Method for analyzing soil structure according to the size of structural elements
The soil structure in situ is the result of cropping history and soil development over time. It can be assessed by the size distribution of soil structural elements such as air-filled macro-pores, aggregates and stones, which are responsible for important water and solute transport processes, gas ex...
Gespeichert in:
Veröffentlicht in: | Computers & geosciences 2015-02, Vol.75, p.96-102 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The soil structure in situ is the result of cropping history and soil development over time. It can be assessed by the size distribution of soil structural elements such as air-filled macro-pores, aggregates and stones, which are responsible for important water and solute transport processes, gas exchange, and the stability of the soil against compacting and shearing forces exerted by agricultural machinery. A method was developed to detect structural elements of the soil in selected horizontal slices of soil core samples with different soil structures in order for them to be implemented accordingly. In the second step, a fitting tool (Eureqa) based on artificial programming was used to find a general function to describe ordered sets of detected structural elements. It was shown that all the samples obey a hyperbolic function: Y(k)=A/(B+k),k∈{0,1,2,…}. This general behavior can be used to develop a classification method based on parameters {A and B}. An open source software program in Python was developed, which can be downloaded together with a selection of soil samples.
•Size distribution of macro-pores, aggregates and stones at CT images in 2D.•Flood fill algorithm for detection of soil structures.•Artificial programming tool Eureqa as a general function finder.•Nonlinear parameter fitting of the hyperbolic function from Eureqa.•Open source software development in Python. |
---|---|
ISSN: | 0098-3004 1873-7803 |
DOI: | 10.1016/j.cageo.2014.11.007 |