— Back to Blogs & Publications Crude Fiber and Ash Content Determination in Feed Rations by NIR Spectroscopy

Crude Fiber and Ash Content Determination in Feed Rations by NIR Spectroscopy

How crude fibre and crude ash are determined by the reference methods, what near-infrared spectroscopy can and cannot do for these two parameters in compound feed, and how an at-line NIR step fits into a feed mill's quality routine.

Why fibre and ash matter at the mill

Compound feed mills in the food and feed industry blend grains, protein meals and co-products into species-specific rations. In the European Union the analytical constituents declared on the label, including crude fibre and crude ash, are subject to official control, and the reference methods for that control are laid down in Commission Regulation (EC) No 152/2009 [1]. Near-infrared spectroscopy has been studied specifically as a tool for checking the mandatory label declaration of compound feeds [2].

What the reference methods involve

Crude fibre is defined operationally as the fat-free organic residue that is insoluble in dilute acid and alkali. The standardised procedure (ISO 6865; Regulation 152/2009, Annex III, Division I) boils the sample for 30 minutes in 0.13 mol/L sulphuric acid and then for 30 minutes in 0.23 mol/L potassium hydroxide, filters and washes the residue, dries it at 130 °C to constant weight and ashes it at 475–500 °C; crude fibre is the loss on ashing [1][3]. Crude ash (ISO 5984; Division M) is the residue after igniting the sample at 550 °C for about three hours, with re-ashing to constant weight [1][4]. Even the ash protocol is not trivial: Quirino et al. show that the 550 °C procedure adopted by ISO 5984 is not fully robust across feed types and can overestimate ash in some matrices [5].

Both procedures take several hours of bench and furnace time per batch, which means the results are normally not available while the batch that produced the sample is still being made.

What NIR can do for these parameters

Near-infrared calibration is an established route to rapid compositional analysis of feeds, and ISO 12099 is the guideline for applying it to animal feeding stuffs [6]. For compound feeds, Pérez-Marín et al. obtained calibrations on commercial feeds with R² of 0.98 for crude fibre and 0.97 for crude protein, with cross-validation standard errors of 5.2 and 5.5 g/kg respectively, and R² 0.87 (4.9 g/kg) for moisture [2]. A conference study by the same group on 287 unground Spanish compound feeds reported R² 0.97 (SECV 0.63 %) for crude fibre but only 0.86 (SECV 0.83 %) for ash [7].

The weaker ash result is expected. Minerals have no near-infrared absorption of their own; Clark et al. showed on forages that mineral calibrations work indirectly, through the association of minerals with organic constituents such as chlorophyll and organic-acid salts, and are usable mainly for the major minerals [8]. Ash calibrations can still be useful: Pojić et al. reached R² 0.94–0.97 for ash in legumes [9], and a 2025 study of a MEMS micro-NIR instrument on poultry feed and ingredients, with 30 PLS models built on 1,437 reference samples, found hold-out RPD above 3 for most protein, fat and moisture models but above 2 for the majority of fibre and ash models [10]. The practical reading is that NIR fibre and ash values are screening values whose accuracy has to be established for each feed type against the reference methods; they complement the reference analyses rather than replace them.

Building and maintaining the calibration

The chemometric workflow is standard. Scatter correction such as the standard normal variate transform reduces the multiplicative effects of particle size and packing on diffuse-reflectance spectra [11]; Savitzky-Golay derivative filtering removes baseline offsets [12]; and partial least squares regression relates the pre-processed spectra to the reference values [13]. Simmler et al. selected the pre-processing for each of their models by cross-validation rather than by a fixed recipe [10]. Two points matter in a mill:

  • Reference data quality bounds the calibration. A NIR model cannot be more accurate than the ISO 6865 and ISO 5984 results it is trained on, so reference analyses should follow the standard procedures [3][4] and the ash protocol should be fixed and documented [5].
  • Co-products need their own populations. Materials such as DDGS vary widely in composition, and model performance is co-determined by that heterogeneity [10]; calibrations should be validated separately for each ingredient class rather than assumed to transfer.

Where USTECH fits

The USTECH MasterLine17EC at-line analyzer is used for finished feeds and raw meals, and the caliX suite provides the SNV, derivative and PLS tools described above [11][12][13]. Which parameters reach the accuracy a mill needs — and whether ash is among them for a given feed type — is established in a feasibility study on that mill's own samples and reference analyses, in line with ISO 12099 [6].

References

  • [1] Commission Regulation (EC) No 152/2009 of 27 January 2009 laying down the methods of sampling and analysis for the official control of feed, Annex III, Divisions I (crude fibre) and M (crude ash). Official Journal of the European Union, L 54, 26.2.2009 (as amended). eur-lex.europa.eu/eli/reg/2009/152/oj
  • [2] Pérez-Marín, D.C., Garrido-Varo, A., Guerrero-Ginel, J.E., Gómez-Cabrera, A. (2004). Near-infrared reflectance spectroscopy (NIRS) for the mandatory labelling of compound feedingstuffs: chemical composition and open-declaration. Animal Feed Science and Technology, 116(3–4), 333–349. doi:10.1016/j.anifeedsci.2004.05.002
  • [3] ISO 6865:2000. Animal feeding stuffs — Determination of crude fibre content — Method with intermediate filtration. ISO, Geneva. iso.org/standard/13377.html
  • [4] ISO 5984:2022. Animal feeding stuffs — Determination of crude ash. ISO, Geneva. iso.org/standard/77807.html
  • [5] Quirino, D.F., Palma, M.N.N., Franco, M.O., Detmann, E. (2022). Variations in methods for quantification of crude ash in animal feeds. Journal of AOAC International, 106(1), 6–13. doi:10.1093/jaoacint/qsac100
  • [6] ISO 12099:2017. Animal feeding stuffs, cereals and milled cereal products — Guidelines for the application of near infrared spectrometry. ISO, Geneva. iso.org/standard/67352.html
  • [7] Garrido-Varo, A., Pérez-Marín, M.D., Gómez-Cabrera, A., Guerrero-Ginel, J.E., de Paz, F., Delgado, N. (2001). NIRS as an essential tool in food safety programs: feed ingredients prediction in commercial compound feeding stuffs. Proceedings of the Korean Society of Near Infrared Spectroscopy Conference, 1153. koreascience.or.kr/article/CFKO200111922080727.page
  • [8] Clark, D.H., Mayland, H.F., Lamb, R.C. (1987). Mineral analysis of forages with near infrared reflectance spectroscopy. Agronomy Journal, 79(3), 485–490. doi:10.2134/agronj1987.00021962007900030016x
  • [9] Pojić, M., Mastilović, J., Palić, D., Pestorić, M. (2010). The development of NIRS calibration for prediction of ash content in legumes on the basis of two different reference methods. Food Chemistry, 123(3), 800–805. doi:10.1016/j.foodchem.2010.05.013
  • [10] Simmler, M., Ishizaki, K., Mulloni, A., Abbate, R., Juste, A., Sroka, A. (2025). Enabling precision feeding: A novel micro-NIRS system for rapid on-site analysis of the nutritional composition of poultry feed and feed ingredients. Smart Agricultural Technology, 12, 101220. doi:10.1016/j.atech.2025.101220
  • [11] Barnes, R.J., Dhanoa, M.S., Lister, S.J. (1989). Standard Normal Variate Transformation and De-trending of Near-Infrared Diffuse Reflectance Spectra. Applied Spectroscopy, 43(5), 772–777. doi:10.1366/0003702894202201
  • [12] Savitzky, A., Golay, M.J.E. (1964). Smoothing and Differentiation of Data by Simplified Least Squares Procedures. Analytical Chemistry, 36(8), 1627–1639. doi:10.1021/ac60214a047
  • [13] Wold, S., Sjöström, M., Eriksson, L. (2001). PLS-regression: a basic tool of chemometrics. Chemometrics and Intelligent Laboratory Systems, 58(2), 109–130. doi:10.1016/S0169-7439(01)00155-1
USTECH Innovations Logo

Discuss this application for your process

Our application team can walk through the method, share the supporting literature and discuss how it applies to your process.

Contact Us

Read Next

Reducing Ingredient Waste in Cattle Feed Production
ARTICLE

Reducing Ingredient Waste in Cattle Feed Production

Why cattle feed is over-formulated against ingredient variability, what the literature says about where that variability comes from, and how NIR at intake and at the mixer narrows the margin.

Read Article
Real-Time Fat Analysis in Ground Beef Processing
ARTICLE

Real-Time Fat Analysis in Ground Beef Processing

Why fat content is the controlling specification for ground beef, what the reference methods involve, and what peer-reviewed on-line NIR studies at grinder outlets and on conveyors actually report.

Read Article