Applying a solar model to LiDAR images of an agrivoltaic pear orchard

Agrivoltaic farming is the practice of growing crops underneath solar panels. Using photovoltaic arrays above tree canopies can potentially save water by limiting transpiration, protect trees and fruit from damage (extreme heat, sunlight, hail), and avoid plant water stress in the hottest periods of the day. The aim of this study was to simulate tree total and localised light interception under different configurations of photovoltaic arrays in relation to control trees (no photovoltaic arrays). The study was conducted in an experimental pear orchard at the Tatura SmartFarm (Victoria, Australia). Trees of the Australian blush pear ‘ANP-0118’ trained to an open Tatura trellis (V) were used for the experiment. LiDAR (laser imaging detection and ranging) and RGB images of trees (on the east and west sides of the row) were obtained with the Leica-BLK360 G1 scanner in three different treatments: a control with no photovoltaic arrays, above-canopy photovoltaic arrays angled at 45° to the west, and above-canopy photovoltaic arrays angled at 5° to the west. A solar radiation model was applied to LiDAR-point cloud images obtained in the three treatments to simulate canopy light distribution. The utility of the sensor for canopy light distribution modelling and for additional uses such as fruit number estimations proved successful. The results showed a significant reduction in canopy illumination under the 5° west treatment and reduced fruit number when compared to the control.