Tuesday, November 6, 2007
183-27
Using KL-Divergence and Water Retention Data to Characterize Soil Structure.
Sung Won Yoon, Environmental Sciences, Rutgers, The State University of New Jersey, 14 College Farm Rd., New Brunswick, NJ 0801 and Daniel Giménez, Department of Environmental Sciences, Rutgers, The State University of New Jersey, 14 College Farm Road, New Brunswick, NJ 08901-8551.
Soil structure can be assessed by differences between properties measured in disturbed and undisturbed soil. The Kullback Leibler Divergence (KL-D) is a measure defined as the entropy of the difference between two density functions. Our objective was to quantify soil pore systems obtained from water retention curves of disturbed and undisturbed soil samples with the KL-D measure. Three soil clods were sampled from each of 24 horizons and used to measure bulk density (saran resin method) and water retention by the hanging column and pressure extractor methods. Bulk soil samples were collected to measure texture and organic matter. Seven soil-water pressure potentials were used in the range from -3 to -100 cm on undisturbed soil samples and 13 pressure potentials in the range from -3 to -15000 cm on disturbed samples. Water retention data was fit with the Kosugi lognormal water retention model and the parameters from the model used to calculate KL-D with an analytical equation. The KL-D measure exhibited distinctive values for some of structural types with the textural and aggregate size classes. The proposed measure could serve as a link between qualitative field description of soil structure and a quantitative measure of the effect of soil structure on pore size distribution.