Soil Organic Carbon Content Estimation with Laboratory-Based Visible–Near-Infrared Reflectance Spectroscopy: Feature Selection

2014 ◽  
Vol 68 (8) ◽  
pp. 831-837 ◽  
Author(s):  
Tiezhu Shi ◽  
Yiyun Chen ◽  
Huizeng Liu ◽  
Junjie Wang ◽  
Guofeng Wu
Keyword(s):  
Feature Selection ◽  
Organic Carbon ◽  
Soil Organic Carbon ◽  
Carbon Content ◽  
Near Infrared ◽  
Reflectance Spectroscopy ◽  
Organic Carbon Content ◽  
Infrared Reflectance ◽  
Near Infrared Reflectance Spectroscopy ◽  
Near Infrared Reflectance

Possibilities of near infrared reflectance spectroscopy for the prediction of organic carbon concentrations in grassland soils

2005 ◽  
Vol 143 (6) ◽  
pp. 487-492 ◽  
Author(s):  
C. VAN WAES ◽  
I. MESTDAGH ◽  
P. LOOTENS ◽  
L. CARLIER
Keyword(s):  
Organic Carbon ◽  
Near Infrared ◽  
Reflectance Spectroscopy ◽  
Infrared Reflectance ◽  
Near Infrared Reflectance Spectroscopy ◽  
Near Infrared Reflectance ◽  
Grassland Soils ◽  
Dry Soil ◽  
Natural Grassland ◽  
Carbon Concentrations

For the determination of soil organic carbon (OC) concentrations, the availability of a fast, low-cost analysis method is required. The aim of the present study was to evaluate the possibilities of near infrared reflectance spectroscopy (NIRS) to build a spectral database and to develop calibrations for the prediction of organic carbon concentrations in grassland soils. NIRS spectra of 1626 soil samples from different grasslands (both agricultural and natural) were collected between 1100 and 2500 nm. NIRS calibrations were developed with modified partial least square regression and tested with independent validation samples. The best equations were obtained with the first derivative of the spectra without scatter corrections. For the global calibration, containing the samples of all origins, the standard errors of calibration (SEC) and of prediction (SEP) were respectively 3·70 g OC/kg dry soil (R2=0·89) and 3·95 g OC/kg dry soil (R2=0·88). The ratio of the standard deviation of the reference validation data to the SEP (RPD), indicating the performance of the calibration, was 2·9. Dividing the samples into groups according to their practice (agricultural or natural grassland), improved SEP by 5·8 and 7·7%, respectively. Dividing the samples into texture groups (clay, silt, sand) improved SEP for agricultural grassland by, on average, 7·4% and for natural grassland by 16·2%.

Sign in / Sign up

Export Citation Format

Share Document