14 March 1996

Target variable areas for best precision

All thats new in the fast moving world of variable input applications was on display at Precision Farming 97, the international event staged by farmers weekly and Fusion Events at the East of England Showground, Peterborough. Charles Abel reports. See pages 82/83 for Machinery.

DONT let precision farming service providers dictate the way you use the technology. That was a key message to emerge from Precision Farming 97.

"You know your fields and you have a good idea where variability is likely to exist," said ADAS consultant David Parish. Instead of getting the whole field mapped on a 1ha grid, target variable areas with more intensive, "intelligent" sampling, he urged.

Results from a detailed study last year underpin his comments. Those showed a 1ha sampling grid did not identify all the variation in the field being studied. Sampling on a 25m grid highlighted significant variations between the sampling points of the 1ha grid, which the smoothed images produced by mapping software could not detect.

"That gave us the evidence to question whether the current 1ha grid needs looking at more closely. There really was quite significant variation between the 1ha sampling points."

This year ADAS will evaluate an "intelligent" sampling approach on its new arable demonstration farm in Bedfordshire. Tests will be made where variation is most expected, possibly going down to sampling every 25m. Where variation is not expected sampling could drop to one per 5ha.

Dick Godwin of Cranfield University supported the idea. Yield maps could help target sampling, which can then check for variations in nutrient supply, soil condition and moisture retention where results are poor, he suggested.

"Knowing your soil will be one of the most useful things in this technology. Getting the JCB out to dig holes after harvest could be one of the most useful things you could do," he said.

All thats new in the fast moving world of variable input applications was on display at Precision Farming 97, the international event staged by farmers weekly and Fusion Events at the East of England Showground, Peterborough. Charles Abel reports. See pages 82/83 for Machinery.

DONT let precision farming service providers dictate the way you use the technology. That was a key message to emerge from Precision Farming 97.

"You know your fields and you have a good idea where variability is likely to exist," said ADAS consultant David Parish. Instead of getting the whole field mapped on a 1ha grid, target variable areas with more intensive, "intelligent" sampling, he urged.

Results from a detailed study last year underpin his comments. Those showed a 1ha sampling grid did not identify all the variation in the field being studied. Sampling on a 25m grid highlighted significant variations between the sampling points of the 1ha grid, which the smoothed images produced by mapping software could not detect.

"That gave us the evidence to question whether the current 1ha grid needs looking at more closely. There really was quite significant variation between the 1ha sampling points."

This year ADAS will evaluate an "intelligent" sampling approach on its new arable demonstration farm in Bedfordshire. Tests will be made where variation is most expected, possibly going down to sampling every 25m. Where variation is not expected sampling could drop to one per 5ha.

Dick Godwin of Cranfield University supported the idea. Yield maps could help target sampling, which can then check for variations in nutrient supply, soil condition and moisture retention where results are poor, he suggested.

"Knowing your soil will be one of the most useful things in this technology. Getting the JCB out to dig holes after harvest could be one of the most useful things you could do," he said.


INTELLIGENTSAMPLING


&#8226 Target sampling towards areas where variation is suspected.

&#8226 Farm knowledge, yield maps, aerial and satellite images can help locate these.

&#8226 Less than 1ha grid may be appropriate.