Tree Information Modeling: A Data Exchange Platform for Tree Design and Management

Abstract

Trees integrated into buildings and dense urban settings have become a trend in recent years worldwide. Without a thoughtful design, conflicts between green and gray infrastructures can take place in two aspects: (1) tree crown compete with living space above ground; (2) built underground environment, the other way round, affect tree’s health and security. Although various data about urban trees are collected by different professions for multiple purposes, the communication between them is still limited by unmatched scales and formats. To address this, tree information modeling (TIM) is proposed in this study, aiming at a standardized tree description system in a high level of detail (LoD). It serves as a platform to exchange data and share knowledge about tree growth models. From the perspective of architects and landscape designers, urban trees provide ecosystem services (ESS) not only through their overall biomass, shading, and cooling. They are also related to various branching forms and crown density, forming new layers of urban living space. So, detailed stem, branch and even root geometry is the key to interacting with humans, building structures and other facilities. It is illustrated in this paper how these detailed data are collected to initialize a TIM model with the help of multiple tools, how the topological geometry of stem and branches in TIM is interpreted into an L-system (a common syntax to describe tree geometries), allowing implementation of widely established tree simulations from other professions. In a vision, a TIM-assisted design workflow is framed, where trees are regularly monitored and simulated under boundary conditions to approach target parameters by design proposals.

Predicting resprouting of Platanus × hispanica following branch pruning by means of machine learning

“…4) After predicting the position of new shoots, our current model did not answer the ongoing growth of those shoots. It is possible to combine a L-system growth simulation ( Boudon et al., 2012 ) with the QSM (see Shu et al., 2022 ). In this way, our model can be integrated as a tool to interrupt a natural growth through branch pruning.…”

“…Following this trend, this study may offer hints about how pruning decisions could be made by the “digital gardener”. For this far vision, an open-source and uniform data platform about trees [e.g., tree information modeling ( Shu et al., 2022 )] is required.…”

“…Furthermore, different approaches have been developed to extract tree structure: skeleton abstraction following occupancy grids ( Bucksch et al., 2010 ; Sun et al., 2022 ); branch direction by eigenvectors of point patches or sections ( Bremer et al., 2013 ; Raumonen et al., 2013 ); skeleton as the Dijkstra’s shortest path from the tree base to ends ( Du et al., 2019 ; Li et al., 2022 , 2022; Wang et al., 2014 ); skeleton redrawn with searching steps ( Hackenberg et al., 2014 ); learning the reconstruction pattern through a neural network ( Liu et al., 2021 ). Overall, this abstracted information about tree architecture is called the quantitative structure model (QSM) ( Åkerblom et al., 2017 ; Shu et al., 2022 ). In this way, every segment of the tree stem or branch can be retrieved, containing its diameter, length, axial direction, and hierarchy in the whole branching structure, as well as the pointer to its parent and child segments.…”

A multilayered urban tree dataset of point clouds, quantitative structure and graph models

“…The topologies of the tree trunks and branches can be abstracted from a 3D point cloud geometry 13 . Shu et al 14 coined the term Tree Information Modelling (TIM) and summarized different approaches: For example, Raumonen introduced TreeQSM 15 to reconstruct tree topology, while PypeTree 16 reconstructes trunks and branches based on skeleton curves. SimpleTree 17 utilizes voxel-grid-based techniques to create cylindrical tree models, and AdTree 18 employs a skeleton-based approach, fitting cylinders…”

TreeML-Data; a multidisciplinary and multilayer urban tree dataset

“…The topologies of the tree trunks and branches can be abstracted from a 3D point cloud geometry 13 . Shu et al 14 coined the term Tree Information Modelling (TIM) and summarized different approaches: For example, Raumonen introduced TreeQSM 15 to reconstruct tree topology, while PypeTree 16 reconstructes trunks and branches based on skeleton curves. SimpleTree 17 utilizes voxel-grid-based techniques to create cylindrical tree models, and AdTree 18 employs a skeleton-based approach, fitting cylinders to point cloud models of individual trees.…”