Simulating Solar Radiation Under Canopy Using Point Clouds and an Hemispherical Photography Simulator

Solar radiation illuminating the canopy and reaching lower strata of vegetation is an important ecological factor that can enhance biodiversity and contribute to species composition in habitats. It is also part of the many factors that make a natural environment a positive contribution to the health and wellbeing of people interacting with it. Aerial survey methods using drones can now provide digital twins with a very high level of detail. In this work, we developed a software, lasPhotoCamSIM, for simulating hemispherical imagery using dense point clouds as input data. The simulated hemispherical images are used to map the diffuse solar radiation reaching the average height of the human eye from the ground. For the sample design, a regular grid with nodes at 0.5 m ground sampling distance and 1.5 m height from the terrain was used. The study area is a historical garden, Villa Revedin Bolasco. It was surveyed via a drone flight with three sensors, two LiDAR sensors (Riegl VUX-120 and Riegl miniVUX-3UAV) and a camera. The camera provided overlapping imagery that was used to create a third point cloud using photogrammetry. The three point clouds were used as input data to lasPhotoCamSIM together with coordinates of the nodes of the grid, resulting in ~330,000 virtual hemispherical images. Gap fraction and estimated solar diffuse radiation was calculated from each hemispheric image, and it was then converted back to a regular raster map.