Modelling software removes guesswork from water utility management

The Canadian engineering company SNC-Lavalin used InfoWorks WS to model the water supply network in Banju, Gambia, in order to better manage the system.

Wallingford Software’s InfoWorks WS helped engineers devise the most effective ways to improve the water supply for the area around Gambia´s capital city, Banjul. The Canadian engineering and construction company SNC-Lavalin used InfoWorks to model the network, help determine better ways to manage the system, and identify priorities for capital investment.

The state utility NAWEC, which is responsible for water supply, sewerage, drainage, solid waste and electricity, contracted SNL-Lavalin to help improve its services given major demographic changes in the past decade.

The West African country of Gambia experienced significant migration of its population from rural to urban areas that led to a sevenfold population growth in its capital city of Banjul. The city grew by 300,000 during this period of time.

For a country only 400-km long and 70-km-wide and with a population of 1.5 million, this migration created serious water supply, sanitation, and drainage problems. The government decided to invest in expanding the water supply system to meet increased demand, improve living standards, and boost the economy. This investment will also improve maintenance and training, and introduce new technologies.

The project will improve water supply, sanitation, drainage and solid waste through a 25-year masterplan, a feasibility study looking 10 years ahead, and detailed design to meet immediate needs.

The feasibility study and InfoWorks WS modeling of the water supply focused on the Greater Banjul area, but the project also included provision for training in how to extend the models to the rest of the country.

The project started in May 2004 so work on creating the InfoWorks WS model begun in November 2005 following development of a three-phase plan. The first phase included a ring main around the area´s well fields and the construction of a new treatment facility with an ultimate capacity of 400 liters per second (l/sec). The smaller second phase includes a new well field to meet growing demand, while the third phase involves the coastal area, which is perfectly located for tourism development.

The existing distribution network includes wells, treatment facilities, water tanks and towers. Nineteen elevated tanks in the Greater Banjul area were sized from 250m3 to 500m3. Treatment includes aeration, chlorination, and use of contact tanks.

Production of the master plan, feasibility study, and detailed design highlighted local factors that influenced the choice of solutions. Gambia does not lack water, but several issues hinder its effective distribution, including a parts shortage and staff training. Maintenance was virtually non-existent.

The InfoWorks WS model provided a reliable mechanism for evaluating earlier studies and preparing the subsequent stages of work for implementation.

Much of the basic equipment is nearing the end of its life. Some water supply networks suffer from low pressure. Water towers leak, and some wells are located close to septic tanks. Long lists of immediate needs at many of the production sites, including generators to operate the boreholes, improved access roads, chlorination equipment, spare parts, and PVC pipes remain undone. Water tanks are often empty and liners tend to fail in the heat of the surrounding metal tank. Replacement funds are rarely available.

New boreholes and extensions to the distribution network are needed for immediate and long-term needs. The project also provided for leakage detection equipment although the analysis showed that this was not needed. The system has low pressure and leakage is less than 10 percent.

The main treatment plant has a capacity of 250 l/s while its full design capacity is 20 percent more; however pumping problems arise from a lack of good pumps.

One of the uses of the InfoWorks model was to study network pressures and determine the optimum operating conditions. Successive runs were carried out to test the effects of adjustments, such as the timing of pump operations. One of the outcomes of the analysis was that there was no need to build two booster stations along the coastal ring that had initially been thought necessary.

Analysis using InfoWorks also showed that the number of wells proposed in a particular area could be reduced from an initial estimate of 140. Further analysis and checking of the data showed that fewer would be required if the existing 152-mm wells were rehabilitated to larger diameters. Furthermore, having a large number of wells would mean having a large stock of spare parts, which was best avoided given the uncertain supply arrangements. These factors led to a reassessment of the potential for rehabilitating the existing wells. This was found to bring significant cost savings in the order of US$80,000 per well. Another advantage was that there would be more stability in the aquifer yield and recharge with no extreme drawdowns or saltwater inflow.

The study found no shortages in water production or distribution, but many of the problems were found to stem from operational and maintenance issues. Investigations showed sufficient water, but staff had no way to know when the reservoir was becoming empty. The utility has received complaints about the modern treatment facility not providing enough water and calls for new water wells.

Existing operations tended to rely on guesswork. Without telemetry, every operational check requires a site visit. Difficult access makes this even harder as roads are often washed away by heavy rain. The country straddles the River Gambia and a lack of bridges hampers daily activities including the operation of the water supply system.

InfoWorks WS was used to develop a model that will enable staff to be trained to take actions, such as when to open a valve. This also opens the way for later implementation of a telemetry system. Gambia’s good mobile network will facilitate the incorporation of data loggers into the network, enabling the information to then be relayed back to the central control.

---

Author’s Note

Zoran Mrdja is senior water resources engineer with Trow Associates Inc. He was formerly with SNC-Lavalin/co Arch-NOVA, the company that carried out the work described in this article. The author can be reached by email at: [email protected]