Premium
Multiphase flow modeling of a crude‐oil spill site with a bimodal permeability distribution
Author(s) -
Dillard Leslie A.,
Essaid Hedeff I.,
Herkelrath William N.
Publication year - 1997
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/97wr00857
Subject(s) - soil science , geology , permeability (electromagnetism) , saturation (graph theory) , transect , spatial distribution , mineralogy , relative permeability , particle size distribution , hydrology (agriculture) , environmental science , porosity , particle size , geotechnical engineering , oceanography , chemistry , paleontology , biochemistry , remote sensing , mathematics , combinatorics , membrane
Fluid saturation, particle‐size distribution, and porosity measurements were obtained from 269 core samples collected from six boreholes along a 90‐m transect at a subregion of a crude‐oil spill site, the north pool, near Bemidji, Minnesota. The oil saturation data, collected 11 years after the spill, showed an irregularly shaped oil body that appeared to be affected by sediment spatial variability. The particle‐size distribution data were used to estimate the permeability ( k ) and retention curves for each sample. An additional 344 k estimates were obtained from samples previously collected at the north pool. The 613 k estimates were distributed bimodal lognormally with the two population distributions corresponding to the two predominant lithologies: a coarse glacial outwash deposit and fine‐grained interbedded lenses. A two‐step geostatistical approach was used to generate a conditioned realization of k representing the bimodal heterogeneity. A cross‐sectional multiphase flow model was used to simulate the flow of oil and water in the presence of air along the north pool transect for an 11‐year period. The inclusion of a representation of the bimodal aquifer heterogeneity was crucial for reproduction of general features of the observed oil body. If the bimodal heterogeneity was characterized, hysteresis did not have to be incorporated into the model because a hysteretic effect was produced by the sediment spatial variability. By revising the relative permeability functional relation, an improved reproduction of the observed oil saturation distribution was achieved. The inclusion of water table fluctuations in the model did not significantly affect the simulated oil saturation distribution.