Solar radiation is a major driver of forest ecosystems and affects the hydrological cycle and energy balance. The vitality of trees strongly depends on the amount of light input to individual trees, and successful forest regeneration demands a certain level of solar radiation. Therefore, characterization of the light regime inside a plant canopy (i.e., quantification and timing of light penetration) is of particular importance. Because these parameters greatly depend on the 3D structure of the forest canopy, Light Detection and Ranging (LiDAR) systems provide suitable technology for solving this task. In this study, an algorithm was developed to estimate the amount of potential solar radiation reaching the forest floor from a point cloud collected in a temperate mixed forest located in Lanzenkirchen (Lower Austria) using terrestrial LiDAR. The path length through the canopy was calculated for 40 forest inventory plots as the distance between the first and last canopy hit along a sun ray, that is, for each day during the vegetation period at an hourly resolution. This distance was easily derived from the z-coordinates of the point cloud after transforming the latter into a coordinate system in which the z-axis ran parallel to the sun rays. Applying Beer's law, the amount of solar radiation on the forest floor was expressed as a function of the solar radiation above the canopy and the path length through the canopy. To validate the results, we used two sets of reference data: (1) hemispherical photographs taken at the same plots, using a camera on a self-levelling mount, which were processed with the HemiView software, and (2) data from a Solariscope survey. Both the Solariscope and HemiView software calculate below-canopy radiation as a percentage of above-canopy radiation, joining direct and diffuse radiation (Total Site Factor, TSF). The TSF values obtained from our modeling procedure were consistent with the Solariscope (R2 = 0.61) and HemiView (R2 = 0.71) results for the entire growing season. Because modern LiDAR-based forest inventories provide all the necessary data for our light modeling approach (the point cloud, the geographical position, and a record of the occurring tree species) without additional requirements, together with the easy applicability of the algorithm, this tool is a promising asset for ecological and silvicultural activities.
