Centre de documentation
scientifique

Gaining insight into the allometric scaling of trees by utilizing 3d reconstructed tree models - a SimpleForest study / Jan Hackenberg (2022)  

Public Titre : Gaining insight into the allometric scaling of trees by utilizing 3d reconstructed tree models - a SimpleForest study Type de document : Article/Communication Auteurs : Jan Hackenberg , Auteur ; Mathias I. Disney, Auteur ; Jean-Daniel Bontemps , Auteur Editeur : BioRxiv Année de publication : 2022 Projets : 1-Pas de projet / Importance : 18 p. Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] données allométriques [Termes IGN] données lidar [Termes IGN] données localisées 3D [Termes IGN] modèle de croissance végétale [Termes IGN] modélisation de la forêt [Termes IGN] volume en bois [Vedettes matières IGN] Foresterie Résumé : (auteur) Forestry utilizes volume predictor functions utilizing as input the diameter at breast height. Some of those functions take the power form Y = a ∗ Xb. In fact this function is fundamental for the biology field of allometric scaling theories founded round about a century ago. The theory describes the relationships between organs/body parts and the complete body of organisms. With digital methods we can generate 3d forest point clouds non destructively in short time frames. SimpleForest is one free available tool which generates fully automated ground and tree models from high resoluted forest plots. Generated topological ordered cylinder models are called commonly QSMs. We use SimpleForest QSMs an build a function which estimates the total supported wood volume at any given point of the tree. As input we use the supported soft wood volume for those query points. Instead of measuring directly the soft wood volume we use as a proxy the number of supported twigs. We argue with the pipe model theory for the correctness of the proxy. We can use the named relationship to also filter our QSMs made of an open data set of tree clouds. The filter corrects overestimated radii. And we compare the corrected QSM volume against the harvested reference data for 66 felled trees. We also found QSM data of TreeQSM, a competitive and broadly accepted QSM modeling tool. Our RMSE was less than 40% of the tree QSM RMSE. And for other error measures, the r2adj. and the CCC, the relative improvement looked even better with 27% and 21% respectively. We consider this manuscript as highly impactful because of the magnitude of quality improvement we do. The relation between soft volume and total volume distributions seems to be really strong and tree data can easily also be used as example data for the generic field of allometric scaling. Numéro de notice : P2022-008 Affiliation des auteurs : LIF+Ext (2020- ) Thématique : FORET/MATHEMATIQUE Nature : Preprint nature-HAL : Préprint DOI : 10.1101/2022.05.05.490069 Date de publication en ligne : 05/05/2022 En ligne : https://doi.org/10.1101/2022.05.05.490069 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101945