Since 2017, precision agriculture enthusiast Mark Branson has been hunting herbicide resistant weeds using a DJI Phantom drone.
Flying at a height of 75 to 90 m across a 46 ha paddock, the drone takes 500 images that are later ‘stitched together’ to give one image of the field.
“It is as simple as ‘where you can see the rows, there’s no weeds’ and ‘where you can’t see defined rows, that’s a weedy patch’,” says Mark. “When you add an NRG filter you can clearly see the extra biomass that results from weed growth.”
Mark (right) and Sam Branson use their drone to identify and map weedy areas in crop on their mixed farming operation at Stockport, SA. (Photo: Vanessa Binks)
Once patches are mapped and ground-truthed to check the species present, Mark then decides on the most appropriate course of action. In the last three years he has cut five patches of about 10 ha each for hay using a cutter bar mounted on the SP sprayer.
The main resistant weeds on the farm are annual ryegrass, wild oats and wild radish. Although it is easy to see the wild radish present in wheat crops the grasses are hard to see from the ground, and it is difficult to accurately map weedy patches.
The idea of using a drone to scout for weeds was sparked in 2015 when Mark’s son Sam bought a drone with a GoPro camera attached to gather video footage. Mark could see the possible application in precision agriculture and in 2016 he bought DGI Phantom 4 to look down into crops. One of the first ‘jobs’ the drone did was to identify an area of poor urea application in a crop.
“The real advantage in identifying the mistake in-crop was that we had the opportunity to take remedial action and fix the problem with an application of liquid nitrogen in the affected area,” says Mark. “Without the drone that mistake would have gone un-noticed and we would have suffered a yield loss in that portion of the paddock.”
When it comes to hunting for resistant weeds, Mark first traces the boundary of the paddock in the drone software package and then, with a set flying height and number of images to collect, the drone takes off and takes the required set on images.
The advantages of the drone imagery over satellite images include being able to collect and use images collected on cloudy days and having full control over the timing of the data collection. The drone is also better suited to the job than the tractor-mounted biomass sensors that can only collect data from a 40 m swathe at a time.
For the drone image capture to be successful it is necessary to have stable light conditions.
“It is fine to fly on both clear and cloudy days, provided the conditions don’t change for the duration of the 20 min flight time,” says Mark. “It is best to fly between 9.30 am and 4 pm to reduce the impact of shadows and I generally prefer cloudy days.”
One of the few advantages of the tractor-mounted biomass sensors over the drone is the ability to collect data at night.
Mark spent a few thousand dollars on his drone but as prices continue to drop, he says a suitable drone can be bought for $1500 or less. He uses a third-party provider, Drone Deploy, to process the images collected into the one seamless image for an annual cost of US990.
“The Drone Deploy subscription is expensive but the quality of the image stitching is excellent and the turn-around is fast,” he says. “It only takes 3 or 4 hours for a set of images to be stitched and returned.”
Using an NRG filter it is possible to clearly see the extra biomass that results from weed growth in-crop.
Mark has three batteries for the drone, each having a flight time of 20 minutes, which means he can easily fly any of his paddocks in a day. He advises any new pilots to be conscious of no-fly zones and to comply with licencing requirements.
“The drone has certainly proved its worth as a good tool for scouting and for identifying and mapping suspicious areas that need investigation and diagnosis,” he says. “I am also working to find an index that will reliably show frost damage.”
Mark and his son Sam crop 1200 ha of wheat, barley, peas, beans, canola, lentils, oats. In November 2015 the catastrophic Pinery fire severely impacted the Branson’s property, destroying crops, machinery and livestock, and they were very fortunate that the fire leapt over their home, leaving it unscathed. Prior to the fire the Bransons ran a self-replacing merino flock of 1000 ewes, three-quarters of which were lost in the fire. They have now built back up to 850 ewes.
Mark was a founding committee member of the Society of Precision Agriculture Australia (SPAA) in 2002 and completed a Nuffield Scholarship in 2005, looking into the economic and environmental impacts of precision agriculture and conservation farming in New Zealand, USA, Canada, UK and France.
“I started marrying up wheeltrack widths in 2002, and then bought a tractor with an RTK steer kit in 2004,” says Mark. “The Nuffield Scholarship confirmed a lot of existing knowledge and understanding, and gave me new direction and confidence to apply precision agriculture to our farm operation. Since then we have used variable rate technology particularly for nutrient management – gypsum, lime, N and P – and seeding rate.”
The Stockport area has suffered two very dry seasons, receiving just 3 per cent of their average growing season rainfall 2018, a less than 1 percentile year. Conditions were very similar in 2019.
Big 6 in action on Branson Farms
The Bransons run a winter cropping rotation and sheep are an important part of weed control program, with the pasture phase being key to driving down annual ryegrass numbers. They have implemented a wide range of tactics from the WeedSmart Big 6.
1. Rotate crops and pastures
The Bransons run a winter cropping rotation and sheep are an important part of weed control program, with the pasture phase being key to driving down annual ryegrass numbers. The crop sequence Mark uses is pasture, canola, wheat, wheat, barley, grain legume, hard or durum wheat (high protein), wheat, [or canola if weeds are building], wheat, wheat, barley, grain legume and back to pasture. The pasture phase is a minimum of two years and preferably three years to ensure all the ryegrass has germinated before returning to cropping.
Within the cropping phase Mark includes a cereal with resistance to cereal cyst nematode (CCN) and no longer uses imi chemistry or imi-tolerant crops due to resistance in ryegrass. TT canola is still a useful strategy for ryegrass control.
The pasture or ley phase is based on cereals mixed with either medic (on alkaline soils) or sub-clover (on acidic soils). Mark’s father used the same system, based on soil colour, but now Mark has the two mixes in the seeder box ready for VR application according to the PA maps from pH surveys.
Mark also uses the sheep to graze crop stubble at high density for a short time after harvest. The sheep benefit from the stubble and graze any weeds present. Although their hooves compact the surface of the soil the layer is easily broken up at seeding.
2. Mix and rotate herbicides
Mark is very aware of herbicide resistance and the need to rotate between mode of action groups. He does not routinely conduct herbicide resistance testing but he is careful to observe any changes in herbicide efficacy. For example, he witnessed one wild oats plant escape Topik (Group A herbicide) and how rapidly this escalated to a patch of herbicide resistant weeds.
He looks for ways to use a range of herbicide options across the cropping phase to control wild radish. Likewise, there are several pre-emergent herbicides used such as simazine in beans, triazine in canola and Sakura in cereals, keeping in mind that the triazine and Sakura need to be applied in seasons where there the likelihood of them being washed off the stubble and into the soil is high.
3. Crop competition
Mark chooses competitive wheats, such as Saintly, and follow them with Scope barley, which is a vigourous variety that provides good early competition. Although he prioritises competitive attributes over extra yield he says that all the varieties are quite good for yield.
The Bransons have been direct drilling since 2002 and currently use a tyned precision seeder that places the seed one inch apart in a parallelogram configuration. The tynes have individual depth control and presswheels.
Mark currently sows crops in 10 inch rows but he would prefer 6 to 9 inch row spacing, which is not practical with his current tyne seeder. He will probably move to a disc seeder to reduce soil disturbance when he is ready to upgrade.
The seeder allows inter-row sowing, which Mark says is the key to stubble management. Currently the rows are 10 inches apart but Mark would prefer 6 to 9 inch row spacing, which is not practical with his current seeder. He will probably move to a disc seeder to reduce soil disturbance when he is ready to upgrade.
Mark uses VRT seeding and nitrogen application for canopy management to ensure enough moisture is available at the end of the season for the crops to finish strongly.
4. Stop seed set
Mark uses both mechanical and herbicide tactics to stop escape weeds setting seed. In the pasture phase, Mark slashes and grazes to prevent seed set and pasture tops with paraquat.
He uses crop topping in pulses and in triazine tolerant canola he sprays glyphosate under the cutter bar, then harvests the windrows and drops crop residue into a narrow windrow for burning. This tactic in canola is an excellent way to really drive down ryegrass numbers in a single year. Mark implements this in the first year out of pasture and also within the cropping phase to extend the length of the cropping phase out to 10 to 12 years. Prior to getting weed numbers down, the cropping phase was only 6 years.
He finds that hay is a good way to deal with weedy patches. So far, he has successfully used hay making to deal with two patches of herbicide resistant weeds identified using the drone. He says wild oats soon becomes a long-term project if it gets out of hand so he’d much rather accept a short-term sacrifice.
5. Double knock
If there is an early break to the season Mark usually implements the double knock pre-seeding.
After harvest, the sheep can also provide the first knock on weeds, which are then sprayed out by 5-leaf stage. He doesn’t usually do a second spray for summer weeds but finds that the sheep do a good job of cleaning up any weeds that survive any herbicide treatment.
6. Harvest Weed Seed Control
Mark has seen the benefits of narrow windrow burning in canola and plans to have an impact mill within the next five years. He hopes that having an impact mill on the harvester for all crops will take the pressure off some of the other practices currently used.