Logo: to the web site of Uppsala University

The distribution of waiting times between time-neighbouring events for a time series obeying the Omori law is examined theoretically and numerically with the aim of understanding the characteristics of these distributions, how these characteristics change (e.g. scale) with the parameters of the Omori series, and thus how empirical waiting time data may be correctly interpreted. It is found that the waiting time distribution, for a single Omori aftershock sequence, consists in general of two power law segments followed by a rapid decay at larger waiting times. The analyses are illustrated using real data from the SIL network on Iceland. This data often shows characteristics predominantly consistent with the Omori law, but there are significant exceptions. We conclude that waiting time distributions and related statistical analysis has meaningful potential for the analysis of earthquake data sets, as a step towards developing physical models of the earthquake process.

Recently, several authors have used waiting time distributions for large earthquake data sets to draw conclusions regarding the physics of earthquake processes. We show, theoretically and by simulation, that a characteristic kink in observed waiting time distributions does not have the physical significance of separating correlated and uncorrelated earthquakes. It also follows from our discussion that the Omori law is not trivially related to a proposed scaling law and that caution must be taken before the spatial scaling exponent of the law is interpreted as a fractal dimension of seismicity.

The Katla volcano, overlain by the Mýrdalsjökull glacier, is one of the most active and hazardous volcanoes in Iceland. Earthquakes show anomalous magnitude-frequency behaviour and mainly occur in two distinct areas: within the oval caldera and around Goðabunga, a bulge on its western flank. The seismicity differs between the areas; earthquakes in Goðabunga are low frequency and shallow whereas those beneath the caldera occur at greater depths and are volcano-tectonic. The seismicity shows seasonal variations but the rates peak at different times in the two areas. A snow budget model, which gives an estimate of the glacial loading, shows good correlation with seismic activity on an annual scale. Data recorded by the permanent network South Iceland Lowland (SIL), as well as by a temporary network, are used to invert for a 3D seismic velocity model underneath Eyjafjallajökull, Goðabunga and the Katla caldera. The tomography resolves a 15 km wide, aseismic, high-velocity structure at a depth of more than 4 km between the Eyjafjallajökull volcano in the west and the Katla volcano in the east. Anomalously low velocities are observed beneath the Katla caldera, which is interpreted as being a significantly fractured area of anomalously high temperature.

Repeating long-period (lp) earthquakes are commonly observed in volcanic regions worldwide. They are usually explained in terms of a volcanic source effect or anomalous propagation through the volcano. Recently, large lp events have also been associated with the motion of massive ice streams. Our joint analysis of climatic and new seismic data shows that small lp events observed at Katla volcano, Iceland, are in fact related to ice movement in a steep outlet glacier and not, as previously thought, to volcanic intrusive activity. The over 13000 lp events recorded since 2000 are consistent in character and magnitude with seasonal changes of the glacier. As the current global warming trend could cause similar earthquake sequences at other glacier covered volcanoes, identifying them as glacial rather than eruption precursors is vital.

Strong and repeating very long period (vlp) signals (10-30 sec) on and around the western Mýrdalsjökull ice cap, which covers the Katla volcano in South Iceland, are recorded simultaneously with long period (lp) events of magnitude >2. The signals are recorded on 6 broadband and 2 higher frequency three-component seismic stations. The vlp signals are linearly polarized and the horizontal polarization direction points to the same source as the lp events, which have recently been interpreted as being generated by ice falls over an escarpment in a steep outlet glacier. The vlp events are registered on several different instrument types, which implies that they must be due either to a generic problem with the instruments or that they are real signals. They do not appear to be consistent with P-, S- or surface wave behaviour.Similar signals elsewhere have been interpreted as the combined effect of static displacement and tilt caused by ground deformation due to magma movements within a volcano. The three-component data are therefore modelled as the combined effects of tilt and vertical translation caused by the ice movements. Simple deformation modelling shows that the observed vlp signals are inconsistent with the ice fall source in the relative magnitudes of the vertical and radial displacements and, in line with results from elsewhere, the magnitudes of the observed effects are large relative to the model response. Thus the origin of the vlp signals remains enigmatic.Our analysis suggests that it is unlikely that the signal is an instrumental artefact. Possible physical origins are discussed.