Inversion of organic matter content in wetland soil based on Landsat 8 remote sensing image

摘要

Using GF-1 and Landsat8 remote sensing images as data sources, combined with the experimental data of wetland soil sampling in Anyi County and Gao'an Research Area, the ability and difference of two remote sensing images in inversion of soil organic matter content were compared. The results show that the reflectivity of the two remote sensing images in the visible and near-infrared bands is significantly correlated with the soil organic matter content, and the correlation is the most in the near-infrared band. The index model established by the near-infrared band of GF-1 is more than the near-infrared band using Landsat8. The power model estimates slightly better. The multi-regression model established by introducing the blue band (dark blue band) and red band has higher inversion precision than the single-band model, especially the improvement effect of Landsat8 remote sensing image. Compared with Landsat8, GF-1 remote sensing image has higher spatial resolution and shorter revisit period, and has similar predictive ability in detecting soil organic matter content, which can replace Landsat8 remote sensing image. (C) 2019 Elsevier Inc. All rights reserved.