Skip to main content
ARS Home » Research » Publications at this Location » Publication #96274
Title: PARAMETER ESTIMATION FOR THE MOBILE/IMMOBILE (MIM) DOMAIN MODEL

Author
item Jaynes, Dan
item SHAO, M - IOWA STATE UNIVERSITY

Submitted to: American Society of Agronomy Meetings
Publication Type: Abstract Only
Publication Acceptance Date: 10/22/1998
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: A simple method has been proposed that is applicable to field soils for estimating the dimensionless MIM model parameters beta, the fraction of mobile to total water content and omega, the solute transfer rate from mobile to immobile domains. The method uses a tension infiltrometer to apply a sequence of conservative, nonreactive tracers to the soil followed by a single soil sample to determine resident concentrations. A critical assumption of the method is that the effect of hydrodynamic dispersion, D, on the resident tracer concentration is small after the solute front has moved past the depth of sampling, thus allowing for a simple log-linear solution to the MIM equations. We evaluated the impacts of this assumption and the use of the log-linear solution on parameter estimates using the complete solution to the MIM equations and synthetic data sets generated for a wide range of solute transport conditions. We found the assumption of negligible D to be valid for large Peclet values (P greater than 100) or small omega (less than 0.01) values. At P less than 0.3, D is too large to assume negligible for accurate parameter estimation at most omega values. At omega greater than 1, the solute transport system degenerates into a single mobile domain and the method does not work for most P values. The log-linear solution to the MIM equations, as well as a modified version that relaxes the assumption that the resident tracer concentration equal the input concentration, provided worse estimates of beta and omega than the complete solution with a small preset value of D over much of the reported range of beta and omega values. However, the three estimation methods gave similar values for beta, and omega values within a factor of 2, when used with measured tracer data.