Using satellite imagery and spectral data to estimate the amount of soil organic carbon in central Zagros forests in Khoozestan

Soil organic carbon (SOC) is one of the most important components of soil physical and chemical properties that prevented soil decay and destruction. The objective of the present study is the evaluation of SOC changes using the remote sensing technique compared with field methods at central Zagros forests in Khoozestan province in Iran over the past 2 decades.

 The soil samples were collected randomly from the soil surface (0-10 cm depth) to estimate the SOC concentrations in the laboratory. Analysis of digital data by using Operational Land Imager (OLI) of satellite Landsat 8 and Enhanced Thematic Mapper (ETM+) sensor of satellite Landsat 7 images in 2016 (the sampling year) was done to estimate surface organic carbon levels of soil. The main objective was to establish soil organic carbon relation with landsat different bands ratios and also herbal and moisture indexes such as NDVI, SAVI, BSCI, NDMI and NSMI corresponding to the SOC values obtained from soil samples of the forest areas in the central Zagros mountain, and for that purpose these data were evaluated using different linear regression methods.

The best fit model of stepwise regression method showed R2 value of 0.435 for landsat 8 and R2 value of 0.501 for landsat 7 and finally based on these results, evaluation of SOC changes occurred in previous years.

Results show the significant relationship between soil organic carbon and the reflectance in the Visible, Near-Infrared and Short-wave Infrared part of the spectrum.