Earthquake induced static stress change in promoting eruptions

The aim of this PhD work is to study how earthquakes could favour new eruptions, focusing the attention on earthquake-induced static effects in three different case sites. As a first case site, I studied how earthquake-induced crustal dilatation could trigger new eruptions at mud volcanoes in Azerbaijan. Particular attention was then devoted to contribute to the understanding of how earthquake-induced magma pathway unclamping could favour new volcanic activity along the Alaska-Aleutian and Chilean volcanic arcs, where 9 seismic events with Mw ≥ 8 occurred in the last century. Regarding mud volcanoes, I studied the effects of two earthquakes of Mw 6.18 and 6.08 occurred in the Caspian Sea on November 25, 2000 close to Baku city, Azerbaijan. A total of 33 eruptions occurred at 24 mud volcanoes within a maximum distance of 108 km from the epicentres in the five years following the earthquakes. Results show that crustal dilatation might have triggered only 7 eruptions at a maximum distance of about 60 km from the epicentres and within 3 years. Dynamic rather than static strain is thus likely to have been the dominating “promoting” factor because it affected all the studied unrested volcanoes and its magnitude was much larger. Regarding magmatic volcanoes, the entire dataset includes a total of 51 eruptions following 9 large earthquakes (Mw ≥ 8); 33 represent first new eruptions occurred at each single volcano. Comparison of the eruption rate before and after each earthquake suggests that 26 out of the 33 first new eruptions have a positive relation with the studied earthquakes; 13 out of 26 represent awakening events, which are first new eruptions occurred at volcanoes with not-continuous eruptive activity that had no eruptions in the five years before the earthquake. I followed a novel approach that resolves the earthquake-induced static stress change normal to the magma pathway of each volcano instead of considering the general crustal volume. I also considered other parameters that may contribute to control eruptions, such as magma composition and viscosity, magma chamber depth and local tectonic settings. The sensitivity analysis performed for the 2010 Chile earthquake shows that the N-S- and NE-SW-striking magma pathways suffered a larger unclamping in comparison with those striking NW-SE and E-W. Magma pathway parallel or subparallel to the strike of modelled faults suffered the highest unclamping, and this was also evident in the Alaska-Aleutian arc. Magma pathway geometry controls the magnitude of the static stress change induced by large earthquakes, with differences of up to 8 times among magma-feeding planes of different orientation at the same volcano. This range of diverse values is larger for the volcanoes closer to the epicentre. The possible error in the estimate of magma chamber depth has a minimum effect on the results since the sensitivity analysis shows that the range of stress changes with depth is about 1.5 orders of magnitude smaller than the range linked to variations in the magma pathway strike. Result suggest that unclamping effect promoted eruptions that occurred at not-continuously erupting volcanoes (Type B) in a range of 157-543 km, while awakening under unclamping occurred in a range of 157-353 km. Regarding the time-gap, unclamping promoted eruptions at Type B volcanoes and awakening in a time window of 2 days-3 years. In the studied cases, it was also noted that unclamping enhanced eruptions preferentially at volcanoes with a deep magma chamber (> 5 km). Regarding the role of magma pathway geometry, results suggest that 4 awakening events occurred along magma pathways parallel or subparallel to the σHmax, and they are always under unclamping. Eleven awakening events occurred along magma pathways that have an angle from 40° to 90° respect to the σHmax, thus unclamping favoured 8 of these events. Finally, based on the results from this work, I suggest that earthquake-induced stress change can favour magma rise by imparting stress field modifications of the deviatoric stress acting on magma pathway in a percentage up to 5-10 %, increasing the efficiency of magma rise and consequent dyke intrusion also where the tectonic regime doesn’t favour magma rise (e.g. in the older and thicker crust in Chile under contractional tectonics).