When does a shortage—or excess—of water become a disaster rather than a temporary inconvenience? The answer may be obvious in prolonged drought or widespread flooding, but not so clear-cut in communities already at health risk from poverty, poor sanitation and limited coordination of health care and other services. Such communities may be unable to restore normality after a water emergency.
Water disasters can be sudden, as in flooding, or progressive and long lasting, as in drought. This affects both the way the disaster is identified and managed, and the timescale of the health effects. The health effects can be classified as:
The long-term health effects of water disasters are usually due to the lack of prompt restoration of public health services and interventions, with the resulting risk of epidemics and other ill health. The health effects of climate phenomena such as the El Niño Southern Oscillation (ENSO) also tend to develop gradually. The recent ENSO during 1997-8 was particularly severe in its effects: the associated natural disasters affected an estimated 160 million people (WHO, 1999). The slow time scale of droughts means that ill health may not be identified until the drought has persisted for months, affecting food supplies as well as the water needed to maintain health.
Whether a water emergency turns into a disaster depends on whether the community can take effective measures without external assistance. One working definition of a disaster is that it causes at least 10 deaths or results in an appeal for outside assistance (WHO 1999). Whatever the definition, disasters involving water are increasing. In recent decades there has been an increase in the numbers of deaths and the numbers of people affected by weather disasters such as droughts and floods (WHO 1999). Climate change appears to be responsible for at least some of this increase: and while global warming has been acknowledged, the term is misleading because it leaves out the key element of water. Floods are the second most frequent cause of natural disaster, after windstorms. The largest cause of deaths through natural disaster is drought, because of the associated severe food insecurity: examples include the high death rate in Sahelian people in Africa in the early 1970s and mid-1980s - and droughts are not only increasing, but also lasting longer.
Too much water: the health effects of floods
The early health effects of floods include death through drowning and accidents such as falls, electrocution and the effect of landslides. People lose their homes and often also lose their source of food and water. If the drinking water supply and sanitation system is already inadequate, flooding poses a further major health threat. Sanitation is a major problem in all flooded areas, as demonstrated by recent floods in Mexico, Ghana and Mozambique(Box 1). Industrial waste, such as engine oil and refuse dumps, adds to the health risks. In tropical countries, the floodwaters provide an ideal breeding ground for mosquitoes and an increased risk of diseases such as dengue, malaria and Rift Valley Fever. They also displace rodent populations, which may cause human outbreaks of leptospirosis and hantavirus infection. The combined effects of open sewage and reduced opportunities for good personal hygiene also favour the spread of infections causing diarrhoea, such as cholera and gastrointestinal viruses. Flooding in the horn of Africa in 1997, associated with ENSO, caused an upsurge in cholera deaths due to the lethal combination of damage to sanitation and contamination of water supplies (WHO, 1998). During flooding in Bolivia and Peru in the mid 1980s, increases in diarrhoeal diseases and acute respiratory diseases were recorded (WHO, 1999). Prolonged heavy rainfall causes less deaths than floods, but the infection risk is just as high in areas of poor sanitation: cholera showed a marked increase after heavy rains in Tanzania, Kenya, Guinea-Bissau, Chad and Somalia in 1997 (WHO 1998).
Box 1: Flood stories 2000: Mexico, Ghana and Mozambique
Mexico: In June 2000, heavy rains ruptured the wall of an open sewer in Mexico. This forced 6,000 people out of their homes in the low-income areas of Chalco valley. Although emergency shelters were available, many residents camped on their roofs to protect their homes from looting. Residents blamed the spill on the local authorities for failing to install piped sewerage.
In northern Ghana, clean drinking water became scarce three months after severe floods. Water sources had been polluted by tons of untreated human and industrial waste. More than 200 dams, wells and boreholes in the upper West Region were reported to be polluted with sewage and used engine oil. In addition, the floodwater had submerged refuse dumps due to rising river levels. The costs of the flooding raised dramatically due to the need to resettle people in other areas and to rehabilitate the polluted dams in the three northern regions.
The widespread floods in Mozambique in February 2000 made international headlines. Coupled with the lack of access to adequate sanitation and drinking water, nearly 800,000 people were put at increased risk of infectious diseases. The dam management was criticised, for example with claims that water had not been released in time, but it is possible that, with such overwhelming floods, better dam management would have had only slight effects. The key issues were to strengthen existing monitoring and early warning systems, to control settlement of flood plains and promote activities to limit human and economic casualties and a new flood is threatening in 2001.
As the examples in Box 1 show, flooding affects health indirectly through the widespread damage to the infrastructure of a community: its roads, buildings, equipment, drainage, sewerage and water supply systems. For example, during flooding in Peru in 1997/98, nearly a tenth of the health facilities were damaged: in Ecuador during the same period 2% of the hospitals were put out of action by a combination of flooding, mud, damage to sewerage systems and contamination of the drinking water supply. The mental distress of flooding may persist long after the floods have receded, because people have lost their homes, their livelihood and their confidence. The severe flooding in China in 1998 killed more than 5,500 people and left at least 21 million homeless (Kriner, 1999).
Too little water: droughts
“There it overtook me that I fell down for thirst, I was parched, my throat burned, and I said, “This is the taste of death”
Anonymous sufferer of drought in Ancient Egypt
According to an old French proverb, they that are thirsty drink silently. They also die silently: drought is a major cause of death worldwide and accounts for about half of the victims of natural disasters (WHO, 1999). Death is mainly due to lack of food and worsening of pre-existing malnutrition, but also through other pathways (Figure). In hot countries or during heat waves associated with drought, mortality may also be directly related to a combination of heat and water shortage.
Food production, such as the grain harvest, is particularly at risk in arid regions with a seasonal rainfall pattern. Loss of livestock is also a major problem during drought: arid regions, such as in SE Asia are particularly vulnerable (Box 2). Early warning systems such as climate forecasts can help communities to prepare for drought: other important early actions include coordinating the supply of water and any necessary rationing, as in the recent drought in Kenya (Box 3). While famine is the biggest killer in drought, the health effects include increased malaria and forest fires. In Venezuela and Colombia, malaria cases increased by more than a third following dry periods associated with ENSO; a fourfold increase in malaria was documented in south west Sri Lanka during ENSO (WHO, 1999). Drought has a major impact on infection because there is less water available for drinking and for personal hygiene: in addition to increasing diseases such as trachoma and scabies, people are more likely to risk drinking unsafe water and its load of infection. Studies have shown than in times of shortage, people tend to use water for cooking rather than for hygiene (WHO, 1999).
Box 2: Drought in SE Asia
In 2000, loss of livestock due to a serious drought in South Asia and the Near East caused the death of many people. In southern Afghanistan, the entire population (300,000 families) of the Registan desert fled when their water sources dried up. In Pakistan, the drought in Baluchistan and Sindh provinces were reported to be the worst in the country's history. This has led to indirect appeals to India to help battle the drought. Forty years ago Pakistan and India signed the Indus Treaty, to officially recognise that the Indus River is the main source of water for both countries. In Iran, 18 of the counties and 28 provinces also faced a severe drought. The Tigris and Euphrates rivers in Iraq also dropped to about 20% of their average flow.
Box 3: Drought in Kenya
Kenya experienced its worst drought in 40 years in 2000 and the President claimed it put 80% of the Kenyan population at risk. The World Food Programme (WFP) confirmed that 3.3 million people were seriously affected. Acute shortages of food, water and insecurity forced 15 primary schools in Kenya's north-central Samburu district to close, according to the local district education officer. Nairobi City Council had to ration water in the city from May.
With water already in short supply in many countries, it is not surprising that drought can also lead to water being used as a political tool, for example the water bargaining in Central Asian countries (Box 4). Water politics also affect attempts to relieve drought in countries in the throes of civil war, such as Ethiopia (Box 5).
Box: 4 Drought in Central Asia and water as a political tool
Shared water resources in the drought-affected nations of Central Asia have been used to bargain between countries. For example, in 1999 Kyrgyzstan succeeded in getting much needed coal from Kazakhstan after closing down water reservoirs. In 2000, Uzbekistan cut water supplies to Kazakhstan, citing non-payment of debt. Kazakhstan asked Tajikistan to release more water to Uzbekistan, in return for Uzbek electricity. The aim of this exchange was the hope that water flowing to Uzekistan would be likely to also flow to Kazakhstan. Meanwhile, a proposed Chinese water diversion project involving the Ertis (Irtysh) River poses more water problems for Kazahstan, as the river provides the drinking water for the industrial northeast region of the country.
Box 5: Water shortages due to drought and war in Ethiopia
A long-term drought as well as the effects of civil war has afflicted the Somali region of Ethiopia. The drought affected 8.3 million people. Heavy rainfall in April 2000 brought some relief, but the continuing conflict has restricted efforts to sort out the water supplies and sanitation in the region. The aid programme includes the installation of water reservoirs and digging or repairing wells. Emergency repairs have been conducted by UNICEF, associated with the distribution of water treatment chemicals and jerry cans throughout the hardest hit areas. ‘Donor fatigue’ and cynicism about the use of aid is a serious barrier to the international relief efforts to improve water supplies in the region.
Flooding after storms
Severe storms and hurricanes may cause water-related health problems through both excessive precipitation (heavy rains) and disruption of the water and sanitation infrastructure. Hurricane Mitch in 1998 was one of the worst recent natural disasters, estimated to have killed more people than any Atlantic hurricane in the last 200 years (NCDC, 1999). More than 3 million people were left homeless and at least 11,000 dead, with thousands of others reported missing. The storm started in late October, producing enormous amounts of precipitation as it travelled west, caused in part by the mountains of Central America. Floods and mudslides destroyed the entire infrastructure of Honduras and devastated parts of Nicaragua, Guatemala, Belize and El Salvador. As well as loss of many homes, approximately 4.5 million people in Honduras (75% of the population) were left without access to clean drinking water: around 1500 rural water mains were destroyed. Following this major natural disaster there were also critical shortages in medicine and food supplies. Fever and respiratory illness were widespread and in the following months increases in malaria, dengue and cholera were reported (NCDC 1999).
Reducing the effects of water disasters
Although the epidemiology of floods and droughts has been well studied (Noji, 1997), we still know more about the effects than how to prevent disasters occurring. Also, the death toll and immediate effects of sudden disasters tend to be given more emphasis than the gradually unfolding ill health following a disaster, or during a lasting disaster such as drought. Countries can reduce the effects of floods and droughts in two main ways: by being prepared for a disaster, and by disaster mitigation.
The aim of disaster preparation is to be able to reduce the immediate mortality and morbidity with a better prepared, well equipped service. The preparation includes early warning systems for seasonal changes in climate, the ENSO, and risk of flood or drought, such as electronic information systems and satellites that can provide information over large regions and continents. Separate systems are needed to cater for the agricultural sector, cities and people in rural or remote communities. The public health infrastructure is particularly important for the immediate measures needed and for public information on reducing the health risks. Being prepared also means thorough disaster contingency plans, covering emergency housing, repairs, replacement of essential equipment and protection of the most vulnerable people in the community: the sick, the very young and the old. Improvement of water supply and sanitation systems is an important way of reducing the effects of a water disaster: countries with a good infrastructure for drainage and disposal of human waste and adequate water supply facilities have far fewer direct health problems during water-related disasters Sanitary inspections are an important tool to assess the water supply and sanitation facilities and these should be conducted systematically. The logistics of the predicted need for health and social services also need to be laid down in advance, including early warning systems to detect health effects. Planning should be on a regional, national and international level and include planning for climate change: as global warming and its water effects will increase the frequency of water disasters. Finally, public information and education can serve two purposes in preparing for disasters: ensuring early warnings to communities at risk; and giving information about how to conserve water and keep it safe from contamination.
Once a disaster has occurred, or has been identified, all the measures in disaster preparedness will be needed and if not in place, outside help is probably needed. At the least, the mitigation efforts must include:
Both disaster preparedness and its mitigation require multisectoral cooperation and joint planning. Both need evaluation after a disaster to reduce the ill effects of later crises. While our world is never likely to be free of water disasters, there is much that can be done to minimise their health effects.
IRIN, 23 June 2000,
- The Nation, 25 May 2000
- More information
- Environment News Service, 3 Feb 2000,
African Eye News Service, 21 Feb 2000
- Floods and dam management in Mozambique, New Scientist, 22 April 2000
Noji E, The public health consequences of disasters. Oxford: Oxford University Press, 1997
WHO, Cholera in 1997. Weekly Epidemiological Record, 1998; 73: 201-8
WHO, El Niño and its health impacts. Weekly Epidemiological Record, 1998; 73: 148-152
WHO, El Niño and Health. Task Force on Climate and Health. Geneva: WHO, 1999. WHO/SDE/99.4