The maths of black-grass control

6 Aug 2015

This is my 150th blog and perhaps too many have been on black-grass, the nation’s most talked about weed. You have to believe me when I say that I was trying to avoid writing about this weed yet again but recent press results have driven me back to the keyboard.

The issue that I would like to raise is the too simplistic conclusions made from field data. These can over-estimate or under-estimate the role of cultural or chemical control of the weed. Good field data is not easily achieved and it is the industry’s responsibility to use it intelligently.

I have written before about the huge fuss made a few years ago about increased crop competition from wheat being able to reduce black-grass seed heads by half. Many took this as the answer to the black-grass problem but the maths of the weed’s dynamics told us differently. Models suggest we have to chemically control 97% of black-grass plants emerging in a continuous wheat crop grown in ‘normal circumstances’ in order to contain populations at current levels. This means that we can let only 3 out of 100 plants make it to seed shedding. Crop competition does not reduce black-grass plant numbers but reduces the ability of the plant to set viable tillers. Hence, reducing seed heads per plant by half through crop competition means that we can allow 6 rather than 3 out of 100 plants to make it to seed shedding. So a 50% reduction in seeding per plant reduces the need for chemical control to 94% rather than 97% control. Every reduction in the reliance on chemical control is useful but this is not the game-changer that was originally claimed. Obviously, greater reductions in seed return through competition would reduce the need for chemical control by a greater percentage. However, more than a 90% reduction in black-grass seed set from increased competition will be required to get down to a chemical control requirement of 70% of plants; a level that is now often achieved by pre- and/or early post-emergence herbicides.

This is an example of when a large percentage figure can over-estimate the value of a control technique. In contrast, recent quotes in press reports seem to be under-estimating the benefit of stacking herbicide products to control the weed. Some say that using more than two products in a mix adds little to the percentage control of plants. These statements can undermine the value of using mixtures of three or more products.

The best way to explain what I mean is to give an example. Let us again assume that there are potentially 100 black-grass plants/m2 that will emerge in a winter wheat crop. A two-way mix applied pre-emergence may control 60 of these plants (i.e. 60% control) and a three-way mix may control 70%. This is ‘only’ a 10% increase at face value. This does not sound a lot and some might say it would be better to apply this third component early post-emergence where potentially it may give a higher level of control. The maths of this situation are illuminating.

In this example, the two-way mix will reduce the numbers in the crop from 100 plants/m2 to 40/m2 and adding the third component of the mix will reduce the numbers from 40/m2 to 30/m2. Reducing survivor numbers from 40 to 30 is in fact 25% control, which is a lot higher than the headline figure of an additional 10% control. Put another way: using a two-way mix pre-emergence would let 40 plants/m2 survive and achieving 25% control from an early post-emergence would provide the same result of a final plant number count of 30 plants/m2 (more than 29 too many!).

I realise that this is a rather simplistic view and herbicide resistance will ‘nuance’ the situation. We know little about the impact a particular pre-emergence application will have on the susceptibility of weeds to a particular early post-emergence application. Are the percent controls of plants achieved by individual products simply additive or does an earlier application pre-dispose its survivors to be more or less difficult to control with an early-post emergence application?

What I am saying is that data need to be carefully interpreted. In practice, it may be that stacks of more than two products applied pre-emergence are particularly relevant to later sown winter wheat crops provided the seedbed and the weather are conducive to good control. Simpler mixes may be more suited to pre-emergence or peri-emergence applications to earlier sown winter wheat crops, particularly when the weather and/or seedbed are not suited to good control, with an expectation that a further application of a mix could well be necessary early post-emergence.