As part of a study to understand factors impacting the efficiency of an artificial recharge pond in Watsonville, CA, a time series of resistivity measurements was made using a permanently installed one-dimensional resistivity probe. Measurements were made in the top 2 m of sediment with data acquired every 30 min. There was an observed diurnal signal in these data due to daily temperature fluctuations in the pond water. By viewing this signal as a thermal tracer, we used the movement of the associated thermal front to estimate infiltration rates from the resistivity data. We developed a wavelet-based method for calculating lag times of the thermal front between measurement locations. As part of this algorithm, we tested the statistical significance of a given signal and automatically rejected calculated lag times that were associated with signals below a given confidence interval. We included a linear inversion routine for calculating the velocity of the thermal front from the calculated lag times. Using the thermal velocity, we estimated an infiltration rate at the resistivity probe that decreased from approximately 3.5 to 1 m d−1 during a period of 18 d. Resistivity data have a distinct advantage over direct temperature measurements: a resistivity measurement is sensitive to changes outside the region disturbed by instrument emplacement. While our processing approach was demonstrated on the presented resistivity data, it is equally valid for use with direct temperature measurements.