Specific storage from sparse records of groundwater response to seismic waves

Rayleigh waves from moderately large earthquakes produce, at all epicentral distances, significant groundwater fluctuations. The direct comparison of seismic Rayleigh waves and associated groundwater oscillations has previously been shown to be useful for evaluating the specific storage of aquifers. However, such methods require measuring water levels on a scale of seconds rather than the more typical scale of minutes employed in most well recorders. A new computation procedure is needed to deal with this sparse amount of water level data provided by the most data loggers relative to seismological data. We show that, given the transmissivity of a confined uniformly porous aquifer, a single water level deflection measurement, if normalized to an appropriately filtered power spectrum of the associated Rayleigh wave motion, can provide a rough but unbiased estimate of specific storage. Given a sufficient number of such discrete observations, specific storage can be computed to the same accuracy as can be found from continuous well records. The precision of the algorithm is strongly dependent on the number of water level measurements available during the passage of Rayleigh waves. However, because each water level measurement is treated independently, data from multiple earthquakes can be combined to ensure a low computational error. Of course, the overall accuracy of the method depends not only on the computation procedure but also on the fit of the aquifer to the initial assumptions and on the extent to which the aquifer transmissivity is known.