The utilization of remote sensing satellite imagery & GIS for estimating soil organic carbon content through vegetation indices

Abstract

The following paper, as a starting point, identifies and outlines the main advantages of soil organic carbon and states the tremendous limitations that global agriculture will face due to ongoing soil degradation. It suggests offering as a solution the estimation of soil organic carbon through remote sensing techniques that implement vegetation indices as predictors of soil organic carbon. In an effort to support these sayings an experiment is also presented related to SOC estimation. 73 soil points were analysed by multivariate correlation and ordinal logistic regression regarding their SOC as the dependent variable and vegetation indices like normalized difference vegetation index, soil adjusted vegetation index, bare soil index, and green normalized difference vegetation index as the independent variables. In the results section, the study outlines the significant correlation coefficients of the statistical analysis that were acceptable and satisfactory as they were mostly equal to 0.9 for the majority of the vegetation indices, with the exception of the bare soil index, which indicated an inversely proportional behaviour towards soil organic carbon. Finally, the paper states the main advantages of the study in its last section and points out how further research can be improved by the implementation of unmanned aerial vehicles.