2008 Joint Annual Meeting (5-9 Oct. 2008): In Situ Assessment of Heat-Pulse and Thermal Profile Methods for Measuring Streambed Recharge/Discharge.

There is a need for effective, continuous, in situ water flux measurements for applications such as streambed recharge or discharge. In this study, streambed fluxes were measured using a water balance method (WBM), which were then compared to measurements from i) a heat-pulse ratio method (HPRM), ii) a temperature-profile method (TPM), and iii) a seepage meter method (SM). The WBM measured flux using upstream (weir) and downstream (flume) water flows differences divided by stream area. The HPRM used ratios of two temperature dissipation curves from a dual-probe to calculate conductive flux at three locations and two depths (3.5 and 7.5 cm). The TPM used equally spaced temperature profiles measurements to calculate flux from the conduction-convection equation at three locations and two depths (4 and 8 cm). Also, the SM used a cylinder inserted into the streambed attached to a balloon to quantify water volumes. The SM was used at three locations. A statistical analysis of HPRM and WBM measurements (n=28-30 for each of the five working locations/depths) indicated correlation coefficients (r) ranged from 0.73 to 0.90 (avg. r = 0.83). However, only one of the five HPRM measurement locations/depths was not statistically different from the WBM measures. For the five TPM measurement locations/depths (n=30 for each location/depth) the correlations to WBM were all negative (r ranged from -0.30¬¬ to -0.53; avg. r =  0.42). All TPM measurement location values were statistically different from the WBM measurements. Only six SM values were measured and all six fell within 95% confidence range of the WBM measurements; however, these SM values did not appear correlated to WBM values. The strong correlations between the HPRM and WBM measurements indicate that the HPRM was directly quantifying fluxes at point locations. Nonetheless, an average stream flux was accurately measured at only one location/depth. This result may have indicated that the point HPRM measurements did not reflect an integrated flux measure of the entire stream crossection. The HPRM was promising and a remedy to more accurate in situ flux measures could be a reconfiguration in HPRM measurements (e.g. perpendicular to stream flow) or perhaps an in situ calibration.