XES Basin

The XES facility is a large experimental basin (13 m x 6.5 m), developed and built with funds from NSF and the University of Minnesota , that permits the formation of stratigraphy through the use of a flexible subsiding floor. The goal is to reproduce the real-world (i.e. spatially variable) kinematics of subsidence, as determined by geophysical modeling and backstripping of real basins. 

The floor is a honeycomb of 432 independent subsidence cells (Fig. 1) through which a gravel "basement" is slowly removed to provide accommodation space for deposition. At the beginning of an experiment, the basin is filled with dry, well sorted commercial gravel. The top of the gravel is covered with a thin rubber membrane. The experimental deposit is formed on top of this membrane. Subsidence is induced by withdrawing gravel from the bottoms of the hexagonal cells. Each hexagon forms the top of a cone that tapers into a standard elbow pipe (Fig. 2). The gravel in the cone rests at the angle of repose in this elbow. Subsidence is induced by firing a pulse of high-pressure water into the gravel in the elbow. A small volume of gravel is knocked out of the elbow and falls into an exhaust line, where it is transported out of the system and stored for later reuse. Each subsidence cell has its own sealed pressure tube that drives the pulses via a computer-controlled solenoid valve. We have refined and calibrated the pulsing so that each pulse produces about 0.12 mm of subsidence: the "earthquake slip" in the experiments. This is about equal to the resolution with which the basement elevation can be read (described below), and also to the typical grain size of sediment in the experiments. Hence the subsidence is effectively smooth and continuous in time. The subsidence is also spatially continuous. The cells are separated only at floor level, so the gravel can flow laterally to accommodate differential subsidence with no breaks at the cell boundaries. Firing a single cell, for instance, produces a smooth bowl-shaped subsidence pattern that extends over the six adjoining cells. Extensive testing has shown that the underlying honeycomb structure is not imprinted on the subsidence at the surface until the rubber membrane (the top of the basement) has been lowered to within about 0.2 m of the honeycomb. This leaves about 1.3 m of usable accommodation space in the basin. As long as the gravel basement is loaded, lateral slopes of up to 60 can be produced between adjoining cells 

Premixed sediment and water can be fed from anywhere along the perimeter of the basin, and the level of standing water is independently set by a computer-controlled head tank mounted outside of the basin. Thus, base level (in effect, eustatic sea level) can be raised or lowered independent of events within the basin. 

During an experiment, the surface flow pattern is recorded using video and still cameras. In addition a topographic scanning system, based on the design of Rice and Wilson (1988) and Wilson (1990), allows us to document separately the 3-D evolution of the surface topography during the run for later comparison with the surface-flow images, the preserved deposits, and theoretical predictions. 

Once the experiment is complete, the tank is pumped dry and the resultant deposits are cut in a series of precise parallel faces, beginning near one edge. Each face is then photographed. At greater intervals, a peel is taken of the cut face. This serial microtome process allows us to build a 3-D image of the deposits by stacking the sequence of photographed slices.