Estimating radiation interception in an olive orchard using physical models and multispectral airborne imagery

This study was conducted to estimate the fraction of Intercepted Photosinthetically Active Radiation (fIPAR) in an olive orchard. The method proposed to estimate fIPAR in olive canopies consisted of a coupled radiative transfer model that linked the 3D Forest Light Interaction Model (FLIGHT) and the Orchard Radiation Interception Model (ORIM). This method was used to assess the estimation of instantaneous fIPAR as a function of planting grids, percentage cover, and soil effects. The linked model was tested against field measurements of fIPAR acquired for a commercial olive orchard, where study plots showing a gradient in the canopy structure and percentage cover were selected. High-resolution airborne multispectral imagery was acquired at 10 nm bandwidth and 15-cm spatial resolution, and the reflectance used to calculate vegetation indices from each study site. In addition, simulations of the land surface bidirectional reflectance were conducted to understand the relationships between canopy architecture and fIPAR on typical olive orchard planting patterns. Input parameters used for the canopy model, such as the leaf and soil optical properties, the architecture of the canopy, and sun geometry, were studied in order to assess the effect of these inputs on the Normalized Difference Vegetation Index (NDVI) and fIPAR relationships. FLIGHT and ORIM models were independently assessed for fIPAR estimation using structural and ceptometer field data collected from each study site, yielding RMSE values of 0.1 for the FLIGHT model, while the specific olive simulation model by ORIM yielded lower errors (RMSE = 0.05). The reflectance simulations conducted as a function of the orchard architecture confirmed the usefulness of the modeling methods for this heterogeneous olive crop, and the high sensitivity of the NDVI and fIPAR to background, percentage cover, and sun geometry on these heterogeneous orchard canopies. The fIPAR estimations obtained from the airborne imagery through predictive relationships yielded RMSE error values of 0.11 when using FLIGHT to simulate both the canopy reflectance and the fIPAR of the study sites. The coupled FLIGHT+ORIM model yielded better results, obtaining RMSE = 0.05 when using airborne remote sensing imagery to estimate fIPAR.