The Impact of Refugees on the Environment and Appropriate Responses
by Gill Shepherd, Forestry Programme, ODI September 1995

For rural populations in developing countries, the natural environment is intimately linked to economic welfare. Populations are dependent on their surroundings for water, food, shelter and medicine. Refugee influxes intensify normal ‘green’ environmental problems – those associated with over-exploitation of rural natural resources due to poverty, rising populations, weak property rights and inappropriate management.

Refugee impact on the environment

Refugee settlements often occur in environmentally sensitive areas. In Africa, refugees have therefore usually been settled in semi-arid, agriculturally marginal areas, or (as in the case of the Rwandese in Zaire) near national parks or forest reserves. Refugee camps tend to be large for both logistical and political reasons. These large camps have a more negative impact on the environment than would be the case if several considerably smaller camps, catering for the same total numbers, were set up. Furthermore, refugees often have to stay in their countries of asylum for extended periods, and the impact on the environment around camps may be prolonged. In the case of unique sites, such as the Virunga National Park, Zaire, the environmental impact of refugees may be irreversible.

The impact of environmental deterioration on refugees and refugee-affected populations

The impact of environmental deterioration on the refugees themselves is intense. Low-quality water affects the health of large numbers of people, in a situation where there is a high risk of infectious diseases multiplying rapidly. Deforestation gradually forces women and children to walk further for wood, putting women in particular in danger of physical assault. Children may have to miss school to help; cooking time is shortened, and drinking water not boiled. Refugees may have to sell part of their food rations in order to obtain the fuel needed to cook the remainder, contributing to increased levels of malnutrition.

Host populations also experience a similar deterioration in the quality of their environment, so that normally available materials and supplies for construction, consumption and fuel are short, and prices for fuel and food in local markets rise. Tensions inevitably result, since host populations are currently made to bear many of the costs of the arrival of refugees in their area without immediate compensation.

Lessons from the environmental impact of the Rwandan refugee camps in Ngara and Kivu

How best to handle the environmental impact of refugees has been an issue brought under the microscope by the Rwanda crisis, because of the ecological importance of the areas into which many of the refugees have had to flee. UNHCR, in particular, is now in the process of reviewing its response.

Deforestation rates

In all, 524,000 people fled to Benaco in the Ngara area, now constituting the second largest ‘town’ in Tanzania, after Dar-es-Salaam. In the first six months to November 1994, tree resources within 5km of the four Ngara camps had been all but expended. By June 1995, the standard radius for getting fuel was 10km or more from the notional centre-point. These are very rapid fuelwood depletion rates. In north Kivu, 850,000 refugees in four camps are located within easy walking distance of the Virunga National Park, and many go there daily to gather fuelwood.

Per capita fuelwood consumption estimates

UNHCR’s early estimates of per capita consumption rates were excessively low. The best commissioned study (ERM 1994) [1] found such widely varying estimates of per capita fuelwood consumption (from 5.86kg/person/day down to 0.22kg/person/day), that it decided to conduct its own survey. It found a daily per capita consumption among local people in Ngara of 2.32kg, and near Karagwe, where wood is less abundant, of 1.18kg (average: 1.75kg). Figures in the camps were higher with a figure of 2.23kg in the northern cluster, and 3.06kg in the southern cluster (average 2.64kg).

Taking a mean figure of 2.6kg, a specific gravity figure of 850kg: 1m3 (a figure based on ERM/CARE calculations) and a figure of 524,000 refugees in the Benaco area, then 1,603m3 a day is needed, or 585,095m3 a year. Mean annual increment in the area is likely to be around 1m3 per hectare, so sustainable off-take ought to spread the refugee demand over 585,095ha or 5,851km2. If this area were a circle (the pattern of fuelwood use around population centres) with all 524,000 refugees at its centre, its radius would have to be at least 43km for sustainable off-take. Instead, all this use is concentrated within a radius of 5-10km. These figures take no account of the needs of local populations resident in the area before the refugees arrived.


There are a variety of short- and longer-term solutions to the need for fuelwood provision on this scale, which were proposed by the agencies involved in the camps, including UNHCR, and by consultants. These are set out in the table on page 9, in descending order of urgency and usefulness.

The most urgent need is to keep per capita consumption of fuel low, and to make fuel available from a wide area and variety of sources so that refugees do not irreversibly damage the area immediately surrounding camps.

On the demand side, the biggest single reducer of per capita consumption of fuel is the provision of food in a quick-cooking form. Maize in the form of maize-meal rather than whole dry ‘popcorn’ maize, for example, takes six to eight times longer to cook. It is theoretically possible to save fuel through the use of fuel-efficient stoves as well, though stove programmes have a depressingly unsuccessful history. A far simpler technology, which greatly reduces fuel-use and cooking time, is the provision of large flat saucepan lids to refugees for covering boiling food and water (high altitude has been a factor in high fuel consumption rates in the Rwanda refugee situation. Cooking times are much slower in highland areas because the boiling point of water is lower).

On the supply side, the simplest way of reducing the impact of refugees (though it is often not politically possible), is to set up a larger number of smaller camps, rather than a tiny number of large ones, so that fuelwood collection is automatically spread over a larger area. If this is impossible, then it is essential for agencies to identify natural stands of forest or plantations, and to organise the delivery of fuelwood to the camps. As time goes by, other sources of fuel may be identified as well. In Tanzania, for instance, both peat and papyrus reeds constitute such sources. A range of other options are inappropriate in this context for the reasons set out in the chart (kerosene, charcoal, briquettes, solar cookers, stoves). At the same time, important trees around the camps (along water courses, large shade trees, etc) can be marked with white paint as not available for felling.

A further area which needs early consideration, from the environmental point of view, is the need for poles and timber. Current refugee shelters provide polythene sheeting, but no wood supports. These have to be cut from the surrounding area. Nor have the agencies themselves been blameless. UNHCR (1994) notes that the implementing agencies cut tens of thousands of poles within easy trucking distance for pit latrines, medical clinics etc. Tents for official purposes, and tent-pole provision, ought to be part of the agencies’ commitment to a refugee situation.

In the longer run, there are three further actions to be taken. Firstly, in the refugee-affected areas, tree-planting programmes with local villagers and with remaining refugees should be a priority.

Secondly, and this is more for future refugee situations than for restoring the environment in current ones, databases for countries in Africa and elsewhere likely to be involved in a refugee crisis in due course, need to be set up to document areas of ample fuelwood resources (if any) available for future need, border areas of each country most unsuitable for the establishment of a refugee camps, and those which ought to be avoided at all costs.

UNHCR’s planned response for the future

In 1995, an internal Working Group was set up to consider UNHCR’s current policy and practice towards the environment. The Group’s final report (an internal document dated July 1995) groups its chief concerns about current shortcomings as follows:

  • Conceptual concerns: sound environmental management is viewed as subordinate to the material and social needs of the refugees, rather than as an integral part of those needs. Environmental rehabilitation is seen as the task of other organisations.
  • Technical concerns: no clear guidance has been developed to allow selection of the most appropriate technical options in each situation.
  • Institutional concerns: no clear comprehensive environmental policies and plans have been developed.
  • Operational concerns: Environmental considerations are not incorporated systematically into UNHCR’s refugee assistance programmes.

The Working Group proposes that in future, more effective environmental planning in the context of refugee camps should be a primary duty of UNHCR and host governments; that both refugees and local populations should be involved in environmental planning of any projects which are instituted; that there should be coordination with other UN agencies and international NGOs and that development funds should be committed where environmental damage is extensive.

All things being equal, prevention of environmental deterioration is preferable to cure, and in many contexts cheaper too, provided that environmental costs have been internalised by UNHCR. This means giving the environment the same weight as water, health and nutrition in mainstream programming. Operationally, it means that the environment must be given a higher priority at two key phases:

  • During the emergency phase, fundamental decisions such as site selection and layout should be taken with environmental considerations in mind, and the emergency team should incorporate these skills.
  • During the ‘care and maintenance’ phase, environmental components should be integrated into programming and implementation, and guidance given on how this is to be effected.

The Working Group sets out three options for the future:

  • The No-Change Option. UNHCR would continue with the ad hoc approach to the environment it has adopted in the past. The Working Group regards this option as irresponsible and suggests that it risks damaging UNHCR’s credibility with host governments.
  • The Two-Pronged Option. UNHCR would focus on only two issues: systematic provision of fuel to refugees; and an environmental rehabilitation programme aimed at attracting development assistance funds to rehabilitate refugee-affected areas.
  • The Mainstreaming Option. This option would lead to the following activities: preventative measures during the emergency phase such as environmentally sound site-planning; changed construction practices and improvements in the sustainability of refugee housing; participation of refugees and local community in planning; fuel supply; controlled fuelwood extraction from forests and reforestation; the introduction and dissemination of fuel-efficient stoves, mills to grind food grains; environmental training, education and awareness building and environmental rehabilitation after repatriation. A user-friendly environmental Source-Book would be produced to provide guidance in all these areas.

Comparing the latter two options, the working group concludes that mainstreaming not only meets the demands of internationally-agreed environmental criteria more closely and is the more responsible, but that it is also more cost-effective. While, in the light of findings from the Rwanda crisis, some of the details of the mainstreaming approach might need to be modified, this is plainly the best way forward.

Under either scenario, environmental expenditure should be budgeted in the General Programme, otherwise it will continue to be seen as a ‘frill’ rather than as a core part of UNHCR’s work.


[1] EMR 1994 Refugee inflow into Ngara and Karagwe districts, Tanzania. Environmental Impact Assessment November 1994. ERM for: Care International and the Overseas Development Administration.