Can Grassland Chemical Quality Be Quantified Using Transform Near-Infrared Spectroscopy?

Area of Science:

• Agricultural Science
• Analytical Chemistry
• Spectroscopy

Background:

• Near-infrared spectroscopy (NIRS) is used for forage and animal feed quality analysis.
• Variability in grasslands (site, year, species) limits traditional NIRS prediction accuracy.
• Need for rapid, reliable methods for analyzing fresh, unprocessed grassland samples.

Purpose of the Study:

• To evaluate Fourier transform NIRS (FT-NIRS) for determining chemical characteristics of fresh, undried, unground grassland samples.
• To assess the predictive capability of FT-NIRS models for key nutritional parameters.
• To determine the feasibility of FT-NIRS for routine grassland analysis.

Main Methods:

• Application of Fourier transform NIRS (FT-NIRS) on fresh, undried, and unground grassland samples.
• Development and validation of predictive models for chemical components.
• Statistical analysis of model performance using R2, Residual Prediction Deviation (RPD), and Range Error Ratio (RER).

Main Results:

• FT-NIRS models achieved high accuracy (R2 > 0.90) for dry matter (DM), crude protein (CP), acid detergent fibre (ADF), neutral detergent fibre (NDF), and acid detergent lignin (ADL).
• Quantitative determination was successful for CP and ADF (RPD > 3, RER > 10), and DM and NDF (RER > 10).
• Screening for ADL was feasible (RPD = 2.4, RER = 8.8), but models for lipid and ash were not practically usable.

Conclusions:

• FT-NIRS is a successful and promising tool for quantifying major chemical parameters in fresh grassland samples without prior preparation.
• The method's efficiency in analyzing CP, ADF, DM, and NDF supports its practical application in animal diet formulation.
• Further research may be needed to improve models for lipid and ash content analysis using FT-NIRS.