In this study, the PIs are analyzing data recorded by the hundreds of seismometers that constitute USArray; however, instead of studying signals from earthquakes, the PIs are analyzing seismic energy that is created by the interference and interaction of ocean waves. This process generates an ambient field of background noise, often referred to as microseisms, that can be used to image Earth's interior, and to detect changes in Earth structure related to volcanic and tectonic processes. In particular, the PIs are using the USArray as a giant, seismic antenna to locate and characterize microseismic sources throughout the world's oceans. Interestingly, many of the source regions occur in the deep waters of ocean basins, far away from the reflecting coastlines that are traditionally considered to be the dominant source of microseisms. Analysis of microseismic energy from these deep water sources has the potential to illuminate those portions of Earth's interior that are poorly sampled by earthquakes, which overwhelming occur in narrow belts along the boundaries of tectonic plates.
This project analyzed seismic data from NSF's USArray Project to better understand the nature of ground motion that is created by wind-driven ocean waves. Although seismometers are commonly deployed to record vibrations created by earthquakes, their extraordinary sensitivity to ground motion means that vibrations from a wide variety of other sources, including ocean waves, are commonly recorded. Seismologists refer to ocean-wave generated vibrations as seismic noise, or microseisms. Over the last ten years microseisms have been exploited to make high-resolution images of the geological structure of Earth's crust and mantle, to monitor monthly-to-yearly changes in the geological structure near volcanoes and hydrothermal systems, and to monitor yearly-to-decadal changes in the ocean wave climate. A better understanding of when, where, and how microseisms are generated helps seismologists refine these techniques. In this project we processed USArray microseism data with a combination of techniques that included power spectral density estimation, principal component analysis, complex trace analysis, and migration. Surprisingly, we found that the amplitude of microseisms is more strongly related to the geologic structure near the seismometer than the proximity of the seismometer to a coastline. For example, the "noisiest" part of North America is the Williston Basin of eastern Montana and the western Dakotas, even though this region is more than 1500 km from the nearest coastline. It turns out that the thick layers of sediment in this and other basins efficiently trap and amplify microseism energy. We also found that individual storms, such as superstorm Sandy that occurred in the fall of 2012, radiate bursts of microseismic energy when they sharply change direction. Another outcome from this project is the discovery that the dominant period of a microseism depends upon the region of the ocean in which it was created. Average periods vary from 7–8 s in the southwestern U.S., to 5–6 s in the northwestern and central U.S., to 4–5 seconds in the Midwestern and northeastern U.S, to 2–4 seconds in the southeastern U.S. The long-term eastward motion of the atmosphere leads to longer-period microseisms on the west coast (windward side) of North America compared to the east coast (leeward side), with the relatively sheltered Gulf of Mexico having the shortest periods. Two or more of these oceanic source regions are often active at the same time, explaining the splitting of the double-frequency microseism spectral peak into distinct sub-peaks that is observed when high-resolution spectra are calculated. These and other results from this project were presented at scientific meetings and submitted to professional academic journals. Our findings were also publicized in traditional media outlets. This project provided partial support for a Ph.D. student, an M.S. student, and an undergraduate researcher.
