Near infrared spectroscopy for enforcement of European legislation concerning the use of animal by-products in animal feeds

• Main Authors: Martnez A., Vicente F., Soldado A., Bautista J., de la Haba MJ., Gmez-Cabrera A., Guerrero JE., Pérez-Marn MD., Garrido-Varo A, de la Roza-Delgado B., Termes S.
• Format: Article
• Language: English
• Published: Presses Agronomiques de Gembloux 2005-01-01
• Series: Biotechnologie, Agronomie, Société et Environnement
• Subjects: Compound feeds; animal by-products; meat and bone meal; animal fats; NIRS; unground sample analysis
• Online Access: http://www.pressesagro.be/base/text/v9n1/3.pdf

The paper summarises the work done in the framework of two R&D projects aimed to demonstrate the contribution of Near Infrared Spectroscopy (NIRS) to help the enforcement of the European legislation governing the use of animal by-products in animal feeds. Three different types of animal feed products were studied: compound feeds (CFs), animal protein byproducts meals (APBPs) and animal fats by-products (AFBPs). The quantitative and qualitative chemometric models produced with a large collection of compound feed samples (n = 1005 ground and 523 unground) have demonstrated, that NIRS can be used for the detection and quantification of the meat and bone meal (MBM) added to compound feeds. Discriminant models produced with unground samples produced 100% of correctly classified samples in two cloned instruments placed in two different locations. The results also show that two dimensions NIR spectra of Animal By-Products (ABP, animal meals and fats) may contain information about the animal species or group of species from which the ABPs were produced. However, further work is needed to enlarge the sample bank and the spectral libraries with well authenticated samples in order to increase the robustness of the quantitative and qualitative NIRS models. The paper opens expectations for using NIRS for the enforcement of legislation concerning the use of ABPs in animal feeds. More research and demonstration efforts have to be done in order to obtain more definitive and robust predictive models and for optimising its implementation either at-line, on-line and in-line in feed factories and inspection laboratories.