Winter wheat biomass estimation based on canopy spectra
The winter wheat aboveground biomass is an important agronomic parameter to estimate the growth status, and evaluate the yield and quality. Spectrum technique provides a nondestructive and fast method for estimating the winter wheat biomass. In order to find the optimum model by analyzing the wheat...
Gespeichert in:
Veröffentlicht in: | International journal of agricultural and biological engineering 2015-12, Vol.8 (6), p.30-30 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The winter wheat aboveground biomass is an important agronomic parameter to estimate the growth status, and evaluate the yield and quality. Spectrum technique provides a nondestructive and fast method for estimating the winter wheat biomass. In order to find the optimum model by analyzing the wheat canopy spectral characteristic during the whole growth period, field trails were conducted at the National Demonstration Base of Precision Agriculture in Beijing Xiaotangshan town. A portable spectrometer (200-1100 nm) was used to collect the wheat canopy spectra of different varieties at the different growth stages (green stage, jointing stage, booting stage, heading stage and filling stage), clipping the winter wheat at ground level at the same time. Regression and correlation analysis were used to establish the winter wheat biomass estimation models in this study. The results showed that the biggest different bands of the winter wheat canopy spectral reflection curves mainly lied along the blue and near-infrared bands. The spectral reflectance at 678 nm in the visible light range had the best correlation with the biomass (correlation=0.724). The monadic regression analysis, the multiple regression analysis and the partial least squares regression analysis were applied to establish the biomass estimation models, among which the partial least squares regression (PLS) model had higher modeling precision. The R2 of the calibration and validation were 0.916 and 0.911, respectively. The root-mean-square error (RMSE) of the calibration and validation were 0.090 kg and 0.094 kg (Sample area 50 cm×60 cm). The results indicated that the PLS model (400-1000 nm) could fully estimate the aboveground biomass in the whole growth period of wheat with a better measurement accuracy. |
---|---|
ISSN: | 1934-6344 1934-6352 |
DOI: | 10.3965/j.ijabe.20150806.1311 |