Scholarlink

<p>The upper elevational limit of vegetation distribution serves as a critical boundary in alpine ecosystems (termed the vegetation line herein), exhibiting high sensitivity to changes in hydrothermal conditions; however, understanding of its spatial patterns and driving mechanisms remains limited. This study, set in the Southwest China alpine-gorge region, introduces a terrain-driven path tracking (TDPT) algorithm that delineates vegetation lines with high precision by tracing stable vegetation communities from non-vegetated cores. Using Sentinel-2 imagery from 2022 to 2024, we extracted 11,574 vegetation line points. Validation confirmed high accuracy (mean absolute error = 9.75&#xa0;m, below the 10&#xa0;m resolution of Sentinel-2; <i>R</i><sup>2</sup> = 0.998). Combining geographically weighted regression (GWR) and light gradient boosting machine (LightGBM), we found that vegetation line elevations (4120–5058&#xa0;m, 5th–95th percentiles) are mainly controlled by growing-season temperature, whereas precipitation effects vary spatially, promoting upward shifts in arid western regions but inhibiting them in humid southeastern areas. LightGBM further identified approximately 140,000 km<sup>2</sup> (22.8% of the study area) of persistent alpine barren zones, concentrated in the central-western high-elevation core. The TDPT algorithm and dual-model framework offer a precise, scalable approach for remote sensing of complex mountain ecological boundaries, advancing quantitative understanding of vegetation line dynamics and spatial heterogeneity, with significant implications for monitoring alpine ecosystems under climate change.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Terrain-driven path tracking: extraction of the vegetation line in the Southwest China alpine-gorge region and attribution analysis of its spatial variability

  • 期刊论文

  • 发表日期：

    2026年3月27日

查看全文
  • Zhongshuo Zhang,
  • Yan Zhang

摘要

The upper elevational limit of vegetation distribution serves as a critical boundary in alpine ecosystems (termed the vegetation line herein), exhibiting high sensitivity to changes in hydrothermal conditions; however, understanding of its spatial patterns and driving mechanisms remains limited. This study, set in the Southwest China alpine-gorge region, introduces a terrain-driven path tracking (TDPT) algorithm that delineates vegetation lines with high precision by tracing stable vegetation communities from non-vegetated cores. Using Sentinel-2 imagery from 2022 to 2024, we extracted 11,574 vegetation line points. Validation confirmed high accuracy (mean absolute error = 9.75 m, below the 10 m resolution of Sentinel-2; R2 = 0.998). Combining geographically weighted regression (GWR) and light gradient boosting machine (LightGBM), we found that vegetation line elevations (4120–5058 m, 5th–95th percentiles) are mainly controlled by growing-season temperature, whereas precipitation effects vary spatially, promoting upward shifts in arid western regions but inhibiting them in humid southeastern areas. LightGBM further identified approximately 140,000 km2 (22.8% of the study area) of persistent alpine barren zones, concentrated in the central-western high-elevation core. The TDPT algorithm and dual-model framework offer a precise, scalable approach for remote sensing of complex mountain ecological boundaries, advancing quantitative understanding of vegetation line dynamics and spatial heterogeneity, with significant implications for monitoring alpine ecosystems under climate change.