In recent years, a myopic view favouring only private well irrigation in preference to canal irrigation is emerging among a few irrigation “experts” in India. We all know that major irrigation projects contributed to expanding irrigated area in the country over the years. A few scholars have recently documented the larger socio-economic (Bhalla and Mukherjee, 2001) and welfare impacts (Perry, 2001; Shah and Kumar, 2008) of large surface irrigation projects. Private well irrigation system in the last three decades witnessed rapid growth surpassing flow irrigation in its contribution to the net irrigated area. This was because of massive rural electrification, heavy electricity subsidies and institutional financing for pump sets. However, the distorted thinking of considering one system superior to the other came because of a poor understanding of determinants of irrigation growth and the fundamental difference between well and surface irrigation.
Canal vs Well Irrigation: Comparing Apples and Oranges
Surface irrigation systems provide more dependable sources of water than groundwater-based systems in most parts. For flow irrigation, there should be a dependable source of water, and a topography permitting flow by gravity to the places of demand. Ideally, the design itself ensures sufficient yield from the catchment to supply water to the command areas for an estimated duty of the design command, or in other words, the design command is adjusted to match the flows available from the catchment. Hence, the design life of the scheme is by far realistic for reliable “dependable yield” estimates, unless major changes occur in the catchment that changes the flow regimes and silt load.
But, in case of groundwater, thousands of farmers dig wells drawing water from the same aquifer. Since they all operate individually, “safe yield” of the aquifer is not reckoned while designing the well. So, the productive life of a well is not in the hands of an individual farmer who owns it, but depends on the characteristics of aquifer, wells and total abstraction. In the entire hard rock peninsular and central India and some parts of western India, as the static groundwater resource is negligible, the wells go dry or yield reduces drastically when aquifer is “over-exploited”, or when monsoon fails. In spite of explosion in well numbers, the well irrigated area has not increased here during the past decade. Experience shows that the bore wells last for only 1-3 years.
Secondly, growth rate in well irrigation is almost decelerated in most parts of India since nineties. Most of well irrigation in India is in the arid and semi arid regions of northern, north-western, western and peninsular India (Table 1). Amongst this, intensive well irrigation in terms of per capita groundwater withdrawal per annum is highest in some of the northern and north western States, viz., Punjab (1729.9 m3/capita/annum), Rajasthan, UP and Haryana and to an extent Gujarat, Tamil Nadu and Andhra Pradesh (Figure 1, Source: Kumar et al., 2008b).
Intensive irrigation could sustain for many decades only in a few pockets such as alluvial Punjab and Haryana and UP. These regions are already saturated in terms of irrigated area. Further, expansion in irrigated area is not possible in these areas. Whereas in Rajasthan, Gujarat, Andhra Pradesh and Tamil Nadu, problems of over-exploitation halt further growth in well irrigation. Most of the untapped groundwater is in eastern Gangetic plains, devoid of sufficient arable land that lies un-watered (Kumar and Singh, 2005; Shah and Kumar, 2008; Kumar et al., 2008b). Peninsular India and central India have a lot of un-irrigated land. Agriculture is prosperous in this part of the country, and demand for water is only going to grow. But, well irrigation is experiencing a “leveling off” and sometimes decline due to “over-exploitation” and monsoon failure.
A recent analysis in the hard rock areas of Narmada river basin in Madhya Pradesh showed that the average area irrigated by a single well has declined over a 25-year period (Table 2 based on Kumar, 2007). Such a phenomenon is occurring due to well-interference, a characteristic feature of hard rock areas, which starts occurring when all the groundwater, that can be tapped, is already tapped. In such situations, an increase in number of wells does not result in increase in total irrigated area (Kumar, 2007). Hence, it is wrong to assume that well irrigation could sustain the same pace of growth in coming years.
This spatial mismatch in resource availability and demand can be effectively addressed only by large surface water projects, and not by groundwater projects1. Water was taken from rich upper catchments of river basins, which formed ideal locations for storages. Surface irrigation can expand in future also with investments in large reservoirs and transfer systems that can take water from the abundant regions of the north and east to the parched, but fertile lands in the south, though their economic viability and social costs and benefits will have to be ascertained. But, the same is not true for wells, as engineering feasibility of transferring groundwater in bulk itself is a questionable proposition.
Outdated Irrigation Management Concepts
Now and then arguments are heard that the government investment in surface irrigation systems should be diverted for better management of our aquifers. They stem from the presumption that surface irrigation systems perform badly. Such arguments are based on obsolete irrigation management concepts, which treated the water diverted from reservoirs in excess of crop water requirement as “waste” (Seckler, 1996). As Seckler (1996) notes, the fundamental problem with this concept of water use efficiency based on supply is that it considers inefficient both evaporative loss of water and drainage. It is not well informed by the water use hydrology of surface irrigation systems. Most of the seepage and deep percolation from flow irrigation systems replenish groundwater, and is available for reuse by well owners in the command (Allen et al., 1998; Seckler, 1996). This recycling process not only makes many millions of wells productive, but also saves the scarce energy required to pump groundwater by lowering pumping depths. This is one reason why well irrigation can sustain in many parts of Punjab and Haryana, Mulla Command in Maharashtra, in the Krishna river delta in AP and Mahi and Ukai-Kakrapar command in south Gujarat.
B. D. Dhawan, one of the renowned irrigation economists, looked at the economic returns from surface irrigation systems in his book wherein he examined the merits of he claims and counter-claims about the benefits of big dams. He had highlighted the social benefits generated by large irrigation schemes through the positive externalities such as improvement in well yields, and reducing incidence of well failures and increasing the overall sustainability of well irrigation by citing the example of Mulla command in Maharashtra (see Dhawan, 1990).
The social benefits these canals generate by protecting groundwater ecosystems are immense (Shah and Kumar, 2008), reduced energy cost for pumping groundwater being one (Vyas, 2001). But, irrigation planners have, by far, nearly failed to capture them in the cost-benefit calculations (Shah and Kumar, 2008). The recent data from government of Andhra Pradesh shows that the command irrigated regions of the state have the lowest number of groundwater “over-exploited” mandals. The tail end regions of the canals have sufficient number of bore wells, which actually reap the benefit of return flows from canals, and thus have good yields. The large reservoirs had raised cereal production to the tune of 42 million ton in fifty year since Independence. The social benefit this had generated by lowering cereal prices in the country was estimated to be Rs. 4300 crore annually (Shah and Kumar, 2008). Added to these are the multiple use benefits that canal water generate such as fish production, water for domestic use and cattle in rural areas.
Groundwater Recharge using Local Runoff: Catching the Crane Using Butter?
It is often suggested that flows from the small canals (Shah, 2008) or small water harvesting/ artificial recharge structure (GOI, 2007; Shah, 2008 as quoted in TOI, 2008) should be used for recharging aquifers. This is fallacious as the arid and semi arid regions, where aquifers are depleting (GOI, 2005; Kumar, 2007), have extremely limited surface water (Kumar et al., 2008a). From the map which shows the over-exploited regions in India (source: GOI, 2005), one can make out that they also coincide with the regions/basins of poor surface water availability. Examples are Western and Central Rajasthan; almost the entire Punjab; north Gujarat; parts of Andhra Pradesh, Madhya Pradesh, Maharashtra and Tamil Nadu. The surface water resources in these basins are extremely limited and are already tapped using large and medium reservoirs (source: GOI, 1999).
Any new interventions to impound water would reduce the d/s flows, creating a situation of “Peter taking Paul’s water”. Such indiscriminate water harvesting is also leading to conflicts between upstream and downstream communities as reported by Ray and Bijarnia (2006) for Alwar in Rajasthan; Kumar et al., 2008a for Saurashtra in Gujarat. Kumar et al. (2008a) shows that in semi arid and arid regions water harvesting/recharge not only has poor physical feasibility and economic viability, but has negative impacts on access equity in water (Kumar et al., 2008a). Kumar et al. (2008a) argued that the central government’s Rs. 1800 crore-scheme to recharge groundwater through open wells in hard rock areas of India, if implemented, would render many small and large reservoirs unproductive, unless bring water from surplus basins in the north and east is brought for recharging the aquifers in peninsular and western India.
Bringing water from water-surplus basins to peninsular India would require large head works, huge lifts, long canals, intermediate storage systems, and intricate distribution networks. As we have argued, recharge schemes using local water are economically unviable. Need for vast precious land for spreading water for recharge, would make it also socially unviable, while further increasing economic costs. Since the aquifers in hard rock areas of India have extremely poor storage capacities, efficient recharge would require synchronized operation of recharge systems and irrigation wells. This would call for advanced hydraulic designs, and sophisticated system operation. Therefore such an approach of using imported surface water for recharge would sound like “catching the crane using butter”. The fact that practicing environmentally sound artificial groundwater recharge is a very expensive affair in these water-scarce hard rock areas is yet to be appreciated by a section of the water community.
Hence, the best option would be for the farmers to use this expensive canal water for applying to the crops in that season of import (mainly monsoon season), and use the recharge from natural return flows for growing crops in next season. Opportunities for using water from “surplus basins” for recharging depleted aquifers exist at least in some areas. Examples are alluvial north Gujarat and north-central Rajasthan. Ranade and Kumar (2004) has shown that use of surplus water from Sardar Sarovar project during years of high rainfall for recharging the alluvial aquifers of north Gujarat through the designated command area in that region (Ranade and Kumar, 2004). They proposed the use of existing Narmada Main Canal, and the rivers and ponds of north Gujarat for this. It can protect groundwater ecology by reducing pumping; reduce the revenue losses in the form of electricity subsidy; and increase the flows in rivers that face environmental water scarcity, apart from giving direct income returns from irrigation. But, one cannot agree more on the point raised by Rath (2006) that only crops having very high water use efficiency will have to be promoted in the commands receiving such waters so as to generate sufficient returns from irrigated production. But, this will be possible only if the price of irrigation water is pitched at a level it starts reflecting the scarcity value of the resource.
Misuse of Statistics
Making choices between surface scheme and groundwater scheme based on crude numbers of “irrigated area by source” is “hydrologically and economically absurd”. Which model of irrigation is best suited for the area in future can be judged by the nature of topography, hydrology and aquifer conditions. For instance, in rocky central and peninsular India, only imported surface water can sustain and expand well-irrigation. Indiscriminately embarking on well irrigation would only ruin the rural economy. Farmers in these regions desperately drill bore holes to tap water with high rates of failure (Kumar and Singh, 2008), and resultant farmer suicides.
At least some scholars have begun to use “declining area under canal irrigation” to build a case for stopping investments in surface irrigation (Shah, 2009; TOI, July 10 & 17, 2008). But, this is a clear case of misuse of statistics. The reasons can be understood if we look at the real factors that influence the irrigation performance of surface systems.
First: As a recent study in Narmada river basin in central India shows, increased pumping of groundwater can significantly reduce stream flows in basins where groundwater outflows contribute to surface flows (Kumar et al., 2006), thereby affecting the inflows into reservoirs. Also, as is evident from the earlier discussions and from some studies, small water harvesting systems are adding to the reduction in inflows into reservoirs (Kumar et al., 2008; Ray and Bijarnia, 2006).
Second: Farmers in most surface irrigation commands install diesel pumps to lift water from the canals, and irrigate the fields. Such instances are increasing with pump explosion in rural India. The better control over water delivery, which farmers can secure by doing this, is the reason for their preference for energy-intensive lifting to gravity flow. Another important reason is the illegal water diversion which is rampant in canal irrigation. The pumping devices enable illegal diversion of water for irrigating plots that are otherwise out of command due to topographical constraints. Such areas get counted as pump irrigated areas in government statistics.
Third: Large reservoirs, primarily built for irrigation in this country, are being increasingly used for supplying water to big cities and small towns as recent studies show. A recent analysis involving 301 cities/towns in India shows that with increase in city population, the dependence on surface water resources for water supply increases, with the dependence becoming as high as 91% for larger cities (Figure 2, Source: Mukherjee and Shah, 2008). Many large cities depend entirely on surface water imported from large reservoirs, built primarily for irrigation. Some examples are Bangalore, Hyderabad, Ahmedabad, Chennai, Rajkot and Jodhpur.
Fourth: Farmers in canal command areas, especially at the head reaches, tend to put more area under water intensive crops, ignoring the cropping pattern considered in the design. This is one of the reasons for shrinkage in the irrigated command area.
Last, but not the least, reservoirs are experiencing problems of sedimentation causing reduction in their storage capacity and life, though as found world-wide in some cases the rates are higher those used at the time of design (Morris and Fan, 1998). The average annual loss of live storage for 23 large reservoirs in India with a total original live storage of 23,497 MCM (23.497BCM) studied by the Central Water Commission was to the tune of 213 MCM, i.e., an annual reduction of 0.91 per cent. Hence, generally for older reservoirs, the loss of storage would be quite significant. Such annual losses can sometimes reduce the effect of additions in storage achieved through new reservoir schemes on expanding irrigation. Therefore, it is likely that with the passage of time, the area under surface irrigation would decline, if nothing is done to revive the reservoirs. It is also therefore quite obvious that with cumulative investments in surface irrigation systems going up with time, there may not be proportional rise in surface irrigated area.
Therefore, at least some of these above facts are compelling reasons for fresh thinking on the planning and implementation of irrigation in India. Clearly, the solution does not lie in completely writing off surface systems for wells as the latter ones are not replacement for the earlier.
Can Wells become the “Poverty Alleviating Machines”?
Over the past few decades, well irrigation has been romanticized by some scholars has a poverty alleviating machine (IWMI, 2007; Mukherjee, 2002). While it is understood and also well documented by many scholars in the past that irrigation has a significant impact on poverty alleviation in rural areas (Bhattarai and Narayanamoorthy, 2003; Hussain and Hanjira, 2003), the over-emphasis on groundwater is somewhat difficult to assimilate. More strikingly, major arguments about the poverty impact of groundwater irrigation are made in the context of eastern India. As Mukherjee argues, “in regions of abundant rainfall and good alluvial aquifers, ground water irrigation can be a powerful catalyst in reducing poverty (IWMI, 2007). Eastern India’s potential for triggering country-wide agricultural growth through a boost in well irrigation is also strongly argued (Shah, 2001; Mukherjee, 2002). Here, one really wonders about the actual effect of rainfall on irrigation demand. Also, one wonders about the effect of irrigation versus land on economic surplus in areas of high water availability. It is a truism that marginal returns from irrigation would be higher in areas of high aridity and low moisture availability, and not in humid/sub-humid areas with high moisture availability. Eastern India falls in the latter.
What is surprising is that in the entire policy discourse on the impact of irrigation on agricultural development, the key factor of production, i.e., “land” does not find a place anywhere. In fact, we would like to vehemently argue here that it is simply fallacious that in eastern India, with plenty of groundwater, there could be a boom in well irrigation, with proper electrification and energy policies. The water demand for irrigation is very low in this region.
The maximum water needed for irrigation is a direct function of per capita arable land and reference evapo-transpiration; and inverse function of effective rainfall, provided the socio-economic conditions are favourable. In eastern India, not only that the rainfall is high, but the ET is comparatively lower than western, north western and southern India. The per capita arable land is lower than that of western, peninsular and north-western India (Kumar et al., 2008b). In Bihar, it is one of the lowest in the country with 0.068 ha against 0.17 ha in Punjab, and only 40% of the net sown area remained un-irrigated in 2000 (source: based on Agricultural Census, Ministry of Agriculture, GOI, 2000).
The groundwater use intensity is already quite high in Bihar and other eastern Indian states like Assam and west Bengal (see Figure 2). This is far higher than the groundwater use intensity in Rajasthan and Andhra Pradesh, which are facing severe problems of over-exploitation. Given that, already more than 60% of the net sown area in Bihar is irrigated. Even if we improve the affordability of irrigation water for millions of poor farmers in this region, what we can achieve is very minimal. Unfortunately, such pampered views dominate the water policy debate in India. The huge opportunity cost of delaying the most essential investments in irrigation, in regions where it matters, is by and large ignored.
Evidence available from both Indo-Gangetic plains and in peninsular India suggests that there is a strong nexus between surface irrigation development and sustainability of well irrigation. Therefore, it is not prudent to invest in well irrigation without investment in large surface reservoirs and conveyance systems in semi arid and arid areas. Also, it is high time for the “die-hards” of well irrigation to understand that water, whether well water or canal water, has to come from the same hydrological system. Promoting aquifer recharge using surface runoff from the same area, to sustain well irrigation is a hydrological and economic nonsense. East India, which has abundant groundwater resources, is not capable of driving growth in well irrigation in future. A greater recognition of the fact that availability of arable land, rather than the availability of groundwater, is a major determinant of regional growth in irrigation demand would change the paradigm of water resource development for irrigation.
1 This, however, does not to trivialize the role of demand management in regions where demand exceeds supplies.
References and Additional Thinking
Allen, R. G., L. S. Willardson, and H. Frederiksen (1998) Water Use Definitions and Their Use for Assessing the Impacts of Water Conservation. Proceedings ICID Workshop on Sustainable Irrigation in Areas of Water Scarcity and Drought (J. M. de Jager, L.P. Vermes, R. Rageb (eds). Oxford, England, September 11-12, pp 72-82.
Bhalla, S and A. Mookerjee (2001) Big dam development: facts, figures and pending issues, International Journal of Water Resources Development, 17 (1) Taylor & Francis Ltd.
Bhattarai, Madhusudan and A. Narayanamoorthy (2003) Impact of irrigation on rural poverty in India: an aggregate panel data analysis, Water Policy, 5 (2003): 443-458.
Dhawan, B. D (1990) Big Dams: Claims, Counterclaims, New Delhi: Commonwealth Publishers.
Government of India (2000) Agricultural Census-2000, Government of India, New Delhi.
Government of India (2007) 2007): ‘Report of the Expert Group on ‘Groundwater Management and Ownership’, Planning Commission, Yojana Bhawan, New Delhi.
Hussain, Intizar and Munir Hanjra (2003) “Does Irrigation Water Matter for Rural Poverty Alleviation? Evidence from South and South East Asia,” Water Policy, 5 (5): 429-442.
International Water Management Institute (2007) Water Figures: Turning Research into Development, IWMI Newsletter, Issue 3, 2007.
Kumar, M. Dinesh and O. P. Singh (2008) How Serious Are Groundwater Over-exploitation Problems in India? Fresh Investigations into an Old Issue, proceedings of the 7th Annual Partners’ Meet of IWMI-Tata Water Policy Research Program “Managing Water in the Face of Growing Scarcity, Inequity and Declining Returns: Exploring Fresh Approaches,” ICRISAT Campus, Patancheru, Hyderabad, 2-4 April, 2008.
Kumar, M. Dinesh and O. P. Singh (2005) Virtual Water in the Global Food and Water Policy making: Is there a Need for Rethinking? Water Resources Management, 19: 759-789.
Kumar, M. Dinesh, Shantanu Ghosh, O.P. Singh and R. Ravindranath (2006) Changing Surface water-Groundwater Interactions in Narmada River Basin, India: A Case for Trans-boundary Water Resources Management, paper presented at the III international symposium on trans-boundary water management, UCLM, Ceudad Real, Spain, June.
Kumar, M. Dinesh, Ankit R. Patel, R. Ravindranath and O. P. Singh (2008a) Chasing a Mirage: Water Harvesting and Artificial Recharge in Naturally Water-Scarce Regions, Economic and Political Weekly, 43 (35): 61-71.
Kumar, M. Dinesh, M V. K. Sivamohan and A. Narayanamoorthy (2008b) Irrigation Water Management for Food Security in India: The Forgotten Realities, paper presented at the International Seminar on Food Crisis and Environmental Degradation-Lessons for India, Greenpeace, New Delhi, October 24, 2008.
Morris, L. M. and J. Fan (1998) Reservoir Sedimentation Handbook: design and management of dams, reservoirs and watershed for sustainable use. McGraw Hill, New York.
Mukherji, Aditi (2003) Groundwater Development and Agrarian Change in Eastern India, IWMI-Tata Comment # 9), based on Vishwa Ballabh, Kameshwar Chaudhary, Sushil Pandey and Sudhakar Mishra, IWMI-Tata Water Policy Research program.
Mukherjee, Sacchidananda and Zankhana Shah (2008) Large Reservoirs: Are they the Last Oasis for Survival of Cities in India?, proceedings of the 7th Annual Partners’ Meet of IWMI-Tata Water Policy Research Program “Managing Water in the Face of Growing Scarcity, Inequity and Declining Returns: Exploring Fresh Approaches,” ICRISAT Campus, Patancheru, Hyderabad, 2-4 April, 2008.
Perry Chris J. (2001) World commission on dams: implications for food and irrigation, Irrigation and Drainage, 50:101–107.
Ray, Sunil and Mahesh Bijarnia (2006): ‘Upstream Vs Downstream: Groundwater Management and Rainwater Harvesting’, Economic & Political Weekly, July 10.
Seckler, David (1996) The New Era of Water Resources Management: from Dry to Wet Water Saving, Research Report 1, International Water Management Institute, Colombo, Sri Lanka.
Shah, Tushaar (2001) Wells and Welfare in Ganga Basin: Public Policy and Private Initiative in Eastern Uttar Pradesh, India, Research Report 54, International Water Management Institute, Colombo, Sri Lanka.
Shah, Tushaar (2009) Taming the Anarchy: Groundwater Governance in South Asia, Resources for Future and International Water Management Institute.
Shah, Zankhana and M. Dinesh Kumar (2008) In the Midst of the Large Dam Controversy: Objectives, Criteria for Assessing Large Water Storages in the Developing World, Water Resources Management, 22: 1799-1824.
Times of India (2008) Wasting $50 billion Dollar in Major Irrigation, Swaminomics, July 10 and 17.
Vyas J (2001) Water and energy for development in Gujarat with special focus on the Sardar Sarovar project, International Journal Water Resources Development, 17(1):37–54.
(M. Dinesh Kumar is a Ph. D in Water Management. He had worked with engineering consultancy organizations, national and international research/academic institutions and NGOs, and had worked very closely with many reputed international and national agencies, viz., UNICEF, the Ford Foundation, the International Development Research Centre (IDRC), the Aga Khan Foundation, New Delhi, Sir Ratan Tata Trust, Mumbai and Arghyam, Bangalore. He is currently the Executive Director of Institute for Resource Analysis and Policy (IRAP). He has nearly 120 publications to his credit, including three books; one edited volume; many book chapters; and several papers in international peer-reviewed journals.
Dr. Sivamohan was a Member of Senior faculty and Chairman Agriculture and Rural Development Area. He served international organizations like Irrigation Research Group at Cornell University, Ithaca USA, National Reserve Institute UK, ICRISAT Hyderabad and International Water Management Institute IWMI, Srilanka. Currently he is Principal Consultant and Member of Governing Board, Institute for Resource Analysis and Policy (IRAP). He has several national and international research publications and books to his credit.
Dr. A. Narayanamoorthy is currently working as NABARD Chair Professor and Professor and Director, Department of Rural Development, Alagappa University, Tamil Nadu. Dr. Narayanamoorthy has published three books, three mimeographs and over 80 research papers in international and national journals. He has completed several research projects sponsored by the Ministry of Agriculture, , Planning Commission, GoI and NABARD. He has been a consultant for the International Water Management Institute, Colombo. Dr. A. Narayanamoorthy also received the prestigious Professor Ramesh Chandra Agrawal Award of excellence for the outstanding contribution in the field of agricultural economics for the year 2009.
The views expressed in the write-up are personal and do not re?ect the official policy or position of the organization.)