Salting the earth: the problem of salinisation

The causes of salinisation

All soils contain salt. That salt is the result of what geologists call 'weathering' - the natural chemical, biological and physical processes which lead to the gradual breakdown of rocks and other geological formations. As those rocks are gradually worn down, so they release their natural salts into the soil, generally to be dissolved in rainwater.

That water either percolates into the underlying groundwater or is washed away into streams and rivers. It follows that all water, like all soil, contains traces of salt. Indeed, even a fresh mountain stream will contain up to 50 parts per million (ppm) of salt - admittedly a minute amount compared with the 35,000 ppm found in seawater, but significant nonetheless. [1]

When the concentration of salts in soil reaches 0.5-1.0 percent, land becomes toxic to plant life. [2] In the dry tropics, that problem is particularly acute since there is not enough rainfall to flush out the salts which accumulate in the soil. Soils in those areas can thus have a natural salt content as high as twelve percent. Equally important, groundwater can contain salts at levels approaching the concentration of those in seawater. [3] Although such natural salt levels are the exception rather than the rule, the generally high salt burden of arid and semi-arid lands renders them particularly vulnerable to salinisation.

As Professor Victor Kovda of the Soviet Academy of Sciences is at pains to point out, groundwater provides "the main reserve and source of salts circulating in the soil profile" and, for that reason, it is essential that the water table under potentially saline soils should be kept at the appropriate level below the surface. [4]

If the water table is permitted to rise to within 2.5 metres of the surface, then the groundwaters are drawn upwards through capillary action - adding still further to their own salt burden on the way, by dissolving the salts in the soils near the surface. [5] In effect, the land becomes waterlogged with increasingly saline water.

Even before that water reaches the surface, it starts affecting crop yields by interfering with the capacity of plants to take up moisture and oxygen. Thus, in China's Shaanxi Province (where the impact of waterlogging on wheat and coffee production has been carefully recorded) it was found that normal yields could still be obtained when the water was 2 to 3 metres below the surface.

When it rose to 1 metre below the surface, and hence into the root zone, wheat yields fell to one fifth of the norm, and cotton yields to one half of the norm. When the water table rose to 0.5 metres and higher, wheat production fell to zero and cotton production fell to between one-fifth and one-twentieth of the norm. [6]

Worse still, once they approach the surface, the by now increasingly saline groundwaters quickly evaporate. The salts they contain are thus left behind to accumulate on the surface. Once again, the dry tropics are particularly vulnerable, evaporation rates in hot arid and semi-arid lands being four to five times higher than those in temperate areas. Under such conditions, it is not long before the whole area becomes covered with a white saline crust.

Where the salts in the water contain sodium or sodium bicarbonate - the result of sodium alumino-silicate minerals being released by the weathering of volcanic rocks - the destruction goes one step further. The lands become alkaline. As Kovda explains,

"The soil undergoes intensive hydrophilisation, losing its structure and permeability. An intensive cementation process gets under way in dry conditions. High alkalinity, cementation and impermeability in the aggregate leads to loss of fertility. The soil turns barren and is very difficult to reclaim". [7]

Such alkalinisation is already affecting parts of Northern India, Pakistan, Armenia, Afghanistan and Iran. Effectively, those lands are now dead forever.

Irrigation and salinisation: the intimate connection

If arid lands are not to become salinised, it is clearly essential to maintain the 'water-salt balance' of the soil. That is to say, the amount of water leaving the soil must be at least equal to the amount entering it. The water should not be allowed to accumulate. So, too, with the salt balance: salt must not be added to the soil - for example, by using irrigation water with a high salt content - unless an equal amount of salt can be flushed out of the land.

Irrigation schemes throw that delicate water-sale balance dangerously out of kilter. Firstly, perennial irrigation invariably raises the water table. According to Professor Gilbert White of the University of Colorado at Boulder, for instance, there are "numerous cases" throughout the world where the water table under irrigated land "has risen within ten years from about 25-30 metres below the soil surface up to 1-2 metres depth". [8] Indeed, in some areas, groundwater tables are rising at a rate of 3 to 5 metres a year.

That rise in groundwater levels is caused, primarily, by the water lost through seepage from irrigation channels. Such seepage losses can be considerable - in some instances, up to 60 percent of the water transported through the canals is lost to seepage. Where the irrigation water is provided by large-scale dams, the problem is compounded by the seepage of water from the dam's reservoir: in some cases, such seepage has raised the level of ground-waters up to 20 kilometres away. Finally, the over-use of irrigation water (a problem common to irrigation schemes the world over) helps to raise the water table and hence further increase waterlogging. As waterlogging sets in, so the inevitable process of salinisation begins.

Secondly, irrigation adds directly to the salt load of soils through increasing the rate of 'evapo-transpiration'. As plants 'breathe', so much of the water taken up by their roots is lost through their leaves to the atmosphere - a process known as 'transpiration'. So, too, a large proportion of the water applied to plants is lost to direct evaporation. Because it is practically impossible to quantify how much water is lost to transpiration and how much to evaporation, the two causes of water loss are treated as a single phenomenon, known technically as 'evapo-transpiration'.

Where land is irrigated, the losses due to evapo-transpiration are particularly high. Not only does irrigation increase the extent of vegetative cover - and hence the rates of transpiration - but it also requires water to be spread thinly over a wide area, thus raising direct evaporation losses.

The inevitable result of high evapo-transpiration is that the natural salt in water becomes concentrated in the soil. On that score, the research of Arthur Pillsbury, Professor of Engineering and (until he retired) Director of the Water Resources Centre at the University of California, Los Angeles, is particularly relevant. Writing in Scientific American, Pillsbury estimates that three-quarters of the water applied each year to irrigated land in the US is lost to evapo-transpiration.

"If, as seems reasonable, the average annual amount of water applied in irrigation in the Western US is equivalent to 3 feet covering the area cultivated, about 120 million acre-feet of water is applied annually to some 40 million acres of land. Roughly 90 million acre-feet of the total volume is lost by evapo-transpiration. The remaining 30 million acre-feet holds essentially all the original salts: a four-fold concentration."

Such water frequently contains more than 2,000 ppm of salt. [9]

That problem is exacerbated when irrigation water is drawn directly from rivers or from the reservoirs impounded by large dams. Thus, in the US, evaporation at Lake Mead (behind the Hoover Dam) and at lake Powell (behind the Glen Canyon Dam) is reported to have increased the salinity of the Colorado River by 100 milligrams per litre.

So, too, Waterbury reports that high evaporation rates at the Aswan High Dam reservoir "have led to 10 percent increases in salinity: that is, water entering the reservoir has about 200ppm, and when it leaves, 220ppm".

He goes on to note:

"Because Upper and Middle Egyptian lands drain back into the main Nile, salinity around Cairo and in the Delta is in excess of 300 ppm. In itself, this is no cause for alarm, but agricultural intensification in Egypt and the Sudan cannot fail to aggravate the problem. Moreover, in developing new water resources from the equatorial lakes and the Jongiei scheme, the White Nile, with a higher salt content than the Blue Nile, will figure prominently in downstream discharge." [10]

That rising salt burden in the waters of reservoirs and rivers leads Kovda to warn that irrigation water is itself now a significant factor in the spread of salinisation.

"Precipitation water has 10 to 30 mg/l, and sometimes 50 mg/l of salts and the water may still be considered practically fresh. The best irrigation water from large rivers contains 200 to 500 mg/l of salts. Supplying 10,000 cubic metres of water on 1 hectare of land during the irrigation season deposits 2 to 5 tons/ha of salts in soils. After 10-20 years of irrigation, this amount becomes enormous - amounting to dozens and even hundreds of tons per hectare." [11]

The extent of the problem

The devastation caused by waterlogging and salinisation is hard to quantify - not least because there are considerable differences of opinion as to when land should be classified as 'saline'. The Pakistan Department of Power and Irrigation, for example, suggests that land should be regarded as 'saline' when crop yields have been reduced by a fifth or more.

Nonetheless, various figures have been advanced. The FAO, for instance, estimates that at least 50 percent of the world's irrigated land now suffers from salinisation. [12] Others put the figure even higher. Thus, Victor Kovda argues that, worldwide, salinisation affects 60 to 80 percent of irrigated land - between 1 and 1..5 million hectares succumbing every year. Significantly, much of that land is "in irrigated crop-lands of high potential production". [13]

In Pakistan, 25 million acres of the 37 million acres under irrigation are estimated to be salinised, waterlogged or both. [14] Of that land, 5 million acres are classified as "severely affected with salinity; 10 million as suffering 'patchy salinity'; and 10 million as being 'poorly drained' ". [15] Overall, 23 percent of the country's land suffers to varying degrees from salinisation or waterlogging - that figure reaching 80 percent in the Punjab.

In the lower Indus, concentrations of salt in the groundwater have been found to reach 30,000 ppm - almost as salty as seawater. Indeed, a recent survey reported that the water from 18 percent of the tube-wells in the area was "unfit for use"; that 76 percent of the wells produced water which, if used, might "salinise the soil profile to a depth of 6 feet within 12 years"; and that only 6 percent had water which could be classified as being "of excellent quality". [16]

All told, an estimated 100,000 acres are lost annually to waterlogging and salinisation in Pakistan - more than 100 hectares a day. [17] This was in fact, the figure provided locally to the Revelle Panel appointed by President Kennedy to study the problem of irrigation and salinisation in what was then West Pakistan. Since, at that time (1962), 23 million acres of agricultural land was irrigated, nearly 5 percent of it was going out of production each year.

The Panel, unfortunately, played down the problem and reported that "between 50,000 and 100,000 additional acres are being affected each year, some of which are passing out of crop production". [17a]

Of the area earmarked to receive water via China's giant Yangtse Diversion scheme, 2.7 million hectares already suffer from salinisation. "Slightly and moderately saline soils prevail on 73.7 percent of the affected area", Guo Huancheng and Xu Zhikang of the Institute of Geography of the Academy Sinica, told a recent conference held to discuss the diversion scheme.

"Here the salt concentration generally ranges between 0.1 percent and 0.7 percent, so it is still possible to grow crops. The remaining 26.3 percent of the affected area has a salt concentration exceeding 1 percent and a seedling retention rate below 30 percent. The land is used mostly for livestock and forestry. The extensive saline area in the region is not only unfavourable to current agricultural production, but also an important problem which must be taken into account in considering a south-to-north water transfer." [18]

Xu Yuexian and Hong Jilian, also of the Institute of Geography, expressed similar concern about the extent of salinisation in the proposed transfer region. Although salinisation levels in the area had fluctuated dramatically over the last 30 years, they told the conference, there was little room for complacency. Overall, the amount of saline land was increasing and, moreover, much of the increase had taken place during the late 1970s - largely, they suggested, as a result of a rise in water tables alongside those rivers which have recently been dammed. In addition to the 2.7 million hectares already affected in the area, a further 4.7 million hectares consists of potentially saline soil "which is most vulnerable to secondary salinisation if affected by detrimental factors". [19]

In Egypt, the problems of salinisation and waterlogging have been described as "grave". John Waterbury writes,

"A few years ago, an FAO study contended that 35 percent of Egypt's cultivated surface is afflicted by salinity and 90 percent by waterlogging. A USAID mission reported in 1976 that 4.2 million feddans (1 feddan=1.038 acres) were undergoing slight to severe effects from inadequate drainage, and unless something were done, all would be severely affected." [20]

Waterlogging alone is estimated to have reduced agricultural productivity by at least 30 percent, although it is claimed (perhaps optimistically) that drainage will restore productivity.

More than 50 percent of the 3.6 million hectares under irrigation in Iraq suffer from salinisation and waterlogging. [21] Worst affected are the middle and lower Rafidain Plains. Indeed, Erick Eckholm - at the time a researcher with the Worldwatch Institute - reports that vast areas of South Iraq now "glisten like fields of freshly fallen snow". [22]

500,000 acres in Syria - half of the country's irrigated land - are waterlogged or salinised. According to M. M. El Gabaly,

"Due to the aridity of the climate, with evaporation exceeding precipitation in many locations, it is estimated that 70 percent of the soils put under irrigation are potentially saline."

Nonetheless, plans are afoot to irrigate a further 1.5 million acres of part of the giant Euphrates Project. [23] Annual crop losses due to salinity and water-logging in the Euphrates Valley alone already amount to $300 million.

In Iran, 15 percent of the irrigated land degree by waterlogging, salinity and alkalinity. 16.8 million hectares of arable land, 7.3 million are estimated to be saline, and 8.2 million are waterlogged. [24]

In India, the amount of land devastated by water and salt has been variously estimated at between 6 million and 10 million hectares - almost a quarter of the 43 million hectares under irrigation. In Madhya Pradesh, affected areas are referred to as 'wet deserts'. [25]

For the Near East as a whole, El Gabaly warns: "In all countries of the region, without exception, salinity is of prime concern in agricultural development". [26] Elsewhere, it is estimated that more than 70 percent of the 30 million hectares of irrigated land in Egypt, Iran, Iraq and Pakistan is now "moderately to severely affected" by salinisation.

In the US, Jan Van Schilgaarde, Director of the US Salinity Laboratory, considers that 25 to 35 percent of the country's irrigated land suffers from salinity - and that the problem is getting worse. "Today, about 400,000 acres of irrigated farmland in the San Joaquin Valley are affected by high, brackish water tables". [27] If no remedial measures are taken, the Valley could lose over a million acres of farmland in the next hundred years.

Worldwide, the annual losses to salinisation and waterlogging are simply staggering. In a survey conducted for the UN Water Conference, Malin Falkenmark and Gunnar Lindh estimate that between 200,000 and 300,000 hectares of irrigated land are taken out of production every year due to the ravages of water and salt. Harold Dregne of Texas Tech puts the figure even higher - at 500,000 hectares. [28]

For its part, the FAO does not propose a precise figure: nonetheless, it admits that "several hundred thousand hectares are abandoned annually as a result of salinisation". [29] Indeed, according to one recent study, as much irrigated land is now being taken out of production due to waterlogging and salinisation as new irrigation schemes are bringing into production. If this is correct, then the rate of salinisation is very much worse than even Dregne's figures suggest.

Can salinisation and waterlogging be avoided?

It is rare indeed to find irrigated areas which have avoided the ecological devastation normally associated with large-scale perennial irrigation schemes. Only in those areas where land is well-drained - as in West Texas, for example, where sub-soils are particularly permeable - can the surplus waters drain away sufficiently fast to prevent waterlogging.

Such soils, however, are exceptional. Indeed, in almost all the arid areas of the world where irrigation is practised - the valleys of the Tigris and Euphrates, the Helmand Valley of Afghanistan or the Imperial and San Joaquin Valleys of California, for example - the sub-soil is relatively impermeable. Down drainage is thus deficient. Inevitably, the surplus waters accumulate; rise; reach the roots of the crops; and, eventually, make their way to the surface where they evaporate, leaving behind their deadly burden of salts.

According to Kovda,

"during many centuries and even millennia, only areas having a free outflow of groundwaters as in Tashkent and Samarkand have not undergone salinisation or waterlogging." [30]

Indeed, irrigated land has been thus degraded with such regularity that Kovda sees "increasing salinity in irrigated soils of arid lands" as being "practically universal". [31] Aloys Michel - whose experience of irrigation projects the world over is vast - goes a step further. Thus he writes:

"Waterlogging and salinity, or both problems, will inevitably arise in all but the truly exceptional surface-water irrigation schemes." [32]

Despite that historical experience, however, governments the world over, still remain committed to extending the amount of land under irrigation. Those who argue that such an extension can only serve to degrade the environment further, are assured that the damage of the past will not be repeated in the future.

Indeed, the promoters of large-scale irrigation insist that salinisation and waterlogging are not the fault of perennial irrigation per se: on the contrary, they claim, they result from technical and administrative 'mistakes' which can easily be corrected in the future. But, can we be so sanguine? Is it really possible to avoid waterlogging and salinisation in lands watered by large-scale irrigation schemes? And if so, how?

One course open to us is to line irrigation canals, thereby reducing seepage. Unfortunately, the cost of lining canals is exorbitant - one reason why lining is rarely installed. Moreover, lining irrigation canals is by no means a certain method of reducing all seepage. For one thing, the lining does not last indefinitely: for another, its efficiency is largely dependent on regular and thorough maintenance.

As we shall see, the standard of maintenance in existing irrigation schemes is extremely low - and there are few signs that it will improve, largely because the problem is social rather than technical in origin. To fight salinisation and waterlogging with technology which requires - indeed demands - regular maintenance in order to work properly would thus seem foolhardy in the extreme.

Another strategy is to dig tube wells in order to pump out ground-water and thus lower the water table. A large number of such wells have been sunk in Pakistan, China and elsewhere. Once again, however, their high cost has often proved an insurmountable problem. In addition, tube wells have a short life span and, with lined irrigation canals, they share the intractable problem of requiring regular maintenance if they are to work.

Moreover, where they have been used, as in China, their successes in bringing down the water table appear short-lived. Thus, in the three provinces of Hebei, Henan and Shandong, tube wells - together with improved drainage - helped to reduce the total area of saline land from 1.9 million hectares in the mid-1950s to 1.4 million hectares in the mid-1970s. Over half a million wells were sunk in Hebei alone but by the end of the 1970s, the area of saline land in the three provinces had begun once again to increase. By 1979, 1.9 million hectares were officially classified as saline.

Ironically, the excessive use of tube wells can, in itself, exacerbate the problem of salinisation. If the salt balance of underground aquifers is to be maintained, then an aquifer must not only have an inflow of fresh water to dilute its salts and replenish its waters, but also an outflow through which the salty water can be evacuated.

Where the water table is lowered too far (as has happened in certain parts of the Southwestern United States) then there is a tendency for aquifers to become so depleted that they are cut off from both their points of inflow and their points of outflow. In such circumstances, the aquifers become closed basins within which used irrigation water simply accumulates.

A third suggested method for reducing the rate of salinisation is the introduction of 'overhead sprinklers'. Such sprinklers are said to minimise water use and thus rule out over-watering as a cause of waterlogging. Nonetheless, they are not without their problems. Discussing the use of sprinklers in Algeria for instance, Taghi Farvar of the Centre for the Biology of Natural Systems, told the 1969 'Careless Technology' conference:

"Much of the water immediately evaporates when it is sprinkled even before reaching the ground. So you have an increase in the salinity of the water by the time it hits the earth, then the rest of it evaporates and salinises the soil." [33]

Although it has been suggested that the problem can be overcome by irrigating during the night (as is done in Israel) Farvar does not consider that this would make much difference in very hot countries. In Algeria, for instance "the night is very often quite hot, particularly when the sirocco blows". In addition to the technical difficulties posed by evaporation, sprinkler irrigation has been found to increase pest outbreaks.

Perhaps the most effective means of combating waterlogging and salinisation however, is the building of drains to remove the excess water. Indeed, irrigation without drainage is now generally accepted to be little more than a recipe for ecological disaster. Aloys Michel, for example, argues that "drainage must go hand-in-hand with irrigation". Indeed "the provision of an artificial drainage system is an inescapable concomitant of providing an artificial irrigation system". [34]

Likewise, Erick Eckholm of the Worldwatch Institute sees irrigation and drainage as "inseparable components of a single system". [35]

Although the essential nature of drainage is now recognised by most of the leading authorities on salinisation, governments throughout the world are still building irrigation schemes without first installing drains. Thus:

Drainage was never installed in the various irrigation projects which have been set up in the Chambal area of Rajhastan and Madhya Pradesh in India. Since the soils of the area consist of heavy clay, waterlogging has quickly developed. In India as a whole, according to M. C. Chaturvedi, only a very small proportion of irrigated land is properly drained. [36]

In Pakistan - where the problems of waterlogging and salinisation are particularly acute - the government announced elaborate plans for digging tube wells and installing drains throughout the country's irrigated lands. Some tube wells were sunk in the Indus Valley: but the programme as a whole never materialised.

In New South Wales, Australia, tiled drains have indeed been installed - but only for those irrigated lands under intensive horticulture. Other irrigated crops go undrained. [37]

Even in America's San Joaquin Valley, some 60 percent of farmers do not have adequate drainage facilities. Instead of tackling the problem of salinisation at source, many farmers have opted for growing shallow-rooting salt-tolerant crops. In the long term, however, those crops can only exacerbate the problem still further: salt-tolerant they may be, but they still require irrigation - and the saline groundwaters of the region are thus continuing their inexorable rise.

In some areas, salinisation is so far advanced that farmers have simply taken their salt-encrusted lands out of production and intensified cultivation and irrigation in their remaining fields in an attempt to make up their losses. Other farmers have turned their most saline fields over to 'evaporation basins' where the water from those lands which are drained can be dumped.

Why no drainage

Undoubtedly, one reason for the reluctance of governments to install drainage lies in the expense involved. The UN Food and Agriculture Organisation (FAO), for instance, estimates that the installation of effective drainage costs between $200 and $1,000 per hectare of land irrigated. [38] At present, the total amount of land under irrigation in the world is perhaps a little over 200 million hectares and is said to be increasing at the rate of 2.9 percent a year - although few see that rate continuing. In effect, about 5.8 million hectares of land are brought under irrigation every year.

To install drainage for all that land at $200 per hectare would cost $1,160 million: at $1,000 per hectare, the cost would be an astronomical $5,800 million. Indeed, in 1978 Egypt's Minister of Irrigation estimated the cost of installing tiled drains in the heavily salinised lands of Upper Egypt and the Delta area would reach some 700 million Egyptian pounds by 1985 - nearly twice the original cost of the Aswan Dam itself. [39] This represents only part of a long-term programme which will not be fully implemented until the year 2000.

Since most of the work would have to be carried out in the Third World, it is unlikely that the vast funds needed to finance the programme will be available. The trouble is that, in order to make a proposed dam project appear viable, the cost is invariably underestimated. To do this means leaving out what (to the uninitiated) are not obviously essential components of the system - drainage, for example. Indeed, if the true cost of drainage were taken into account, then many water projects would cease to appear in any way economic.

As a result, when building work nears its completion, there is often little money left to spend on what to politicians, economists and engineers often seems an irrelevance. Carl Widstrand makes the point forcefully:

"Costs for drainage are always under-estimated and when irrigation schemes overrun their budgets - which they always do - there is little money or interest left for drainage." [40]

"The legacy of this continued defiance of reality is a stupendous loss of global agricultural output." [41]

In succumbing to that "defiance of reality", governments and their advisors have been abetted by what has come to be known as 'the anti-drainage lobby'. Thus, when Egypt announced in 1958 that it intended to introduce main and field drains to all of Egypt's cultivated land, it was persuaded not to do so.

According to Azim Abulata, the programme was opposed on the grounds that "the cessation of the flood phenomenon due to the storage by the High Dam would lead to a fall in water levels from Aswan to Cairo", which would thus improve the natural flow of drainage water back to the riverbed. It was also claimed that the 'low-lift' irrigation system to be employed "would permit excessive use of water thus reducing the quantity of drainage water". [42]

This reminds one of the reaction among promoters of intensive agriculture in the South Great Plans of North America to the erosion and desertification of the late 1930s. Farmers were assured that "erosion actually improved farmland by exposing fresh layers of soil". [42a]

All this was sheer wishful thinking. Indeed, if one wishes to look for the real reason why the government was willing to abandon the programme, one need seek no further than the estimates of the cost of the project. In fact, it was admitted that funds were not available to cover the massive expense of draining all the lands to be brought under perennial irrigation.

Perhaps still more to the point is Waterbury's terse comment: "No-one ever built a monument to themselves by installing tile drains". [43] Given the essentially political motives for building the Aswan Dam, that thought must have weighed heavily with Nasser and his fellow officers in the government.

The anti-drainage lobby has not only been at work in Egypt. According to I. P. Gerasimov of the Institute of Geography at Moscow University, for example, that same lobby was largely responsible for the failure to install drainage in the irrigation schemes of the Golodnaya Steppe, southwest of Tashkent. [44]

So too, Aloys Michel notes the hand of the lobby in the Irrigation Branch of the Punjab Public Works Department during the heyday of the British Raj. The department's experts insisted that drains were unnecessary in the area because lowering the water table would prove detrimental to the working of hundreds of Persian wheels operating in shallow wells. Moreover, they claimed, high water tables permitted the regeneration of water supplies by seepage during the dry season. [45]

More recently, Victor Kovda has singled out the US Salinity Laboratory in Riverside for similar criticism. Pointing to the successes of the Soviet Union in ameliorating salinised lands in Azerbaidzhan, Uzbekistan, Tadzhikistan and Turkmenia, he berates the laboratory for ignoring the Soviet experience.

"For 25 to 30 years, this laboratory has rejected indisputable conclusions of the geochemical theory of salt accumulation. It has underestimated the importance of groundwater level and mineralisation of the groundwater and properties of saline soils. The concept of critical level, critical regime and critical mineralisation of groundwater is either rejected or ignored. Secondary salinisation of soils is attributed mostly to salts of irrigation water, which, in fact, are of secondary importance. The necessity of leaching salts from salty soils independently of seasonal watering and the necessity of desalinisation of salty groundwaters are rejected or even misunderstood. The importance of ionic composition of the soluble soils is ignored and a 'cult' of electroconductivity as the way to study soil salinity has been followed ... Publications of the Riverside Laboratory by-pass all these problems, are actually cultivating the idea of a permanent domination of downward flux and over-irrigation in order to suppress ascending capillary solutions and are supporting the anti-drainage assumptions of the advocates of cheap 'drainageless irrigation', which, in fact, leads to waterlogging and salinisation." [46]

Indeed, David Sheridan, writing in Environment, warns that one million acres of the San Joaquin Valley could become unproductive by 2080 unless sub-surface drainage systems are installed. [47]

The activities of the 'anti-drainage' lobbies in other countries can only lead to similar results unless they are successfully countered.

Salinisation: the historical experience

In reality, it has been known for a very long time that large-scale water development schemes (such as the building of dams and canals) will, in hot countries, lead to the salinisation of the soil. Unfortunately, however, we seem incapable of learning from the experience.

To make the point, let us briefly consider the Indian experience with large-scale water schemes during the 19th century. Our information is derived from Elizabeth Whitcombe's seminal book, Agrarian Conditions in Northern India.

After the Indian Mutiny, there began in India a period of large-scale economic development, an important constituent of which was the construction of the Lower Ganges canal. It opened in March 1874 and commanded an area of 375,800 acres. At the same time, however, canals were built commanding smaller acreage. Those water development schemes gave rise to precisely the same problems that have been described in this report. Among these problems was salinisation, a problem referred to in India as Reh.

The increasing incidence of Reh was reported as early as 1871, even before the Lower Ganges Canal was opened by C. H .T. Croswaithe. At the time, the then officiating commissioner of Agra Division, G. H. M. Ricketts, argued that the evils of salinisation demanded an immediate remedy.

Seven years later, the Reh problem was carefully examined by Lt. Col. A. F. Corbett. By this time it had become very much more serious, especially in parts of Aligarh, Meerut and throughout the Kali Nadi Valley. Hundreds of acres had already been put out of cultivation, which, as Elizabeth Whitcombe points out, was particularly serious in the highly populated areas of the Ganges Valley where it affected the livelihood of thousands of people.

In each case the damage was correctly imputed to excessive irrigation by canal water. A Reh Committee was set up and reported in 1878 under Sir Edward Charles Buck, director of Agriculture in the United Provinces. He warned in his report that the causes of salinisation - in particular waterlogging - were "the first and earliest outcome of the introduction of a canal system". [49] He also pointed out that the same disturbing influences might be at work in many areas. The Reh Committee report confirmed Corbett's earlier statements.

The committee also attributed the increase in the amount of Reh affected land (referred to as Usar) to the swamping of the fields for irrigation. Too much water was available. The true remedy was to aim for an economy in the distribution of water. This could only be achieved by charging more for it.

The Canal department could not comply with this recommendation. Its concern was a purely economic one: the price was already high and to increase it further would only have led to a reduction in the revenue accruing from canal charges. The Committee realised that prices would never be increased and, therefore, recommended that wherever possible 'lift irrigation' should be substituted for flush irrigation, thereby increasing the amount of work required in obtaining water and encouraging people to use it more economically.

On the basis of experiments that had started in 1874, the Committee also recommended that Usar tracts be reclaimed by careful watering, accompanied by intensive manuring. Unfortunately, it did not appear economic to reclaim land through such measures - at least not on any extensive scale - and the problem simply got worse.

Thus, in 1891, Dr. Voelker, a chemist appointed by the government to report on India's agrarian conditions, writes of finding "enormous tracts, especially in the plains of Northern India" affected by Reh. [50] In the Northwest Provinces alone, between 4 and 5,000 square miles were by this time already salinised.

Significantly, Voelker pointed out that in the midst of this desolation could be seen patches of "valuable crops" (by which was meant cash crops for export including opium, sugar cane, wheat, castor oil plants and cotton). These stood out "like oases in the salt covered desert around them". [51] Small wonder, perhaps, for the canal-irrigated land was almost invariably used for plantation agriculture rather than for producing crops for local consumption.

At the same time, the building of the West Jumna Canal created similar problems in the area commanded by it. The Chief Comissioner of the Punjab, for instance, reported that

"for some time past, it has been known that many villages on the banks of these canals in the Paneeput, Delhi and Rohtuk Districts, have been suffering from a destructive saline efflorescence. The accounts of poverty in some of these villages have been quite distressing ... the mischief is increasing yearly and that it would soon attain to very considerable proportions and would entail fiscal loss to the government and suffering to the people." [52]

The causes of salinisation were already well understood. Thus, in a note prepared for the Journal of the Royal Asiatic Society, H.B. Medlicott, then Professor of the Geological Survey, pointed out that although the canal waters were themselves an exacerbating agent, they were not the prime cause of the problem. It was the salt in the soil itself which caused the greatest degree of salinisation: moreover, he maintained, inefficient drainage and water accumulation was leading to those salts being dissolved and thus rising, through capillary action, to the surface where they were left after the irrigation waters had evaporated. It was, in fact, a very modern analysis of the process of salinisation. [53]

Medlicott went on to suggest that the solution lay in using the canal waters to soak the soil thoroughly in order to wash away the salts and establish a groundwater connection. Although, today, that solution would be frowned upon for ecological reasons, it was then rejected - despite the pressing nature of the salinisation problem - simply because it would have meant reducing the amount of land under irrigation so as to make more water available for flushing out the salts.

As Elizabeth Whitcombe points out, Medlicott's blueprint for desalinising the area was of little interest to the authorities because it "clearly conflicted with the official canal policy of distributing canal water over an increasingly wide command area for revenue purposes". Indeed, the Jumna experience clearly illustrates a recurring theme in this report: namely, just how suicidal it is - ecologically, socially and, indeed, economically - to allow short-term economic ends to dominate public policy.

Medlicott's diagnosis of the salinisation problem was subsequently confirmed by both T. E. B. Brown, then chemical examiner for the Punjab, and by Dr Thomas Anderson, Professor of Chemistry at Glasgow University. Thus, Anderson reported that the cause of the problem was the presence in the soil of

"some minerals rich in alkaline, the decomposition of which is prompted by the irrigation water.... A large quantity of these substances are converted into a soluble form and gradually accumulate until they become so abundant as to become noxious to plants." [54]

For Anderson, the solution lay in introducing drainage.

Those early reports - together with the correspondence that followed them - reveal, according to Whitcombe, the first "collected documentation of the problem (of Reh) and its scientific analysis". [55] Copies of the various documents were sent to the Secretary of State and three boxes of Reh-affected soil were sent to be examined in the United Kingdom.

A further report of Reh-affected lands was then produced by the officiating Superintendent Engineer of the Punjab, together with the Executive Engineer (Delhi Division) of the West Jumna Canal. Specimens collected during the research for that report were examined at the Royal School of Mines by W.J. Ward.

Technical reports were also sent to provincial administrations in India, and it was recommended that experiments in leaching out salts by applying sufficient water to 'wash' the soils should be undertaken. In addition, it was urged that a drainage scheme in the Jumna area should be initiated forthwith. Unfortunately, the Governor General - though apparently in agreement with the thrust of the reports - felt he could do no more than sanction a modest research programme. As he put it at the time, "the whole operation must necessarily be of an exceedingly simple character". [56]

Experiments were duly carried out. Unfortunately, however, it was deemed 'uneconomic' to take any long-term action to reclaim salinised soils - a further illustration of the principle that no remedial action is ever likely to be taken so long as we adhere to conventional economic dogma.

Thus, although it was already well established that salinisation was a serious problem; that it could only get worse; that, in the long run, it would inevitably affect government revenues and cause increased poverty among the populace; and, finally, that its primary cause lay in the development of large-scale irrigation schemes: the government, no doubt hamstrung by the Exchequer, persuaded itself that money spent on reclaiming salinised lands would be money down the drain - forgetting, of course, that there were no drains, and that, therein, lay the main cause of the Reh crisis.

Although the role that undrained irrigation schemes played in furthering salinisation was uncontested - being the result of observations and experiments carried out by the most prestigious authorities of the time - the government still embarked on the building of yet further irrigation projects. Thus, the building of the Lower Ganges Canal went ahead between 1873 and 1878, in the clear knowledge that the canal's command areas were affected or threatened by Usar and Reh. No measures were taken, however, to prevent further salinisation: indeed, the experience with previous schemes does not even appear to have influenced the design of the canal in any way.

The Reh Committee presented its report in 1878, the year that the lower Ganges Canal was opened. H. B. Medlicott, in his contribution to the report as Superintendent of the Geological Survey, concluded that the cause of salinisation was due to the

"combination of deforestation and evaporation together with the presence and movement of excessive quantities of moisture in the soil." [57]

Denzil Ibbeston, another contributor, insisted, on the other hand, that the solutions Medlicott proposed for the Reh problem were not practical. In particular, he pointed out that it was difficult to restore forest cover. Moreover, he argued, flush irrigation - which Medlicott deplored - was official canal policy. It could not, therefore, be changed. [58]

Like previous government investigations of the problem of salinisation, the Reh Committee concluded with a proposal for a modest series of reclamation experiments. As a result, attempts were made to reclaim Usar patches by planting trees for fuel and fodder. Those experiments were rarely successful. Indeed, since the Reh problem could not be solved without compromising short-term economic priorities, practically nothing was done. With that inaction came a legacy of salinised land, much of which was never to be reclaimed.

That legacy alone is a damning indictment of the economic system of the day. sadly, it is an economic system which still persists - albeit in modified form - and, moreover, it is still wreaking havoc in the arid lands of the world.

Salinity and downstream agriculture

Paradoxically, the very means of reducing salinisation and waterlogging at a local level only serve to exacerbate the problem for those living downstream of the irrigation schemes. The reason is clear enough. Lining irrigation canals, digging tube wells and introducing drainage are all measures undertaken to ensure that salts are flushed away in the water from irrigated land. But that salty water must go somewhere: when - as is generally the case - it is returned to the nearest river, the river's salt content will inevitably be increased.

In the past, that increased salinity did not pose the problem it does today, argues Arthur Pillsbury.

"Before man began harnessing the rivers, the seasonal floods were highly effective in carrying salts to the ocean and keeping the river basin in reasonable salt balance. Today, with river flow being regulated by storage systems, and with high consumptive use of the released water, there is not enough waste flow left to achieve salt balance. The salt is being stored, in one way or another, within the river basins." [59]

For the downstream farmer, the problem is obvious: he must irrigate his land with increasingly saline water. Moreover, if there are cities upstream of him, that water is likely to be also contaminated with domestic waste and industrial chemicals. Nor do the downstream farmer's problems stop there. In heavily developed river basins, much of the river's water will have been extracted even before it reaches him.

As a result, the flow of the river will have been considerably reduced - in many cases to the extent that it can no longer prevent the intrusion of seawater into the delta. Indeed, in some river basins, seawater has been known to intrude up to 100 kilometres inland. This Is particularly so in Bangladesh. Small wonder, perhaps, that downstream farmers in arid lands the world over, are finding their livelihood increasingly threatened. The following examples suffice to make the point:

Mexico, Colorado and the Rio Grande - Northern Mexico is partly dependent for its water supply on two 'shared' rivers - the Colorado and the Rio Grande, both of which must first flow through the South-western United States. Unfortunately for Mexico, it is precisely the South-western US which - since the Second World War - has seen the greatest expansion in irrigated agriculture.. Approximately 40 million acres are now under irrigation and - given the high evaporation rates - even good quality water can only be used twice before it becomes too brackish for agriculture.

Disposing of that brackish water is proving an increasingly intractable problem.. One method of doing so is to divert excess water into 'evaporation basins' where the salts can be left to accumulate. In that respect, the farmers of Southern California are particularly fortunate: much of the saline water from the farms in the Imperial valley (where more than 500,000 acres are under irrigation) is channelled for 80 miles via the All-American Canal into the naturally salty, inland Salton Sea. That sea - whose waters are almost as salty as sea water - also receives water from the 65,000 acres irrigated by the Coachella Canal. All told, 90 percent of the surface water entering the Salton Sea is waste water from agricultural land.

Few agricultural areas have a Salton Sea at their disposal. Where no natural salt 'sink' exists, therefore, artificial evaporation basins have been built. Those basins, however, do not - and cannot -provide a lasting solution to the salt problem. As Pillsbury notes,

"Such schemes, designed to store salts in the river basins themselves, may work for a few years or decades but they are bound to be disastrous in the long term." [60]

In particular, they are almost certain to result in the contamination of groundwaters. That problem, says Pillsbury, cannot be avoided. For one thing, saline water rapidly breaks down soils which are impermeable to fresh water, thereby rendering them permeable. Thus, even building evaporation basins on impermeable land will not prevent the long-term contamination of groundwaters. Nor, says Pillsbury, can that problem be circumvented by lining basins with an impervious material such as rubber or plastic, asphalt or even special concrete.

"Conceivably, such linings will be effective for as long as 50 years but, ultimately, one expects them to fail. In all probability, their lifetime when they are exposed to saline water will be shorter than their lifetime is when they are exposed to fresh water, for which they are normally tested." [61]

Finally, the damage caused through groundwater contamination cannot (as some have argued) be minimised by placing evaporation basins over already saline groundwaters.

"Every ground-water basin with a flow gradient must have an outlet somewhere near its lower end. The saline water in the evaporation basin will serve to increase the 'head', or hydraulic pressure, on the saline waters below, and will thereby increase the rate of discharge at the natural outlet, wreaking havoc in downstream groundwaters and downstream lands."

Apart from evaporation basins, the other principal means of disposing of waste irrigation waters is via long-distance drainage canals. Thus, in Southern California, a 290-mile long 'master drain' has been half-built - at an estimated cost of over 1.2 billion dollars - in order to take the waste waters from the San Joaquin Valley directly to the Pacific. That canal will serve 500,000 acres and will have the capacity to move more than 3 million tons of salt every year. Inevitably, the dumping of such massive quantities of salt will cause untold ecological damage within the bays and estuaries where the canals meet the sea.

Moreover, it is a moot point as to whether long-distance drainage canals actually avoid the further salinisation of water tables. Indeed, Pillsbury argues that such salinisation is only avoidable where the water table is semi-perched - that is, isolated from the deeper, main body of groundwater.

"Under those conditions there is little opportunity for the irrigation water, enriched in salts, to percolate downward and degrade the deep ground water, which remains available for irrigation or other uses."

In the valley lands, however,

"the subterranean structure is such that near-surface water cannot be isolated from deeper water, with the result that ditch or tile-drain systems are powerless to preserve the quality of the ground water."

In the meantime, the Colorado River and the Rio Grande - which, between them, receive most of the drainage water from the surrounding agricultural lands - are becoming increasingly saline. Salt concentrations in the Rio Grande, for instance, have increased from 221 to 1.691 ppm in recent years. As for the River Colorado, if salinity levels continue to rise at their present rate, then by the year 2010 the economic cost in terms of lost production and declining water quality will have exceeded $1.24 billion. [62]

Under the terms of a joint treaty signed in the early 1970s between Mexico and the United States, the American government agreed to reduce the salinity levels of waters entering Mexico. To that end, a massive desalinisation plant is being built on the Mexican border at Yuma in Arizona. The plant is designed to treat 107,000 acre-feet of water (with an average salinity level of 2,800 ppm) a year: 92,000 acre-feet of water with a salinity of less than 800 ppm will thus be provided to the Mexicans.

Originally priced at 300 million dollars, the plant is now expected to cost more than one billion dollars. At that price, irrigation water provided by the plant will cost some $800 per acre-foot - more than 35 times the current cost of irrigation water in the Imperial Valley.

It is now quite clear that the Yuma desalinisation plant will not in itself even begin to solve the water and salt problems facing the Southwest. Indeed, several other much more ambitious schemes have been proposed for the area. Although those schemes are certainly intended to provide water for flushing out excess salts from the soil, their primary purpose is undoubtedly to further the extension of irrigated agriculture.

All involve importing vast quantities of water from other parts of the US; and, due largely to the Government's concern for its escalating budgetary deficit, all have run up against financial problems. Indeed, with local tax payers unwilling to meet even a part of the vast costs of the projects - and federal funds less and less available - two of the schemes have already been vetoed as too expensive. No doubt, however, those vetoes will be reconsidered when - or, rather, if - economic conditions ever become more propitious. The schemes include the two following projects:

• The Peripheral Canal project, which would bring water from Northern California rivers around the San Francisco Bay Delta and down to Southern California, at a cost initially of between $700 million and $1.3 billion, was voted down in a statewide referendum in 1982.
• The Texas Water System, which is intended to provide water to the arid and semi-arid West of Texas. The scheme involves the building of a system of reservoirs and inter-basin water transfer facilities in the Eastern and Central parts of Texas: a coastal aqueduct stretching some 400 miles from the Sabine River to the Lower Rio Grande Valley; and a trans-Texas canal stretching mainly uphill from North-East Texas to the High Plains, with a branch running to New Mexico and another to the Trans-Peco area.
    Much of the 17.3 million acre-feet of water which would run annually through those canals would come from the Mississippi (some 12-13 million acre-feet per year) and would have to cross Louisiana before entering Texas. Because the course of the canal would be largely uphill, almost two-fifths of the state's total electricity supply (as of 1970) would be required to pump the water along the canal. Like the Peripheral Canal Scheme, the Texas Water System has been vetoed by local taxpayers, who simply refused to finance a $3.5 billion bond issue to fund the project.

Another proposal is the North American Water & Power Alliance (NAWAPA). Proposed by the Ralph M. Parsons Company of Pasadena, California, the NAWAPA scheme is one of the most ambitious of all time. It would divert water from Alaska and Northern Canada to various parts of Canada, the US and Mexico. Hydro-electric plants built along the giant canal would provide - over and above what would be required for pumping - a surplus electricity capacity of 70,000 megawatts, the equivalent output of 70 large nuclear power stations. The drainage area of the scheme would be 1.3 million square miles and 160 million acre-feet of water would be diverted southwards for irrigation and 'water-way control'. The estimated cost of the project is $200 billion: if, however, the experience of similar projects is anything to go by, the final cost could well be 3 to 4 times higher.

It goes without saying that the environmental destruction which would be caused by such wholesale interference with the ecology of the regions through which the NAWAPA scheme would pass would be little short of disastrous. Moreover, the scheme is likely to be strenuously opposed by Alaska and Canada who do not take kindly to the idea of their waters being diverted to the Southwest. Indeed, it seems that there is little chance of the NAWAPA project ever getting off the drawing board.

Nonetheless - and this is particularly significant - Pillsbury insists that, without the NAWAPA scheme, the future of the Southwest is extremely precarious. In his opinion, it is quite simply

"the only concept advanced so far that will enable the lower reaches of western rivers to achieve the salt balance necessary for the long-term health of western agriculture, on which the entire US and indeed the world has much dependence. Unless the lower rivers are allowed to reassert their natural function as exporters of salt to the ocean, today's productive lands will eventually become salt-encrusted and barren." [63]

Sind, Pakistan - Geography has condemned Pakistan's Sind province to being a 'sink' for the whole Indus river valley. As a result of economic development within both Afghanistan and the Punjab - the two upstream states which also share the Indus - its waters are increasingly polluted. Although the Tarbela Dam is intended to ensure a supply of 92 million acre-feet of good quality water to the Sind, few experts expect the dam to provide a permanent solution to the problem. Indeed, even with the dam, it will be necessary to mine the extensive groundwater reservoirs beneath the Punjab.

After the year 2000, however, that supply is likely to start running out - although some 20 million acre-feet per year of groundwater recharge will still be available. At that point, it will only be a question of time before the inevitable happens:

"Sooner or later, the concentration of salts, due to repeated capillary rise and evaporation followed by repeated irrigation and leaching, is bound to increase ... downstream." [64]

The only solution - or, rather, palliative - would be to 'export' that highly saline water directly to the sea or to allow it to accumulate in sinks along the desert margins. The cost of either undertaking would be enormous. Not surprisingly, Aloys Michel concludes that such wasteways 'are likely to be postponed'. In that event, it is hard to see how further salinisation can be avoided in the area.

Iraq and the Euphrates - Like Sind Province, Iraq is at the tail-end of a shared water supply. Thus, the River Euphrates must pass through both Turkey and Syria before it crosses into Iraq. Until the end of the Second World War, only Iraq abstracted water in any quantity from the river. Since then, however, both Syria and Turkey have put up large dams - the Keban Dam in Turkey, for instance, and the Tabaqua Dam in Syria - in order to exploit the waters of the Euphrates fully.

In the 1960s, both countries together abstracted 16,000 million cubic metres a year, which, at that time, represented 45 percent of the average annual discharge of the Euphrates into Iraq. Should plans to build new dams in all three riverine states go ahead, then, according to Professor Peter Beaumont of Bangor University, "the likely demand for water will be in excess of the available flow of the river". Inevitably, Iraq will suffer most: very little water will be left for her use and what water there is will have high salt and pollution loads. One consequence will be a corresponding fall in the quantity and quality of food produced in the area.

South Australia and the Murray River - South Australia is one of the driest states in what is probably the driest continent in the world. Sixty-six percent of its water supplies are derived from the Murray River - with that figure rising to approximately 83 percent in a dry season. Before it reaches South Australia, the Murray flows first through the States of Victoria and New South Wales, which together contribute 64 percent of the 1.1 million tons of salt carried by the river each year.

As a result of abstraction for irrigation, domestic consumption and industrial use, the amount of water reaching South Australia has been drastically reduced: moreover, what water does arrive is seriously polluted with agricultural and industrial wastes. South Australia has no say in controlling that pollution - nor is there any inter-state body which could enforce pollution controls upon the two offending upstream states.

Thus, the Murray River Committee - to which all three states send representatives - only has responsibility for allocating the amount of water used by each of the riverine states: it has no brief to ensure the quality of the Murray's waters. Indeed, in 1980, the new South Wales representative on the Committee officially stated:

"Water pollution control in South Australia is a matter for that state alone." [66]

The groundwater in South Australia is naturally saline - in some cases, it is saltier than sea water. Inevitably, as irrigation water seeps back into the Murray, it brings with it much of the salt in that groundwater. In an attempt to overcome that problem, tiled drains were installed underneath irrigated lands, the water being pumped from them into evaporation basins on the river flats.

Those basins were not watertight, however, and highly saline water is already seeping out of them into the Murray. Furthermore, the basins do not have the capacity to hold all the saline water from the State's irrigation schemes: the rest is thus released directly into the river. Saline water is also reported to be seeping from the basins into the water table, leading to the formation of 'groundwater mounds' from which there is further seepage into the Murray.

With ever greater demands being made on the waters of the Murray River for domestic, agricultural and industrial purposes, the river's flow is inevitably being reduced. Since the amount of salt carried by the river tends to remain constant - at about 3,000 tonnes a day - salinity can only increase. The prospects are grim, according to M. Butler, a geographer at Adelaide College.

"In an uncertain economic climate and in the face of rising salinity levels and increasing demand for good water for metropolitan Adelaide, the farmer's future looks decidedly shaky ... Irrigated lands will eventually be abandoned and farmers will lose their way of life." [67]

That last comment could apply to any one of the examples we have considered. Indeed, by opting for technological solutions to what are essentially ecological problems, the further salinisation of lands throughout the world is ensured. In effect, we have become trapped on a technological treadmill, which can only result in long-term ecological destruction.

In that respect, the experience of the US Southwest is, as we have seen, particularly eloquent. Thus, in their thirst for water, the inhabitants of the Southwest have sunk tuba-wells and built huge reservoirs. In their fight against salinisation, America has spent a fortune on technological measures of a type which less prosperous countries can ill-afford. Thus, they have lined irrigation canals, dug horizontal drains and built evaporation basins.

Now that those measures have failed to solve the Southwest's water and salinisation crisis, the search for new 'technical fixes' has become increasingly desperate: river basin transfers and the development of genetically-engineered salt-tolerant crops have become the order of the day but at what financial - let alone ecological - cost?

Sooner or later, the technical fixes will run out: even now, as we have seen, many are proving too costly to implement - witness the massive water transfer schemes which have been proposed for the area. The future is thus bleak for the US Southwest - as, indeed, it is for Sind, Iraq and South Australia. How long will it be before vast areas of those regions are abandoned, their best farmlands being transformed into uninhabited salt encrusted deserts.

References: