The coal resources in China are mainly concentrated in the arid and rain-less northern areas. Open-pit mining and transportation can easily cause the diffusion and pollution of coal dust. After the dust diffuses, some of them settle and cover the surrounding vegetation under gravity, leading to the common phenomenon of dust falling on vegetation. When remote sensing is used to monitor vegetation growth and health status, the dust fall effect will affect the spectral purity of vegetation so that the signal obtained by remote sensing is the mixed spectral signal of vegetation and dust fall, which seriously affects the quantitative remote sensing accuracy of vegetation. In order to study the dust retention effect of coal dust in vegetation remote sensing, the experiment of spectral measurement for dusty leaf was carried out, and the change of spectrum and vegetation index was studied. On this basis, sentinel-2A remote sensing images with red-edge bands were used to compare vegetation indices in the coal dust affected area and control area in Huolinhe open-pit mining area, the Inner Mongolia Autonomous Region, respectively, to verify the experimental results of ground spectral measurement.In this experiment, vegetation leaves were selected from a coal mining area in Shenyang. The coal dust was evenly sprayed on the leaf surface with a level difference of 2 g·m-2 to simulate the effect of dust retention on the actual leaf surface. Spad-502 chlorophyll meter was used to measure the spectrum of dust retention on the leaf surface, and the spectrum and vegetation index variation rules were explored.Since the red edge band was used for the vegetation index in this experiment, sentinel-2A remote sensing images were selected. Huolinhe Open-pit mine in NeiMengGu province and its surrounding areas were selected as the verification area. The coal-dust-affected area, and the control area were selected for vegetation index comparison to verify the experimental results of ground spectral measurement. The results show that, with the increase of dust retention (0～36 g·m-2), coal dust on the surface of the leaves will gradually reduce the overall reflectance of the leaves, and the change amplitude of the reflectance at the peak of the leaf spectrum (560, 720, 860, 1 680, 2 220 nm) is significantly higher than that at the trough (445, 681, 1 940 nm).With the increase in dust retention, NDVI (normalized difference vegetation index), SR705 (simple ratio), and ND705 (normalized difference index) decreased significantly. However, MTCI (the medium resolution imaging spectrometer terrestrial chlorophyll index), mSR705(modified simple ratio) and mND705 (modified normalized difference index) are unchanged, showing the characteristics of coal dust resistance. The reflectivity at 445 and 681 nm in these indices play an important role. By using Sentinel-2A remote sensing image in Huolinhe open pit coal mine area and comparing the coal dust affected area with the control area, MTCI, mSR705, and mND705 indices show the characteristics of coal dust resistance, verifying the results of ground experiments. The study would lay a foundation for vegetation remote sensing in the area of coal dust pollution and ensure the accuracy of vegetation remote sensing inversion.