Using parallel computing for seismo-volcanic event location based on seismic amplitudes

Volcanoes are very complex geophysical systems where fluids of different nature interact with porous rock at different physical conditions and within a complex matrix of conduits. Two types of seismicity are generated by this complex interactions. The first type is characterized by fracture of the elastic media, in which we have the volcano-tectonic events (VT) that produce two distinctive phases: a compressional phase (P wave) and a shear wave (S wave) that travels with different velocities within solid media. The second type is characterized by low frequencies, in which we have a wide variety of long period events (LP) and volcanic tremors. These signals are produced by fluid motion within restricted paths and have normally emergent onsets and no distinctive P or S wave phases. Classical earthquake source location procedures take advantage of the distinctive phases and their different propagation velocity. However, for LP events and tremors, those procedures can not be used. Therefore, complex algorithms have to be applied, demanding much more computer resources and time than the classical location methods. In this work, we present the analysis and design of a LP and tremor location application based on amplitude decay. We demonstrate that the algorithm is highly parallelizable allowing to develop a parallel implementation using the Python programming language and the de-facto standard for parallel computing, the MPI standard. We show experimentally that it exhibits almost linear scalability with respect to the number of events and the number of cores.