Soil Organic Carbon Variation in Alpine Landscape (Northern Italy) as Evaluated by Diffuse Reflectance Spectroscopy
2014
Colombo, Claudio | Palumbo, Giuseppe | Di Iorio, Erika | Sellitto, Vincenzo Michele | Comolli, Roberto | Stellacci, Anna Maria | Castrignanò, Annamaria
A better understanding of soil organic C (SOC) distribution in mountain soils is important for quantifying C emissions–removals involved in land use change processes, such as afforestation, reforestation, and deforestation activities. The highest part of Valchiavenna, a Northern Italy alpine area, is characterized by a great variety of landscapes owing to the different parent material, climatic conditions, topography, vegetation, and geomorphology. The interaction of these features has produced many soil types, such as Leptosols, Regosols, Cambisols, Umbrisols, Podzols, and Histosols, with large differences in total organic C (TOC) content. To predict TOC content, diffuse reflectance spectroscopy (DRS) in the visible and near-infrared (VIS-NIR) range was applied. Nine soil profiles were sampled to represent a range of pedologic profiles and were analyzed with DRS in the wavelength interval of 350 to 2500 nm and for some chemical and physical parameters. Spectral data were transformed and then analyzed through partial least square regression (PLSR) and multiple linear regression. Soil parameters were also considered as covariates. The results indicate that the DRS technique linked with PLSR and applied to smoothed spectra was able to predict TOC and highlight the relationships between TOC and Al and Fe amorphous minerals. The wavelength in the visible range, with particular regard to the 520- to 560-nm interval, were the most influential in explaining TOC variation compared with NIR (700–1400 nm) and short-wave infrared (SWIR) bands (1400–2500 nm). This may suggest that changes in reflectance in the visible range can be caused by a significant variation in TOC in the alpine soil profiles. This calls for a closer examination of pedologic processes in mountain environments owing to advanced chemometric models to predict SOC content in mountain soils.
显示更多 [+] 显示较少 [-]