Benchmarking and multivariate data analysis techniques for improving the efficiency of irrigation districts: An application in spain

https://doi.org/10.1016/j.agsy.2007.07.010Get rights and content

Abstract

Performance indicators are a powerful tool for identifying deficiencies in irrigation district management and determining which measures should be taken to improve them. This process is known as Benchmarking. Until now, analysis has been based on direct comparisons of performance indicators from different irrigation districts. However, this procedure does not provide an overall view of the actual performance of each district in relation to others. Furthermore, on some occasions irrigation districts are compared with very different ones and best practices cannot be adapted to organisations having lower performance.

In this paper, a methodology to analyse performance indicators is developed and applied to nine irrigation districts in Andalusia (Spain). The methodology is based on multivariate data analysis (cluster analysis), thereby enabling irrigation districts to be classified into statistically homogeneous groups. The irrigation districts have also been ranked according to an index developed in this work which aggregates all the methodology. This classification allows each irrigation district to be compared to another with similar characteristics. Our results demonstrate great differences in terms of performance between districts with open channel water delivery systems and those with pressure water delivery systems. In districts where users are charged per unit of irrigation water consumed, water use was found to be more efficient.

Introduction

Benchmarking is defined as “a process of learning from your own past performance and the performance of others in pursuit of continuous improvement” (Malano and Burton, 2001).

The application of benchmarking techniques to improve irrigation district performance is a relatively recent phenomenon. The main objective of this technique is to enhance the performance of a given irrigation district by comparing its current performance with that of other districts. In this way, it is possible to determine which practices lead to better performance in a district and subsequently adapt these practices to irrigation districts that perform less efficiently. Similarly, irrigation districts that perform more poorly will be able to determine which aspects are in need of improvement and take the necessary steps to achieve better performance.

Performance indicators are the main tool in a benchmarking process. A performance indicator is a ratio that relates variables (i.e. irrigated area, volume of irrigation water applied or productivity) in such a way that a large amount of information can be reduced to a single number. By comparing performance indicators it is possible to determine when an irrigation district is more or less efficient than another and take the necessary measures to correct any existing deficiencies.

Although a great number of researchers have developed and applied sets of performance indicators to measure the efficiency and sustainability of irrigation systems (Rao, 1993, Schoups et al., 2006, Gorantiwar and Smout, 2005, Smout and Gorantiwar, 2005, Molden et al., 1998, Garcés, 1983, Alexander and Potter, 2004, Burt and Styles, 2000, Malano et al., 2004), there are few examples in the literature of improving efficiency by comparing several irrigation districts by means of performance indicators. Nevertheless, performance indicators are becoming an increasingly important tool in irrigation district management. This is the case, for example, of Australia where performance indicators have been applied to irrigation water management since 1996 (Hydro Environmental, 2002). However, sometimes very different districts are compared and best practices cannot be adopted by less efficient ones.

In this paper, a further step is taken in the analysis of irrigation districts using performance indicators. In order to do so, a methodology was developed that allows the state of the districts to be studied and districts with similar characteristics to be compared using multivariate data analysis techniques (principal components analysis and cluster analysis). The results of the principal components and cluster analyses are then aggregated into a single quality index (QI).

By combining benchmarking and multivariate data analysis, policy makers are able to detect the weak points of each irrigation district and apply the necessary actions to correct performance gaps (which may involve major investments in new irrigation systems or simply improvements in the management of current infrastructures). Moreover, by comparing performance before and after modernisation actions, policy makers can ensure that the aim of the investments has been properly achieved.

This methodology is applied to several irrigation districts in Andalusia (Southern Spain). With 3.3 Mha (Del Campo, 2002), Spain has the largest irrigated area in all of Europe; while Andalusia, with almost 0.9 Mha, is the region with the most extensive irrigated surface in the country (Junta de Andalucía, 2002). In the last century, irrigated agriculture grew rapidly in Andalusia (Rodríguez Díaz et al., 2007), thus putting available water resources under increasing pressure. With a view to conserving supply, the government has implemented far-reaching programmes for modernisation to replace old gravity irrigation systems with more efficient pressurised schemes using trickle or sprinkler irrigation. As a result, trickle irrigation has recently become the most widespread system in the region. However, there is no clear methodology to aid in determining where investments should be made, what sort of actions are needed in each irrigation district or to evaluate the outcome of the investments.

In the future, methodologies to ensure that resources are used efficiently will be important for the Spanish irrigation sector if it is to comply with current legislation such as the water framework directive of the European Union (EU, 2000), which takes into account the principle of costs recovery or the effects of climate change, where it is predicted that irrigation needs will rise in the country (Rodríguez Díaz et al., 2006).

Section snippets

Study area

The methodology was applied in the region of Andalusia. The climate in the region is predominantly Mediterranean, with rainfall mainly in the autumn and spring and dry spells in the summer. Due to the vagaries of the Mediterranean climate, irrigation is necessary to produce higher value summer crops which would be impossible under rainfed conditions and to assure yields in periods of drought. For this reason, there are marked differences in productivity between irrigated and non-irrigated lands

Methodology

The methodology developed in this work is divided into four steps. The first step is to select a suitable set of performance indicators that can be used to characterise the irrigation districts. Secondly, the initial sample of indicators is reduced to a smaller number using principal components. The irrigation districts are then divided into statistically homogeneous groups. Lastly, all the methodology is synthesised in a complex index developed from the previous steps. Each step is discussed

Characterisation of the irrigation districts and selection of the most representative indicators

Historical series of performance indicators for periods of up to six years were obtained previously for each irrigation district (Rodríguez Díaz, 2004). As the performance indicators are highly correlated, a principal components analysis is performed prior to the cluster analysis. The coordinates of the 21 IPTRID indicators for the first four principal factors (which explain more than 75% of the variance) are shown in Table 3. The components are ranked in descending order according to the

Conclusions

Benchmarking and multivariate data analysis techniques are powerful tools to evaluate efficiency in irrigation districts. These techniques make it possible to group the initial sample in an objective way according to the main characteristics of the irrigation districts under study.

The methodology developed has been aggregated into a quality index. This index could be a useful tool for policy makers to monitor the actual performance of irrigation districts, analyse their strong and weak points,

References (25)

  • N.M. Holden et al.

    Definition of agroclimatic regions in Ireland using hydro-thermal and crop yield data

    Agricultural and Forest Meteorology

    (2004)
  • P. Alexander et al.

    Benchmarking of Australian irrigation water provider businesses

    Irrigation and Drainage

    (2004)
  • Burt, C.M., Styles, S.W., 2000. Modern water control and management practices: impact on performance. Water Report 19,...
  • Corominas, J., 1996. El Regadío en el umbral del siglo XXI: Plan Nacional de Regadíos y Plan de Regadíos de Andalucía....
  • Del Campo A., 2002. El futuro de los regadíos españoles en el contexto de la directiva europea del agua. International...
  • EU, 2000. Water Framework Directive of the European Parliament and of the Council (2000/60/EC),...
  • Garcés, C., 1983. A methodology to evaluate the performance of irrigation systems. Application to Philippine national...
  • S.D. Gorantiwar et al.

    Performance assessment of irrigation water management of heterogeneous irrigation schemes: 1. A framework for evaluation

    Irrigation and Drainage Systems

    (2005)
  • J.F. Hair et al.

    Multivariate Data Analysis

    (1998)
  • Hydro Environmental, 2002. Australian irrigation water provider, benchmarking report for 2000/01. Australian National...
  • A.K. Jain

    Statistical pattern recognition: a review

    IEEE Transactions on Pattern Analysis and Machine Intelligence

    (2000)
  • Junta de Andalucía, 2002. Actualización del Inventario y Caracterización de los Regadíos de Andalucía,...
  • Cited by (0)

    View full text