Herbicides are the foundation of weed control in commercial crop-production systems. However, herbicide-resistant (HR) weed populations are evolving rapidly as a natural response to selection pressure imposed by modern agricultural management activities. Mitigating the evolution of herbicide resistance depends on reducing selection through diversification of weed control techniques, minimizing the spread of resistance genes and genotypes via pollen or propagule dispersal, and eliminating additions of weed seed to the soil seedbank. Effective deployment of such a multifaceted approach will require shifting from the current concept of basing weed management on single-year economic thresholds.

The farm-level economics of conservation agriculture (zero tillage, mulching and crop rotation) are described, reviewed and modelled. The economics are defined broadly to include not just shortterm financial benefits and costs, but also the whole-farm management context, constraints on key resources such as labour and capital, risk and uncertainty, interactions between enterprises, and time-related factors, such as interest rates and the urgency of providing for the farm family. A wealth of evidence shows that these economic factors and variables related to them have significant influences on farmers' decisions about adoption of conservation agriculture. Literature on the farmlevel economics of conservation agriculture for resource-poor farmers is reviewed. There is not a large body of high-quality relevant studies. Those that have been published highlight that the economics are highly heterogeneous and need to be considered on a case-by-case basis. Their results tend to indicate that it would be profitable to adopt conservation agriculture or components of it (although not in all cases). This contrasts with disappointing adoption in many of the regions of interest. Potential reasons for this disparity are discussed. A general model of the farm-level economics of conservation agriculture and its components is presented, and used to illustrate influences on the overall economic attractiveness of conservation agriculture. Key factors that would tend to discourage adoption in situations that otherwise look favourable include: the opportunity cost of crop residues for feed rather than mulch, the short-term reduction in yields under zero tillage plus mulching in some cases, combined with short planning horizons and/or high discount rates of farmers, farmer aversion to uncertainty, and constraints on the availability of land, labour and capital at key times of year. Good quality economic analysis should be used more extensively to guide research and extension in this area, particularly in relation to the targeting of effort, and adaptation of the system to suit local conditions.

In 2003, a random survey was conducted across the Western Australian wheatbelt to establish the frequency and distribution of herbicide resistance in ryegrass populations infesting crop fields. Five hundred cropping fields were visited at crop maturity, and ryegrass seed was collected in 452 of these fields. Subsequently, each crop field population was screened with herbicides of various modes of action that are commonly used for ryegrass control in Australian cropping systems. Most of these ryegrass populations were found to be resistant to the ACCase-inhibitor herbicide diclofop-methyl (68%) and the ALS-inhibitor herbicide sulfometuron (88%). A comparison of resistance levels in the same agronomic zones surveyed 5 years earlier determined that there had been an increase of 20 percentage points in the frequency of resistance over this 5-year period. This survey also determined that the majority (64%) of populations were found to be multiple resistant to both diclofop-methyl and sulfometuron. The distribution patterns of the collected populations indicated that there were higher frequencies of resistant and developing resistance populations occurring in the intensively cropped regions of the wheatbelt, which had greater herbicide selection pressure. Of concern is that 24% and 8% of populations were found to be developing resistance to trifluralin and clethodim, respectively. Currently these herbicides are heavily relied upon for control of ACCase and ALS herbicide resistant ryegrass. Nearly all populations remain susceptible to glyphosate. Ryegrass across the WA wheatbelt now exhibits multiple resistance across many but not all herbicides, posing severe management and sustainability challenges.

No abstract