Effect of cropping sequences on soil nitrogen levels, rice growth, and grain yields

Abstract

The effect of 7 different crop and pasture sequences and cereal stubble management on soil nitrogen (N) levels and growth and grain yield of 2 subsequent rice crops was studied over 6 years (1986-92) on a Birganbigil clay loam soil at Yanco Agricultural Institute. After harvest of an initial rice crop in 1987, the following crop sequences were imposed: continuous annual rice cropping; 2 triticale crops, 2 seasons of pasture; pasture for 4 seasons; a second rice crop, 3 seasons of pasture; a second and third rice crop, 2 seasons of pasture; winter cereals; double cropped (winter-summer) on raised beds. The growth and productivity of 2 subsequent rice crops (1990-91, 1991-92) was then evaluated. Stubble retention rather than burning was practiced on half of each plot. Half of each pasture plot was sown to annual pasture (Trifolium subterraneum), while the other half was sown to perennial pasture (Trifolium repens). All plots returned to rice in 1990-91 and N rates up to 400 kg N/ka were applied. No fertiliser was applied to the 1991-92 crop. Anaerobic incubation ammonium content of soil, sampled before sowing of the 1990-91 rice crop, ranged from 28 to 85 pg/g oven-dried soil. The 1990-91 rice crop reduced soil ammonium by 30 y g/g in all sequences except continuous annual rice cropping. These differences in soil ammonium content were significantly correlated with rice growth and yield. Dry matter at panicle initiation and at physiological maturity and grain yield all increased significantly with N fertiliser application for all crop sequences. This indicates the benefit of N fertiliser, even after a 4-year pasture phase. Modelled grain yield from the unfertilised 2-year subterranean clover pasture (10.4 t/ha) was equal to that from the continuous rice sequence receiving 212 kg N/ha. Applying 320 kg N/ha to continuous rice plots produced yields similar to that of the 2-year subterranean clover pasture sequence, which received 98 kg N/ha. This indicates that the major effect on grain yield of different crop sequences could be alleviated through high N applications. Stubble management had little effect on rice growth or grain yield. Sequences with pasture phases of 2, 3, and 4 years produced similar rice grain yields. Pasture phases still had a positive effect, compared with continuous ricegrowing, on the second rice crop; however, crop growth and grain yield were about half that obtained from the unfertilised plots of the first rice crop. We conclude that continuous, high-yielding, annual ricegrowing can be maintained provided the N supply is adequate, that highly productive legume-dominant pasture phases of 2, 3, or 4 years contribute eqivalent amounts of N to succeeding rice crops, and that stubble management technique or pasture type has little effect on grain yield.