 |
| Titel |
Charactering Spatial Variability of Soil Properties Measured on a Transect by Multifractal Analysis |
| VerfasserIn |
A. Paz González, M. Valcarcel Armesto, J. Dafonte, J. M. Miras Avalos, R. Da Silva Días, M. A. Marinho, C. A. de Abreu |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250064418
|
|
|
|
|
|
| Zusammenfassung |
| Spatial variability of soils in landscapes has been studied in different ways, for example in
terms of soil survey reliability, soil development and erosive processes. Due to the advent of
site-specific management in the 1990s, there is now an increasing interest in measuring the
amount of soil variability within a field. Methods for assessing spatial variability also include
use of transect techniques to sample soil sequences. On the other hand, over the past few
decades fractal and multifractal models have been applied in the evaluation of the spatial
variability of soil attributes. Therefore, the aim of this study was to characterize the spatial
variability of general soil properties and extractable nutrients measured along a transect by
means of multifractal analysis. The field work was conducted at the experimental farm of
CIAM located in Mabegondo, A Coruña, Spain on a gently slope. The soil was loamy
textured. Soil samples were taken at 66 points located 0.8 m apart along a transect of 52 m.
Samples were analyzed for pH, organic matter content (OM), exchangeable K,
Mg and Ca, exchangeable H+Al, and DTPA extractable Fe, Mn, Cu and Zn. In
addition, sum of bases (SB), cation exchange capacity (CEC) and percent base
saturation (V) were calculated from exchangeable cations. For all the studied statistical
moments the logarithm of the normalized measures varied linearly (r2 > 0.87) with the
logarithm of the measurement scale, meaning that the distribution of the measure
could be considered as a fractal. The scaling properties of the soil properties studied
were further characterized to determine if the scaling types was monofractal or
multifractal. To this effect selected indices were calculated from the generalized
dimension function, Dq. So for a distribution with a monofractal tendency values
of the correlation dimension D2 and the entropy dimension, D1, become similar
to the capacity dimension, D0, however D0 >D1 > D2 if the distribution has a
tendency to multifractal behaviour. For pH, H+Al, CEC and DTPA extractable Fe,
D0 -D1 -D2 -1.00 (value which corresponds to a one-dimensional spatial series), meaning
a monofractal scaling type. However, for OM, exchangeable K, Mg and Ca, percent
base saturation (V) and DTPA extractable Mn, Cu and Zn, D0 >D1 > D2, which
implies a multifractal nature where low and high density regions of the variable scale
differently.
Acknowledgments. This work was funded in part by Spanish Ministry of Science and
Innovation (MICINN) in the frame of project CGL2009-13700-C02. |
|
|
|
|
|
|