Changes in the acidity and fertility of a red earth soil under wheat–annual pasture rotations

Abstract

This paper reports the effects of 6 wheat–annual pasture rotations over 18 years on soil N, organic C, P, and pH in a red earth soil at Wagga Wagga (35° 03′ S, 147° 21′E), in southern NSW. There were 3 cropping intensities (33, 50, 67%) with pastures dominated by subterranean clover (Trifolium subterraneum L. cv. Bacchus Marsh) and annual ryegrass (Lolium rigidum Gaud. cv. Wimmera). Rotations were long (6-year) or short (2- or 3-year).

Initial soil N and organic C concentrations (0–10 cm) were low, 1300–1400 kg N/ha and 0·7–0·9 g organic C/100 g. The rate of increase of total N in the top 20 cm was the same on short and long rotations, and increased with the proportion of pasture in the rotation from 2·0 to 12·1 to 20·7 kg N/ha · year for pasture to crop ratios of 0·33, 0·50, and 0·67. Estimates of the amounts of N fixed and the measured accumulation of N per pasture year varied within the narrow ranges of 95–113 and 45–64 kg N/ha · pasture year. Organic C increased faster as the proportion of pasture in the rotation increased and there was no evidence that steady-state concentrations were achieved by Year 18.

Estimates of the average amount of N leached below 30 cm varied in the range 22–29 kg N/ha · year. Analysis of the individual crop and pasture effects on soil N in the surface 10 cm indicated that net nitrate leaching was greatest in the second pasture year or in the first crop year following 1 year of pasture. A significant amount of N leached during the first 2 or 3 pasture years in a rotation was recovered by the first wheat crop or by the third and fourth year pastures. Second to fourth cereal crops depleted soil N by an amount similar to that removed in the grain. Average grain N% for the rotation treatments was closely described (R² = 0·96) as a function of the length of the pasture phase, the pasture to crop ratio, and the interaction pasture to crop ratio number of preceding wheat crops.

In the top 30 cm the pH changed at a rate near –0·04 units/year on all treatments, equivalent to addition of 2·3–2·8 kmol H⁺/ha · year. The acid addition rate, and hence the long-term lime requirement (50 kg lime/kmol H⁺), did not vary with pasture to crop ratio or with the length of the rotation. The proportion of the acid added to the top 30 cm of soil that was contributed from the N cycle (nitrification followed by nitrate loss by leaching below 30 cm or by run-off) was 0·65 for rotations with 67% pasture and 0·80 for rotations with 33% pasture. Carbon cycle acids, produced during organic matter accumulation and the synthesis of products that were subsequently removed, accounted for the remainder.

Individual crop and pasture effects on soil pH were near the overall mean of –0·04 units except in years preceding and following the transition from pasture to cereal phases of the rotations. In cereal-dominated rotations the last pasture year was strongly acid (pH decrease 0·13–0·17) and the following cereal year was alkaline (pH increase 0·05–0·08). In pasture-dominated rotations the effects were reversed, the last pasture being alkaline (pH increase 0·07–0·12) and the following cereal being acid (pH decrease 0·13–0·19). In the 50% rotations, effects were intermediate.

Organic and inorganic forms of soil P in the surface 10 cm increased linearly with time, accounting for 38% of the applied fertiliser P. Of the applied P, 88% was accounted for by the sum of P accumulated in the surface 20 cm of soil and by removal in products and waste products. The remainder may have been lost by erosion or accumulated in forms resistant to extraction by 0·1 M H₂SO₄ after ignition at 550°C. There was a slightly greater rate of increase of organic P as the proportion of pasture in the rotation increased. The annual addition of 11·8 kg P/ha·year marginally exceeded the amount required to maintain the available P concentration.