Applying multivariate methods to soil–solution interactions in carbonate media

• Published in: Geoderma Vol. 137; no. 3; pp. 352 - 359
• Main Authors: González, C., Quintana, J.R., Moreno, L., Vázquez, A., Lafuente, A.L., Romero, A.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 15.01.2007; Elsevier
• Subjects: Agronomy. Soil science and plant productions; Biological and medical sciences; Calcaric Fluvisols; Earth sciences; Earth, ocean, space; Exact sciences and technology; Experimental columns; Fundamental and applied biological sciences. Psychology; Multivariate statistical analysis; Soil solution; Soils; Surficial geology; Calcaric Fluvisols; Experimental columns; Multivariate statistical analysis; Soil solution; atmosphere; Information; Total nitrogen; displacements; Motion study; soils; Fluvisols; Gradient; processes; Organic clay; Redundancy; Interaction; Property of soil; pressure; carbonates; water; correlation; Characteristics; solid phase; techniques; ecosystems; leachate; statistical analysis; Contaminant
• ISSN: 0016-7061; 1872-6259
• DOI: 10.1016/j.geoderma.2006.08.016

Soil solution composition depends on the solid phase properties of soils. Therefore it is necessary to determine how soil and solution interact in order to understand the processes that influence ecosystems, such as plant nutrition or the movement of contaminants in soils. As the product of many interrelated factors, these interactions are very complex. We studied the properties of leachates of 10 calcaric Fluvisols subject to displacement with de-ionized water, in columns at atmospheric pressure, and investigated their relationship with properties of the soil. To determine these interactions we used two complementary statistical techniques: bivariate correlations and a multivariate direct gradient canonical ordination method (Redundancy Analysis). The two statistical techniques used to relate the characteristics of the solution and the properties of the soil gave similar results. The two techniques allowed the conclusion that clay (%), organic carbon (g kg − 1 ) and total nitrogen (g kg − 1 ) were the properties of the soils that determine the composition of the soil solution. Also, the multivariate methods employed have been useful to synthesize information when a large number of variables are used.