Abstract: In the present study, GF-1 and Landsat 8 remote sensing data for Yi'an County and Baiquan County were adopted along with laboratory tested data of soil sample to reveal the feasibility and differences of application of GF-1 and Landsat 8 in soil organic matter inversion. It was shown that the reflectivity of remote sensing images was significantly associated with soil organic matter content in the visible and near-infrared band, and the correlation coefficient reached the maximum in the near-infrared band. The exponential model based on near-infrared band of GF-1 was better than the power model based on near-infrared band of Landsat 8 in estimating soil organic matter content. Blue (deep blue) and red band were introduced into multiple regression model to enhance the ability of estimating organic matter content, and it was especially beneficial for Landsat 8. In short, GF-1 exhibited a higher spatial resolution, shorter revisit cycle and similar predictive ability as compared to Landsat 8, and thus was able to replace Landsat 8 in soil organic matter inversion.

Key words: GF-1, Landsat 8, soil organic matter, quantitative inversion