Satellite Visible–Near Infrared Reflectance Correlates to Soil Nitrogen and Carbon Content in Three Fields of the Thessaly Plain (Greece)
2013
Stamatiadis, Stamatis | Evangelou, Lefteris | Blanta, Anna | Tsadilas, Christos | Tsitouras, Alexandros | Chroni, Christina | Christophides, Calliopi | Tsantila, Eleftheria | Samaras, Vasilis | Dalezios, Nikolaos | Dimogiannis, Dimitris
Surface visible–near infrared (NIR) reflectance of bare soil by remote sensing devices has been used to infer topsoil properties such as organic matter, soil texture, water content, salinity, and crop residue cover. Spectral mapping of soil properties can be ultimately used as a tool for the implementation of site-specific management practices at the field scale or for soil–landscape modeling at a regional scale. The accuracy of prediction of soil properties with satellite imagery is affected by conditions and properties of the soil surface, by radiometric and atmospheric effects, and by spatial and spectral resolutions. In this study, a high-resolution World View 2 image was used to map soil reflectance in three 10-ha fields of differing soil types and textures that were located in different sections of the east Thessaly Plain. Radiance data from four visible-NIR channels were extracted from the same coordinates that soil samples were taken at two soil depths within each field. Point radiance values were correlated to soil organic matter, total carbon (C) and nitrogen (N) contents, their isotopic composition, carbonate content, nitrate content, pH, electrical conductivity, and soil texture that were analyzed in the laboratory. Strong correlation coefficients emerged between green/NIR image reflectance and total soil N, organic matter, and carbonate content across the three fields in both soil depths. The greatest negative correlation coefficient (R ² = 0.77) was obtained between satellite NIR reflectance and soil N content. More data are needed to verify these relationships, but the results indicated the potential of high-resolution satellite imagery to quantify within-field and regional-scale variability of soil N and C in the Thessaly Plain.
显示更多 [+] 显示较少 [-]