Controlled traffic is a new technique aimed at reducing energy while restricting soil compaction to a limited area of each field. However, the extent and magnitude of different degrees of compaction on infiltration, runoff, soil water and crop yield are not clearly understood. This thesis describes soil physical changes and the related water balance and crop response, associated with alternative traffic and tillage (residue cover) management options. These soil changes are described in terms that allow treatments to be represented in a soil-crop system model. This allows for interactions relevant to controlled traffic farming of vertosols to be explored.
Self-mulching, black vertosols are a major proportion of the cropped soil area in sub-tropical Australia, where dryland crop production is usually limited by available moisture. In this environment rainfall is unreliable, summer dominant and sometimes of high intensity. Conservation tillage has been developed to protect the surface from rainfall drop impact and to reduce runoff and soil erosion. Tillage, soil and crop establishment systems have been explored in detail and incorporated into soil-crop simulation models.
Traffic and tillage effects on runoff, soil water and crop yield were monitored for 5 years under natural rainfall to provide parameters for model simulation. A major and consistent reduction in runoff (35%) was achieved with controlled traffic compared with wheeled soil, and a smaller but consistent reduction was achieved under zero tillage (high residue cover). Improved infiltration resulted in a significant increase in plant available water and crop yield under controlled traffic zero tillage.
Rainfall simulator experiments were used to independently characterise variation in infiltration and runoff associated with changes in axle load and wheelslip. Simulated rainfall provided independent estimates of soil parameters for the PERFECT model. Curve number (CN) values of 78 and 93 were determined for controlled traffic and wheeled soils (bare at average antecedent moisture content) respectively. The maximum reduction in CN due to residue cover for controlled traffic soil was 23, but only 3 for wheeled soil, indicating that compacted soil layers were restricting infiltration.
Measured daily runoff, soil water and crop yield for traffic and tillage treatments were used to test the PERFECT soil-crop simulation model. The model already incorporates tillage and fallow management and has been tested on a range of soil types and cropping systems in Australia, India and China, but does not explicitly capture compaction effects. Calibration of this model has normally been achieved by optimising CN, while saturated hydraulic conductivity (Ksat) was held constant for the soil profile. In this study experimental measurement was used as the starting point of a calibration optimisation procedure to estimate CN and Ksat for surface and subsurface layers, respectively. The good correlation between measured and predicted data indicates the model's success in capturing traffic and tillage effects. Subsurface Ksat under controlled traffic increased 4 fold.
Long-term predictions showed that controlled traffic zero tillage, applied in an opportunity cropping system, was the optimum dryland cropping practice for utilising rainfall, reducing runoff and soil erosion, and increasing soil water storage and crop yield. In contrast, wheeled stubble mulch in a continuous wheat monoculture gave poor rainfall utilisation, plant available water, crop yield, and increased runoff and soil erosion. The PERFECT soil-crop simulation model is a useful tool for exploring the interactions between traffic, tillage, soil and crop, and provides insight into the long-term effects of improved soil management and crop rotation options.
Rainfall energy has been commonly used as an index of the surface seal development associated with raindrop impact. Wheeling energy is proposed here to quantify the effects of wheeling on soil infiltration characteristics. Measurements taken during this study indicated that wheeling energy is a simple and versatile index of damage to subsurface layers. Further study is needed to incorporate tillage energy, wheeling energy, rainfall energy and appropriate natural and management repair functions into a soil-crop simulation model.