RECENT ADVANCES IN SEEPAGE METERS DESIGNED TO QUANTIFY TEMPORAL AND SPATIAL VARIABILITY OF GROUND-WATER SURFACE-WATER EXCHANGE IN FLUVIAL SETTINGS

Numerous studies have demonstrated large spatial variability in hyporheic exchange and several have indicated substantial short-term temporal variability where sediment beds are mobile. Characterizing and quantifying spatial and temporal variability in seepage exchange remains one of the most difficult challenges for hyporheic research. Standard half-barrel seepage meters provide perhaps the most direct measurement of flux across the sediment-water interface, but their use in fluvial settings often is discouraged because the hydraulic disturbance presented by the seepage cylinder, as well as the velocity-head gradient at the flexible surface of the seepage bag, corrupt the attempted seepage measurement. Enclosing the bag in a shelter greatly reduces the effect of moving water on the seepage bag, but the problems of hydraulic disturbance and sediment scour associated with the seepage cylinder remain. Relatively simple modifications in seepage-meter design can minimize these problems and allow direct measurements of hyporheic exchange. Low-profile seepage cylinders greatly reduce the hydraulic disturbance of the flow field. Separating the seepage-bag shelter from the cylinder and placing the shelter in a lower-energy portion of the stream allows seepage-meter measurements in surface-water velocities as large as 0.6 m/s. Temporal variability also can be quantified, either with frequent bag measurements or a fast-response flowmeter connected to the seepage cylinder. Seepage cylinders can be ganged together to integrate spatial heterogeneity and focus seepage volumes large enough to exceed the minimum measurement threshold of the flowmeter. These modifications were tested in a controlled-seepage environment and coefficients were developed to account for seepage-meter inefficiency and any remaining hydraulic disturbances. Tests conducted in a short reach of a sand-and-gravel-bed river indicated that hyporheic exchanges were large and directionally variable and, where the gravel bed was mobile, varied temporally on a scale of minutes to hours.