Virtual Geographic Simulation of Light Distribution within Three-Dimensional Plant Canopy Models

• Main Authors: Liyu Tang, Dan Yin, Shuwei Chen, Chongcheng Chen, Hongyu Huang, Ding Lin
• Format: Article
• Language: English
• Published: MDPI AG 2017-12-01
• Series: ISPRS International Journal of Geo-Information
• Subjects: virtual geographic environment; three-dimensional model; light distribution; canopy structure; radiosity
• Online Access: https://www.mdpi.com/2220-9964/6/12/405

Virtual geographic environments (VGEs) have been regarded as an important new means of simulating, analyzing, and understanding complex geological processes. Plants and light are major components of the geographic environment. Light is a critical factor that affects ecological systems. In this study, we focused on simulating light transmission and distribution within a three-dimensional plant canopy model. A progressive refinement radiosity algorithm was applied to simulate the transmission and distribution of solar light within a detailed, three-dimensional (3D) loquat (Eriobotrya japonica Lindl.) canopy model. The canopy was described in three dimensions, and each organ surface was represented by a set of triangular facets. The form factors in radiosity were calculated using a hemi-cube algorithm. We developed a module for simulating the instantaneous light distribution within a virtual canopy, which was integrated into ParaTree. We simulated the distribution of photosynthetically active radiation (PAR) within a loquat canopy, and calculated the total PAR intercepted at the whole canopy scale, as well as the mean PAR interception per unit leaf area. The ParaTree-integrated radiosity model simulates the uncollided propagation of direct solar and diffuse sky light and the light-scattering effect of foliage. The PAR captured by the whole canopy based on the radiosity is approximately 9.4% greater than that obtained using ray tracing and TURTLE methods. The latter methods do not account for the scattering among leaves in the canopy in the study, and therefore, the difference might be due to the contribution of light scattering in the foliage. The simulation result is close to Myneni’s findings, in which the light scattering within a canopy is less than 10% of the incident PAR. Our method can be employed for visualizing and analyzing the spatial distribution of light within a canopy, and for estimating the PAR interception at the organ and canopy levels. It is useful for designing plant canopy architecture (e.g., fruit trees and plants in urban greening) and planting planning.