Predicting the Evolution of Pasture Quality by NIRS: Perspectives for Real-Time Pasture and Grazing Management

Abstract

Pasture quality monitoring is a key element in the decision making process of the farm manager. Laboratory reference methods for assessing pasture quality parameters such as crude protein (CP) or neutral detergent fibre (NDF) require cutting, collection and analytical procedures involving technicians, time and reagents, making them laborious and expensive. The objective of this study was to evaluate the potential of near infrared reflectance spectroscopy (NIRS) combined with multivariate data analysis as a rapid method to predict and monitor the evolution of pasture quality parameters (CP, NDF and a pasture quality index, PQI=CP/NDF). During the 2018 and 2019 growing seasons a total of 398 composite pasture samples were collected from 9 biodiverse pastures, representing a wide range of botanical composition and phenological states. These samples were scanned with a FT-NIR spectrometer: 315 (collected in 2018) were used in the calibration phase and 83 (collected in 2019) were used during the validation phase. Calibration and validation models were developed and regression equations between predicted and laboratory reference values of CP, NDF and PQI were established. Were used as evaluation parameters the coefficient of determination (R2 ), the residual predictive deviation (RPD) and the root mean square errors (RMSE). The best results obtained were: (i) for CP prediction model (R2 =0.844; RPD=4.0; RMSE=1.622); (ii) for NDF prediction model (R2 =0.826; RPD=2.4; RMSE=4.200); and (iii), for PQI prediction model (R2 =0.808; RPD=3.2; RMSE=0.066). The results show the practical interest of portable spectrometry, associated with GNSS, as expeditious tools for monitoring pasture quality. Good prospects and opportunities open up for technology-based service providers to develop remote sensing-based computer applications from satellite imagery that enable dynamic management of animal grazing.