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.