Cremeans, M.M., J.F. Devlin, T.C. Osorno, U.S. McKnight, and P.L. Bjerg.
Groundwater Monitoring & Remediation 40(1):24-34(2019)

Four tools to measure discharge rates in a sandy streambed were compared at six locations spaced 3 m along a transect near the north bank of a stream in Denmark: mini-piezometers, streambed point velocity probes (SBPVPs), temperature profilers, and seepage meters. All identified a similar trend of low to high groundwater discharges moving westward along the transect. Differences between discharges estimated from Darcy calculations (using the mini-piezometers) and SBPVPs were not statistically different from zero at the 90% confidence level. Seepage meter estimates were consistently lower than those of the other two methods but compared more reasonably with the application of a correction factor. In contrast, discharges estimated from temperature profiling to a depth of 40 cm were about an order of magnitude less than those determined with the other methods, possibly due to interferences from horizontal hyporheic flow. Method-specific sources of bias were hypothesized to cause statistically different discharge estimations at the same location. Seepage meters and the SBPVP achieved the least variability in measuring flow across the groundwater-surface water interface. The accuracy of the seepage meter depended on a calibrated correction factor while that of the SBPVP did not.