Lg attenuation, frequency dependence and relative site response of the western United States as revealed by the EarthScope Transportable Array

Lg attenuation of the western United States is estimated using regional waveforms recorded by EarthScope's Transportable Array, the Berkeley Digital Seismograph Network and the Caltech Regional Seismic Network. Lg attenuation is determined using the two-station method (TSM). 292 events ranging from magnitude 4 to 6.5 in a time range for the years between 2005 and 2010 are used. We have tomographically mapped variations in effective Lg Q at narrow bands with central frequencies of 0.5, 1, 2 and 3 Hz. The relative site responses of individual stations are determined using a reverse TSM at these central frequencies. Both the two-station and reverse TSMs have the advantage of effectively removing site and source effects without requiring a priori models while mapping lateral heterogeneities in the crust. Regions of high attenuation are found along the Snake River Plain, along the northern California-Oregon border, throughout the Basin and Range, Walker Lane, Wasatch Fault and the Rio Grande Rift. Regions of low attenuation are found in the Colorado Plateau, Rocky Mountains, central Nevada and the Columbia Plateau. Overall, Lg Q was found to have a power-law frequency dependence, with an increase in central frequency resulting in an increase in Q. Relative site responses were found to vary strongly with local geology. Hard rock areas such as the Sierra Nevada Mountains and the Snake River Plain show a consistent pattern of negative site response (i.e. deamplification), while areas with a higher degree of sediments, such as the grabens of the Basin and Range, show positive site response (i.e. amplification). Finally, site responses were found to vary with frequency, becoming less variable and more pronounced as frequency increases.