关键词: 钱江源-百山祖国家公园候选区, 亚热带森林, 激光雷达, 冠层结构, 高光谱, 植被指数, 人为干扰

Abstract

Aim: National parks prioritize the conservation of their unique and representative vegetation ecosystems. Monitoring these ecosystems is crucial for evaluating conservation effectiveness and formulating sustainable management strategies. This study utilized remote sensing data to characterize canopy characteristics in Qianjiangyuan section of the Qianjiangyuan-Baishanzu National Park candidate area (Qianjiangyuan section). By establishing direct links between canopy features and disturbance regimes, we aimed to evaluate the effectiveness of forest biodiversity protection. 

Method: We analyzed airborne LiDAR (light detection and ranging) and hyperspectral data for 21,820 100 m × 100 m pixels and naturalness survey data for 6,977 subcompartments within Qianjiangyuan section. The three-dimensional canopy structural characteristics were derived from airborne LiDAR data, while vegetation indices were calculated from hyperspectral data. These canopy characteristics were then correlated with the naturalness of forest subcompartments to assessed differences across various naturalness levels. 

Results: Our findings revealed that: (1) Canopy characteristics within Qianjiangyuan section exhibited a wide range of variability, with considerable differences across regions of the park. (2) Canopy structural and spectral characteristics varied significantly between areas with varying degrees of anthropogenic disturbance, reflecting the success of conservation efforts. Specifically, heavily disturbed secondary and plantation forests exhibited the lowest canopy height, leaf area index, vertical structural complexity, and light resource utilization efficiency, with a higher proportion of shorter and sparser vegetation, increased stress levels, and higher leaf senescence. In contrast, moderately disturbed secondary deciduous broad-leaved forests and lightly or undisturbed old-growth evergreen broad-leaved forests demonstrated elevated canopy heights, leaf area index, and vertical structural complexity. However, old-growth forests experienced lower stress levels, higher light use efficiency, and reduced leaf senescence compared to moderately disturbed forests. 

Conclusion: The canopy characteristics in Qianjiangyuan section vary widely depending on the level of anthropogenic disturbance, with such disturbances negatively impacting canopy attributes. Our findings indicate that airborne LiDAR and hyperspectral imaging are effective in quantifying the three-dimensional structure and spectral attributes of the canopy, offering valuable insights into conservation effectiveness. Our study serves as a scientific foundation for the protection and management of subtropical forests within Qianjiangyuan section. Furthermore, it highlights the potential of airborne remote sensing technologies in monitoring changes in canopy characteristics to evaluate conservation efficacy, thereby supporting the development of informed conservation strategies and sustainable management practices.

Key words: Qianjiangyuan-Baishanzu National Park candidate area, subtropical forest, LiDAR, canopy structure, hyperspectral, vegetation indices, anthropogenic disturbance