基于NDVI和EVI不同植被指数表征的粤港澳大湾区植被空间格局驱动因子影响力比较分析

为研究归一化植被指数（normalized difference vegetation index,NDVI）和增强植被指数（enhanced vegetation index,EVI）的表征差异是否会造成有关植被研究的结果差异,分别在采用2005—2020年MODISNDVI、MODIS-EVI 2种不同遥感植被指数表征粤港澳大湾区植被空间特征的基础上,以同时期17个自然因子和人为因子作为驱动因子,通过地理探测器模型方法,计算各驱动因子对基于NDVI和EVI的植被空间特征的影响力。结果显示,虽然粤港澳大湾区南亚热带-热带植被在NDVI和EVI的表征下,其结果存在差异,但在不同植被指数下,通过地理探测器模型方法计算各驱动因子影响力量化及排序结果基本一致,未受不同植被指数表征差异的影响。在驱动因子中,土地利用类型、高程均是最主要驱动因子,对植被空间分布影响力均超过50%。因子之间均表现出双因子增强作用。土地利用类型协同人口分布因子对NDVI表征下的植被空间分布影响力最强;高程协同人口分布因子对EVI表征下的植被空间分布影响力最强。研究结果表明,虽然NDVI、EVI在表征植被覆盖特征方面存...

Influence of driving factors under different vegetation indices of NDVI and EVI in Guangdong-Hong Kong-Macao Greater Bay Area

To investigate whether differences in the representations of the two commonly used vegetation indices, NDVI and EVI, affect vegetation-related studies under different index representations. This study is based on the use of two different remote sensing vegetation indices, MODIS-NDVI and MODIS-EVI, from 2005 to 2020 to characterize the spatial features of vegetation in the Guangdong-Hong Kong-Macao Greater Bay Area. Using 17 natural and anthropogenic factors from the same time period as the driving factor, the effect of the 17 driving factors on the spatial features of the vegetation based on the different indices NDVI and EVI is calculated separately using Geo-detector. The results show that although there are differences in the results of the southern subtropical-tropical vegetation in the Guangdong-Hong Kong-Macao Greater Bay Area as characterized by different vegetation indices in NDVI and EVI, the quantitative and ranking results of the influence of driving factors under different indices calculated by the Geo-detector module are consistent, and are not affected by the differences in index representation. Among the drivers, land use type and elevation are the dominant drivers. Their effect on the spatial distribution of vegetation is greater than 50% under different metrics. Most of the factors have a two-factor enhancement effect on the spatial features of vegetation under different representation metrics. Synergies in population distribution and land use type have the strongest effect on NDVI-based vegetation cover; however, the synergistic effect of the population distribution and elevation has the strongest effect on the vegetation cover based on the EVI index. This study concludes that while the NDVI and EVI indices differ in representing vegetation cover characteristics, the results of the quantitative analysis of drivers remain consistent.