Page 17 - Caldera Volcanism Analysis, Modelling and Response

xvi Preface

consequence, not the cause of the collapse. At Kilauea Volcano in Hawaii, the
concept of caldera formation by post-eruptive collapse was also recognised by
Dutton (1884).
At the beginning of the 20th century, some examples of eroded calderas
(cauldrons) were first interpreted as subsided blocks associated with igneous activity,
e.g. Glen Coe in Scotland (Clough et al., 1909) and the eroded calderas in the
Tertiary San Juan volcanic field, Colorado (Burbank, 1933). Subsequent
investigations recognised a collapse origin for Crater Lake, Oregon (Diller and
Patton, 1902), and identified the Toba caldera in Indonesia and the post-collapse
resurgent uplift of its floor (Van Bemmelen, 1929). A few years later, large
Pleistocene calderas associated with voluminous explosive volcanism were
recognised in southwestern Japan (Matumoto, 1943).
In 1941, Williams presented an insightful analysis of calderas and their origin.
He assumed the reverse relation as suggested by Fouque ´ (1879), that most large
calderas form by collapse after the eruption of magma during pyroclastic volcanism.
In the 60s and 70s, several studies (Smith, 1960, 1979; Smith and Bailey, 1968)
provided a conceptual framework for stratigraphic studies of ash flow sheets, a
model of collapse resurgence within ash flow calderas and a petrologic framework
for ash flow magmatism. Also important were the studies of pyroclastic deposits
(e.g. Sparks and Walker, 1977; Walker, 1980; Wilson and Walker, 1982) that
showed the way for correlating the physical properties of pyroclastic deposits with
their emplacement mechanism.
The tremendous progress made since the 1960s in the related fields of
pyroclastic flow analysis, caldera geology and caldera dynamics are difficult to
summarise. However, after these pioneering works, collapse calderas have been the
subject of studies of diverse disciplines. In fact, during the last decades, analogue
modelling has become a useful tool in the study of caldera-collapse processes. The
first experiments were performed by Ramberg (1967, 1981), who carried out
centrifuge experiments using putty (the magma chamber analogue) under a roof of
clay (the host rock analogue). These experiments and posterior works (e.g. Komuro
et al., 1984; Komuro, 1987; Martı ´ et al., 1994; Roche et al., 2000; Roche and
Druitt, 2001) focused on understanding caldera-collapse mechanisms and resulting
caldera structures. The use of analogue models for caldera studies is currently
increasing, and new questions such as the influence of regional tectonics and the
effect of pre-existing topography (volcanic edifice) on the formation of collapse
calderas are arising (Acocella et al., 2002, 2004).
Simultaneously, theoretical models based on the application of thermodynamics
and solid and fluid mechanics have progressively become an indispensable tool
for the investigation of caldera-forming processes. These theoretical studies are
oriented towards different aspects of the collapse caldera formation, mainly the
behaviour of magma (Druitt and Sparks, 1984; Bower and Woods, 1997; Martı ´
et al., 2000; Roche and Druitt, 2001) or the behaviour of host rocks (Komuro
et al., 1984; Gudmundsson et al., 1997; Gudmundsson, 1998; Folch and Martı ´,
2004).The geophysical disciplines have also significantly contributed to our
understanding of calderas in recent years: first, by providing unprecedented insights
into their subsurface structure using in particular seismic signals to construct