| Page 1 |
GLOBALIZATION AND WATER RESOURCES MANAGEMENT:
THE CHANGING VALUE OF WATER
AUGUST 6 -8 AWRA/IWLRI-UNIVERSITY OF DUNDEE INTERNATIONAL S PECIALTY CONFERENCE 2001
A REVIEW OF THE GANGES TREATY OF 1996
A. Nishat1 and M. F. K. Pasha2
ABSTRACT: The critical review of the Ganges Treaty of 1996 (Treaty ’96) between Bangladesh and India
regarding the water flow in the dry season (January to May) through the river Ganges has been made using an
analytical approach based on an hydrological frequency analysis of flows released to Bangladesh at Farakka (West
Bengal of India) and the Gumbel’s extreme-value distribution. It also includes a simple analysis of flow data at this
point. First, two years of data (1997 and 1998) were studied just after the Treaty’96 was signed. After an analysis of
the treaty, it is observed that the treaty was followed more effectively in 1998 than in 1997. Since it is concluded
that the sharing of flows fully depends on the availability of water at Farakka, the upstream reach of the Ganges,
both countries (Bangladesh and India) would do better to augment the flows upstream for their own interests though
compliance with the terms of the treaty. A copy of the Ganges 1996 treaty is attached herewith for a clearer
understanding of this paper.
KEY TERMS: Ganges River, international water law, treaties, international rivers, water allocation.
INTRODUCTION
The Ganges is an international river with its basin spread over China, Nepal, India, and Bangladesh; it
originates at an elevation of about 23,000 feet in Gangotri on the southern slope of the Himalayan range. From there
the river traverses south and southeastward in India for about 1,400 miles. About 11 miles below Farakka, India, it
forms the common boundary between India and Bangladesh and continues about 63 miles before finally entering
Bangladesh near Rajshahi. The total course of Ganges before discharging into the Bay of Bengal is about 16,000
miles (2,500 km) (Abbas, 1984). Since water is a scarce resource and the Ganges is an international river, there
needs to be a treaty or agreement to assure the proper utilisation of water.
When there was no barrage on Ganges at Farakka, India, there was no question about sharing the water. At
that time water flowed naturally through the Ganges and its distributories. After constructing ) the Farakka Barrage
(in India) in 1975, vital questions arose as to who would control the gate of the barrage, and for what purposes and
how much water would be used, and how water would be shared between Bangladesh and India. To get answers to
these questions, the need for a treaty or agreement was felt by each basin state and so the agreement 77, MOU 82,
MOU85 and lately the Treaty’96 are made to share the water between Bangladesh and India.
1Professor, Department of Water Resources Engineering, BUET, Dhaka, Bangladesh, Country Representative,
IUCN, Bangladesh.
2Department of Environmental Engineering and Pollution Control, Shahjalal University of Science and Technology,
Sylhet, Bangladesh.
BRIEF HISTORY OF DIFFERENT AGREEMENTS, MEMORANDA OF UNDERSTANDING,
AND THE TREATY OF 1996
The Ganges dispute was recognised in 1951 when Pakistan protested to India about the proposed Farakka
Barrage. Negotiations had been pursued between India and Bangladesh, and prior to 1971 with Pakistan, on this
issue for the last 40 years. Instead of converging towards an acceptable sharing arrangement, the issue had become
more complex with many other issues related to a sharing arrangement. There have been several short-terms sharing
agreements: the first Ganges Water Treaty was worked out in 1977. Subsequently, the Memorandum of
Understanding of 1982 (MOU82), and MOU85 were developed. The most recent agreement is Treaty’96. Since the
proposal for diversion of Ganges water was first disputed in 1951, discussions and negotiations can be divided up
into the following seven phases (Nishat, 1996).
Table 1: Different phases of discussions and negotiations
Phase (Period)
Focus
Phase 1
(1951-74)
How to finalise respective shares before commissioning the barrage. Discussions
on respective claims and their justifications.
Phase II
(1974-76)
Issue of flow augmentation raised. Failure in bilateral negotiations. Barrage
operation begins in 1975 with concurrence of Bangladesh. Unilateral withdrawal
of waters by India in 1976. Issue taken to UN by Bangladesh
Phase III
(1977-82)
Ganges Water Agreement (1977-82) signed and implemented Discussion on
augmentation fails. Treaty not renewed despite provisions to do so.
Phase IV
(1982-88)
Memoranda of 1982 and 1985 implemented. Provisions similar to
1977 Agreement except it contains a guaranty clause. All regional rivers brought
to the discussion table. Both sharing and augmentation
options discussed. Both sides review augmentation proposal.
Phase V
(1988-92)
Negotiations continue but without success. Divergence in approach. Relationship
between sharing arrangements and augmentation proposal becomes a very critical
issue.
Phase VI
(1993-1995)
No dialogue and India has unilateral control over the Ganges.
Phase VII
(1996- Present)
Treaty is signed to share water resources according to the terms of the treaty.
Before the commissioning of the Farakka Barrage, an interim arrangement was made for a test run to
release a flow of 11,000 to 16,000 cusecs through the Feeder Canal from 21 April to 31 May 1975 and releasing the
remainder for Bangladesh. After 1975, there was no renewal of this arrangement and unilateral withdrawal by India
continued causing disastrous effects on the Ganges-dependent areas of Bangladesh, mainly in the south-west region.
Repeated protests against the continued unilateral withdrawals were of no avail, and Bangladesh took the matter to
the United Nations. Through intensive efforts of the Non-aligned countries at the United Nations, the General
Assembly adopted a consensus statement on 24 November 1976, in which the parties decided inter alia, to have an
urgent ministerial level meeting to negotiate a fair and expeditious settlement.
Pursuant to the General Assembly’s consensus statement, negotiations continued between the two
countries, and the Ganges Waters Agreement was signed on 5 November 1977 for sharing the Ganges water during
the dry months from January to May each year for a period of five years (1978-1982). The Agreement provided
34,500 cusecs for Bangladesh and 20,500 cusecs for Calcutta port in India and guaranteed a minimum flow of
27,600 cusecs for Bangladesh during the lowest 10-day average flow period between 21-30 April. The Agreement
specified inter alia, a schedule for sharing of the flow on a 10 day basis; the guarantee of a minimum of 80 percent
of the amounts mentioned in the schedule for Bangladesh to protect against withdrawals from upstream reaches; the
formation of a joint Committee comprising representatives of both the governments who would implement sharing
arrangements; the carrying out investigations leading to augmentation of the Ganges flow at Farakka within three
years by the Joint Rivers Commission (JRC) created in 1972; and a review of the Agreement by the two
Governments at the end of the third year. During the Agreement period, Bangladesh and India exchanged their
proposals for the augmentation of the Ganges flow, which were reviewed jointly by both parties. The Bangladesh
proposal envisaged optimum utilisation of the water sources of the Ganges basin by the construction of storage dams
in the upper reaches of the Ganges in India and its tributaries in Nepal to conserve the monsoon flows for
augmenting dry season flows. India’s proposal was to transfer water from the Brahmaputra through a link canal
across Bangladesh to the Ganges above Farkka. The proposal of each country was not acceptable to the other, and
on the grounds that the obligation under the Agreement was not fulfilled, the Agreement was not renewed in
November 1982. However, the governments of Bangladesh and India did sign a Memorandum of Understanding
(MOU) on 7 October 1982.
The MOU of 1982 included provisions for burden sharing instead of achieving a guaranteed minimum
flow. Besides sharing, the MOU asked for submission of pre-feasibility studies of the augmentation schemes
proposed by both parties. Accordingly in 1983, the countries exchanged their updated proposals for augmentation,
based on the “Report on Pre-feasibility Study of the India and Bangladesh Proposals for Augmenting the Dry Season
Flows of the Ganges at Farakka” that was completed in 1984. In view of the differences of opinion, it was not
possible to make any recommendation acceptable to both countries with regard to the optimum solution for
augmentation of the dry season flows of the Ganges at Farakka, which could be rapidly implemented. Thus, the two
sides rejected the proposals in March 1984. The MOU expired in 1984 and as no formal arrangements for sharing
the flows in future years were made, Bangladesh made serious efforts to reach a formal agreement but without
success.
In the absence of the agreement, there was no sharing during the dry season of 1985. After painstaking
efforts, the two countries signed another MOU on 22 November 1985 for sharing the waters of the Ganges at
Farakka for three dry seasons commencing in 1986. Like its predecessor, the MOU of 1985 included provision for
burden sharing instead of guaranteed flows. It also provided for negotiation on the flows of all common border
rivers between the two-country including the Ganges. The sharing of the Ganges flows under the MOU of 1985
ended in May 1988. Between 1988 to 1996, in the absence of any sharing arrangement, Indian’s unilateral
withdrawal of the Ganges waters during the dry season at Farkka continued at an alarming rate which was evident
from the dry season flows of the Ganges at Hardinge Bridge. The Gorai, which is the main tributary carrying water
to the south-west region of Bangladesh, became totally dry at the beginning of the lean period. The progressive
utilisation of the waters upstream in India, coupled with the massive unilateral withdrawal of Farakka Barrage,
culminated in the lowest recorded flow a Hardinge Bridge of 13,521 cusecs in 1992, compared with historical
average flows of 75,000 cusecs during the last ten-day period of March.
During the New Delhi summit, held on 29 September 1988 between the heads of governments of
Bangladesh and India, the Bangladesh Secretary of Irrigation and the India Secretary of Water Resources were
assigned to work out an integrated formula for the permanent, long-term sharing of the flows of common rivers
between Bangladesh and India. In order to break the stalemate in the sharing arrangement, the Secretaries
Committee held six meetings alternately at Dhaka and New Delhi over three years from April 1990 to February
1992. They emphasized the need for immediate allocation of the Ganges and Teesta (another common river of
Bangladesh and India) waters on a priority basis, including the sharing of waters from other common rivers as
mandated. However, they could not decide on a general principle for the sharing of the waters at any of the common
rivers, including the Ganges (Nishat, 1996).
BASIS OF THE TREATY OF 1996
Treaty ‘96 is based on the principle of reasonable and equitable sharing of water and the river basin approach.
Non-Navigational Laws and the Helsinki Rules of 1966 are followed accordingly to formulate the treaty. Treaty’96
adopts Article IV of the Helsinki Rules: “Each Basin State is entitled, within its territory, to a reasonable and
equitable share in the beneficial uses of the waters of an international drainage basin.” According to the treaty, the
total availability of water will be measured and shared at Farakka (India) on the basis of the previous 40 years of
historical 10-day average flows. There are also some provisions in the treaty for both parties to discuss options for
the augmentation of the Ganges flow during the dry period and the way in which optimum allocation of waters of
other common rivers between Bangladesh and India is possible These are guided by the principles of equity,
fairness, and no harm to either party.
METHODOLOGY
In this study, hydrological analysis has been done using some control curves at Farakka to analyse and
compare the flows of different years. A control curve is a line of some availability of flow plotted against a
particular duration. Control curves for 10, 50, 75 and 90 percent flow duration were drawn based on Gumbel’s
method. Gumbel (1941) introduced the extreme value distribution, and it is commonly known as Gumbel's
distribution (Gumbel’s method). It is one of the most widely used probability-distribution functions for extreme
values in hydrologic and meteorological studies and is used to estimate the magnitudes of probably flood peaks,
maximum rainfalls, maximum wind speed, and other exteme events.
According to his theory of extreme event, the probability of occurrence of an event equal to or larger than a
value is
0
x
y
e
e
x
XP
−
−
−
=
≥ 1)
(
0
y
in which is a dimensionless variable given by
)
(ax
y
−
= α
)
(ax
y
−
= α
where
x
xa
σ
45005
.0
−
=
and
x
σ
α
/
2825
.1
=
.
Noting that the return period T = 1/P and the value of the variate X with a return period T is
x
T
Kx
x
σ
+
=
where
n
n
T
sy
y
K
)
(
−
=
in which, T
y
= reduced variate, a function of T and is given by
)]1
/(
ln.[ln
−
−
=
TT
yT
or
)],1
/
log(
log
303.2
834.0[
−
+
−
=
TT
yT
n
y
= reduced mean, a function of sample size N, for N (( n
y
(0.577, n
s
=reduced standard deviation, a function of
sample size N for N (( (1.2825. In practice the value of the variate X with a recurrence interval T is used as
n
s
1
−
+
=
n
T
Kx
x
σ
where
1
−
n
σ
= standard deviation of the sample (Subramanya,1994)
ANALYSIS AND RESULT
Control curves of 10, 50, 75 and 90 percent available flow have been drawn and the discharges of 1978, 1983,
1986, 1997 and 1998 at Farakka to Bangladesh are compared with these curves in figure 1. The top most control
curve is for the flows available 10 percent of the time and the bottom one is for flows available 90 percent of the
time. The figure shows that flows in 1978 and 1986 are above the 75 percent availability line whereas in 1983 and
1997 flow lines are below the average available 75 percent of the time, and at times below the 90 percent available
flow. The 1998 flow line is partially above and below than 75 percent available flow.
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
JA
N
I
JA
N
II
JA
N
III
FEB
I
FEB
II
FEB
III
MA
R
I
MA
R
II
M
A
R
III
APR
I
APR
II
A
PR
III
MA
Y
I
MA
Y
II
MA
Y
III
Periods: 10 days interval
Flow
in cusecs
10% available flow
50% available flow
75% available flow
90% available flow
Flow line 1978
Flow line 1983
Flow line 1986
Flow line 1997
Flow line 1998
Figure 1: Comparing the flow line with control curves at Farakka
The Agreement in 1977 was made based on 75 percent availability of flow, which was followed by MOU82
and MOU85. So accordingly, the Agreement of 1977 is well implemented in the year of 1978 but not in 1983. The
flows in 1997 and in 1998 are somewhat less than 75 percent availability of flow, but it can not be concluded that
the Treaty’96 is not well implemented because it is based on the historical average flow of the Ganges at Farakka
from 1949 to 1988. But it is true that in these years the flows are sometimes less than that of 1978, just considering
this point it can’t be said that the Treaty’96 is not well implemented in these two years as the share of both countries
Table2: Flow at Farakka in 1997 and 1998
PERIOD
Actual flow
release to
Bangladesh
in 1997
(cusecs)
Share
of
Bangladesh as
per
formula
given
in
Annexure-I of
1996 Treaty in
1997
(cusecs)
Difference in
flow between
actual and
share
of
Bangladesh
in 1997
(cusecs)
Actual flow
release
to
Bangladesh in
1998
(cusecs)
Share
of
Bangladesh as
per
formula
given
in
Annexure-I of
1996 Treaty in
1998
(cusecs)
Difference in
flow between
actual and
share
of
Bangladesh
(cusecs)
1
2
3
4=2-3
5
6
7=5-6
January 01-10
62019
62180
-161
164763
164797
-34
January 11-20
49556
49635
-79
135591
135566
25
January 21-31
48884
48672
212
105881
105866
15
February 01-10
45604
45604
0
88181
88186
-5
February 11-20
41029
41015
14
61831
61841
-10
February 21-28
38387
37399
988
54711
54738
-27
March 01-10
33489
33085
404
45322
45323
-1
March 11-20
35028
35000
28
37323
35967
1356
March 21-31
16528
13487
3041
36557
35000
1557
April 01-10
30137
35000
-4863
40474
38588
1886
April 11-20
25613
19526
6087
55952
50955
4997
April 21-30
35065
35000
65
47876
47901
-25
May 01-10
31722
31728
-6
72185
62203
9982
May 11-20
33021
33028
-7
82062
82062
0
May 21-31
31643
31654
-11
81218
81220
-2
Total flow
557725
552013
5712
1109927
1090213
19714
(Bangladesh and India) depends on the total available flow at Farakka and the basis of this treaty is different from
the agreement of 1978 . To comment about the degree of implementation of the treaty it would be a fair comparison
if a table which can directly compare between the shared value and released value of flow is drawn herewith.
Presentation of such a table (table2) of flows could provide a clear understanding about the flow condition in 1997
and 1998.
Table 2 clearly shows that in 1997 the total flows released to Bangladesh exceed that required by the treaty
by 5712 cusecs, although in January I, II, April I and in all of May releases were less than the quantity fixed by the
treaty. The same situation occurred in 1998. So the study shows that during the first two years after the treaty,
Bangladesh received an excess amount of water on the whole according to the Treaty’96 based upon the total
available water at Farakka. Since the share of both countries depends upon the water available at Farakka flows need
to be augmented upstream as the released amounts of water are not sufficient during dry periods for some specific
uses. It is also inadequate to restore the environment and other conditions and to mitigate the harmful effects of
reduced flows for a long time.
CONCLUSION
Water is a sensitive and scare resource in the 21st century, and the distribution of it among the co riparian
states is a complicated and difficult task. A river system has to be considered as a whole from its source to its mouth
for the optimum development of its water resources to get the best results. There has to be a broad outline of a plan
covering the entire drainage basin to ensure the coordinated and harmonious development of the various works in
relation to all the reasonable possibilities of the basin. Concisely, it can be concluded that theTreaty’96 has
performed well during the first two years, though the flow reaches a very low level in March and April because of
the inadequate available flow at Farakka. After reviewing the treaty 96’, it is obviously that there is no way but to
consider the entire drainage basin as a whole to ensure the optimum allocation of water for the benefit of each co-
riparian state. Doing so, can mitigate the problems that have become increasingly serious over the years as a result
of reduced flow and can guarantee a healthy economy of the co riparian states.
ACKNOWLEDGEMENTS
The authors thank the Joint River Commission for providing the 10-day average flow data for the Ganges at Farakka
and at Hardinge Bridge
REFERENCES
Abbas, B. M. A. T., 1984. The Ganges Water Dispute. University Press Limited, Dhaka.
Crow, B., 1995. Sharing the Ganges; the Politics and Technology of River Development. University Press Limited,
Dhaka.
Nishat. A., 1996. From Ganges- Brahmaputra to Mekong, pp. 60-80. In Hashimoto, T., and A. K. Biswas, editors,
Asian International Waters. Water Resources Management, Series: 4, Oxford University Press: Bombay, Delhi,
Calcutta, Madras. ,
Subramanya, K., 1994. Engineering Hydrology, Tata McGraw-Hill Publishing Company Limited, New Delhi,
APPENDIX
THE TREATY OF 1996
TREATY BETWEEN THE GOVERNMENT OF THE PEOPLE'S REPUBLIC OF BANGLADESH AND THE
GOVERNMENT OF THE REPUBLIC OF INDIA ON SHARING OF THE GANGA / GANGES VATECS AT
FARAKKA.
THE GOVERNMENT OF THE PEOPLE'S REPUBLIC OF BANGLADESH AND THE GOVERNMENT OF THE
REPUBLIC OF INDIA.
DETERNNED to promote and strengthen their relations of friendship and good neighbourliness.
INSPIRED by the common desire of promoting the well being of their peoples.
BEING desirous of sharing by mutual agreement the waters of the international rivers flowing through the territories
or the two countries and of making the optimum utilisation of the water resources of their region in the fields of
flood management irrigation, river basin development, and generation of hydro-power for the mutual interests of the
peoples of the two countries.
BEING desirous of finding a fair and just solution without affecting the rights and entitlement of either country
other than those covered by this Treaty, or establishing any general principles of law or precedent.
HAVE AGREED AS FOLLOWS:
ARTICLE-I
The quantum of waters agreed to be released by India to Bangladesh will be at Farakka.
ARTICLE-II
(i) The sharing between India and Bangladesh or the Ganga / Ganges waters at Farraka by ten day periods from the
1st January to the 31st May every year will be with reference to the formula at Annexure I and an indicative schedule
giving the implications of the sharing arrangement under Annexure I is at Annexure II.
(ii) The indicative schedule at Annexure II as refereed to in sub Para (i) above is based on 40 years (1949-1988) 10-
day period average availability of water at Farakka. Every effort would be made by the upper riparian to protect
flows of water at Farakka as in the 10 years average availability as mentioned above.
(iii) In the event flow at Farakka falls below 50,000 cusecs in any 10-day period, the two governments will enter
into immediate consultations to make adjustments on an emergency basis, in accordance with the principles of
equity, fair play and no harm to either party
ARTICLE-III
The waters released to Bangladesh at Farakka under Article I shall not be reduced below Farakka except for
reasonable uses of waters, not exceeding 200 cuseces, by India between Farakka and the point on the Ganga/Ganges
where both its banks are in Bangladesh.
ARTICLE-IV
A committee consisting of representatives nominated by the two Governments in equal numbers (hereinafter called
the joint Committee) shall be constituted following the signing of this treaty. The joint committee shall set up
suitable teams at Farakka and Hardinge Bridge to observe and record at Farakka the daily flows below Farakka
Barrage, in the Feeder Canal, and at the Navigation Lock, as well as at the Hardinge Bridge.
ARTICLE-V
The joint Committee shall decide its own procedure and method of functioning.
ARTICLE-VI
The joint Committee shall submit to the two Governments all data collected by it and shall submit a yearly report to
both the Governments. Following submission of the reports the two Governments will meet at appropriate levels to
decide upon such further actions as may be needed.
ARTICLE-VII
The joint Committee shall be responsible for implementing the arrangements contained in this Treaty and examining
any difficulty arising out of the implementation of the above arrangements and of the operation of Farakka Barrage.
Any difference or dispute arising in this regard, if not resolved by the joint Committee, shall be referred to the Indo-
Bangladesh Joint Rivers Commission. If the difference or dispute remains unresolved, it shall be referred to the two
Governments, which shall meet urgently at the appropriate level to resolve it by mutual discussion.
ARTICLE- VIII
The two Governments recognize the need to cooperate with each other in finding a solution to the long-term
problem of augmenting the flows of the Ganga/Ganges during the dry season.
ARTICLE-IX
Guided by the principles of equity, fairness and no harm to either party, both the Governments agree to conclude
water sharing Treaties / Agreements with regard to other common rivers.
ARTICLE-X
The sharing arrangement under this Treaty shall be reviewed by the two Governments at five years interval or
earlier, as required by either party and needed adjustments, based on principles of equally, fairness, and no harm to
either party made thereto, if necessary .It would be open to either party seek the first review after two years to assess
the impact and working of the sharing arrangement as contained in this Treaty.
ARTICLE- XI
For the period of this Treaty, in the absence of mutual agreement on adjustments following reviews as mentioned in
Article X, India shall release downstream of Farakka Barrage, water at a rate not less than 90 percent (ninety
percent) of Bangladesh’s share according to the formula referred to in Article II, until such time mutually agreed
flows are decided upon.
ARTICLE - XII
This Treaty shall enter into force upon signature and shall remain in force for a period of thirty years and it shall be
renewable on the basis of mutual consent.
IN WITNESS WHEREOF the undersigned being duly authorised their to by the respective Governments, have
signed this Treaty.
DONE at New Delhi 12th December, 1996 in Hindi, Bangla and English languages. In the event of any conflict
between the texts, the English text shall prevail.
(SHEIKH HASINA)
(H.D. DEVE GOWDA)
PRIME MINISTER,
PRIME MINISTER,
PEOPLE'S REPUBILIC OF
REPUBLIC OF INDIA.
BANGLADESH.
ANNEXURE - 1
Availability at Farakka
Share of India
Share of Bangladesh
70,000 cusecs or less
50%
50%
70,000- 75,000 cusecs
Balance of flow
35,000 cuscs
75,000 cusces or more
40,000 cusecs
Balance of flow
Subject in the condition that India and Bangladesh each shall receive guaranteed 35,000 cusecs of water in
alternative three 10-day periods during the period March 1 to May 10
ANNEXURE – II
Schedule
(Sharing of waters at Farakka between January 1 and May 31 every year)
If actual availability corresponds to average flows of the period 1949 to 1988, the implication of the formula in
Annex -I for the share of each side is:
Period
Average of total flow
1949-88 (cusecs)
India's share (cusecs) Bangladesh share
(cusecs)
Jan
1-10
11-20
21-31
107,516
97,673
90,154
40,000
40,000
40,000
67,516
57,673
50,154
Feb
1-10
11-20
21-28
86,323
82,859
79,106
40,000
40,000
40,000
46,323
42,859
39,106
March
1-10
11-20
21-31
74,419
68,931
64,688
39,419
33,931
35,000*
35,000
35,000*
29,688
April
1-10
11-20
21-30
63,180
62,633
60,992
28,180
35,000*
25,992
35,000*
27,633
35,000*
May
1-10
11-20
21-31
67,351
73,590
81,854
35,000*
38,590
40,000
32,351
35,000
41,854
(* Three ten day periods during which 35,000 cusecs shall be provided)
__._,_.___
