← Back to Sessions

Importance of Accurate Earth Models and Network Geometry for Earthquake Location

Description: 

For more than two decades induced seismicity has been an important topic due to the increased number of Oil and Gas well operations (Saltwater disposal and Hydraulic Stimulation). Recently, induced seismicity has become an important issue related to geothermal and Carbon Capture and Storage projects, being one of the top concerns of the public. One of the most challenging problems is the strong dependence of hypocenter location, especially depth, on the velocity model used for the subsurface and the seismic network geometry (i.e. density and azimuthal coverage). At a country/state seismic network operations level it is almost impossible to overcome these problems since, in general: (a) seismic networks do not have a station density equal to the target depth of the human activities, and (b) accurate 3D velocity models for basin and basement structures are not available.

We have developed an approach using 3D models based on sonic log, check shot data and tomography results for the Midland basin, West Texas, to investigate the sensitivity of earthquake location and depth to network geometry and the velocity structure. In a subarea of the basin where the network is dense (interstation distance ~5km) and using at least 5 stations to locate an event, we find that including stations at distances greater than ~12 km shows a strong effect of the basement velocity on the locations. For example, reducing velocities for the basement and using all station data, hypocenters move deeper in the basement. When there are high-impedance velocity contrasts either inside the basin structures or at the basin basement interface, streaks of earthquake location occur, as expected. Due to lack of a dense network in the Midland basin, and in order to avoid the streaks and large time residuals we have incorporated an approach based on simplified wide angle refraction seismology to optimize the sub-basement velocity to create an improved reference 3D velocity model. We used this 3D reference velocity model to relocate the seismicity in the Midland basin area and identify the bias on hypocentral depth compared to 1D velocity models.

Session: Advances in Reliable Earthquake Source Parameter Estimation - I

Type: Oral

Date: 4/16/2025

Presentation Time: 08:15 AM (local time)

Presenting Author: Alexandros

Student Presenter: No

Invited Presentation: 

Poster Number:

Authors