Shoot and stalk dynamics and the yield of sugarcane crops in tropical and subtropical Queensland, Australia

Yield constraints have become an important feature of sugarcane farming systems in Australia, with significant yield increases obtainable by manipulating farming system components (e.g. crop rotation). However, to identify key crop targets for new farming systems, an improved understanding of the physiological mechanisms mediating these yield responses is required. Data from experiments investigating impacts of crop rotation, plant population, tillage and trash management have been analysed to determine the impact of treatments on shoot and stalk dynamics and crop yield. Crop rotation or soil fumigation resulted in significantly improved crop establishment, with production of greater numbers of both primary and higher order shoots. Manipulation of soil N status in these experiments had limited effect on the initiation of secondary tillers, regardless of whether the N was the result of a rotation treatment or of fertiliser N applications within a common rotation history. In all cases, effects of N were minimal compared to those of other components of soil health affected by fumigation (primarily detrimental soil biota) and rotation (differing elements of physical and chemical fertility and soil biota). Survival of tillers until final harvest was strongly influenced by the population density of established primary shoots and the growing conditions during the latter part of the growing season. Reduced water stress and application of N fertiliser near the time of maximum tiller numbers increased tiller survival. The relative importance of final stalk population density and individual stalk dry mass to variation in yield in each experiment, varied widely. The cultivar Q 117, which was used in experiments in northern regions of Mackay, Burdekin and Tully, achieved relatively low final stalk densities (4-8 stalks m(-2)) and yield variation in these experiments was primarily due to treatment effects on this attribute. Experiments in the south at Bundaberg used a number of cultivars (Q124, Q141, Q188) which all achieved higher stalk population densities (9-12 stalks m(-2)) with yield variation primarily affected by individual stalk mass. Experiments designed to broaden the range of plant densities tested revealed a negative relationship between individual stalk mass and stalk population density. These data show that treatments that increase stalk population densities at harvest without also increasing rates of dry matter accumulation during stalk filling are unlikely to increase yields. Improvements in soil health afforded by crop rotation and soil fumigation deliver yield benefits by achieving both higher population densities and heavier stalks. (c) 2005 Elsevier B.V. All rights reserved.