 |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Methods of matching varying supply and demand have been discussed above. Actual operation involves forecasting of inflows and water demands, planning operational schedules, monitoring the supply/demand situation as the season progresses, and modifying delivery schedules where necessary. In addition to these planning functions, the project facilities must be operated from day to day, and regular and periodic maintenance must be carried out. As will be apparent from earlier discussion, operation can be largely supply-oriented and relatively simple, or demand oriented and more complex The first type of system requires fewer operational staff and has least exposure to interference with structures. The second requires more staff and is more susceptible to interference with, or damage to, structures. On the other hand it can tailor supply more closely to demand.
In the following, selected technical or administrative subjects relevant to O and M arc discussed, with particular attention to problem areas.
Poor performance of an irrigation system is often due to needed repairs long left untended and a general deterioration in effectiveness of supply. Cultivators take the law into their own hands under these circumstances and a state of operational anarchy develops, with further deterioration of the infrastructure. Poor design in the first place may be a contributing factor, with the system incapable of delivering as planned, but the primary source of the problem is usually budgetary. Funds provided are barely sufficient to meet staff salaries, with little remaining to meet essential repairs. Water charges are insufficient to cover O and M costs, and in any case they are paid into Treasury, not to the irrigation agency. Funding for O and M comes from annual appropriations from general revenues. It is usually paid reluctantly and is subject to official "norms" regarding annual cost per hectare, regardless of actual needs.
Government reluctance to spend money on repairs where the damage has been inflicted by the cultivators (e.g. regulating structures with gates repeatedly broken) is understandable, particularly in view of cultivator unwillingness to pay due water charges. On the other hand, cultivators are reluctant to pay for water supply which is unreliable due to lack of maintenance.
Funding for new construction is, in fact, easier to obtain than funding for O and M, and this also applies to financing by international agencies, which do not usually cover recurrent costs. Specific agreement regarding budgetary provisions for O and M, from one source or another, should be prerequisite to new irrigation construction or improvement projects.
A major item of cost in canal maintenance can be removal of silt. In the design of early canals in the Indus basin much attention was given to achieving a balance between deposition of silt and erosion from the canal-bed ("regime" flow). However, factors other than silt balance often determine the design of canal systems, and in projects diverting directly from major rivers without intervening storage siltation of canals remains a problem. Rivers originating in the Himalaya are particularly heavily silt-laden in the high-flow season. Provision of de-silting basins at primary canal head-works can be a solution in major projects, although still posing a problem of disposal of silt deposited in the basin. More frequently, periodic de-silting of the canal system will be necessary.
Intentionally or not, a proportion of the finer fraction of the silt entering a canal system (unless removed by headworks de-silting) will eventually find its way on to the fields. The silt is generally beneficial, but in some cased it is not. Examples are highly micaceous silts, which can have a very adverse affect on surface soil structure, and certain organic sediments which can be toxic when excavated from canal bed and disposed of on adjacent fields. Except where the potential silt problem is very obvious, practice has generally been to observe the extent of siltation which develops, before taking remedial action. The one provision which should be made in initial designs, however, is access for prospective canal-side silt clearing equipment. This requirement may be in conflict with policies regarding "public forestry" canal-side tree plantations, or may require restriction of such planting to one bank only.
Depending upon local circumstances weed control can range from a relatively minor problem to a major one, almost prohibitive in some cases. The offending vegetation may be rooted in the canal bed, floating, or it may be canal-side phreatophytes. The problem can be aggravated by the presence of lateral storage (otherwise highly desirable), which provide still-water conditions and an ideal nursery for aquatic plants which then spread into the canal system.
A degree of control can be affected by un-watering of canals for a month or more in the hot dry season, a very desirable practice from the maintenance viewpoint, but not always practical if perennial crops are being cultivated, unless alternative source of water, such as groundwater, is available. Furthermore, in conditions where watertable is relatively high throughout the year, annual un-watering of canals will not materially assist in control of canal-side phreatophytes. The latter are particularly troublesome with smaller canals (secondaries) as the proportionate reduction in water-way due to encroachment from canal-side vegetation is greater than with large canals. A 50% reduction in carrying capacity is common. The phreatophytes are frequently deep-rooted hardy plants which rapidly recover after cutting unless all roots are removed, a very difficult operation in most circumstances. Control by herbicide spray could probably be effective, but is not usually practiced in the South Asian area due to high cost and the need for frequent retreatment. Manual cutting is the only course, presenting a departmental budgetary problem unless the work is undertaken by cultivator organizations at their cost.
Bottom-growing weeds can be controlled biologically under certain conditions by fish, notably the species of Tilapia. However, in addition to their sensitivity to temperature and other conditions, and the need for re-seeding from nursery, Tilapia are edible. In a developing countries situation, this ensures their capture for home consumption. Weed control by Tilapia has not yet proved practical in the region under discussion, except under closely-controlled conditions.
Control of water-hyacinth, the most widespread of the floating plants, continues to be the subject of intensive research. To date, however, it remains a problem, with mechanical or manual removal the usual expedient. This can be a major task in heavily infested areas where flow-regulating structures can be blocked and virtually submerged by floating masses of the plant. Water hyacinth remains an ecological problem of the first magnitude.
The earlier discussion of alternative methods of operation of canal systems was in the context of completed projects. However, every major project, for which construction and development may extend over ten years or more, goes though a stage of supplying upstream portions of the service area while construction of the remainder of the canal system continues downstream. During this interim period the amount of water available in relation to the size of the area as yet under irrigation can be much more than when the full project area comes into service. The issue is whether cultivators in the upstream area should be given the interim use of this temporarily surplus water.
The case for such practice is that additional food production may thereby be achieved, a compelling argument in periods of scarcity. The case against such practice is that upstream cultivators may adopt high water-use crops such as paddy during the interim period, or poorly efficient irrigation methods, and may be very reluctant to give up the use of the surplus water in favor of downstream cultivators when the latter finally come under irrigation. Past experience indicates that this reluctance may translate, at worst, into acute political pressures by the cultivators, also technical anarchy and loss of control of the distribution system. No satisfactory solution to this problem, a commonly occurring one in greater or lesser degree, is yet apparent.
This is a perennial issue. Irrigation at night has many disadvantages. Field distribution of water is less efficient than in the daytime due to lack of visibility, although evaporative losses may be lower. Such irrigation is also most unpopular with cultivators for a variety of reasons, and it may be dangerous (dacoits, wild animals, snakes trodden on in the darkness etc). Reluctance to remain in the fields at night contributes to the poor field application efficiency, as irrigation streams are often left untended, running to waste or flooding.
With the disadvantages of night irrigation so evident the obvious question is how can it be avoided and at what cost. It is largely a question of storage, hydraulic control on the canal system, and canal capacity. In a reservoir-controlled scheme the reservoir itself could theoretically provide the overnight storage of inflows, and with sufficient control structures the canal system could (again theoretically) be made instantly responsive, permitting shutting down the entire system at sunset and re-commencing deliveries at dawn. However the cost of such a system would be prohibitive in most situations, and other solutions are sought, notably providing storage within the canal system, further downstream. This would also be necessary with a run-of-river scheme, which has no upstream storage, in which case diversion into the primary canal must continue through the night or water will be lost to the project.
The most convenient location of over-night storage, from some points of view, would be in terminal pondages supplying each tertiary or in the tertiaries themselves. As previously discussed, however, sites suitable for terminal or lateral pondages with capability of gravity inflow and outflow are rare. Pumping either into or out of the pond would generally be required, and the pondage created by embankment construction, occupying a significant area. Both requirements would be substantial obstacles to this solution.
Storage within the tertiaries, by increasing their sectional area to provide elongated overnight pondage, has been tried but only once. This was in the Gezira scheme in the Sudan. Topographic conditions (near-flat gradients) in that area were favorable to such a system, but it has nevertheless not been repeated. In more typical rolling topography it would not be practical.
Night storage within the secondary canals, associated with cyclic operation of the primaries (except in run-of-river situations) or night storage within the primary canals themselves, remain technical possibilities which could warrant investigation in particular situations. However, cost would be high. The traditional reluctance of irrigation engineers to seriously consider such systems is not simply due to their unawareness of the problems of night irrigation, but to acute awareness of the technical problems and cost of avoiding it.
In areas of water deficiency, night irrigation is in fact regularly practiced by cultivators, probably at reduced efficiency. It is in situations where the need for canal irrigation is not so pressing that cultivators may simply forgo their night-time share or headend cultivators may illegally extend their hours of day-time irrigation at the expense of tailend cultivators who then are forced into use of the night-time supply rejected by the headenders. This is in some respects a solution to the problem, except the question of night-time irrigation efficiency.
Monitoring of an irrigation project covers several different functions at successive stages of its development. In the construction phase, progress is monitored for contract management purposes and as an input to "project supervision" and control of disbursement of loan funds by the lending institution. On conclusion of the work, a completion report is filed, and later, when the project is nominally in full operation, the lending institution may prepare an ex-post evaluation report dealing primarily with economic performance and the social impact of the project (Malhotra 1987).
The area of monitoring discussed here is the technical performance of the project, including water delivered, area irrigated, irrigation efficiency, etc. Of particular importance is the feed-back of such information and its utilization in management of system operation and maintenance. For a number of reasons, the latter function is notably absent in most South Asian irrigation projects. First, such data collection is generally assumed to be for purposes of criticism and is given only token support by operational staff. Second, information on timeliness and reliability of water deliveries at the farm level is subjective at best and requires considerable judgement on the part of the collector. Thirdly, analyses of the information gathered, usually by a mid-level officer is unlikely to be particularly candid for fear of giving offence to higher level in the same Department. Information is usually simply filed, and the results of analyses, if ever carried out, are of retrospective interest only. They are not available as an input to current operations. In any case operational decisions are taken only at senior departmental levels not by monitoring staff. If senior operational staff are not sympathetic to the process of monitoring and evaluation, it becomes ineffective unless pressure is exerted by the financing agency, but that agency is no longer associated with the project after completion of construction. Notable successes in monitoring and evaluation do occur, but only where top-level management is convinced of its value and takes personal interest in its execution.
Computer application to the operation of major Western irrigation systems may be highly complex, involving main-frame computers for near-instantaneous analyses of dynamic flow situations in large canals or for their actual control. Such major systems are also found in developing countries and could be candidates for such computer installations. However, they are the exception and are outside the scope of present discussion. More pertinent is the question of how the computer may assist in the operation of more common less complex schemes.
Such assistance can cover several areas. The most obvious is in recording and processing of data, which even the simplest desk-top computer does very well. The information may be physical data on all components of the canal system, the service area of each canal and outlet, actual water deliveries, rainfall, etc. It may also include inventories of equipment and spare parts, as well as service and maintenance records of all irrigation facilities and equipment. Finally, it may include complete property ownership records and billing and accounting data In this function of data recording the computer is simply replacing the written record, but with the facility for instant recall and also for processing the data. It can also reproduce in quantity the forms required for reporting the data, in the manner required for entry into the computer.
In addition, the computer may be used as a calculator, including for instance, the estimation of consumptive water use for a range of crops and crop-mixes and climatic conditions. This information may or may not be used directly in scheduling of water releases, which may be influences by supply considerations, but it is obviously required for planning of such operations. Calculations may also include generation of hydraulic profiles in canals under various steady state conditions. With appropriate software all of the above functions can be operated by regular staff with little special training and in view of the great convenience which they provide they are likely to continue to be used if the computer facilities are provided.
In a more sophisticated category is the analysis of varying-flow situation in main canals, possibly extending to computer modeling of the complete hydraulic system. Such applications is likely to be the exception. It would require special programming and considerable computer skills on the part of the operator.
Finally, the computer, with appropriate software, may be used for economic optimization exercises, covering a range of agricultural and irrigation options and inputs as well as market conditions. Commonly referred to as "computer games", such exercises are not expected to be routinely used as an input to day-to-day system operation, but they undoubtedly are of value in promoting understanding of the economics of irrigation at project and farm levels.
In addition to technical factors, social and political pressures may also have profound influence on the functioning of an irrigation project. The question of individual and group interests in relation to operations within the tertiary command has already been referred to, and will be discussed later in connection with cultivator organizations.
There are two other factors, much less discussed, which can have a major influence on system operation. The first is political pressure, exerted by local elected representatives on field staff of the irrigation department, to secure operations favorable to their constituents. Such pressures can be acute during periods of deficiency when it is necessary to ration, restrict supply to certain types of crop, or delete supply to portions of the command. The pressure is reinforced by the ability, at political level, to secure desirable posts or transfers for Departmental staff and equally the threat of undesirable transfers.
The second factor is the existence in some areas of a parallel unofficial system of water levies. Together with kick-backs from contractors at established rates, the funds are reported to flow upwards for disbursement at various levels in accordance with traditional percentages. This system is not everywhere practiced, but where it is its functioning is well organized.
The ethics of such practices are not part of this discussion, but rather their implications regarding system operations. Imposing an unofficial levy for the supply of water implies the ability to withhold supply if the levy is not forthcoming. This requires the existence of control structures. In fact, the more sophisticated the control system the more susceptible it is to such mismanagement. This observation is not intended as an argument for universal basic simplicity in system design, but it does underline the need for consideration of the factors discussed, where they apply, in deciding upon a particular operational system.
 |
 |
