Optimising fertiliser nitrogen for modern wheat and barley crops

Abstract

In each of the three harvest years 2005, 2006 and 2007 ten N response trials were conducted on winter wheat and five on spring barley. Trials were distributed from Kent to Aberdeenshire; each one tested two 'old' (new in the 1980s) and two 'new' (from 2000s) varieties at six N rates from nil to 166% of the amount recommended. For the determination of N optima, grain yields were related to N applied by fitting a linear plus exponential function for each variety and grain N% was described by the better fit of a 'normal with depletion' or a linear function. Mean grain yields with optimum N were 8.69 and 9.98 t/ha for old and new wheat varieties respectively, and they were 5.03 and 5.90 t/ha for old and new barley varieties. In each year and for each species 20% of the trials showed nil or a very small response to fertiliser N. At the other sites differences in N optima of wheat varieties related to their differences in grain yield, the slope being about +20 kg fertiliser N per tonne grain. However, the same did not apply to spring barley varieties for which better N Utilisation Efficiency completely compensated for their better grain yield. The yield response curves indicate that N optima decrease for each point increase in the break-even ratio (the price ratio of fertiliser N to grain) by 11 kg/ha for wheat and by 8 kg/ha for barley. Mean grain N (% DM) with optimum fertiliser N (break-even ratio 6:1) was the same for old and new wheat varieties at 1.98%, whilst it was 2.09% and 1.96% for old and new barley varieties, respectively.

The 45 trials provided a test of current recommendations. To maximise average profit from feed grain production, current recommendations (RB209 7th edition published in 2000) had to be increased by 18 kg/ha N for modern wheat varieties and by at least 40 kg/ha N for modern barley varieties. The Field Assessment Method (FAM) used in the recommendations to predict soil N supply (SNS; 'true' values being estimated from grain N at harvest with nil N applied) did not perform satisfactorily either at wheat or barley sites, and recommendations based on the FAM gave no more average profit than use of a fixed N amount at all sites (185 kg/ha N for wheat and 162 kg/ha N for barley). Soil mineral N to 90 cm depth, corrected for over-winter leaching (SMN) was reasonably well related to soil N supply at the wheat sites (R2 = 0.52), with SNS showing equivalence (at least) with estimated SMN amounts, and recommendations based on SMN improved average profit by the value of 15 kg/ha N at the wheat sites. SMN did not relate to SNS at the barley sites, and recommendations based on SMN only improved average profit by the value of 6 kg/ha N.

Results were inconclusive on whether early N applications improved alcohol production from wheat, because the test year (2007) provided inappropriate (dry spring and low-yielding) conditions.

The results have been used to inform the concurrent revision of RB209.