Near infrared spectroscopy for world food security

The Food and Agriculture Organization predicts that 70% more food will be required to ensure food security by the year 2050 and that cereal production must increase from 2.1 to 3 billion tonnes per year. Timely, reliable and inexpensive data will be vital for farm managers to be able to maximise crop yields. Near infrared spectroscopy is now the analytical technology of first choice to obtain these data. Quantitative near infrared spectroscopy requires calibrations to be developed using reference data from traditional methods. Today these calibrations are available for a wide range of constituents in soils, plants and products and in turn they can be related to the yield and quality of food produced. Calibrations continue to be reported for a wider range of food stuffs and their constituents and we see near infrared being adopted both on-farm and in-factory across a variety of food sectors. To make a tangible contribution to food security, the data generated from these calibrations must facilitate crop management practices which maintain or promote yield. In this article, I present examples from my own experience to illustrate why it is important to understand the basis of a model developed from near infrared spectra and why errors due to sampling and poor reference values can negate the benefits from using near infrared spectroscopy.