Investigation of the Suitability of Simple and Rapid Techniques for Leakage Management in Water Distribution Systems: A Case of Blantyre Water Supply Area, Malawi

Abstract

Water demand continues to increase due to rapid population growth, changes in standard of living, and industrial development among others. Water demand levels could be greatly reduced if water losses are reduced particularly leakages, which normally form the large part of total water losses. Leakage management has been a challenge for most developing countries due to financial constraints, lower levels of skills, and technology. For this reason, a study was carried out in Blantyre, Malawi to investigate suitability of simple and quick techniques that could improve leakage management so as to reduce and maintain non revenue water (NRW) levels within acceptable limits. The study included determination of NRW levels in Blantyre water supply area, investigation of leakage levels through analysis of minimum night flows (MNF), investigation of pressure relationships, examination of pipe failure modes, and investigation of prediction tools for pipeline replacement. Historical monthly data for water produced and sold was collected for the period from July 2001 to January 2008 to determine the trend of NRW for the entire water supply area. The specific study areas, Chinyonga and BCA district metered areas (DMAs) were selected in order to investigate leakage levels as well as relationships between pressure, leakage, burst frequency, and MNF. Data loggers were installed on the inlet points of the specific study areas to record pressures, while flows were recorded by a ultrasonic flow meter. The number of pipe bursts was obtained by field observations on a daily basis for investigation of pressure and burst frequency relationships. Broken pipe samples were also collected and analysed on their mode of failures so as to establish a technique for quick estimation of leakage flow rates in the study area. Additional information for the entire water supply area such as population, length of pipelines, cost of water were gathered from documentation reviews for the purpose of investigating a suitable decision tool for pipeline replacement. The NRW was found to be 52% of total water produced (or 423 m3/conn/yr) by the end of 2007 and 72% of which were leakage losses. A general increase in NRW by an average of 2.5% per annum was found from 2001 to 2007. Analysis of MNF figures showed that excess night flows ranged from 1.26 m3/hr to 3.54 m3/hr for Chinyonga and 1.05 m3/hr to 3.50 m3/hr for BCA. In addition to the determination of leakage levels, the SANFLOW model was also found to be a suitable tool for estimating expected number of pipe bursts in water distribution systems. It was also established that the mean coefficient relating pressure and leakage (leakage exponent, N1) was 1.6 and 1.0 for Chinyonga and BCA DMAs respectively. Furthermore, in the investigation of the relationship between pressure and burst frequency, mean burst frequency exponent, N2 (coefficient relating pressure and burst frequency) was found as 1.9 for both specific study areas, which compares well with values found elsewhere that ranged from 0.5 to 6.5. However, the mean N2 exponents found had huge standard deviations, indicating that there is a relationship but not very strong using the provisional equation that was adopted. Investigation of a relationship between pressure and MNF confirmed that there is also relationship. Regarding the examination of pipe failure modes, it was found that specific pipe material had specific predominant failure mode, which can be used to estimate leakage flow rates. The study also established that an ECONOLEAK model is a suitable prediction tool for pipeline M.M. Chiipanthenga, 2008 ii replacement in the Blantyre water supply area. In conclusion, it was investigated that the simple and quick measures as recommended by International Water Association (IWA) task force members on water losses are suitable and could prove to be necessary techniques for reduction of NRW in the study area. It is also recommended that the utility in the study area should commence active leakage control and pressure management in addition to pipeline replacement activities as soon as possible so as to reduce and maintain NRW levels within acceptable limits.