Boulder transport models have been used to develop a historical record of hydrodynamic events. This is done by evaluating the size, location, and composition of boulders that have been removed from their region of formation. Boulder transport is difficult to predict, even under laboratory conditions.
Experiments were conducted to evaluate analytic boulder transport models under dam break flow conditions. The experiment was conducted using a flow tank designed specifically for generating dam break flow over a soil model in a centrifuge. The initiation of movement of the boulders was measured by proximity sensors. Flow depth was measured with pressure transducers, flow velocities were estimated using two frame comparison of images. The hydrodynamic forces applied to the boulders were estimated based on the flow characteristics, then compared to the measured movement of the boulders.
The estimated hydrodynamic forces predict that the boulders would begin the transport process before any measured movement. This discrepancy is attributed to the flow depth relative to boulder size. Existing boulder transport models may reduce error by incorporating the impact of relative water depth on estimated hydrodynamic forces.