Tsunamis from prospected mass failure on the Marsili submarine volcano flanks and hints for tsunami hazard evaluation

The Marsili Seamount (Tyrrhenian Sea, Italy) is the largest submarine volcano in the Mediterranean Sea, located in the middle of the Marsili Basin, facing the Calabrian and Sicilian coasts on its eastern side, and the coasts of Sardinia on the opposite side. It has erupted in historical times, and its summit crest is affected by widespread hydrothermal activity. This study looks at mass failures taking place at different depths on the flanks of the volcano and estimates their associated tsunamigenic potential. Mass failure, tsunami generation, and propagation have been simulated by means of numerical models developed by the Tsunami Research Team of the University of Bologna. In all, we consider five cases. Of these, three scenarios, one regarding a very small detachment and two medium-sized ones (between 2 and 3 km3 failure volume), have been suggested as possible failure occurrences in the published literature on a morphological basis and involve the north-eastern and north-western sectors of the volcano. The two additional cases, one medium-sized and one extreme, intended as a possible worst-case scenario (volume 17.6 km3), affecting the eastern flank. Results indicate that small-volume failures are not able to produce significant tsunamis; medium-size failures can produce tsunamis which dangerously affect the coasts if their detachment occurs in shallow water, i.e., involves the volcano crest; and extreme volume failures have the potential to create disastrous tsunamis. In all the simulations, tsunami waves appear to reach the Aeolian Islands in around 10 min and the coasts of Calabria and Sicily in 20 min. This study highlights that there is a potential for dangerous tsunamis generation from collapses of the Marsili volcano and as a consequence a need to intensify research on its status and stability conditions. More broadly, this investigation should also be extended to the other volcanic seamounts of the Tyrrhenian Sea, since their eruptive style, evolution, and tsunamigenic potential are still poorly known.

Internal Erosion of Soil Pipes: Still More Questions Than Answers

<p>Internal erosion of soil pipes can be a very important process in gully erosion as well as other mass failure events such as sinkholes, landslides and levee/dam breaching. Flow through preferential flow paths such as macropores can be rapid enough to exceed the soil critical shear stress and cause detachment of particles from the walls of the flow path, i.e. internal erosion. Development of a soil pipe from enlargement of a macropore results in more rapid flow and thus greater internal erosion, particularly mass failure of aggregates from pipe walls and roofs. If the sediment transport capacity of the pipe is exceeded, the pipe will plug causing back-pressure to build up within the soil pipe, which can foster hillslope instability. However, limited research has been conducted on particle and aggregate detachment within soil pipes as well as transport of sediment through soil pipes. The objective of this paper is to present observations of little known and poorly described processes involved in pipeflow and the resulting internal erosion of soil pipes. Many of the processes involved in internal erosion of soil pipes are assumed based upon processes observed in surface and stream erosion studies but are so poorly quantified for soil pipes that they are yet to be transferable. For example, the role of solution chemistry on sediment detachment from pipe walls has been quantified to a limited extent but little has been done on the effects of seepage forces on particle detachment in pipes and even less done on sediment transport capacity of soil pipes. Recent advances have included: development of suspended sediment and bedload rating curves for soil pipeflows but the results are crude and warrant further study. Quantification of the interactive effects of surface flow in channels with flow through soil pipes below channels on headcut migration and gully widening is in its infancy. Other processes, such as air-entrapment in creating or temporarily plugging pipes, have been suggested as important but lack quantification. These processes and others combine to result in internal erosion of soil pipes but they must be better understood and quantified in order to develop the next generation of soil erosion models and landscape morphology prediction technologies.</p>