4.5 Dye tracer test
Uranine dye was released continuously at GS1 during 10:53–11:20 on June 26, 2019, and it took 4 minutes for the tracer front to travel 152 m downstream to F1, and 6.5 minutes to F2. A plateau concentration of 37-43 ppb was reached at F1 and 31-36 ppb at F2. The range in the plateau concentrations measured at each fluorometer are attributed to an inconsistent pumping rate. This is likely due to the battery not providing a consistent voltage and the decreasing head difference between the tracer solution and the stream which made pumping harder.
The breakthrough curves were analyzed using OTIS-P. A model was generated in which both F1 and F2 were considered (Figure 12). A storage zone area (As , Equation S1-2) of 0.11 m2 and 0.18 m2, and an exchange coefficient (α, Equation S1-1 and S1-2) of 2.2×10-4s-1 and 3.0×10-4 s-1 were estimated for the respective sub-reaches. A dispersion coefficient of 1.2 m2 s-1 and 1.3 m2s-1, and stream cross sectional area of 0.33 m2 and 0.36 m2 were also estimated. The average stream cross sectional area from stream survey measurements was 0.36 m2. This means the storage zone area was between one-third and one-quarter the stream cross sectional area. The rate of longitudinal dispersion was comparable to estimates by Haggard et al. (2001) where flow conditions were similar. This was on the higher end of values reported in the literature (0.05 – 1.3 m2 s-1) (Lautz & Siegel, 2007). The RMSE of the model was 0.93 ppb. The plateau concentrations were used to account for tracer dilution. Discharge increased by 0.042 m3 s-1 or 25 % between GS1 and GS2. Discharge increases of 0.018 m3 s-1and 0.024 m3 s-1 were determined for the respective sub-reaches. Discharge obtained from the GS1 and GS2 rating curves showed an increase of 28% along the reach (Figure 3) which is similar to the estimates obtained through tracer dilution. The average time that solutes spent in the storage zone (ts ) was 18 minutes suggesting that transient storage represents storage in the hyporheic zone, which can retain some solutes for 10-30 minutes, rather than stagnant side channel. This is complemented by field observations of turbulent flow with few stagnant pools.
Hyporheic exchange in the channel therefore did not have a noticeable effect on stream temperatures and represented a negligible flux compared to the energy fluxes included in the heat transfer equation (Equation S2-1). Furthermore, hyporheic exchange is likely captured in part by the bed conduction flux (Equation S2-11).