| As the largest terrestrial ecosystem,the carbon sink function of forests plays an important role in combating climate change.Accurate estimation of forest carbon stocks is of great practical significance for the assessment of environmental benefits of forest carbon sinks and carbon trading.With the rapid development of technology,compared with traditional remote sensing methods,UAV remote sensing technology has become one of the important earth observation methods on small and medium-scale scales with its advantages such as high precision,low cost and flexibility,and provides important technical support for scientific understanding and research of forest ecosystems.In this study,we took Arborvitae in typical mixed forest stands in Mount Taishan as the research object,and UAV remote sensing technology was used as the main data acquisition method to explore the quantitative relationship between canopy volume and tree carbon storage,constructed a carbon storage remote sensing inversion estimation model and automatic extraction tool,and used the above methods and tools to extract carbon storage and study the heat of carbon distribution in the study area.The main research results are as follows:(1)Canopy space model study.In view of the characteristics of the tree canopy,the concept of canopy three-dimensional spatial carbon density(CDCS)(the density relationship between the aboveground space volume of the canopy and the carbon content of the whole tree)and the canopy quasivoxylation projection volume(CQVPV)(canopy space volume composed of canopy DSM as the top surface and canopy plane projection area as the bottom surface)and quasi-canopy occupation volume(QCOV)(subtracting the high part of the subbranch space volume on the basis of CQVPV)were proposed on the basis of the canopy "quasivoxelization" integral geometric volume model;(2)Research on aerial survey methods of forestry UAVs.On the basis of the traditional geostationary flight orthophoto mode,the oblique photography mode is integrated,and the UAV forestry aerial survey method of "photogrammetry 3D(nine-directions)"(PG-3D-N)under the ground-imitation condition is proposed,which improves the accuracy and integrity of UAV remote sensing on canopy spatial structure and understory terrain;(3)Target tree species identification research.According to the characteristics of northern trees,the Ex G-Ex R vegetation index and height threshold were fused,and the forest characteristics were extracted from the horizontal and vertical dimensions,which avoided the classification errors caused by similar feature interference and splicing errors in the process of feature classification,realized the accurate identification of the target tree species,and improved the accuracy of canopy height model(CHM)extraction;(4)Research on remote sensing inversion estimation model of carbon storage.Based on the four groups of canopy volume parameters of density,area,height and volume extracted by the target tree species CHM,a regression model between the target tree species carbon storage and CQVPV was constructed through multiple linear regression,and based on this model,the carbon storage of target tree species in the study area was estimated and spatial distribution was studied.Under the background of emission reduction and sink increase,exploring the new metho d of UAV remote sensing technology in the fine assessment of forest carbon sink has importan t practical significance for forestry management and production.Compared with the traditiona l remote sensing carbon storage extraction method based on crown and tree height measureme nt factors,this study has simple application,high efficiency and ideal accuracy(the highest R~2 is 0.973),which has certain promotion significance in forest resources investigation and fores try carbon asset assessment,and has good practical guiding value for the fine assessment of fo rest carbon sink at small and medium-scale forests. |
PDF Full Text Request