Article taken from Agronomist & Arable Farming, May 2018.
Drone technology has been evolving rapidly in recent years, to the extent that even relatively modestly priced machines can now deliver crystal clear images of crops, highlighting in-field variation to a degree that would have been impossible a couple of seasons ago.
""By flying crops several times a season, a range of key agronomic parameters can be accurately assessed, including establishment, plant counts, biomass, weeds and disease presence," says Agrovista's head of precision technology Lewis McKerrow.
Simple snapshots can be useful, highlighting areas suffering from pests, poor drainage or nutrient deficiency.
However, in the right hands, the technology offers much more. Using powerful MapIT Pro software, Agrovista can quickly process the masses of data collected by a drone's high-resolution camera and accurately assess the in-field variation to create field zone maps.
""When overlaid with other data such as soil, soil nutrient and yield maps, zone maps can be used as a basis for variable rate drilling, nitrogen application and patch spraying of weeds," says Mr McKerrow.
Agrovista's commercial drone service offers flight packages carried out by the company's approved operators for less than £5/ha depending on area, or a processing service ' ' Zone for farmers who have their own drones for about £1/ha.
The company's fleet of DJI Inspire quadcopters can fly for 15-20 minutes per battery, covering 40-50ha in still conditions. At least l00ha/hour is easily achievable in practice allowing for battery changes, and about two-thirds of that in more breezy conditions.
These machines are fitted with a 12-megapixel camera that delivers 3.5cm resolution at 120m, and down to 1cm at 40-50m. A typical cropscanning flight will produce l-2GB of data. A couple of hours of flying would create more data than the average office computer could handle, which is where Agrovista's processing service comes in.
In 2017, Agrovista made more than 500 drone flights covering 7,000ha. Since then, the quadcopter fleet has been expanded and several more of the company's agronomists have undergone the training required for commercial flights.
Drone service offers increased precision
Aberdeenshire grower Andrew Booth has several years' experience with precision farming. He sees drones as a natural progression to improve the quality of the data on which maps and, ultimately, agronomic decisions are based.
The heavy land operation at Savock Farms, near Ellon, covers l,400ha of owned, rented and contract-farmed land, growing wheat, malting barley, oilseed rape, rye, triticale, oats and grass. About 60% of the land grows anaerobic feedstock.
""We have been soil mapping for at least 12 years and yield mapping for 10, so we have built up a good historical record," says Mr Booth.
Variable rate fertiliser application (N, P and K) started in 2012, based on soil and yield maps and tractor-mounted N-sensing equipment.
""We needed to increase the precision of our nutrient applications. Since we started using Agrovista's drone service in 2016, we've been able to tailor N rates before going into the field, rather than finding out afterwards what the sensor did.
""Using pre-loaded prescription maps derived from drone biomass images, we cut back N rates on areas with poor yield potential, whereas the sensor would have applied higher rates.""
Mr Booth is also using the technology to vary seed rates. The heavy soils vary in their workability, so optimum plant populations differ across fields.
He began variable rate sowing in 2016 on crops of wheat. Seed rates were determined using soil maps, his own knowledge and Google Maps taken in the dry spring of 2012/13, which highlighted variation in plant establishment. This resulted in three to four zones for each field.
The drone was used to check the success of this approach. Flights showed that the target seed rate did not always match the zones and plant populations tended to be too high.
This season the process has been refined by using last year's establishment drone images to help drive this year's seed rate decisions understand what is going on in the soil and how the crop is establishing and growing in that soil," says Mr McKerrow.
""All this information, together with eventual crop yields, gives us a great basis to tailor our agronomic advice to generate the most costeffective return from each field.""
Mr Booth has also used the service to check field boundaries, pinpoint pest damage and to highlight potential drainage problems. This season, he is also using it to target nutrient applications on his grass silage, which will go to the local AD plant.
Mr Booth believes the service offers good value for money. Information is turned around quickly, with maps available the morning after fields were flown.
""The information is available on Axis/ MapIT Pro, which is very user-friendly. It is easy to use in the field, as the maps are simply programmed into the kit.
""We have received excellent support from the Agrovista team, which gives me confidence to increase the area on which we use the drone service.""
Satellite vs drone
With more satellites in orbit enabling more regular image capture, the chances of getting a good cloud-free image when required are improving. That, coupled with more and more providers offering low-cost satellite images, poses the question: Why the need for drones?
""Satellite data can be a very accurate and low-cost tool for a number of uses, such as variable N. However, it is very difficult to sense-check image quality without a close inspection of the field," says Mr McKerrow.
With high-resolution drone visuals, anomalies such as shadows from trees or standing water can be identified by looking at the visual image.
The satellite visual resembles a blurry photograph so it is almost impossible to pick out any features.
Normalised difference vegetation index (NOVI) images, which use infra-red to analyse green crop biomass, do show similar trends however, so the satellite image could suffice if a simple variability map was required.
""However, the satellite NOVI image shows a red strip along the north side of the field that isn't an accurate reflection of what the crop is actually like in that area. But there is no way of checking this without a field visit, whereas with the drone you can clearly see every inch of the field from the visual image," says Mr McKerrow.
""In this example the higher definition of the drone image allows you to easily identify field drains which are not possible to see from the satellite.""
Drone technology has been evolving rapidly in recent years, to the extent that even relatively modestly priced machines can now deliver crystal clear images of crops, highlighting in-field variation to a degree that would have been impossible a couple of seasons ago.
""By flying crops several times a season, a range of key agronomic parameters can be accurately assessed, including establishment, plant counts, biomass, weeds and disease presence," says Agrovista's head of precision technology Lewis McKerrow.
Simple snapshots can be useful, highlighting areas suffering from pests, poor drainage or nutrient deficiency.
However, in the right hands, the technology offers much more. Using powerful MapIT Pro software, Agrovista can quickly process the masses of data collected by a drone's high-resolution camera and accurately assess the in-field variation to create field zone maps.
""When overlaid with other data such as soil, soil nutrient and yield maps, zone maps can be used as a basis for variable rate drilling, nitrogen application and patch spraying of weeds," says Mr McKerrow.
Agrovista's commercial drone service offers flight packages carried out by the company's approved operators for less than £5/ha depending on area, or a processing service ' ' Zone for farmers who have their own drones for about £1/ha.
The company's fleet of DJI Inspire quadcopters can fly for 15-20 minutes per battery, covering 40-50ha in still conditions. At least l00ha/hour is easily achievable in practice allowing for battery changes, and about two-thirds of that in more breezy conditions.
These machines are fitted with a 12-megapixel camera that delivers 3.5cm resolution at 120m, and down to 1cm at 40-50m. A typical cropscanning flight will produce l-2GB of data. A couple of hours of flying would create more data than the average office computer could handle, which is where Agrovista's processing service comes in.
In 2017, Agrovista made more than 500 drone flights covering 7,000ha. Since then, the quadcopter fleet has been expanded and several more of the company's agronomists have undergone the training required for commercial flights.
Drone service offers increased precision
Aberdeenshire grower Andrew Booth has several years' experience with precision farming. He sees drones as a natural progression to improve the quality of the data on which maps and, ultimately, agronomic decisions are based.
The heavy land operation at Savock Farms, near Ellon, covers l,400ha of owned, rented and contract-farmed land, growing wheat, malting barley, oilseed rape, rye, triticale, oats and grass. About 60% of the land grows anaerobic feedstock.
""We have been soil mapping for at least 12 years and yield mapping for 10, so we have built up a good historical record," says Mr Booth.
Variable rate fertiliser application (N, P and K) started in 2012, based on soil and yield maps and tractor-mounted N-sensing equipment.
""We needed to increase the precision of our nutrient applications. Since we started using Agrovista's drone service in 2016, we've been able to tailor N rates before going into the field, rather than finding out afterwards what the sensor did.
""Using pre-loaded prescription maps derived from drone biomass images, we cut back N rates on areas with poor yield potential, whereas the sensor would have applied higher rates.""
Mr Booth is also using the technology to vary seed rates. The heavy soils vary in their workability, so optimum plant populations differ across fields.
He began variable rate sowing in 2016 on crops of wheat. Seed rates were determined using soil maps, his own knowledge and Google Maps taken in the dry spring of 2012/13, which highlighted variation in plant establishment. This resulted in three to four zones for each field.
The drone was used to check the success of this approach. Flights showed that the target seed rate did not always match the zones and plant populations tended to be too high.
This season the process has been refined by using last year's establishment drone images to help drive this year's seed rate decisions understand what is going on in the soil and how the crop is establishing and growing in that soil," says Mr McKerrow.
""All this information, together with eventual crop yields, gives us a great basis to tailor our agronomic advice to generate the most costeffective return from each field.""
Mr Booth has also used the service to check field boundaries, pinpoint pest damage and to highlight potential drainage problems. This season, he is also using it to target nutrient applications on his grass silage, which will go to the local AD plant.
Mr Booth believes the service offers good value for money. Information is turned around quickly, with maps available the morning after fields were flown.
""The information is available on Axis/ MapIT Pro, which is very user-friendly. It is easy to use in the field, as the maps are simply programmed into the kit.
""We have received excellent support from the Agrovista team, which gives me confidence to increase the area on which we use the drone service.""
Satellite vs drone
With more satellites in orbit enabling more regular image capture, the chances of getting a good cloud-free image when required are improving. That, coupled with more and more providers offering low-cost satellite images, poses the question: Why the need for drones?
""Satellite data can be a very accurate and low-cost tool for a number of uses, such as variable N. However, it is very difficult to sense-check image quality without a close inspection of the field," says Mr McKerrow.
With high-resolution drone visuals, anomalies such as shadows from trees or standing water can be identified by looking at the visual image.
The satellite visual resembles a blurry photograph so it is almost impossible to pick out any features.
Normalised difference vegetation index (NOVI) images, which use infra-red to analyse green crop biomass, do show similar trends however, so the satellite image could suffice if a simple variability map was required.
""However, the satellite NOVI image shows a red strip along the north side of the field that isn't an accurate reflection of what the crop is actually like in that area. But there is no way of checking this without a field visit, whereas with the drone you can clearly see every inch of the field from the visual image," says Mr McKerrow.
""In this example the higher definition of the drone image allows you to easily identify field drains which are not possible to see from the satellite.""