Selective spraying will be an essential tool for prairie farmers in their struggle to keep increasingly difficult weeds in check.

Green-on-brown selective spraying that detects and sprays weeds during burn-down applications is already offered by a few companies, while green-on-green applications that target weeds in-crop is only a few years away from being available to Canadian farmers.

Once this technology is widely available, the door will open for new chemistries to be developed that were not feasible for blanket applications.

Daniel Ebersold, Head of Global Project House at BASF Digital Farming, said the company will likely customize future crop protection products to fit the practice of selective spraying.

“Starting with specific formulations, and in the long term also developing specific active ingredients, which, for example, we did not develop for cost reasons for a flat application,” Ebersold said.

“If you only apply 10 percent of the area, that changes completely the cost proposition of these new kinds of new molecules, which opens new spaces for new modes of action for farmers.”

BASF Digital Farming also develops machine learning algorithms through its subsidiary Xarvio, a digital ag platform it acquired from Bayer.

These computer programs can identify weeds in images of crops in real time, with onboard processors that do not require broadband connectivity.

BASF works closely with Bosch, which has developed sensing and spraying technology capable of targeting individual weeds with large field sprayers.

Ebersold said the partnership with Bosch will help BASF meet farmers’ agronomic needs, while reducing the environmental impact of large-scale grain production.

“How can we offer a solution for farmers without compromising his outcome? (A solution) that fulfills the farmers’ needs in terms of combating difficult weeds, resistant weeds, different weed pressures under different weather conditions in different crops, and at the same time reducing potential environmental impact drift problems, and showing the political environment, the society farmers are living in, a use of less but more effective chemical products?”

BASF and Bosch have been working together for years to develop their Smart Spraying product, and the companies recently received approval by merger control authorities to set up a new joint company called Bosch BASF Smart Farming (BBSF).

Smart Spraying combines Bosch’s camera-sensor technology, boom-mounted lights and spraying hardware with Xarvio’s machine-learning-based platform that offers automated and real-time weed identification.

Jens Koenig of Bosch said the system will typically be set to detect plants as small as 10-by-10 millimetres in North America, while in Europe it will be set to detect plants as small as six-by-six mm.

The boom-mounted cameras are set up to scan the field one metre in front of the boom in Europe, and a metre and a half in front of the boom in North America.

“We have seen that it’s useful to have different (image) resolutions from the ground. In Europe there are no GMO crops, or less GMO crops used, but in North America there are GMO crops,” Koenig said.

“You have different kinds of herbicides, different application strategies. Really high-resolution images are less useful (in North American crop production), but you want to drive faster, for example. So, we optimize the system for Northern America to the size of plants.”

There will also be a level of customizability of Smart Spraying.

“Depending on the weeds, if you have resistant weeds or weeds with a lot of seeds, which create a high seed bank for the next year, you would rather hit them all. Or if you have smaller weeds with low crop damage and the low multiplication potential, you could approach it differently,” Koenig said.

“If you want to optimize the performance or if you want to optimize savings and return on investment. So, it will be the farmer’s choice to select how we will help to set up the equipment.”

Users can change the sensitivity threshold by using Xarvio’s automated economic intelligence system that will change settings based on user inputs.

This is done by logging on to a field manager account though a mobile app to set the desired performance level for a particular field.

The Bosch hardware includes boom-mounted lights and cameras sensitive to red and infrared wavelengths that are used to calculate indexes, which, in turn, are used to identify living material.

The rows are also used, because the crop typically takes up most of the space within the rows.

The system also looks within the rows for plants with a different shape compared to the crop. Originally, this selective spraying approach was developed for row crops on 30-inch spacing or greater, but the company is improving the system to target weeds in much smaller row spacing, including cereal crops.

However, this is much more complicated because an extremely low false negative rate of weed detection is required so the crop isn’t inadvertently targeted.

“It’s easy from a computer and computing algorithm to say, OK, we achieve 80 percent, perhaps 90 percent of the right classification. But if you go over the different fields in the world from Europe, Canada, U.S., South America, they look different and it’s really hard to train (the algorithms),” Koenig said.

He said it will take millions of pictures to train the algorithm to be able to achieve an acceptable hit-rate threshold for broadacre crops with narrow row spacings.

For now, BBSF Smart Spraying will be rolled out for row crops, but the company does have several machines conducting trials and taking pictures of crops with narrow row spacings in several countries including the United States.

COVID-19 has challenged the expansion of this research and BBSF is not currently experimenting with its Smart Spraying system in Canada.

The Smart Spraying system does not require prior field images to target weeds because the sprayer can detect the weeds.

However, prior scouting is useful to know which weeds are present so that the appropriate products are used.

One of the prototypes BBSF is testing is an Amazone pull-type sprayer equipped with a Smart Spraying system.

It has two product tanks and two independent product streams that increases farmers’ options.

“The front tank, in this case of this pull-type sprayer, is used for spot spraying and the bigger tank is for the flat applications,” Koenig said.

Koenig said BBSF is targeting a two-to-four-year payback period for growers who get Smart Spraying installed on their sprayers, which will come from more efficient use of herbicides.

BBSF plans to introduce its Smart Spraying product in Brazil by the end of this year, and it’s also working with OEMs that are running proof of concept and field trials.

There is no clear date when the technology will be available in Canada.