Schätzung von Inhaltsstoffen einheimischer Leguminosen mit Hilfe der Fourier-Transform Nah-Infrarot-Reflektions-Spektroskopie (NIRS)

Kurzfassung

Legumes as protein feedstuffs could solve the problems of protein supply in organic farming. To evaluate the quality of these feedstuffs the NIRS could be used to predict the main ingredients. The basic for a successful application of the NIRS are stable calibration equations for the prediction of the ingredients. Therefore reference analyses are necessary. In the presented investigation 200 legumes were analysed by classical chemical methods and also by NIRS. The measured reference data were used to develop NIRS equations. Using two-thirds of the samples as a calibration set and the remaining samples as a validation set NIRS equations were calculated by partial least square regression and different pretreatments of spectra. Prediction was satisfactory for dry matter, crude protein, ether extract and crude fiber. NIRS prediction was less accurate for crude ash. On the improvement of the NIRS equations should be further worked.