use was reported to decline for villages that continue to access the wells, and to completely cease in villages that abandon the wells in favor of the available surface water stored in the seasonal ponds.  During the rainy season, surface water was found to be an essential component of water-use activities in all of the villages.  It should be noted that the population of the village may increase by 30-50% during the rainy season due to the return of the young men to the village to work the fields, and so the recorded decline in per capita water-use is roughly offset by the increase in overall household size meaning that the overall water demand for domestic use remains relatively constant throughout the year.

Table 1. Daily Per Capita Domestic Water Use


Daily Per Capita Water Use (L/c/d)

Dry Season

Rainy Season


26 ± 6

16 ± 4


29 ± 7



28 ± 5

15 ± 5

link to page top


Figure 2 shows the results of the assessment of domestic water use by activity in the three villages of study.  It can be seen that bathing was the activity that consumed the most water.  Drinking, represented approximately 11% of the total water use and ranked as the fourth most important activity in respect to the quantity of water used.  The most significant variance in water use behavior by activity was found in the livestock component and can be related to water scarcity during the driest months of the year. 

Figure 2: Average Domestic Water Use by Activity, Dry and Rainy Seasons, 2004
(click here for full-size image)

link to page top

The source of groundwater that people were choosing to access in these three villages depended upon the individual household’s choice and the current state of the available hand pumps.  In the village of Yadianga, 24-hour surveys of one primary large-diameter well and the two functioning hand pumps, showed that 80% of the village’s water for domestic consumption was being assured from the large-diameter wells.  The remaining 20% was being pumped from the two India-Mali hand pumps (Table 2).  Similarly, a survey of the hand pump managers revealed that only 33 households (17% of the village) were currently contributing to the hand pump fund and therefore permitted to use the hand pumps.  Even at this low use (20% of domestic water needs), it was observed that the two pumps were being used nearly to capacity during approximately 12 hours per day.  It is interesting to note that by the end of the study, these two pumps were also broken and all of the water needs of the village were being met by the large-diameter wells.

Table 2. Daily Water Withdrawn (Domestic Use)
Yadianga, February 2004

Water Source

Domestic Water (L/day)

Percent of Total Domestic Demand (%)

Manual Pumps (2)



Large-diameter Wells (4)



link to page top 


Table 3. Daily Water Withdrawn (Total)
Yadianga, February 2004 

Water Source

Total Water


Percent of Total (%)

Pumps (2)



Large-diameter Wells (4)



link to page top

The hand pumps were limited by several factors.  The main reason stated for not using the pumps was a lack of economic resources to pay the initial contribution or the required maintenance costs.  Women frequently stated that they prefer to use the pumps but lack the funds to do so.  In the case of Yadianga, frequent breakdowns of the pumps have led many men and women to believe that it is not worth the continued investment.  Another primary reason for the limited use of hand pumps in Yadianga was that the flow rates from the hand pumps could not meet the demands of the entire population even if the residents were to use the pumps for 24 hours a day.  The maximum pumping rate from the India-Mali hand pumps was found to 14.5 L/min which is above the WHO minimum acceptable limit of 13 L/min (Brikké 2003), but this was only true of one of the hand pumps, whereas the other pump had an observed maximum rate of 9.3 L/min.  In comparison, the maximum withdrawal rate from the large-diameter wells was observed to be 66.3 L/min.  The inadequate pumping rates from the hand pumps necessarily leads to the limited use of the hand pumps and the continued reliance upon large-diameter wells to meet domestic daily water requirements.  Other reported reasons for not using the pumps include a preference for the taste of well water, the belief that it was good luck to drink the well water, and that the distance to the pump was too great.  Thus it can be seen that cultural factors as well as practical constraints are also important determinants of water use behaviors in the communities studied.  

Livestock Use      

The Gondo-Seno plain of Mali supports one of West Africa’s most numerous populations of livestock (mainly cattle, sheep, goats, camels and donkeys).  During the rainy season, the transhumant Fulani herders move the herds out of the village as they can rely exclusively on surface water to water their herds and to meet their own personal needs as well.  The livestock which stay year-round in the village (small herds of sheep and goats as well as the draught cattle), are watered at nearby ponds during the rainy season.  By the end of the rainy season however, these ponds begin to dry up, surface water becomes harder to find and the transhumant herders begin to return to the outskirts of the villages in search of water for their herds.  For the next seven to eight months, the drinking requirements of the livestock will be met exclusively by groundwater.  In the villages of study, water for the livestock was obtained exclusively from the large-diameter wells.      

Typical per head estimates for daily drinking water requirements in the Sahel are 27 liters for cattle, 5 liters for sheep and goats, 16 liters for donkeys and 50 liters for camels (FAO 1986).  Using these estimates in a village like Yadianga, which supports a population of approximately 900 cattle, 3000 sheep and goats, 250 donkeys and 25 camels[2], it is predicted that the village will need to provide approximately 44,550 liters of water per day to satisfy the needs of the livestock.  Observations at the large-diameter wells confirmed these estimates and showed that livestock used nearly as much water per day as the human residents (Table 4).

Table 4. Daily Well Water WithdrawnYadianga, February 2004

Water Use

Water Withdrawn (L/day)

Percent of Total Water Demand








link to page top



One of the major issues raised by the village councils and residents of both Ogodouroukoro and Yadianga was that the livestock not only exert significant pressure upon the supply of available water, but they also adversely affect the quality of the water.  The livestock come to the wells once or twice a day to drink from a reservoir adjacent to the well, usually a large bucket or half of a 200-liter barrel, placed not more than 5 meters distant from the well.  Many wells are equipped with adjacent concrete reservoirs for the livestock, but due to design flaws and their inconvenience to the herders, none of these were found to be in use.  As the cattle wait for their turn to drink, they meander about near the well, kicking up sand and defecating.  Some of this sand and manure is inadvertently swept up by the wind or the ropes being used to pull water, and a portion ends up in the bottom of the well.  The manure left by the cattle does have some positive implications for the village as the majority of it is collected by the residents for use on their fields as fertilizer, but the residents stated that it was not good to have the cattle so close to a well which is also used to provide drinking water for the human population.


In the district of Koro, small-scale gardening (shallots, garlic, tomatoes, peppers, potatoes, lettuce, cabbage, etc.) is a principal economic activity during the dry season (November – May).  The most extensive gardens (10 – 15 hectares) were found in Benebourou, where the gardeners have access to the shallow groundwater (~5 meters) throughout the entire dry season.  As the water table becomes deeper throughout the district, the extent and duration of the gardens become significantly smaller.  In Yadianga the total area occupied by gardens was reported to be approximately 1 hectare, even though the village has 35% more residents than Benebourou, and no gardens were found in the month of February due to the scarcity of water at that point of the dry season.  No gardens were found in Ogodouroukoro and the residents expressed that they had no intention of trying to garden.  No pump water was found to be used for gardening purposes in the district.

 Water Supply Infrastructure Preferences      

Perspectives on the type of water supply infrastructure currently in-use and what is needed for the sustainable development of the Koro district were gathered via formal and informal surveys from village members, the NGO community, local governments and an entrepreneur specializing in reinforced wells.

Community members      

Preferred types of water supply infrastructure varied along gender lines and water table depths, and also according to previous experience with various technologies.  In Benebourou, the village with the easiest access to groundwater, an overwhelming majority (94%) of the women indicated that they would choose a hand pump.  This reflects the fact that water scarcity is not an issue in the village and that there exists a sufficient number of large-diameter wells.   As the water table decreases, preferences became more balanced, and the women of Yadianga chose hand pumps to wells by a 73% to 27% margin.  The increased percentage of women preferring wells was attributed to concerns over pump costs and the lack of sustainability of the pumps.  Many women in Yadianga also mentioned that they had seen solar pumps and wind pumps in the surrounding villages and that for Yadianga to truly advance, and for women’s suffering to decrease, the village needed to contribute to a larger project that would bring in a solar- or wind-powered pump.  In the village with the deepest water, Ogodouroukoro, only 45% of the women preferred pumps over wells and those women who chose pumps qualified their statements by stating that they would only want a pump if it was easier to use than the UPM pump currently in the village.  Considering the incredible hardship that the women endure in drawing water from such depths, it was not surprising that the women of this village were the most adamant in demanding a solar-, wind-, or diesel-powered pump.         

Whereas the women based their decisions mainly upon ease of access, the men typically answered based upon the potential costs of maintaining the water access point and also by looking at other issues not entirely related to the provision of household water.  In the village with the easiest access to water, Benebourou, the men spoke of constructing several large-diameter wells at the eastern edge of the village so that residents of neighboring Bondo could come to draw their water there.  They also stated that the village had a plenitude of water access points and that the village needs a large pond for pisciculture.  In the village of Yadianga, the men were frustrated with the performance and sustainability of the hand pumps and most men stated that they would not be interested in a hand pump but that they would prefer a fifth large-diameter well that could help meet the demands of the village’s livestock.  They felt that large-diameter wells were a more appropriate long-term solution and that it was necessary to have a sufficient number of wells before beginning to put their resources into repairing pumps.  The village council made it clear that they would prefer to invest in the construction of a water tower that would provide running water to several village standposts.  The council also expressed interest in constructing a large pond (2 hectares) at the edge of the village for gardening purposes in an effort to improve the food security of the village.         In the village with the deepest water, Ogodouroukoro, the village council expressed that they would not be interested in another hand pump, because the UPM had proved to be far too difficult for the women, and they were not confident that a new hand pump could perform any better than the UPM.  Some members claimed that the village still needed at least one more large-diameter well in order to meet the livestock demands and also to provide the women with another alternative source during the driest months, while others were interested in improving the three existing large-diameter wells and would only want to invest in a new large-diameter well that would not dry up during the months of May and June.  As for their vision of a true solution to the problem, there was a general consensus, in agreement with the women’s, that a powered pump (diesel or solar) would be the only real option considering the depth of the water.

Local Government Perspectives      

Analysis of the three-year development plans for the rural sectors of the counties of Bondo and Koro in regards to water resources, found that the top five priorities and budget allocations were as follows

  1. Construction of large-diameter wells
  2. Installation of village-wide water distribution systems
  3. Installation of solar pumps
  4. Reinforcement and rehabilitation of existing large-diameter wells
  5. Construction of large ponds (mares) to meet the demands of livestock

These priorities were developed through detailed multi-day evaluations with the various villages of the counties and these views were meant to reflect the needs and priorities of the villages themselves.  The most notable point to make about this list is the absence of any mention of hand pumps.  When the mayor of Bondo was asked why hand pumps were not a priority and yet the village of Bondo had recently contributed money for the installation of two hand pumps as part of WAWI, his response was that “our priorities have changed” and that in a water-scarce village like Bondo, the chief cannot refuse any type of assistance that is offered.        

In relation to the second priority, village-wide distribution systems, the mayor of Koro highlighted the fact that the national agenda regarding water supply development states that any village with more than 3,000 residents should have a village-wide distribution system in place.  This, he said, was his goal as well, but that financial constraints were making the execution of this goal impossible.  Nevertheless, he felt that when dealing with the larger villages of the county such as Yadianga, it would be more desirable to introduce a water tower and general distribution system than to continue to pour resources into the temporary solution offered by hand pumps.  The village council of Yadianga agreed with this and expressed the need for a large-scale solution to their water supply to page top


Non-Governmental Organizations      

Interviews in 2003 and 2004 with various regional directors and field agents of both CARE International (CARE) and World Vision revealed that there is a difference between the organizations’ official views on rural water supply and the staff’s personal views.  The regional director for CARE who has been working in the water development sector for over 20 years, expressed his belief that financial realities in rural Mali make large-diameter wells the most appropriate solution.  He stated that he sees a significant problem between the money-lending institutions, the implementing agencies (typically NGOs), and the villages that are to be served.  The money-lenders are often capable of obligating an NGO to promote and install a certain type of technology which leads to a situation in which an NGO enters into a village to provide something that the money-lenders have assumed to be desirable for the local population, but which is not.  However, due to the extremely water scarce conditions in this region, the villagers have no real choice except to accept the NGO’s proposed solution.  Eventually, the village ends up with the responsibility for managing an intervention they did not necessarily want in the first place and the village’s own resources are exhausted in the process.        

The regional director for World Vision has also been involved in the water supply sector for many years and explained that the current water supply development approach is aimed at providing potable water to the rural communities and that WAWI is a positive initial step in meeting the lasting needs of the rural populations.  The installation of hand pumps in these villages, along with the concurrent creation of pump maintenance groups, hygiene groups and the initial financial contribution of the village, are essential steps in leading to the long-term success of the project.  However, the director also pointed out that, in his view, long-term water accessibility should be complemented with the construction of large-diameter wells to ensure a low-maintenance option in case of financial or logistical difficulties related to the hand pumps.  The reality in the field is that the potability of pump water, while desirable by all parties, takes a back seat to the simple existence of an adequate quantity of water which can more effectively be provided via large-diameter wells.  This view was widely held among the field staff of World Vision as well.  A senior hygiene and sanitation agent for World Vision summed up the situation as follows:  the villages of this district face extremely difficult conditions of water scarcity.  The most important thing that needs to be provided for these villages is an adequate supply of water.  If there is not enough water, it is unreasonable to expect the villagers to be able to adequately comply with even the most basic hygienic practice in regards to the domestic drinking water supply.  With the installation of hand pumps, the supply is almost never sufficient, and as a field agent it is incredibly difficult to influence the hygiene behavior of the villagers when they have such limited supplies of water. However, if the village is given a large-diameter well, the supply is typically much greater, and this makes the villagers much more likely and capable of complying with the recommended hygiene practices.  That is to say that it is much more effective to teach somebody how to be hygienic with an adequate supply of well water than a limited supply of pump water.

Entrepreneur (Large-Diameter Well Specialist)      

A private entrepreneur based in Koro, Abdoulaye Beloum, who had previously worked for 17 years with the national large-diameter well initiative (Opération Puits), continues to be the primary specialist in his field for the entire Koro region.  He has worked in wells that are over 100 years old and he believes that if a well is properly constructed, it should last for at least 100 years.  His experience has shown that the water quality from large-diameter wells can be extremely good if they are lined with concrete cylinders all the way to the bottom and the upper margin is adequately constructed (reinforced well).  He also stated that growing up in the region, has led him to understand that the essential need of the region is an adequate and plentiful supply of water.  Considering the low-maintenance requirements of large-diameter wells, these are, in his opinion, the best way to ensure an adequate and permanent supply of groundwater.

Water Supply Development Approach       

The current approach being used in the Koro district by World Vision in the WAWI project begins by contacting the local mayor’s office to present their objective of improving water access conditions in the region.  World Vision asks for a list of the villages that are in most need of assistance and uses this list as a guide to decide where to install pumps.  The final decisions are typically based upon three general criteria:

  1. the absence of a pump in the village,
  2. village population, and
  3. in the case of an already existing pump, the degree of successful pump management.

Once the list is finalized by World Vision, the villages are contacted and asked if they are willing to contribute 100,000 FCFA to begin the process of drilling a borehole and installing an India-Mali Mark II hand pump in their village.  No other water supply infrastructure alternatives are offered to the villages, and every village receives an India-Mark II hand pump regardless of the specific groundwater conditions of the village.  Pump committees are established with a focus on women’s participation, local mechanics are trained to perform routine maintenance on the pumps, spare parts are made available through the World Vision office, and a hygiene group is established.

Sustainability of Manual Pumps       link to page top


A survey of all 28 villages with manual pumps in the counties of Koro and Bondo was conducted in the month of June, corresponding with the latter stages of the dry season.  Data was gathered regarding the current condition of the existing 39 manual pumps and 5 more advanced pumping systems.  Background data on the boreholes was obtained through the DNHE.  Pumps that were being used on a daily basis were considered to be ‘in-use,’ regardless of the flow rates from the pumps.  For the two counties, 41% of the manual pumps were found to be in-use (Table 5).

Table 5.  Total Number of Manual Pumps In-Use for
the Counties of Koro and Bondo, June 2004

Pump Type

Success Rate

Success Rate (%)













link to page top



It should be noted that many of the pumps were being used under very difficult circumstances.  The two UPM pumps in Dangatene for instance, required four people working together to pull down the pumping lever and the India-Mali pumps in this village were barely functioning (4-5 L/min).  The minimum acceptable flow rate established by the World Health Organization’s guidelines for rural water supply sources is 13 L/min.  Applying this standard to the pumps of the two counties, it can be seen that only 10% of the manual pumps were operating above the WHO guidelines (Table 6).  Further, if the Vergnet foot pumps are not considered, this figure dropped to 3%. 

Table 6.  Manual Pumps Meeting WHO Guidelines
(Flow Rate > 13 l / min), June 2004

Pump Type

Success Rate

Success Rate (%)













link to page top



The DNHE and local residents presented several reasons for the low success rates.  Absence of spare parts and specialized labor, poor pump management, and a critical lack of basic information on behalf of the users were the primary reasons given for the failures.  In addition, many of these pumps were installed as part of region-wide experimentation efforts that used materials that turned out to be poorly suited to the environment leaving the users therefore unfairly overburdened with poorly adapted pumps.  Across the two counties, the pumps were typically operating at static water levels of an average depth of 37 meters, meaning that the pumps were operating at depths near or beyond their design capacity (45 meters, maximum design depth for India-Mali and Vergnet pumps).  Under these conditions, pumping forces are near the maximum recommended by the pump manufacturers and for the pump to properly function it would need to be in near perfect condition.  It was reported that none of the pumps had ever been serviced other than when there is a breakdown.  This implies that no routine maintenance was being performed on the pumps, and consequently the probability of breakdown was high.  Local NGO staff, including hydro-geologic experts, elaborated on this difficulty.  They explained that hand pumps were not designed to operate at the groundwater depths of this district and that even with proper maintenance, the pumps were being stretched beyond their capacities and pump failure was likely.        

Another contributing factor to the low success rate was reported to be poor management of the pumps at the village level.  Discussions with Yadianga’s women’s group and village hygiene group showed that many women of the village were frustrated with the current management of the hand pumps and there was a consensus among the women that they could do a better job of managing the pumps than the men.  In Yadianga, the current pump managers were older men who had no previous pump management experience, three of the four did not even know the name of the pump (India-Mali) and they had been chosen based upon the fact that they lived in the general vicinity of the pump and that they were old enough not leave the village throughout the year.  Not only were the women not involved in the official management of the pumps, but they were also dependent upon their husbands to provide funds for the pump.  The women of Yadianga stated that if they had more financial independence, the pumps would be fixed much sooner.  They also stressed the fact that the men of the village have not placed enough importance on the value of the hand pumps and this has led to a situation in which the village’s financial resources have been invested in sectors other than water supply.  If the women were to have more financial responsibility, they claim that more money would be spent on the water supply sector, and therefore the actual financial capacity of the village to deal with water supply issues would be higher than under the current management scheme.  This raises the issue of the gender dimension of sustainable pump management, as well as addressing the fact that within the existing patriarchal and ethnically segregated hierarchal system in the villages of the Koro district, management decisions are typically left to the elderly men, autochthonous to the village.       

Commenting on the issue of local economic capacity, Ibrahim Togola of the Mali Folkecenter in Bamako, who has spent over ten years working on water supply issues across the Sahelian areas of West Africa, spoke of the importance of correlating the type of pumping infrastructure with the financial capacity of the village.  One way to ensure the financial sustainability of a water supply development project is to fix a price on the pumped water to begin to get away from the idea that water is a free resource.  Although the current study did not rigorously address this issue, all parties of the water supply sector have constantly discussed this issue and its importance seems to be indisputable. 

Hygiene and Sanitation      

Household drinking water was always kept in a communal water jar, typically made of clay or occasionally plastic.  The women were typically the managers of these jars and they were responsible for keeping them clean and full.  Some smaller families were able to keep very small water jars (~15 liters); while the larger families would often have much larger water jars (~75 liters).  Typically, water was served from the jar by a plastic or metal drinking cup.  Household members and visitors would typically use the same cup for drinking, which practically leads to continual contamination of the water in the jar.                  

Water that was going to be used for cooking, bathing, washing dishes, or household activities other than drinking was typically stored in a very large, uncovered water jar.  These jars were typically not nearly as clean as the jars used for drinking purposes and algal growth, small worms, and large debris was commonly found in these jars.  Residents claimed that the water from these jars was never used for direct consumption, but men, women and children were all observed to drink water from the jars at some point during this study.      

The issue of proper water storage was something that both World Vision and the regional government hygiene staff were interested in promoting.  They both were conducting village-level training programs stressing the importance of covered water jars and point-of-use treatment using bleach.  In order to assess the impact of these programs, a survey of water jars in the three villages of study was undertaken.  The results found a significantly higher proportion of covered water jars in the villages where recent training sessions had been given (Benebourou, 84% and Yadianga, 45%).  In Ogodouroukoro, where no training had been given, only 22% of the water jars were covered.

Table 7.  Water Jar Survey Results
Dry Season, 2004

Type / Condition of Water Jar

Water Jars (% of Total)




Small-Mouth / Covered




Small-Mouth /  Not Covered




Large-Mouth / Not Covered




link to page top


Point of Use Water Treatment      

During the formal and informal interviews, two methods of household water treatment were found to be in practice in the region of study: filtering and bleaching.  The women claimed that due to the prevalence of Guinea worm during the rainy season, they filtered the pond water directly at the ponds before carrying the water home.  At the home, women reported that they filtered their water, but this was rarely observed.  If the household had a filter it was typically torn or broken and most women claimed to be waiting for the government to bring them new filters, rather than trying to repair the filters.   Local hygiene groups mentioned that they were interested in encouraging filtration with clean fabric to replace the typically torn original strainers and they asked for additional support of the NGO community to assist them in conducting pond water treatment training seminars in their villages.  Recent evidence of one new case of Guinea worm found in a village next to Ogodouroukoro for this rainy season, August 2004, demonstrates that Guinea worm is still a problem in the region and that fighting for its eradication should be continued.      

The second method of treating household water was direct disinfection by using a small quantity of bleach.  This technique was being promoting by World Vision and the District Hygiene Department of Koro.  Most women were convinced that this was an adequate way of treating their household water, yet very few women actually claimed to be using bleach.  The outside agencies had given away small bottles of bleach to the villages in an effort to promote point-of-use disinfection.  These bottles were used until they were empty and then in most cases, bleaching was abandoned.  The women complained that bleach was not locally available and that they could not often make the trip to Koro where they could buy it. 

link to page top


The observations from this 10-month study and a review of the current literature show that there are several essential factors that need to be duly considered to ensure a sustainable and effective water supply development project.  The main factor contributing to project sustainability is community involvement during all phases of the project, especially the planning and design phases.  Community involvement imparts ownership of the project to the community, reduces the impact of external influence on choice-of-technology decisions, and requires that spare parts and technical expertise be locally available.  To ensure the effectiveness of the project, not only does the project need to be sustainable, but aspects of hygiene and sanitation need to be equally considered to ensure that the health impacts of the improved water sources are as great as possible.  The following discussion will address these issues as related to the Koro district of Mali.Water Supply Development Approach      

Due to the historically poor sustainability of manual pumps in the region of study and in the greater region as a whole, local and international efforts have been dedicated to finding a better and more sustainable approach to water supply development.  The current approaches to water supply development in the region are attempting to adopt a more participatory approach to target the appropriate villages and to ensure the continued maintenance and sustainability of the water supply infrastructure.  This approach is a tremendous improvement upon previous projects, as women are being included in the pump management committees, spare parts and technical expertise are being assured at the local level, and the importance of hygiene is also being recognized.  However, certain aspects of this approach can still be improved upon.        

Currently, the local populations are completely excluded from the choice-of-technology decision and this can be seen as detrimental to the potential sustainability of the project.  In water scarce regions like the district of Koro, villages are highly unlikely to refuse any water supply project that attempts to bring assistance, regardless of the type of technology offered and how this fits with their own development philosophy.  This was clearly the case in Bondo, where the community had outlined their development objectives and manual pumps did not make the list.  Yet when the village was offered the manual pumps, due to a severe lack of water in the village they were in effect obligated to accept them.  The willingness of local villages to put forth initial financial contributions for the offered type of technology does not necessarily reflect the village’s preference for such a technology, but simply affirms the fact that the village is in need of increased access to its groundwater resources.      

An additional concern with the current development approach that may lead to non-sustainability issues is that implementing agencies stock the spare parts for the pumps, rather than the local spare parts suppliers that have been trained and designated by the regional governments.  This type of arrangement seems to be common in the international NGO field, as local government agencies are often times resource-lacking and therefore judged as incompetent.  By supplanting the local government agency, however, the NGO relegates the government agency to an inferior position, further reinforcing the concept that local government agencies are inefficient and lack credibility.  This serves to undercut the existing state agencies (Carter et al. 1999, Akouoko-Asibey 1997).  The NGO becomes the primary point of contact for the local population and the government is viewed as a secondary option.  In the potential case where the implementing agency is no longer operating in the region, as was recently the case with the withdrawal of World Vision from the Gao region, or the loss of funding for a CARE well project in the Koro district, a lack of spare parts situation could emerge.  Considering that government will always exist regardless of the presence of NGOs, it is important to improve the capacity of the local government institutions to more effectively meet the needs of the local populations rather than to fill the existing gap with the resources of often transitory NGOs.

Choice-of-Technology Decisions and Tradeoffs       

The current study has further demonstrated that the type of technology considered appropriate by the local population is highly varied and depends upon such factors as groundwater depth, perceived water scarcity in the village, gender of the user, convenience, village size, livestock population sizes, previous experience with various technologies, and the economic capacity of the users.  In general, villages with deep groundwater are typically more concerned about water scarcity issues than villages with shallow groundwater.  In the villages of Ogodouroukoro and Yadianga, domestic water supply was the top development priority of the village, whereas in Benebourou it was given a much lower priority.  While the priority of water supply development differs for various villages throughout the region, the in-depth water resource surveys showed that the readily available quantity of water was considered to be more important than the source water quality.  This suggests a general preference for water supply technologies that provide a sufficient supply of water, typically associated with large-diameter wells.  This view was expressed and enumerated in the development plans of several village councils, local government agencies, and was favored by NGO staff and many of the village residents.  However, on the other hand, the women of the region, who are responsible for assuring that the domestic water needs of the villages are met, were primarily concerned with reducing the work required to draw water on a daily basis.  For this reason, most of the women of the three study villages were in favor of the installation of solar pumps or other highly technical solutions that would make fetching water more convenient.  Given these differing views within the villages, how can the villages and development agencies reach a consensus on how to best provide easy access to a sufficient supply of water on a sustainable basis?      

Considering that the differences in opinion are evident from the original choice-of-technology preference through to the management strategy of the resource, community involvement should necessarily exist from the outset of the project.  Recent studies have shown that sustainability of water supply projects improves when communities are allowed to take a central role during all stages of the project, including design and planning (Mangin 1991, Williams 1998, Bah 1992).  In the case of previous projects (e.g., FED7), the observed failure of the pumps can in part be attributed to the project’s failure to involve the communities in the design, planning and management phases.  In the case of the FED7 project, although it was widely known by the DNHE and the pump suppliers in the region that the UPM pumps require regular replacement of the pistons, none of the surveyed villages with a UPM pump were aware of this.  If the villages are unaware of the expected future requirements of the water supply infrastructure, it cannot be expected that they will be able to maintain them.  This transfer of information and training regarding the technical and financial requirements of the various technological options rests with the implementing agencies (both governmental and non-governmental).  A transparent and total transfer of information is necessary to support a truly community-based decision making process.        

During the planning phase, an exhaustive list of technological options, highlighting the tradeoffs of the different technologies, should be created in a collaborative manner with both the village and the implementing agency.  Technical information can be provided by the implementing agencies that have the access to this type of knowledge.  An example of the manual pump component of such a list, comparing the two most common manual pumps found in the region, was compiled by the Direction Régionale de l’Hydraulique et de l’Energie – Mopti in 2003 (Table 8).      

In addition to the list of tradeoffs, water supply infrastructure in neighboring villages can be assessed and evaluated by all of the stakeholders in the current project.  Once the village is provided with enough information regarding the various tradeoffs for all of the possible water supply options (from large-diameter wells to solar pumps), the village can make an informed and appropriate decision.  Using this type of approach, the responsibility of the decision is transferred to the village, which allows the implementing agency to play a supporting role in helping the village to realize their own objectives.  A supporting role for the implementing agency is preferable to the common scenario in which they have chosen and installed their choice of water supply technology and their role becomes convincing the village to take care of something that was essentially imposed upon them. 

Table 8.  Manual Pump Comparisons
DRHE – Mopti, 2003


Pump Type

India-Mali Mark II


Pump Mechanism



Type of Pump

Hand pump

Foot pump

Maximum Design Depth

>100 meters

>100 meters

Pump Durability

Very robust

Less robust

Ease of Maintenance

Difficult (trained technician)

Easy (pump users)

Annual Maintenance Cost

~35,000 FCFA

~25,000 FCFA

Replacement Parts

Replace used pump parts once per year

Replace used pumps parts once every 1-2 years

Flow Rate (peak)



Initial/Replacement Cost

2,500,000 FCFA

1,100,000 FCFA

link to page top



By offering options it is more likely that the chosen type of water supply technology will show benefits in terms of the community values, not just water quality improvements that the donor agencies typically stress, but also convenience, time savings, improved access, etc. (Carter et al. 1993, Kendie 1992, Mu et al. 1990).  This type of approach requires a high degree of institutional flexibility on behalf of the implementing and donor agencies in order that they remain capable of responding adeptly to the diverse needs of their client villages.  In the case of the World Vision approach in the WAWI project, the choice-of-technology decisions are predetermined by the donor agency and therefore solely reflect the preferences of the implementing agency, inherently ignoring the actual village preferences.  As long as choice-of-technology decisions are predetermined and influenced in such a way by the implementing agency, the community demands cannot be adequately met (Narayan 1995).  If large-diameter wells, for example, are not considered to be an appropriate water supply technology and therefore removed by the implementing or donor agencies from the list of possible options, community involvement in choice-of-technology decisions will be marginalized and subsequent non-sustainability issues may result.

Manual Pumps as a Centerpiece of Rural Water Supply Development      

Considering the historically poor performance of manual pumps in the region of study, it may be worth re-examining the appropriateness of manual pumps in general.  Village size may prove to be the single most important parameter to consider when determining the appropriateness of manual pumps.  By using the maximum observed pumping rate in Yadianga (14.5 liters/minute) and limiting pump use to daylight hours, it can be found that it would be necessary to have eight (8) properly functioning pumps to adequately serve the human population of Yadianga (pop. 2,473, est. 2001).  This estimate ignores the livestock requirement and considering that the estimated population of Yadianga in 2004 was 3,200 residents, the minimum number of manual pumps required to meet the human needs would increase to ten (10) and would subsequently increase incrementally with every ~300 new residents.  In this case, it can be seen that manual pumps may not be a feasible alternative.  Manual pumps may have their place in meeting the needs of a small community (<500 residents) or in meeting the immediate short-term demands of a larger community during a time of crisis.  However, manual pumps may not be the most appropriate choice of technology to put forth as a focal piece of sustainable rural development, but rather should be seen as one among a range of strategic options based on technological, ecological and social fit. 

Health: Water Availability and Water Quality      

The latest development philosophy for safe and potable water sources, such as boreholes equipped with manual pumps, presupposes that water quality is more important than water quantity.  This view seems to be in opposition to that of a large proportion of the population in the region of study, as well as to much of the recent research (Nyong and Kanaroglou 2001,Cairncross 2003) which has shown that improvements in water quality alone are not sufficient, and that ready access to water and the resulting increases in the quantity used for hygiene can have a greater impact on health than water quality improvements.  Village councils, local mayors, World Vision staff and community members (both men and women) in the villages of study indicated that if they had access to more water, rather than cleaner water, then they would be able to practice better hygiene than is possible in their current limited water supply situation.  In a region like the district of Koro, which is extremely water scarce, the simple provision of an adequate supply of daily water is a struggle for much of the population.  Considering the water scarcity situation and the established notion that an adequate water supply is more beneficial to the health of the communities than just the provision of a safe and potable water supply, it seems reasonable that meeting this initial deficiency in water supply should be the first priority.  Once a permanent, sustainable source of water is provided to meet the local demands, efforts can be realistically focused on source water quality improvements.        

Until the communities are at a stage where they can invest in source water quality improvements, it is in their interest to keep the available water as potable as possible.  One technique being promoted by the NGO and local government agencies is the use of bleach in the water jars as a point-of-use disinfectant.  The success of these efforts will, in large part, be determined by the hygiene practices of the communities, as the effectiveness of bleach is not guaranteed.  Evidence suggests that disinfection with chlorine or bleach is only partially effective in water having turbidity values greater than 30 NTU (WHO 2004).  Most well water and pump water exceed this turbidity criterion, and the effectiveness of using bleach in such waters could be much less effective than desired.  In highly turbid water, disinfection cannot replace filtration and relying solely upon point-of-use bleaching may not be effective.  The introduction of bleach as a disinfectant may (in the absence of community education programs) in fact have the unintended and undesirable effect of convincing many residents that hygiene is no longer important, because the water has been disinfected by bleach and that re-contamination is no longer possible.

Hygiene and Sanitation      

Hygiene improvement and the construction of adequate sanitation facilities have been shown to play a more essential role in improving the health of the local population than source water quality improvements alone (Brikké and Bredero 2003, Gasana et al. 2002, Van Derslice and Briscoe 1995).  Most diarrhea is infectious in origin and spreads from person to person, especially when personal hygiene is poor (Gasana et al. 2002).  Considering that benefits from water quality improvements at the source are minimized by a deterioration of water quality from the source to the point of use (Trevett et al. 2004, Wright et al. 2004, Genthe et al. 1997), it is important to look at current attempts to improve local water handling and storage methods.  This study looked at the local efforts to encourage village residents to cover their household water jars.  Success was much more frequent in villages that had been given training and education on the topic than in those villages that had not received similar training.  However, studies have shown that covering the water jar does not appear to have any effect on the bacteriological water quality (Trevett et al. 2004, Lindskog and Lindskog 1988).  The more important factor may be the amount of contamination that results from the continual dipping of the drinking cup and some part of the hand into the water jar.  Several studies have found that water quality is better from containers designed to prevent hand-water contact, as in the case of a water jar fitted with a tap (Roberts et al. 2001, VanDerslice and Briscoe 1993, Empereur-Bissonnet et al. 1992).  Given the inevitable hand-water contact that exists in every community in the Koro district, hand washing can be seen as a crucial practice that is needed to maintain clean water in the jars.  This will require community education efforts and it should be noted that, as World Vision staff pointed out, these efforts can only be successful if the most basic condition of a permanent and adequate supply of water is being met in the village.  Efforts to improve hand washing and other hygiene practices may be greatly compromised if this most basic condition is not met.  link to page top


Long-term Sustainability of Groundwater Resources in the Koro District                

As a final point of discussion, the issue of overall sustainability of the groundwater resources in the Koro district should be addressed.  Seasonal fluctuations in the large-diameter wells suggest that some degree of local recharge is taking place.  The amount of drawdown over time of the aquifers was estimated by the DNHE as being on the order of 5 mm per year, but there is not enough available data to accurately quantify the annual drawdown.  However, due to the absence of large-scale agricultural or industrial extractions of groundwater in the region, it can be assumed that the annual local recharge and the longer term recharge from the cliffs of the falaises is sufficient to meet rural domestic and livestock needs for the near future.  The most important groundwater extraction occurs in Pomorodoudiou to supply the city of Koro (pop. 9,300, est. 2001).  To date, the source aquifer has been able to satisfy the demands of the growing urban population in Koro, but concerns over the viability and longevity of the aquifer were expressed by the local mayor’s office and the residents of Koro, and this is something that needs to be monitored and researched further. 

link to page top


This research has contributed insight into the complex nature of the rural water supply situation in the Koro region of Mali.  Choice-of-technology preferences vary according to several factors including local perception of water scarcity, individual water use requirements, and previous experience with various technologies.  Sustainability of various types of water supply infrastructure is dependent upon the degree to which the technology corresponds to the needs of the local community and the community’s ability to maintain and repair it over time.  Considering the poor state of the manual pumps observed in the district of Koro, it is apparent that efforts need to be made to ameliorate the situation.  Learning from previous development projects, the latest approaches address the problems of the limited availability of spare parts, the absence of trained technicians at the local level and the limited role of women in the pump management scheme.  Dedicating more time and resources to the maintenance and management aspects of rural water supply development is a positive action and should help to improve sustainability of newly installed water supply infrastructure.  However, the continued lack of community involvement in the decision-making phase of water supply development projects, deeply entrenched patriarchal structures that continue to constrain leadership and decision-making participation of women and minorities, principal stakeholders in water use, and the potentially global inappropriateness of manual pumps as a sustainable long-term option remain to be addressed.   

Recent studies have shown that sustainability of water supply projects improves when communities are allowed to take a central role during all stages of the project, including design and planning (Mangin 1991, Williams 1998, Bah 1992).  Limited success of many such projects is due to the communication gap between the local communities and the project planners, as well as the over-emphasis among implementing agencies upon population coverage rather than a clear focus on facility design and system support in cooperation with local institutions to ensure long-term functioning and use of the water supply infrastructure (Chandler 1989).  Community involvement should include a complete and total transfer of information from all of the parties involved to ensure that tradeoffs are adequately explained and decisions can be made based on the fullest possible set of data available at the time.  In order for the array of choice-of-technology options to be as inclusive as possible, institutional flexibility on behalf of the funding agencies, the implementing agencies, and the community members necessarily needs to exist.      

The need for this type of collaborative, participatory project planning has in some ways been circumvented by current development philosophy that argues that the community’s willing contribution of financial resources to a project adequately demonstrates their preference for the selected technology.  However, due to the water scarce geographical context in which these communities find themselves, this may not necessarily be the case.  In these communities, community leaders are for all practical purposes obligated to financially contribute to any type of water supply project, manual pumps or otherwise, that is made available to the community.  Given that the implementing agencies typically offer only one technological option, the community’s financial contribution does not necessarily reflect their preference for the selected technology but rather reiterates the fact that they live in a water scarce region and are willing to contribute financially, even if this is a significant sacrifice on their part, to any project that may help them ensure access to their water resources.      

The reluctance of many funding agencies to consider the large-diameter well as a suitable component of rural water supply development limits the number of options available to the implementing agencies and communities alike.  This reluctance is based on the fact that large-diameter well water is not considered to be potable by the international funding agencies as it is typically not as free of bacteria as water from a closed source such as a manual pump.  However, recent evidence continues to show that hygiene practice and adequate sanitation facilities are of equal importance as source water quality in improving the health of the local population (Brikké and Bredero 2003, Gasana et al. 2002, Van Derslice and Briscoe 1995).  Ready access to water and the resulting increases in the quantity used for hygiene can have a greater impact on health than water quality improvements alone (Nyong and Kanaroglou 2001,Cairncross 2003).  These findings, along with the significant livestock water needs of the district, suggest that the most elemental objective should be to provide easy access to an adequate supply of water, rather than limited access to a more potable source of water.  Considering the poor state and low flow rates of the manual pumps in the district, it does not seem likely that a water supply development project focusing on hand pumps will prove to be a sustainable, long-term solution.  Depending on the village, large-diameter wells or a large-scale solar pumping system may be considered to be the sustainable and preferred choice-of-technology option to provide access to an adequate supply of water.  In the case of WAWI, one step that the donor agency could make, would be to allow World Vision more flexibility in choosing the type of water supply infrastructure that they can provide, which in turn will allow World Vision to offer the communities a real choice in the choice-of-technology decision making process.       link to page top


Regardless of the community’s technological choice, the importance of hygiene and sanitation will need to be continually emphasized in an effort to reduce the occurrence of contamination from the point-of-collection to the point-of-use.  Tangible village-wide health improvements are difficult to achieve and the interdependence of source water quality, hygiene practice, sanitation facilities and adequate water supply cannot be ignored.  A recent study in Rwanda has concluded that, “there is no best way to achieve health improvements, except that it must be by the people themselves.  Interventions in water supply and sanitation infrastructure, together with hygiene education and the extension of primary health care services, need to be implemented within an integrated multidisciplinary framework.” (Gasana et al. 2002, 87)  Once again, it can be seen that the community members themselves are placed at the forefront of this multidisciplinary effort.  In the water scarce district of Koro, community members, NGO regional managers, field staff and government officials all stress the importance of permanent access to an adequate supply of water stating that this will create the conditions to improve hygiene and eventually improve the health of the community.  If these multiple voices can be heard and if the funding and implementing agencies are able to respond to these voices, sustainability of water supply development projects and their positive impact on the health of the community should improve.


Figure 3: Condition of Manual and Powered Pumps,
Counties of Koro and Bondo, June 2004
(click here for full-sized image) link to page top

Table 9.  Pump Inventory Results and Specifications, June 2004.

(click here for full-sized image)link to page top



Water Quality Testing Results (Bacteriological Analysis)      

To determine the effect of water storage behavior on household water quality, a bacteriological analysis was performed by the DNHE on seven samples taken from various source waters (hand pumps, large-diameter wells) at the points-of-collection and at the points-of-use (Table 10).  Visual inspection of the well water at the time of sampling suggested that total suspended solids content was much higher than for the pump water.  After analysis, visual inspection of the testing plates representing the wells found that they were much more densely colonized by non-coliform colonies than the plates from the pump samples.  Most notably, the plates representing the water jars filled with well water had much denser bacterial colonies than the plates representing the well itself.  This suggests that the conditions inside the water jars may be favorable for certain bacteriological growth and further testing may provide important insight into this issue.    

The coliform analysis found that no fecal coliforms were found to be present in any of the samples.  No total coliforms were found to be present in the samples taken from the large-diameter well (okorokana) or the water jars filled with water from the same well.  Total coliforms were however found to be present in the water taken from the Vergnet foot pump and in one of the water jars filled with this water.  The levels were on the order of 20 colonies per 100 mL and the water from all samples was deemed by the DNHE to be of an acceptable quality (Eau de qualité bactériologique acceptable).  

Table 10.  Bacteriological Analysis of Source
and Household Water Samples, April 2004 

DNHE ref #

Origin of Water

Description of Sample

Total Coliforms

Fecal Coliforms



well (okorokana)

drinking cup - uncovered water jar



~100% covered w/ red colonies, very dense and large



drinking cup -covered water jar



~100% covered w/ red colonies dense, small diameter



drinking cup - covered water jar



~100% covered w/ red colonies dense, small diameter



puisette from the well



least dense of the well samples.  markedly so.  small diameter



directly from the pump



22 green colonies (~10% coverage) + few large red colonies



drinking cup - covered water jar



1/4 as many red colonies as sample 24



drinking cup - covered water jar



similar to 25 but with 21 green colonies as well

link to page top



Akuoko-Asibey, A. (1997) Views of selected government officials on the impact of a rural water supply program in Ghana. Evaluation and Program Planning 20(2), 225-230.

Bah, O.M. (1992) Community participation and rural water supply development in Sierra Leone. Community Development Journal 27(1), 30-41.

Cairncross, S. (2003) Editorial: Water supply and sanitation: some misconceptions. Tropical Medicine & International Health 8(3), 193-195.

Carter, R.C., Tyrrel, S.F. and Howsam, P. (1999) The impact and sustainability of community water supply and sanitation programmes in developing countries. Journal of the Chartered Institution of Water and Environmental Management 13(4), 292-296.

Carter, R.C., Tyrrel, S.F. and Howsam, P. (1993) Lessons Learned from the UN Water Decade. Journal of the Chartered Institution of Water and Environmental Management 7(6), 646-650.

Chandler, C.G. (1989) Beyond ‘coverage’: An integrated approach to rural water supply and sanitation projects and programmes. Natural Resources Forum 13(1), 53-58.

Feachem, R.G. (1980) Community participation in appropriate water supply and sanitation technologies: The mythology for the Decade. Proc. of the Royal Society of London B209, 15-29.

Food and Agriculture Organization of the United Nations (1986) Water For Animals AGL/MISC/4/85. Web address: Last visited April, 2005.

Gasana, J., Morin, J., Ndikuyeze, A. and Kamoso, P. (2002) Impact of Water Supply and Sanitation on Diarrheal Morbidity among Young Children in the Socioeconomic and Cultural Context of Rwanda (Africa). Environmental Research Section A 90, 76-88.

Genthe, B., Strauss, N., Seager, J., Vundule, C., Maforah, F. and Kfir, R. (1997) The Effect of Type of Water Supply on Water Quality in a Developing Community in South Africa. Water Science Technology 35(11-12), 35-40.

Kendie, S.B. (1992) Survey of water use behaviour in rural North Ghana. Natural Resources Forum 16(2), 126-131.

Lindskog, R.U.M and Lindskog, P.A. (1988) Bacteriological contamination of water in rural areas: an intervention study from Malawi. Journal of Tropical Medicine and Hygiene 91, 1-7.

Livingstone, A. and McPherson, H.J. (1993) Community Management of Rural Water-Supplies – Lessons for Developing-Countries from a Western Canadian Experience. Water International 18(4), 225-232.

Mangin, J.M. (1991) Rural water supply in southern Ethiopia: Failures and Alternatives. Canadian Journal of Development Studies 12(2), 297-312.

MacRae, D. and Whittington, D. (1988) Assessing Preferences in Cost-Benefit Analysis – Reflections on Rural Water-Supply Evaluation in Haiti. Journal of Policy Analysis and Management 7(2), 246-263.

Mu, X., Whittington, D. and Briscoe, J. (1990) Modeling village water demand behavior – A discrete choice approach. Water Resources Research 26(4), 521-529.

Narayan, D. (1995) The contribution of people’s participation: Evidence from 121 rural water supply projects. The World Bank: Environmentally Sustainable Development Occasional Paper Series, No. 1.

Nyong, A.O. and Kanaroglou, P.S. (2001) A survey of household domestic water-use patterns in rural semi-arid Nigeria. Journal of Arid Environments 49(2), 387-400.

Roberts, L., Chartier, Y., Chartier, O., Malenga, G., Toole, M. and Rodka, H. (2001) Keeping clean water clean in a Malawi refugee camp: a randomized intervention trial. Bulletin of the World Health Organization 79, 280-287.

Toulmin, C., Leonard, R., Hilhorst, T. and Diarra, D. (2000) Mali Poverty Profile: Report prepared for SIDA. IIED Drylands Programme, May 2000 , 6-7.

Trevett, A.F., Carter, R.C. and Tyrrel, S.F. (2004) Water Quality Deterioration: A Study of Household Drinking Water Quality in Rural Honduras. International Journal of Environmental Health Research 14(4), 273-283.

United Nations. (2000) UN Millennium Development Goals (MDG). Web address: Last visited March, 2005.

VanDerslice, J. and Briscoe, J. (1995) Environmental Interventions in Developing Countries – Interactions and their implications. American Journal of Epidemiology 141(2), 135-144.

VanDerslice, J. and Briscoe, J. (1993) All coliforms are not created equal: A comparison of the effects of water source and in-house water contamination on infantile diarrheal disease. Water Resources Research 29(7), 1983-1985.

Williams, T.O. (1998) Multiple uses of common pool resources in semi-arid West Africa: A survey of existing practices and options for sustainable resource management Overseas Development Institute (ODI) 38.

Wright, J., Gundry, S. and Conroy, R. (2004) Household drinking water in developing countries: A systematic review of microbiological contamination between source and point-of-use. Tropical Medicine and International Health 9(1), 106-117.

[1] Formally presented by Bara Kassambara of Winrock International and repeatedly stressed by various community members, NGO and local government authorities.

[2] Official 2001 figures from the Koro Mayor’s office for the village of Yadianga reported 339 cattle, 1531 sheep, 9 horses, 129 donkeys and 3 camels.  These figures were acknowledged by the mayor’s office and everyone else in the region as gross underestimates due to severe under reporting by the population.  The figures used here are more realistic estimates of actual population sizes in 2004.

link to page top