Waterlogging and Salinity in the Indus Plain:/ '-. o (S
A Critical Analysis of Some of the
Major Conclusions of the Revelle Report
In September 1962, President Kennedy's Science Advisory Committee,
headed by Dr. Roger Revelle, then Science Adviser to the United States Secre-
tary of Interior, submitted its Draft Report on Waterlogging and Salinity in West
Pakistan . An analysis of the main recommendations of this report was
published in an earlier issue of this Review [9, pp. 251-279]. Much material
was presented in that review which will not be repeated here.
The draft was considerably modified and final version of the report was
prepared in January 1964 after a number of visits by Dr. Revelle and his collea-
gues to Pakistan . The report was conveyed to President Mohammad Ayub
Khan of Pakistan by President Lyndon B. Johnson of the United States with his
letter of March 25, 1964. This article analyses the major recommendations
of the final Revelle Report and argues that the solution presented in the Revelle
Report is i) exceedingly costly compared to feasible alternatives, and ii) is not
likely to bring about the necessary increases in agricultural production assumed
in the Revelle Report. We argue that, on the basis of the evidence now available,
far greater increases in agricultural production can be realized at significantly
lower cost than those recommended in the Revelle Report by alternative me-
thods of land and water development. These alternatives are presented in Sec-
tion III of this article.
Background to the Revelle Report
The soils of the Indus Plain are alluvial. As is true of most alluvial valleys
in arid climates, salts were deposited in varying degrees when the alluvium
was laid. The low rainfall in the area has not been sufficient to leach these salts.
Many of the irrigated lands also contain calcium carbonate nodules (locally
termed as kankar) in the surface profile. These appear to have been formed
since irrigation was introduced in the Plain.
Construction of perennial I canals in the Punjab was started in the latter
part of the nineteenth century; and by 1915, the major part of the presently
irrigated area had been covered with a close network of canals. Due to con-
*The author is a Senior Research Economist at the Pakistan Institute of Development
Economics. He is indebted to Drs. Eric Gustafson and Stephen R. Lewis, Jr., Research Advisers
in the Institute and Mr. Carl H. Gotsch of the Harvard Advisory Group in the Planning
Commission for their valuable comments on the earlier drafts. However, responsibility for the
views expressed and for any errors is that of the author alone.
358 The Pakistan Development Review
tinuous seepage of water from these canals, their branches, distributories and
watercourses, the watertable has been rising at an average rate of about one
to one-and-a-half feet a year so that by 1940 it had come to within ten to
fifteen feet offthe ground surface in the major part of the perennially irrigated
areas. Since then, the watertable has been rising but at a slower rate.
Irrigation practices have also contributed to salt accumulation. One cusec2
of water is generally supplied for 333 acres in West Pakistan compared with
one hundred acres or less in the western part of the United States and other
countries under similar climatic conditions. Water applied is, thus, less than the
evapotranspiration requirements of crops; hence, the water does not wash down
much beneath the rootzone, and salts are left behind in the upper surface of
the soil. Because of the remarkably low salt content of the canal waters, this
practice does not do much harm in a few years; but over many decades it leads
to damaging salt accumulation.
More serious is the capillary rise and evaporation of the underground water
that occurs when the watertable comes to within ten feet of the ground surface.
The salts left behind by evaporation are deposited on the fields and within a
few years the salt content of the soil builds up to a level that retards and ulti-
mately prevents crop growth.
Many attempts have been made in the past to reclaim salty and waterlogged
lands. Among the various measures tried were lining of selected reaches of canals,
lowering of the level of water in the canals, closure of canals in the winter,
and construction of drains in severely affected localities. In 1918, eighty miles
of main drains were dug. In 1925, an area of 3,600 acres, which had been
abandoned due to the rise of the watertable and salinity, was selected on the
Lower Chenab Canal (at a place called Charkanwali), and various system of
drains were tried. Open field-drains and open collector-drains were found to
be more beneficial and economical than other types of drains. The whole area
was reclaimed in a few years and all crops began to be grown [26, p. 189].
As a result of the recommendations of E. Mackenzie Taylor and his co-wor-
kers, construction of open shallow-drains for removal of monsoon rainwater
was started in 1933 . This was based on the conclusion that the main cause
of the rise in watertable was the monsoon rainfall. By 1947, about 2,300 miles
of drains had been constructed in the Rechna and Chaj Doabs3.
1 Canals which supply water to land throughout the year and are fed from a permanent
barrage or diversion dam spanning the source-river are called perennial canals.
2 Cusec means one cubic foot per second. It is the most commonly used measure of
irrigation watersupply in West Pakistan. One cusec of water used for one year will deliver
720 acre-feet of water.
3 The term doab refers to an area of land located between two rivers.
Ghulam Mohammad: Waterlogging and Salinity 359
In 1954, a detailed investigation of the problem was started by the Irrigation
Department, Punjab, and later by the West Pakistan Water and Power Develop-
ment Authority (WAPDA), with the help of the International Cooperation
Administration4. This effort, along with work carried out between 1954
and 1959, culminated in the preparation of the Salinity Control and Reclamation
Project Number One (SCARP 1) in the Rechna Doab .
In order to make a detailed investigation of the waterlogging and salinity
problem throughout the province, the West Pakistan WAPDA, in 1959, appoin-
ted two firms of consultants, namely, Hunting Technical Services Limited (for
work in the Southern Zone) and Tipton and Kalmbach, Inc. (for work in Nor-
thern Zone). A number of detailed reports had been prepared or were under
preparation by these consultants [15, 16, 17, 36, and 371, when in
1961, the President of Pakistan called a meeting to consider measures to cope
with the problem. At this meeting, the President requested a Masterplan for
the control of salinity and waterlogging in West Pakistan. Drawing upon the
extensive studies which had been going on for many years, the West Pakistan
WAPDA produced such a plan in May 1961 . This plan called for cons-
truction of 31,000 tubewells, 7,500 miles of major drainage-canals and 25,000
miles of supplementary drains at a total cost of rupees 5,900 million. The entire
irrigated area of West Pakistan was to be divided into twenty-six project areas
of different sizes for waterlogging and salinity control. The first project under
this plan, SCARP 1 in Rechna Doab, went into operation in 1960/61.
The WAPDA Masterplan was discussed by President Mohammad Ayub
Khan during his state visit to the United States in 1961 with President John F.
Kennedy. As a result of this discussion, President Kennedy assembled a Panel
of experts composed of specialists in agriculture, soil science, hydrology, geology,
engineering and social sciences under the chairmanship of Dr. Roger Revelle.
the then Science Adviser to the Secretary of Interior. They submitted a draft
report in September 1962 [311 and the final report (referred to hereafter as the
Revelle Report) in January 1964 .
In Section II of this paper, we summarize the major recommendations of
the Revelle Report. In Section III we give our comments on these recom-
mendations and offer some alternatives for land and water development for
H: MAJOR RECOMMENDATIONS OF THE REVELLE REPORT
The Revelle Report starts with the assumption that waterlogging and salinity
is destroying the fertility of much of the irrigated area of West Pakistan; how-
ever, this is seen as only one of the problems besetting the agriculture in the
4 Now the Agency for International Development (AID).
360 The Pakistan Development Review
Indus Plain. Among the other deficiencies cited are shortage of irrigation water,
the system of land holdings, primitive methods of cultivation, lack of credit
and marketing facilities, and inadequate transport and supporting technical
services in the rural areas. The Revelle Report, therefore, recommends an
integrated programme for the provision of additional watersupply and drainage
by tubewells, use of more fertilizer, pure seed of improved varieties of crops,
pest and disease control and better cultural practices.
The total irrigated area in the Indus Plain is twenty-three million acres,
for which an average of seventy-four million acre-feet5 (MAF) of water was
annually diverted into canals during the five years ending 1956/57. Because of
conveyance and application losses, only thirty-four to forty-one MAF of water
is available for crop use. Consequently, an average of 1.5 to 1.8 acre-feet per
acre of canal water has been available for crops
The average annual diversion of water at canal heads is expected to increase
from the eighty MAF a year that can be obtained with the construction of
new barrages and link canals of the Indus Settlement Plan, to about ninety-
two MAF a year after the construction of Mangla and Tarbela Dams. Out of
this, about forty-eight MAF a year will go to the Punjab and Bahawalpur canals
and about forty-four MAF a year to Sind canals. Out of forty-eight MAF of
average diversion in the Punjab and Bahawalpur, the watersupply actually
available for corps is estimated at 24.3 MAF a year-the balance being lost by
nonbeneficial evapotranspiration and leakage from canals and branches [44,
Recharge of groundwater in the Punjab and Bahawalpur is estimated
a) Leakage from canals and branches ... ... 13.9 MAF
b) Seepage from rivers ... ... ... 2.0 MAF
c) Seepage from rainfall ... ... ... 1.0 MAF
d) Seepage from link canals ... ... 3.1 MAF
Total ... 20.0 MAF
The Revelle Report also makes the basic assumption that out of thirty million
acres of land in the Punjab and Bahawalpur, twenty-three million acres have
nonsaline groundwater containing an average of 700 parts per million (p.p.m.)
of dissolved salts and the remaining seven million acres have saline ground-
water with an average of 6,000 p.p.m. of dissolved salts [44, pp. 281-82]6.
5 One acre-foot is the amount of water required to cover one acre to a depth of one
foot. It is the most commonly used measure of irrigation watersupply.
6 In the Draft Revelle Report of September 1962, the area having nonsaline ground-
water was estimated as nineteen million acres and that having saline groundwaters was esti-
mated as nine million acres. In the Final Reelle Report the extent of saline groundwater areas
has been decreased and that of nonsaline groundwater areas has been increased. No reason
for this change has been given in the Final Revelle Report.
Ghulam Mohammad : Waterlogging and Salinity 361
The Revelle Report proposes that the groundwater be pumped extensively
i) to capture the whole of the recharge and ii) to mine the aquifer so as to
lower the watertable by one hundred feet in thirty years in the nonsaline areas
and by fifty feet in the saline areas [44, p. 273].
The amount of water to be pumped is estimated at 40.6 MAF a year in
the nonsaline area and 8.9 MAF a year in the saline area or a total of 49.5
MAF a year [44, p. 281]7. When this amount is added to the canal water-
supply of 24.3 MAF at the outlet head, the total water will become 73.8 MAF
a year. Out of this about 15 per cent or eleven MAF will be recycled8 and 3.9
MAF of highly saline water will be exported out of the area. The remaining
fifty-nine MAF will be available for use by crops. The net acreage that would be
cultivated in the Punjab and Bahawalpur would amount to 16.4 million acres9
of which 13.6 million acres would lie in the nonsaline area and 2.8 million acres
would be in the saline area. The average depth of water will be 3.5 acre-feet
per acre in the nonsaline area and 4.0 acre-feet per acre in the saline areas.
Modification of the canal system to transport an increased volume of river
water for long distances into areas overlying salty underground water is not
considered economical by the Revelle Panel. Even without such an additional
supply of canal water, it is believed that the average salinity of irrigation water
in the saline areas can be held as between 1,200 and 2,000 parts per million [44,
Tubewells will be installed in the whole of the thirty million acres; 56 per
cent of the pumps would lie within the cultivated area, whereas 44 per cent
would lie in the uncultivated area [44, p. 322] 10.
Salinity of groundwater will increase when the salts present in the soil are
washed downward. Moreover, there is an influx of salt from the canal water.
To maintain the salt balance 1 at a satisfactory level for agriculture, about
7 In the Draft Revelle Report if September 1962, the amount of water to be pumped was
estimated at 34.6 MAF a year in the nonsaline area and 11.4 MAF a year in the saline area,
or a total of 46.0 MAF a year.
8 The term recycling means that this amount of water after being pumped out will again
seep down into the groundwater aquifer.
9 In the Draft Revelle Report the total area to )e irrigated in the Punjab and Bahawalpur
was estimated at 15.4 million acres of which the area actually irrigated at that time was 14.1
10 In the Draft Revelle Report tubewells were proposed to be installed only inside the
canal-irrigated areas and not in the uncultivated areas.
11 The term salt balance concerns the ratio between the quantity of dissolved salts in the
irrigation water delivered to an area and the quantity removed from the area by drainage.
If irrigated areas are to be kept in continuous production, the outflow of salts must equal or
exceed the inflow.
362 The Pakistan Development Review
10 per cent of the total water used for irrigation will eventually have to be ex-
ported. However, it is stated that for the next ten or even twenty years, the
amount of water to be exported does not need to be more than one million
acre-feet a year12 [44, p. 305].
Out of forty-four MAF of canal water available for the Sind, about
eleven MAF a year may be lost by canal and watercourse leakage; about six
MAF a year by nonbeneficial evapotranspiration. Around twenty-seven MAF a
year will be available for consumptive use of crops. About four to twelve
MAF a year can be obtained by pumping from tubewells near the Indus. The
total water available for crops will, thus, be about thirty-one to thirty-nine
MAF a year. At the rate of three-and-one-half to four acre-feet per acre, this
would cover a maximum of eight to eleven million acres. This is more than the
present cultivated area but less than the twelve million acres under the canal
commands [44, pp. 283-285].
Surface and Subsurface Drainage
The Revelle Report states that surface and subsurface drains are used in
controlling the watertable and the build-up of salinity in many agricultural
regions elsewhere in the world. It would be possible to use them on the Indus
Plain in place of tubewells. In the Punjab, however, the Revelle Report states
that according to the Panel's calculations, such horizontal drains are not only
much more expensive than tubewells for eliminating waterlogging and salinity,
but they do not provide the outstanding advantages of the latter, namely, the
increase and regulation of the irrigation watersupply13. The Revelle Report
merely acknowledges that in parts of Sind and a few other places where vertical
drainage by tubewells may not be practicable, recourse must be had to hori-
zontal drainage by surface and subsurface drains [44, p. 269]. In the major
part of Sind, however, tubewells will be installed and water pumped into
conveyance channels to return saline water to the Indus [44, p. 325].
The capital cost for installation of tubewells and for other agricultural
development in the Punjab and Bahawalpur is estimated at eighty-one
million dollars, or 385 million rupees for each project area of one million
12 At another place the Revelle Report says that for the next twenty or thirty years the
amount of water to be exported does not need to be more than a million acre-feet per
year [44, p. 14].
13 No details of the calculations showing that the drains are more expensive than tube-
wells are given anywhere in the Revelle Report.
Ghulam Mohammad : Waterlogging and Salinity 363
acres14. The major components of this cost are shown in Table I. For Sind,
the capital cost is estimated at 100 to 110 million dollars for each project area1 5.
CAPITAL COST OF LAND AND WATER DEVELOPMENT FOR ONE
MILLION ACRES IN THE PUNJAB AND BAHAWALPUR
Components of cost million million
Tubewells and their electrification 41.0 195
Drainage system (return flows and floods) 5.0 24
Salt export system (wells for pumping salt water plus
conveyance channels) 6.6 31
Transporting pumped water from uncultivated area
to cultivated areas or its equivalent 9.1 43
Subtotal: tubewells and drains 61.7 294
Fertilizer plant and distribution facilities 12.0 57
Pest control and seed treatment 2.0 10
Facilities for education, research, extension and
management 5.0 24
Total 80.7 385
Source: [44, p. 165, Table 3.4].
The estimated cost is higher for Sind because a grid of conveyance channels
must be constructed to carry off saline pumped waters and canal capacity must
be enlarged to bring in additional water [44, p. 11].
14 In the Draft Revelle Report, the capital cost for each project area was estimated at
54.5 million dollars or 260 million rupees, as follows:
Wells and electrification 20.0
Drainage and redistribution 5.0
Fertilizer plant 10.0
Pest control and seed treatment 2.0
Education, research, extension 5.0
The major difference in the Final Revelle Report is in provision for drainage and redistribution,
which has been increased from 5.0 million dollars in the Draft Report to 20.7 million dollars in
the Final Report. The cost of tubewells has also been increased by 8.5 million dollars.
15 At another place in the Report, the cost of a project area in Sind is given as 110 to 150
million dollars [44, p. 142].
364 The Pakistan Development Review
Total capital cost for 16.4 million acres in Punjab and Bahawalpur and
9 to 10 project areas in Sind would be 2.3 billion dollars or 11 billion rupees
[44, p. 11]. Exclusive of the fertilizer, plant protection, and improved seed, the
cost would be 8.5 billion rupees.
The annual operating cost per net cultivated area is estimated at 107 rupees
per acre out of which 55 rupees will be required for running cost of the tube-
wells; the remaining 52 rupees are allocated for fertilizer and other agricultural-
The present value of crops in a typical" million-acre tract in the Lower
Chenab Canal area of the Punjab is estimated at 152 million rupees. With the
installation of tubewells, more than two MAF of water is expected to be avail-
able for each project area. This is expected to result in a gross increase of 236
million rupees17 and net increase of 204 million rupees in the value of agricul-
tural production as shown in Table II. The difference between gross value
EXPECTED INCREASE IN PRODUCTION FROM INCREASED USE OF
TUBEWELL WATER AND AGRICULTURAL MODERNIZATION IN MILLION-ACRE
TRACT IN THE LOWER-CHENAB-CANAL AREA
Source of increased production Gross Net
Increased depth of water on present crops 8 8
Desalination and reclamation 43 43
Increased gross crop acreage 79 73
Nitrogen fertilizer 62 44
Improved seed and plant protection 44 36
Total potential increase 236 204
Present value 152 142
Total potential value 388 346
Source: [44, Table 5.10].
16 In the Draft Revelle Report, when the tubewells were located inside the cultivated area
only, the operating annual cost was estimated at fifty-five rupees per acre, out of which
seventeen rupees were estimated for operation of tubewells and thirty-eight rupees for fertilizer
and other agricultural improvement facilities.
17 In thp Draft Revelle Report, the gross increase was estimated at 256 million rupees
and net increase as 215 million rupees in the value of agricultural production in each project
Ghulam Mohammad : Waterlogging and Salinity 365
and net value is small because the tillage costs are considered to be minor. It
is stated that "they comprise, principally, four elements. (1) Rental value of
additional land cultivated. This can be neglected since it is not a social cost,
the land would be simply unused and useless without the additional water.
(2) Additional labour for cultivation. . There is a considerable surplus of
farm labour which can be applied to land without sacrificing valuable output
elsewhere. (3) Additional labour of draft animals. . even after taking into
account the need for more bullock feed, the value of milk and meat produc-
tion for the fodder and straw grown on the additional acreage would be .. above
the cash value of total new fodder production. (4) Additional expense for pur-
chase of fertilizer, seed, plant protection and other agricultural requisites. Under
current methods of tillage these costs appear to be nominal except for a few
crops such as sugarcane. .. It will undoubtedly be necessary to fertilize most of
the new land brought under cultivations, but this will also produce additional
benefits" [44, p. 193].
The Report sets the value of present crops at 203 rupees per net cultivated
acre. This will rise to a predicted 471 rupees per acre, an increase of 126 per
cent over the existing value, after the installation of tubewells and agricultural
Total value of crops in the irrigated areas of the Punjab and Bahawalpur
in 1960 was 2,400 million rupees on 15.4 million acres of net area sown. When
the net area sown increases to 16.4 million acres, and the value of crops by
126 per cent on each acre sown, the total value of production could rise to
5,600 million rupees after the installation of tubewells and modernization of
agriculture [44, p. 201].
The Revelle Panel estimates that if tubewells are used for the recovery of
recharge water only, with the watertable under the well fields held at an average
of fifty feet, the capital costs would be 57 per cent of the design in which the
groundwater is mined to 110 feet in thirty years. Operating costs would also be
57 per cent of those in the latter case. However, by mining the groundwater it
is, the Panel members conclude, possible to add 4.8 million acres to the culti-
vated area. According to the Report, the net benefits of mining, discounted
to the present time, more than offset the discounted additional costs involved
[44, pp. 323-325].
The Revelle Report states that although the purely technical possibilities
for the improvement of agriculture in Pakistan are fabulous, there are organi-
zational and intangible impediments which must be overcome in order to re-
366 The Pakistan Development Review
alize the increases in agricultural production. These impediments are mobiliza-
tion and motivation of governmental personnel, need for better guidance and
education of the farmers, improved marketing facilities, improved agricultural
credit and continued land reforms [44, p. 201].
The Revelle Report, therefore, recommends a reorientation of strategy to con-
centrate efforts on limited project areas. It recommends that the major part of
the irrigated area of the Indus Plain be divided into some twenty-five to thirty
project areas of roughly one million acres, each manned by a competent and
adequate staff under the supervision of a Project Director. At the provincial
level, a Land and Water Development Board is recommended for getting ap-
proval of the government for development of various areas and to coordinate
the work of various departments in the project areas. The Revelle Report recom-
mends that the Project Director should have authority to supervise and direct
the project personnel and not merely to coordinate their activities. A large staff
is recommended for each project area consisting of 60 senior administrative
and technical men, about 300 junior staff and about 1900 foremen, technicians,
clerks and labourers [44, p. 164]. It is stated that the operating expenditure of
the development organization will be high throughout the Revelle-Plan period
and that cost of these, for the first few years, should be regarded as capital
expenses [44, p. 158].
III: COMMENTS ON THE REVELLE REPORT AND SOME SUGGESTED
ALTERNATIVES FOR POLICY ORIENTATION
The major recommendation of the Revelle Panel is that tubewells be in-
stalled in the whole of the Indus Plain. The Panel recommends that tubewell
waters be used for crop production in the whole of the Punjab and Bahawalpur,
both in the nonsaline and in the saline groundwater areas. The Panel further
proposes that in Sind, part of the saline pumped water be used for crop pro-
duction, but that the major part of the tubewell water be pumped into conveyance
channels and be disposed of in the Indus river. The Panel further recommends
that the project directors of twenty-five to thirty project areas into which the
Indus Plain is to be divided should have responsibility for modernizing the
agriculture of their project areas and that they should have a large and com-
petent staff for this purpose.
We consider that most of the recommendations of the Revelle Panel are
questionable on technical as well as economic grounds. A summary of our
comments on the major recommendations of the Revelle Report and some of
the alternatives suggested by us is given below.
Ghulam Mohammad: Waterlogging and Salinity 367
A. SUMMARY OF THE CONCLUSIONS OF THIS PAPER
1) Development of Nonsaline Areas
The Revelle Panel has considered the quality of irrigation water mainly
on the basis of total salt content. The only mention of sodium in the Punjab and
Bahawalpur groundwaters is contained in the assumption that one-third of
the tubewells have an effluent with an excessive salinity or sodium absorption
ratio such that they require dilution with surface water in the ratio of 1:1. No
mention is made of the excessive bicarbonate found in the groundwaters. An
examination of tubewell waters in the Punjab and Bahawalpur and in SCARP-1
area indicates that many of the groundwaters considered as fit for irrigation by
the Revelle Panel have excessive amounts of sodium and an unfavourable ratio
between calcium plus magnesium and bicarbonates. Use of such water would
cause the soils to become impermeable and alkaline. Such waters can be used
for irrigation only by mixing it with large quantities of canal water or with
heavy applications of gypsum18 and heavy drainage rates to prevent eventual
development of high alkalinity. Canal water is not likely to be available in ade-
quate supply to mix with all of the groundwaters; and the cost of the necessary
gypsum added on to the heavy cost of pumping would make the use of such
waters uneconomic. In the long run, there would, therefore, be a decrease
in crop production rather than an increase with the use of such waters. We,
therefore, recommend that tubewells should not be installed all over the Punjab
and Bahawalpur but only in such areas where the sodium content and the bicar-
bonates in relation to calcium plus magnesium content are low so that the use
of such waters does not cause the soils to become impermeable and alkaline.
2) Development of Saline Areas
For areas where groundwaters are saline, and contain excessive sodium,
deep open-surface drains would be better than tubewells for eliminating water-
logging and salinity. The deep open-surface drains can be supplemented with
open and/or tile-field drains to remove the excess of water and salts from the
waterlogged and saline soils. Such deep open-main drains with open and/or tile-
field drains are a common feature of almost all irrigation projects in the United
States, the Soviet Union, and Egypt. Provision of these drains along with an
appropriate increase in the canal water to these areas would result in far greater
increases in agricultural production than is possible by the pumping and use of
18 Gypsum is hydrated calcium sulphate. When applied to the soil containing high sodium
content, it reacts to form the less harmful and more soluble sodium sulphate and lime, a
desirable soil constituent.
The Pakistan Development Review
3) Private Tubewells
In nonsaline areas, where tubewells are to be installed, the main objective
should be to supply additional irrigation water for crop use. Drainage in these
areas becomes of secondary importance when the watertable is lowered. Such
tubewells are being installed privately by the farmers in the Punjab and Bahawal-
pur at a much lower cost than the tubewells installed by the government in
Salinity Control and Reclamation Projects. Based on experience to-date, addi-
tional area brought under crops and the increase in agricultural production are
much higher in areas where tubewells are being installed privately by the farmers
than they are in the areas where these are installed by the government. This
would seem to indicate a policy of maximum assistance to farmers for installa-
tion of private tubewells. Instead of planning the whole reclamation-and-salinity
control programme in the public sector, the government should concentrate on
providing electric-transmission and special credit facilities in the nonsaline areas
so that tubewells can be installed by the farmers.
4) Construction of Drains
While the farmers are themselves able to install tubewells with the help of
private drilling-concerns, they need considerable technical assistance in the
layout of field drains, whether open or covered. This assistance will have to be
provided by the government in areas in which field drains are required. In addi-
tion, all collector drains and deep main drains will have to be dug by the gov-
ernment. Most of these can be dug by hand-labour under the Rural Works
Programme at a very low real cost to the economy.
In the following sections, we give our detailed comments on major recom-
mendations of the Revelle Report and expand upon alternative proposals for
elimination of waterlogging and salinity and for increasing agricultural produc-
tion in West Pakistan.
B: CHEMICAL COMPOSITION OF GROUNDWATER
1) Quality of Water and Tubewells
Suitability of pumped water for irrigation depends upon four factors [33,
i) total concentration of soluble salts;
ii) sodium concentration in relation to calcium plus magnesium concen-
iii) carbonate and bicarbonate concentration;
iv) boron concentration.
Ohulam Mohammad : Waterlogging and Salinity 369
i) Total Salt Concentration: The total salt concentration is usually expressed
in terms of electrical conductivity19 (E.C.) as part per million (p.p.m.) of
dissolved salts or as milliequivalent per litre20 (m.e/1) of dissolved salts. Ac-
cording to the United States Salinity Laboratory (USSL) at Riverside in Cali-
fornia, irrigation waters with electrical conductivity of less than 250 micromhos
per cm at 250C or less, that is, with less than 160 p.p.m. of dissolved salts are
considered as excellent. Those waters with electrical conductivity of 250 to
750 (i.e., 160 to 480 p.p.m. of dissolved salts) can be used if a moderate amount
of leaching occurs. With these waters, plants with moderate salt tolerance can
be grown without special practices for salinity control. High-salinity water having
electrical conductivity of 750 to 2,250 micromhos per cm (480 to 1,440 p.p.m. of
dissolved salts) cannot be employed on soils of poor drainage while waters of
very high salinity (more than 2,250 micromhos per cm or 1,440 p.p.m. of dis-
solved salts) are not suitable for irrigation under, ordinary conditions. These
can only be used with very salt-tolerant crops and a high degree of leaching.
ii) Sodium: Sodium percentage of water is another important index of its
quality. High sodium-content adversely affects the physical properties of the soil,
makes the soil difficult to irrigate, and reduces the crop yields. The permissible
sodium concentration depends upon the relative concentration of calcium and
magnesium. The sodium hazard of irrigation water is usually expressed in terms
of sodium-absorption ratio (SAR)21.
In practice, total-salinity and sodium-absorption ratio must both be app-
raised, because the limits of sodium hazard are affected by total salinity.
iii) Carbonates and Bicarbonates: Carbonates and bicarbonates affect the soils
in an indirect way. They cause precipitation of dissolved calcium and magnesium
which increases the relative concentration of sodium. To evaluate the effect of
carbonate and bicarbonate, the concept of 'residual sodium carbonate' is used22.
19 Electrical conductivity is measured by the capacity of ionized inorganic salts in
water solution to conduct an electrical current and is expressed in terms of specific conductance
20 Milliequivalent is one-thousandth of an equivalent. Equivalent is the weight in grams
of an ion or compound that combines with or replaces one gram of hydrogen.
Milliequivalent per liter (m.e/1) is one-thousandth equivalent weight of an ion or salt
per one million gram of solution or soil.
21 Sodium-absorption ratio expresses the relative activity of sodium ions in exchange
reactions with the soil. It is calculated from the following equation:
SAR = /Calcium + Magnesium ions
22 Residual sodium carbonate is defined as the milliequivalent of carbonate and bicar-
bonate ions per litre of water remaining after the m.e/1 of calcium and magnesium are sub-
tracted from m.e/1 of carbonates and bicarbonates in the irrigation water. If this calculation
results in a negative value, the water is said to be free from residual sodium carbonate.
370 The Pakistan Development Review
It is generally accepted that water with more than 2.5 m.e/1 of residual sodium
carbonate are not suitable for irrigation, waters containing 1.25 to 2.5 m.e/1l
are considered marginal and those containing less than 1.25 m.e/1 of residual
sodium carbonate are considered as safe.
In evaluating the sodium hazards of groundwaters having high amounts
of carbonates but no residual sodium carbonate, Dr. C. A. Bower, Director of
the United States Salinity Laboratory at Riverside, California, has proposed the
use of the concept of calculated 'exchangeable sodium percentage' (ESP)2 3.
He considers that waters for which calculated ESP is less than 10 are probably
safe for direct use, those for which the calculated ESP ranges between 10 and
20 may be considered as marginal, and those for which the calculated ESP
exceeds 20 are definitely hazardous unless diluted with sufficient water to decrease
the calculated ESP to less than 20 [3, p. 59].
2) Composition of the Punjab and Bahawalpur Groundwaters
The Water and Soils Investigation Division (WASID) of WAPDA nas,
with the help of technical experts of the United States Geological Survey, carried
out 2,600 complete chemical analyses of water samples from 800 test holes
drilled between 1955 and 1962 [10, pp. 50-65]. Results of these studies are
a) TotalSalt Concentration: According to the analysis carried out by WASID
of WAPDA, the salt content of the groundwater varies in different parts of the
Punjab and Bahawalpur. The differences are largely due to the pattern of circula-
tion that existed prior to the introduction of canal irrigation. At that time,
groundwater moved from the rivers, and from upstream areas where precipita-
tion was a factor of recharge, downstream into areas of progressively diminishing
precipitation towards the south-western parts of the doabs where stagnation and
discharge through evaporation were the dominant factors in the regime. With
increasing distance from the areas of recharge and active circulation, ground-
watei in transient storage became progressively more mineralized. In areas of
more or less active recharge and circulation, the mineral content increased
gradually from less than 500 p.p.m. down the gradient to about 2,000 p.p.m.
chiefly as a result of solution of material from the sediments. This trend gave
way in the central and lower parts of the doabs to a rather abrupt transition
into zones of highly mineralized groundwater where the mineral content was
enhanced by the effects of stagnation and evaporation from the watertable. The
23 The calculated ESP of an irrigation water is claimed to be the equilibrium exchange-
able sodium percentage that will be attained by a soil irrigated with the water.
Ghulam Mohammad : Waterlogging and Salinity 371
concentration of groundwater increased to about 20.000 p.p.m. in the lower
reaches of Chaj and Rechna Doabs and Bahawalpur and to about 10,000
p.p.m. in the lower reaches of the Bari and Thai Doabs [10, pp. 52-53].
Harza Engineering Company have tabulated the aerial extent of the
Punjab and Bahawalpur areas having different concentrations of salt. Their
results are given in Table III. Of the thirty-four million acres of gross area in the
GROUNDWATER SALINITY IN THE PUNJAB AND BAHAWALPUR
Salinity concentration in parts Gross area Percentage
per million in of the
million acres total area
Below 500 10.7 31
500-1000 9.5 28
1000-3000 6.8 20
More than 3000 5.1 15
Unknown 1.9 6
Total 34.0 100
Source: [13, p. 37].
Punjab and Bahawalpur, nearly four million acres are located in the Thai desert
and in the Bahawalpur desert. Out of the remaining thirty million acres, twenty-
three million acres have groundwaters with less than 3,000 p.p.m. of dissolved
salts and seven million acres have groundwaters with more than 3,000 p.p.m.
of dissolved salts. It seems that areas with less than 3,000 p.p.m. of dissolved
solids are regarded as nonsaline areas by the Revelle Panel.
We will later revert to this and show that this assumption by the Panel is
not correct, particularly when chemical composition of the groundwaters is
considered along with the total salt content.
b) Chemical Composition: The chemical composition of the Punjab and
Bahawalpur groundwaters as determined by WASID of WAPDA is given in
Table IV. It is only in groundwaters with less than 500 p.p.m. of dissolved solids
The Pakistan Development Review
COMMON RANGES OF THE PRINCIPAL CATION AND ANION
CONCENTRATIONS IN THE NATIVE GROUNDWATERS
Total CATIONS" ANIONSa
solids Sodium Carbonate
Calcium Magne- and and Sulphate Chloride
sium Potas- Bicarbo-
(p. p. m .) (.................................per cent......... ...................)
Rechna Below 500
Bahawalpur Below 500
30-45 15-30 40-50 40-70 10-25 15-25
Source: [10, p. 62].
a) The principal salt constituents of the irrigation water are con-
sidered to exist in solution as electrolytes or dissociated ions.
Some ions are basic and are called cations; others are acid and
are called anions. When common salt, sodium chloride, is
dissolved in water it supplies sodium ions as cations and chloride
ions as anions.
Ghulam Mohammad: Waterlogging and Salinity 373
that concentration of calcium and magnesium equals or exceeds the concentra-
tion of sodium. All other groundwaters have an undesirably high concentration
of sodium. On this basis groundwaters with less than 500 p.p.m. of dissolved
solids should be safe for direct irrigation use. However, these waters have
an excessive proportion of bicarbonates. This is because the river water, which
is of the calcium-bicarbonate type, upon entering groundwater circulation,
gradually becomes more mineralized and is modified to the sodium-bicarbonate
type within a few miles from the rivers. An increase in sodium content is also
common in progressively deeper water samples. The relative increase in sodium
at the expense of calcium appears to be the result of base exchange in the clays
The transition in chemical quality downgradient is largely the result of
chemical reaction between the groundwater and the soil particles. The increased
concentration chiefly involves the accretion of more soluble constituents of
the rock particles, that is, sodium and chlorides and sulphates. Thus, while the
groundwater in the range of concentration of dissolved solids between 500 and
1,000 p.p.m. is generally a sodium-bicarbonate water, that with 1,000 to 2,000
p.p.m. may be of the mixed type having sodium bicarbonate, sodium chloride
and sodium sulphate.
Sodium and chloride account for much of the increase in concentration for
waters having 2,000 to 4,000 p.p.m of dissolved salts as the character of ground-
water evolves from a mixed type to a sodium-chloride type. Increasing con-
centration of dissolved solids above 4,000 p.p.m. is marked by a further increase
in the relative concentration of sodium and chloride and in many of the highly
mineralized waters sodium chloride accounts for about 75 per cent or more of
the total dissolved salts [10, pp. 61-65].
To sum up, practically all of the Punjab and Bahawalpur groundwaters
have undesirably high concentrations of sodium and/or bicarbonate. When
such waters are used for irrigation, carbonates in the irrigation water precipitate
as calcium carbonates or kankar and this precipitation enhances the tendency
of the soil to accumulate exchangeable sodium from the water. As exchange-
able sodium accumulates, the alkalinity of the soil further increases, leading to
further calcium precipitation and repetition of the above process. Ultimately
complete deterioration of the soil occurs in time [3, pp. 50-60].
Tubewells should not, therefore, be installed in areas where concentration
of sodium and/or bicarbonates are very high.
374 The Pakistan Development Review
R. L. Hausenbuiller, M. A. Haque and Abdul Wahab of Agricultural
College, Lyallpur determined the extent of exchangeable sodium accumulation
with the use of high bicarbonate Punjab groundwaters both in the field and in
the laboratory. They found that irrigation with four of the eleven groundwaters
over a period of four or more years resulted in accumulation of excessive amounts
of exchangeable sodium in the soil, and that the extent of sodium accumulation
was highly correlated with the residual sodium carbonate content of the water.
They further found that soil irrigated with waters containing no residual sodium
carbonate may accumulate exchangeable sodium in excess of that predicted
by the sodium absorption ratio (SAR) of the water. They suggested that consi-
derable risk may be involved in irrigating with waters containing as little as
1.25 m.e/1 of residual sodium carbonate [14, pp. 357-364].
Dr. F. M. Eaton, an FAO soil-salinity expert, examined the analyses of
tubewell waters taken at different depths by the Agricultural Engineer, Lyallpur.
Dr. Eaton came to the conclusion that "practically all these groundwaters
would require heavy drainage rates and heavy applications of gypsum to
prevent eventual development of high alkalinity and black alkali". He con-
cluded that the high cost of gypsum, imposed on the much greater cost of pum-
ped water, would limit the possibilities of pumping as universally applicable
means of increasing agricultural production in West Pakistan [5, pp. 15-16].
C. R. Maierhofer, Chief of the Drainage and Groundwater Engineering of
the United States Bureau of Reclamation, after a visit to West Pakistan, consi-
dered that the chemical analysis of waters and the chemistry of the soils that had
been irrigated for long periods indicated that the soluble sodium percentage and
calcium-plus-magnesium ratio to bicarbonate in the water may not be favourable.
Under such condition when groundwater is used to supplement the irrigation
supply, calcium precipitates as carbonate in the form of nodules (kankar) and
the soil solution becomes increasingly concentrated with sodium. These con-
centrations lead to complete deterioration to alkaline status in time [23,
3) Chemical Composition of SCARP-1 Groundwaters
The Water and Soils Investigation Division of WAPDA is undertaking a
chemical analysis of all SCARP-1 groundwaters. Harza Engineering Company
and the United States Salinity Laboratory at Riverside in California have exa-
mined the chemical analyses of the groundwaters in the SCARP-1 area as
Ghulam Mohammad: Waterlogging and Salinity 375
determined by Water and Soils Investigation Division of WAPDA. The
results of their examination are given in Tables V and VI.
ANALYSIS OF 729 TUBEWELL WATERS IN SALINITY CONTROL AND
RECLAMATION PROJECT NUMBER ONE IN RECHNA DOAB
Total dissolved salts Residual sodium carbonate
Per cent Per cent
Parts per million of Milliequivalents of
samples per litre samples
Less than 500 39 Zero or less 24
500-750 27 Zero to 2.5 37
More than 750 34 More than 2.5 39
Sourcc: [22, p. 153].
ANALYSIS OF 74 TUBEWELL WATERS OF SCARP-1 AREA BASED
ON RESIDUAL SODIUM CARBONATE AND CALCULATED
EXCHANGEABLE SODIUM PERCENTAGE
Calculated exchangeable sodium Residual sodium carbonate
per cent Milliequivalents Per cent
Range of ESP of per litre of
Less than 10 (safe) 32 Less than 1.25 45
10-20 (marginal) 28 1.25-2.5 19
More than 20 (hazardous) 39 More than 2.5 36
Source: [3, p. 59].
Using the traditional standard that waters with amounts of residual sodium
carbonate exceeding 2.5 m.e/1 are hazardous for irrigation use, Maasland and
his associates concluded that 39 per cent of the Rechna-Doab tubewells were
hazardous for irrigation use (see, Table V). Using the new method evolved by
the United States Salinity Laboratory at Riverside in California, Bower and
Maasland also found that 39 per cent of SCARP-1 groundwaters were
376 The Pakistan Development Review
hazardous on the basis of calculated ESP. They concluded as follows: "If the
74 tubewell waters considered in this study are representative of a substantial
fraction of the Punjab groundwaters, then by either the "residual Na2CO3"
[sodium carbonate] concept or the empirical equation a potential sodium hazard
is involved in the use of many of these waters for irrigation. It seems evident
that if some of the groundwaters are used as the sole source of water for irri-
gation, the soil will accumulate injurious amounts of exchangeable Na [sodium]
with time. On the other hand, if hazardous groundwater is used to supplement
surface water for irrigation by dilution or by alternate use, no excessive ac-
cumulation of exchangeable Na may occur. Rainfall, where appreciable and
effective, may also be expected to have a retarding effect on the accumulation
of exchangeable Na from hazardous waters. It is evident that the Na hazard
of the Punjab groundwaters needs further study with particular attention being
given as to how surface waters, groundwaters, and soils can be managed so as
to prevent harmful accumulations of exchangeable Na. In any case, changes in
the ESP and permeability of soils over time resulting from the use of ground-
waters of various qualities should be measured as a means of acquiring informa-
tion on the Na hazard of the groundwaters." [3, p. 60].
For the seventy-four samples for which results are given in Table V, we
have worked out the concentration of total dissolved salts in p.p.m. and com-
pared this with the calculated ESP. We find that:
i) most of the waters which are safe for direct use on the basis of calcu-
lated exchangeable sodium percentage (ESP less than 10) have less
than 500 p.p.m. of dissolved salts;
ii) most of the samples which are marginal on the basis of calculated
exchangeable sodium percentage (ESP between 10 and 20) have 500
to 750 p.p.m. of dissolved salts;
iii) most of the samples which are hazardous on the basis of calculated
exchangeable sodium percentage (ESP more than 20) have more
than 750 p.p.m. of dissolved salts.
4) Classification of the Punjab and Bahawalpur Groundwaters
Accepting the classification of groundwaters on the basis of calculated
ESP as evolved by the United States Salinity Laboratory at Riverside in California
and the very close relationship found between ESP and total salt content in
SCARP-1 groundwaters, we have classified the Punjab and Bahawalpur ground-
waters on a rough basis. We have assumed that groundwaters containing less
than 500 p.p.m. of dissolved solids or calculated ESP of less than 10 are likely
to be fit for irrigation, those with 500 to 750 p.p.m. of dissolved salts or cal-
Ghulam Mohammad : Waterlogging and Salinity 377
culated ESP of 10-20 are likely to be marginal and those which have more than
750 p.p.m. of dissolved salts or calculated ESP of over 20 are likely to be hazar-
dous for irrigation use.
Salinity of the Punjab and Bahawalpur groundwaters given in Table III
shows that 31 per cent of the waters have less than 500 p.p.m. of dissolved salts
and 28 per cent have 500 to 1,000 p.p.m. of dissolved salts. As an approxima-
tion, one-half of 28 per cent may be assumed to contain 500 to 750 p.p.m. of
dissolved salts. Similarly, we will assume that one-half of the 6 per cent waters
for which no chemical analysis is available contain 500 to 750 p.p.m. of dissolved
salts. The Punjab and Bahawalpur groundwaters may be compared, on this rough
and ready basis, with SCARP- I1 groundwaters (Table VII). On the basis of total
SALINITY OF THE PUNJAB AND BAHAWALPUR GROUNDWATERS
COMPARED WITH SCARP-1 GROUNDWATERS
Salinity concentration SCARP-1 Bahawalpur
(p.p.m.) (..........per cent..........)
Below 500 39 31
500-750 27 17
More than 750 34 52
Sources: Col (1): Table V.
Col (2): Table III and the text.
salt content, it appears that groundwaters over the Punjab and Bahawalpur as
a whole are somewhat inferior to the groundwaters in SCARP-1 area. On the
basis of chemical analysis of the Punjab and Bahawalpur groundwaters given
in Table IV, it appears that the content of sodium and bicarbonate in relation
to total salt content in the Punjab and Bahawalpur groundwaters is about the
same as that of Rechna-Doab groundwaters. Therefore, about half of the Punjab
an < Bahawalpur groundwaters are likely to be hazardous for irrigation use on
tht b asis of the calculated exchangeable sodium percentage24. About 17per cent
of ih em (i.e., those containing 500-750 p.p.m. of dissolved salts) are likely to be
m tg final; but they could be used for irrigation when mixeu with canal water
24 According to Harza Engineering Company, "it is unrealistic to assume that more
than 60 per cent of the low salinity (so called "fresh") ground waters can be used safely for
irrigation without dilution or mixed use" [13, p. 11-23]
378 The Pakistan Development Review
or with heavy applications of gypsum to the soil. Application of gypsum would
be too costly for use with waters having excessive sodium and carbonate con-
tents (even with waters having less than 750 p.p.m. of dissolved salts)2 5. All of
these marginal waters should, therefore, be used in conjunction with canal water
as is in fact being done by WAPDA and by the farmers of the Punjab and
There is some question regarding the exact correlation between salt content
and ESP. Milton Fireman and M. A. Haque, who studied the effect of tube-
well waters on soils in SCARP-1 area, stated: "It is interesting to note that the
salt content of these waters increases with calculated ESP; therefore this index
(ESP) also serves to group the waters on the basis of salinity" [7, p. 2]. On the
other hand, V. E. Hansen, C. A. Bower and G. J. Williams, who examined
the analytical data for Sind soils, found that ESP and salinity level of Sind soils
tended to be positively correlated but the correlation was not sufficiently good
to satisfactorily estimate the ESP from the salinity level. They concluded that
an independent evaluation of ESP will, therefore, be needed [11, p. 35]. If the
correlation between salt content and ESP in the Punjab and Bahawalpur ground-
waters is not as strong as that in SCARP 1, it is possible that some of the waters
having more than 750 p.p.m. of dissolved salts may not have so high an ESP
as to preclude them from irrigation use. A part of the 17 per cent of ground-
waters having salt content between 750 and 1,000 p.p.m. may be found to be
fit for development. A detailed chemical analysis of the individual areas parti-
cularly with regard to bicarbonate and sodium should be carried out. Maps
showing areas with high-sodium groundwaters and high-bicarbonate ground-
waters should be prepared before tubewells are installed in areas having more
than 750 p.p.m. of dissolved salts.
There is another reason why we should be very cautious in selecting areas
for tubewell installation. Several factors will tend to deteriorate the quality of
groundwater overtime. First, the leaching of soil profile will add appreciable
amounts of salts to the groundwater in storage. Secondly, in the cycle of recir-
culation of water from aquifer to fields and back to the aquifer, most of the salts
will remain in solution whereas most of the water will be lost to evaporation.
Thirdly, there will be an annual increment of salts from canal water to the
aquifer. Finally, chemical reaction between the percolating water and the sedi-
ments will bring more salt into solution [10, pp. 91-92]. The effects of these
factors will be mitigated somewhat by dilution with recharge by seepage from
25 See, for example, M. Maasland, C.E. Priest and M.S. Malik. They state: "It has been
proposed to overcome the adverse effects of the high carbonate contents of tubewell waters by
dissolving gypsum in the tubewell water or applying gypsum to the land. The former appears
impractical because of the relatively low solubility of gypsum (30 m.e/1). Both appear
infeasible in view of the high quantity of gypsum that is required". [22, p. 137].
Ghulam Mohammad: Waterlogging and Salinity 379
canals, rivers and rainfall. But the quality will continue to deteriorate. Tube-
wells should, therefore, be installed only in areas where groundwaters have
low salt-content, so that no great deterioration of pumped waters takes place
The Revelle Panel developed a mathematical model called the Salt-
Flow Model" for digital computer simulation and investigated the rate of increase
of groundwater salinity with tubewell pumping. They found that with no initial
salt on the ground surface the concentrations of applied irrigation water remained
under 1,100 p.p.m. during the first fifty years. But with initial salt concentration
of sixty tons per acre in the surface soil (average value for large area of saline soils
in the Punjab and Bahawalpur), the concentration of applied irrigation water fell
above the 1,100 p.p.m. level all or part of the time during the first fifty years.
They concluded that surface drainage of about 10 per cent of the tubewell pum-
ping over a 50-year period was needed to prevent eventual excessive salt accu-
mulation in the rootzone of crops [44, pp. 301-306].
We have two comments on this. First, as pointed out by Maasland, Priest
and Malik, if the watertable is lowered from 10 feet to 110 feet below ground
surface with 250-foot tubewells, the salinity will increase approximately to
(250/150) x 100 = 167 per cent of that estimated by the Panel [22, p. 149]. There-
fore, either a drainage rate of more than 10 per cent will have to be provided or
the salinity of the mixed water will fall above 1,100 p.p.m. Second, the basic
assumption by the Revelle Panel in the above Salt-Flow Model" that con-
centration of mixed canal-and-tubewell water can be accepted as 1,100 p.p.m.
is highly questionable when sodium and bicarbonate content of the ground-
waters is taken into consideration. Very probably, the high concentration of
sodium and/or bicarbonates in all the Punjab and Bahawalpur groundwaters
would cause the soil to become completely alkaline much before the salt con-
centration of the mixed canal-tubewell waters reaches 1,100 p.p.m. This is
already being experienced in SCARP-1 area as described in the following sub-
5) Effect of SCARP-1 Groundwaters on Soils
Fireman and Haque classified the mixed canal-and-tubewell waters from
SCARP-1 area into "good", "fair" and, "poor" on the basis of their ESP, and
studied their effect on saline-alkaline soils and on normal soils during 1962/63
and 1963/64. According to their classification, "good", "fair", and "poor",
waters had total salt content and ESP as shown in Table VIII. Comparing the
thirty-four mixed waters of SCARP 1 with the seventy-four tubewell waters of
SCARP 1 (Table VI), it will be seen that after mixing with canal water
the percentage of safe (good) waters slightly increased and that of marginal
380 The Pakistan Development Review
CLASSIFICATION OF MIXED CANAL-AND-TUBEWELL WATERS STUDIED
FOR THEIR EFFECT ON SOILS IN SCARP-1 AREA
Total salt Per cent of
Mixed water content in ESP mixed waters
Good 250-350 4-8 35
Fair 400-600 11-19 26
Poor 600-1000 21-43 38
Source: Calculated from [7, Tables 1 and 2].
(fair) and hazardous (poor) water slightly decreased. However, the im-
provement in water quality was very small. Percentage of hazardous waters
decreased from 39 per cent in the unmixed tubewell waters to 38 per
cent in canal-tubewell mixed waters. It is, therefore, unlikely that poor
groundwaters all over the Punjab and Bahawalpur could be made fit
for irrigation by mixing with canal water. The mixing ratio required to
lower the calculated ESP to a safe limit of 10 for the 39 per cent of
hazardous groundwaters in SCARP 1 has been calculated as 3.8 parts of canal
water to 1 part of tubewell water. It is obvious that canal water on this scale
cannot be provided. The total availability of canal water at outlet head is esti-
mated by the Revelle Panel to be 24.3 MAF a year against 49.5 MAF a year
of tubewell water. Tubewell installation will, therefore, have to be much more
With regard to their effect on soils, Fireman and Haque found that with
the use of mixed canal-and-tubewell waters in saline-alkaline soils, both salinity
and alkalinity (exchangeable sodium percentage) decreased significantly in the
two-year period. When these waters were used on normal soils, there was a
decrease in the salinity and in the exchangeable sodium percentage of the soils
irrigated with "good" and "fair" waters. These decreases were attributed to
the increased availability of irrigation water. However, there were small but
uniform increases in the average exchangeable sodium percentage of the normal
soils irrigated with the "poor" waters. In these cases, increased water applications
did not completely offset the effects of poor quality. They concluded that a mild
alkali problem would develop in six to ten years by irrigation of normal soils
by these "poor" waters (which are mixed canal-and-tubewell waters). However,
Fireman and Haque state that where "poor" waters are used for irrigation,
application of gypsum to the soil every few years at the rate of about one-half
ton per year would prevent significant soil deterioration.
Ghulam Mohammad: Waterlogging and Salinity 381
Results studied by Fireman and Haque are corroborated by the farmers
of Sheikhupura district. Some of them complain that their lands irrigated with
mixed canal-and-tubewell waters are becoming stiff and that germination of
the young seedlings is badly affected 26. This is the result of two-year irrigation.
As more and more of these mixed waters are used over time, much more deterio-
ration of the soil may be expected and the cost of gypsum added to the high cost
of pumping may make the use of such waters uneconomic.
From the results of studies by Eaton, Maierhofer, Hausenbuiller, Haque,
and Wahab, Maasland, Priest and Malik, Bower and Maasland, and Fireman
and Haque, it is evident that tubewell waters cannot be used all over the Punjab
and Bahawalpur area even when mixed with canal water for a long-term sus-
tained agricultural development.
C. IMPLICATION OF GROUNDWATER CLASSIFICATION FOR TUBE-
1) Areas Where Tubewells should be Installed
About 19.5 million acres of culturable land is commanded by canals in the
Punjab and Bahawalpur. Tubewells can, for the present, be installed in one-half
of this area, that is, about 9.5 million acres which have less than 750 p.p.m. dis-
solved salts. In addition, there is a gross area of 1.7 million acres (with culturable
area of probably 1.5 million acres) in the Sialkot and Gujrat Plain which is not
irrigated by canals but is irrigated by open-surface wells and receives high
rainfall [22, p. 148]. The salt content of the groundwaters in this area is low,
and, therefore, tubewells can be installed in this area. About one million acres
of land which receive water from river overflow and has nonsaline ground-
waters may also be developed by tubewell installation [22, p. 148]. Thus, in all,
tubewells can safely be installed at the present time in about twelve million acres
(with the net cultivated area of about ten million acres) against thirty million acres
proposed in the Revelle Report. Total water requirement for this area will be
about thirty-five MAF a year at the rate of 3.5 acre-feet per acre27. Out of this
about four MAF will be met by recharge from rivers and rains28. The balance
(thirty-one MAF) must ultimately come from canals-some of this through direct
26 This information was supplied to the author by the West Pakistan Department of
Agriculture in August 1964.
27 According to Harza Engineering Company International, the water requirement for
crops for different canal systems in acre-feet per acre are: Upper Jhelum, 3.3; Lower Jhelum
and Lower Chenab, 3.8; Marala Ravi, 3.0; Raya, 3.4; Upper Chenab, 3.5; Central Bari, 3.8;
Lower Bari, 4.2; Upper Dipalpur, 3.7; and Lower Dipalpur, 4.3 [14,p. 11-22].
28 Total rainfall and river seepage for the Punjab and Bahawalpur is estimated by Harza
Engineering Company International as 5.7 MAF. As rainfall is heavy in the upper parts of
doabs more than half of this or about four MAF of recharge from river and rain may be
taken for the nonsaline areas [14, p. II-31].
382 The Pakistan Development Review
use and remaining through wells capturing the recharge of groundwaters by
Watersupply for the whole of the Punjab and Bahawalpur is estimated by
the Panel as forty-eight MAF a year [44, p. 270]. One-half of this or about twen-
ty-four MAF would be available for nonsaline areas. This should be increased by
about seven MAF to provide thirty-one MAF of canal water. This can be accom-
plished in the next ten years. In the meantime, the farmers can safely mine about
seven MAF of water in addition to capturing all the fresh-water recharge. The
water budget for nonsaline areas is shown in Table IX. An estimate of cost for
WATER BUDGET FOR NONSALINE AREAS
First 10.years After 10 years
Canal water at canal head 24 31
Recharge from rain and river 4 4
Mining of groundwater 7 -
increasing the canal capacity by about seven MAF a year should be prepared
by the WAPDA or by the West Pakistan Irrigation Department. On the basis
of a programme prepared by the West Pakistan Irrigation Department for
increasing the capacity of canals during the third-plan period, the cost of in-
creasing the capacity of canals by seven MAF a year is likely to be of the order
of about 400 million rupees.
2) Areas Where Tubewells should not be Installed
In the saline areas of the Punjab and Bahawalpur, the Revelle Panel
propose to install tubewells to pump 8.9 MAF of groundwater with an average
salt content of 6,000 p.p.m., to export 0.5 MAF of this and to mix the remaining
with canal water and use the mixed water with a maximum of 2,000 p.p.m. of
dissolved salts [44, pp. 274-2801. The average rate of application of water
recommended for this area is 4.0 acre-feet per acre. The Panel does not recom-
mend any increase in the supply of canal water for these areas [44, p. 14].
We consider that when the high sodium content of the groundwaters are
taken into consideration, it will not be possible to maintain a long-term agricul-
ture with an application of 4.0 acre-feet per acre of water containing 2,000
p.p.m. of dissolved salts. The potential evapotranspiration of the crops for
Ghulam Mohammad: Waterlogging and Salinity 383
these areas are estimated by the Panel to about 5.0 acre-feet per acre29. At
another place, the Panel estimate the drainage rate to be 25 per cent to 67 per
cent for crops of high-to medium-salt tolerance with waters containing 2,000
p.p.m. of dissolved salts [44, p. 117]. The water requirements, including leach-
ing, for these areas should, therefore, be 6.2 to 8.2 acre-feet per acre instead
of 4.0 acre-feet per acre. Obviously, it would not be possible to continue
agriculture for a long period with 4 acre-feet per acre of water having 2,000
p.p.m. of dissolved salts particularly when sodium content of these waters is
taken into consideration.
We consider that either the total area under cultivation will have to be
reduced or application of surface water will have to be increased to these areas.
Harza Engineering Company International consider that total river-water
diversions to the Punjab and Bahawalpur canals could be increased to sixty-one
MAF a year by 1975 [13, p. 55]. If thirty-one MAF of this goes to the
tubewell-canal-irrigated areas, the balance of thirty MAF a year could be deli-
vered to the saline canal-irrigated areas where tubewells are not installed. Adding
1.7 MAF of recharge from rain and rivers would raise the total watersupply to
31.7 MAF a year. Assuming a water requirement of 5 acre-feet per acre and a
drainage requirement of 10 per cent, 31.7 MAF a year of canal water would
be sufficient to provide irrigation to a net cultivated area of about 5.8 million
acres or a gross area of about 7.5 million acres. Total culturable canal-com-
manded area having more than 750 p.p.m. of dissolved salts is about 9.5 million
acres, and the net cultivated area is about 7 million acres. Thus, about two
million acres of culturable canal-commanded area or about 1.2 million acres of
net cultivated area will have to go out of cultivation. The worst-affected area
at the lower ends of the doabs where the salinity of groundwater is as high as
20,000 p.p.m. may be allowed to go out of cultivation. This area can be con-
verted into evaporation flats or drainage wastes for dumping of saline effluent
out of the drains installed in the remaining area of seven million acres.
3) Other Issues of Consequence
a) Development of Cultivable-Uncultivated Waste Land: In the Draft Report
the Revelle Panel had proposed that 70,000 acres of uncultivated land in each
project area (1.4 million acres in 20 project areas) be brought under crops in the
Punjab and Bahawalpur [31, pp. 79-80]. In our previous analysis of Draft
Report, published in an earlier issue of this Review, we had pointed out that
as the watersupply was not sufficient for year-round cropping on the areas
29 In Appendix A-1 to the Revelle Report, the potential evapotranspiration of crops for
Multan are estimated to vary between fifty-three inches and seventy inches by various
methods. The Panel recommends a figure of 59.6 inches [44. p. 411].
384 The Pakistan Development Review
already under cultivation, no new land should be brought under cultivation
[9, p. 273]. In the final report, the Panel again recommends that about
forty-four thousand acres of land uncultivated at present be brought under
crops [44, p. 192].
We reiterate that it is not in Pakistan's interest to bring any new unculti-
vated area under cultivation, so long as year-round watersupply is not provided
to the area already under cultivation. Hansen, Bower and Williams state: "It
is a gross error to continue to spread water over more land rather than to inten-
sively irrigate a portion of the available land. Marginal lands and marginal
development schemes should be abandoned to provide an adequate watersupply
to areas already being irrigated. Intensive irrigation is a prerequisite to per-
manent arid-region agriculture" [11, p. 10]. In order to keep the movement of
salts constantly downward, the intensity of cropping should be raised from
167 per cent proposed by the Panel, when sugarcane is counted in both kharif
and rabi, to about 200 per cent, before any uncultivated area is brought under
b) Depth of Mining: In our previous review of the Draft Report, we had
suggested that the watertable should be lowered to about 30 to 40 feet below
the ground surface instead of the 110 feet proposed in the Revelle Report [9,
pp. 268-270]. We had suggested this in order to keep pumping costs to the
minimum and to keep the proportion of the mined saline water in relation
to the fresh-water recharge low in the pumped water.
In the final report, the Panel states that the capital costs as well as the
pumping costs would be 57 per cent of the planned total outlay if
only fresh recharge water were to be pumped and if the watertable were to be
held constant at a 50-foot depth. However, by mining the groundwater to 110
feet, it would be possible to add 4.8 million acres to the total irrigated land.
The present value of the net benefits of such a programme was estimated by
the Panel as being 8 per cent greater with mining than with pumping of
recharge only [44, p. 324]. On this basis, the Panel recommended mining down
to a depth of 110 feet.
In calculations of benefits, the Panel has assumed the discount rate to be
4 per cent. It has further assumed that water mined from 50 to 110 feet depth
would have the same salinity and the same value for crop production as the
fresh-water recharge. These assumptions by the Panel can be questioned on
several counts. First, the calculations are extremely sensitive to the rate of dis-
count. For example, if a discount rate of 5 per cent was used instead of 4
per cent used by the Panel, the present value of net benefits with mining would
Ghulam Mohammad: Waterlogging and Salinity 385
be 96 per cent of that obtained with pumping of recharge only instead of 108
per cent estimated with 4 per cent discount rate. With a discount rate of
6 per cent, the net benefits of mining would be only 86 per cent of that
with pumping of recharge only. Secondly, if the quality of water was
taken into consideration, the present value of net benefits with mining
would be further reduced. The salt concentrations of shallow groundwater
are generally less because of dilution by canal seepage [10, p. 82]. Therefore,
the greater the proportion of mined water in relation to the fresh recharge in the
pumped water, the less will be the value of the pumped water for crop produc-
tion. Thirdly, over time the quality of the pumped water will deteriorate and
with greater pumping a larger amount of saline water will have to be exported
to the Indus which would further decrease the value of crop production in
Sind. Finally, mining to 110 feet depth will greatly increase the cost of pump-
ing after 30 years, because the recharge will have to be pumped from that
depth over an indefinite period. If all those points are taken into consideration,
the calculations result in a negative present value of net benefits with mining
when compared with pumping of fresh-water recharge only.
The actual economic depth of pumping can be determined by farmers by
experience. But regardless of the depth to which the farmers find it economic
to pump water, eventually, there must be an equilibrium of water pumped with
the amount of recharge to the groundwater. It is, therefore, essential to increase
the capacity of the canals to supply more fresh water directly from the canals
to the outlets and to increase the fresh-water recharge to the groundwater.
c) The Salt Balance: The introduction of salt-balance concept some two
decades ago was an important step in the study of salt accumulation in the
soils. The salt-balance concerns the ratio between the quantity of dissolved salts
in irrigation water delivered to an area and the quantity removed from the
area by drainage. If irrigated areas are to be kept in continuous production, it
is necessary to maintain a favourable salt balance, in which the outflow of the
salt equals or exceeds its inflow.
On an average, during the five years ending 1956/57, about forty-four MAF
of water were diverted into the Punjab and Bahawalpur canals [44, p. 69]. The
average salt content of the canal waters is about 250 p.p.m. The total salt added
to the Punjab and Bahawalpur lands and groundwater from canal water is, there-
fore, some thirteen million tons annually. The Revelle plan tubewell-waters will
add about ninety-eight million tons of salt to the soil every year (Table X). Most
of this will be from the salt originally present in the soil and in the groundwater
or that received from irrigation water during the last fifty to sixty years.
The Pakistan Development Review
SALT WHICH WOULD BE ADDED TO THE PUNJAB AND BAHAWALPUR
LANDS UNDER THE REVELLE PLAN
Area Water Salt content Total weight of
(MAF) (p.p.m.) (million tons)
Whole area 44.0 (canal) 250 13
Nonsaline 40.6 (tubewell) 700b 34
Saline 8.9 (tubewell) 6,000c 64
Sources : a) [44, p. 275].
b) [44, p. 272].
c) [44, p 272].
The total addition to the salt content of the soil and groundwater will be
111 million tons a year. Assuming that 40 per cent of the canal water and 15
per cent of the tubewell water seeps down to the watertable about twenty million
tons of salt will go to the groundwater and about ninety-one million tons will
remain in the soil. Against this, the Revelle Panel proposes the removal of one
MAF of highly saline water during the first ten or twenty years and 3.9 MAF a
year thereafter30. This water is expected to have a salt content of 4,000 p.p.m.
[42, p. 282]. Thus, only about five million tons of salt will be removed annual-
ly from the Punjab and Bahawalpur soils during the first ten or twenty years
and about nineteen million tons a year thereafter 31. Thus, it is clear that
removal of salt is only a fraction of its accumulation in the soil. The Punjab
and Bahawalpur lands cannot be kept in continuous production under this
For the Khairpur project area (0.3 million acres) in Sind, which is proposed
to be reclaimed through drainage tubewells, Hunting Technical Services state
that 356 thousand tons of salt arrive in the canal and 1,377 thousand tons leave
in the Rohri canal annually. There is, thus, a favourable salt balance for the
Khairpur area. But the salt content of the Rohri Canal which serves by far the best
area in Sind will increase. On this subject, Hunting Technical Services state that :
30 At another place, the Revelle Report gives a figure of 10 per cent of 59 MAF or 5
MAF of saline water to be exported [44, p. 305].
31 Or twenty-four million tons if five MAF of saline water is exported.
Ghulam Mohammad : Waterlogging and Salinity 387
" We do not propose to allow the salinity of the Rohri canal water when mixed
into tubewell water to exceed 600 p.p.m. and to keep within this limit, we would
temporarily reduce the drainage facilities for the Khairpur Feeder East system
if this should ever be necessary" [18, p. 23]. Thus, there is no gain in removing
the salt from Khairpur and adding it to Rohri canal. Total salt content of the
Sind and Khairpur as a whole would increase and not decrease.
D. DRAINAGE OF SALINE LANDS
If tubewells are not installed in the saline areas, how should these areas
be reclaimed from excessive salinity and waterlogging? For these areas,
we recommend the installation of a system of deep open main drains
combined with open and/or tile field-drains32. The open drains in varying
degrees are, as we have seen, a common feature of almost all irrigation projects
in the United States, Egypt and the Soviet Union [23, pp. 1-30; 32, pp. 51-72;
35, pp. 1-53]. Open drains are usually supplemented by tile drains which
collect drainage waters from the fields and discharge them into the main open
drains. In many irrigation projects in the Soviet Union and in most of Egypt,
field drains are also open and not tiled or covered [32, p. 10; 35, p. 361.
1) Drains Versus Tubewells in Saline Groundwater Areas
Although tubewells are extensively used for irrigation purposes in the United
States, they are specifically used for drainage only in a few irrigation districts
like the Salt River Valley Irrigation District of Arizona and in the San Joaquin
Valley of California. In both these places, the success of tubewell programme
in reclaiming waterlogged and saline land is attributed to two major factors,
namely, ample layers of water bearing gravel in the subsoil strata and the very
low electric power costs. The tubewells provide valuable supplementary water
and the drainage value of these wells is now incidental to their irrigation value.
In other places like the Imperial Valley of California and the Rio Grande
Valley of Texas, tubewells were installed for drainage purposes but had to be
abandoned [35, p.87]. Except for experimental purposes, no tubewells are ins-
talled for drainage purposes either in the Soviet Union or in Egypt [32, p. 18 and
32 A drain is a conduit or channel either natural or artificial for carrying surplus surface
or groundwater. The drains are of various kinds.
Tile drain is a pipe of burnt clay or concrete, in short lengths, usually laid with open joints
to collect and remove drainage water.
Collector drain is the smallest category of open drain. Its direct function is to maintain
the watertable at a desired level and to provide for movement of leaching water.
Subdrain is one into which collector drains flow. As proposed for the Lower Indus Plain,
the subdrains are aligned between canal distributories and minors.
Branch drain collects water from two or more branch drains.
Main drain collects water from two or more subdrains.
388 The Pakistan Development Review
35, p. 44]. There is no reason why these should be installed for pumping of saline
water in West Pakistan where electric power is much more expensive than in
the United States, the Soviet Union and Egypt. There is no reason why deep
open main drains combined with open and/or tile drains should not be cons-
tructed in West Pakistan where labour is much cheaper and most of the drains
can be dug with hand labour. Maierhofer considers that: "Other pressures in
the form of indefensible promises of benefits to be derived from the purchase
and installation of certain well screens, casing and pumping equipment have
detracted from progress" [23, p. 277].
The Revelle Report does not recommend a system of open and/or tile drains
in West Pakistan. They give two reasons for this:
i) The West Pakistan irrigation plain is remarkably level and on account
of the mild slope it is difficult to utilize the surface drainage extensively
for the return of drainage flow to the rivers [44, p. 259].
ii) For the Punjab, the Revelle Panel's calculations show that the surface
and subsurface drainage is not only more expensive than tubewells
for eliminating waterlogging and salinity, but they do not provide the
outstanding advantage of the latter, i.e., increase and regulation of the
irrigation supply [44, p. 2691.
Regarding mild slope and movement of drainage waters on level plains,
Maierhofer has pointed out that if water can be moved by gravity in irrigation
canals, it can also be moved by gravity in the drains in the same area [23, p. 15].
Drainage is certainly difficult in such flat areas, but it can be accomplished
economically if some power is used for pumping to supplement existing gra-
With regard to cost, the Panel might have based its judgement on tile drains
being more expensive than tubewells on the basis of information contained in
Feasibility Report on SCARP 2 (Chaj Doab) prepared by Tipton and Kalmbach
Inc., Consultants to WAPDA [37, pp. A-1 to A-6]. In this report Tipton and
Kalmbach assume that:
i) Both methods (tubewells and tile drains) require an electrification
network throughout the area to be reclaimed with power service sup-
plied to a point within each 600 to 700 acres of land. For 2.3 million
acres of land in SCARP 2, 3,550 sump pumps would be required with
tile drains. Alternatively, 3,311 tubewells would be required for the
Ghulam Mohammad : Waterlogging and Salinity 389
ii) Both methods require, in addition to the facilities for drainage of sub-
soil, adequate surface drainage works to take care of storm runoff
and overland flooding.
iii) The drains would be six feet deep and would vary in spacing from
250 feet in fine-textured soils upto 650 feet in coarse-textured sands.
iv) Average life of both tubewells and the tile drains would be forty years.
v) Level of crop production under a tile-drainage system could be expected
to be about one-and-one-half times the present level. With tubewells
for irrigation and drainage, crop production could be expected to be
about double the present level.
We consider that most of the above assumptions by Tipton and
Kalmbach are incorrect:
i) While a tubewell-drainage system does need an electrification network
for pumping of water, a tile-drainage system does not need one. If
water can move in the Lower Jhelum Canal by gravity throughout
the SCARP-2 area, it can certainly move in the drains by gravity in
the same area [23, p. 15]. The idea of having 3,550 sump pumps with
tile drains is based on the wrong idea of the efficiency of drains. The
tile drains should discharge into collector drains which should discharge
into deep main open drains and not into pumpage sumps. Lands
having equally flat gradients are being drained in the Lower Rio Grande
Valley of Texas and other irrigated areas of south western United
States with only minor pumping required in some cases [23, p. 15].
ii) A system of surface drains is needed to remove the pumped water
from tubewells in saline areas. It is not needed to take care of storm
runoff. Rain water being free from salts is very valuable. It should
percolate down into the soil and move through the tile drains into
main deep drains.
iii) Most of the areas in the Chaj Doab are underlaid at shallow depth by
medium to coarse sands. For such areas, the average spacing for tile
drains laid at eight to ten feet depth is estimated as 800 to 1,500 feet
instead of 200 to 650 feet assumed by Tipton and Kalmbach [23, p. 19],
iv) The average life of tubewells in saline groundwater areas will be much
less than forty years. Harza Engineering Company International expect
the life of tubewells to be only twenty years [13, p. 491. On the other
hand, tile drains made of clay are chemically inert and have an almost
indefinite life [32, p. 25]. Some of the clay tiles laid over one hundred
years ago in the United States are still working successfully [21, p. 3].
390 The Pakistan Development Review
v) With tubewells pumping saline water, the salt content of the mixed
canal-and-tubewell water is expected to be 1,100 p.p.m. of dissolved
salts in the saline part of the SCARP-2 area [37, p. 487]. Such water
can be used if 11 to 67 per cent additional water over the consumptive
use of crops is provided 144, p.117]. Water application in these cases
should, therefore, be about four to six acre-feet per acre instead of two
acre-feet per acre proposed for SCARP-2 area (4.5 MAP on 2.3 million
acres). With this depth of water and with high sodium and bicar-
bonate content of the pumped water, it will not be possible to main-
tain agricultural production at the existing level. There is no possibi-
lity of doubling agricultural production in the saline groundwater
areas included in SCARP 2 over a long-term period. On the other
hand, tile drains will progressively desalinize the soil and demineralize
the groundwater through continuous removal of saline effluent out of
the area if additional canal water is provided to leach out the salts.
Under such conditions production can increase much more than the
50 per cent assumed in the Feasibility Report.
Whatever material was available and considered by the Panel, the
material supplied by Tipton and Kalmbach to prove that tile drains were more
expensive than tubewells in the Punjab, was based on incorrect assumptions in
The Revelle Panel was aware of the surplus farm labour available in Wesi
Pakistan which could do the digging of open field and collector drains at near
zero cost to the economy. For when they calculate the cost of cultivation of
an additional 380 thousand acres to be brought under crops by tubewell water
in each project area (9.5 million acres in 25 project areas) the cost of labour is
regarded as zero because, according to the Revelle Panel, "there is considerable
surplus of farm labour which can be applied to land without sacrificing valuable
output elsewhere" [44, p. 193]. An application of the same principle to the
labour required for digging of drains, would have shown that open drains are
not only more economical than tubewells, but that their cost is only a small
fraction of the cost of tubewells installed by the government through foreign
contractors using imported materials. Dr. C. A. Bower, Director of the United
States Salinity Laboratory, Riverside, California, Dr. V. E. Hansen, Director,
Engineering Experiment Station, Logan, Utah, and Dr. F. J. Williams of Rice
Experiment Station, Stuttgart, Arkansas, visited the Lower Indus Basin of West
Pakistan in July and August 1962. They recommend a phased construction of
drainage system. They state "a drainage system may initially consist of only the
major lines, with branches to drain only surface accumulation of water. The
Ghulam Mohammad : Waterlogging and Salinity 391
farm drainage might under some circumstances be justifiably deferred until the
needs become more acute" [11, p. 19]. They further state that "providing an open
drainage system with gates so that water level in the channels can be controlled
has several benefits for rice area of Sind. With small land slopes encountered
in the Indus Basin the elevation of the water table can be controlled to fit cropp-
ing needs.... The drainage system should be designed to permit maximum reuse
of drainage flows and adequate mixing of saline groundwater with canal water.
By mixing, the total water supply will be increased, drainage conveyance
costs will be reduced, the economics -of lowering saline water tables will be im-
proved and storage capacity for underground recharge will be increased" [11,
A system of providing gates in the drains as recommended by Hansen,
Bower and Williams, is already being used in the Soviet Union where the water-
table is raised by checks in the drains during periods of water shortage to over-
come drought conditions.
2) Cost of Construction of Open Drains in Pakistan
With regard to the cost, the Panel does not give their calculations to show
how the open drains were found to be more expensive than tubewells in the
Punjab. Open field-drains and open collector-drains have been provided in a
small area of about 3,600 acres at Chakanwali on the Lower Chenab Canal in
the Punjab. This area had been abandoned due to waterlogging and salinity but
has been successfully reclaimed by open field-drains dug about 4 feet deep
and 220 feet apart. According to the Director, Land Reclamation, West Pakistan,
the cost of construction of such drains under the present rates is seventeen
rupees per acre for tiled drains and twenty rupees per acre for collector drains
or a total of thirty-seven rupees per acre [26, p. 190].
Hunting Technical Services estimated the cost of open drains in Khairpur
and in the perennial areas of Ghulam Mohammad Barrage (GMB) as 220 and
300 rupees per acre respectively.
For Khairpur, the cost was estimated as follows [15, p. 50] 33
Collector drains rupees 113 per acre
Main drains rupees 108 per acre
Total rupees 221 per acre
33 In January 1961, Hunting Technical Services proposed drains for Khairpur area
estimated to cost 221 rupees per acre [15, p. 50].
In June 1961, Hunting Technical Services stated "our conclusion is that the area east of
Rohri Canal should be drained by open drains and that area west of Rohri should be drained
by Tubewells". It was confirmed that area east of Rohri had very saline groundwater. Capital
( Contd. on page 392 )
The Pakistan Development Review
For Gaja perennial area of Ghulam Mohammad Command, the cost was
estimated as follows [16, p. 45]:
Collector drains rupees 214 per acre
Main drains rupees 87 per acre
Total rupees 301 per acre
The cost of collector drains was high in Gaja command of the GMB be-
cause the soil is heavy and collector drains were to be located at 900 to 1,100
feet apart compared to 1,300 feet for Khairpur area. On the other hand, the cost
of main drains was lower in the Gaja command because of the shorter distance
to which the water was to be moved.
The cost of the major items of the collector drains dug 1,100 feet apart in
the Gaja command of GMB is estimated at 142 rupees per acre (Table XI).
COST PER ACRE OF COLLECTOR DRAINS IN THE GAJA COMMAND OF
GHULAM MOHAMMAD BARRAGE
Item Collector Subcollector per acre of
drains drains gross area
(................. rupees per acre............)
Earthwork 63 35 98
Structures 2 2 4
Land 17 10 27
Miscellaneous 8 5 13
Total 90 52 142
Source : [16, p. 42, Table 26].
Where the drains were to be laid at 900 feet apart, the total cost was estimated
at 169 rupees per acre. The average for the two came to 156 rupees per acre.
To this cost was added contingencies at 5 per cent, engineering cost at 15
per cent and interest during construction at 4 per cent. The total cost then
came to 196 rupees per acre of the gross area. This was equal to 214 rupees
per acre for the culturable commanded area.
( Contd. from page 391 )
cost of tubewells was estimated as 137 to 163 rupees per acre, whereas that of open drainage
system was estimated at 214 rupees per acre [17, p. 7].
In 1962, Hunting Technical Services presented recommendations for a tubewell project
for the whole area of high watertable in Khairpur. This was to include areas which had been
confirmed to have very saline groundwater. The capital cost of the tubewell project was
estimated at 481 rupees per acre. In this report, the cost of drains was raised to 959 rupees
per acre [18, p. 9].
Ghulam Mohammad: Waterlogging and Salinity 393
For the main drains, the distribution of cost in rupees per acre was estimated
as under [16, pp. 48-50]: (rupees per acre)
To this was added contingencies, engineering cost and interest and the
total cost came to eighty-seven rupees per acre. It will be seen that major
component of cost is earthwork, equal to about one hundred rupees per acre
for collector drains and fifty rupees per acre for main drains. This is equal
to about sixty man-days per acre. As is wellknown, underemployment is
widespread in the rural areas of Pakistan. According to the Planning-
Commission estimates, unemployment and underemployment during the
third-plan period will amount to nearly eight million manyears per year
[30, p. 25]. Digging of drains for each five million acres would provide employ-
ment to one million-men for a year. This can be arranged through Rural Works
Programme for the main and branch drains while all field and collector drains
can be completed by the farmers in their spare time in a few years without any
real cost to the economy.
The next major item is land (about thirty rupees per acre). Total area
under the present canal command is much more than the available water
supply can support. Hansen, Bower and Williams recommend that marginal
lands should be abandoned to provide an adequate watersupply to the areas
irrigated [11, p. 10]. Even if drains pass through some good lands, it will
not result in decrease in production, but the application of adequate water
to the remaining area will increase total production. Land without water in
West Pakistan is useless. The cost of land can, therefore, be neglected from the
cost of drains.
The only real cost is the cost of structures which is about four rupees per
acre for collector drains and about fourteen rupees per acre for main drains.
In addition, cost of engineering services would come to about thirty rupees per
acre. The real cost of the drains is, thus, about fifty rupees per acre.
3) Cost of Construction of Tile Drains
Dr. Eaton estimated the cost of tile drains for Pakistan on the basis of
400-foot spacing and 1950 costs in California as 50 dollars per acre or 250
rupees. This is based on the assumption that the work will be done by im-
ported tile-laying machines [5, p. 15].
394 The Pakistan Development Review
Maierhofer gives the cost of tile field-drains plus deep open main drains
in the United States as 50 to 200 dollars, or 250 to 1,000 rupees per acre,
the difference being primarily dependent upon land gradients and soil profiles
capacity for vertical and lateral water conductivity. For Pakistan, Maierhofer
considers that it may be possible to accomplish the task at much less cost, as
hand labour would be appreciably cheaper for doing the work than would
power equipment [23, pp. 17-20]. He cites the example of another (unnamed)
country where wages were twenty cents per hour for unskilled workers and
fifty cents per hour for skilled tile-layers. At these wages, the costs were about
one-third those of comparable work by power equipment. In Pakistan rural
areas, wages for unskilled workers are only five cents an hour (two ruppees
per day) and those for skilled workers about fifteen cents an hour (six rupees
per day). It would, therefore, be possible to lay tile drains and dig open
collector and main drains with hand labour at much less than 50 to 200 dollars
per acre if the government were prepared to forget the hopelessly unrealistic
task of accomplishing all drainage works with imported machinery.
4) Disposal of Drainage Waste
Under ideal conditions, the saline drainage-waste should discharge into
the sea or natural depressions. This can be done for the lower Indus Plains. But
for the Punjab and Bahawalpur disposal of saline waste into the sea would be
too costly in view of the long distances involved.
Drainage water can be disposed of into drainage wastes or evaporation
flats in the lower reaches of the doabs. Evaporation from free water surface in
these areas is of the order of eighty to ninety inches a year [4, p. D-3 to D-81.
As much as seven MAF a year of drainage effluent could, therefore, be
evaporated from one million acres of evaporation flats.
Total area for which drains are to be provided in the Punjab and Bahawal-
pur is about seven million acres with a net cultivated area of about 5.8 million
acres. Total watersupply for this area is estimated as thirty MAF a year. If 10
per cent of this is removed in the drains, the total drainage effluent will be
about three MAF a year. This can be evaporated from about half a million
acres of waste land. This land can be selected out of the noncultivated area or
from the abandoned areas in the lower reaches of the doabs where salt concen-
tration of the groundwater is the highest.
Contrary to proposal of the Revelle Panel, this will not increase the salinity
of the irrigation water in any canal in the Punjab and Bahawalpur and will not
cause any increase in the salinity of the Indus water entering Sind.
Ghulam Mohammad : Waterlogging and Salinity 395
With provision of drains, there will be rapid and progressive improvement
in the salinity of the soils and groundwater in these areas. Eventually, we should
reach a stage when the groundwater will be far less saline and could be reused
for irrigation. In the perennial area of the Ghulam Mohammad Command,
groundwater salinities are higher than the salinities of the saline groundwater
areas of the Punjab and Bahawalpur. For Ghulam Mohammad Command, Hunt-
ing Technical Services proposed field drains spaced 900 to 1,500 feet apart for
a water level six feet below ground surface. Irrigation supply for this area was
estimated at 4.4 acre-feet per acre at the canal head. With this water and surface
drains six feet deep, they expected the groundwater to have less than 2,500 p.p.m.
of dissolved salts. Under our proposals, the tile drains, laid at eight to ten feet
deep, in the saline areas of thePunjab and Bahawalpur with canal watersupply of
5.5 acre-feet per acre, should remove the salt from the soil as well as from the
groundwater up to about fifteen to twenty feet below ground-surface when
the surface water percolates down and enters the tile lines from below along
with the saline groundwaters. This has been accomplished in ten to fifteen years
in the Soviet Union [32, pp. 62-64].
5) Increase in Crop Production on Drained Land
The increase in agricultural production on land drained by open and/or
tile drains would be much higher than that obtained by the installation of tube-
wells and the use of saline water for irrigation. As an example, in the Uzbekistan
Republic of the Soviet Union, the yield of cotton has increased from 16.3 maunds
per acre in 1940 to 31.7 maunds per acre in 1959 due to provision of open main
drains combined with other improved practices [32, p. 101]. For those areas
where open drains have been supplemented with tile field-drains, the yield of
cotton has increased to forty to fifty maunds per acre or even more [32, p.100].
Similarly, the Imperial Valley of California had never been prosperous before
the installation of tile drains. Families repeatedly lost all of their savings and
moved away [5, p. 28]. In December 1945, the Operation Division of the Soil
Conservation Service began, in cooperation with the district authorities, pre-
paring drainage plans for the farmers[35, p. 24]. Several thousand miles of tile
drains were installed in the Imperial Valley lands in the next few years. The
results were so striking and practical that immediately thereafter the Federal
Land Bank began to make loans on tile-drained land whereas no loans were
granted previously [5, p. 28]. On the other hand, in the Khairpur project area
in Sind where tubewells are proposed to be used for reclamation of saline and
waterlogged lands, Hunting Technical Services State "We do not think that
drainage by itself will result in very large increases in crop yields. Our basic
assumption in calculating the benefits of drainage are that the immediate impact
of drainage, with no change in methods of farming or increases in other inputs,
396 The Pakistan Development Review
should result in an approximate 25-per-cent improvement in crop yields and
that this drainage accompanied by a successful extension and research service
and the progressive removal of social and economic barriers to increased pro-
duction will raise crop yields to 100 per cent of their present value." [18, p. 10].
As we have previously pointed out, any increase in production by pumping of
saline water in one area is likely to be offset by a decrease in production in down-
stream areas where the saline pumped waters are used after mixing with canal
water. In our judgement, the best method for saline groundwater areas would
be drains and not tubewells.
6) Water Economy on Drained Land
One great advantage of tile drainage is that it materially reduces the water
requirements of crops to be met from irrigation. This is because roots of most
crops grow in the direction of increasing soil moisture. The greater the depth
of drained soil, the greater will be the root penetration, feeding area, available
plant food supply and drought resistance of crop [39, p. 9].
Tipton and Kalmbach regard this as a disadvantage rather than an advan-
tage. According to W. W. Donan, a drainage engineer engaged by Tipton and
Kalmbach, tile drains and open drains can lower the watertable to four feet
below ground surface [4, p. D-8]. At this depth, there is an evaporation oppor-
tunity34 of about 0.12 inches per day or 43 inches per year. On the premise that
about half of this evaporation opportunity might actually occur, there could be
a loss of about eighteen inches of water from a tile-drained land. On the other
hand, by pulling the watertable to below ten feet with tubewells, the loss of
water could be decreased to about three to four inches a year, thus saving a
considerable volume of water.
We choose to put the same proposition in the opposite way, If tiles are
laid at eight to ten feet depth, the watertable will remain at about six to nine
feet depth. When crops are grown on such tile-drained lands throughout the
year, the underground water could supply about twelve inches or more of water
requirements of crops when the groundwater has been demineralized in a few
years. Consequently, less canal water would be required. This is what is being
done in the Soviet Union where the tile drains are laid about six-and-half to
eight feet deep. The watertable is kept at five to six-and-half feet below sur-
face. Under these conditions, cotton plant obtain 30 to 50 per cent of their water
requirement from the groundwater reservoir and only 50 to 70 per cent of the re-
quirements have to be applied in irrigation water.
34 Evaporation opportunity refers to the amount of water that could evaporate from a
soil surface under a given set of climatic conditions and with watertable held at a given depth
below the soil surface.
Ghulam Mohammad : Waterlogging and Salinity: 397
This view is supported by Hansen, Bower and Williams. They state as
follows: "Care should be exercised in lowering the watertable much below the
maximum root zone of the crop. A considerable portion of the consumptive
use requirements will be met from groundwater under favourable conditions.
Lowering the watertable below the reach of mature roots will increase propor-
tionately the amount of water that must be applied by irrigation from limited
surface water supplies. Thus, a water table supplying water to the lower portions
of the root zone will be very beneficial from the view point of water supply, but
will have disadvantages due to increased evaporation from the soil and increased
accumulation of salts. A watertable between the two extremes is the most de-
sirable depth. However in applying this concept, care must be exercised to insure
that adequate salinity control results. If good control of the watertable cannot
be assured a deeper table may be advisable." [11, pp. 13-14]. They further add:
"The depth from which roots can extract moisture from the soil should be deter-
mined. With the high rates of consumptive use, depths would be expected to
exceed those encountered in more temperate climates. Thus, crops requiring
two months to mature may have roots penetrating to three feet, crops requiring
four months to mature may have roots six feet deep, and a six-month crop may
have roots extending to ten feet." [11, p. 231.
The above estimates of Hansen, Bower and Williams regarding the depth
from which roots may extract water are confirmed by preliminary experiments
carried out by the Irrigation Research Institute at Lahore. The Institute has
carried out experiments on the evapotranspiration of crops with watertable held
at different depths. It was found that a large part of the water requirements of
crops could be met from the groundwater storage when watertabie was held
at five to nine feet depth. The results of their work are summarized in Table
XII. In the case of cotton more than 40 per cent of the evapotranspiration
requirement was met by the groundwater at nine feet depth. In the case of wheat,
45 to 80 per cent of the total requirement was met from groundwater main-
tained at five to six feet below ground surface. In the case of barley, gram and
lentil, 70 to 95 per cent of the total water requirement were met from
groundwater and only a small fraction of water requirement had to be supplied
as surface irrigation.
The work on evapotranspiration of crops at the Irrigation Research Insti-
tute, Lahore has been in progress since one or two years only. The experiments
were carried out in small lysimeters or hume pipes. It is not definitely possible to
conclude that under field conditions as much as 40 to 95 per cent of water re-
quirement of various crops will be met from the groundwater when water is
held at five to nine feet depth. There is no doubt, however, that, as predicted
398 The Pakistan Development Review
EVAPOTRANSPIRATION OF DIFFERENT CROPS AND THE REQUIREMENT MET
FROM SURFACE IRRIGATION, RAINFALL AND GROUNDWATER
Watertable MET FROM of evapot-
Crop below Evapotrans- --- ranspira-
ground piration Surface Rainfall I Ground tion met
surface irrigation [ water from gro-
(feet) (inches) (............. inches............)
Cotton 9.0 40.3 16.0 6.5 16.9 44
Wheat 1962/63 5.7 19.9 3.1 1.0 15.8 79
Wheat 1963/64 5.0 26.3a 15.8 1.3 11.6 44
6.0 22.6b 15.8 1.3 10.0 44
Barley 4.7 17.9 4.1 1.0 12.8 71
Gram 5.0 18.8 1.0 17.8 95
Lentil 6.0 13.3 3.1 1.0 9.3 70
Notes: Source : [27, pp. 13-22].
a) Of this 2.4 inches infiltrated to the watertable.
b) Of this 2.1 inches infiltrated to the watertable.
by Dr. Bower and his associates, a considerable part of the water requirements
of crops can be met from the groundwater when a system of deep open main
drains combined with open and/or tile field-drains is installed. In the Soviet
Union, they are already meeting 30 to 50 per cent or more of the water
requirement of cotton and other crops from the groundwater. We can probably
do the same in the saline groundwater areas when the salts have been removed
from the soil and upper part of the groundwater demineralized by installation
of a system of drains.
E. ORGANIZATION FOR LAND AND WATER DEVELOPMENT
The Revelle Panel proposes that the major part of irrigated area of West
Pakistan be divided into twenty-five to thirty project areas of about one million
acres each. It further proposes that each project area be put under the charge
of a project director with competent and adequate staff for operation of
tubewells and modernizing the agriculture of his area. The Panel proposes that
new project areas should be brought into the programme at the rate of about
one every year and the entire programme should be completed in twenty-five to
Ghulam Mohammad : Waterlogging and Salinity : 399
We consider that it is not necessary to open project areas and provide a
large staff for operation of tubewells in such areas. As we have pointed out
previously, it is not desirable to install any tubewells in the saline groundwater
areas in the Indus Plain. These areas should be developed by providing deep
open main drains combined with open and/or tile field-drains and increased
canal watersupply. Tubewells should be installed only in good groundwater
According to a survey carried out by the Pakistan Institute of Development
Economics, private tubewells are being installed in these areas by the farmers
at a far less cost than the government tubewells. Farmers are saving and invest-
ing considerable funds and rural capital formation, so badly needed for econo-
mic development, is taking place. Further, the farmers are getting much larger
increases in agricultural production where they have installed their own tube-
wells than in the areas where the government has installed the tubewells. For
details readers are referred to the last issue of this Review in which results of
this survey were published [8, pp. 233-243]. We, therefore, recommend that the
West Pakistan Government should concentrate on providing electric-transmis-
sion facilities to the good groundwater areas, extend long-term credit, and
provide other ancillary facilities required by the farmers for the installation of
tubewells. If these facilities are provided, the farmers may complete the installa-
tion of tubewells in all good groundwater areas within a period of five to ten
Drainage for the development of saline groundwater areas through deep
open main drains and open and/or tile field-drains calls for government attention.
The main and branch drains will have to be laid first. A masterplan for these
should be prepared by the government and work executed under the Rural
Works Programme. Field drains will have to be dug by the farmers themselves,
but technical assistance must be provided by the government. Layout of
field drains should be marked by the government and farmers should be asked
to dig the part of the drain passing through their holdings. As one field drain
may run over the holding of many cultivators, the government will have to
assume power to ask any reluctant land holder to undertake the digging of the
drain marked through his holding. When the farmers see the increase in agri-
cultural production brought about by drainage they will surely take up the job
without much difficulty. As soon as laying of open field-drains is started, part
of these should be covered tile drains, again the government cooperating
with and helping the farmers in this job.
400 The Pakistan Development Review
IV : CONCLUDING REMARKS
In sum, the information presented in this paper, on the chemistry of
groundwaters and relative benefits and costs of the tubewell programme
presented in the Revelle Report, and a feasible alternative programme of deep
open main drains combined with tile field drains, raises serious questions
about the major recommendations of the Revelle Panel. Information available
on the chemistry of the groundwaters indicates that active implementation of
the programme recommended by the Revelle Panel and now under execution in
SCARP areas should be postponed. Pending further investigations, tubewell
installation should be limited to the best groundwater areas. If the farmers
are encouraged to play an active role in the installation of private tubewells,
the programme would both cost drastically less and achieve a great deal more.
A large expansion of the electrification programme to cover all nonsaline
groundwaters should be undertaken to expedite the installation of private
tubewells. On the other hand, in saline groundwater areas, installation of a
system of drains with adequate canal watersupplies would bring far greater
increases in production at far less real cost to the country than the installation
of tubewells for pumping and use or export of saline water. A programme
for diversion of additional river water and increasing the capacity of canals,
both in the saline and nonsaline groundwater areas should be initiated for this
It is to be hoped that the Panel's recommendation will be reconsidered
and modified along the lines indicated in this paper before these are
Ghulam Mohammad : Waterlogging and Salinity 401
 Agricultural Development Bank of Pakistan. Annual Report and Statement
of Accounts for the Year ending June 1962. (Karachi: Agricultural Development Bank of
 Beanchamp, Keith H. "Tile Drainage-Its Installation and Upkeep", Water, The
Yearbook of Agriculture, 1955. (Washington, D.C.: US Department of Agriculture, 1955).
 Bower, C. A. and Maasland, M. "Sodium Hazard of Punjab Groundwaters", Sy-
mposium on Waterlogging and Salinity in West Pakistan. (Lahore: West Pakistan Engineering
Congress, October 1963).
 Donnan, W. W. "Evaporation of Ground Water," Feasibility Report on Salinity
Control and Reclamation Project Number Two, Chaj Doab. (Denver: Tipton and Kalmbach,
 Eaton, Frank M. Report to the Government of Pakistan on Certain Aspect of Salinity
in Irrigated Soils. (Rome: Food and Agriculture Organization, September 1953).
 Edminster, T. W. and Reave, Ronald C. "Drainage Problems and Methods", Soil,
The Yearbook of Agriculture, 1957. (Washington, D.C.: US Department of Agriculture, 1957).
 Fireman, Milton and Haque, M.A. "Progress Report on Effect of Tubewell Waters
on Soils in SCARP 1", Note supplied to the author by M. Maasland of Harza Engineering
Company International in August 1964.
 Ghulam Mohammad. "Some Strategic Problems in Agricultural Development in
Pakistan", Pakistan Development Review, Vol. IV, No. 2, Summer 1964.
 and Beringer, Christoph. "Waterlogging and Salinity in
West Pakistan: An Analysis of the Revelle Report", Pakistan Development Review, Vol. III,
No. 2, Summer 1963.
 Greenman, D. W., Swarzenski, W.V. and Bennet, G. D. The Groundwater Hydro-
logy of the Punjab, West Pakistan, Bulletin No. 6 of the Soils and Investigation Division of
WAPDA. (Lahore: West Pakistan Water and Power Development Authority, 1963).
 Hansen, V. E., Bower, C.A. and Williams, G. J. Salinity Control and Reclamation
Programme in the Lower Indus Basin. (Lahore: Harza Engineering Company International,
 Harza Engineering Company International. Programme for Water and Power De-
velopment in West Pakistan, 1963-1975. Supporting Studies. An Appraisal of Resources and
Potential Development. (Lahore: Harza Engineering Company International, September 1963).
 Programme For Water and Power Development in West
Pakistan through 1975. (Lahore: West Pakistan Water and Power Development Authority,
 Hausenbuiller, R. L., Haque M.A. and Wahab, Abdul. "Some Effects of Irriga-
tion Waters of Differing Quality on Soil Properties", Soil Science, Vol. 90, pp. 357-364, quoted
by C. A. Bower and M. Maasland .
 Hunting Technical Services Ltd. Sukkar-Guddu-Ghulam Mohammad Drainage
and Salinity Control Project, Report No. 1, Khairpur Command. (London: Hunting Technical
Services Ltd., January 1961).
 Report No. 2, Gaja Perennial Area. (London: Hunting Tech-
nical Services Ltd., May 1961).
 Report No. 3, Khairpur Command Supplementary Report.
(London: Hunting Technical Services Ltd., July 1961).
[ 18] -. Report No. 3, KhairpurProjectPlanning Report. (London:
Hunting Technical Services Ltd., 1962).
 Report No. 4, Ghulam Mohammad Barrage Command.
(London: Hunting Technical Services Ltd., July 1961)
402 The Pakistan Development Review
 Report No. 5, Sukkur-Guddu-Right-Bank Command.
(London: Hunting Technical Services Ltd., October 1961).
 King, James A. and Lynes, W. S. Tile Drainage. (Mason City: Mason City Brick
and Tile Co., 1946).
 Maasland, M., Priest, C. E. and Malik, M. S. "Development of Groundwater
in the Indus Plain ", Symposium on Waterlogging and Salinity in. West Pakistan. (Lahore: West
Pakistan Engineering Congress, October 1963).
 Maierhofer, C. R. Reconnaissance Report: Draintage, Waterlogging and Salinity
Problems of West Pakistan. (Denver: US Bureau of Reclamation, October 1952).
 Majid Hasan Khan. Wells and Tubewells. (Lahore: Supdt. Government Printing
West Pakistan, 1959).
 Appraisal of Work Done by the Agricultural Machinery
Organization, Northern Zone. (Lalhore: Supdt. Government Printing West Pakistan, 1960).
 Husain, Mohammad and Nishat, H. A. "Review of Reclamation Activities and
Methods and Suggested Measures foT Waterlogging and Salinity Control", Symposium on
Waterlogging and-Salinity in West Pakistan: (Lahore: West Pakistan Engineering Congress,
 Nazir Ahmad, Akram, Mohammad and Ahmad, Shabir. A Study of Evapotranspi-
ration of Crops under Varying Conditions of Watertable and Soil. (Lahore: Irrigation Research
 and Sheikh, Mohammad Sadiq. A Statement of Coir
String Strainer 7ubewells Installed by Farmers in the Ex-Punjab and Bahawalpfir Area.
(Lahore: Irrigation Research Institute, 1963).
 and Qureshy, M.A. A Study of the Performance of Diffe-
rent Types of Strainers Installed in an Area Between Upper Chenab Canal and Sheikhupura Com-
monly known as Shadman and Chichoki Mallian Reclamation Schemes. (Lahore: Irrigation
Research Institute, 1964).
 Pakistan, Planning Commission. Outline of the Third Five Year Plan (1965-70).
(Karachi: Planning Commission, August 1964).
 President Kennedy's Science Advisory Committee. White House Interior Panel,
Draft Report on Waterlogging and Salinity in West Pakistan. (Washington, D.C.: Supdt. Doc.,
 Qureshi, M. S. Control of Salinity and Ground Waters in the Agricultural Lands of
U.S.S. R. A Report of the Delegation of Pakistan on Waterlogging and Salinity Control Prob-
lems in U.S.S.R., together with an Account of Practices in Cotton Cultivation, Collective and
State Farms, Sheep Breeding, Afforestation and Associated Farms, Sheep Breeding, Afforesta-
tion and Associated Activity. (Karachi: Ministry of Industries and Natural Resources, 1963).
 Richards, L. A. (ed), Diagnosis and Improvement of Saline and Alkali Soils.
(Agriculture Handbook No. 60). (Washington, D.C.: US Department of Agriculture, 1954).
 Taylor, Mackenzie, Malhotra, J. K. and Mehta, M. L. An Investigation of the Rise
of Watertable in the Upper Chenab Canal Area. (Lahore: Irrigation Research Institute, 1933).
 Thomas, Sir Roger, Drainage and Reclamation of Irrigated Lands in Pakistan in-
cluding A Report on Visits to USA and Egypt, 1949. (Karachi: Times Press, 1949).
 Tipton and Kalmbach. Review of Project Number One Salinity Control Programme
in West Pakistan. (Denver: Tipton and Kalmbach, June 1959).
 Feasibility Report on Salinity Control and Reclamation
Project Number Two. (Chaj Doab). (Denver: Tipton and Kalmbach, 1960).
 Feasibility Report on Salinity Control and Reclamation
Project Number 3, Lower Thal Doab. (Denver: Tipton and Kalmbach, April 1963).