Scale Effects on Water Use Efficiency and Productivity: A Case Study from UPRIIS, Philippines

Increasing water productivity of irrigated rice remains the most promising challenge to tackle the increasing water scarcity problems. The aim of this study is, therefore, to measure scale effects on water productivity through water accounting coupled with remote sensing and geographic information system (GIS) in rice-based irrigation system of District I of the Upper Pampanga River Integrated Irrigation System (UPRIIS), Philippines. The water accounting is applied at five different spatial scales to study water savings and impacts on different scales. The spatial scale ranges from a micro scale at the size of farmers field (area of 1 ha), to sub-irrigation canal scale (10,000 ha) and a system scale (area of 18,000 ha) covering the entire District I area. Daily measurements of all surface water inflows and outflows, rainfall, evapotranspiration, and amounts of water internally reused through check dams and shallow pumping are summed into seasonal totals, from November 19, 2000 to May 18, 2001 for five spatial scale units ranging from 1 to 18,000 ha. Water accounting technique is applied to measure performance indicators of the water productivity at each spatial scale. Results from a field scale show that water productivity per unit of delivered water is higher than all spatial scales due to best management practices and high input of fertilizers (180 kg/ha). The process fraction of gross inflow is very high which shows that farmers put lot of efforts to make full use of irrigation water and rainfall. These observations at field scales indicate that farmers are very effective in capturing and utilizing all water input. However, the amount of net surface water input (rainfall plus irrigation) per unit area decreases and the process fraction, depleted fraction, water productivity, and amount of water reuse increases with increasing spatial scale. In total, 57% of all available surface water is reused by check dams and 17% by pumping. The process fraction of all surface water input (irrigation and rainfall) is very high (0.71) at the field scale and it is relatively low (0.15) at TRIS-L scale but gradually increases with spatial scale up to 0.22 for all scale units combined. The major reason of improving water productivity is due to large volume of re-used water (30 % of water lost through ETa of rice) from 15 small check dams, hundreds of small farm ponds and 1451 pumps installed in District I. Water productivity with respect to ET of rice varies between 1.75 and 0.8 kg grain m(-3) water at field and District scale). The only option to increase WPET at large spatial scales is again crop protection measures such as pest and disease control and reduced post-harvest losses. The overall water productivity with respect to available water (WPavailable) is 0.45 kg grain m(-3) water, which compares well with the average of 0.4 kg grain m(-3) water for rice at the field level. At the field scale, the WPavailable is 1.25 kg grain m(-3) water, suggesting that there is considerable scope for improvement in the study area. The results show that water re-use plays a dominant role in the growth of a rice crop during the dry season. The finding shows that scale effects are important for understanding, planning for water saving and for planning appropriate measures to increase water productivity. The results clearly indicate that the quantification of volumes of water re-use is crucial for understanding and finding real water saving possibilities at the irrigation system level. These findings would lead to an improvement in the water use efficiency and water productivity of irrigated rice systems.