Title: Sensitivity analysis of tracer transport in variably saturated soils at USDA-ARS OPE3 field site Authors
Submitted to: American Geophysical Union
Publication Type: Abstract Only
Publication Acceptance Date: September 27, 2010
Publication Date: December 13, 2010
Citation: Guber, A.K., Pan, F., Pachepsky, Y.A., Yakirevich, A., Gish, T.J., Nicholson, T., Cady, R. 2010. Sensitivity analysis of tracer transport in variably saturated soils at USDA-ARS OPE3 field site. American Geophysical Union. Technical Abstract: The objective of this study was to assess the effects of uncertainties in hydrologic and geochemical parameters on the results of simulations of the tracer transport in variably saturated soils at the USDA-ARS OPE3 field site. A tracer experiment with a pulse of KCL solution applied to an irrigation plot was implemented to measure the chloride concentration at three sampling depths of 12 observations wells at the OPE3 site. The soil material spatial distribution was obtained from soil cores. The hydraulic conductivity for each material was estimated based on the soil pore size distribution using the ROSETTA software. Saturated and residual soil moisture contents, and van Genuchten parameters were obtained by fitting the van Genuchten model to the measured water retention curves. A three-dimension flow and transport model was set up to simulate the flow and chloride transport for the tracer experiment at the OPE3 site. Sensitivity analysis was conducted for three parameters (hydraulic conductivity, longitudinal and transverse dispersivity) through variations of these parameter values by one standard deviation from the base-case values at each of eight soil materials. The parameter variations resulted in 256 parameter sets. Sensitivity index of each parameter for each material was calculated based on the 256 parameter-set modeling results with regard to groundwater level, peak tracer concentration, and peak travel time. The importance of parameter variations on the tracer transport at the OPE3 site was evaluated through the values of sensitivity indices. The results showed that the variations in hydraulic conductivity at sandy loam soils caused larger uncertainty of tracer transport than the ones at other soil materials. Stochastic simulations of the spatial distribution of sandy loam subsurface structural units can contribute to the evaluation of possible improvements in monitoring designs. The outcome of this study can provide the information for the future data collection and monitoring efforts to further reduce the uncertainty of tracer transport in variably saturated soils.