| dc.description.abstract | Production from the Groningen gas field has resulted in induced seismicity due to reactivation of existing faults in the subsurface ever since 1991. In the same year, the Royal Netherlands Meteorological Institute (KNMI) deployed a pilot borehole with a number of geophones at depth. With time the network has evolved in a dense geophone network in the past 5 years. A denser network allows for higher accuracy in determining hypocenter locations. These locations are especially important because they indicate which faults in the subsurface are seismically active and thus can result in an improved seismic hazard analysis. In the hypocenter method that is used by the KNMI to determine hypocenter locations, the focal depth is fixed to 3 km as this is the average gas reservoir depth. In this research, we explore the viability of hypocenter relocation in the Groningen region by means of the double-difference method. This method is based on the similarity of ray paths if the epicentral distance between two events is small relative to the event-station distance, for a common seismic station. With this, travel time differences can be attributed to relative hypocenter distances between those events within the source area, thus minimizing the effects of travel time variations caused by unknown subsurface structures. HypoDD from Waldhauser and Ellsworth (2000), a double-difference relocation program based on ray theory with a 1D velocity model, is used for this purpose. We have found that under some constraints, hypocenters can be relocated with an estimated uncertainty of 240 m in both the vertical and horizontal plane. Compared to other methods, the double-difference method has proved to be a viable hypocenter relocation method. Additionally, our results indicate that the gas source rock, gas reservoir, cap rock and a shallower anhydrite layer likely all are seismically active, with faults present in the gas reservoir extending into deeper formations. | |
