img

官方微信

高级检索

中国沙漠 ›› 2025, Vol. 45 ›› Issue (5): 78-91.DOI: 10.7522/j.issn.1000-694X.2025.00030

• • 上一篇    下一篇

19862023年吕梁山区归一化植被指数(NDVI)动态及驱动力

徐德华1(), 刘强2,3, 李端1, 李继彦1()   

  1. 1.太原师范学院 地理科学学院/汾河流域地表过程与资源生态安全山西省重点实验室,山西 晋中 030619
    2.天水师范学院 资源与环境工程学院,甘肃 天水 741000
    3.中国科学院水利部水土保持研究所,陕西 杨凌 712100
  • 收稿日期:2025-01-20 修回日期:2025-03-22 出版日期:2025-09-20 发布日期:2025-09-27
  • 通讯作者: 李继彦
  • 作者简介:徐德华(1998—),男,甘肃景泰人,硕士研究生,主要从事风沙地貌、环境遥感研究。E-mail: zgxudehua@163.com
  • 基金资助:
    山西省基础研究计划项目(202103021224304);山西省基础研究计划项目(202303021212257)

Changes and driving forces of NDVI in the Lvliang Mountain area from 1986 to 2023

Dehua Xu1(), Qiang Liu2,3, Duan Li1, Jiyan Li1()   

  1. 1.School of Geography Science / Shanxi Key Laboratory of Earth Surface Processes and Resource Ecology Security in Fenhe River Basin,Taiyuan Normal University,Jinzhong 030619,Shanxi,China
    2.College of Resources and Environmental Engineering,Tianshui Normal University,Tianshui 741000,Gansu,China
    3.Institute of Soil and Water Conservation,Chinese Academy of Sciences and Ministry of Water Resources,Yangling 712100,Shaanxi,China
  • Received:2025-01-20 Revised:2025-03-22 Online:2025-09-20 Published:2025-09-27
  • Contact: Jiyan Li

摘要:

吕梁山区位于黄河流域山西段的西部,是典型的生态环境脆弱区域。已有研究的时间跨度较短导致对该区域植被覆盖的长期变化趋势及其驱动力分析仍显不足。本研究利用GEE平台提供的Landsat系列卫星影像,计算并研究1986—2023年吕梁山区归一化植被指数(NDVI)时空变化,并利用地理探测器分析驱动因素。结果表明:(1)1986—2000年吕梁山区NDVI年均值呈不显著波动下降趋势,而2001—2023年年均值显著上升。1986—2023年NDVI整体向好,显著增加区域占94.92%。增速最大区域主要位于吕梁山区的忻州、吕梁、临汾市中部和太原市南部,而吕梁山区中东部、南部边缘因人类活动增强NDVI显著减少。(2)各因子对NDVI变化的解释力排序为干燥度>年实际蒸发量>年降水量>年均土壤湿度>高程>坡度>年均风速>年均气温>年均土壤温度>夜间灯光指数。其中,干燥度与年实际蒸发量q值最大,解释力均在35%以上,年降水量、年均土壤湿度、高程的解释力均大于20%。(3)1986—2023年驱动因子呈现气候因子解释力显著或不显著下降、人为因子与坡度解释力显著增强的趋势。1990—2023年土地利用变化表明吕梁山区耕地面积显著减少、林地面积显著增加,说明了人类活动在吕梁山区NDVI变化中的重要影响。

关键词: 吕梁山区, NDVI, GEE, 地理探测器

Abstract:

The Lvliang Mountain area, located in the western part of the Shanxi province within the Yellow River Basin, represents a typical region characterized by fragile ecological conditions. While existing studies have focused on relatively short-term periods, the analysis of the long-term trends and driving forces of vegetation cover in the region remains insufficient. This study utilizes Landsat satellite imagery provided by the GEE platform to calculate and analyze the spatio-temporal variation of NDVI in the Lvliang Mountain area from 1986 to 2023. Additionally, the Geodetector model is employed to identify the key driving factors. The findings reveal the following insights: (1) From 1986 to 2000, the annual mean NDVI of vegetation in the Lvliang Mountain area exhibited a non-significant declining trend, whereas from 2001 to 2023, it showed a significant increasing trend. Overall, during the period from 1986 to 2023, vegetation NDVI demonstrated an improving trend, with areas of significant increase accounting for 94.92% of the study area. The regions with the most rapid NDVI growth were primarily concentrated in the study area of central Xinzhou, Lvliang, Linfen, and southern Taiyuan. In contrast, significant NDVI declines were observed along the central-eastern and southern margins of the study area, mainly due to intensified human activities. (2) The analysis of the driving factors reveals that their explanatory capacity, in descending order, is as follows: drought index > actual evaporation > precipitation > soil moisture > elevation > slope > wind speed > temperature > soil temperature > nighttime light. Among these, the drought index and actual evaporation have the highest q-values, each contributing more than 35% to the explanation of NDVI changes. Precipitation, soil moisture, and elevation each account for more than 20% of the explanatory capacity. (3) From 1986 to 2023, the explanatory capacity of climatic factors for vegetation NDVI change exhibited a significant or non-significant decreasing trend, while the explanatory capacity of anthropogenic factors and slope showed a significant increase. Meanwhile, land use changes in the study area from 1990 to 2023 indicated a significant reduction in cropland area and a substantial increase in forested land, highlighting the critical role of human activities in influencing NDVI variations in the Lvliang Mountain area.

Key words: Lvliang Mountain area, NDVI, GEE, Geodetector model

中图分类号: