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

Introduction: Resprouting is a crucial survival strategy following the loss of branches, being it by natural events or artificially by pruning. The resprouting prediction on a physiological basis is a highly complex approach. However, trained gardeners try to predict a tree's resprouting after pruning purely based on their empirical knowledge. In this study, we explore how far such predictions can also be made by machine learning. Methods: Table-topped annually pruned Platanus × hispanica trees at a nursery were LiDAR-scanned for two consecutive years. Topological structures for these trees were abstracted by cylinder fitting. Then, new shoots and trimmed branches were labelled on corresponding cylinders. Binary and multiclass classification models were tested for predicting the location and number of new sprouts. Results: The accuracy for predicting whether having or not new shoots on each cylinder reaches 90.8% with the LGBMClassifier, the balanced accuracy is 80.3%. The accuracy for predicting the exact numbers of new shoots with the GaussianNB model is 82.1%, but its balanced accuracy is reduced to 42.9%. Discussion: The results were validated with a separate dataset, proving the feasibility of resprouting prediction after pruning using this approach. Different tree species, tree forms, and other variables should be addressed in further research.

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

The significance of urban trees in promoting human health and well-being has been amplified by urbanization and the climate change effects. Simultaneously, advancements in remote sensing techniques have enhanced the opportunities for studying urban trees. The TreeML-Data has been compiled to support these efforts. It consists of labelled point clouds of 40 scanning projects of streets in Munich, 3,755 leaf-off (scans in winter) point clouds of individual trees, quantitative structure models (QSM), tree structure measurements, and tree graph structure models of these trees. The dataset offers valuable data for generating and evaluating models in various scientific disciplines, which include remote sensing, computer vision, machine learning, urban forestry, urban ecosystem, green architecture, and graph analysis. To ensure its quality, the tree structure measurements and QSM have been crosschecked. For instance, the tree diameter at breast height (DBH) in the sample dataset exhibits a deviation of approximately 1.5 cm (4.3%) when compared to manual measurements. In conclusion, the quality checks confirm its reliability for subsequent studies when compared to manual measurements.

A circuit analogy based girth growth model for living architecture design.

Architecture with and from living trees (Baubotanik) is a promising approach to sustainable, climate-adapted construction. Shaping and grafting allows one to create resilient structures that combine the ecological performance and aesthetics of trees with the functions of buildings. In order to design and engineer such living structures, it is necessary to predict the growth of different tree segments, especially when trunks, branches or roots are bent and jointed into a complex inosculated network. To address this, we have developed a tool to forecast the relative girth growth of different segments in such structures based on topological skeletons, the pipe model theory and circuit analogy. We have validated our results with a set of (scaled) photographs of inosculated tree structures of the so-called 'Tree Circus', covering over 80 years of their growth. Our model has proven to predict the relative girth growth with sufficient accuracy for conceptual design purposes. So far, it does not allow the simulation of absolute growth in circumference over the course of time that is necessary to predict quantitative technical aspects, such as mechanical performance at a given time. We conclude by briefly outlining how this could be addressed in future research.

Conjoining Trees for the Provision of Living Architecture in Future Cities: A Long-Term Inosculation Study.

Faced with the environmental challenges posed by climate change, architects are creating nature-based solutions for urban areas, such as transforming living trees into artificial architectural structures. In this study, we have analyzed stem pairs of five tree species conjoined for more than eight years by measuring the stem diameters below and above the resulting inosculation and by calculating the respective diameter ratio. Our statistical analyses reveal that Platanus × hispanica and Salix alba stems do not differ significantly in diameter below inosculation. However, in contrast to P. × hispanica, the diameters of the conjoined stems above inosculation differ significantly in S. alba. We provide a binary decision tree based on diameter comparisons above and below inosculation as a straightforward tool for identifying the likelihood of full inosculation with water exchange. Moreover, we have compared branch junctions and inosculations by means of anatomical analyses, micro-computed tomography, and 3D reconstructions showing similarities in the formation of common annual rings that increase the capacity for water exchange. Due to the highly irregular cell arrangement in the center of the inosculations, cells cannot be assigned clearly to either of the stems. In contrast, cells in the center of branch junctions can always be attributed to one of the branches.

Representing living architecture through skeleton reconstruction from point clouds.

Living architecture, changing in structure with annual growth, requires precise, regular characterisation. However, its geometric irregularity and topological complexity make documentation using traditional methods difficult and presents challenges in creating useful models for mechanical and physiological analyses. Two kinds of living architecture are examined: historic living root bridges grown in Meghalaya, India, and contemporary 'Baubotanik' structures designed and grown in Germany. These structures exhibit common features, in particular network-like structures of varying complexity that result from inosculations between shoots or roots. As an answer to this modelling challenge, we present the first extensive documentation of living architecture using photogrammetry and a subsequent skeleton extraction workflow that solves two problems related to the anastomoses and varying nearby elements specific to living architecture. Photogrammetry was used as a low cost method, supplying detailed point clouds of the structures' visible surfaces. A workflow based on voxel-thinning (using deletion templates and adjusted p-simplicity criteria) provides efficient, accurate skeletons. A volume reconstruction method is derived from the thinning process. The workflow is assessed on seven characteristics beneficial in representing living architecture in comparison with alternative skeleton extraction methods. The resulting models are ready for use in analytical tools, necessary for functional, responsible design.

Living bridges using aerial roots of ficus elastica - an interdisciplinary perspective.

Here we report on a pilot study of the Living Root Bridges (LRBs) in the Indian State Meghalaya, which are grown with aerial roots of Ficus elastica, a facultative hemiepiphyte developing abundant aerial roots. Locals use these aerial roots to build living bridges, which strengthen themselves over time due to adaptive secondary growth and their capacity to form a mechanically stable structure via inosculations. An extensive inventory of LRBs in Meghalaya including data of location, altitude, approximate age and bridge length was performed in field studies. Root morphology was characterised by measurements of cross-sectional area and shape-related parameters and analysed in relation to the orientation of the roots. LRBs are found to occur mainly in the mountainous limestone rainforests where F. elastica may be native or traditionally cultivated. They cover an altitude range of 57-1211 m a.m.s.l. and display a length of 2 to 52.7 m. Some bridges are several hundreds of years old. Horizontally and vertically trained roots differ significantly in shape and cross-sectional area when approximately even-aged roots are compared. The results are discussed from an interdisciplinary perspective, considering the adaptive traits in the natural life cycle of F. elastica and possible applications in living architecture (Baubotanik).