ETH, Zurich, Switzerland

see Abstract Volume

see Abstract Volume

see Abstract Volume

Forecasting Seismicity on Local to Regional Scale

Solving the Woes of Statistical Seismology: The Community On-Line Resource for Statistical Seismicity Analysis (CORSSA)

Collaboratory for the Study of Earthquake Predictability

A variety of seismic signals representing different physical mechanisms precedes volcanic eruptions. The most meaningful signals are high- and low-frequency earthquakes and volcanic tremor that have tentatively been related to fracturing and magma transport in the volcanic edifice. We provide experimental support for this association by reproducing magma migration while recording seismic signals. Opening fractures emit high-frequency acoustic events, while the switch to low frequency and harmonic tremor accompanies the flow of the melt in the fractures. Discerning between these seismic signals in nature can signifi cantly refine volcanic hazard evaluation.

Observations of seismic surface waves provide the most important constraint on the elastic properties of the Earth’s lithosphere and upper mantle. Two databases of fundamental mode surface wave dispersion were recently compiled and published by groups at Harvard (Ekström et al., 1997) and Utrecht/Oxford (Trampert and Woodhouse, 1995, 2001), and later employed in 3-d global tomographic studies. Although based on similar sets of seismic records, the two databases show some significant discrepancies. We derive phase velocity maps from both, and compare them to quantify the discrepancies and assess the relative quality of the data; in this endeavour, we take careful account of the effects of regularization and parametrization. At short periods, where Love waves are mostly sensitive to crustal structure and thickness, we refer our comparison to a map of the Earth’s crust derived from independent data. On the assumption that second-order effects like seismic anisotropy and scattering can be neglected, we find the measurements of Ekström et al. (1997) of better quality; those of Trampert and Woodhouse (2001) result in phase velocity maps of much higher spatial frequency and, accordingly, more difficult to explain and justify geophysically. The discrepancy is partly explained by the more conservative a priori selection of data implemented by Ekström et al. (1997). Nevertheless, it becomes more significant with decreasing period, which indicates that it could also be traced to the different measurement techniques employed by the authors.