中国草地学报2025，Vol.47Issue(3)：31-44,14.DOI:10.16742/j.zgcdxb.20240114
基于NDHD、NDVI及EVI光谱植被指数的锡林郭勒盟草地冠层高度遥感反演对比研究
Grassland Canopy Height Retrieval Using Remote Sensing Vegetation Indices(NDHD,NDVI,and EVI)in Xilin Gol
摘要
Abstract
Grassland canopy height is a key indicator of phenological dynamic and surface spatial hetero-geneity,reflecting vegetation growth and grassland cover changes.The study focused on the natural grass-lands of Xilin Gol League,Inner Mongolia Autonomous Region,China.Moderate resolution imaging spec-troradiometer(MODIS)optical multi-angle canopy observation data were used to reconstruct the plant structural characteristics of the grassland population's bidirectional reflectance distribution function(BRDF)based on the RTLSR kernel-driven model.Canopy hotspot and darkspot reflectance were calcu-lated using BRDF data,and the normalized difference hotspot and darkspot(NDHD)vegetation index was derived for the study region.Further,a dynamic inversion model of grassland canopy height(GCH)was developed using the NDHD,NDVI and EVI vegetation indices and a general regression neural network(GRNN).The GCH inversion results showed as follows:(1)Based on the kernel-driven model coeffi-cients(fiso,fvol and fgeo),the reflectance of grassland vegetation canopy at hotspot and darkspot was calcu-lated.The volume and geometric optical scattering characteristics of the vegetation population effectively described the spectral dynamic response associated with vegetation growth.(2)The red and near-infrared NDHD vegetation index,derived from canopy BRDF data,effectively captured structural changes and spa-tial heterogeneity in grassland growth dynamics.(3)The retrieved GCH spatial distribution in the study area exhibited a pattern of low-to-moderate values in the west and higher values in the northeast.Temporal variations in GCH aligned with the phenological progression of grassland growth.(4)The R2 values of the NDHD-based temporal and spatial model were 0.58 and 0.89,respectively,significantly higher than those obtained from traditional vegetation indices(0.44 and 0.36).The contribution ratio of the red and near-infrared NDHD vegetation indices were 3.32%and 3.57%,demonstrating their improved spatial heteroge-neity representation and retrieval potential.In summary,the grassland canopy height inversion model,developed using the RTLSR model and NDHD vegetation index with GRNN,achieved high numerical inversion accuracy in both temporal and spatial dimensions.This approach provides a valuable theoretical and methodological reference for practical applications in grassland monitoring and management.关键词
草地植被/多角度遥感/二向反射分布函数/归一化热点和暗点植被指数/冠层高度模型Key words
Grassland vegetation/Multi-angle remote sensing/Bidirectional reflection distribution function/Normalized difference hotspot and darkspot vegetation index/Canopy height model分类
信息技术与安全科学引用本文复制引用
兰春阳,郭利彪,谭维贤,黄平平,李苏和,马铭泽..基于NDHD、NDVI及EVI光谱植被指数的锡林郭勒盟草地冠层高度遥感反演对比研究[J].中国草地学报,2025,47(3):31-44,14.基金项目
内蒙古自然科学基金面上项目(2021LHMS04002) （2021LHMS04002）
内蒙古教育厅基本科研业务费项目(JY20220198,JY20220072) （JY20220198,JY20220072）
国家自然科学基金联合基金项目(U22A2010) （U22A2010）
内蒙古工业大学科研项目(BS2021078) （BS2021078）
