Fluid pressure and effective stress in sandbox models

We have done a series of simple tests on sandpacks, involving upward flow of compressed air through the pores and its effect on the yield strength. The ultimate objective is to model deformation coupled with fluid how in sedimentary basins. For all tests, we used a single batch of Fontainebleau sand, sieved to a grain size between 0.200 and 0.315 mm, poured into a cylindrical container and then fluidized. The density of this sand was 1.585 g/cm(3), irrespective of sand thickness. The lithostatic pressure was proportional to the thickness of the sandpack. A yield envelope was obtained by shearing the sandpack horizontally. Compressed air entered the base of the sandpack and flowed upwards through the pore spaces. For 69 measurements on air flow without shearing, a plot of discharge velocity versus gradient of fluid pressure is close to a straight line, verifying Darcy's law and yielding an intrinsic permeability of about 1.7 darcy for the sand. For 72 tests on simultaneous shearing and fluid flow, the estimated effective stress (lithostatic stress minus estimated pore fluid pressure) ranged from 0 to 1600 Pa and the pore fluid ratio (between air pressure and lithostatic pressure) from 0.0 to 1.0. A plot of shear stress versus effective stress at failure is almost linear, verifying Terzaghi's principle of effective stress. The best-fit slope (coefficient of internal friction) is about 0.55 and the intercept on the shear stress axis (cohesion) is less than 85 Pa. The tests show that it is feasible to use compressed air within sandpacks, as a means of modelling deformation coupled with fluid flow. The next step will be to build sandbox models of layered sequences and to investigate detachments caused by abnormal fluid pressures. (C) 1999 Elsevier Science B.V. All rights reserved.