Estimation of grassland biomass and nitrogen using MERIS data

This study aimed to investigate the potential of MERIS in estimating the quantity and quality of a grassland using various vegetation indices (NDVI, SAVI, TSAVI, REIP, MTCI and band depth analysis parameters) at a regional scale. Green biomass was best predicted by NBDI (normalised band depth index)...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:ITC journal Jg. 19; H. 1
Hauptverfasser: Ullah, G, Si, Y, Schlerf, M, Skidmore, A.K, Shafique, M, Iqbal, I.A
Format: Journal Article
Sprache:Englisch
Veröffentlicht: 01.10.2012
Schlagworte:
aboveground biomass
band-depth analysis
branta-leucopsis
broad-band
canopy
hyperspectral data
quality
red edge position
terrestrial chlorophyll index
vegetation indexes
ISSN:0303-2434
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This study aimed to investigate the potential of MERIS in estimating the quantity and quality of a grassland using various vegetation indices (NDVI, SAVI, TSAVI, REIP, MTCI and band depth analysis parameters) at a regional scale. Green biomass was best predicted by NBDI (normalised band depth index) and yielded a calibration R2 of 0.73 and a Root Mean Square Error (RMSE) of 136.2 g m-2 (using an independent validation dataset, n = 30) compared to a much higher RMSE obtained from soil adjusted vegetation index SAVI (444.6 g m-2). Nitrogen density was also best predicted by NBDI and yielded a calibration R2 of 0.51 and a RMSE of 4.2 g m-2 compared to a relatively higher RMSE obtained from MERIS terrestrial chlorophyll index MTCI (6.6 g m-2). For the estimation of nitrogen concentration (%), band depth analysis parameters showed poor R2 of 0.21 and the results of MTCI and REIP were statistically non-significant (P > 0.05). It is concluded that band depth analysis parameters consistently showed higher accuracy than vegetation indices, suggesting that band depth analysis parameters could be used to monitor grassland condition over time at regional scale.
ISSN:0303-2434
DOI:10.1016/j.jag.2012.05.008