Estimation of maize evapotranspiration and yield under different deficit irrigation on a sandy farmland in Northwest China

A field experiment was conducted on a sandy farmland in Northwest China to estimate on the response of maize evapotranspiration and yield to deficit irrigation. The five irrigation treatments consisted of specific combinations of full irrigation and limited irrigation in different crop growing phases (I, from elongation phase to heading; II, from heading phase to milk; III, from milk phase to physiological maturity) were designed. And for estimation of maize evapotranspiration, reference crop evapotranspiration (ET0), basal crop coefficient (Kcb), soil evaporation coefficient (Ke), and water stress coefficient (Ks) in different treatments were calculated. Results showed that; 1) the crop actual evapotranspiration (ETc) for treatments SII (deficit irrigation in phase I), ISI (deficit irrigation in phase II), IIS (deficit irrigation in phase III), SIS (deficit irrigation both in phase I and III), and III (full irrigation) were 570, 604, 579, 542, and 607 mm, respectively. (2) The phase II was the most sensitive phase to water deficit, with reductions in leaf area index (LAI), biomass, yield, irrigation water productivity (IWP), and harvest index (HI). In this phase, the effect of water stress on Ke and Ks was slight, and the evapotranspiration has no obvious difference between full irrigation and limited irrigation. (3) Deficit irrigation in phase I can slow down the crop development in early phase, and can also reduce maize biomass and yield. In this phase, water stress obviously reduced Ke and Ks, and the evapotranspiration in limited irrigation treatments were obviously lower than full irrigation treatment in this phase. (4) However, deficit irrigation in phase III has no significant effect on height and leaf area of maize, and did not also significantly reduce maize biomass and yield. In this phase, the evapotranspiration in limited irrigation treatments were also obviously lower than full irrigation treatment. It can be concluded that it was possible to reduce water consumption and maintain the maize yield by adopting deficit irrigations from milk to physiological maturity, then from elongation to heading, but not from heading to milk in this sandy farmland regions.   Key words: Evapotranspiration, deficit irrigation, sandy farmland, maize.