In a town scarred by war

 Pier Giorgio Nembrini  has a Ph.D. in analytical chemistry, specializing in water treatment and environmental analysis. He was head of the ICRC's Water and Sanitation Unit from 1983 to 1995, and until 1997 he coordinated the ICRC's activities in this field in the Horn of Africa and the Great Lakes. He is now working as an independent consultant.  

 Riccardo Conti  is a hydrogeologist and mining engineer, with experience in hydrogeophysical techniques and mining exploration. He worked for 10 years in the private sector before joining the ICRC in 1988. Having carried out many missions in Africa, Asia, Europe and the Middle East, he is now based at the organization's headquarters in Geneva, Switzerland, where he is responsible for the Water and Habitat Unit.  

 The uncontrolled privatization of water distribution, which followed several chaotic war years in Mogadishu, Somalia, has benefited its 1 million residents in many ways - responding to water needs, keeping down costs and creating wealth. However, such a process has its own dangers. Giorgio Nembrini and Riccardo Conti describe them and the ICRC's water and sanitation programme in the city.  

By the end of the 1980s, Mogadishu's water supply was already on the verge of collapse. Fuel shortages, looting and technical breakdowns - which went unrepaired for weeks for lack of spare parts a nd because of security constraints - made water distribution through the poorly maintained network very erratic. Complete breakdowns could last for weeks and large areas of the town were deprived of a single drop of water for years. The situation was even more precarious in the 1990s owing to the damage inflicted by the civil war. In the aftermath of the 1991 war, an attempt was made to run the system, when some agencies helped the local water board to maintain minimum distribution by providing it with fuel, new generators and spare parts.

It soon became clear, however, that the city's water problem would not be solved simply by supplying fuel and spare parts. Moreover, it would take years for the country to reach a political settlement that would restore the donors'confidence sufficiently for them to make the huge investments necessary for such a town. In the meantime, the inhabitants had to survive using hand-dug wells and boreholes located within urban areas and drilled in response to industrial and economic expansion. Many hand-dug wells were also equipped with motorized pumps.

 Improving the water supply  

The total collapse in mid-1995 of the distribution network that depended on the Afgoi well-fields convinced the ICRC of the need for a programme to drill and rehabilitate boreholes, in addition to the usual maintenance of hand-pumps. The aim of the exercise was to increase the water supply to most of the city. This was done at the sites of former productive boreholes, which were either not operational or in poor condition, but which were dug sufficiently far from the sea to avoid salinization problems. Other agencies, such as ACF [1 ] , were concentrating on rehabilitating hand-dug wells supplying water to internally displaced people. UNICEF [2 ] was supplying the chemicals required to disinfect most of the wells during cholera outbreaks.

The ICRC's programme financed the drilling and equipment of the new boreholes, and the " owner " of the borehole was then responsible for operating it and distributing the water. Between the beginning of 1995 and the end of 1997, six new boreholes were drilled, tested and each equipped with a submersible pump. The diesel generators to power the pumps were installed in protected premises (University Residence Camp, Gupta, Muuri, Yakshid, Daynile and Medina). Four boreholes (Black Sea, Livestock, Big Pipe and Gulwadayasha) were cleaned and equipped, and 12 hand-dug wells were equipped with submersible pumps powered by diesel generators (two in Medina, two in Bermuda, four in Mogadishu North and four in Mogadishu South). Some boreholes located in critical locations, like the one supplying water to Digfer Hospital, were completely rehabilitated. Others, such as the one in Sheikh Adan Camp, had been running without problems since 1992.

 Water quality and quantity  

At the beginning of 1997, the ICRC carried out a survey aimed mainly at monitoring the effects of saline intrusion related to possible overharvesting. The results (Figure 1) showed that the inland movement of the saline wedge, expressed as the 3,000 microsiemens/cm boundary, was close to 2 km since the last record in 1980. [3 ]

The situation was not considered dramatic as far as salinity was concerned [4 ] . However, salinity not only influences public health (overly saline water will not be used for drinking) but also affects the price of water, reducing its economic value.

In Figure 2 the location and type of wells surveyed per district were positioned on a geo-referenced map using GIS (Geographical Information Systems) software, and the wells used by the donkey cart water vendors (the traditional method of distribution in Somalia) were selected. It immed iately became obvious that only wells with an acceptable level of salinity were used, i.e. the ones located fairly far away from the seashore, the best being those situated close to or north of the 21 October road (about 6 km from the shore). Most of the vendors obtained their water from the wells between 3 and 5 km away.

From the data it is also possible to calculate the quantity of water abstracted every day from the aquifer and tentatively assess the overall income from the city's water trade. If we make the same assumptions as UNICEF [5 ] - i.e. one borehole produces 15 m3/hour, a hand-dug well equipped with a motorized pump about 6 m3/hour, a hand-dug well (with or without a hand-pump) about 1 m3/hour, with wells in use for approximately 10 hours a day - the estimated abstraction is currently close to 24,000 m3/day, and probably slightly higher if we bear in mind that some boreholes have an output close to 20 m3/hour and operate for more than 10 hours a day.

According to these results, the number of donkey cart loads distributed per day is close to 45,000, about 9,000 m3/day, probably with some 4,000 donkey carts in operation.

 Figure 1  

 Figure 2  

 The water business  

Many well-owners had to equip their wells with a pump to cope with their own needs, and also to improve their income. The water is sold to the donkey cart vendors at a fairly standard price, varying from 500 to 750 Somali shillings per 200-litre drum [6 ] , up to a maximum of 1,000 shillings. The water is then sold to the consumers according to the carting distance, at around 2,500 shillings and up to 3,000 shillings a drum. It is nevertheless distributed free to users living close to the wells, provided that the containers are not too big (between 5 and 10 litres).

Since the 1995 UNICEF survey mentioned above, the way in which wells are equipped has changed in keeping with the demand. The number of hand-dug wells equipped with motorized pumps has increased from 75 to 264. The number of wells used to supply the donkey cart vendors has also certainly risen, as shown in Figure 2, even if precise data on this increment are lacking. With the exception of the wells drilled or equipped by the ICRC, the majority have been equipped by their owners and are privately managed. Owners have invested resources to run their systems, and expect to earn enough to recover their investment and buy the necessary spares for maintenance and repair purposes. Data on Middle Shabelle from ETC [7 ] show that the minimum annual operation and maintenance cost for a borehole, calculated for a 50-m-deep borehole at a pumping rate of 12 m3/hour for a mean dai ly operation of 7 hours, is close to US$ 10,000. Total annual income is some US$ 17,000, with a net annual income of about US$ 7,000, which decreases the deeper the level at which the water is struck. This calculation assumes a private enterprise scenario and includes drilling costs, installation of the pumping set, interest rates on capital borrowing and operation and maintenance costs.

The traditional method of water distribution in Somalia. (ref. SO-D31/17)

By analogy, the calculation for hand-dug wells at a pumping rate of 4.2 m3/hour for 8 hours/day from a 50-m-deep well shows a lower annual income but a higher net unit income (income/hour/m3), the capital cost of digging and equipping the well being lower. This means that it is tempting for private-sector owners to dig and equip highly productive hand-dug wells instead of boreholes, provided that the demand for water is high. This was, and certainly still is, the case in Mogadishu and explains why there has been such an increase in the number of wells equipped with a motorized pump. Borehole drilling or equipment is left to humanitarian organizations like the ICRC, which can restrict their action to places in dire need of water, but with the risk that they may create conflicts with people who are already running commercial water supply systems. The intervention must be carefully analysed, balanced and community-oriented. For instance, leaders may be helped so that they can, in turn, care for displaced people, as in Sheick Aden, or hospitals may be prioritized, as in Benadir or Digfer, where water is essential and must be provided in large quantities. There is, of course, always a temptation for well-owners to obtain some support from any organization, to cover part of the capital cost or part of the operation and maintenance costs, in order to enhance their net unit income.

Increased harvesting is, of course, exacerbating the movement of the saline wedge inland and this may drastically raise the salinity, thus decreasing the value of the water for sale. For the time being, this situation is not giving any cause for alarm, and the 1997/98 floods have certainly helped to restore the previous boundaries. However, overabstractions may significantly increase the salinity of the water in neighbouring wells and reduce their commercial value for years, thereby raising tensions between owners.

 Table 1. Wells in Mogadishu and donkey cart loads per day (by district), 1997  

 Spread of disease  

In the four cholera epidemics that struck Somalia from 1994 to 1997, over 55,000 people fell ill and 2,100 died. Drinking water was assumed to be the major source of infection in 1994, when Vibrio cholera 01 was cultured from water samples collected from wells in cholera-affected areas. However, changes in vibrio transmission modes were suspected, as the vibrio was found in only one water sample taken in 1995 and in none of those collected in 1997.

Mogadishu has been hit several times by cholera epidemics and was the point of origin for the last three. The course of the 1997 epidemic was well documented, even if it did not reach the proportions of the previous ones [8 ] . Mogadishu suffered a further epidemic at the end of 1997 and beginning of 1998, certainly linked with excessive rainfall and extensive flooding assumed to be due to the El Niño phenomenon. The floods triggered population displacement, which in turn exacerbated contamination of the water supply. Thousands of people were admitted to several cholera treatment centres scattered throughout the city and on its outskirts.

To identify transmission modes, the WHO [9 ] and MSF carried out a study [10 ] in March and April 1997. Contaminated drinking water was still considered to play an important role in transmitting the disease, even if other factors also contributed, including the consumption of food and drink bought from street vendors and insufficient water for adequate hygiene practices, especially hand-washing. Despite huge efforts to chlorinate wells, the chlorine levels found in water samples taken from wells and household storage were far below those recommended, and many water samples showed faecal contamination. Water is also contaminated after collection from the source, but several samples of water from the same well were found to be clean one day and contaminated the next, probably because of the erratic way in which t hey were being disinfected with chlorine. Most of the samples from the wells were bacteriologically contaminated and with the onset of the epidemics most of the unprotected wells could be expected to become contaminated as well, putting an even greater proportion of the population at risk.

Mogadishu does not have a central waste-water evacuation network and most buildings dispose of their waste in septic tanks. Permanently displaced people use communal latrines, when available, or pit latrines. Percolation through preferential infiltration routes may lead to heavy contamination of the well water, particularly when these wells are poorly protected. This is the case for most wells in the displaced people's settlements located in the suburbs close to the seashore, where the higher salinity also allows the vibrio to survive for longer periods. [11 ]

The costs of epidemics are high, and not only in lives lost. Although treatment is fairly simple, the logistics required to provide rehydration fluids can be expensive. During the 1997 epidemic, the weekly supply through the ICRC to the treatment centres at Benadir Hospital and in Bermuda was about 420 litres of Ringer Lactate, which had to be brought in by plane together with other drugs and medical items. It can be assumed that the total weekly quantity supplied by all the organizations together was about twice that amount.

 Need for vigilance  

The uncontrolled privatization which has come about as an inevitable consequence of the chaos in Mogadishu triggers other dangers, as can be seen from the above data. Organized control of water use is more necessary then ever and this can only be done by an official service, usually the local water board.

In addition, the uncontrolled equipment of wells and boreholes with powerful pumps and the sinking of new wells has modified the position of the saline/soft water wedge, which has shifted inland, increasing the salinity of the water in a large number of wells. Such developments must be monitored and managed, as the future of the water supply to the whole city could be threatened. The uncontrolled equipment of water sources has further collateral consequences, as it may contribute to poor protection of the water sources and increase the spread of waterborne or water-related diseases.

Monitoring will help to manage the underground aquifer and also assist the various players in deciding on their priorities if they want to reduce the high risk of contamination of the water and the recurrent outbreaks of cholera within the city. During the period of unrest, several humanitarian organizations tried to maintain minimum monitoring of the water quality situation, but no measures can be enforced until law and order prevails. Guidelines can be established, but it will always be impossible to impose measures to manage private initiatives and avoid overharvesting practices or poor protection of this essential resource. Once the political situation is settled and centralized distribution resumes, the conflicting interests will have to be taken into account, and it may not be possible or even desirable to condemn the current practices. Only a precise study of the capacity of the aquifer will enable a new policy to be defined which will reconcile private interests with the common good.

Benadir (ICRC); Bermuda (ICRC); Afgoi (Fed.); Martini; Fortanini (MSF-B); Fairground (ACF); Medina (ACF); Bal'ad (Fed.)

Fed.: International Federation of Red Cross and Red Crescent Societies

MSF: Médecins sans frontières  

 Notes  

1. ACF: Action contre la faim  

2. UNICEF: United Nations Children's Fund

3. .A. Gibbs and Partners, Consulting Engineers, Source investigation for Mogadishu water supply expansion , Vol. 1, Technical Report, 1980.

4. P.G. Nembrini, R.G. Conti, " Water for Mogadishu: Water supply in a war-torn town " , Bull. angew. Geol. 2/2 , December 1997.

5. Results of the recent assessment of water sources in Mogadishu city , UNICEF, 1995.

6. US$ 1 = 7,000 Somali shillings as at March 1997.

7. Evaluation of water projects and assessment of water needs in Hiraan and Middle Shabelle regions , Environmental Technical Consultants, Kenya Consultants, November 1996.

8. M.T. Schick, R. Shoo, M. Neira, Cholera control in complex emergencies: Lessons from Somalia, 1994-1998 , WHO, personal communication.

9. WHO: World Health Organization

10. M.T. Schick, Cholera transmission in Mogadishu, Somalia, A case control study, WHO/MSF, Spain, June 1997.

11. R.G. Feachem et al. , " Sanitation and disease, Health aspects of excreta and waste water management " , World Bank studies in water supply and sanitation, John Wiley & Sons, 1983