Premium
Estimation of nitrogen concentration and in vitro dry matter digestibility of herbage of warm‐season grass pastures from canopy hyperspectral reflectance measurements
Author(s) -
Starks P. J.,
Zhao D.,
Brown M. A.
Publication year - 2008
Publication title -
grass and forage science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.716
H-Index - 56
eISSN - 1365-2494
pISSN - 0142-5242
DOI - 10.1111/j.1365-2494.2007.00619.x
Subject(s) - reflectivity , dry matter , grazing , linear regression , canopy , pasture , zoology , spectroradiometer , forage , nitrogen , environmental science , agronomy , dry season , mathematics , biology , chemistry , botany , statistics , ecology , physics , organic chemistry , optics
Remote sensing of nitrogen (N) concentration and in vitro dry matter digestibility (IVDMD) in herbage can help livestock managers make timely decisions for adjusting stocking rate and managing pastures during the grazing season. Traditional laboratory analyses of N and IVDMD are time‐consuming and costly. Non‐destructive measurements of canopy hyperspectral reflectance of pasture may provide a rapid and inexpensive means of estimating these measures of nutritive value. Using a portable spectroradiometer, canopy reflectance was measured in eight warm‐season grass pastures in the USA in June and July in 2002 and 2003 to develop and validate algorithms for estimating N concentration and IVDMD of herbage. Nitrogen concentration of herbage was linearly correlated ( r = 0·82; P < 0·001) with a ratio of reflectance in the 705‐ and 1685‐nm wavebands ( R 705 / R 1685 ) and IVDMD was correlated with R 705 / R 535 (r = 0·74; P < 0·001). Compared with simple linear regressions of N concentration and IVDMD in herbage with two‐waveband reflectance ratios, multiple regression, using maximum r 2 improvement, band‐depth analysis with step‐wise regression, and partial least‐squares regression enhanced the correlation between N concentration and IVDMD of herbage and canopy reflectance values (0·81 ≤ | r | ≤ 0·90; P < 0·001). Validation of the prediction equations indicated that multiple regression only slightly improved accuracy of a model for predicting N concentration and IVDMD of herbage compared with simple linear regression of reflectance ratios. Results suggest that the N concentration and IVDMD of herbage of warm‐season grass pastures can be rapidly and non‐destructively estimated during the grazing season using canopy reflectance in a few narrow wavebands.