Trees are spread worldwide,as the watchmen that experience the intricate ecological effects caused by various environmental factors.In order to better understand such effects,it is preferential to achieve finely and f...Trees are spread worldwide,as the watchmen that experience the intricate ecological effects caused by various environmental factors.In order to better understand such effects,it is preferential to achieve finely and fully mapped global trees and their environments.For this task,aerial and satellite-based remote sensing(RS)methods have been developed.However,a critical branch regarding the apparent forms of trees has significantly fallen behind due to the technical deficiency found within their globalscale surveying methods.Now,terrestrial laser scanning(TLS),a state-of-the-art RS technology,is useful for the in situ three-dimensional(3D)mapping of trees and their environments.Thus,we proposed co-developing an international TLS network as a macroscale ecotechnology to increase the 3D ecological understanding of global trees.First,we generated the system architecture and tested the available RS models to deepen its ground stakes.Then,we verified the ecotechnology regarding the identification of its theoretical feasibility,a review of its technical preparations,and a case testification based on a prototype we designed.Next,we conducted its functional prospects by previewing its scientific and technical potentials and its functional extensibility.Finally,we summarized its technical and scientific challenges,which can be used as the cutting points to promote the improvement of this technology in future studies.Overall,with the implication of establishing a novel cornerstone-sense ecotechnology,the co-development of an international TLS network can revolutionize the 3D ecological understanding of global trees and create new fields of research from 3D global tree structural ecology to 3D macroecology.展开更多
Accurate acquisition of forest spatial competition and tree 3D structural phenotype parameters is crucial for exploring tree-environment interactions.However,due to the occlusion between tree crowns,current UAV-based ...Accurate acquisition of forest spatial competition and tree 3D structural phenotype parameters is crucial for exploring tree-environment interactions.However,due to the occlusion between tree crowns,current UAV-based and ground-based LiDAR struggles to capture complete crown information in dense stands,making parameter extraction challenging such as maximum crown width height(HMCW).This study proposes a canopy spatial relationship-based method for constructing forest spatial structure units and employs five ensemble learning techniques to train 11 machine learning model combinations.By coupling spatial competition with phenotype parameters,the study identifies the optimal fitting model for HMCW of Chinese fir.The results demonstrate that the constructed spatial structure units align closely with existing research while addressing issues of incorrectly selected or omitted neighboring trees.Among the 10,191 trained HMCW models,the Bagging model integrating XGBoost,Random Forest(RF),Support Vector Regression(SVR),Gradient Boosting(GB),and Ridge exhibited the best performance.Compared to the best single model(RF),the Bagging model achieved improved accuracy(R^(2)=0.8346,representing a 1.6%improvement;RMSE=1.4042,reduced by 6.66%;EVS=0.8389;MAE=0.9129;MAPE=0.0508;and MedAE=0.5076,with corresponding improvements of 1.63%,1.49%,0.1%,and 7.06%,respectively).This study provides a viable solution for modeling HMCW in all species with similar structural characteristics and offers a method for extracting other hard-to-measure parameters.The refined spatial structure units better link 3D structural phenotypes with environmental factors.This approach aids in canopy morphology simulation and forest management research.展开更多
基金The work was financially supported by the National Key Research and Development Program of China(No.2022YFE0112700)the National Natural Science Foundation of China(No.32171782 and 31870531).
文摘Trees are spread worldwide,as the watchmen that experience the intricate ecological effects caused by various environmental factors.In order to better understand such effects,it is preferential to achieve finely and fully mapped global trees and their environments.For this task,aerial and satellite-based remote sensing(RS)methods have been developed.However,a critical branch regarding the apparent forms of trees has significantly fallen behind due to the technical deficiency found within their globalscale surveying methods.Now,terrestrial laser scanning(TLS),a state-of-the-art RS technology,is useful for the in situ three-dimensional(3D)mapping of trees and their environments.Thus,we proposed co-developing an international TLS network as a macroscale ecotechnology to increase the 3D ecological understanding of global trees.First,we generated the system architecture and tested the available RS models to deepen its ground stakes.Then,we verified the ecotechnology regarding the identification of its theoretical feasibility,a review of its technical preparations,and a case testification based on a prototype we designed.Next,we conducted its functional prospects by previewing its scientific and technical potentials and its functional extensibility.Finally,we summarized its technical and scientific challenges,which can be used as the cutting points to promote the improvement of this technology in future studies.Overall,with the implication of establishing a novel cornerstone-sense ecotechnology,the co-development of an international TLS network can revolutionize the 3D ecological understanding of global trees and create new fields of research from 3D global tree structural ecology to 3D macroecology.
基金funded by Fundamental Research Funds of CAF(CAFYBB2023PA003)Science and Technology Innovation 2030-Major Projects(2023ZD0406103)National Natural Science Foundation of China(32271877).
文摘Accurate acquisition of forest spatial competition and tree 3D structural phenotype parameters is crucial for exploring tree-environment interactions.However,due to the occlusion between tree crowns,current UAV-based and ground-based LiDAR struggles to capture complete crown information in dense stands,making parameter extraction challenging such as maximum crown width height(HMCW).This study proposes a canopy spatial relationship-based method for constructing forest spatial structure units and employs five ensemble learning techniques to train 11 machine learning model combinations.By coupling spatial competition with phenotype parameters,the study identifies the optimal fitting model for HMCW of Chinese fir.The results demonstrate that the constructed spatial structure units align closely with existing research while addressing issues of incorrectly selected or omitted neighboring trees.Among the 10,191 trained HMCW models,the Bagging model integrating XGBoost,Random Forest(RF),Support Vector Regression(SVR),Gradient Boosting(GB),and Ridge exhibited the best performance.Compared to the best single model(RF),the Bagging model achieved improved accuracy(R^(2)=0.8346,representing a 1.6%improvement;RMSE=1.4042,reduced by 6.66%;EVS=0.8389;MAE=0.9129;MAPE=0.0508;and MedAE=0.5076,with corresponding improvements of 1.63%,1.49%,0.1%,and 7.06%,respectively).This study provides a viable solution for modeling HMCW in all species with similar structural characteristics and offers a method for extracting other hard-to-measure parameters.The refined spatial structure units better link 3D structural phenotypes with environmental factors.This approach aids in canopy morphology simulation and forest management research.
