Prediction of forage chemical composition by NIR spectroscopy

Abstract

Near-infrared spectroscopy (NIR spectroscopy) has been used in analytics for more than 50 years. The aim of this review is to present statistical indicators of the developed calibration models for predicting forage chemical composition by NIR spectroscopy, which have been published over the last 15 years. This paper presents statistics for predicting of forage dry matter (DM), crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), ash, and pH value of forage at different pre-scan processing level (fresh, dried / ground forage) and different forage types such as grass monocultures, legumes, grass-clover mixtures (GCM), semi-natural pasture, straw, maize, hay, silage and haylage. Due to wider applicability of NIR calibration model for prediction of chemical composition of forage, the development of calibration includes forage originating from various agricultural production technologies, cultivation climates, varieties and vegetation seasons, etc. In order to develop more reliable calibration models for prediction of forage chemical composition, calibrations are developed for individual plant species, cultivars, harvest during the vegetation season, as well as for individual microclimates of cultivation. NIR spectroscopy has high potential for predicting the content of DM, CP, ABSTRACT Chemical composition of forage is of a high importance in animal nutrition, plant breeding programmes, identifying possible animal health problems, etc. Near infrared spectroscopy (NIR spectroscopy) is an alternative method for classical chemical analysis to predict forage chemical composition. For this purpose, NIR spectroscopy is based on a combination of laboratory chemical analysis data and spectral data to predict the each of the individual chemical parameters. Compared to classical chemical analysis, NIR spectroscopy is an environmentally friendly, multi-analytical, physical, rapid and non-destructive method. The use of NIR spectroscopy to predict the chemical composition of forage may be equally accurate but significantly cheaper compared to wet chemistry. Near infrared spectroscopy (NIR spectroscopy) has been used in analytics for more than 50 years. The aim of this review is to present statistical indicators of the developed calibration models for predicting chemical composition of forage by NIR spectroscopy, which have been published over the last 15 years. This paper presents statistics for predicting forage dry matter (DM), crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), ash, and pH value of forage at different pre-scan processing level (fresh, dried / ground forage) and for different forage types such as grass monocultures, legumes, grass-clover mixtures (GCM), semi-natural pasture, straw, maize, hay, silage and haylage. A coefficient of determination (R 2 ), standard error of calibration (SEC), standard error of cross-validation (SECV) and standard error of prediction (SEP), as a basic calibration statistics are presented for each of the calibration model. Due to wider applicability of NIR calibration model for prediction of chemical composition of forage, the development of calibration includes forage originating from various agricultural production technologies, cultivation climates, varieties and vegetation seasons, etc. In order to develop more reliable calibration models for prediction of forage chemical composition, calibrations are developed for individual plant species, cultivars, harvest during the vegetation season, as well as for individual microclimates of cultivation. The decision on the type of calibration and the procedure to be followed when developing calibrations depend on the acceptable deviations of the results of analysis and the extent to which they can compensate the speed of obtaining results and the cost of the analysis. The basic statistic indicators related to the applicability and the accuracy of the developed NIR estimation model show a high potential for predicting the content of DM, CP, NDF, ADF, ash and pH value in forage.