Shear velocity of the Rotokawa geothermal field using ambient noise

Abstract:

Ambient noise correlation is an increasingly popular seismological technique that uses the ambient seismic noise recorded at two stations to construct an empirical Green's function. Applications of this technique include determining shear velocity structure and attenuation. An advantage of ambient noise is that it does not rely on external sources of seismic energy such as local or teleseismic earthquakes. This method has been used in the geothermal industry to determine the depths at which magmatic processes occur, to distinguish between production and non-production areas, and to observe seismic velocity perturbations associated with fluid extraction. We will present a velocity model for the Rotokawa geothermal field near Taupo, New Zealand, produced from ambient noise cross correlations. Production at Rotokawa is based on the "Rotokawa A" combined cycle power station established in 1997 and the "Nga Awa Purua" triple flash power plant established in 2010. Rotokawa Joint Venture, a partnership between Mighty River Power and Tauhara North No. 2 Trust currently operates 174 MW of generation at Rotokawa. An array of short period seismometers was installed in 2008 and occupies an area of roughly 5 square kilometers around the site. Although both cultural and natural noise sources are recorded at the stations, the instrument separation distance provides a unique challenge for analyzing cross correlations produced by both signal types. The inter-station spacing is on the order of a few kilometers, so waves from cultural sources generally are not coherent from one station to the other, while the wavelength produced by natural noise is greater than the station separation. Velocity models produced from these two source types will be compared to known geological models of the site. Depending on the amount of data needed to adequately construct cross-correlations, a time-dependent model of velocity will be established and compared with geothermal production processes.