The fine structure of double‐frequency microseisms recorded by seismometers in North America

We performed a frequency‐dependent polarization analysis on ambient seismic energy recorded by 1768 USArray Transportable Array (TA) seismometers for the time period of 1 April 2004 through 31 October 2014. The seismic energy has strong seasonal variations in power and polarization at essentially all stations; however, the annual variation is much smaller. One year of data is sufficient to determine the average properties of the ambient seismic wavefield at a particular site. The average power and dominant period in the double‐frequency (DF) microseism band, defined here as periods of 2–10 s, vary significantly and coherently across North America. Proximity to a coastline generally leads to increased DF microseism amplitude, but site geology is much more important, with sedimentary basins having especially large DF amplitudes. The western U.S. as a whole has longer dominant DF periods than the central and eastern U.S., with the southeastern U.S. having the shortest dominant DF periods. Power spectral density estimates at many TA stations show a splitting of the DF microseism peak into two distinct subpeaks. This has been observed previously in data recorded by ocean bottom seismometers, with the shorter‐period DF peak attributed to the local sea and the longer‐period DF peak attributed to more distant, coastally generated microseisms. In the case of the land‐based TA data analyzed here, the DF splitting arises from simultaneous microseism generation at various source areas (Pacific Ocean, Atlantic Ocean, and Gulf of Mexico) with distinct, preferred excitation frequencies. DF microseism source properties derived from global models of ocean wave interaction support this interpretation.