Experimental investigation of the thermal time‐series method for surface water‐groundwater interactions

Abstract

Diel temperature fluctuations have been used to quantify vertical water flow through saturated sediments with 1‐D analytical heat models. The underlying transport equation relies on assumptions that could be violated using field temperature records. To test the capability of this method a hydraulic laboratory experiment was designed. A mass of fully water‐saturated homogeneous sand was exposed to 19 different uniform pressure gradients inducing steady state Darcy velocities 0 < q < 25 m d−1. An areal heating grid generated steady sinusoidal thermal forcing. Multipoint sediment temperature responses demonstrated increasing spatial variability for increasing velocities. This introduced horizontal temperature gradients. Velocities calculated from heat tracing were compared to velocities independently obtained from solute slugs. Heat‐ and solute‐derived velocities agreed for q < 3 m d−1, but heat‐derived velocities were consistently larger at higher velocities. Temperature amplitude‐ and phase‐derived velocities revealed significant scatter when compared to solute velocities. This scatter reduced when amplitude‐ and phase‐derived velocities were compared for each sensor pair. The variability in heat‐derived velocities therefore represents a spatially variable flow field in the sand. However, amplitude‐ and phase‐derived velocities deviated from a 1:1 relationship at higher velocities. This can partly be explained by longitudinal thermal dispersivity, and partly by enhanced thermal spreading due to horizontal temperature gradients originating from nonuniform flow. This is surprising given the homogeneous sand and transition zone transport conditions (Pet < 0.7). These findings have implications for the quantification of velocities from field records because field conditions are likely more heterogeneous, which would exacerbate the effects found in this study.