TITLE: Water Supply Coverage and Water Loss in Distribution Systems “The case of Addis Ababa”

AUTHOR: Welday Berhe Desalegn

OBJECTIVES:

1. The main objective of the research is to evaluate the supply coverage and explore the water

loss in city water supply distribution and suggest a method to better identify and reduce the

loss.

2. To evaluate the domestic water supply coverage and distribution

3. To evaluate the total loss of water (unaccounted for water) at city level.
4. The evaluate and compare the total water loss in selected sample areas (sub-systems)
5. To explore the possible causes of water losses and possible solutions
6. To suggest a methodology to support a city water supply

SUMMARY:

 Both the average water supply coverage and the intra city distribution were evaluated

based on the daily per capita consumption and level of connection using the

population data of the city. The average water supply coverage of the city is found to

be 16.65 litre/person/day. This average per capita consumption is found much lower

compared with other developing cities like the southern African cities and even is

lower than that of the standard set by UN-Habitat as a basic need (20l/per/day). The

average number of connections per family of the city which is equivalent to 28% in-

house or yard connection, is also far below the African cities’ average of 43%.

 The intra city distribution of the water supply coverage was also using a correlation

between number of population and yearly consumption of each Kebele and the

correlation coefficient is found to be positive and strong (+0.749). This positive and

strong correlation on the one hand shows that although there is overall shortage of

water in the city, predominantly the existing amount of water is fairly distributed

among the different localities except few kebeles that consumed much water although

their number of population is either low or moderate. On the other hand, except one

Kebele (Wereda 19 & Kebele57), no Kebele is found consuming extreme low

amount of water while compared to its number of population.

  Despite the low water supply coverage of the city, the total water loss is found to be

high enough (up to 41%). The total water loss was computed by subtracting the

consumption from the water supplied. Three approaches were used to compare the

loss among the sub-systems, (i) the UFW expressed as a percentage, (ii) loss per

length of pipes and (iii) loss per connection. Comparison using the percentage has

reversed the results of the comparison using the loss per length of mains and loss per

number of connection.

 From the water loss analysis of the sub-systems, higher water loss has been found in

Colfe Core sub-system which is located relatively at lower elevation in comparison to

other sub-systems and has higher elevation differences with in the sub-system itself.

From this it can be concluded that pressure resulting from significant ground

elevation differences (potential pressure) have an impact on the rising of water loss

due to leakage. These also give an indication that the predominant cause of the loss

might be that of leakage.

 The seasonal water loss comparison that was intended to be included in the analysis

was only done for one year duration due to data limitation. Although the total water

loss of the city level analysis show a relatively higher percentage of loss during the

rainy season, it might be difficult to reach this conclusion only from one year’s data.

At leas two to three years data is necessary to observe the trend if the higher

percentage of loss in the rainy season is repeating itself in the next rainy seasons.

Yet, from the local experts’ point of view, due to the difficulty of tracing leaking

pipes, water loss is expected to be higher in the rainy season.

 The other issue addressed in the analysis was that of the impact of pipe age on water

loss. On the one hand the loss was found to be higher in the sub-system where pipes

of relatively older age are located. On the other hand despite the fact that overall pipe

network seems to be of a young age, the total water loss is higher. This signifies that

besides to the loss caused as a result of leakage, other non-physical losses may also

be expected to be higher. To this effect, as illegal connection is not noticed as major

problem, loss due to meter errors especially under recording of meters is expected to
be higher.

Recommendations

 While the higher magnitude of the loss is evaluated in conjunction with the amount

of water which is extremely low, the magnitude of the loss is expected to be much

higher had there been enough water supply and pressure in all areas. Therefore due

attention should be given while additional water sources are planned for the future

that a proper management of the existing infrastructure in general and the water

network in particular is paramount.

 One of the major limitations of this study was the quality of the consumption data

especially the customer meter readings. The contract number of the customer meters

which is supposed to be the unique identifier has been found to be repeated for some

of the contacts. As the data set is very large it may even difficult to know which

contracts are active at present. The other problem observed was, as each of the

twelve months data of the customer meters are not linked together it will be very

difficult to compare for any significant reading differences among consecutive

period’s meter readings. Therefore for better management of the billing system in

general and better evaluation of water losses comparing with the billed water

consumption, the problem stated in 3.3.3.2 of the Methodology chapter need a further

study.

 Updating of the network which is undergoing by the GIS unit of AAWSA is

appreciated, but this need to be integrated with the land information system (LIS) of

the city as well as information on hydraulic flow of the water network. Operation

and maintenance data including pressure records need also be integrated spatially

with the network Therefore, introducing geographic information system (GIS) is

timely as it may facilitate the updating of the networks and support to perform related

spatial analysis. The recently prepared digital elevation model (DEM) of the city may

support to divide the network in to manageable smaller hydraulic zones or district

meter areas.
METHODOLOGY