Snow cover is an important indicator of global climate change， and its surface high reflection and rapid ablation change characteristics affect the energy and water vapor exchange between the earth’s land surface and the atmosphere. Therefore， accurate identification of snow ablation information is crucial for local climate research and water resources management. The traditional methods of snow cover monitoring at meteorological stations and field measurements are not only time-consuming and laborious， but also limited in observation range， which cannot fully reflect the characteristics of snow cover at regional scale. Nowadays， satellites images with large-area simultaneous observation capabilities have become an important data source for snow cover monitoring. Optical remote sensing can reliably extract snow coverage， but it is easily affected by weather factors such as clouds and fog， and is not sensitive to dry/wet snow distinction， making it difficult to identify wet snow information. Synthetic aperture radar （SAR） not only overcomes the influence of atmospheric conditions， but also is sensitive to changes in dielectric constant caused by snow ablation， and can accurately identify wet snow. However， the common microwave wavelength is much larger than the particle size of dry snow， making it difficult to directly identify dry snow cover.In this paper， a snow cover ablation recognition method is proposed by combing SAR and optical remote sensing data. Taking Babao River basin as the research area， firstly， Sentinel-2 simulation data simultaneously with Sentinel-1 were obtained to extract the snow cover range by using the ESTARFM fusion model with MODIS and Sentinel-2 images. Wet snow cover was extracted based on a multi temporal and multi polarization SAR change detection algorithm. Then， by making a difference between the wet snow extracted from SAR and the snow cover extracted from Sentinel-2 simulation data， the distribution of wet and dry snow was obtained during the ablation period （February to May） in 2020. And the uncertainty of dry and wet snow obtained is corrected through two methods： （1） the wet snow outside the snow coverage area is divided into snow free pixels based on the snow cover extracted by Sentinel-2； （2） by calculating the average elevation of wet snow on semi-shady slope， semi-sunny slope and sunny slope， all the dry snow below this elevation was corrected as wet snow. Finally， the distribution of dry and wet snow in 18 days during the ablation period in Babao River basin was obtained. At the same time， GF-2 images was used to verify the accuracy of the snow cover range， and the Sentinel-2 image during the severe ablation period were used to verify the accuracy of the wet snow distribution. The results show that the method can quickly identify the characteristics of snow ablation in Babao River basin， and the overall classification accuracy OA is as high as 99%， and the Kappa coefficient is as high as 0.86.Experiments show that the spatial distribution of dry and wet snow cover changes drastically with time in the Babao River basin during the ablation period in 2020. In the initial stage of ablation the wet snow at the is mainly concentrated in the low-altitude area of the valley， while dry snow is mainly distributed in surrounding high altitude mountainous areas. Subsequently， the wet snow area extracted from the descending orbit and ascending orbit data shows an increasing trend， while the snow cover area is basically stable. As the temperature further increases， the snow cover in high-altitude areas begains to melt. And due to the difference in satellite transit time， the area of wet snow extracted from descending orbit data is smaller than that extracted from ascending orbit data during the entire ablation period. This method can fully utilize the advantages of SAR and optical remote sensing images to quickly monitor snow cover ablation changes except for forest cover areas， and provide reliable basic data for research on climate change， water resource management， and other related fields.