Page 339 - Caldera Volcanism Analysis, Modelling and Response

314 Agust Gudmundsson

is located in a single, thick layer with stiffness different from that of the other layers. The
crustal segment hosting the chamber is either 20 or 40 km wide but has a constant
thickness of 20 km. The loading conditions considered are crustal segments subject to:
(a) underpressure (lack of magmatic support) of 5 MPa; (b) tensile stress of 5 MPa;
(c) excess magmatic pressure of 10 MPa at the bottom of the crustal segment (doming of
the volcanic field containing the chamber) and (d) combination of tension and doming. In
all the models, the magma-chamber top is at 3 km depth; the diameter of the sill-like
chamber is 8 km (its thickness is 2 km) while that of the spherical chamber is 4 km.
The main results are as follows: (1) Underpressure and excess pressure in a shallow,
crustal chamber normally result in dyke injection rather than caldera formation. (2) For
doming or tension, a spherical magma chamber normally favours dyke injection rather
than ring-fault initiation. However, when the spherical chamber is located in a very soft
(10 GPa) layer, the local stress field may be suitable for caldera-fault formation. (3) For a
sill-like chamber in a 20-km-wide volcanic field, a ring fault may be initiated either
during horizontal tension or a combination of tension and doming. (4) For a sill-like
chamber in a 40-km-wide volcanic field, doming alone is sufficient to initiate a caldera
fault. The results indicate that the local stresses in composite volcanoes most likely to
initiate caldera faults are associated with sill-like chambers subject to tension, doming
or both.

1. Introduction

In the last few decades it has become generally realised that collapse calderas
are not confined to volcanic regions on Earth; they are common structures on
several other solid planets and satellites. Known calderas in the solar system range in
diameter from about 1.6 km to nearly 300 km and have been observed on Earth,
Venus, Mars and Io (a satellite of Jupiter). Calderas may exist on many other
planetary bodies, for example the Moon and Mercury, where most of the circular
structures, however, are impact craters.
Traditionally, collapse calderas have been defined as circular or elliptical volcanic
depression with a maximum diameter similar to or larger than about 1 mile or
1.6 km (Macdonald, 1972). When this lower limit is used, pit craters on Earth,
which rarely exceed 1 km in diameter, can be distinguished from calderas (Okubo
and Martel, 2001). While most calderas are actually elliptical in plan view, and many
have rectangular margins along parts of, or the entire, periphery, the slip plane
along which the main caldera subsidence takes place will here be referred to as a
ring fault (Figure 1).
Clearly, the ring fault of a collapse caldera is a fracture; more specifically it is a
shear fracture, that is, a fault. And as such it can only form under certain special
stress conditions. Thus, to be able to forecast the likelihood of caldera formation or
slip during an unrest period, we must understand the stress conditions that favour
ring-fault formation. These stress conditions are still poorly understood and remain
the main unsolved problem regarding collapse calderas.
Since ring faults are faults, we might expect that their formation could be
understood within the general framework of earthquake mechanics. But there appear