Abstract: This study takes Hengshui Lake National Nature Reserve as the study area, and uses object-oriented support vector machine classification method to achieve the ecological land classification based on the GF-2 satellite multi-spectral data. The results indicated that the GF-2 satellite data could achieve satisfactory performance on remote sensing classification of ecological land in the study area with the overall classification accuracy of 92.62% and Kappa coefficient of 0.875. Therefore, the GF-2 satellite data have good performances in ecological land classification in nature reserves and object-oriented support vector machine classification method is a suitable strategy, which provides references for further application of GF-2 satellite data.