• DAS
• DFOS
• seismic location
• waveform coherence methods

Distributed Fiber Optic Sensing (DFOS) technologies provide unprecedented capabilities in
sampling the seismic wavefield, enabling new approaches for seismicity data analysis.
Among the different seismic data analysis procedures, earthquake location is one of the tasks
that can benefit from the high spatial density of DFOS datasets. In particular, coherence-based
techniques are the ideal approaches for such kind of data.
Despite this potential, several factors affect the direct application of coherence-based earthquake
location to DFOS data, including spatial sampling, limited azimuthal coverage, sources of noise,
and the physical quantity measured (strain/strain rate).
To evaluate the feasibility of coherence-based earthquake location on DFOS data, a series of
tests have been implemented using synthetic simulations mimicking a standard microseismic
monitoring infrastructure with three wells. Specifically, the tests analyze the impact of differ-
ent spatial sampling, acquisition geometries, real noise contamination, on coherence matrices.
Moreover, different characteristic functions are implemented for the same goal. Overall, the
focus is on assessing the accuracy of event location under these assumptions.
Results indicate that for well-based monitoring systems, spatial sampling is not a serious limiting
factor, as locations are retrieved using a reduced number of sensors. This observation suggests
stacking neighbor channels to lower computational and memory costs, such as for expensive 3D
traveltime lookup table computations. Moreover, we show how coherence-based approaches are
robust against real noise contamination, with the hyperbolic cosine function providing better
coherence images compared to standard stacking functions.
Overall, these initial tests show the need for automatic data selection techniques that reduce
problem dimensionality while ensuring good geometrical coverage. Furthermore, the results
highlight the potential of coherence-based techniques, also exploiting different characteristic
functions of raw data, as promising location methods for next-generation hybrid networks that
integrate both fiber and conventional seismometers.