| Summary: | Soil organic carbon (SOC) is an important soil property. It is the main constituents of soil organic matter and a good indicator of soil quality. The estimation and mapping of SOC content could be used to select potential agricultural areas in the Arctic areas. However, conventional analysis of SOC are time consuming and expensive. They involve a lot of sample preparation, and chemicals and are destructive. Near infrared spectroscopy (NIRS) in the range between 400 and 2500 nm is an alternative method for SOC analysis. It is fast and non-destructive. The aims of this study where to test the feasibility of using NIRS to estimate SOC content on a landscape and field scale in Greenland. Partial Least squares regression models were built to correlated soil spectra and their reference SOC data to develop calibration models. Very good predictive ability for both landscape and field scale were obtained with r2 values of 0.96 and 0.88, and RMSEP of 2% and 0.45%, respectively. NIRS has proven to be a reliable method for SOC determination in Greenland soils with the potential for supplementing or substituting the expensive conventional SOC analysis.
|
|---|