Monitoring steam content in the crust with seismic noise

Seismic noise interferometry (SNI) is based on the computation of seismic response by extracting correlated wavefields from continuous seismic recordings. This seismic response is understood as the Green’s function of the propagation medium and it is typically used to image the subsurface. Furthermore, a regular retrieval of this magnitude in time enables the identification of changes in the structural and mechanical properties of the subsurface and therefore, it can be also used for monitoring purposes.

Here, we use SNI to monitor the steam content in geothermal reservoirs. The massive extraction of fluids commonly causes a pressure drop in the rock matrix and land subsidence in the surroundings of production areas. Furthermore, the boiling point decreases yielding decompression boiling. Estimating the steam fraction at depth is challenging especially in reservoirs with two-phase fluids. In this work, we focus on the Hengill geothermal area (Iceland), where the steam ratio has increased by around 30% in the last 10 years due to the harnessing of geothermal energy.

In this area, we measure the land deformation with synthetic aperture radar interferometry (InSAR) and estimate seismic velocity changes with SNI. We observe that both magnitudes linearly decrease in the long term accordingly to the energy production. Besides, we model the expected seismic velocity drop with in-situ borehole data (temperature, pressure, and steam ratio) and conclude that the seismic velocity drop might be related to the steam evolution within the reservoir. These results offer an innovative way of estimating the steam content in the crust with a surface and non-invasive technique.