ABSTRACT Self-supply is widely reported across various contexts, filling gaps left by other forms of water supply provision. This research assesses low-cost household groundwater supply technologies in markets in developing country contexts of sub-Saharan Africa and Latin America, with a focus on the potential for improving Self-supply technology implementation and markets in sub-Saharan Africa. Specifically, a mature and unsubsidized Self-supply market for Pitcher Pump systems (suction pumps fitted onto hand-driven boreholes) is studied in an urban context in Madagascar, EMAS low-cost water supply technologies are assessed in Bolivia, and a technical comparison is completed with manual EMAS Pumps and family versions of the Rope Pump in Uganda. In Madagascar, locally manufactured Pitcher Pump systems are widely provided by the local private sector, enabling households to access shallow groundwater. This market has developed over several decades, reaching a level of maturity and scale, with 9000 of these systems estimated to be in use in the eastern port city of Tamatave. The market is supplied by more than 50 small businesses that manufacture and install the systems at lower cost (US$35-100) than a connection to the piped water supply system. Mixed methods are used to assess the performance of the Pitcher Pump systems and characteristics of the market. Discussion includes a description of the manufacturing process and sales network that supply Pitcher Pump systems, environmental health concerns related to water quality, pump performance and system management. The research additionally considers the potential of EMAS low-cost household water supply technologies in accelerating Self-supply in sub-Saharan Africa, and consists of a field assessment of EMAS groundwater supply systems (handpumps on manually-driven boreholes) and rainwater harvesting systems as used at the household level in Bolivia, focusing on user experiences and the medium/long-term sustainability of the pump (cost, functionality, etc.). The EMAS Pump is a low-cost manual water-lifting device appropriate for use at the household level. Developed in the 1980s, the EMAS Pump has been marketed extensively for local manufacture and use at the household level in Bolivia, and marketed to a lesser extent in other developing countries (mainly in South and Central America). The simple design of the EMAS Pump, using materials commonly found locally in developing countries, allows for it to be fabricated in many rural developing community contexts. Its capability for pumping from significant depths to heights above the pump head makes it quite versatile (e.g. for pumping to household tanks, reservoirs at higher elevations, or for installing multiple pumps on wells). A survey/inspection of 79 EMAS Pumps on household water supply systems in areas of three regions of Bolivia (La Paz, Santa Cruz and Beni regions) showed nearly all EMAS Pumps (78 out of 79) to be operational. 85% of these operational pumps were found to be functioning normally, including 72% that were reported to have been installed eleven or more years earlier. It is shown that rural households in Bolivia are able to maintain EMAS Pumps. The EMAS Pump can be installed and repaired by local technicians, and numerous examples were seen of small groups of local technicians that operate small businesses installing and repairing such systems. The cost of a new EMAS Pump was reported by users to be US$ 30-45. Maintenance and repair costs of the EMAS Pump were found to be reasonable, with pump valve replacement (the repair most commonly reported by users) costing an average of US$9 (materials and labor). The Rope Pump has some similar attributes to the EMAS Pump, in that it is can be made locally from materials commonly available in developing communities, it has a relatively low cost, and is simple to understand. The Rope Pump is well-known among international rural water supply professionals, and thus serves as a good baseline to compare the lesser-known EMAS Pump. A technical comparison completed in Uganda of the EMAS Pump and the Rope Pump considered performance (flow rates and energy expended, pumping from various depths), material costs, and requirements for local manufacture. The study concluded that, based on its relative low-cost (material costs ranging from 21-60% that of the family Rope Pump, dependent on depth and pumping pipe size), similar technical performance to the Rope Pump when pumping from a range of depths, and the minimal resources needed to construct it, the EMAS Pump has potential for success in household water supply systems in sub-Saharan Africa. Combined with the conclusion from the research in Bolivia, it is believed that there is considerable potential for the EMAS Pump as a low-cost option for Self-supply systems in sub-Saharan Africa. Recommendations for further research focus on: (1) improvements to the Pitcher Pump system (focusing on reducing risk of water contamination); (2) formative research to identify factors that have led to the sustainability of the Pitcher Pump market in eastern Madagascar, and (3) development of the Self-Supply Market in Madagascar beyond Pitcher Pump systems.

With 663 million people still without access to an improved drinking water source, there is no room for complacency in the pursuit of Sustainable Development Goal (SDG) Target 6.1: “universal and equitable access to safe and affordable drinking water for all” by 2030 (WHO, 2017). All of the current efforts related to water supply service delivery will require continued enthusiasm in diligent implementation and thoughtful evaluation. This cannot be over-emphasized in relation to rural inhabitants of low-income countries (LICs), as they represent the largest percentage of those still reliant on unimproved drinking water sources. In that lies the motivation and value of this thesis research- improving water supply service delivery in LICs. Manually operated suction pumps, being relatively robust, low cost, and feasible to manufacture locally, are an important technology in providing access to improved drinking water sources in LICs, especially in the context of Self-supply. It seems widely accepted that the water-lifting limit of suction pumps as reported in practice is approximately seven meters. However, some observations by our research group of manually operated suction pumps lifting water upwards of nine meters brought this “general rule of thumb” limit into question. Therefore, a focused investigation on the capabilities of a manually operated suction pump (a Pitcher Pump) was conducted in an attempt to address these discrepancies, and in so doing, contribute to the understanding of this technology with the intent of providing results with practical relevance to its potential; that is, provide evidence that can inform the use of these pumps for water supply. In this research, a simple model based on commonly used engineering approaches employing empirical equations to describe head loss in a pump system was used to estimate the suction lift limit under presumed system parameters. Fundamentally based on the energy equation applied to incompressible flow in pipes, the empirically derived Darcy-Weisbach equation and Hydraulic Institute Standards acceleration head equation were used to estimate frictional and acceleration head losses. Considering the theoretical maximum suction lift is limited to the height of a column of water that would be supported by atmospheric pressure, reduced only by the vapor pressure of water, subtracting from this the model was used to predict the suction lift limit, also referred to herein as the practical theoretical limit, assuming a low (4 L/min) and high (11 L/min) flow rate for three systems: 1) one using 1.25-inch internal diameter GI pipes, 2) one using 1.25-inch internal diameter PVC pipes, and 3) one using 2-inch internal diameter PVC pipes. In all considered cases, with an elevation equal to sea level, the suction lift limit was estimated to be over nine meters. At a minimum, the suction lift limit was estimated to be approximately 9.4 meters for systems using 1.25-inch internal diameter pipe and 9.8 meters for systems using 2-inch internal diameter pipe, with essentially no discernable effects noticed between pipe material or pipe age. Additionally, laboratory (field) trials using a Simmons Manufacturing Picher Pump and each of the aforementioned pipe specifications were conducted at the University of South Florida (Tampa, FL, USA) to determine the practical pumping limit for these systems. Results from the pumping trials indicated that the practical pumping limit- the greatest height at which a reasonable pumping rate could be consistently sustained with only modest effort, as perceived by the person pumping- for a Pitcher Pump is around nine meters (9 meters when using 1.25-inch internal diameter GI or PVC pipe and 9.4 meters when using 2-inch internal diameter PVC pipe). Therefore, results from this research present two pieces of evidence which suggest that the practical water-lifting limit of manually operated suction pumps is somewhere around nine meters (at sea level), implying that reconsideration of the seven-meter suction lift limit commonly reported in the field might be warranted.

Rural improved water supply coverage in Uganda has stagnated around 64% for a number of years and at this point more than 10 million rural people do not have access to an improved drinking water source. It has been recognized that progress toward improved water supply coverage and increased service levels may be gained through Government and nongovernmental organization (NGO) support of private investment in household and shared water supplies, commonly known as Self-supply. Self-supply can be promoted by introducing and building local capacity in appropriate and affordable water supply technologies such as hand-dug wells, manually drilled boreholes, low-cost pumps, and rainwater harvesting. Support can also be focused on technical support, marketing, financing, and strategic subsidies that promote and enhance user investment. The Uganda Ministry of Water and Environment has embraced Self-supply as a complementary part of its water supply strategy while government and NGO programs that support Self-supply have emerged. The EMAS Pump is a low-cost handpump appropriate for use in household water systems in the developing world. There are more than 20,000 in use in Bolivia, with many constructed through Self-supply. The EMAS Pump is constructed from simple materials costing about $US 10-30, depending largely on installation depth, and can be fabricated with simple tools in areas with no electricity. The EMAS Pump is used with low-cost groundwater sources such as hand-dug wells and manually drilled boreholes or with underground rainwater storage tanks. It can lift water from 30 m or more below ground and pump water with pressure overland or to an elevated tank. The objectives of this research were to conduct an assessment of the EMAS Pump that considers pumping rates, required energy, and associated costs, to characterize the EMAS Pump for its potential for use in household water systems in Uganda, and to make relevant recommendations. The potential of the EMAS Pump was assessed through testing its use with 2 subject participants (male and female) on wells of 5.1 m, 12.6 m, 17.0 m, 18.4 m, 21.1 m, and 28.3 m static water levels as part of a side-by-side comparative assessment with the Family Model version of the Rope Pump, a more widely known low-cost handpump that has recently been introduced and promoted in Uganda. Shallow and deep versions of each pump were tested on selected wells for 40-liter pumping trials. The status and feasibility of low-cost groundwater development and underground storage tanks were also explored in order to help characterize the potential of the EMAS Pump as an option for low-cost household water systems in Uganda. In general, it was observed that the EMAS Pump performed comparably to the Rope Pump in terms of pumping rates for shallow depths, but the Rope Pump outperformed it on deeper wells. It was determined that the EMAS Pump required more energy for pumping during nearly all trials. A study of relevant supply chains in Uganda concluded that the EMAS pump has a material cost that is less than 50% of the Rope Pump for most applications and 21% of the cost for shallow wells. It was also determined that the EMAS Pump could feasibly be produced nearly anywhere in the country. There are indications that low-cost wells and underground rainwater tanks are applicable in many parts of Uganda and could be paired with an EMAS Pump to achieve significant affordability for Self-supply household water systems. Recommendations are provided in terms of the feasibility of introducing the EMAS Pump as a part of Self-supply strategy in Uganda.