Importance of mineral nitrogen in the subsoil to yield and uptake of nitrogen by wheat in southern New South Wales

Abstract

Dry matter and nitrogen uptake of wheat tops at flowering, dry matter and nitrogen of wheat grain at maturity, and soil mineral nitrogen (0-90 cm) at sowing and flowering, were measured at 68 sites (1 experiment per site in 5 Shires) in southern New South Wales to test the hypotheses that: (i) mineral nitrogen below 30 cm would improve the prediction of wheat yields, (ii) soil mineral nitrogen would be better indicated by wheat yields at flowering than those at maturity, and (iii) soil mineral nitrogen would be better indicated by nitrogen uptake by wheat than by dry matter yields. Mineral nitrogen concentrations in soil at depths greater than 30 cm did not improve the prediction of wheat attributes, but hypotheses (ii) and (iii) were validated. Curvilinear regressions, significant (P< 0.05) on 2 occasions, were not important in this study. The best regression of wheat dry matter at flowering against soil mineral nitrogen at sowing was a single straight line, but the best models for the other 3 wheat variables were all bilinear. The best of the latter related the uptake of nitrogen by wheat at flowering to mineral nitrogen in the soil at sowing as follows: FNUH = (31.6 ¦ 5.9) + (0.892 ¦ 0.110) TMNS30 and FNUL = (9.7 ¦ 7.3) + (0.892 ¦ 0.110) TMNS30 where FNUH is nitrogen uptake by wheat at flowering (kg/ha) in 1960, 1964 and 1966 (when Shire wheat yields were above the Shire's long term average), FNUL is nitrogen uptake by wheat at flowering (kg/ha) in 1961, 1965 and 1974 (when Shire wheat yields were below the Shire's long term average), and TMNS30 is total mineral nitrogen (0-30 cm) (kg/ha) at sowing.