D.A. Derksen and K. McGillivray. AAFC, Box 1000A RR#3, Brandon, MB. R7A 5Y3

Herbicides are an important component of crop production in the northern Great Plains region. When used appropriately they reduce crop yield losses caused by weeds; however, herbicide usage does not necessarily ensure optimal yields or net-returns. In 1993, it was estimated that on the Canadian prairies economic losses from weeds were $612 million after herbicide application (Swanton et al. 1993). How can this be? In Canada, herbicides must get greater than 80% control to be registered with many products having 90% control, therefore, if weed densities are high even 90% control can leave high enough densities to cause economic loss. The key to weed control is to use cropping systems to keep weed densities low enough to get maximum benefit from herbicides or to avoid the use of herbicides when possible.

Herbicides represent 20-30% of input costs for most producers. An estimated $850 million were spent on herbicides on the Canadian prairies in 1999. If herbicide usage could be reduced by 10%, without reducing crop yield or increasing weediness, $85 million in savings could be had by producers. To put this in context, this is more money that all producers in Saskatchewan netted in 1998. Reducing herbicide input costs can be achieved by either avoiding the necessity for herbicide usage by keeping weed densities below economic threshold levels or by reducing herbicide rates.

To be of long-term benefit to producers, reduced herbicide usage can not result in greater weediness over time, therefore, reducing usage can only be done in the context of an integrated weed management (IWM) program. The least controversial method of reducing usage is to avoid the need for some herbicides by using crop rotations and varied seeding dates. For example, wild oats are seldom problems in late seeded crops, thereby eliminating the need for wild oat herbicides. Reducing herbicide rates is more controversial. If reducing rates is to be effective, efficacy must be similar to full rates or high enough for final weed population densities to be below threshold levels. In Canada, off label usages of herbicides are illegal; however, to be prosecuted for reducing rates it would have to be proven that there was an increased safety risk to the user, consumer, or environment. In some countries, reduced rates are not only legal but information is provided to producers as to how to reduce rates and achieve the same efficacy levels. The real issue with reduced rates is liability. Currently, herbicide companies and suppliers do not provide support for off labelled usages.

Reducing herbicide usage without using other weed management strategies will likely lead to failure. Using the principles of weed management is critical, that is, keep weed densities low when crops are establishing, give the crop an advantage over weeds whenever possible (high seeding rates, selective fertilization, competitive varieties, etc), and keep weed communities off balance with a diversified production system (rotate: dates of seeding, herbicide windows of application (include pre-harvest), and summer, winter, and perennial crops). Finally, make sure that densities of weeds are below threshold levels before deciding to skip a herbicide application and are low when reducing herbicide rates.

Reducing the usage of the expensive herbicides will have the greatest impact on net-returns. In the past, the high price of glyphosate led to research on means of reducing rates; however, since prices have fallen four to five fold, there is not as great an incentive to reduce rates. Currently, the herbicides which are the most expensive are the grassy weed herbicides and efforts are being made to reduce the need for them. This is particularly important given the high levels of weed resistance in grassy weeds. Reducing the number of times a grassy weed herbicide is required in a field is the best way to delay the onset of resistance.

In the long-term, using the best management strategy for control specific weeds will reduce overall herbicide usage. For example, although Canada thistle can be suppressed with pre-seeding and in-crop herbicide applications, the best long-term cost effective control comes from pre-harvest applications of glyphosate. Furthermore, this strategy is a good means of controlling many perennial weeds, and using pre-harvest or post-harvest glyphosate every few years in a crop rotation will reduce overall expense for perennial weed control.

Although weeds emerge throughout the growing season, most species have definite emergence patterns. For example, the peak emergence of wild oats occurs early in the spring while that of green foxtail occurs later. Getting a crop through the stage when it is susceptible to yield loss with weed densities below threshold values means that herbicides will not be required. This stage is called the critical period. O’Donovan has shown that wild oats emerging 5 days after barley do not reduce yield regardless of their density (1985). Wheat is susectable to yield loss in the 1-4 leaf stage. Research on the critical period for canola is currently conducted by Van Acker and Martin at the University of Manitoba. Therefore, timing seeding so that a specific weed, such as wild oat, is not present during the critical period will reduce herbicide usage. This can be done in two ways. Firstly, seeding early so that the weed emerges much later than the crop or by seeding after the weeds have emerged and have been removed with pre-seeding burn-off herbicides.

Research by Van Acker and Marginet at the University of Manitoba is being done to determine the emergence patterns of common weeds in zero and conventional tillage systems with the goal being to predict emergence time based on heat unit accumulation. Determining if herbicide application is necessary and timing seeding or herbicide applications would be aided by the knowledge of weed emergence patterns.

Most grassy weed herbicides used in the NGP have one or two rates on the label. In Denmark and Sweden, producers are given information to use reduced rates. The Denmark decision support system allows producers to chose their desired level of control (eg:90%) as well as input growing conditions, and weed and crop densities. The program tells producers the appropriate rate to use whether 25% or 100% of the recommended rate for all registered herbicides and crops. Note that the reduced rates provide the same efficacy as full rates! The Swedish approach is similar, but instead of using a computer system, they use a simple chart to obtain similar information, in this case only for phenoxy herbicides. Past attempts to implement reduced rates in the NGP has met with great resistance by industry. Given the current farm economic crisis, pressures to reduce herbicide use for environmental reasons, and interest in using variable rate technology in precision farming, it is time for industry and farmers to come together on recommending reduced rates. Otherwise, producers do not have the appropriate information to get the most out of herbicide applications and run into poor performance problems when indiscriminately using reduced rates.

Research reported in 1999 at MANDAK, indicated that overall using 100% and 66% of registered rates of Horizon plus Target in wheat and Roundup in Roundup Ready canola did not lead to reduced yields or increased weediness after three years of research at Brandon, Melfort, or Beaverlodge. However, low rate plots were weedier where fertilizer damage occurred to suppress wheat crops at Brandon in 1999 (Derksen et al. 1999).

In other research, the success of reducing Assert rates for wild oat control in barley was found to depend on barley variety. Virden and Lacombe were the most competitive varieties while Falcon and Candle were the least over 2 years and 4 site years of data (Feschuk and Derksen unpublished). Concerns about reduced rates leading to increased weediness were not borne out in this work as indicated by similar wild oat dockage in full and reduced rates (40-60%).

Conclusions

Reducing herbicide input costs is possible when done when using systems level weed management strategies. It is possible to save money, maintain yields, and not incur increased weediness. However, reducing herbicide usage, especially reducing rates, in a poorly diversified farm with high weed densities could be disastrous. Research is required to put together decision support systems, similar to those used in other parts of the world, that will provide producers with the information need to make the decisions on when reducing herbicide usage will pay. To do this, producers and industry need to come together to chart a future course that works for both sides of agribusiness.

References

Swanton, C.J., K.N. Harker, and R.L. Anderson. 1993. Crop losses due to weeds in Canada. Weed-technol. 7 (2) p. 537-542..