The Other Water Problem

Climate Change may not only be about rising sea levels but also about reduced freshwater availability. The New York Times has an article about the USA. I wonder if we have such studies and popular articles about India.

Water Fiction

In earlier posts I have referred to data about water availability and utilisation without looking at it in any great detail. The estimates, made by government bodies in turn form the basis of many papers, essays and analysis by administrators, researchers and others (see 1 and 2). In a recent report the Planning Commission too uses the same numbers.

While I had also pointed out several conceptual deficiencies in the estimates I had not examined the data. It is time to do so.

It is instructive here to quote from A Vaidyanathan at some length. In his latest book, India’s Water Resources, while discussing surface flows as estimated by the CWC, he writes:

“………….These estimates have gaps and limitations. The flow observations sites are known to be relatively few, located along the main rivers of each basin and, until recently measurement devices were relatively crude. Locations for which long time series are available are even fewer. Releases from all major and medium reservoirs are supposed to be measured and recorded regularly. However, this information is not available for all reservoirs and in such cases utilisation is estimated on the basis of reported irrigated area.

More importantly, extensive changes have taken place in land use patterns: forest area has declined and the quality of the forest cover has deteriorated; extension of cultivation to marginal lands and soil erosion has led to degradation of land; and natural drainage channels (have) been disrupted by the expansion of road and rail networks and urbanization. These changes are likely to make a significant difference to the quantum, duration, and seasonal profile of surface flows. But their impact on the above has not been adequately and systematically studied. ………….”(pg. 34)

While noting that CWC estimates of utilizable surface flows relate mostly to large storages, he remarks about small structures:

“……….Their extent has been increasing and there is reason to believe that official statistics understate the area irrigated and volume of water harnessed. ……”(pg. 34)

And about Central Ground Water Board estimates of natural recharge from rainfall and recharge from irrigated areas of groundwater, he says:

“..… The Commission (NCIWRD) draws attention to the limitations of these estimates and the need for further research and observation to improve them. While it does not modify the estimate of overall recharge, their projections assume, rather arbitrarily and without citing any reason, that only two-thirds of the total recharge is utilizable in all the basins. ……………”(pg. 34-35)

It would thus seem that the estimate of aggregate water availability and utilisation as well as its categorisation as surface flows and groundwater is based on faulty or inadequate measurements and untenable assumptions.

For several purposes, water flows (irrespective of source and sink) can be treated as one system as the distinction between groundwater and surface flows may not be relevant. Remember water continually flows between the two – surface and ground – and this is not merely natural but human induced as well.

While the overall quantum of water that falls in the form of rain and snow (the starting point of water estimates) is known with a great degree of certainty, evaporation rates and return-flows, to name two important variables need to be measured adequately to estimate aggregate availability and utilisation. However, in several instances it is not only flow quantities but also withdrawals from stocks that matter as for instance if groundwater levels are falling over large tracts of land area.

In other words not only we don’t quite know how much of the total water is flowing on the surface and how much of it is seeping into the ground but that the estimates of both aggregate availability and utilisation too are fraught with severe problems of data that it seems little more than guess work.

Precise estimates as are put out by official agencies should be taken with more than a pinch of salt.

Elementary Errors Compounded

In the Handbook of Water Resources in India (2007) several authors bemoan the low per capita water storage in India. These include John Briscoe and RPS Malik (pg. 2), A Sekhar (pg. 63) and RPS Malik (pg. 142), well known and influential names.

Mr. Sekhar, Advisor to the Planning Commission, even says that one of the main reasons for water problems in the country is the low per capita storage. He adds that India has no option but to go ahead with its dam construction programme. This is quite incredible for in the vast number of articles in book (including by these authors and others) the problems discussed and the solutions offered are quite different.

Nevertheless, it may well be that the authors are merely inconsistent in their views. The question may still be posed. Is there merit in the augmenting water storage?

The reason proffered by these authors is that rainfall is concentrated in a few monsoon months making storage inevitable. Further, the per capita water storage in India at about 265 cubic metres (see Tables 1.8 &1.9 and Chart 4 of this report) is much less than that of other countries. While Briscoe and Malik say that arid-rich countries like USA and Australia have over 5,000 cubic metres (m^3), middle-income countries like South Africa, Mexico, Morocco and China have a storage capacity of about 1,000 m^3. Interestingly, Mr. Malik in a later chapter offers somewhat different figures – 1964 (USA) and 753 (South Africa). He also points out that dams on the Colorado (USA) and Murray-Darling (Australia) can store 900 days of river flow; India can store only about 30 days of rainfall.

The comparison is with a select list of countries (why only these and not other countries make it to their list and why global averages are not presented is a moot point but I will let that pass for the moment). The interesting that is that most of the countries mentioned - Australia, Spain, China, USA and South Africa - face varying levels of acute and chronic water-related problems including that of scarcity. Water storage hardly seems such a silver bullet after all.

Mr. Sekhar also suggests the need for a storage capacity of 750–1,000 m^3 per capita though the rationale of that figure is not explained.

In any case, what matters is the ability of a country to meet its requirements - to provide water - and not its storage capability. India’s per capita availability of 1,700 m^3 per capita per annum is considered quite comfortable. As I pointed out earlier, several countries with higher storage have not solved their problems while several countries (e.g. in the Gulf) with less storage manage quite well. In fact if a country can meet it use without storage so much the better.

Since rainfall in India occurs over a few months it is, of course, imperative to store water. Let’s examine this more closely.

Note that the Indian storage estimates – approximately 300 billion cubic metres (BCM) (265 m^3 per capita) - relate to major (more than 10 million m^3) and medium facilities with the former accounting for almost all of it. This excludes the tens of thousands of small storage structures in the country that collectively store considerable quantities of water.

A major source of surface storage is water in the form of snow, something not considered by the authors at all. This is a free and valuable form of storage and the melting of snow provides water in a very regular and dependable manner, though admittedly the water is not accessible the way it is from other sources. Yet it is a form of storage that hardly deserves to be excluded. In another essay in the book (pg. 184) a figure of 200 BCM of annual water flows from snowmelt is mentioned and one can derive a figure of 800 BCM of storage in the form of snow (700 m^3 per capita), a figure hardly to be scoffed at.

Nor is water stored only above the ground. It is stored below the ground (in shallow and deep aquifers) and which provides near-free and low evaporation capability. As is well known by now groundwater is a considerable source of supplies to irrigation, domestic and industrial users. Aggregate storage below the ground would dwarf surface facilities by a large margin.

(Now it is true that other countries storage capacities too will increase if we include all sources but as I have pointed out overall storage per capita may not be a useful way of looking at things.)

There is also the whole issue of the source of water for storage. If water is not being stored today where is it going? Is it flowing to the seas? Recharging groundwater? Or is it evaporating? Surface storage in large dams would make sense only in case of the last of the three. For as the authors themselves recognise water flowing to the seas is not a waste. And if it were to be captured from what goes to recharge groundwater it would be a zero-sum game.

Finally, it is in the Ganga-Brahmaputra-Meghna system that the bulk of the storable potential exists but water shortages are endemic in the arid parts – select regions of north, western, central and southern India. Enhanced water storage is unlikely to benefit these areas.

(This post may be read along with the others on Water.)

Water Woes in Chennai: Quick Comments on RWH and Desalination

In a recent post I had discussed the potential for desalination as a solution to shortages of water and had made a reference to Chennai. The intention was not to discuss the water problems of the city at any length. The city’s water woes are well known but not well studied.

One attempt to redress the lacuna is the work of A. Vaidyanathan (done jointly with J. Saravanan). Readers may read the relevant chapter in this book. Based on a survey of households carried a few a years ago it provides information on several aspects such as consumption, sources of supply and costs. Care must be taken in interpreting and drawing conclusions from the study as it had several limitations, which the authors themselves highlight. It does provide a broad overview of the water issues including discussion on conservation, rainwater harvesting (RWH) and other supply augmenting measures.

While the work is useful it lacks conceptual clarity. If suffers from most of the errors I had pointed out in an earlier post. We don’t get a water balance for the city that would take into account the way water is received, stored, used and disposed.

For example, while discussing RWH the authors don’t tell us where does the water go if it is not harvested. Does it flow out to the seas or to tanks or lakes that dot the city or in to the marshes or other natural bodies? Does RWH increase overall water availability or does it just redistribute it? Does local availability increase? Is it a zero-sum game?

(This is an important issue especially in urban areas. In rural areas local harvesting has much stronger rationale though the issues are relevant there too. See, 1 and the responses to it - 2 and 3.)

There is also the issue of RWH on individual structures. Even if water is to be harvested locally must it be done on each and every building rather than in a collective enterprise? Making RWH compulsory as was done in Chennai is also likely to lead to corruption, or people putting up token structures that are not effective to begin with or then failing to maintain them.

Desalination is dismissed in one paragraph, which is surprising as the book has been written in 2006 by when a lot of initiatives in Chennai had been taken up. As mentioned in an earlier post this augmentation measure may have great potential and impact.

Chennai is interesting not only because it suffers acute and chronic water problems; relies on water from surrounding and far-off areas with attendant problems; but also because it is on the coast where desalination can be an attractive proposition.

Desalination in India: Some Comments

It is not for nothing that Earth is called a blue planet. Not only is 71 % of the surface covered by seas, the water they hold is so unfathomably large (in relation to what circulates in the hydrological cycle) so as to be considered infinite. See here.

Removing salt from water is rather easy - boil it and then condense the vapours in another vessel, - and in the bargain get salt, as well. The issue has always been of the cost – primarily energy.

Developments in the past decades have dramatically reduced these costs – present estimates range from 50 to 80 cents (Rs. 20-32) per cubic metre (1,000 litres) of water. (See 1, 2 and 3), though in India a figure of Rs. 50 is also quoted. Costs are most sensitive to the level of salt in water (the lower the salt content the cheaper it is to desalt - so treating brackish water is cheaper) and energy costs. Note that the above costs are that of desalinisation and don’t include those of distribution. Transporting water over long distances (which may also entail lifting it) can increase considerably the final delivered cost of water.

There are two major technologies for desalting water – reverse osmosis (RO) and multi-stage flash (MSF); the former is increasingly more popular. RO is modular in nature and capacity ranges are wide. Another point to the noted is that while the cost of desalination is going down that of conventional water is going up as fresh supplies come from deeper aquifers or water is transported over longer distances. (See above citations for details including on costs, technologies and other matters. See also the references in the sources cited above.)

Given that so much water is available, costs are falling, traditional sources are turning dearer and more difficult to tap, and that there are severe shortages in several coastal location (Chennai is the example that springs to mind) is there a case for desalting water on a large scale?

Apart from Chennai, several other cities on the lower east coast and in Gujarat would seem to hold immediate potential. At present Chennai is building several desalination plants for both domestic use and for industries. See 1, 2 and 3.

Another interesting possibility is that if coastal areas can develop their own independent water supplies it may relieve pressures on upstream water resources that currently supply water to these locations. In other words, water which is now required for downstream users can be saved and used upstream. Hence, desalted water may have a role in helping upstream water users too!!

Desalted water is undoubtedly more expensive, say 2-5 times (numbers are illustrative) the cost of conventional water and it is feared that overall costs of water would shoot up considerably if it were to be adopted on a wide scale. The purpose of this post is not to discuss the costs in any detail but I would suggest a perusal of the links cited above. Instead I wish to make some more general observations and clarify certain matters in this regard.

To begin with it is not as if all water supplies would be met from desalted water. Only incremental supplies will be. So if say, 10 % of water is to be met from desalination and it costs 5 times as much as conventional water overall total costs go up only 50 % and don’t become 5 times. (A scenario analysis using various assumptions on the cost of desalting water and its contribution to overall water supplies is encouraged)

However, comparing the marginal cost of desalted water with the average of conventional is not the right way to go about it. Marginal costs ought to be compared with the marginal cost of conventional water supplied. Since the latter is likely to be closer to Rs. 50 and not Rs.10 or so the difference between the two sources narrows down considerably. (The average cost of the RO is nearly the same as its marginal cost).

In short, desalination may be cheaper, relatively, than what appears at first glance.

It is also argued that the energy costs of RO are considerable but that is already included in the higher costs of desalting water and highlighting them separately is wrong and if done unthinkingly may end up in double counting.

Note, that as a practical matter residents of Chennai and industries around the city already pay Rs. 50 or more per cubic metre, the very high end of the cost of desalted water. And if one adds the opportunity cost of time, disease, additional investments in pumps and storage, desalination is not more expensive, probably a cheaper alternative. The extra burden, assuming all costs are to be recovered from users would hardly burn a hole in the pockets of the residents. Compare the monthly expenditure on water with items such as telephony or entertainment. (Tamil Nadu seems rich enough for its government to give free colour TV sets to the needy!!)

The other major concerns with desalination are its environmental impact.

One of them is the loss of marine life during the intake as organisms get sucked in and die. This is rather a minor problem to solve and is preventable by the suitable placement of intake pipes, meshes and beach-wells.

The major worry has been the effluents generated during purification. Note, however, that the common notion that hot water generated during the process can damage the marine ecosystem is not true. Hot water is not generated during RO but brine (highly concentrated salt solution) definitely is.

It has been argued that brine can be discharged at appropriate places and diluted with water to lessen its impact. Another suggestion is to solidify the wastes and dispose them in say, abandoned mines or such places.

But the best possible solution would be to sell it. After all salt is a major input for many chemical industries and maybe it can even be made good enough (after treatment) for human consumption. I am reminded about flyash (generated from thermal power plants) and which was such a problem many years ago. Now cement companies clamour to gain access to it to make blended cement. They are willing to pay for it.

It is also curious that papers such as that of the WWF cited above make no discussion of environmental costs of existing water supplies. After all costs are relative. Groundwater depletion, energy use by borewells, tankers plying all over the city, water transported over long distances, are all environmental costs associated with conventional supplies.

Overall it seems that desalination of water has a promising role to play. It ought to begin small but if economic and environmental costs are reasonable it can be expanded to more locations, and water conveyed inland.

Finally, desalination is not a substitute for demand-side measures. It is sometimes argued that we should rely on the latter rather than the former to solve our problems. Of curse, we should. But where supplies need to be augmented, desalting would be as good a bet as withdrawing water from the ground or bringing it from distant places. Nor is desalination likely to be relevant for the whole of India. It is also not a panacea for the myriad ills of India’s water system but it could play a considerable role in supplying clean water to select locations at low rates with minimal damage to the ecosystem.

Elementary Errors in Analysing Water

Water analysis usually starts with estimates of water availability. According to India’s Central Water Commission:

"Precipitation (including snowfall) is the source of all water on the earth. The average annual precipitation over the country is estimated at 4000 BCM of which a part goes towards increasing ground water storage, a part is lost as evapo-transpiration and the remaining appears as surface water. The water resources potential of the country which occurs as natural run off in the rivers is estimated as about 1869 BCM, considering both surface and ground water as one system. Due to various constraints of topography, uneven distribution of resource over space and time, and geographic [sic] only about 1122 BCM of the total potential can be put to beneficial use, 690 BCM through surface water resources and 432 BCM by ground water."(pg. 13)

These numbers are widely used (see CSE and Iyer (2007)) and rather uncritically.

Note that just over 25 % of the precipitation is estimated of being put to beneficial use. So even a small increase in the utilisation percentage can lead to a big jump in available water.

The paragraph quoted above, however, is factually inaccurate, misleading and incomplete. To begin with the statement about precipitation being the source of all water is erroneous as oceans (97%), glaciers and polar icecaps (2.4%) hold the bulk of surface water. See here. Shallow and deep aquifers (in the aggregate) hold enormous quantities of freshwater.

The CWC statement can perhaps be re-read to indicate an estimate of sustainable water availability as water from rains is replenished every year. However, even so their estimate is incorrect, as we will see below.

Firstly, rainwater which seeps in to the ground is also (potentially) available for use so it should not be deducted from total precipitation. Secondly, India has commitments to supply (let water flow) to neighbouring countries and in turn it receives water from outside its boundaries. The net figure has to be deducted from overall precipitation. Finally, flow of water in rivers and out in to the seas serves many critical ecological and socio-economic functions, so even if all water could be captured and stored one wouldn’t do so.

So beginning with the annual precipitation over the country, a proper analysis must deduct the quantity of evapo-transpiration (strictly speaking this is the only quantity not available for use) and India’s net commitments to neighbouring countries. Water, which needs to flow to the seas to fulfil ecological and other functions, too needs to be subtracted.

Potentially all other water is available for use. However, it is not quite practicable to store all the water that falls as precipitation and much of it flows to the seas. The storable potential is not fixed and has and can be increased. Note, however, that this increase in storage capacity doesn’t necessarily have to come from the construction of large dams. Small storage structures and increasing ground water storage through increase in percolation of rainwater through the soil, to name just two measures, can be just as effective.

Even this analysis is incomplete. For water can be and is used again and again. This is true of the three major water-using sectors – domestic, agricultural and industrial. Return flows, as they are termed are extremely important and the bulk of water used is returned back to the hydrological cycle. Most of this happens naturally but can be enhanced by human efforts. A multiplier is at work here and recycled and reused water may increase manifold the effective utilisable water.

For an extremely illuminating discussion on the above see IWMI especially the section on Water Balance Analysis and Appendix A.

Finally, water supplies can be augmented by desalination of seawater. This is, of course, limited to coastal location and largely for industrial and domestic use but with a coastline of 7,500 kms this need not be a trivial source of supply. Such water is now available at very competitive rates without severe environmental damage (this is not the place to go into details, but I will discuss this in a later post). And rather than take out all the water from our rivers it may be far more sensible to let water flow into the seas and then desalt it.

The CWC analysis (which forms the basis for many others) seriously underestimates water availability in the country. I don’t have the model or the data to estimate the revised numbers but they must surely be much more than present CWC estimates.

{Note: BCM is billion cubic metres. I cubic metre = 1,000 litres. India’s average annual rainfall is 1200 mm (1.2 mts) and multiplied by the area of 328 million hectares gives an approximate figure of 4,000 BCM since a hectare = 10,000 sq. mts. Also 4,000 BCM=400 million hectare metres=4,000 cubic kms.}

Essays, essays, everywhere, not one to read

Waterscapes
The Cultural Politics of a Natural Resource
Edited by Amita Baviskar

The book comes in the midst of the flood season in India. The floods have, as usual, taken their toll of human and animal life and property. So I quickly flipped the pages to see what this book may have to say on the important subject and found a solitary article.

Rohan D’Souza’s piece is about flood control in deltaic Orissa alongside comparisons with similar schemes in the USA and China. The author points out the conflicting views on controlling floods by constructing embankments and dams and how the views swung from seeing flood waters as a calamity to seeing it as a nutrient-rich resource. The history is not quite up-do date as it stops with the 1950s. What lessons does it have for contemporary matters relating to flood control?

Precious little, one would think. Whatever may have the situation earlier matters have changed since the 1850s, from when the author traces the history. Two major changes can be pointed out. One, that nutrients to enrich soils are now increasingly supplied by artificial means and with greater effectiveness and control.

The second and the major change, however, is in the impact of floods. Urban areas and populations (both absolute and in relation to rural areas) have increased dramatically. Now urban areas don’t derive any benefits from floods but bear huge costs. Even in rural areas population densities have increased. What earlier may have been a relatively simple expedient of moving away when waters flooded homes is these days nothing short of a nightmare (even if it were simple to shift in the past it must have been a terrible experience though the author doesn’t discuss this aspect).

In contemporary India we are quite familiar with what happens when floods strike and images of people marooned and forced to spend days and nights on tree-tops under rain with scarcely any food or amenities are grim reminders of the heavy costs that people face. Even the army needs to be called in at many places to rescue people. Unfortunately, waters don’t confine themselves to farm land but enter houses and offices destroying life and damaging possessions. Not only is life at risk but even in rural areas large scale inundation is extremely disruptive of economic, social and educational activities. These days there is more infrastructure (roads, banks, schools), more material possessions (household durable goods, for example) and more equipment (tractors, pumps, engines) whose submergence is hardly an inconsequential matter.

In other words India can’t afford to have large tracts of land inundated with water every year. The author quotes a source that 12-20% of the land area of the country may be flood-prone. If true, it would imply a population at risk in such areas of 100-200 million.

So the notion of floods as a resource is now merely an academic curiosity. This, however, doesn’t and shouldn’t be construed as a plea or a case for building dams or embankments. There are other important aspects to inundation and hydrology that the author doesn’t discuss - the most prominent omission is drainage – and which are critical to understanding, regulating and preventing floods and the terrible consequences they bring.

Most of the papers in the book are not about surfeit but scarcity of water. On groundwater there are two articles – both about Gujarat and oddly enough David Hardiman’s paper relies a lot on the work of the other contributor - Navroz Dubash. Not only does it make the essays repetitive but one also misses perspectives from other states. For example, even communist-ruled West Bengal has fairly extensive groundwater markets.

Both Hardiman and Dubash are critical of and reject work claiming competitive water markets. They don’t, however, demonstrate that prices deviate from competitive outcomes but merely assert that this is so. This is clearly not sufficient.

On the other hand they do claim roles for institutional factors and historical development in water markets. Dubash makes the case that kinship and caste-based understandings play an important role in determining outcomes. Hardiman also claims that in one village (of the 2 studied by Dubash) a price cartel has been at work.

In that particular village the dominant community makes up 64% of the households and owns 97% of the land and 100% of the wells. Hence the thesis that kinship and caste play an important role is a trivial conclusion and not particularly insightful. After all given the ownership pattern is it a surprise that all decisions and outcomes will be internalised within the particular community? And if as Hardiman says they have formed a price cartel it is not clear who the cartel is exploiting or targeting. After all, the cartel is itself the market!!!!

The second village is more diverse and market forces play a greater role and caste and kinship forces seem to be weaker. The thesis thus propounded by these two authors suffers from lack of evidence and seems quite tenuous.

Overall, in the book, several important areas such as pollution, urban water scarcities, waterlogging and salinity of irrigated land are missing and while no book can be expected to be comprehensive this one has a lot of articles of questionable relevance if not questionable scholarship.

Take the editor’s own article. It reads like a diary – a travel diary – of her journey through Jhabua evaluating its watershed programme. With little else apart from her own impressions and of those she met how seriously can one consider it? And one searches in vain for something on water, and then finally the last paragraph of the paper begins “Finally, water………..”

Other essays, for example in the section entitled “Imagining Communities” are not that neglectful of water. However, and in spite of what the editor says in the introduction there is no ecological specificity about these essays. They are not water-distinctive at all, they could have been about any resource, indeed about anything. For example substitute school for the tank in Arun De Souza’s piece and it would still read right. Or if in R Brara’s it were electricity instead of water it would still ‘enlighten”. So whatever be the other merits of these essays they don’t deepen our understanding of water issues.

The redeeming feature of the book is the essay on south Indian tank irrigation by David Mosse. It questions the idea of a pre-colonial ‘ideal’ water and irrigation regime and associated conceptions and roles of the state and the community. But it doesn’t stop there. Instead it also draws implications relevant for our times and can be read as a critique of attempt to recover or recreate the past. One does miss, however, what could have been a dialogue with another contributor with different views on the subject.

The book’s jacket has a colourful water painting titled “Small Pond, Many Fish”. I wonder if the artist is alluding to the billions of humans on planet Earth. Interesting all the fish are of the same size - no small fish, no big fish - and no fish is eating other fish!!! The rest of the landscape is verdant with trees and birds and women around a handpump giving plentiful water. Has the artist not read the book or is she making a point of her own?