25 February 2000


Yield mapping is only the start to a precision approach

to agronomy claims a pioneering Lincs grower.

Andrew Swallow reports

SIX years of precision farming has boosted yields and seen seed, nutrient and spray costs tumble on one Lincs farm.

"We have cut input costs by 40% in the past five years," says Robert Pask of Pask Farms, Grantham. In the same period the five-year average yield has risen 12%.

Cheaper buying prices have helped, he admits. But better input targeting through precision farming techniques has added to savings and raised yields.

Mapping yields has been the keystone, he maintains. Data now goes back to harvest 1994 when the Massey 40RS combine first mapped the farms crops.

"Watching the yield monitors we could see there was massive variation. But yield maps show variation is beyond what you could believe, as low as 4t/ha to 12.5-13t/ha in the same field.

"When we looked at it from a gross margin point of view we got really excited. In the same field we were making serious amounts of money in some parts and making a loss on others."

The yield maps prompted questions, resulting in improved agronomy. Solutions on some poor yielding areas have been as simple as moling and sub-soiling.

Nutrient mapping followed by variable rate application of phosphate and potash has been thoroughly tested. But after three seasons of GPS sampling Mr Pask has concluded current 100m grid sampling techniques are too inaccurate.

"Looking back we never once had a correlation between nutrient maps and yield maps. A 100m grid masks one hell of a lot of variation."

Now P and K are applied at variable rates according to the previous seasons crop off-take. "So where we had a 10t/ha crop, we apply fertiliser for a 10t crop, but where we only had 4t/ha we only apply for 4t," he says.

The instructions for the variable applications are loaded on to a PCM card by Mr Pask in the farm office. That is then transferred to the FieldStar system in the tractor cab, which automatically varies the spreading rate according to the tractors position in the field.

"I can load up to 14 different jobs on to the smart card, which keeps the driver going for at least a day," he says.

Seed rates are now varied in a similar way, but the investment in GPS automation was only made after experimenting with manual rate changes. Standard seed-rates were raised 10% on patches of heavy soil, and cut 10% on pockets of light land by manual switches in the cab. In extreme cases, at the operators discretion, a 20% adjustment might be made, on a compacted headland for example, he says.

More even plant stands resulted and Mr Pask has been varying seed rate by GPS for two seasons now.

"I can pre-determine where the seed-rate changes across the field beforehand, in the office, which takes the onus off the operator. But he can still over-ride the pre-set rates with, say, an extra 10% of seed if he sees fit. We are still learning."

Hydro did trials with the N-sensor system at Heydour Lodge in 1998 and Mr Pask had one fitted to his own tractor last spring. The sensor reads the crop colour and applies a fertiliser rate accordingly, based on a Hydro calibration.

After a standard 40kg/ha (32 units/acre) nitrogen application to wheat in seed-rate trials, variable rate nitrogen applications were made at GS32 and GS39. On the low seed-rate plot Mr Pask believes the GS32 application was too low.

"I think they got that calibration wrong, but it is the future without a doubt. A variable rate application as the second nitrogen on our winter barley improved margins by £20/ha," he says.

Total nitrogen used across the barley field was nearly the same as the farms standard 140kg/ha. "But there was a massive variation from the shallow soil areas to the heavy clays. The sensor applied a lot more nitrogen where we get higher yields. That has got to have an influence on quality, it should be more uniform," he says.

Milling wheat samples could also be more consistent following variable rate applications, he reckons.

Being able to offer yield mapping with combining and variable rate application with fertiliser commands an extra £5-£7/ha for contract work, he adds.

Variable rate spray application will be the next step at Heydour, but as yet the technology is not commercially available, he says. Variable rate fungicide applications could be made using real-time sensing for disease pressure but herbicides are more likely to be applied according to GPS weed maps.

"There is big potential to save a lot of money patch spraying herbicides. We spray whole fields, which is just not necessary. If you do not need to apply it, why apply it?" he asks. &#42


&#8226 370ha in three blocks.

&#8226 Soils: limestone brash, heavy blue clay, heavy clay loam.

&#8226 Crops: W wht, w bly, w OSR, s bly, s beet, s beans.

&#8226 Two full-time staff.

&#8226 Three Tractors, 410hp total.

&#8226 Massey 40RS combine with auto-levelling & contour flow.

&#8226 GPS yield mapping, N-spreading and drilling.

Modest Investment

So far Mr Pasks spend on precision farming technology is modest. "The biggest investment was on the combine – £7000 for the FieldStar unit. But it is transferable to the tractors."

A controller each for the drill and the fertiliser spreader adds £4600, bringing the total to just £11,600. "Besides the benefits on our own farm, we have picked up a lot of contract work we would never have done otherwise," he says.


&#8226 Yield mapping core element.

&#8226 Simple solutions first.

&#8226 Variable N great potential.

&#8226 Tuning variable seed-rate use.

&#8226 P & K mapping discontinued.

&#8226 Investment: £11,300, four-year write-off.

Precision farming enthusiast Robert Pask enters seed rate instructions on to the computer in the office (1) and a smartcard transfers the data to the FieldStar GPSsystem in the tractor (2). A linear motor on the side of the Amazon combination drill adjusts seed rate as the drill moves across the field, to give a more even plant count despite changing soil types (3). Here Ragland series clay (4) had 20% more seed than the limestone brash (5).

See more