PERCOLATION THRESHOLD AND ITS EFFECT ON COMPUTING PERMEABILITY FROM GRAIN SIZE

Grain size data are usually more readily available than permeability data. Furthermore, permeability can be estimated using the Kozeny-Carman equation and an effective grain diameter. Koltermann and Gorelick (1995) concluded that the effective grain diameter should be computed using the geometric mean for distributions in which the coarser pores are not completely filled by finer grains (fine packing), and the harmonic mean for distributions in which coarser pores are completely filled by finer grains (coarse packing).

Recent work has suggested that openings in coarser pores cease to percolate even when they are incompletely filled with finer grains. A percolation threshold should then be used as the basis for choosing between the geometric mean and the harmonic mean when computing the effective grain diameter. A series of bimodal sediment mixtures was created with grain size distributions in which the finer grains did not completely fill the pores created by the coarser grains. Permeability values for these mixtures are being both measured in the laboratory and separately calculated with the Kozeny-Carman equation. Measured permeability values are being compared to two sets of calculated values, those that used the geometric mean and those that used the harmonic mean when computing effective grain diameter. The results will be used to test the idea that a percolation-based threshold needs to be considered when choosing between the geometric and harmonic mean for effective grain diameter.