Scholarlink

<p>With advances in remote sensing technology, vegetation indices derived from optical properties have been widely employed to monitor regional vegetation dynamics. However, the limited historical archive of remote sensing data imposes temporal constraints on vegetation monitoring. Tree-ring data, serving as a natural proxy with high temporal resolution, provide the opportunity to extend vegetation index records into the pre-satellite era. Nevertheless, most vegetation index reconstructions based on tree rings have focused on forested, mountainous regions and primarily on the Normalized Difference Vegetation Index (NDVI). The effectiveness of long-term reconstructions of tree-ring-based vegetation indices in arid desert conditions still remains a gap. In this study, we systematically evaluated the relationships and reconstruction performance between ring-width index (RWI) chronologies and six classic vegetation indices (NDVI, Enhanced Vegetation Index (EVI), Ratio Vegetation Index (RVI), Perpendicular Vegetation Index (PVI), Global Environment Monitoring Index (GEMI), and Soil Adjusted Vegetation Index (SAVI)) from two sites of <i>Zygophyllum xanthoxylon</i> in the Tengger Desert. The PVI outperformed the other indices, showing the strongest correlation with RWI at both the interannual (<i>R</i> = 0.48) and monthly scales, reaching up to <i>R</i> = 0.67 in the optimal month. EVI and SAVI also showed good sensitivity at the monthly scale, whereas NDVI and RVI showed weaker correlations in some months. In addition, the reconstructed PVI series from 1889 to 2024 reveals a gradual increasing trend in vegetation cover at the study sites, with an average rise of 0.0003 per decade. Four major low-vegetation periods were observed during 1889–1899, 1913–1932, 1935–1945 and 1946–1962, while three periods of high vegetation were identified during 1899–1909, 1962–1972 and 1979–1995. Cyclical fluctuations with a dominant 4-year periodicity were also noticed. These results demonstrate the feasibility of reconstructing vegetation indices from shrub tree rings in desert environments, providing a novel approach and valuable proxy data for long-term vegetation monitoring and tracking the dynamic processes of desertification in dryland ecosystems.</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

Vegetation index reconstruction and suitability evaluation in arid desert areas based on shrub tree rings

  • 期刊论文

  • 发表日期：

    2026年6月18日

查看全文
  • Yu Ren,
  • Shengchun Xiao,
  • Xiaomei Peng,
  • Aijun Ding,
  • Quanyan Tian,
  • Jiakang Wang

摘要

With advances in remote sensing technology, vegetation indices derived from optical properties have been widely employed to monitor regional vegetation dynamics. However, the limited historical archive of remote sensing data imposes temporal constraints on vegetation monitoring. Tree-ring data, serving as a natural proxy with high temporal resolution, provide the opportunity to extend vegetation index records into the pre-satellite era. Nevertheless, most vegetation index reconstructions based on tree rings have focused on forested, mountainous regions and primarily on the Normalized Difference Vegetation Index (NDVI). The effectiveness of long-term reconstructions of tree-ring-based vegetation indices in arid desert conditions still remains a gap. In this study, we systematically evaluated the relationships and reconstruction performance between ring-width index (RWI) chronologies and six classic vegetation indices (NDVI, Enhanced Vegetation Index (EVI), Ratio Vegetation Index (RVI), Perpendicular Vegetation Index (PVI), Global Environment Monitoring Index (GEMI), and Soil Adjusted Vegetation Index (SAVI)) from two sites of Zygophyllum xanthoxylon in the Tengger Desert. The PVI outperformed the other indices, showing the strongest correlation with RWI at both the interannual (R = 0.48) and monthly scales, reaching up to R = 0.67 in the optimal month. EVI and SAVI also showed good sensitivity at the monthly scale, whereas NDVI and RVI showed weaker correlations in some months. In addition, the reconstructed PVI series from 1889 to 2024 reveals a gradual increasing trend in vegetation cover at the study sites, with an average rise of 0.0003 per decade. Four major low-vegetation periods were observed during 1889–1899, 1913–1932, 1935–1945 and 1946–1962, while three periods of high vegetation were identified during 1899–1909, 1962–1972 and 1979–1995. Cyclical fluctuations with a dominant 4-year periodicity were also noticed. These results demonstrate the feasibility of reconstructing vegetation indices from shrub tree rings in desert environments, providing a novel approach and valuable proxy data for long-term vegetation monitoring and tracking the dynamic processes of desertification in dryland ecosystems.