April 24, 2014

Water losses: exceeding gains?

Published as Chapter 5 of The Social and Environmental Effects of Large Dams: Volume 1. Overview. Wadebridge Ecological Centre, Worthyvale Manor Camelford, Cornwall PL32 9TT, UK, 1984. By Edward Goldsmith and Nicholas Hildyard.

« previous chapter · contents · next chapter »


In hot, dry areas, the loss of water from a dam’s reservoir and from its accompanying channels can be staggering. Indeed, in many regions – the Colorado River Basin, for example – evaporation alone can result in such high water losses that, according to Professor William Ackermann, “increased storage reservoir development may reach a point beyond which the reduction of water yield . . . surpasses the possibility of increasing low-flow discharges from reservoir storage.” [1]

If that is so, then in many cases, one of the major benefits claimed for dams – namely the storage of water against times of low rainfall – is not as clear cut as it might appear. The more so when one adds to evaporation losses the loss of water through seepage and evapo-transpiration. Let us look at those three causes of water loss in more detail: how far do they undermine the benefits of having a permanent store of water on tap? And how – if at all – can they be avoided?

Back to top

Losses to evaporation

In arid and semi-arid countries, the loss of water to evaporation is – as Mohammed Kassas is quick to tell us – “inevitable”. [2] That evaporation can be considerable. In Guyana, for example, evaporation losses from open surfaces reach 139.7 cms. a year; in Burma, 114 to 152 cms; whilst, in some regions of India, it is common for up to 300 cms of water to be lost. [3]

Indeed, in certain areas of the world, evaporation rates can exceed local rainfall. Such areas are said to suffer from a ‘water deficiency’ – a deficiency which can be temporary, lasting only a month or two a year, or permanent. 56 million hectares of India’s 160 million hectares of cultivated land, for example, receive less than 750 mms of rainfall annually, and are thus classified by the Ministry of Agriculture as ‘drought affected’. [4]

Building vast reservoirs in areas where evaporation rates are high is, thus, to invite trouble. In Egypt, Lake Nasser loses a minimum of 15 billion cubic metres a year to evaporation – enough water, as Kassas points out, to irrigate two million acres of farmland. [5]

Although, in theory, such evaporation losses can be reduced by spreading a thin film of chemicals over the reservoir’s surface, that solution presents insuperable environmental risks: its cost is also exorbitant. It would seem, therefore, that we must resign ourselves to the inevitable loss of water from reservoirs to evaporation.

Back to top

Losses to transpiration: the problem of aquatic weeds

By creating a vast artificial reservoir, one inevitably transforms a terrestrial ecosystem into an aquatic one. Inevitably, too, that new aquatic environment will favour the growth of very different types of vegetation to that which existed before the reservoir was filled. Various phytoplankton and aquatic weeds will thus, sooner or later, make their appearance. Those aquatic weeds pose a particular problem where water losses are concerned for they increase dramatically the rate of transpiration. Indeed, several recent studies have shown that such water losses are 2, 3 or even 6 times higher in reservoirs covered in weeds than they are in open waters. [6]

Of these aquatic weeds, the most infamous is undoubtedly the water hyacinth (Eichornia crasspipes ). Not only does it establish itself very quickly but it has proved itself capable of thriving in almost every area into which it has been introduced: and, moreover, it spreads at a phenomenal rate. It has been estimated, for example, that a pond infested with one hectare of water hyacinth will produce up to 1.8 tonnes of dry mass a day. That rate of reproduction alone makes the weed almost impossible to control.

Despite the known effects of water weeds on increasing the rate of evapotranspiration, the planners of large dams rarely take their contribution to water losses into account. Yet it is now accepted that the invasion of reservoirs and their accompanying water channels is almost inevitable. Indeed, a recent study undertaken as part of UNESCO’s Man and the Biosphere (MAB) programme noted that canals and distribution systems “which are rich in organic matter and nutrients, but unsatisfactorily maintained, invariably are invaded and sometimes choked by dense growth of algae and aquatic weeds.” [7]

Since just about all canals and irrigation channels in hot dry areas are ‘rich in organic matter and nutrients’ (such as soil and vegetation from eroding banks, raw sewage from nearby settlements, and the run-off of artificial fertilisers) the problem of weeds would seem thoroughly predictable. The more so, given the low standards of maintenance which – as we shall see in Chapter 12 – prevail throughout irrigation schemes in the Third World.

Small wonder, then, that few reservoirs have escaped the menace of water weeds. In Surinam, for example, aquatic weeds invaded the Brokopondo reservoir almost as soon as it was filled in 1964. Within two years, more than 50 percent of the surface of the reservoir was infested, an area of some 410 square kilometres. [8] So, too, within three years of the Congo River being dammed, aquatic weeds had spread throughout 1,600 kilometres of the river. Egypt has also suffered from a plague of water weeds: thus, by 1974, a variety of aquatic weeds had infested more than 80 percent of the water courses fed by Lake Nasser and much of the River Nile was itself affected. Meanwhile, in the Sudan, an estimated 3,000 square kilometres are infested by water weeds.

Once established, water weeds are difficult – if not impossible – to eradicate. The Egyptian government, for instance, has dosed its canals and irrigation drains with massive quantities of herbicides – at unknown ecological cost – in its battle to rid itself of water weeds. Nonetheless, it is admitted that those spraying operations are little more than a holding operation:

“Because of the viability of the (water hyacinth) and its ability to reproduce sexually and asexually, it is believed that the best that can be hoped for is to strike a balance of control and utilisation methods that will contain the plant and reduce its effects to manageable proportions.” [9]

According to Al Ahram al Iqtisadi (22 June 1981), the Ministry of Irrigation accepts that evapotranspiration caused by aquatic weeds leads to a loss of water equivalent to 40 percent of the gain obtained by the High Dam at Aswan.


Back to top

Losses due to seepage and over-use of water

Water losses do not only occur through evaporation and transpiration: seepage from irrigation canals is also a major contributor to the problem. The extent of seepage varies considerably according to climate, soil, and the length and type of distribution system. Almost always, however, the extent of seepage is underestimated by the planners of large-scale irrigation schemes: indeed, in some cases, seepage rates have been underestimated by 100 percent. [10] The following examples are indicative of the extent of seepage and its contribution to overall water losses. Thus:

  • A 1967 study, undertaken by the International Commission on Irrigation and Drainage (ICID), reported that between 13.1 and 19.15 percent of the water transported along India’s Upper Bari Doab Canal was lost to seepage. In the plains of Uttar Pradesh and the Punjab, such losses were as high as 36 percent. [11]
  • In Egypt, during the summer, the main irrigation canals are estimated to lose some 1,500 million cubic metres of water through seepage every year – approximately 10 percent of the water available for irrigation. [12] Over the entire irrigation system, losses to conveyance, seepage and extravagance in water utilisation amount to 17 percent of the water delivered.
  • In many areas of the Middle East, anything between 10 and 70 percent of the total volume of water conveyed through irrigation canals is lost to seepage. [13]
  • In Pakistan, conveyance losses in the waterways fed by the massive Tarbala Dam are estimated to be ‘of a size corresponding to the content of the dam’s reservoir’ – some 4 to 7 million acre-feet per year. [14]

Other losses can occur from damaged dams and from inefficient operation. Water is thus wasted when too much is applied to the land: the plants cannot take it up and, consequently, a considerable proportion simply seeps into the subsoil. According to the US General Accounting Office, for example, half the water used in American agriculture is lost to over-watering and seepage. [15] In Taiwan, the loss is 30 percent: in other countries of South and South East Asia, it is said to range from 50 to 80 percent.

Indeed, the overall ‘water use efficiency’ of modern irrigation schemes is notoriously low. In the US, for example, it ranges from below 25 percent to 70 percent. [16] In Third World countries, the figures are even worse. Thus, the Indus River has a total of 142 million acre-feet of water available for use. 86 million acre-feet of that total are diverted into irrigation canals by way of dams and headworks. Of that figure, some 64 million acre-feet reach the farming areas via same 80,000 separate water courses, serving 3.2 million farms on 32 million acres. At the farm level, an estimated 26 million acre-feet are added by 125,000 private and public tube-wells.

But of the combined total of 90 million acre-feet, 35 are lost in the poorly maintained earth channels en route from the public distributary to the farmers’ fields. Thus, 55 million acre-feet or 1.7 acre-feet per household are left for the farmers, of which some 24 million acre-feet are lost due to field application losses. Effectively, therefore, only 30 million acre-feet out of a total supply of 170 million acre-feet are ever really utilised. [17]

Can the losses due to seepage and over-watering be avoided? In theory, the answer is ‘Yes’. The problem, however, lies in the vast expense involved. Ideally, for instance, closed tube conduits would be introduced, thus preventing both seepage and evaporation. Unfortunately, the cost is astronomical, particularly where large amounts of water are being discharged – which, according to Professor Holy, is the main reason why closed conduits have not been used on any appreciable scale in the world’s large irrigation systems. [18]

The second best method for avoiding losses from seepage is to line irrigation canals. The most efficient – and expensive – technique is to install a concrete lining. Cheaper methods include the use of plastics, oil films and even mirrors; and cheapest of all is a process known as ‘colmation’ which involves sealing porous soils with small soil grains which enter the pores during watering. [19]

It is by no means clear how effectively those cheaper methods work. What is certain, however, is that few irrigation canals are ever lined; one reason, undoubtedly, being the astronomical expense involved. Moreover, those schemes which have installed lined canals frequently suffer from poor management and technical foul-ups which drastically reduce the efficiency of the lined canals.

Over-watering can certainly be reduced by the introduction of sprinkler irrigation which makes possible very accurate water distribution. Once again, however, the cost of sprinkler irrigation is extremely high and, like lined canals, the sprinklers require a high level of maintenance and skill on the part of their operators. As we shall see in the next chapter, they also introduce other problems of their own making – not least the creation of a permanent, moist ecosystem which favours pest infestations.

« previous chapter · contents · next chapter »

Back to top


1. W. B. Langbein, ‘Water Yields and Reservoir Storage in the United States’, US Geological Survey Circ. 409, 1959. Quoted by William Ackermann in William Ackermann et. al. (Eds) Man-Made Lakes: Their Problems and Environmental Effects, American Geophysical Union, Washington, D.C. 1973. p.12.

2. Mohammed Kassas, ‘Environmental Aspects of Water Resource Development’, in Asit K. Biswas et. al. (Eds), Water Management in Developing Countries, Pergamon, Oxford, 1980, p.69.

3. Milos Holy, ‘The Efficiency of Irrigation Systems: An Overview’, in E. Barton Worthington (Ed) Arid Land Irrigation in Developing Countries: Environmental Problems and Effects, Pergamon, Oxford, 1977, p.347.

4. K. L. Rao, India’s Water Wealth, Orient Longman, New Delhi. Quoted by Malin Falkenmirth, ‘Water and Land: Interdependent but Manipulated Resources’, in Carl Widstrand (Ed), Water Conflicts and Research Priorities, Pergamon, Oxford, 1980, p.54.

5. Mohammed Kassas, op.cit. 1980, p.69.

6. W. T. Penfound and T. T. Earle, ‘The Biology of Water Hyacinth’, Ecological Monographs, Vol. 18, pp.447-72. Quoted by Mohammed Kassas, op.cit. 1980, p.69.

7. Gilbert F. White (Ed) Environmental Effects of Arid Land Irrigation in Developing Countries, MAB Technical Notes No. 8, UNESCO, paris, 1978, p.40.

8. Asit K. Biswas, ‘Environmental Implications of Water Development for Developing Countries’ in Carl Widstrand (Ed), The Social and Ecological Effects of Water Development in Developing Countries, Pergamon, Oxford, 1978, p.292.

9. Agricultural Research Council Staff Summary Report, Regional Workshop on Aquatic Weed Management and Utilisation in the Nile Valley, Khartoum, 1975. Quoted by John Waterbury, The Hydropolitics of the Nile Valley, Syracuse University Press, 1979, p.237.

10. Asit K. Biswas, ‘Water: A Perspective on Global Politics’, Water Resources Journal, December 1980. Quoted by Warrey Linney and Susan Harrison, Large Dams and the Developing World: Social and Environmental Costs and Benefits – A Look at Africa, Environment Liaison Centre, P.O.Box 72461, Nairobi, Kenya, 1981, p.43.

11. Milos Holy, op.cit. 1977, p.347.

12. Ibid, p.347.

13. M. M. Elgabaly in Asit. K. Biswas et. al. (Eds), Water Management for Arid Lands in Developing Countries, Pergamon, Oxford, 1980, p.62.

14. Gilbert L. Corey and W. Clyma, ‘Improving Farm Water Management in Pakistan’, Water Management Technical Report, no. 37, Colorado State University, 1975. Quoted by Carl Widtsrand (Ed), op.cit. 1980, p.88.

15. Bruce Stokes, Bread and Water: Growing Tomorrow’s Food, Worldwatch Institute, Unpublished manuscript, circa. 1980, Section 3, p.8. Other studies have not found the rate to be as high. Nonetheless, even the Department of the Interior acknowledges that 10 to 15 percent of irrigation water is wasted.

16. M. N. Langley and D. C. N. Robb, ‘Irrigation Water Use Efficiency’, Paper presented at the Fourth Technical Conference on Irrigation, Drainage and Flood Control, Phoenix, Arizona, March 27-29 1969. Quoted by Carl Widstrand, op.cit. 1980, p.88.

17. Max K. Lowdermilk, D. M. Freeman, A. C. Early, Social and Organizational Factors for Farm Irrigation Improvement: A Case Study, Colorado State University, 1977. Quoted by Carl Widstrand, op.cit. 1980, pp.88-89.

18. Milos Holy, op.cit. 1977, p.348.

19. Ibid, p.347.

  • Twitter
  • Facebook
  • Digg
  • Reddit
  • StumbleUpon
  • Diaspora
  • Identi.ca
  • email
  • Add to favorites
Back to top