Unrstanding Growth in Apple Orchards using FSPM Lidar-Derived Measurements

By Dick Kuijper

Abstract
Functional-Structural Plant (FSP) models are useful tools for understanding the complex dynamics of tree growth. Integrating accurate tree trait data into these models can significantly enhance their predictive power. The role Terrestrial Laser Scanned (TLS) time series can have in the integration of tree trait data is not well understood. This thesis evaluates the extent to which TLS time series derived tree traits can be used as inputs or as complementary data in FSP models for cultivated apple trees. Tree traits were derived from TLS time series of a cultivated apple tree orchard, including branch angle, length, and vertex count. These traits were assessed for direct and indirect use in FSP models. The study identified branch angle as a direct input for FSP models, despite its significant variability over time. Dormant winter scans and summer scans returned stationary patterns, showing significant similarity in branch length and vertex count across scenarios, which underpins the potential for TLS time series as indirect input through stochastic modelling to refine FSP model functionality.

The study demonstrates the viability of using TLS time series derived tree traits to discern patterns in apple tree growth, enhancing FSP models. However, the dynamic nature of tree growth and the influence of various factors limit the direct comparison of TLS-derived parameters with synthetic parameters.

Keywords: Terrestrial Laser Scanning; Functional-Structural Plant Models; Tree architecture; Apple Trees; Tree Trait Dynamics.