Experimental Study on Tree Shape Identification and Interaction Among Trees

Abstract

Forests play a crucial role in mitigating global warming and maintaining climate balance, necessitating a comprehensive understanding of forest dynamics. However, many optical characteristics related to forests remain poorly understood, particularly models for determining forest parameters such as tree species, shape, and inter-tree distances. Although several models exist for estimating leaf area index and photosynthetically active radiation, fewer models address structural forest parameters. This research aims to develop a model using visible and near-infrared radiometer data from Earth observation satellites to estimate forest parameters, including tree species, shape, height, and spacing. Results show that as the distance between trees increases, the impact of multiple reflections decreases significantly. Elliptical tree shapes exhibit approximately three times higher multiple reflection effects compared to conical shapes, indicating potential for distinguishing tree shapes through radiometric data. For canopy shapes, shorter, thicker trees experience more significant reflection effects than taller, thinner trees, suggesting the feasibility of estimating tree height. Overall, the impact of multiple reflections ranges from a few to 10% of TOA radiance, necessitating its consideration when calculating forest reflectance to ensure accurate forest parameter retrieval from satellite observations.