Maps and detection key to patch spray progress
Maps and detection key to patch spray progress
Spraying weeds only where
they occur makes obvious
sense. But putting patch
spraying theory into practice
is easier said than done.
Louise Impey asked
Silsoes Paul Miller how the
technology is progressing
WEED detection and mapping techniques are now the key to making patch spraying a widespread commercial reality, says a leading researcher.
Altering spray delivery across a field has been possible for some time, points out Silsoe Research Institutes Paul Miller.
But adoption of patch spraying as a regular technique for herbicide application will depend on the development of reliable weed detection methods and successful integration of all the component parts, he says.
"The priority is finding the best way to detect and map the weeds accurately. Being able to interpret the maps is also important, so that problem areas in fields receive the right treatment."
Work being done by Silsoe and IACR-Rothamsted is helping provide that knowledge. Weed species must be mapped at different times of the year, and an understanding of the factors which affect weed patches is also required.
"We have looked at automated weed detection methods, such as video cameras mounted on tractors, as well as the use of manual methods of mapping. It is perfectly possible for the farmer or an agronomist to map weeds from an ATV or tractor, but it is very time intensive."
Voice recognition weed mapping systems, tested in cereals during the past 18 months, seem preferable to touch-sensitive computer screens, says Prof Miller.
"It has produced good results. It can be operated from a vehicle, to make the process reasonably quick, and allows one person with a trained eye to map weeds."
Blackgrass and wild oats are the main mapping concerns, but cleavers have been added to the trial this year.
Automated detection in fully established cereal crops is almost impossible, but at early growth stages, or in row crops, the task is much easier. "Guidance devices have been used successfully in vegetable row crops. They can identify spacings and rows, so weed detection is possible.
"Growers are understandably nervous about only spraying patches of competitive weeds, so being confident of a machines ability to detect them is essential," he adds.
Cleavers can be mapped at most times of the year, but blackgrass should be done in the spring and wild oats during the summer, he says.
Once detected and a weed map produced, the challenge is to interpret it and compose a treatment map. Sprayer technology exists to vary the spray applied to a field but this does not yet allow a 24m boom to apply products at different rates at the same time. "That would require two GPS signals coming into the cab, which adds a complication factor as well as extra cost.
"The agronomy input is also important. We are still a long way from having a device fitted to the tractor which both detects the weeds and determinees the best course of action."
Potential cost savings and environmental benefits are driving the research and the farmer interest, but Prof Miller believes there are other advantages. "Being able to better target the use of pesticides should help to extend their life and give growers greater choice."
Record keeping is another potential benefit. "If the sprayer control system also records operations, this will help growers with their legislative responsibilities."
Remote sensing may also have a role in coming years but images of the field still require investigation, he says. *