Cover Image
close this bookPriorities for Water Resources Allocation (NRI)
close this folderPriorities and conflicts in water resource development
View the documentPaper 1 Demographic trends: implications for the use of water
View the documentPaper 2 Fortunately there are substitutes for water: othetwise our hydropolitical futures would be impossible

Paper 1 Demographic trends: implications for the use of water

Michael Chisholm

Professor of Geography, University of Cambridge

Summary: World population is growing at about 1.nnually; even though the growth rate is falling, we can expect the total to rise from about 5.3 billion to over 8 billion in 2020. Virtually all of the increase will be located in the poorer countries, especially within the tropics, and furthermore most of the increase in the foreseeable future will be urban, not rural. Therefore, the management of water resources will become increasingly complex, especially in regions where precipitation is low, highly seasonal and/or very variable. To balance rural (agricultural) needs, urban needs and water requirements to maintain ecosystems will require the skills of the engineers and sensible water pricing policies, plus hard thought about the priorities for water use.

Pressures on water supplies

Several trends are putting increasing pressure on water supplies: population and income growth, the trend to spatial concentration of people, the widening range of uses to which water is put, and the need to disperse and to transport waste products of all kinds. To begin to consider these matters, something must be said about the growth and the location of population, at present and for the foreseeable future.

Table 1 presents some recent estimates for the total world population and its distribution by broad geographical region. Most of the population increase will take place in the less developed regions, and especially in Africa, Asia and Latin America. Although these estimates are subject to error, it is highly unlikely that over the next 30 years or so the broad picture conveyed by Table 1 will be altered, unless some catastrophic and unforeseen event occurs. Thus, we can say with fair confidence that between 1990 and 2020, the less developed countries will gain about 2.7 billion people - a gain well in excess of 50% - all of whom will need water for the food they eat, for personal use, for the industries in which they work, etc.

Table 1 Estimated and projected population (milllions)

The less developed countries are, practically by definition, relatively poor, as is shown in Table 2. If we compare Table 1 and Table 2, it is immediately clear that the bulk of future population growth in the less developed countries will be located in the low income nations which, in 1989, had an average GNP per head of only $ 330. If we assume, as is reasonable, that income per person will rise in these countries, at 1-2% per year it will take a long time for the absolute level of income to rise appreciably. At a growth rate of 2% annually, it would require 35 years for individual incomes to double.

Table 2 Population estimates and GNP per caput (actual and projected data)

A very large proportion of the world's existing population, and the great bulk of the prospective increase, is located within the tropics and therefore in hot climates, which range from the semi-desert and desert conditions of the African Sahel to the extremely wet conditions of Bangladesh. But even where there is an overall abundance of water, even excess, seasonal drought is common. Furthermore, much of the tropical area of the globe is characterised by extremely heavy precipitation, with the implication that the control, and hence use, of water is much less tractable than is the case in the temperate climate of Western Europe.

Every bit as important is the rapid urbanisation that is going on in the less developed countries of the world. Table 3 provides estimates for past and future years of the proportion of the population residing in urban areas. This table shows that there is a long-standing urbanisation trend which is manifest in all the major world regions, and that the most rapid urbanisation is oacurring in the less developed countries. Although these broad averages hide much diversity, the estimates available for 200 separate countries show that increases in urbanisation into the next century are to be expected in all but six countries - and all of these are already 100% urban (e.g. Bermuda, Cayman Islands and Singapore) (United Nations, 1991). So strong is the urbanisation trend that in the foreseeable future all of the increment in world population will live in towns. The United Nations (1991) estimates that the total rural population in the world will reach a peak in 2010 at 3.13 billion and will thereafter begin to decline slightly Table 4). This decline has been in evidence in the more developed regions since 1950; in the less developed regions, the absolute decline of rural population will occur after 2015. By major geographical area, it is only in Africa that the rural population will continue to rise in absolute terms throughout the period to 2020; even in Asia, the peak will be reached in 2005 and thereafter the absolute number of rural dwellers in that region will decline.

Table 3 Percentage of population residing in urban areas

Table 4 Rural population (millions)

Urban growth

One of the striking features of urban growth in the less developed countries is the emergence of very large cities - usually known as mega-cities. If we take a threshold of eight million for an agglomeration that can be called a mega-city, then in 1950 there were just two - London and New York. The number in the more developed regions of the world increased to six by 1990 and is expected to remain at that number until 2000. In contrast, by 1990 there were 14 in the less developed countries, and that number is expected to rise to 22 just 10 years later (United Nations, 1991). Table 5 gives details of the growth expected for those 20 cities which in 1990 had already reached eight million, from which several points quickly emerge. First, the very biggest cities are already huge, at something like 20 million, or just under two-fifths of the United Kingdom population and nearly three times the inhabitants of London. Second, the fastest growth will be in the less developed countries, and very rapid it will be. In addition, other cities, all in the less developed regions, are rapidly heading for mega-city status - Dhaka, Lagos and Bangkok, for example.

Table 5 Urban agglomerations with 8 million inhabitants or more in 1990

The poverty of many of the developing countries is reflected in the large scale of squatter, or illegal, settlement which characterises much of the explosive urbanisation in Africa, Asia and Latin America. This phenomenon has been documented elsewhere (Chisholm, 1992). The point of main relevance here is the enormity of the task which is faced merely to provide the most basic systems for delivering potable water and removing wastes; such provision is essential if minimum standards of private and public hygiene are to be achieved, standards which are essential if the health of the population is to be improved.

Priorities for water use

The key fact with which policy-makers will have to wrestle is the need to assess priorities for water use as between rural and urban needs. Some indication of the problems faced can be inferred from Table 6. At first sight, if only 15% or less of the available water is being used there is plenty of scope to expand supplies by suitable engineering works. However, supplies are unequally distributed both in space and time, so that it will never be possible to exploit the full potential. We know, also, that in substantial parts of Europe and North America in particular, demand already exceeds available supplies, while in some of the oil-rich states of the Middle East water usage actually exceeds local supplies - Qatar, Saudi Arabia and the United Arab Emirates for example - the difference being made up by desalination plants. Table 7 lists all the countries in the world in which water usage already equals or exceeds 20% of available supplies. While it is true that the greatest pressure is being felt in dry countries, it is notable that several important developed countries are included. Furthermore, if consumption were to double in the next few decades in countries such as India and China, their consumption of water would rise to well in excess of 30% of available supplies.

Table 6 The use of the world's freshwater in the 1980s: broad geographical regions

Table 7 The use of freshwater in the 1980s: selected countries

These aggregate consumption figures pay no attention to several factors of importance. First, they ignore medium- to long-term changes in precipitation which are attributable to natural oscillations in world weather systems and man-induced climatic change. Second, the main growth in demand will be geographically concentrated in urban areas, which implies major engineering works to gather and transport water, and then to cope with the resulting wastewater which, by virtue of its geographical concentration, should not be released back into rivers, lakes or the oceans without proper treatment. In Europe, 72% of sewage is treated in some degree, in the Mediterranean it is only 30%, in the Caribbean under 10%, and in the following regions the percentage is almost zero - Southeast Asia, South Asia, South Pacific, and West and Central Africa. The growth of the urban population will exacerbate this problem (World Resources Institute and United Nations Development Programme, 1992; for further details, see United Nations, 1989).

In addition, however, water plays a vital part in the maintenance of ecosystems and it is quite clear that were anything approaching the total 'available' water to be used, there would be serious and potentially devastating consequences in many situations. The ecosystems of rivers and their associated wetlands, estuaries and deltas are vital for many forms of wildlife and in addition provide important sources of gainful employment. It cannot possibly be assumed that all the water which ultimately 'escapes' to the sea in this manner is 'wasted' and should be otherwise used. It is a matter for investigation and judgement as to how much water must be retained in the natural ecosystems to ensure their maintenance. My own belief is that, too often, these limits have been exceeded with scant regard for the long-term consequences. One aspect of ecosystems that is often overlooked is the fact that, in addition to water, rivers carry dissolved minerals and sediments. Estimation of the latter is subject to very wide margins of error (Douglas, 1990; cf. World Resources Institute and United Nations Development programme, 1992). Douglas gives estimates for 26 rivers which together transport about 5.2 billion tonnes of sediment each year, an average of just over 200 million tonnes per river. Globally, the mean annual discharge of river sediment to the sea has been variously estimated between 13.5 billion tonnes and 24.0 billion tonnes. All of this is sediment which has entered the river systems and which, if not discharged to sea, will accumulate along the length of the river in one of two ways: by filling reservoirs; or by aggrading the course of the river and so building the river above the areas over which it flows until its banks cannot contain a major peak discharge. Meantime, in estuaries and delta areas, the absence of sediment input may lead to serious erosion, as in the case of the Nile delta (the problems being compounded by the penetration of saline water into the groundwater).

Renewable water resources

If we remember the considerable error of estimation involved, the volume of annually renewable water may be compared with the volume which is discharged to the sea by rivers (Table 8). The river discharges have been obtained by summing the totals for the individual rivers for which data are reported; although these include most of the major rivers, the aggregates so obtained represent minimum estimates. In a rough-and-ready way, we may say that the larger the river discharge is relative to total water availability, the greater is the scope for further river abstractions without seriously impeding the transport of sediment.

Table 8 Renewable water resources (in cubic kilometres)

Aggregate figures of the kind mentioned above do not give guidance as to what it is feasible to do in particular situations, but they do suggest that great care must be taken in the exploitation of water resources. However, they do show that the greatest pressures are likely to arise in supplying urban populations, on account of their growth, geographical concentration and the problem of dealing with effluents. Water will of course be needed to raise the crops to feed the urban dwellers, but the growth in demand for water will almost certainly be greatest in the urban areas. It will be essential to ensure that the short- and medium-term urban needs for water do not result in projects which are unsustainable in the long run and damage rural environments irretrievably.


Consequently, urgent thought must be given to the following issues:

· Ensuring that water is not used extravagantly.

· Ensuring that polluted waters are not returned to hydrological systems to carry infections and harmful pollutants to downstream populations.

· Seeking to re-cycle as much water as possible.

· Giving consideration to the uses to which water is put. Water-borne sewerage may be a luxury that cannot be afforded, and other means may have to be found to dispose of excrete safely.

These conclusions are broadly consistent with the view of the World Bank (1992) that the major problems of water supply relate to health and hygiene rather than increased agricultural usage. Therefore, given the demographic trends outlined, urban problems are going to gain in prominence, as is the need for the proper pricing of water as an increasingly scarce resource.


CHISHOLM, M. (1992) Demography and urbanisation. In: A Global Strategy for Housing in the Third Millennium. ALLEN, W. A. et al. (eds) Chapman and Hall, London (for the Royal Society).

DOUGLAS, I. (1990) Sediment transfer and siltation. In: The Earth as Transformed by Human Action. TURNER, B. L. et al. (eds) Cambridge University Press, Cambridge.

UNITED NATIONS (1991) World Urbanization Prospects 1990. UN, New York.

UNITED NATIONS CENTRE FOR HUMAN SETTLEMENT (1989) Urbanization and Sustainable Development in the Third World: an Unrecognized Global Issue. UN, Nairobi.

WORLD BANK (1992) World Development Report 1992. Development and the Environment, Oxford University Press, New York.



The discussion opened with consideration of whether it was feasible or desirable to resist increasing urbanisation. It was concluded that, in spite of some attempts to stem the flow from country to town, this trend would have to be accepted as a basis for planning. On the question of the returns to the use of water in different sectors, it was pointed out that there are gross distortions adopted in the pricing of water supplied by publicly owned institutions, and in the pricing of the crops grown under irrigation. This is true in the developed world as much as in the developing world.

It was concluded that only by more efficient use of water and reallocation between sectors could future needs be met. This would be assisted by realistic pricing and would require a radical change in the conventional wisdom on appropriate technology e.g. the use of non water-borne sanitation for large cities.