An Evaluation of Hypocenter Location Techniques With Applications to Southern Utah: Regional Earthquake Distributions and Seismicity of Geothermal Areas

D. J. Wechsler, Department of Geology and Geophysics, University of Utah
R. B. Smith, Department of Geology and Geophysics, University of Utah

IDO/DOE/ET/28392-32 Author's work performed under Contract no. DE-AC07-78ET/28392 Prepared for Department of Energy, Division of Geothermal Energy

Abstract

Three techniques for the computation of earthquake hypocenter locations were compared through empirical results of synthetic test cases utilizing four different computer programs. The three approaches: (1) the single event method, (2) the master event method, and (3) the joint hypocenter determination method, were analyzed with respect to their application to regional (epicenter-to-station distances from 10 to 500 km) and local (epicentral distances from 0 to 70 km) seismic network recording stations. It was found that the joint hypocenter technique can significantly correct for inadequacies in assumed velocity models by simultaneous computation of station adjustments. Earthquakes located using the joint hypocenter technique are located more precisely with respect to each other. The joint epicenter location technique was then applied to relocations of epicenters in southern Utah below 40 degrees N latitude with the intent of resolving some spatial relationships of epicenter occurrence on this regional scale. Earthquakes in Utah occur in a diffuse zone of 150 km to 200 km wide that coincides with the physiographic boundary of the Colorado Plateau. Swarm activity is prevalent where the zone changes from a north-south to a southwest orientation, but except for a cluster northwest of Cedar City, the only major alignments of epicenters are north-south.

On a local scale an objective was in determining some relationships of epicenters to two known geothermal resource areas: Cove Fort and Roosevelt Hot Springs, Utah. The differences in the character of these two areas as expressed by the earthquake occurrence was emphasized by the jointly determined relocations. Areas of possible active east-west and northeast faulting near Cove Fort were delineated, but it was verified that seismic activity in the Milford area is sparse. Attention was given to possible effects of lateral inhomogeneities in the velocity model on earthquake locations. Evidence for substantial lateral variations in the velocity models used in the locations was contributed from an analysis of P-wave residuals from broadside refraction data. Arrival times of P-waves were found to be early near the Opal Mound fault, possibly due to local siliceous cementation of alluvium. Contacts beetween the gneissic basement and the granitic pluton could account for some of the observed delays in P-wave arrival times. Effects of lateral velocity variations on earthquake locations were minimized by precise relocation using joint hypocenter determination, thus aiding is defining seismically active areas. Results of earthquake relocation for the Cove Fort-Roosevelt Hot Springs KGRA's and adjacent areas indicate that methods of systematically and precisely locating earthquakes can better contribute to understanding the seismicity and velocity structure and the relationships of geothermal areas to regional and local stress fields and geologic characteristics.