Optimizing water and fertilizer input using an elasticity index: A case study with maize in the loess plateau of china

Abstract

Matching fertilizer rates with available water supplies in water-scarce environments remains a major challenge for improving water use efficiency and crop yield. The objectives are to (i) develop a new approach to characterizing interrelations of yield ( Y), evapotranspiration (ET), water use efficiency (WUE), and soil fertility using an elasticity index, and (ii) to further derive optimal-coupling domains of water and fertilizer inputs using maize data of 1997 and 1998, as an example. The experiment was an incomplete factorial design with two factors (water supply and fertilizer input) with five levels each, and had a total of 13 treatments with three replicates each. A maize cultivar (Zhongdan 2, Zea mays L.) was grown in a loessial silt loam in the hilly region of the Loess Plateau of China. Irrigation was hand applied at predetermined amounts as needed, and fertilizers including nitrogen, phosphate, and yard manure were applied at planting and jointing at predetermined rates. Approaches on how to use the crop–water production function and elasticity index (EI) to characterize the interrelations of Y, ET, and WUE were presented, and further extended to derive the optimal-coupling domains of water and fertilizer inputs. Yield responses to water and fertilizer inputs followed a quadratic function with a positive interactive term. When constrained by local maximum yields, the optimal-coupling domain took a half-ellipse form with the global maximum WUE and Y (or maximum ET) corresponding to the left and right end points on its long axis. As water supply increased, WUE reached its maximum before yield did. If water supply is limiting, fertilizer rates that maximize WUE rather than yield should be used; otherwise, seeking maximum yield may be desirable. For irrigation management, total water supply to maize should not exceed 550 mm in the region. Furthermore, the optimal domain can be used to determine optimal fertilizer rates for any given water supply, which may be estimated from seasonal climate forecasts in the case of dryland farming or based on available water supply for future irrigation. For a given water supply, fertilizer rates should be between the rate of reaching local maximum WUE and the rate of reaching local maximum yield.