关键词: 黄河三角洲, 光合生理, 三维空间分布, 辐射传输模型, 无人机激光雷达, 无人机高光谱

Abstract: The threedimensional spatial heterogeneity of leaf chlorophyll content (LCC) at the individual tree canopy is critical for evaluating the physiological status and light use efficiency of trees, and for analyzing interspecific interactions in mixed forests. However, previous studies often focus on the stand scale and overlook the vertical variation of LCC within the canopy, which limits in-depth understanding of ecological processes in mixed forests. To address this gap, we conducted plot surveys and leaf biochemical analyses in artificial mixed forests in the Yellow River Delta. Unmanned aerial vehicle (UAV)-borne LiDAR (UAV-LiDAR) was employed for individual tree segmentation and 3D spatial partitioning, while hyperspectral imagery (UAV-HSI) data were processed using a coupled framework of PROSPECT-D leaf optical model and LESS canopy radiative transfer model to classify tree species and retrieve individual canopy LCC. The results showed that: (1) The fusion of UAV-LiDAR and UAV-HSI data enabled effective discrimination of different tree species in the mixed forests (Overall Accuracy=91.7%, Kappa=0.887), and the coupled PROSPECT-D-LESS model achieved accurate retrieval of individual canopy LCC (R²=0.809, RMSE=4.973 μg·cm^-2). (2) Species mixtures were associated with significant shifts in LCC. Compared to monocultures, the LCC of Robinia pseudoacacia increased by 13.6% (P<0.05) in R. pseudoacacia-Fraxinus velutina mixtures but decreased by 39.4% (P<0.05) in R. pseudoacacia-Ulmus pumila mixtures. (3) In the horizontal direction, canopy LCC was generally higher on the eastern and southern sides than on the western and northern sides. Vertically, LCC increased with the height of R. pseudoacacia and F. velutina monocultures. The trend was opposite in mixed stands. Notably, the vertical LCC gradient of R. pseudoacacia inverted when mixed with U. pumila, while U. pumila exhibited a consistent decreasing trend with height in both monoculture and mixed settings. Our results can provide an effective method for the inversion of individual tree canopy LCC and the analysis of its spatial variations in mixed forests, and offer a scientific basis for evaluating tree physiological performance in mixed forests and the guidance of regional ecological restoration strategies.

Key words: Yellow River Delta, photosynthetic physiology, 3D spatial distribution, radiative transfer model, UAV-HSI, UAV-LiDAR