农业科学研究2026，Vol.47Issue(1)：20-30,11.DOI:10.13907/j.cnki.nykxyj.2026.01.003
基于BRT模型的黄河流域NDVI时空变化及其驱动力分析
Spatiotemporal dynamics of NDVI and their driving forces in the Yellow River basin based on a BRT model
摘要
Abstract
The Yellow River basin serves as a critical ecological security barrier in China but remains threatened by vegetation degradation and soil erosion under the combined influences of climate change,geographical conditions,and human activities.To systematically assess trends in vegetation coverage and their driving forces,this study obtained the NDVI for 2000-2020 from the MOD13Q1 product on the Google Earth Engine(GEE)platform,analyzed spatiotemporal trends across 11 ecological functional zones using Theil-Sen slope estimation and the Mann-Kendall test for abrupt changes,and quantified the contributions of meteorological,topographic,geomorphological,and demographic factors with a Boosted Regression Tree(BRT)model.Results show that:① NDVI exhibited an overall increasing trend from 2000 to 2020,with higher values in the southeast and lower values in the northwest.The annual median NDVI ranged from-0.12 to 0.61,with a basin-wide mean of 0.28.Areas with NDVI<0 were concentrated in regions experiencing overgrazing and frequent drought,indicating a risk of vegetation degradation.② Each ecological functional zone displayed one or two significant change points,all occurring before 2018.Across the basin,areas with improving NDVI accounted for 85.22%,stable areas for 9.08%,and degrading areas for 5.70%.③ Climatic factors were the dominant drivers of NDVI change,with precipitation contributing more than temperature;topographic factors(elevation,slope,aspect)and population density ranked next.These findings provide essential data support and scientific evidence for ecological conservation and vegetation management in the Yellow River basin.关键词
黄河流域/归一化植被指数(NDVI)/提升回归树分析/GEE云平台Key words
Yellow River basin/normalized difference vegetation index(NDVI)/Boosted Regression Tree(BRT)analysis/Google Earth Engine分类
资源环境引用本文复制引用
张新,田佳,王逸飞,马乐鹏..基于BRT模型的黄河流域NDVI时空变化及其驱动力分析[J].农业科学研究,2026,47(1):20-30,11.基金项目
宁夏自然科学基金项目(2024AAC03089) （2024AAC03089）
宁夏重点研发计划项目(2023BEG02039) （2023BEG02039）
大学生创新创业项目(G202310749034) （G202310749034）
