Revisiting the Alternative Restructuring of Sardar Sarovar:

Suhas Paranjape and K. J. Joy

The deadlock over Sardar Sarovar continues. In a book published in 1995 we had put forward a proposal for a restructuring of the Sardar Sarovar Project (SSP) which we believed was `capable of satisfying the legitimate interests and demands of the main parties to the long standing dispute over the project, and can thus help break the current deadlock between the supporters and the critics of the project’. That was in 1995. In the five years and more since we started working on that alternative, a lot has happened and yet, one thing is quite clear -- the deadlock continues.

As we return to some of the themes of that book in this short paper, we are no longer so sanguine as to as to what it can do to resolve that deadlock. For it is quite clear that the conflict has turned into a war of attrition and both sides have dug themselves in with no middle ground conceded. A war of attrition, by its very logic, is most often forced by the side that feels itself stronger and more powerful in the hope that it will wear down the other side. What is remarkable is the fortitude and resolve displayed by the adivasis in the valley in this struggle, in an unequal struggle forced on them by a belligerent and intransigent, powerful adversary. It is only this tremendous resolve and the gathering opinion in their favour which may force a reconsideration of the project itself.

We are writing this short paper for a related, but different objective. We are writing it to highlight some of the different approaches which made it possible for us to work out an alternative at all and which gave us the strength to dare to present an alternative at all. For those are relevant, not only in respect of the SSP, but in a much wider context. And whatever the public stands adopted, we have found a positive attitude and keen interest in these ideas in a wide spectrum of both the pro- and anti-dam camps, a middle ground that the war of attrition does not allow to consolidate. In that context, the SSP alternative functions more as an illustration, a definite case in which they have been applied. In the book, they have been all presented as coming together to form an integrated alternative. Here we desegregate them and present them as worth consideration, each in its own light.

A summary of the alternative

Let us begin with a recapitulation of the main features of the alternative as compared with the current plan for the SSP.

Table : Comparison of Alternative Plan and the Current Plan



Alternative Plan

Current Plan


Storage level at SS dam

107 m (90 m baseline level)

140 m


Total Submergence

10,800 ha

36,000 ha



Drastic Reduction in displacement

1.5 lakh people displaced



Within the same area with assured share of Narmada water

Uprooted, rehabilitation in new area


Upstream Service Area

more than 1 lakh ha



Gujarat Service Area

41 lakh ha

18 lakh ha

(of which)


13.1 lakh ha (32%)

3.9 lakh ha (22%)


4.0 lakh ha (10%)

0.4 lakh ha ( 2%)

North Gujarat

14.7 lakh ha (36%)

3.1 lakh ha (17%)

Rest of Gujarat

8.9 lakh ha (22%)

10.6 lakh ha (59%)

. . . . contd.


contd. . . .

New electricity generation

850 MW
2600 MU

1400 MW
3600 MU

Consumed in the project

1646 MU

1138 MU

Peak load capacity

1200 MW

1400 MW

Gas-solar hybrid generation out of saving

200 MW
1750 MU



Surplus energy

at least 4410 MU (26.3 MT) produced as biomass

Not planned


Equitable water distribution & sustainable development

Basic issue

Not planned


Total cost

12,920 crores

13,000 crores

of which

Expenses on local employment & services

3,620 crores



Cost recovery

Based on distinction between basic and economic service

No such plan


Gujarat's total share of Narmada water




Loss of Forest

3,000 ha by submergence and 10,000 ha low grade forest for rehabilitation

13,700 ha substantial prime quality forest


Permanent vegetative cover in service area

11 lakh ha(23,000 ha in upstream contiguous to forest area)

No provision

The alternative proposal also ensures a delivery of 9 MAF of Narmada waters to Gujarat, but does so at a much smaller height and brings down the submergence to one-third of the current value. It retains the power benefit at levels very close to that of the present SSP plan and within the same order of cost. It greatly increases the share of Narmada water delivered to the drought prone regions of Gujarat without a substantial reduction in the incremental irrigation benefit to rest of Gujarat. This is made possible because it provides for the harnessing of local water almost equal to the water drawn from the Narmada which allows it to extend the service area from 18 Mha to 41 Mha. It provides for equitable and sustainable water use in the service area and for permanent tree cover over a little less than one-third of the service area of the project. And, lastly it provides for a dispersed biomass energy surplus of 26.3 MT coal equivalent which can become the basis for a transition to a dispersed industrial system.

Much more important are the various components of the alternative strategy which make such an alternative possible. We have discussed them in the book in three parts: the first related to the central conveyance and delivery system, the second related to the equitability and sustainability of the local system and the third related to an energy perspective which will allow for transition to a dispersed industrial system. We discuss them broadly in the same sequence below.

Dispersal of storage: integration of large and small

One of the important components of the alternative is the belief that exploitation of large sources is necessary if small systems are to provide a service with a high degree of reliability. It does not see the conflict as one of large versus small, but as a question of what is the relation between them. Planned and used properly, large sources can support smaller and local systems and increase their reliability as well as sustainability. The alternative proposal incorporates this feature.

Most large source planning identifies storage behind the dam as the main storage and systems are usually planned to make behind the dam storage equal to, or at least comparable to the total planned water use from the large source. The alternative breaks from this practice and suggests a dispersal of storage, with behind the dam storage mainly having the function of a regulatory storage. This means that dispersed local storages and their refilling serves as the main storage. This greatly reduces the storage behind the dam and brings down submergence while allowing utilisation of much larger quantities of water from the large source.

Integration of local and exogenous water

This also implies integration of local and exogenous water sources. In fact, the alternative provides for the development of local water resources and their integration to allow for a extension of the service area to almost double the planned service area. The basic principle behind this not new, as is the case with almost every aspect of the alternative. Indeed, we claim no originality in this respect. All we have done is integrate these aspects into planning of large systems. The so-called `system tanks’ in Tamil Nadu are an example of such integration. They are rainfed tanks which are refilled from large sources, and in fact studies have found that is greatly increases their reliability. What we have done is extend this principle and make it the basis of the alternative system. All that we do is to see to it that the local storages are filled and/or refilled by the start of rabi to mid-rabi season.

Dual role of small systems

What it does is to allow small systems to act in their dual capacity, as means to harness local water resources and as receptacles and as dispersed storage elements of water from the large source. Taking into consideration that local systems need to be built in their own right, the alternative then uses these systems as storage systems for the water from the large source, thus avoiding the duplication of cost which is involved in creating behind the dam storage for this component and keeping it separated from the large system by the conceptual as well as administrative Chinese wall that separates small and large systems. This in fact is the greatest cost saving measure of the alternative. It starts with an integrated view of the large and the small and makes the large system a supporting and strengthening system for the small system.

Feeder canals

Obviously, this needs a different kind of delivery and conveyance system. Instead of a large centralised `command area’ it needs a collection of dispersed `service areas’ built around local sources to which water in conveyed by `feeder canals’. They are large canals and their main function is speedy conveyance to the local storage systems. The sphere of the large system thus does not/need not extend to the individual farmers fields, but is limited to supplying the water to the local and smaller system, cutting down on the top-heavy, centralised command area system. Here too there is a simplification and rationalisation of the system.

Local storage potential is very large

It is not often realised that local storages built are mainly limited by yield. Actual storage potentials are much higher. If exogenous water is available, storages with lower dependability can be accommodated, since filling and refilling from exogenous water is possible. In the alternative we had limited ourselves to smaller systems of less than 100 Mm3 in estimating the recipient storages. In a truly integrated system this constraint is not necessary, and even medium projects can be used as recipient storages with extension of their service areas. If we do so, recipient surface storage potential itself will be very large.

Run-of-the-river operation of hydro power plants

The second important change proposed in the alternative is in respect of the power plant operation at SSP. In the conventional operation of the hydro plant as a peak load plant based on behind the dam storage is maintained, then obviously almost the entire power benefit would be lost, both due to the virtual elimination of behind the dam storage in the alternative, as well as due to the reduction in the height of the dam. However, the alternative suggests a run-of-the-river (RoR) operation of the power plant during the monsoon season, and a pumped hydro operation for peak load in the post-monsoon season. This fully preserves the peak load benefit from the plant, and comes close to the new energy generation benefit of the present SSP.

Here again, we should emphasise that we have not incorporated this aspect in order to preserve the energy benefit at SSP alone. Monsoon RoR operation combined with post-monsoon peak load pumped storage operation has its own independent benefit. In fact, our suggestion, irrespective of SSP, would be to add an RoR element to all irrigation projects in all size ranges which at the moment have storages and considerable monsoon flows but have no power project component. Such a measure would build a truly large, dispersed peak load capacity for the power sector.

A basic change in this respect would also help absorb some of the costs of switching over to the alternative. It would be necessary to modify some of the power equipment so as to adapt it to the change in head which would result from the lowering of the dam height. Though such changes would be quite easy to incorporate without substantial change in cost if the equipment were to be ordered it is very difficult and costly, if not impossible to incorporate these changes in the equipment already fabricated according to earlier specifications and awaiting delivery. However, it would not be difficult to incorporate this equipment individually for the proposed RoR at other sites in irrigation projects where heads are comparable, monsoon flows are adequate at least for capacity peak load operation but the project at present has no power project component.

Implication for post-monsoon flows: bound and unbound flows

Both the elements described above, conveyance and delivery to local storages through feeder canals as well as monsoon RoR followed by post-monsoon pumped storage peak load operation have important implications for the riparian environmental regime. Since a major portion of the water is conveyed to the local systems in the late monsoon and immediate post monsoon season, and all of it by the mid-rabi season, a major portion of the post-monsoon flows are left unbound. Similarly, since the peak load operation in the post-monsoon season is planned on pumped storage and is not dependent on the post-monsoon riparian flows, power generation too leaves the post-monsoon flows unbound. This means that the post-monsoon environmental regime downstream of the dam is largely preserved, which is an important consideration in its own right.

Impact on rehabilitation

The problem of rehabilitation with the present SSP plan is simply intractable. And the recent seemingly progressive step taken by the authorities where they assure the oustees that they are prepared to buy at market prices any piece of land for shows this up. For this is a tacit admission that they cannot find or acquire the requisite land for the oustees, and that it is now for the oustees to locate the land and to negotiate a settlement for acquiring the land. It is, in short, an abdication of responsibility and an admission of the intractability of the problem. In fact, the present SSP plan is unacceptable simply on the basis of the scale of the human, social and economic cost of rehabilitation, a cost that has to be borne by the adivasis in the valley.

The alternative, first of all, brings down the scale of the problem to make it tractable. It brings down submergence by almost 70%. More important is the impact it has on displacement as distinct from submergence alone. The major human cost of rehabilitation is directly related to the uprooting it causes, the number of people who lose all or major portion of their lands, and have to be resettled in an entirely new area -- the number of people who are, literally, uprooted. It grows rapidly as the contiguous area of submergence increases and increases, or decreases, much more rapidly than the increase, or corresponding decrease, in area. Thus the reduction in the number of people uprooted would be expected to be much more than 70%, and though it is difficult to say how much more until we have very detailed data, we would say that in comparison with the present SSP plan, the people uprooted from their lives would come down by almost 90%.

The second aspect of the alternative which relates to impact on rehabilitation is where the oustees are to be rehabilitated. It is usual for the project affected, necessarily from upstream areas, to be given land for rehabilitation in the downstream command areas. The objective behind this is laudable enough. They have to be given irrigated land from the same project and it has to be acquired from the beneficiaries. And generally, if the oustees in the upstream areas and the beneficiaries in the downstream command are bound together by ties of kith and kin and culture, there is a favourable environment for tackling the issue, though nothing can finally replace the organised struggle and resolve of the oustees. The less they share the sharper the conflict becomes, and when these boundaries coincide with the boundaries which demarcate the dominant and the dominated, the better off and the oppressed the problem becomes intractable. The rehabilitation of oustees in the downstream areas then means not only an uprooting but also their further dispersal and their being thrown into a hostile environment.

For this reason, the alternative aims at a rehabilitation of the oustees in the upstream area itself, in their own socio-cultural milieu and sphere. But for this the numbers have to be on a scale where this becomes possible. It is literally impossible to do this if we consider the requirements of rehabilitation of the present SSP plan. By bringing down the submergence and the number of people displaced, the alternative first of makes it possible to visualise rehabilitation within their own socio-cultural milieu. The book provides estimates and makes out a case for such a rehabilitation.

For that obviously, brining down submergence is necessary but it is not enough. There have to be other positive measures which have to be thought of. Why should the people in upstream areas accept this kind of rehabilitation when they themselves are finding it hard to make ends meet, when they are already living at the margins, in an environment that is fast getting degraded, when their own livelihoods are being eroded? And, since it is usual practice to restrict benefit to downstream command areas which can be irrigated by gravity flows, there is virtually no benefit to the upstream areas. In such a situation, there is likely to be equal hostility to the rehabilitation of oustees in the upstream areas as well.

The alternative, therefore provides for an influence zone of one lakh hectares of the upstream area within which the rehabilitation is to take place. It makes a provision of Narmada water on more favourable terms for this entire influence zone including the oustees and provides for pumping energy as project running cost from the energy generation. The water is drawn from the MP share and the cost is charged to Gujarat who is the main beneficiary. In effect, rehabilitation is made part of a positive sum game for this entire zone. It should be emphasised, as we have done in the book, that the parameters for the choice of this influence zone are only minimally technical (in the sense that it should constitute a sufficiently large contiguous upstream area including the submerged area and its immediate surroundings). How large this zone should be and what the positive measures for it should be are a matter of social choice and negotiation and our assumptions in this respect are illustrative and represent only what is minimally necessary.

Equity and sustainability

Equity and sustainability are issues which transcend the whole question of large versus the small. Small systems are no less iniquitous or unsustainable because more often than not they are informed by the same production paradigms. Inequity between those who have and those who do not have the means to access and abstract water merges with socio-economic inequity as, if not much more, closely in small systems than in large systems. Unregulated water use is as common in small as in large systems. And small systems engaged in a pumping race in water scarce areas have caused serious groundwater depletion, often accentuated further by the provision of electricity at subsidised tariffs unrelated to intensity of use. Small systems stand as much in need of well articulated policies pertaining to equity and sustainability as the large and medium systems and are not automatically equitable and sustainable.

Role of exogenous water

Further, our extensive work in the drought prone regions has led us to the firm belief that in most of them limited but definite quantities of exogenous water are sorely needed if livelihoods in those regions are to be stabilised and assured with an adequate degree of dependability, which we assume to be around 80%. This can only come about if exogenous water supplies are planned to provide this support and supplement to local water resources in these areas. Unfortunately, water from larger exogenous sources is planned independently, without reference to and without integration with local resources and this is a cause of inequity and unsustainability in both systems. The alternative is an effort to combine them in a way that brings about equity and sustainability.

Conditions of exogenous water use

The first step in this direction, we feel, is to make exogenous water available on certain conditions to be fulfilled by the users and the state mutually in respect of equity and sustainability. At present, no such conditions are required and we have a peculiar situation in which exogenous water is robbed of the strengthening and supplementing role that it can play in respect of local water resources, so that they always fall short of the requirements of sustainable livelihoods and increase the pressure on the environment in areas which they do not serve, while in the areas they do serve they become a license to waste and to environmental degradation arising from it.

We therefore make provision of Narmada water conditional on the fulfilment of four conditions: a) that the Narmada water will be available to the local systems in proportion to the local resources they harvest and harness (for most regions we have proposed a ratio of 1, that 1 m3 of Narmada water to be provided for every m3 of local resource created); b) Equitable water access, that is the minimum livelihood water to be assured to all families in the service area and only after that is water to be provided for extra, economic service; c) one-third of the service area to be brought under permanent cover (the provision of water at b) above makes this feasible) and d) self management of the system by the users to ensure equitable and sustainable use.

Each of these provisions has a definite purpose. The first provision at a) ties together the development of local and exogenous resources and ensures that local systems will not simply die and be replaced by exogenous water when it enters the area, which is what happens today. The second provision at b) ensures that within the service area of the project the right to water is not dependent on the users’ landholding, but is part of the right to livelihood vested in the individuals and not in their property and in proportion to it. The third provision at c) is made possible by the provision at b) and it ensures that a minimum basis of environmental upgradation is established, in the entire service area. And the fourth provision at d) is what makes it possible for the local water users community to fulfil these conditions. This provision is as important as the others, if not more so, because unless the users themselves and not the anonymous functionaries of a top-heavy bureaucracy exercise control over their own water use they are in no position to fulfil these conditions.

These state too has corresponding obligations in respect of each of these conditions above. The state has to provide the requisite funds necessary for the harvesting and harnessing of local resources mentioned at a) above. In fact, this is why in the alternative minimal watershed treatment of an area equivalent to twice the service area is included in the project cost itself. It implies that any irrigation project that comes up must walk on these two legs and make provisions for both of them. The next provisions at b), c) and d) imply that the state has to enable the local communities to ensure that they are sufficiently empowered to satisfy these conditions. This implies enabling legislation, policies and incentives.

In this short paper which is a reiteration of some of the themes the alternative pursues and tries to bring forward, there is no scope for discussing these in any great detail. The book discusses these issues in some detail and there is an ongoing discussion on many of these issues which is available in various places. There is a need to continue this discussion and bring it to a point where water users, present and prospective, come to a consensus and jointly struggle for those policies. What we would emphasise here is that this aspect of the alternative is relevant, not only to the SSP, but to all irrigation projects in the country, to the water system as a whole.

Energy : The final currency

One important dimension which runs through all the three parts of the alternative is the question of energy and is crucial to it. The first departure from unwritten conventions takes place when it provides for pumping, for the lifts to Saurashtra and Kutch, to the upstream areas of the dam as well as to the areas upstream of the so-called gravity command of the canals. This violates the usual, unwritten dogma in project planning by which all canal deliveries, barring small and exceptional areas, take place by gravity. This unwritten dogma has a very large part to play in the iniquitous nature of canal commands. First it automatically excluded all areas upstream of canal alignments. Also it favoured people with holdings in lower areas over people with holdings in the higher reaches, which most often means favouring the holdings of the relatively better off. More seriously, it obstructs any attempt to construct equitable access based on watershed or village as unit because it inevitably cuts across them. Equitable water access therefore presupposes either that everyone shares in the command like the old phad systems in Maharashtra or that water is conveyed to everyone’s field, even if it involves pumping. In that sense redistribution of access involves costs, both energy and economic costs and an unwillingness to pay those costs means a denial of equity.

The alternative provides for this cost and at the same attempts to see to it that this cost is sustainable. A prime question in this respect is the question of where is the energy to come from. The alternative provides for this energy in two ways. Firstly, it earmarks priority use of the power generation to meet the pumping costs of the conveyance and delivery systems. Secondly, it recommends a biomass perspective, a perspective that takes into account that all the water use is resulting in the production of biomass, potentially, biomass energy. The book provides estimates which go to show that for every unit of water pumped in this manner to provide equitable access, only a small proportion of the incremental biomass production is needed for even large lifts. In that sense every unit of water pumped produces an energy surplus that more than pays for itself. Provided, of course, that sufficient cognisance is taken of biomass as energy and policies are planned around it. In fact, once one-third of the service area is brought under permanent cover, the wastes from that area, the shrubwood and brushwood can also pay for the pumping energy.

Energy is of course more than a simple cost. It is the underlying currency of all modern industry and it also opens up a path to a dispersed industrial system. The magnitude of this resource can be estimated variously, and what we have adopted is to count the biomass surplus which remains after fulfilment of all the basic needs which can be met from the local biomass resource as a measure of that potential. We have estimated it in the least effective use it has, as fuel, where it has the lowest energy replacing effect -- and the estimated annual surplus comes to almost 4500 MU of power, more than the maximum transitional energy generation at SSP. Put to better uses this can represent an energy resource of anything up to three times it fuel replacement value, that is almost 5 times the energy generation at SSP itself.

The modules for the biomass energy generation cum production units which we have presented in the book have come a long way since. Much better efficiencies of hybrid systems, solar collectors along the lines we have outlined have been achieved by our colleagues and that whole section would have to be rewritten today and would stand on much firmer ground that it did then. But this is not the place to go into those details. In general, we have devoted much less space here to the discussions of Parts Two and Three of our book: firstly, because of the limited scope of this short paper, and secondly, because they are of much wider significance and are being discussed in many more forums where they can be dealt with adequately and in detail. What we would again point out, before we conclude this paper, is their importance in their own right.

A last plea

We cannot , however, avoid the temptation of making a last plea before we conclude. We had pointed out in our book that though we had made all efforts in the alternative to incorporate as much of the cost incurred in the project into the alternative, as the project progresses it makes many more of its features irreversible and difficult to accommodate and more and more of the social cost becomes unsalvageable. A crucial stage in this respect is when the dam reaches a height of 90 m, the baseline figure for the alternative. With every metre of construction beyond that point, the alternative will have to be recalibrated to the new level and for every metre it will lose that much of its optimality.

We would therefore, like to make a last plea as we stand on the edge of this possibility. The main driving force behind the alternative was to bridge the interests of the drought prone regions and those of the adivasis in the valley. The present SSP has a regional allocation of Narmada water which leaves it open to the charge that it is nothing but an attempt to use the needs of the drought prone regions in order to provide water to the already water rich regions of Central Gujarat. This happens in two ways. The lions share of the Narmada water goes not to the drought prone regions but to the rest of Gujarat. Secondly, because the delivery to the drought prone regions is from the tail end of the central Gujarat command, deliveries to those regions take place, even in the planned schedules more than a decade after the deliveries to the rest of Gujarat begin. And if we take into account the experience of large commands where deliveries begin from the head reaches, the tail portions of the command at best receive much less than their allocation or at worst do not receive water at all.

Another of our concerns has been to ensure that deliveries to the drought prone regions have independent and simultaneous channels of delivery. We have in our book suggested a modified proposal for deliveries to Kutch and Saurashtra. This provides for the use of the already built canal network up to Shedhi (or Mahi), releasing the Kutch and Saurashtra allocation into the Shedhi (or the Mahi), making use of the natural water course of the Shedhi-Sabarmati (or Mahi) system and picking it up a suitable low point from the Sabarmati (or Mahi) and conveying it to Kutch and Saurashtra. This allows simultaneous deliveries to be made to all the regions of Gujarat service area (taking into account the Kadana replacement suggested in the alternative).

We are quite aware of the feeling in Gujarat that all the attempts to get the SSP reviewed are but excuses to stall the project in one way or the other. While this is neither fair nor correct, there are historical reasons for it. What we would suggest is that the alternative conveyance to Kutch and Saurashtra should be built speedily, so that at the dam height of 90 m, deliveries can begin to the different regions of Gujarat. So the project will already be in a stage where deliveries can be made to any area of Gujarat if it is so decided. Thus the project is no longer `stalled’ and a commitment to deliver Narmada water to all areas of Gujarat would have been demonstrated. But we would enjoin on everyone concerned, and especially on the people of Gujarat, to take a pause and seriously reconsider and review the project, see to it that an optimal plan which satisfies the legitimate interests of the people in the drought prone regions of Gujarat as well as of the adivasis in the Narmada valley emerges with their consent. If this does not take place, we are afraid, their will be no winners, neither the people in the drought prone regions of Gujarat nor the adivasis in the Narmada valley. The SSP will become nothing more than a monument to our colossal callousness.

Revisiting the Alternative Restructuring of Sardar Sarovar p. 7
Suhas Paranjape and K. J. Joy