ADVANCES IN THE AUTOMATION OF DIURNAL SIGNAL ANALYSIS FOR STREAMBED PROPERTIES AND VERTICAL WATER FLUX

The theory underpinning the use of heat as a tracer to estimate vertical fluid exchange between surface water and groundwater was developed in the 1960s. In recent years there has been a rapid expansion in the use of heat as a quantitative tracer of water flow, based on improved temperature sensors, signal processing techniques, new analytical solutions, and computer software to apply them. In particular, analytical solutions based on amplitude ratio and phase shift output (or both) extracted from temperature time series offer the opportunity to estimate extended time series of vertical fluid fluxes, thermal diffusivity, and streambed scour at a sub-daily time step.

Under certain hydraulic conditions, such as strong groundwater upwelling, the modeled vertical flux rates are highly-sensitive to streambed thermal diffusivity. Localized, in-situ measurements of this parameter derived from diurnal signal amplitude attenuation and phase shift relationship can increase flux measurement accuracy. In systems with a highly mobile streambed, thickness above a consistently buried sensor can be estimated, yielding a fine-scale (sub-daily, sub-cm) record of dynamic scour and deposition. Further, it has recently been demonstrated that the thermal diffusivity and streambed scour equations can be used to identify time periods when flux estimates are unreliable. Fundamentally, the researcher must choose between amplitude and phase-based fluid analytical flux models. In this talk, we provide an overview of diurnal temperature signal methods and discuss recent advances in the automation of vertical flux estimation using the VFLUX programs, allowing informed model selection. In particular we will show how estimates of thermal diffusivity (using amplitude and phase information) that reproduce the known sensor spacing can be utilized to reduce the uncertainty in flux estimates due to uncertainty in thermal properties of porous media.