Abstract
With the launch of the Sentinel-2 satellites, a European capacity has been created to ensure continuity of Landsat and SPOT observations. In contrast to previous sensors, Sentinel-2′s multispectral imager (MSI) incorporates three additional spectral bands in the red-edge (RE) region, which are expected to improve the mapping of vegetation traits. The objective of this study was to compare Sentinel-2 MSI and Landsat-8 OLI data for the estimation of leaf area index (LAI) in temperate, deciduous broadleaf forests. We used hemispherical photography to estimate effective LAI at 36 field plots. We then built and compared simple and multiple linear regression models between field-based LAI and spectral bands and vegetation indices derived from Landsat-8 and Sentinel-2, respectively. Our main findings are that Sentinel-2 predicts LAI with comparable accuracy to Landsat-8. The best Landsat-8 models predicted LAI with a root-mean-square error (RMSE) of 0.877, and the best Sentinel-2 model achieved an RMSE of 0.879. In addition, Sentinel-2′s RE bands and RE-based indices did not improve LAI prediction. Thirdly, LAI models showed a high sensitivity to understory vegetation when tree cover was sparse. According to our findings, Sentinel-2 is capable of delivering data continuity at high temporal resolution.
Highlights
Most of a plant’s atmospheric exchanges happen through the leaves [1]
Sentinel-2 models performed slightly better than Landsat-8 models in predicting LAIe
The best Sentinel-2 model based on stepwise selection explained 59.3% of the variance in LAIe and had an root-mean-square error (RMSE) of 0.888, whereas the best Landsat-8 model explained only 48.7% of the variance, but had a lower RMSE of 0.813 (Figure 3)
Summary
Most of a plant’s atmospheric exchanges happen through the leaves [1]. The amount of leaf area is related to many plant–atmosphere processes, such as photosynthesis [2], evaporation and transpiration [3], rainfall interception [4], and carbon flux [5]. The quantity of that exchange is strongly correlated with the total amount of leaf area [1]. The leaf area of plant canopies is usually measured or estimated in the form of leaf area index (LAI), which is defined in broadleaf forests as the one-sided total green leaf area per unit ground surface area (m2/m2) [6]. LAI is considered by the United Nations Framework Convention on Climate Change (UNFCCC) and the Intergovernmental Panel on Climate Change (IPCC) as an essential climate and biodiversity variable [7]. Measuring LAI at local to global scales is still an active area of research (e.g., [8,9])
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call