Search

A multi-variate approach was undertaken to identify the major components of the cropping system that interact to affect yield and water use efficiency (WUE) in wheat in northern NSW. The major components identified from an on-farm monitoring program were rainfall, soil plant available water (PAW) and nitrogen (N) at sowing, grass weeds (GW), crown rot (CR) and common root rot (CRR). Combined, these factors accounted for 80% and 60% of the variation in yield and WUE, respectively. Principal components analysis showed that the region has two seasons; summer and winter, and cluster analysis identified six rainfall zones. The three northern zones have summer dominant rainfall patterns, while the three southern zones have uniform patterns. In general, rainfall declines and variability increases moving in a westerly direction. There were 12 major soil types identified and plant available water capacity (PAWC) ranged from 45 mm (red kurosol) to 221 mm (brown vertosol). While PAWC is largely determined by clay content, sub-soil sodicity (based on exchangeable Na percentage = ESP) and salinity also reduced the amount of plant available water (PAW). The impact of ESP on PAWC in the vertosols varied, as the negative effect of high ESP was off-set by electrolyte concentrations and the fact that these soils can ‘self—repair’ by cracking. In the low to medium clay content sub-soils that have no shrink-swell capability, high ESP caused major reductions in PAWC and this effect started at ESP values below 6%. A multiple regression model, set at mean in-crop rainfall of 263 mm, predicted yields (kg/ha) for the maximum, or best observed values, for each variable (holding the other four at their mean value) as: soil N at sowing 3164, PAW at sowing 3124, CR severity 2966, CR severity 2990 and GW at tillering 3120. Yield differences between these predicted values and the mean (2847 kg/ha) were 317, 277, 119, 143 and 273 kg/ha, respectively. Yield gains for each of these components were minimal. While PAW and N at sowing were the main components that drive the system, these model outcomes show that optimising only one component if the others are sub-optimal will not realise a substantial yield gain. Holding PAW at sowing and rainfall at average values, with N not limiting (200 ng/ha), eliminating grass weeds and disease resulted in a predicted yield of 3983 kg/ha which is a gain of 40% over the mean. This equates to a WUE of 12 kg/ha/mm. To maximise yield, growers need to target all of these components within a cropping system. Fallows that retain stubble residue now dominate the cropping system and N use has risen substantially in the last 15 years. The interaction between water, N, CR and CR on wheat yield and water use was examined in a replicated experiment. The incidence of the CR pathogen, Fusarium pseudograminearum (F.ps) increased under drought conditions and the application of N, while the incidence of the CR pathogen, Bipolaris sorokiniana (B.so) increased under non—water limited conditions. F.ps led to a reduction in grain number per head and grain weight, while B.so reduced yields through the reduction of tiller density and grain weight. Yield loss due to the interaction between these disease components in the absence of severe plant water stress is discussed. Zero-tillage cropping system strategies are proposed that target high and low PAWC soils and N fertility. How growers might manage these diseases under these different cropping system regimes is discussed.