Quantifying and benchmarking irrigation scheme performance with water balances and performance indicators.

Abstract

South Africa is a water scarce country. As pressure on available water resources increases, irrigation, the largest consumer of water, has to find ways of improving water use efficiency. Benchmarking in the irrigation sector has been identified as a suitable technique to implement this improvement. Benchmarking can be broadly defined as the identification and application of organisation specific best practices with the goal of improving competitiveness, performance and efficiency. A South African sugarcane irrigation scheme was identified to investigate a proposed benchmarking methodology. The scheme was unique in that electromagnetic flow meters were utilised and monitored on a daily basis. This facilitated an in depth study into irrigation water use at the scheme. The project focused on three different objectives. The first objective was to determine the losses, and consequently the efficiency, with which the irrigation scheme was able to deliver irrigation water from the water source to the farm boundary during the years 2004 and 2005. This was achieved by completing the water balance for the scheme with specified geographic and temporal boundaries. Results indicated that the scheme was very efficient with a delivery efficiency of 83.4 and 94.0 % for 2004 and 2005 respectively. These efficiencies were above the accepted South African Department of Water Affairs and Forestry (DWAF) standard of 80 %. The temporal distribution of the delivery efficiency was also investigated to identify periods within each year when inefficiencies occurred, and to better understand the nature of potential losses. It was concluded that the investigations into the temporal distributions be utilised together with the water balance approach in future studies into the performance of irrigation water delivery infrastructure at other South African irrigation schemes. The second objective was to calculate a set of internationally applied external irrigation benchmarking indicators. External indicators from the International Water Management Institute (1WMI), the International Program for Training and Research in Irrigation and Drainage (IPTRID) and the Irrigation Training and Research Center (ITRC) were reviewed for application in a South African context. The external indicator analysis highlighted that at a scheme level, insufficient irrigation was occurring to effectively meet the irrigation demand. It was also found that the scheme infrastructure was not the limiting cause of this observation. The external indicator results highlighted the need for additional schemes for comparison purposes. The results from this component of the study also emphasized the importance of stakeholder confidentiality concerns when attempting to implement a benchmarking initiative. The third objective was to rank individual farm performance of all the farms in the scheme, in terms of total farm sugarcane yield and seasonal irrigation water use. Farm yield and irrigated area were obtained to investigate the relationships between yield and irrigation water application. There were substantial variations in total farm yield and water use for both the 2004 and 2005 seasons, indicating much potential for improvement by many farmers relative to each other. The individual seasonal farm water use was also compared to a simulated irrigation demand, as determined with the SAsched irrigation systems and crop yield model. Simulation results with the SAsched model, using representative soils and climate data for the scheme, showed that the majority of farms were under irrigating relative to the simulated demands, especially in the late spring/early summer period. From on-farm irrigation system evaluations that were performed, it was found that irrigation system capacity constraints were not limiting irrigation applications in the majority of farms. Further research in the form of selected soil water monitoring is required to investigate these observations further.