On reliability in irrigation service preliminary concepts and application

Without doubt the performance indicator that has so far received least attention in irrigation is the reliability of irrigation service. A review of reliability approaches as reflected in the literature is made and the analysis leads to propose a new conceptual framework for reliability of irrigation.The reliability of irrigation service is defined as the degree to which the irrigation system, and its water deliveries, conform to the prior expectations of its users. The perception of the user is central to the process of defining expectations and to the process of making strategic and tactical choices for the cropping pattern, the quantity of inputs, etc.It is proposed here to use a composite reliability approach, combining two key aspects: the variability and the predictability of the service characteristics. The methodology consists firstly of defining a set of three, non-overlapping, first-order indicators which fully characterize irrigation deliveries. These indicators are: Adequacy, Timeliness and Steadiness. Then the variation (in time or in space) of these indicators leads to Second-order indicators of performance. One second-order indicator is the well-known Equity, which assesses the spatial distribution or the distribution between shareholders.Variability and predictability are the two main aspects of the proposed concept of reliability. The variability is the temporal variation of the first-order indicators while the predictability defines the extent to which irrigation can be forecast. Low variability means high predictability, whereas high variability does not necessarily mean low predictability, as patterns of variation can possibly be deduced from past events.The reliability is inversely proportional to the frequency and the magnitude of failure. Here the magnitude of failure in irrigation service is measured by the deviation of the actual from the expected delivery.Existing approaches and the proposed conceptual framework are applied to a case study in the Punjab-Pakistan. It shows that reliability indicators proposed so far are not doing any better than the classical approach of adequacy. The new proposed approach for reliability, based on the combination of variability and predictability, seems more promising.Finally this preliminary analysis also identifies several key domains for which further investigations should be undertaken to increase the general understanding of reliability.