Some aspects of South Asia's groundwater irrigation economy: analyses from a survey in India, Pakistan, Nepal Terai and Bangladesh

292 villages were covered by the original sample; however, complete data sets are available only for 278 villages.

There were some parts of some zones–Balochistan and Kashmir, Jammu and Kashmir, and hill areas of Nepal–which were not covered by the survey for a variety of reasons. These were excluded from the analysis.

In hindsight this question may have been interpreted differently by different farmers and may have evoked different responses. The two most likely interpretations are: (1) When did you first acquire the well/tubewell in use now? (2) When did you first establish a mechanized well/tubewell. Under the circumstances, the best interpretation of Fig. 3 pertains to establishing the year during in which the respondent first acquired a mechanized well/tubewell.

Open or tubewells mounted with either diesel or electric pumps.

A variety of estimates are in circulation about the groundwater contribution to India's irrigated area. (NABARD 2005).

The National Sample Survey of 78,990 rural households around India, however, estimates the irrigated areas cultivated with the five major field crops (which account for 93% of the entire cultivated area) to be 66% (NSSO 1999a).

The Minor Irrigation Census (GOI 2001b), which covered all the Indian states except Gujarat, Maharashtra, Tamil Nadu and Karnataka, offers a rough reality check; however, this census reports data which is nearly 10-years old. According to this census, in 1993–1994, 38% of the net cultivated area was irrigated. The census computed that major and medium projects accounted for 39.4% of the irrigated areas while groundwater explained 50%; other surface sources accounted for 10.6%. The IWMI-Tata survey (in 2002) suggests that the structure of Indian irrigation is changing rapidly with canal dominated areas shrinking at a worrying pace. Recent large-scale surveys carried out in India by the National Sample Survey Organization lend indirect support to this contention that groundwater has become the mainstay of India's irrigation economy. The NSS 54th Round, which interviewed 78,990 rural households throughout India, estimated that while 61.2% of rural households are cultivators, 36% of all rural households interviewed (i.e., some 60% of cultivators) used irrigation (NSSO 1999a). NSSO (1999b) estimates that, of the irrigating households, some 40% depended on their own wells, another 27.2% relied on purchased pump irrigation, and 32.8% obtained irrigation from public/community surface and groundwater sources such as government canals, tanks, rivers, streams and public tubewells. This suggests that two-thirds of India's irrigation economy is dominated by private wells.

This does not tally with a much larger sample survey carried out by India's National Sample Survey Organization (NSSO 2003) in which 4,646 villages were asked about the kind of irrigation facilities they had. Of these villages, 76.2% reported some kind of irrigation, implying that 23.8% of the villages had no irrigation source whatsoever. Of the villages that reported irrigation facilities, 64.3% reported wells and tubewells as their source of irrigation, 17.3% reported the availability of canals, and 16.7% reported irrigation from streams, tanks and other sources. While this survey suggests the existence of a larger number of rain-fed villages than the smaller IWMI-Tata survey does, it supports the IWMI observation that wells and tubewells are the dominant source of irrigation in Indian villages. Moreover, if one were to expand this sample to cover Bangladesh, Nepal Terai and Pakistan Punjab and Sindh, it is quite likely that the proportion of villages with canal irrigation would fall and those with tubewells as the only source would increase.

That is, the quantity, quality and infrastructure of the electrical supply to agriculture have all progressively deteriorated resulting in farmers replacing electric pumps with diesel pumps.

The region also has much surface water. However, the lower Ganga Basin offers few sites for building large-scale storages. As a result, flood waters of Ganga and its tributaries cause much socio-economic and livelihood loss due to recurrent floods and acute surface-water logging. It has long been recognized that intensive groundwater development in the region may significantly ease flooding by curtailing the rejected recharge during and after a monsoon.

\( {\rm Discharge}\,({\rm m}^3 {\rm /ha)} = \frac{{{\rm HP} \times {\rm hours}/{\rm ha} \times 75 \times {\rm pump\, efficiency}\,(\%)}}{{1,000 \times {\rm head}\,({\rm m})}}.\) where 1,000 is the specific weight of the water. The average energy efficiency of Indian irrigation pump sets is widely recognized to be abysmally low. Patel and Pandey (1996)who rectified 3,600 irrigation pumps in Gujarat noted that “the levels of consumption of electricity or diesel in the farmers’ pump sets are 150–200% of the desired/achievable limits (p. 13)” and can be improved by simple rectification with minimal investments. Also see (Sant and Dixit 1996; Operations Research Group 1991, and TERI, 1996).

The Sardar Sarovar Project on the Narmada River in western India was one of the largest river-valley projects planned by India after it became independent. The project acquired worldwide notoriety because of opposition to it from environmental groups.

This survey, as mentioned earlier, covered area cultivated with the five major field crops, which together accounted for 93% of the entire cultivated area.

A new paper by Sharma and Sharma (2004) on the functioning of groundwater markets in the arid and semi-arid zones of Rajasthan, however, suggest a different pattern than what the IWMI aggregate picture suggests. In this water scarce region, groundwater markets should have a relatively low breadth and depth. However, this paper shows that 103 of the 137 households interviewed in the semi-arid Alwar and Jaipur districts participated as water buyers and/or sellers; and 115 out of 143 interviewed in arid Jodhpur and Nagaur districts participated as buyers and/or sellers. These statistics suggest participation rates comparable to Bangladesh’s. The Sharma and Sharma study also shows that in Rajasthan, where aquifers are deep and tubewells quite large, the modal seller is typically the large farmer, whereas over 90% of the buyers are small farmers with less than 2 ha of arable land.