For the effective application and better understanding of correlation between NDVI (Normalized Difference Vegetation Index) and ST (Surface Temperature) and the conversion process between them to better monitor and evaluate the land cover change, especially for explaining the situation with NDVI being saturated based on the parameters of the land cover and soil moisture inversion, the relation of NDVI to ST and the application of different characteristicspace NDVIST models in drought monitoring are analyzed. The results indicate that the arbitrariness in the simulation of the NDVIST characteristic space increased when the land cover was high, which influences the accuracy of drought monitoring. The uncertainty of the effects of multifactors on ST and NDVI also affected the accuracy of drought monitoring. Suggestions are given: the analysis of the variation regularities of the NDVIST space under different land cover and different moisture conditions should be based on the multitemporal remote sensing and different resolution images and focused on the modification of the models of the NDVIST characteristic space and the characteristic changes of drywet edges applied in drought monitoring, especially in the higher land cover situation, so to best facilitate the application of the NDVITs characteristic space method in drought monitoring.