Premium
Estimation of the Saturated Hydraulic Conductivity in Clay Soils from Soil Moisture Retention Data
Author(s) -
Messing I.
Publication year - 1989
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1989.03615995005300030003x
Subject(s) - soil water , hydraulic conductivity , water content , soil science , scaling , clay soil , porosity , exponent , water retention curve , mathematics , mineralogy , chemistry , geology , geotechnical engineering , geometry , linguistics , philosophy
A generalized Kozeny‐Carman equation was employed to relate saturated hydraulic conductivity ( K s ) to effective porosity (φ e ) in clay profiles in Sweden with clay content >15% were analyzed. The value empirical constants. The term φ e was defined as porosity (φ) minus soil‐water content at −30 kPa. Around 500 horizons from 60 soil profiles in Sweden with clay content >15% were analyzed. The value of n was tested with two statistical methods. According to the first method, a regression test on the log‐log relationship between K s and φ e with least‐squares lines fitted through the data, the exponent n was found to lie around 2 for soils with high clay content (>40%), and 3 for soils with lower clay content (15–40%). The correlation coefficients were moderate, ranging from 0.51 to 0.71 for different clay soil groups. In the second method, the cumulative frequency distributions of scaling factors for K s determined from measured values of K s were compared with scaling factors for K s determined from φ e . Reasonable agreements were obtained with n = 4 to 5 for soils with clay content >40%, and with n = 5 to 6 for soils with clay content 15 to 40%. The second method gave n ‐values that correspond well with other recent published work. The results indicate that the spatial distribution of K s is estimated reasonably well from φ e values, whereas the individual values of K s are not estimated as well.