Tillage and crop residue management affect Vertisol properties and grain sorghum growth over seven years in the semi-arid sub-tropics. 3. Crop growth, water use and nutrient balance

The effects of fallow surface management treatments on subsequent growth and yield of grain sorghum were studied during seven cropping periods on a grey Vertisol near Biloela in central Queensland, Australia. Treatments were disc (D), blade (B) and zero (Z) tillage, each with stubble or residue from previous crops either retained ( + ) or removed ( − ) at the start of the fallow period. Treatments were planted to grain sorghum between December and February, following a 7–8 month fallow period.

Plant establishment was significantly (P<0.05) lower in Z than in D and B treatments in five of the seven crops. Average plant establishment over 7 years was about 20% lower in Z treatments. Early crop growth was generally more vigorous in D and B than in Z treatments, but this trend became less pronounced at later stages of growth. Surface cover (stubble+crop) ranged from 46% in Z + to 26% in Z− 5 weeks after planting and from 68% to 57% at anthesis and 62% to 46% at maturity in these same treatments. For the seven crops, annual mean grain yields over all treatments ranged from 1650 to 4490 kg ha⁻¹. Z− had a significantly (P<0.05) lower mean grain yield over the seven crops than all other treatments, which did not differ significantly (P>0.05). Where stubble was retained, yields of D+ and B+ never differed significantly. However, yields from Z+ were significantly higher in three crops than those from D+ and B+ and lower in three other crops. Stubble retention consistently resulted in significantly higher yields than stubble removal in the Z treatment. Crop water use was consistently lower in Z− than in all other treatments. Of the total roots to a depth of 1.6 m, a mean of 28% were found at 0–0.1 m. At 0.1–0.2 m and in deeper 0.2 m increments, the mean ranged from 6 to 11%.

Over 7 years, decreases in soil organic carbon at 0–0.1 m with stubble retained were less than with stubble removed. The differences corresponded to 4.1, 10.0 and 6.2% of stubble carbon inputs in D, B and Z treatments, respectively. Nitrogen removal in grain and stubble, expressed as a fraction of soil total nitrogen at 0–0.2 m plus nitrate-nitrogen at 0–1.6 m, ranged from 43% for D+ to 82% for Z+. Decreases of bicarbonate- and acid-extractable phosphorus at 0–0.1 m each corresponded to between 31 and 50% of the phosphorus removal in grain and stubble.