Premium
Determination of the exponential unsaturated hydraulic conductivity function by parameter identification
Author(s) -
BRANDYK T.,
ROMANOWICZ R.,
SKAPSKI K.
Publication year - 1989
Publication title -
journal of soil science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.244
H-Index - 111
eISSN - 1365-2389
pISSN - 0022-4588
DOI - 10.1111/j.1365-2389.1989.tb01271.x
Subject(s) - exponential function , hydraulic conductivity , soil science , levenberg–marquardt algorithm , function (biology) , water content , identification (biology) , muck , mean squared error , mathematics , geotechnical engineering , soil water , environmental science , computer science , engineering , mathematical analysis , statistics , artificial intelligence , artificial neural network , botany , evolutionary biology , biology
SUMMARY The parameter identification technique for the evaluation of the exponential unsaturated hydraulic conductivity function is described. The identification is performed using the indirect approach with an output error criterion for steady soil moisture flow through peat‐muck soil columns. The identification procedure uses the Levenberg‐Marquardt algorithm to the solution of the stated least‐square optimization problem. Examination of the evaluated parameters from laboratory experiments justifies the use of this simple identification method for the whole soil profile when the variation of the soil properties in the soil profile is not too large. For layered soil profiles with relatively large differences of soil properties between layers, the parameters of the exponential function must be evaluated separately for each layer.