Chickpea in wheat-based cropping systems of northern New South Wales. III. Prediction of N2 fixation and N balance using soil nitrate at sowing and chickpea yield

Functions quantifying relationships between N2 flxation by legumes and other factors would be useful to farmers in the management of legumes and nitrogen in their production systems. The 2 most critical factors regulating N2 flxation are legume yield and soil nitrate and both should be included as independent variables in the functions. Data from 9 experiments in northern New South Wales on soil nitrate at sowing, yields of shoot and grain dry matter (DM) and N of chickpea, and - 15 N of shoots of chickpea and non N2-flxing reference crops (wheat, barley, and uninoculated chickpea) were used to determine the percentage of chickpea N derived from N2 flxation (Pflx), total N2 flxed, and N balance (flxed N2 minus grain N). Data were then subjected to simple and multivariate regression analyses with Pflx and total N2 flxed as the dependent variables and soil nitrate, shoot N, and grain yield as the independent variables. Simple regression coe-cients (r 2 ) for Pflx were 0¢26 with shoot N as the independent variable, 0¢59 with soil nitrate, and 0¢62 with grain yield. Coe-cient values were increased in the 2-factor (multiple) regressions to 0¢74 (P<0¢001) for soil nitrate plus shoot N, and 0¢82 ( P< 0 ¢001) for soil nitrate plus grain yield. For total N2 flxed, the regression coe-cient was lower at 0¢68 ( P< 0 ¢001), using soil nitrate plus grain yield. We tested the functions against an independent data set with best prediction of Pflx involving the nitrate and grain yield equation (r 2 =0 ¢83; P< 0 ¢001). Total N2 flxed was not well predicted. We concluded that further development of such functions is warranted to reflne both accuracy and precision for chickpea and to extend the approach to other species. Additional keywords: N2 flxation regressions, natural 15 N abundance.