Spectral statistics study

A statistical analysis of visible and near-infrared (vis-NIR) spectroscopy has demonstrated the feasibility of quantifying oil, protein and total glucosinolate content of three cultivars of rapeseed.

Rapeseed (Brassica napus), also known as rape, oilseed rape, rapa, rappi, and in the common North American commercial cultivar, Canola, is a bright yellow flowering member of the family Brassicaceae (mustard or cabbage family). Rapeseed is an increasingly important crop plant in the production of animal feed, vegetable oil for human consumption, and biodiesel. It is grown extensively across the European Union, North America, Australia, China, and India.

Rapeseed is thus an important crop not least because it is very high lipid and medium protein content. Nutritional profiling of the crop is significant to its use in human and animal food production and as an alternative energy source. Indeed, it has several advantages over petroleum-derived fuels in that it is non-toxic, potentially renewable, and leads to lower emissions of polluting gases. Within the EU, targets have been set to see the blending of 20% of biofuel in general vehicle fuels until 2020. Rapeseed is also being used in other non-food applications such as in lubricant, surfactant, paint, ink, polymer, and pharmaceutical production.

C. Petisco, B. García-Criado, B.R. Vázquez-de-Aldana, and A. García-Ciudad of the Institute of Natural Research and Agrobiology of Salamanca, IRNASA-CSIC, working with A. de Haro of the Institute of Sustainable Agriculture, IAS-CSIC, in Córdoba, Spain, have demonstrated how to save time in evaluating intact rape seeds.

The researchers have turned to non-destructive and rapid vis-NIR spectroscopy in order to quickly quantify oil, protein and total glucosinolates in different strains of rapeseed and a cultivar of Ethiopian mustard (Brassica carinata A. Braun). Animal feed is produced once oil for human consumption is pressed from the seeds. However, while the presence of glucosinolates or erucic acid in rapeseed act as protection against pests, diseases, and weeds, they are anti-nutritive factors with various physiological effects in both humans and in animals.

Nevertheless, while protein and glucosinolate content is important in animal feed production, it is the yield of rapeseed oil that is the most important quality control parameter. This is especially true in the European Union and elsewhere, where rapeseed production is being driven hard by an increasing demand for biodiesel as an alternative to conventional fossil fuels.

The team recorded spectra in the range 400 to 2498 nm on intact seeds from 189 samples at 2 nm intervals. The anticipated that they would be able to accurately predict composition from a statistical analysis of the NIR data. If this approach proved viable, they suggest that, "NIR technology could be very useful for the rapid quality evaluation of intact rapeseeds avoiding the need for grinding and saving time."

The team used global calibration equations on the entire Brassica database developed by means of modified partial least squares (MPLS) regression analysis of conventional laboratory analysis values using different data pre-processing treatments. They applied different external validation sets to check the accuracy of the calibration model and found that in all cases the second derivative pre-processing showed the best results.

Actual data revealed that constituents ranged from 34.1 to 48.4 percent for oil, from 14.5 to 31.2 percent for protein and from 15.8 to 97.9 micromoles per gram for total glucosinolates.

"In view of the multiple applications and the present increasing importance of the oilseed rape crops, robust NIR calibrations are necessary to estimate with accuracy the chemical composition of heterogeneous samples including all possible variability due to different factors (plant taxonomy, variety, site, fertilization, sampling date/year)," the team says. "The NIRS analysis of whole rapeseeds in an extremely short time period could allow the processing of samples immediately after harvest and to sow materials selected according quality criteria," they conclude.