Influence of solar zenith angle on the enhanced vegetation index of a Guyanese rainforest

Research output: Contribution to journalArticleAcademicpeer-review

8 Citations (Scopus)

Abstract

In this study, the effect of solar zenith angle () on enhanced vegetation index (EVI) of a Guyanese tropical rainforest was studied. For this sub-crown resolution, hyperspectral data have been collected with an unmanned aerial vehicle (UAV) at five different solar zenith angles in a 1-day period. The hyperspectral data were used to simulate Moderate Resolution Imaging Spectroradiometer (MODIS) spectral bands and generate EVI. The linear trend of EVI with solar zenith angle at nadir viewing conditions was found to be –0.00285 (). The direction of this trend was in agreement with earlier studies, but with a differing magnitude. Analysis of EVI images with sub-crown resolution pointed to strong influence of canopy shadows on EVI, which is supported by other studies. Additionally, the EVI–solar zenith angle trend was investigated in the semi-empirical RossThick-LiSparse-Reciprocal (RTLSR) model implemented in the MODIS MCD43 product suite. A database of model parameters has been created and the EVI–solar zenith angle trend was modelled with each set of parameters. The linear approximated trend was found to be –0.00219 on average, only slightly weaker compared to the trend derived from the UAV. Further analysis of the relationship between the single RTLSR model parameters and the EVI–solar zenith angle trend showed that the RTLSR produces the trend for the right reason, namely canopy shadowing expressed by the near-infrared geometric kernel. In total, this study delivers further evidence that EVI is dependent on solar zenith angle and this effect is mediated through EVI’s sensitivity to within-canopy shadows.
Original languageEnglish
Pages (from-to)972-981
JournalRemote Sensing Letters
Volume6
Issue number12
DOIs
Publication statusPublished - 2015

Fingerprint Dive into the research topics of 'Influence of solar zenith angle on the enhanced vegetation index of a Guyanese rainforest'. Together they form a unique fingerprint.

Cite this