Premium
Determining Soil Hydraulic Properties from One‐step Outflow Experiments by Parameter Estimation: I. Theory and Numerical Studies
Author(s) -
Kool J. B.,
Parker J. C.,
Genuchten M. Th.
Publication year - 1985
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/sssaj1985.03615995004900060004x
Subject(s) - outflow , hydraulic conductivity , residual , mathematics , soil water , nonlinear system , nonlinear regression , estimation theory , sensitivity (control systems) , water content , soil science , mechanics , geotechnical engineering , environmental science , regression analysis , geology , statistics , physics , algorithm , engineering , oceanography , quantum mechanics , electronic engineering
The numerical feasibility of determining water retention and hydraulic conductivity functions simultaneously from one‐step pressure outflow experiments on soil cores by a parameter estimation method is evaluated. Soil hydraulic properties are assumed to be represented by van Genuchten's closed‐form expressions involving three unknown parameters: residual moisture content θ, and coefficients α and n. These parameters are evaluated by nonlinear least‐squares fitting of predicted to observed cumulative outflow with time. Numerical experiments were performed for two hypothetical soils to evaluate limitations of the method imposed by constraints of uniqueness and sensitivity to error. Results indicate that an accurate solution of the parameter identification problem may be obtained when (i) input data include cumulative outflow volumes with time corresponding to at least half of the final outflow and additionally the final outflow volume; (ii) final cumulative outflow corresponds to a sufficiently large fraction (e.g., >0.5) of the total water between saturated and residual water contents; (iii) experimental error in outflow measurements is low; and (iv) initial parameter estimates are reasonably close to their true values.