Large-scale instability of volcanic edifices, which influences eruption style and timing,
is dependent on the strength and fracture state of the composite rock. We can estimate the
range of rock strength using the Geological Strength Index, which is based on the character
and distribution of fractures and discontinuities in a rock. This is usually based on qualitative
observations of an in-situ rock mass. Small-scale strength of a rock sample can also be
assessed in the laboratory via uniaxial compressive strength (UCS) testing. This small-scale
strength determined in the lab does not necessarily reflect the rock mass strength as it does
not include the possibility of fractures in the rock mass.
Numerical models have used different methods of upscaling these laboratory values
to generate appropriate large-scale properties, such as bulk reduction of UCS values, and
incorporation of natural fracture states that reduce the UCS. Appropriate strength of volcanic
rock is important to create reliable, large scale models of volcanic instability. Here, we create
2D models in Particle Flow Code, where we incorporate a discrete fracture network that
reproduces qualitative observations from GSI. We validate the relationship between GSI and
UCS using modeled laboratory testing. Then we incorporate fracture networks into
large-scale volcanic instability models to assess the impact of fracture distribution on
volcanic deformation.