ARCHIVED REPORTS PHASE II FIELD INVESTIGATION AND LNAPL MOBILITY EVALUATION REPORT

Home Browse Search

1 ' Phase II Field ARCADIS Investigation and LNAPL ' Mobility Evaluation Report Union Pacific Railroad Company ' Former Maintenance Facility Tracy, California other analyses discussed herein such that confidence in the overall conclusion is ' not diminished due to the lack of representative bail-down testing results. 4.1.4 Undisturbed Soil Cores LNAPL fluid properties and petrophysical aquifer properties are estimated from the undisturbed soil cores and the LNAPL sample described above(Sections 3.4 and 3.6). ' The LNAPL fluid properties and aquifer characteristics are required as inputs for the LNAPL mobility evaluation described in Section 5 below. Specifically,ARCADIS developed a spreadsheet to calculate LNAPL velocity that requires the following inputs: ' • Site Characteristics: hydraulic gradient, hydraulic conductivity(described in Section 2) ' • Fluid Characteristics: LNAPL and water density, air-water-LNAPL interfacial tensions (included in Table 5 and described in more detail in Section 5.1.3) ' • Aquifer Characteristics: porosity, soil capillarity parameters(van Genuchten N and alpha), irreducible water saturation, residual LNAPL saturation (included in Table 6 and described in more detail in Section 5.1.3) ' Soil capillarity was estimated using Retention Curve Program (RETC)for Unsaturated Soils with both the van Genuchten model and the Brooks and Corey model. These models are documented in API Bulletin 4760: LNAPL Distribution and Recovery Model Volume 1:Distribution and Recovery of Petroleum Hydrocarbon Liquids in Porous Media(API Bulletin 4760;API 2007). The soil capillarity data generated by the AWC ' test(see Section 3.4)were used with the RETC program to generate van Genuchten N and alpha fitting parameters and the Brooks and Corey displacement head pressure. Appendix E documents the RETC program runs that generated the capillarity fitting ' parameters used in this analysis. In the van Genuchten model, a soil moisture retention model (either Burdine or ' Mualem)must be selected in order to calculate relative permeability appropriately. There is little information available in the literature addressing selection of the appropriate model, and generally it is assumed that coarser-grain soils adhere to the ' van Genuchten-Burdine model, and finer-grain, more cohesive soils are better approximated using the van Genuchten-Mualem soil moisture retention model. ' To aid in model selection for this LNAPL mobility evaluation, a relative permeability test was performed in the laboratory on the soil core sample from LIF-02. Multiplying 1 1 Mtp4 Y-bWl MONRY(I1-2a2010)4W 14