Defra - Department for Environment, Food and Rural Affairs.

Science Search

Science and Research Projects

Return to Science Search homepage   Return to Project List

Validation of Near Infrared Spectrometry (NIRS) for the prediction of DOMD (Digestible Organic Matter Determination) in Lolium perenne. - VS02009

Description
The UK NL VCU forage system needs to be cost effective and produce data relevant to industry practice. Current UK NL VCU quality assessments for Lolium perenne varieties concentrate on DOMD values (D-values) that relate the digestibility of the variety to the metabolisable energy using standard chemical analyses.

NIAB has been investigating the use of NIRS to predict D-values of grass samples from NL trials as a cheaper alternative to current gravimetric chemical analyses since 2001. Calibrations to relate NIRS spectra to D-values have been demonstrated to produce similar outcomes in terms of standard error of prediction and correlation when data is analysed in a single set. However, data accumulated to date, highlight a significant site/method interaction, evident in the regression equations between wet chemistry and NIRS at each site. This has led to concerns as to whether NIRS measures would support comparable decision making within the NL VCU, which is a critical test for the acceptability of NIRS in this context.

Our original experiments were designed to test the feasibility of NIRS as a replacement method to standard wet chemistry approaches. In early 2003 the NL VCU committee requested a re-analysis of NIAB data to provide guidance as to the effect changing to NIRS methods would have had on decision making. All wet chemistry data for the two harvest years used in our original study was converted to ‘surrogate NIRS’ data by use of regression equations specific to each site. The resulting data matrix showed very limited differences between wet chemistry and NIRS on variety ranking across a range of cuts. There was some concern that the original samples had not been selected to test the possibility of site interaction and therefore not all varieties had been tested at all the NL sites.

A NL VCU forage committee meeting in October 2003 requested that NIAB extend its NIRS investigations to analyse a sub set of varieties at every site and every cut in order to validate the findings in the original study. This proposal seeks to meet this request.

This project will allow Defra and the NL VCU herbage committee to assess whether it is feesible to improve the cost effectiveness of digestibility testing in the NL VCU system by moving to an NIRS method without significant loss of accuracy or reliability.


Project Documents
• Final Report : Final Report - Validation of Near Infrared Spectrometry (NIRS) for the prediction of DOMD (Digestible Organic Matter Determination) in Lolium perenne.   (465k)
Time-Scale and Cost
From: 2004

To: 2004

Cost: £4,500
Contractor / Funded Organisations
National Institute of Agricultural Botany
Keywords
Plant Varieties and Seeds              
Plants and Animals              
Variety Testing              
Fields of Study
Plant Varieties and Seeds