8 March 1996

Research casts doubt over soil N test accuracy

SOIL N testing may not be as useful as some would have it, according to a comparison of the technique with standard recommendations.

Speaking at last weeks Ulster Arable Society conference at Greenmount College, near Belfast, Mike Carver, director of Arable Research Centres, produced the dramatic trial results shown below.

He stressed that samples had been taken from winter barley fields as recommended and sent to two companies offering a soil N testing service.

The advice that came back ranged from a recommended rate of 40kgN/ha to 175kgN/ha. By contrast traditional techniques suggested a uniform rate of 125kgN/ha (100units/acre) suited all four sites.

Applications were made in line with the advice and yields taken at harvest. When translated into gross margin terms the errors were compounded and huge financial penalties revealed.

Overwhelming losses

In one case soil testing gave a higher gross margin. But that £71/ha (£29/acre) uplift was overwhelmed by losses of £6-£345/ha (£2-£140/acre).

Mr Carvers message was clear – "be careful – this technology may not be sophisticated enough to use without other checks."


Soil N testing – advised rates and gross margins

ARCCo 1Co 2

Site(kgN/ha)(£/ha)(kgN/ha)(£/ha)(kgN/ha)(£/ha)

South 1125(877)75(871)40(790)

South 2125(761)135(608)94(416)

East125(867)60(845)85(938)

North125(795)175(795)85(681)

Average125(825)111(780)76(706)