Lawrence Berkeley National Laboratory

This project successfully implemented an approach to jointly image time-lapse changes in water and steam concentrations and subsurface flow in a geothermal reservoir, using data from small earthquakes and magnetotelluric observations. The project advanced the technology by imaging time-lapse changes of the two data sets, based on different physical properties, for the first time. The project demonstrated this technology at The Geysers geothermal field in Northern California over an area of approximately 75 square kilometers, where seismic and magnetotelluric data were collected over several years. The project team collected seismic data from over 280,000 earthquakes and collected three magnetotelluric surveys to generate images of water and steam volumes as well as flow paths and barriers in the geothermal reservoir. Correlation of the geophysical images with known water injection and steam production volumes allowed the team to calibrate the data and to gain confidence in the results, which can now be applied throughout the reservoir, where borehole data are unavailable. The results of the joint imaging, together with reservoir data derived from observations in boreholes, allow interpretation of the images to identify water and steam saturated zones, as well as fluid pathways and barriers. This information allows the reservoir operator to improve its drilling program by minimizing drilling of unsuccessful wells, resulting in reduced costs and lower electricity rates for California ratepayers.