Managing late N applications to meet wheat protein market requirements using pre-harvest near infrared (NIR) sensing (LK0927)

Abstract

Optimising late nitrogen (N) fertiliser application to achieve breadmaking wheat protein targets remains a problem for growers. Correct use of N fertilisers is imperative for financial and environmental reasons. Given the increasing costs of N fertilisers, there is an urgent need to improve the efficiency of their use. Studies have suggested that the most effective boost to protein comes from using foliar urea at the milky-ripe stage of grain development (Growth Stages 71-75). However, the current basis for late N decisions is inadequate, as only 1 in 4 of the crops benefit directly from extra N in attaining the 13% protein premium. Unfortunately, growers are unable to assess the N status of their crops in order to target fertiliser decisions.

This 4-year project aimed to develop a rapid method for on-farm measurements using new portable FT-NIR (Fourier transform near infrared) technology for optimising N input. In addition, the project sought to increase the scientific understanding of environmental and genetic factors that influence the deposition of proteins that govern end-use quality of wheat.

Wheat was sampled at around GS 70 and at harvest, from trials sown at Boxworth, Terrington, High Mowthorpe, Rosemaund and Essex from 2002 to 2005. The crops received no late N fertiliser and included nabim Group 1- 4 varieties, featuring a range of protein contents. The Matrix-I NIR instrument (Bruker Optics) was evaluated for measuring coarsely chopped immature ear and whole plant material. The need for sample drying and, in particular, the effect of microwave drying was investigated. Reference moisture and protein contents (Kjeldahl method) were determined. Protocols were developed for field sampling, as well as for presenting samples for NIR measurements. NIR calibrations were developed and validated for the different forms of wheat material.

Results showed that:

• Using the Matrix-I NIR instrument it was possible to make rapid determinations of protein and moisture contents of fresh (undried), developing ear and whole plant material.
• The most accurate prediction of protein content was generally achieved from the use of ears rather than whole plant material.
• The most accurate prediction of protein (but not moisture content) was generally achieved from the use of dried material.

It was concluded that such NIR assessments could form the basis of an integrated crop monitoring system enabling accurate decision making for the targeted application of late N for boosting grain protein. NIR forecasting of grain protein content at harvest may be improved by using models taking account of the state of grain development and the likely crop yield. This could improve the farmer's ability to meet high quality wheat targets consistently, and improve the marketability and sustainability of the UK arable sector.