Page 18 - Caldera Volcanism Analysis, Modelling and Response

Preface xvii

subsurface images (Aprea et al., 2002; Nishimura et al., 1997), and second, by
recording time series to capture the dynamic behaviour of calderas (Tikku et al.,
2006; Gottsmann et al., 2007). Ground deformation monitoring has evolved into
a standard technique by either land-based or air-borne instrumentation. Combined
with gravimetry (to detect mass changes at depth), geodetic investigations are
regarded important tools for the quantification of precursor signals during unrest
(Rymer and Williams-Jones, 2000).
However, some important aspects of the dynamics and the structure of collapse
calderas still remain poorly understood. A collapse caldera is defined as the volcanic
depression that results from the disruption of the geometry of the magma chamber
roof due to down faulting during the course of an eruption (Williams, 1941; Smith
and Bailey, 1968; Williams and McBirney, 1979; Smith, 1979; Lipman, 1997,
2000). This definition embraces a diversity of structures, processes and deposits,
but we are still far from having a complete understanding of the causes that drive
a shallow magmatic system to caldera formation. A number of problems are still
unresolved:
Do the different caldera morphologies observable in the field really represent
different genetic processes related to the shallow magmatic system or are they
merely a product of variations in the collapse mechanism due to differences in the
mechanical behaviour of host rocks and tectonic structures?
How much does a geophysical image of the internal structure of a restless caldera
correspond to the original subsurface architecture prior to the caldera-forming
event?
Is caldera unrest controlled by the deep caldera structure or by shallow structures,
or a combination of both?
Is a present shallow magmatic system beneath an active caldera an analogue to
the magmatic system during the caldera-forming events, or are both systems
completely unrelated?

These are just a few of the crucial, yet unresolved problems pertaining to
collapse calderas and to which we do not yet have satisfying answers.
One of the most important future tasks for volcanologists is to identify
precursors to potentially devastating caldera-forming eruptions. Any cluster of
active volcanoes should be evaluated as a possible site of a growing upper crustal
magma chamber that could lead to a future caldera formation (Lipman, 2000).
However, herein lies the crux in determining whether caldera-collapse is an
inevitable result of an ensuing eruption; caldera-formations are rare events in
human timescales and thus our current knowledge of precursory signals is very
limited if not absent. Not enough data or information exist to clearly define
precursory events as indicators for impending major eruptions and critical
parameters indicating whether a caldera-collapse will or will not occur during
a major eruption are still mostly unknown. The same applies to the question of
characteristic time intervals between the occurrence of anomalous (precursory)
signals and an eruption. From the experience of several important periods of caldera
unrest over the past decades (e.g., Yellowstone, Long Valley, Campi Flegrei) it is
safe to state that most periods of unrest do not terminate with a volcanic eruption,