Predicting Total Organic Carbons and Nitrogens in Grassland Soil Using Wavelet Analysis and Hyperspectral Technology

Abstract

The assessment of greenhouse gas emissions from soils requires an accurate knowledge on the fate of carbon and nitrogen in soils. Traditional analysis can not be used to assess carbon and nitrogen over large geographical areas. For this reason, hyper spectral remote sensing techniques for predicting soil organic carbon (SOC) and total nitrogen (TN) on a large scale have received much attention. This study mainly focused upon capturing the feature values of soil organic carbon and total nitrogen, and predicting SOC and TN by applying wavelet analysis to reflectance spectra. Results indicated that the maximum correlation coefficient between SOC, TN, and wavelet coefficient were more than 0.96 compared to the relationship between SOC, TN, and spectral reflectance (r=-0.79 for SOC, r=-0.40 for TN), especially for TN (the maximum negative correlation coefficient r=-0.964). For SOC+TN and SOC/TN, due to SOC contents accounted for a large proportion of soil composition compared to TN, their spectral feature were affected by SOC in soil samples. In addition, wavelet analysis also enhanced the features of SOC+TN and SOC/TN obviously. These results suggested that wavelet analysis was a better method for capturing the absorption features of soil composition using hyper spectral remote sensing data, and predicting the changes of C and N in terrestrial ecosystems.