 |
| Titel |
Prediction of soil organic carbon in forest areas of the Piedmont region, Northern Italy, using environmental variables: vegetation and topographic patterns effect. |
| VerfasserIn |
Ines Oueslati, Paola Allamano, Eleonora Bonifacio, Pierluigi Claps, Giovanni Laguardia |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250042440
|
|
|
|
|
|
| Zusammenfassung |
| Soil organic carbon (SOC) is one of the most important parameters affecting soil hydraulic
properties. It is easily measured by chemical analyses, but it is highly variable in space.
Therefore the definition of a methodology allowing for SOC spatial prediction with a
reasonable accuracy is crucial in large scale studies.
This study aims at predicting the spatial variability of the soil organic carbon
concentration (%SOC) in forest topsoils in Piedmont (North-western Italy) using
spatially referenced environmental factors related to terrain morphology, climate, and
vegetation.
In this region 122 soil profiles were available with soil organic carbon concentrations at
depths of 0 –10 cm, 10 –20 cm, 20 –30 cm, 0 -20 cm and 0 –30 cm and in the horizons A, B
and C.
For each point terrain attributes were derived from a 50 meters pixel digital elevation
model (DEM), using the SAGA geographic information system. In addition basic NDVI
statistics, such as the mean, minimum and maximum values, and the Fourier series
phases (F1 and F2) and amplitudes (A1 and A2) for the 12 months and 6 months
periods were evaluated from the long term monthly average NDVI series obtained by
SPOT-Vegetation data. Mean annual precipitation estimates were also available for each
sample.
A multiple regression analysis were applied to investigate the relationship between the
%SOC in different layers and horizons and the environmental descriptors. The relationships
that we found show that the NDVI parameters and the precipitation are statistically significant
predictors (P |
|
|
|
|
|
|