Emergency Water Supplies for Rwandan Refugees
Outline
In 1994, the civil war between Hutus and Tutsis in Rwanda caused 700 000 Hutu refugees to flee to Goma in E Zaire and a similar number to NW Tanzania. This Oxfam programme aimed to provide emergency water and sanitation facilities to the Rwandan refugees who where returning from the camps in Eastern Zaire and Tanzania towards the end of 1996. The whole programme, for Oxfam, ran from October 1996 to February 1997 and was financed mainly by the Belgium government. The whole repatriation programme was co-ordinated by UNHCR who also oversaw the watsan (water and sanitation) activities. In addition to Oxfam, the other implementing partners primarily included MSF, ICRC, COOPI, IRC and Trocaire.
This paper is based on a review of watsan activities undertaken by UNHCR and its implementing partners involved in the repatriation of the refugees from Goma, Zaire and complemented by additional experiences of working on water supplies for the subsequent repatriation of refugees from Tanzania.
Activities
The general strategy was to provide water points along the roadside at 5-8km intervals, which consisted of a storage tank of between 10 and 30m3 capacity and a tapstand with six or 12 taps. The interval was reduced in particularly arduous areas, such as steep terrain and increased in easy walking areas. Where possible, water points were connected and supplied by existing water systems, such as gravity-fed water from captured springs. In other areas, water points were supplied by water tankers provided by the implementing partners and filled from emergency and permanent water treatment works close to the route. The fleet of water tankers had a combined capacity of 90m3. This low capacity was acceptable due to the number of water points connected to existing systems. The tankers consisted of a mixture of cleaned petrol tankers and flatbed trucks with bladder tanks mounted on the back. Large pumps with a capacity of 30 litres/s were set up at water treatment plants for rapid loading while truck mounted pumps of 7 litres/s capacity were used to discharge at the water points. As the refugees were permanently moving, the water demand constantly varied at each water point. This resulted in some tankers being used as mobile water points. While existing opinion is that water tankers should only be used to transport water from a supply to a storage tank, this assumes that the location of the demand is static and, in general, the transport is from one supply to a central piped distribution network. Two tankers were fitted with a tap bar on the back of the truck with six or eight taps. Drivers were instructed to deliver water to the water points, but if they were prevented by the number of people on the road, they should stop and open the valve on the tap bar to supply water directly from the tanker until they were able to proceed down the road.
Successes
The scale of water supplies required on both refugee repatriations, from Zaire and Tanzania, were too large to be managed by one NGO. The success of each operation was only achieved through the close co-operation between all agencies to achieve a common goal. The review of the water and sanitation activities towards the end of the first repatriation assisted the planning and operation of the activities from the Tanzanian repatriation.
Constraints
Following the start of the repatriation from Goma, it rapidly became apparent that the Rwandan authorities would not allow large numbers of people to congregate in one area, resulting in the frequent forcible emptying of the transit centres (resting areas). These centres which had large water storage capacities (70-100m3) connected to a piped supply, often became redundant and a wasted resource.
Each family generally had one or two 20-litre jerry cans, which had been provided by NGOs in the camps during the previous two years. These jerry cans were carried by people of all ages, from five years old upwards. While this size jerry can may be ideal in refugee camps, where water only has to be carried a few hundred metres, they proved unsuitable, especially for children on a long walk. Due to the lack of a secondary road network in NW Rwanda, the water trucks were forced to travel along the main road being used by the refugees. During the main influx, the road became too congested with people to be used by vehicles which resulted in water points with lower capacities running dry by mid-afternoon.
There were areas where a large proportion of water was being taken by the local population, as a water point was located nearer to their home than their usual supply. It was clearly politically unacceptable to prevent these people from taking water, which resulted in a higher than expected demand being supplied in some areas.
Lessons learned
To ensure an efficient water-trucking operation, one person should control all the water trucks in a given area. This co-ordinator and all the truck drivers should each have a radio. Ideally, ersonnel at water points and treatment works should also be in radio contact with the coordinator. Where implementing agencies wish to participate in water-trucking operations, they should ensure that they procure not only suitable trucks, but also water pumps to enable the loading and discharging of water from the trucks. Due to the relative rapid changes in concentration of demand, it was often necessary to install water points within a couple of hours. Water tanks which can be rapidly set up, such as bladder or onion tanks, are the most suitable, unless more permanent tanks can be set up at critical points in advance.
Mobile water trucks were very effective. The rapid deployment of these mobile water points counteracted the negative effects of the low volumes of water, which they transported.Collapsible plastic water containers, which hold up to 5 or 10 litres, are already manufactured. They have a number of advantages over the ‘traditional’ 20 litre rigid jerry cans:
1. They were not as bulky as the 20 litre rigid containers which made them easier to carry by young children
2. Their volume was only that of the water they contained which also made carrying easier
3. As they were supplied collapsed a large number could be transported on each delivery trip. (For example, a Land Rover pick-up could carry 1000 containers in two packing crates rather than 80-100 rigid 20 litre cans – cf. airfreight costs as well).
These water containers were viewed as a vital commodity, even when empty. Scuffles and fights could break out where they were handed out. Distribution arrangements must be able to cope with the number of people passing at any time. The most effective points were near the top of a hill where people were spread out and it was easier to arrange a queuing system.
The collapsible containers were generally handed out empty. However, some were filled and oral rehydration salts (ORS) added. These containers were handed out to people who were identified as suffering from dehydration, ensuring that they had the means to continue drinking water.
Due to the number of water transfers and receptacles which were involved during the trucking of water and the cleanliness of the refugees water containers, it was discovered that a residual chlorine concentration of between 1.0 and 1.2 mg/l was necessary for water leaving the treatment plants to give a residual chlorine concentration of 0.2-0.5 mg/l in refugees’ jerry cans.
The high level of inter-agency co-operation in this operation highlighted the different equipment pecifications used by different agencies. It would be highly desirable if the major public health NGOs could agree common water equipment pecifications. Alternatively, increased supplies of ‘adaptation’ fittings, other than old tyre inner tubes, should be deployed to future major relief operations, where interagency collaboration is likely to occur.
