Diagnostic analysis and some approaches
for improving water delivery performance in the Bhakra canal command

N.K. Tyagi

Abstract

The modernization of an irrigation system generally aims at removing or relaxing the constraints that undermine the performance of the system in respect of the designed objectives. It involves cost and has to be undertaken after establishing the nature and degree of improvement required and its technical and economic feasibility. Evaluation of the hydraulic performance of the irrigation system at the watercourse and farm levels was undertaken in the command area of a branch canal, the Fatehabad branch of the Bhakra canal system in Haryana. Equity (in terms of uniformity coefficient and modified inter-quartile ratio), adequacy (in terms of relative water supply) and water productivity, which reflected both adequacy and timeliness, were evaluated.

The equity of water distribution decreased with the size of the watercourse (flow rate), and the average value of the modified inter-quartile ratio was 1.85. With an average relative water supply value of 0.64 across watercourses and the seasons, the system had highly deficient supply. Water productivity value was only 0.51 across watercourses and seasons. Lower values of water productivity as compared to seasonal adequacy reflected a time mismatch between supply and demand. Equity in water distribution along the watercourse and irrigation efficiencies on the farm can be substantially improved through the proper design of the unit command area size using the procedure that has been developed for this purpose. Variable time warabandi and provision of tube wells in tail reaches would also minimize inequity. Intra-seasonal modifications in the existing water delivery schedule, based on the simulation model, indicated a substantial improvement in water productivity. Further improvements were possible through the provision of auxiliary storage at the head of watercourses.

Introduction

During the last few decades there has been very a rapid expansion of irrigation facilities all over the world. In India, which has chosen irrigation development as a vehicle for time-targeted progress, the rate of irrigation development has been the highest. At present, an area of some 51 million ha is irrigated by different sources. Through large and medium-sized projects in several states, including Tamil Nadu, Punjab, Haryana and Rajasthan, the irrigation potential has grown by 70 percent or more.

Provision of irrigation facilities has raised agricultural production, has improved productivity and has brought some sort of stability to Indian agriculture. In spite of these substantial gains, there is a growing perception that the performance of the irrigation system has been less than satisfactory. The shortcomings that are frequently mentioned include:

• smaller than expected increase in productivity;
• low irrigation intensities;
• higher than anticipated costs of construction and maintenance;
• inequity in water distribution; and
• environmental degradation in the form of waterlogging, salinity, erosion, spread of disease, etc.

The scope for further increase in irrigation potential at a reasonable cost is not very high. Also, allowing the present state of affairs to continue for long may have bearing on the long-term sustainability of irrigation in many regions. The conviction is spreading that a major breakthrough in irrigated agriculture is possible only through the modification of existing practices after diagnosing the causes through performance analysis (Government of India, Ministry of Water Resources, 1987). Meaningful performance appraisal is possible only if there is a clear understanding of how we define the irrigation system, its management and the objectives for which it has been created. The vastness of the subject also makes it necessary to define the boundaries of the proposed exercise along with anticipated outputs and future beneficiaries.

Physically, an irrigation system may include (1) capture and storage, (2) conveyance, (3) bulk distribution, (4) delivery, (5) application and (6) removal of water from agricultural land. These physical entities, which are distinct hydraulic levels fall under different management domains (Kellar et al, 1988). The management at hydraulic levels 1 and 2 is exclusively dealt with by irrigation project authorities such as irrigation departments or management boards, whereas the management at hydraulic levels 5 and 6 is in the domain of farmers or of the officials of the agriculture department. Hydraulic levels 3 and 4 fall in the intermediate zone where both irrigation department and command area development authorities interact.

Scope of the study

This study aims at determining the scope of improving the performance of an irrigation system at hydraulic levels 4 and 5 (watercourse and farm) through structural and operational changes in the system. The need for improvement will be established through a diagnostic analysis of the system performance. The type and nature of the interventions will be decided on the basis of an analysis of the strengths and weaknesses of the system.

Project area

The project area is a part of the Bhakra canal system in Haryana and it covers about 0.28 million ha between the latitudes 29º 00' to 30º 55' N and longitudes 73º 02' to 77º 28' E in the Ghaggar river basin, which is a part of the Indus basin.

The Fatehabad Branch canal, which branches off the Bhakra Main Branch canal, was selected for the study. Four pairs of watercourses (one lined and the other unlined) branching off the Gorakhpur and Khajuri distributaries in the head reach, Adampur in the middle reach and Kutiyana and Sheranwali in the tail reach, were selected for investigation. In addition, two watercourses on the Fatehabad distributary in the tail reach near Fatehabad were also chosen. The salient hydraulic data for these watercourses are given in Table 1. In this irrigation command, a three-stage water distribution system is in use. The first stage consists of main and branch canals and the second stage has the network of distributaries and minors. The distribution network in the first two stages is owned and operated by the state irrigation authorities. Watercourses which distribute water beyond the canal outlet constitute the third stage, and these are owned and managed by groups of farmers. The watercourse draws water through an adjusted proportionable module.

Table 1. Salient hydraulic data for different watercourses

The capacity factor (actual discharge/designed discharge) during the winter season is only 0.72 and is indicative of inadequate supply in general.

The problems

The problems of distribution and application in the Fatehabad branch canal are representative of similar problems on other projects in the region. Appraisal of the relevant documents, a walk through the system and discussions with the farmers and officials of the irrigation department indicated the following problems:

• Rigid irrigation water delivery schedules with almost the same frequency throughout the growing season cannot meet crop demands.
• The water supplied is scarce and the effects of scarcity are more severe in areas with highly saline/sodic groundwater.
• In unlined watercourses about one third of the area toward the tail does not get any water; this leads to the unauthorized practice of sale of canal water.
• The command areas of watercourses seem to have been fixed entirely on the basis of topography without much consideration for the soil infiltration rate.

Diagnostic analysis of irrigation performance

Chambers (1984) has listed some of the perceptions of good performance by different disciplines and sections of the society. An in-depth criterion of the objectives is given in Small and Svendsen (1990). Since we are limiting our scope of performance evaluation at the hydraulic level of watercourses and farms, success can be measured in terms of equity, adequacy and timeliness, and efficiency in application, distribution and storage on the farm.

On-farm irrigation system performance

In the present study only hydraulic performance of the field irrigation was evaluated. Graded borders are the most common way of applying water to crops. The test borders were specified along the entire length of the Fatehabad branch canal, in the command of selected watercourses. In general, three borders on each watercourse with location in head, middle and tail reaches were selected. The detailed procedure prescribed by Meriam (1978) was adopted. The application efficiency in most cases in quite high, whereas storage efficiency is low (Table 2). Because of higher stream size, the water spreads quickly and the irrigation is terminated before the required quantity has been diverted into the border. The distribution uniformity is also poor (less than 60 percent) in the majority of cases. Thus there is scope for improvement in the design of water application practices.

Table 2. Field irrigation efficiency in head, middle and tail reaches of watercourses

AE = application efficiency; SE = storage efficiency; DU = distribution uniformity; Ic = cumulative infiltration, depth and equity

Conveyance losses at different points were measured to compute the water being supplied to different farms and the equity, as represented by the Christiansen uniformity coefficient (C_u), and modified inter-quartile ratio (IQR) were computed (Table 3).

The values of C_u range from 0.63 to 0.95 for different watercourses with an average value of 0.8. So, if C_u were chosen as the criterion parameter for equity, the values of equity are apparently quite high. Of course, there is a decrease in equity as the size of the watercourse increeases. The distribution looks more non-uniform when one computes IQR, which represents the ratio of the average depth in the most favoured quarter to the average depth in the least favoured quarter. The IQR value at the lowest discharge of 19.5 l/s is 1.33 and it increases to 2.58 at the highest discharge of 68 l/s, with an average value of 1.85. In other words, the farms located in the head reaches of the watercourses receive nearly twice the water supply going to the tail-end farms.

Table 3. Values of equity measures in different watercourses

Relative water supply

The relative water supply is the ratio between the water supplied and the demand in an irrigation unit over a period of time. The concept is related to the available water supply, demand and management intensity in an irrigation system. The capacity of the system to control water is determined both by physical resources and the institutions. The degree to which capacity is actually realized is called intensity of management. By varying the intensity, it is possible to match supply and demand. As shown by Oad and Podmore (1989), a low relative water supply requires a more intense management.

The relative water supply varied across the watercourses, the seasons and the reaches. Watercourse 9400L (Khajuri) and 179415L (Adampur) (Table 4) had relatively higher adequacy, with values of about 80 percent, than watercourse 5000L (Gorakhpur) with 49 percent and 2000R (Sheronwali) with 58 percent. Watercourse 179415L was actually drawing more water than its designed discharge due to the inaccurate installation of an adjusted proportion module outlet. The sill level of the adjusted proportionate module was lower than provided for in the design. Relative water supply values during the summer, when crop water requirements are partly met by rainfall, were higher by 8-12 percent than in winter. There was marked variation in the relative water supply along the watercourse, with values at the head exceeding those at the tail by 25 percent. For example, the average value at the head of watercourses in winter was 0.65 as against 0.50 in the tail reach. Such large differences obviously call for immediate remedial measures.

Table 4. Relative water supply in different watercourses

L = left-hand side; R = right-hand side

Productivity of the water delivery system

Irrigation systems are meant to provide water to increase land productivity by maximizing water use. The excess or deficit of the water supply has an impact on crop yields, though the effect varies with each stage. There are several approaches to simulate the effect of a water supply regime (Bhirud et al, 1990; Vijayaratna, 1988); the most elegant, which has potential for large-scale application, is due to Lenton (1984). Lenton's water delivery performance takes into account both the adequacy and the timeliness of the water supply and essentially represents productivity on a 0-1 scale. The potential productivity of Abernathy (1987) is similar to Lenton's (1984) water delivery performance index and it produces a number in the range of 0 to 1 obtainable under a given water supply regime.

The computed values of productivity are given in Table 5. There are similarities as well as differences with the results obtained in the relative water supply analysis: the productivity of water delivery is higher in the head reaches of all the watercourses, the average value during summer being 0.61 in the head as against 0.48 in the tail reaches; but the relative productivity potential is higher in winter than during the summer. This may be due to the occurrence of rain during supply periods in the summer. In such cases, the supply becomes a surplus. The higher sensitivity to moisture of crops grown during the summer could also be the reason for the low water productivity.

Another important difference is that, in general, the values of productivity are lower than those of seasonal relative water supply. This implies a mismatch in time terms between supply and requirement. In other words, the water supply is wanting in timeliness. Timeliness and reliability both carry the implications of some external demand or need to be fulfilled. Small and Svendsen agree that in the absence of any readily identifiable distinction between reliability and timeliness, the single concept of timeliness would be more useful. In the above analysis, the need for improving the timeliness of water supply at different growth stages is clearly indicated.

Table 5. Water delivery performance in terms of relative productivity potential

Interventions for improving the performance

The problems of non-uniform and inadequate water application at the farm, inequitable distribution along the watercourse, and rigidity of irrigation schedules that lower water productivity can, to some extent, be overcome by making certain modifications in the system. The detailed description on how the interventions are to be designed is given elsewhere (Tyagi et al, 1995) and cannot be reproduced here for want of space. However, suggestions and modifications can be briefly mentioned.

• Improvement in water distribution equity and efficiency. Equity in water distribution along the watercourse and irrigation efficiency on the farm can, to a large extent, be resolved through the proper design of the unit command area size. A model has been developed for designing an optimal unit command area incorporating the concepts of equity, efficiency and productivity. It is validated with data from watercourses on the Sudkan distributary. The design of watercourses by adopting the suggested procedure led to improvement in three performance criterion parameters (equity, efficiency and productivity), thereby clearly establishing the need and scope for modifying the existing design standards.

Further scope for improving the water delivery performance lies in variable time warabandi from head to tail of the watercourse or through the installation of tube wells toward the tail reach of the course. The design procedures for these interventions need, however, to be established.

• Relaxing the rigidity of the delivery schedule. It is possible to make intra-seasonal modification in the existing rigid water delivery schedules without involving structural changes. A mathematical model has been formulated to evaluate the different rotational and on-demand water supply schedules with a view to suggest intra-seasonal changes. Application of the model to the Gorakhpur distributary of the Fatehabad branch has suggested the possibility of obtaining 94 percent of on-demand yield with intra-seasonal change in the irrigation interval after 100 days of sowing of wheat with only 66 percent of the on-demand water requirement. The suggested model provides a rationale and the procedure to modify water delivery schedules.
• Improving timeliness. The problem of time mismatch between supply and demand may be solved by providing auxiliary storage at the head of the watercourse and allowing for intra-seasonal variations in water delivery schedules. A procedure has been outlined to estimate the availability of canal water for auxiliary storage and to establish its economic feasibility. For the demand and supply conditions obtaining in the Adampur distributary of the Fatehabad branch canal, provision of auxiliary storage could improve wheat yields by 20 percent with a benefit-cost ratio of 1.3.

References

Bhirud, Sanjay, Tyagi, N.K. & Jaiswal, C.S. 1990. A rational approach for modifying rotational water delivery schedules, Irrigation and drainage engineering, ASCE, Vol. 116(5): 632-644

Chambers, R. 1984. Irrigation management: ends means and opportunities. In Productivity and equity in irrigation systems (ed. Niranjan Pant), Ashish Publishing House, New Delhi

Lenton, R.A. 1984. A note on monitoring productivity and equity in irrigation systems. In Productivity and equity in irrigation systems, op. cit.

Ministry of Information & Broadcasting. 1990. India 1990 - A reference manual, Research and Reference Division, Government of India, New Delhi, p 433

Oad, R. & Podmore, T.H. 1989. Irrigation management in rice-based agriculture: concept of relative water supply. ICID Bulletin, 38(1):1-12

Small, L.E. & Svendsen, M. 1990. A framework for assessing irrigation performance, irrigation and drainage systems, 4(4):283-312

Tyagi, N.K., Bhirud, S., Kaushal, R.K., Ambast, S. and Mishra, A.R. 1995. Improving canal water delivery performance: some approaches, Research Bulletin No 246, Central Board of Irrigation and Power, New Delhi, p 69

Vijayaratna, C.M. 1986. Assessing irrigation system performance: a methodological study with application to the Gal Oya scheme, Sri Lanka, PhD dissertation, Cornell University, Ithaca, USA