Computer package to predict consequences
Computer package to predict consequences
SCIENTISTS at IACR-Rothamsted, ADAS, SAC and Silsoe Research Institute are working on a computer package to allow growers to assess weed control consequences with much greater certainty.
That will give them confidence to leave non-yield threatening, beneficial weed species, believes IACRs John Cussans.
"The maths of a single weed density to yield loss relationship is fairly simple. But in practice, with multiple species and seed return to consider, only a computer can handle the numbers."
But the DEFRA, HGCA and industry SAPPIO LINK-funded Weed Management Support System (WMSS) will not be available until 2005.
In the meantime growers best guide is to examine the yield loss (see table) and seed return data for individual species and balance that with the increasing amount of information on which weed species are beneficial environmentally.
"Then the idea is to try to pick a herbicide strategy that leaves some of the non-competitive, beneficial species. But at the moment it is quite difficult to do that because products are promoted on what they control, not on what they do not."
While yield loss estimates by weed species have been adjusted to err on the side of caution, growers should be wary of interpreting the 5% yield loss level as an absolute control threshold, he warns.
"For some of the most competitive weeds a sensible threshold density for control is approaching the limit of detection."
WEEDS IN WHEAT
Weed species Competition density for 5% yield loss
(Index) (plants/sq m)
Cleavers 3 1.7
Wild oats 1 5.0
Ryegrass 0.6 8.3
Blackgrass 0.4 12.5
OSR 0.4 12.5
Chickweed 0.2 25
Groundsel 0.06 83
Field Pansy 0.02 250
Parsley piert 0.02 250
Weed Competition Density
species index (plants/
sq m)
for 5%
yield loss
Cleavers 3 1.7
Wild oats 1 5.0
Ryegrass 0.6 8.3
Blackgrass 0.4 12.5
OSR 0.4 12.5
Chickweed 0.2 25
Groundsel 0.06 83
Field pansy 0.02 250
Parsley piert 0.02 250
IACR-ROTHAMSTED is staging a "Competition Competition" at Cereals 2002 to test growers and agronomists perceptions of the penalties from different weed species.
"We have a series of plots with known populations of weeds in them and will ask visitors to estimate the yield loss attributable to each," says researcher John Cussans.
Plots will be taken to yield, compared with clean control plots, and the competitor whose estimate is nearest the actual difference will be the winner.