Groundwater-lake interactions: I. Accuracy of seepage meter estimates of lake seepage

Abstract

The use of seepage meters to identify nearshore seepage patterns and to quantify seepage in lakes was evaluated with a Monte Carlo simulation model. The model simulated seepage flux as would be derived from seepage meter measurements along a transect extending from the shore of a hypothetical lake to 40 or 100m offshore. Along the transect, simulated seepage velocities decreased exponentially with distance from shore according to patterns measured at Narrow Lake, Alta., and Lake Sallie, MN. To determine statistical parameters needed in the model, seepage flux was measured in situ with closely spaced seepage meters at four different sites in Narrow Lake. Seepage velocities within a small area of lakebed were log-normally distributed, and the variance was positively correlated with mean seepage flux. The modeling indicated that the most sensitive parameter affecting the accuracy of seepage meter estimates of seepage patterns and average seepage flux along the transect was the variability in the spatial distribution of seepage flux within a small area of lakebed. There was little improvement in the accuracy of estimates of seepage patterns or transect flux when more than ten seepage meters were simulated along the transect, when the transect was “sampled” more than twice, or when seepage meters along the transect were simulated to follow a stratified rather than a systematic design.