Water management in irrigated rice - Rice Knowledge Bank ...

given in Table 1.1. Water losses by seepage andpercolation account for about 25–50% of all waterinputs in heavy soils with shallow groundwater tablesof 20–50-cm depth (Cabangon et al 2004, Donget al 2004), and 50–85% in coarse-textured soilswith deep groundwater tables of 1.5-m depth ormore (Sharma et al 2002, Singh et al 2002). Thoughseepage and percolation are losses at the field level,they are often captured and reused downstream (justlike overbund flow; Chapter 5.1). The actual amountof water reuse in rice-based irrigation systems isnot known and is expected to vary widely amongirrigation systems.Daily seepage and percolation losses from theponded water do not occur in dryland crops suchas wheat and maize. Percolation of water below theroot zone can also occur in dryland crops when theamount of water infiltrating into the soil (either afterheavy rainfall or after irrigation) is larger than thestorage capacity of the root zone, but this is not adaily water flow as in lowland rice. Also, evaporationfrom ponded water surfaces is higher than fromsoil surfaces (as in dryland crops). Therefore, it isthe relatively large water flows by seepage, percolation,and evaporation that make lowland ricefields heavy “water users.” Total seasonal waterinput to rice fields (rainfall plus irrigation) can beup to 2–3 times more than for other cereals such aswheat or maize (Tuong et al 2005). It varies fromas little as 400 mm in heavy clay soils with shallowgroundwater tables (that directly supply waterfor crop transpiration) to more than 2,000 mm incoarse-textured (sandy or loamy) soils with deepgroundwater tables (Bouman and Tuong 2001,Cabangon et al 2004). Around 1,300–1,500 mm isa typical value for irrigated rice in Asia. Table 1.1lists some values for water inputs and daily seepageand percolation rates for lowland rice fields inChina and the Philippines.It is useful to distinguish between the wateroutflows from rice fields that can, in principle, bereused and those that cannot be reused. Nonreusableoutflows are called “depleted water” (Molden1997) and are evaporation and transpiration. Overbundflow, seepage, and percolation are generallyreusable flows. Only when these flows enter verydeep or saline groundwater (or saline surface water)from where they cannot be recovered are theynot reusable any more and they become depletionflows as well.1.4 Groundwater under rice fieldsThe role of groundwater in providing water to riceplants may be large, but it has been neglected inmost studies of the rice water balance. Recent datacollection suggests that through the (decade- toage-old) practice of continuous flooding, the largeamounts of percolating water have raised groundwatertables to very close to the surface. This isespecially true in soils with a heavy texture that arepoorly drained in the subsoil, as is the case in manytraditional irrigated rice environments.Figure 1.3 gives the groundwater table measuredunder flooded rice fields at some sites in Chinaand the Philippines. When the groundwater is lessthan 20 cm deep, it provides a “hidden” source ofwater to the rice crop as the roots of the plants candirectly take up water from the groundwater. Whenfor some reason fields are not flooded (Chapter3), capillary rise may reach into the root zone andagain provide extra water to the crop. In mostwater balance studies, the effect of groundwateron water supply is not taken into account, and thebeneficial effect of water-saving technologies canbe overestimated. With shallow groundwater, cropgrowth with a small irrigation water supply can stillbe good because of the “hidden” water supply ofgroundwater.1.5 Rice water productivityWater productivity (WP) is a concept of partialproductivity and denotes the amount or value ofproduct (in our case, rice grains) over volume orvalue of water used. Discrepancies are large in reportedvalues of WP of rice (Tuong 1999). These arepartially caused by large variations in rice yields,with commonly reported values ranging from 3 to8 tons per hectare. But the discrepancies are alsocaused by different understandings of the denominator(water used) in the computation of WP. Toavoid confusion created by different interpretationsand computations of WP, it is important to clearlyspecify what kind of WP we are referring to andhow it is derived. Common definitions of WP areWP T: weight of grains over cumulative weightof water transpired.WP ET: weight of grains over cumulative weightof water evapotranspired.WP I: weight of grains over cumulative weightof water inputs by irrigation.