Page 301 - Caldera Volcanism Analysis, Modelling and Response

276 J. Martı ´ et al.

Theoretical models demonstrate that ring faults controlling caldera collapse may
originate from magma chambers with either overpressure or underpressure. Nature
exhibits calderas which may have formed by either condition: (i) there are examples
of large calderas, usually located in continental settings, where the analysis of the
caldera deposits supports the idea of a rapid accumulation of caldera-forming
deposits without any preceding central vent eruption that could have resulted
in an a priori chamber decompression (Sparks et al., 1985; Martı ´, 1991; Petrinovic,
1999; Aguirre-Dı ´az and Labharte-Herna ´ndez, 2003). (ii) There are also numerous
examples of caldera collapses preceded by plinian central vent eruptions that
account for a significant decompression of the magma chamber prior to the
initiation of caldera subsidence (Williams, 1941; Mahood, 1980; Bacon, 1983;
Heiken and McCoy, 1984; Heiken et al., 1990; Hildreth and Fierstein, 2000).
There are still some aspects that need to be studied in more detail in order to
confirm the validity of the numerical models. One of the main uncertainties
corresponds to the rheological behaviour of the host rock. Most of the existing
theoretical models assume a purely elastic behaviour for the magma chamber walls.
This is clearly an oversimplification, particularly in long-standing volcanic systems
with a series of inflation and deflation episodes and ensuing thermal effects of
magma replenishment on host rocks. Development of theoretical models using
more realistic rheologies is definitively needed in order to better constrain the
mechanics of caldera formation.

5.3. Implications of geophysical images on the assessment of caldera
processes

Despite their limitations as outline above, geophysical images provide invaluable
insights into the interior of calderas and, thus, represent an important contribution
to our understanding of subsurface dynamics. While geophysical imaging is a
universal and widely applied tool for assessing the structure of and dynamics in oil
or gas reservoirs, images of the interior of collapse calderas are however still rare,
given the fact that more than 100 calderas have shown signs of unrest in the past
decades (Newhall and Dzurisin, 1988). As a result, our knowledge on the
subsurface is far from complete, and many questions remain regarding the timescale
and amount of magma replenishment beneath active calderas. Other questions
pertaining to short-wavelength anomalies, for example, the extent of fault zones or
hydrothermal systems or the existence of fluid (?magma)-filled pockets, are difficult
to answer with a caldera-wide distribution of recording devices but could be
answered by dedicated high-precision surveys at selected areas that have proven to
show marked anomalies during conventional surveys. Recent archaeological studies
may serve as an example (Cardarelli and de Nardis, 2001).
Another key implication of geophysical imaging concerns the question of the
origin of calderas. A current subsurface image of an active caldera rarely mirrors
the subsurface structure prior to the formation of a collapse caldera. Although it is
tempting to adopt a geophysical image of a caldera-wide magmatic reservoir of
given shape and depth as a mirror image of the magma chamber causative for the
collapse, in order to validate results from other investigations, we have to be