International Workshop on Earthquakes in North Iceland

Húsavík, North Iceland, 6-8 June 2013

The thermodynamics of fissure swarms and fault zones: application to

the Husavik-Flatey Fault
Agust Gudmundsson1, Nahid Mohajeri2
1
Department of Earth Sciences, Royal Holloway University of London (rock. fractures@googlemail.com)
2
Department of Geography, University College London (mohajeri.nahid@gmail.com)

Fault zones are open thermodynamic systems. They receive energy, primarily elastic energy, as
input from the surroundings and transform it partly into surface energy and heat (due to friction)
during earthquake rupture. Earthquake ruptures are characterised by unstable fracture growth or
propagation. Unstable fracture propagation occurs when the energy released during the propagation
is greater than the energy input needed to overcome the surface energy of the main fault plane as
well as the surface energies of the associated fractures in damage zone and core of the fault zone
(e.g., Gudmundsson, 2011). The elastic energy associated with fault propagation has mainly two
sources. One is the strain energy that accumulates in the fault zone and its surrounding prior to the
earthquake rupture and is primarily related to the plate movements. The other source is the work
done by the displacement of the boundaries or margins of the fault zone during the rupture
propagation (Gudmundsson, 2012). Both sources contribute to the elastic energy input that is
available to drive the rupture propagation.

Fault zones such as the Husavik-Flatey Fault are complex systems whose growth and general
evolution are still rather poorly understood. It is generally accepted that faults normally initiate
from 'flaws' or weaknesses in the rocks, but their subsequent development and growth and their
seismogenic activity are not well understood. Here we present new results on the slip/displacement-
length scaling relations, both for faults and co-seismic ruptures. Using these data, together with data
from the literature and analytical and numerical models, we provide a general growth model for
faults (cf. Gudmundsson et al., 2013) and apply the results to the Husavik-Flatey Fault and
associated fissure swarms in Iceland.

Following an earthquake in a fault zone, the co-seismic rupture length and the slip are commonly
measured. Similarly, in a structural analysis of major faults, the total fault length and displacement
are measured when possible. Typical rupture length - slip ratios are orders of magnitude larger than
typical fault length-displacement ratios. So far, however, most of the measured co-seismic ruptures
and faults have been from different areas and commonly hosted by rocks of widely different
mechanical properties (which have strong effects on these ratios). The compilation of new data
presented here, mostly from areas with the same mechanical properties, shows that the average
rupture-slip ratio is about 150-times larger than the length-displacement ratio in the same fault
zones. We propose that the differences between the length-slip and the length-displacement ratios
can be partly explained by dynamic Young's modulus of fault zone being 101-2-times greater than its
static modulus. In this model, the dynamic modulus controls the length-slip ratios whereas the static
modulus controls the length-displacement ratio. We suggest that the common aseismic slip in fault
zones is partly due to adjustment of the short-term seismogenic length-slip ratios to the long-term
length-displacement ratios. Fault displacement is here regarded as analogous to plastic flow, in
which case the long-term displacement can be very large so long as sufficient shear stress
concentrates in the fault. This model explains, partly at least, the difference in the
slip/displacement-length scaling relations between co-seismic ruptures and faults and also explains
slow earthquakes and aseismic slip, features that are now known to be very common in active fault
zones.

When applied to the Husavik-Flatey Fault, we would expect much of its slip to be aseismic ('creep').
This follows particularly because the fault has more than one very thick (ca. 20 m) fault cores
composed of comparatively soft breccias, in addition to a damage zone that reaches several
kilometres in thickness (Gudmundsson, 2007). Part of this creep is, according to the model above,
adjustment of the length-slip ratio, controlled by the dynamic Young's modulus, to the length-
displacement ratio, which is controlled by the static Young's modulus.

We have also applied the thermodynamic principles and energy considerations to the fissure
swarms of Iceland (Mohajeri and Gudmundsson, 2012; Gudmundsson and Mohajeri, 2013), such as
the swarms in North Iceland that are connected with the Husavik-Flatey Fault. The fractures
(tension fractures and normal faults) of the swarms show heavy-tailed or power-law length
distributions. Thermodynamic principles, while forming the basis of fracture mechanics, have
previously not been much used in connection with power-law size distributions of rock fractures.
Here we provide power-law scaling exponents and the calculated entropies of fracture networks
from Holocene fissure swarms and show that the associated fracture networks can be divided into
populations based on 'breaks' (abrupt changes) in the scaling exponents (as determined from log-log
plots of fracture lengths). The breaks, we suggest, are related to the comparatively long and deep
fractures changing from tension fractures into normal faults and penetrating the contacts between
the Holocene lava flows (commonly several hundred metres thick) and the underlying and
mechanically different Quaternary rocks. The results show a strong linear correlation (r = 0.84)
between the population scaling exponents and the associated entropies. The correlation is partly
explained by the entropy (and the scaling exponent) varying positively with the arithmetic average
and the length range (the difference between the longest and the shortest fracture) of the populations
in each fracture network. We propose that the power-law size distributions of fractures are a
consequence of energy requirements for fracture growth. More specifically, only those fractures that
are favourably oriented in relation to the time-average plate-boundary stress field receive enough
energy input to become large; the other fractures remain short.

References

Gudmundsson, A., 2007. Infrastructure and evolution of ocean-ridge discontinuities in Iceland. J. Geodyn., 43, 6-29.
Gudmundsson, A., 2011. Rock Fractures in Geological Processes. Cambridge University Press, Cambridge.
Gudmundsson, A., 2012. Strengths and strain energies of volcanic edifices: implications for eruptions, collapse calderas
and landslides. Nat. Hazards Earth Syst. Sci., 12, 2241-2258.
Gudmundsson, A., De Guidi, G., Scudero, S., 2013. Length-displacement scaling and fault growth. Tectonophysics (in
press).
Gudmundsson, A., Mohajeri, N., 2013. Thermodynamic aspects of the development of fracture networks. Bull. Geol.
Soc. France (in press).
Mohajeri, N., Gudmundsson, A., 2012. Entropies and scaling exponents of street and fracture networks. Entropy, 14,
800-833.