Sensitivity Screening the van Genuchten/Mualem Soil Hydraulic Parameters
Abstract
A screening evaluation, including a Morris one-factor-at-a-time (OAT) sensitivity analysis (SA), was performed to evaluate the relative elementary and secondary effects for six of the seven van Genuchten Mualem (VGM) hydraulic model parameters used in predicting water retention and movement through a coarse-textured soil profile under atmospheric boundary conditions. The tortuosity parameter (l) was not evaluated in the OAT analysis because it is generally considered a fitting parameter with little physical significance, and the lack of reliable data concerning the parameters value and distribution as a function of soil texture. Simulated data sets were created based on coarse textural classes using the HYDRUS-1D code, a one dimensional finite element water and solute transport model based on Richards’ equation, to eliminate bias and uncertainty associated with site heterogeneity and measurement error inherent to field data, as well as evapo-transpiration. For each textural class, the simulated data set consisted of hourly estimates of volumetric water content values at 30 and 60 cm soil depths, assuming a homogeneous 100 cm soil profile with atmospheric boundary conditions based on hourly precipitation for the month of February. Initial water distribution profiles were based on average sensor readings for the field site recorded at the beginning of the simulation interval. For the Morris OAT analysis, a set of random input parameter values was chosen from within the upper and lower parameter bounds defined by the extreme 95% confidence intervals reported for coarse-textured soil materials assuming an even parameter distribution. Each parameter was then varied independently consistent with the Morris method. Model results for the new parameter set were then compared to the ideal simulated data for each soil texture, i.e., RSSE. Of the five parameters evaluated, SA indicated that n and θsat were the most sensitive parameters, and α was found to be the least.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2009
- Bibcode:
- 2009AGUFM.H23F1027S
- Keywords:
-
- 1846 HYDROLOGY / Model calibration;
- 1873 HYDROLOGY / Uncertainty assessment;
- 1875 HYDROLOGY / Vadose zone;
- 1894 HYDROLOGY / Instruments and techniques: modeling