| Estimation of leaf area index in Zagros forests using SENTINEL-1 and SENTINEL-2 data |
| Paper ID : 1224-SMPR |
| Authors: |
|
Sasan Vafaei *1, Omid Fathizade2, Hadi Fadaei3 1Tehran university 2Ahar Faculty of Agriculture and Natural Resources, University of Tabriz, Ahar, Iran 3Assistant Professor of Amin Police University, Geography group |
| Abstract: |
| Using remote sensing data is an applied method to estimate leaf area index (LAI). In this study, satellite radar data of sentinel-1, with the dual polarization (VH, VV) and the optical data of Sentinel-2, has been used to estimate the LAI in a part of the Zagros forests, Ilam province, Iran. In order to measure LAI, 61 sample plots were laid out and fisheye images were collected using hemispherical photography in each ground sample plot. The backscattering coefficients at VH and VV polarization, which obtained from original and synthetic bands of optical images in two different combinations of radar (S1) images and optical images (S2) were used as inputs of the multiple linear regression (MLR) models. So, the spectral values of corresponding plots were extracted from S1 and S2 dataset. Finally, the relationship between the two datasets and ground LAI were analyzed using multiple regression with 75% of the samples and Validated using the rest of control plots (25%).The results showed that coefficient of determination (R2) and RMSE to predict LAI based on sentinel-2 dataset (optical dataset) were 0.66 and 0.29, respectively. The mentioned parameters were estimated to be 0.02 and 0.56, respectively, with SENTINEL-1 (radar dataset). We conclude that there is an appropriate potential of using the sentinel-2 images in estimating the forest LAI. |
| Keywords: |
| Leaf area index (LAI), Multiple linear regression (MLR), SAR Data, SENTINEL, Zagros Forest |
| Status : Paper Accepted (Poster Presentation) |
