HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Figures Only Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Biewer, S. Articles by Wachendorf, M. PubMed Articles by Biewer, S. Articles by Wachendorf, M. Agricola Articles by Biewer, S. Articles by Wachendorf, M. Related Collections Forage Management Spatial Variability Field evaluation techniques

FORAGE & GRAZINGLANDS

Determination of Dry Matter Yield from Legume–Grass Swards by Field Spectroscopy

Dep. of Grassland Science and Renewable Plant Resources, Univ. of Kassel, Steinstr. 19, D-37213 Witzenhausen, Germany

^* Corresponding author (mwach{at}uni-kassel.de).

An efficient and accurate detection of dry matter (DM) yield of legume–grass mixtures can facilitate a targeted and site-specific management of legume-based swards. The major objective of this study was to examine the relationship between spectral signatures of legume–grass swards and DM yield across a wide range of legume species (white clover [Trifolium repens L.], red clover [T. pratense L.], alfalfa [Medicago sativa L.], birdsfoot trefoil [Lotus corniculatus L.]), legume proportion (0–100% of DM), and growth stage (beginning of tillering to end of flowering). Modified partial least squares (MPLS) regression, stepwise multiple linear regression (SMLR), and the vegetation indices (VIs), simple ratio, normalized difference vegetation index, enhanced vegetation index, and red edge position were used for analysis of the hyperspectral data set (350–2500 nm). Compared to common calibrations, legume-specific models achieved better results, indicating that each legume species had its own spectral characteristics. Modified partial least squares and SMLR gave the best R² values, ranging in cross validation from 0.74 to 0.92 with a standard error below 92 g DM m^–2. The DM yield prediction by VIs resulted in unsatisfactory accuracies. Prediction accuracy for MPLS and SMLR models were still acceptable even with a reduced spectral data set (630–1000 nm), a finding that could facilitate an application of field spectroscopy in practice.

Abbreviations: DM, dry matter • EVI, enhanced vegetation index • MPLS, modified partial least squares • NDVI, normalized difference vegetation index • PLS, partial least square • REP, red edge position • RPD, residual predictive value • SEC, standard error of calibration • SECV, standard error of cross-validation • SMLR, stepwise multiple linear regression • SR, simple ratio • VI, vegetation index • 1-VR, coefficient of determination of cross-validation