Prediction from Linear Regression Equation for Nitrogen Content Measurement in Bentgrasses leaves Using Near Infrared Reflectance Spectroscopy

Near Infrared Reflectance Spectroscopy(NIRS) is a quick, accurate, and non-destructive method to measure multiple nutrient components in plant leaves. This study was to acquire a liner regression equation by evaluating the nutrient contents of ‘CY2’ creeping bentgrass rapidly and accurately using NIRS. In particular, nitrogen fertility is a primary element to keep maintaining good quality of turfgrass. Nitrogen, moisture, carbohydrate, and starch were assessed and analyzed from ‘CY2’ creeping bentgrass clippings. A linear regression equation was obtained from accessing NIRS values from NIR spectrophotometer(NIR system, Model XDS, XM-1100 series, FOSS, Sweden) programmed with WinISI Ⅲ project manager v1.50e and ISIscan® (Infrasoft International) and calibrated with laboratory values via chemical analysis from an authorized institute. The equation was formulated as MPLS(modified partial least squares) analyzing laboratory values and mathematically pre-treated spectra. The accuracy of the acquired equation was confirmed with SEP(standard error of prediction), which indicated as correlation coefficient(r²) and prediction error of sample unacquainted, followed by the verification of model equation of real values and these monitoring results. As results of monitoring, r² of nitrogen, moisture, and carbohydrate in ‘CY2’ creeping bentgrass was 0.840, 0.904, and 0.944, respectively. SEP was 0.066, 1.868, and 0.601, respectively. After outlier treatment, r² was 0.892, 0.925, and 0.971, while SEP was 0.052, 1.577, and 0.394, respectively, which totally showed a high correlation. However, r² of starch was 0.464, which appeared a low correlation. Thereof, the verified equation appearing higher r² of nitrogen, moisture, and carbohydrate showed its higher accuracy of prediction model, which finally could be put into practical use for turf management system.