RESUMO

An analytical model has been developed to describe the initiation of episodic percolation in a variable-aperture fracture or channel. The model describes two primary processes: (1) the accumulation of water above a capillary barrier that is induced by variations in aperture sizes and (2) the subsequent drainage of the accumulated water once the capillary rise height has been exceeded. The transient head and velocity terms are derived for the intermittent flow processes. An episodicity factor, e, is also derived that represents the fraction of time in which flow actually drains through the fracture or channel for a specified infiltration boundary condition. A stochastic Monte Carlo analysis is performed using uncertainty distributions for the model input parameters to simulate the capillary rise height, volume of accumulated water, average episodic flow rate, drainage time, episodicity factor, and ratio of episodic flow rate to a steady infiltration rate. The simulated drainage events, while generally short in duration, are large in magnitude relative to the infiltration boundary condition. The distributions of simulated results spanned several orders of magnitude due to the uncertainty in the input parameters, and sensitivity analyses using stepwise linear regression are performed to identify the input parameters that are most important in propagating the uncertainty in the simulated results.