Page 428 - Caldera Volcanism Analysis, Modelling and Response

Hydrothermal Fluid Circulation and its Effect on Caldera Unrest 403

seismic events and volcanic tremor at Redoubt, Alaska (Chouet, 1996); Aso
volcano, Japan (Kaneshima et al., 1996; Yamamoto et al., 1999; Kawakatsu et al.,
2000); Phlegrean Fields, Italy (Bianco et al., 2004); Rabaul, Papua New Guinea
(Blong and McKee, 1995; Gudmundsson et al., 2004) and Mt. Spurr, Alaska
(De Angelis and McNutt, 2005). A connection between seismic swarms and
long-term pulsating degassing of a shallow magma body was also proposed for La
Soufrie `re, Guadaloupe, Lesser Antilles (Villemant et al., 2005). At Long Valley,
California, the ascent of deep, CO 2 -rich fluids (later responsible for tree-killing)
was recently indicated as the possible trigger for the long seismic swarm recorded in
1989 (Hill and Prejean, 2005). Evidence of hydro-fracturing, associated with the
ascent of hydrothermal fluids, was also recognised in the focal mechanism of small
micro-earthquakes recorded in 1997 (Foulger et al., 2004).
Hydrothermal fluids also generate a variety of geophysical signals that can
be detected at the surface. This is well known in the geothermal industry, where
such signals are collected to monitor reservoir properties during exploitation.
Circulating fluids induce changes in electrical potential as they move with respect
to the host rock (electro-kinetic effect). Thermo-electric and electro-chemical
effects are also known to arise from thermal and chemical gradients (Zlotnicki and
Nishida, 2003). Self-potential (SP) anomalies are commonly identified upon fluid
injection or production in geothermal reservoirs (Darnet et al., 2004). In active
volcanic systems, anomalies up to several hundreds of mV are commonly associated
with thermal and hydrothermal features, and their temporal evolution is known to
reflect the evolution of the volcanic system (Zlotnicki and Nishida, 2003).
Phase transition or displacement of liquid water can occur within the hydro-
thermal system and modify the subsurface density distribution. The resulting
gravity change can be detected at the surface by accurate microgravity measure-
ments. Gravity changes are recorded in geothermal fields to monitor reservoir
properties during fluid production and to constrain numerical modelling of
reservoir exploitation (Hunt and Kissling, 1994; Nortquist et al., 2004). Gravity
changes are also commonly observed in active calderas during episodes of ground
deformation (Brown et al., 1991; Berrino et al., 1992; Rymer, 1994; Murray et al.,
2000; Battaglia et al., 2003; Battaglia and Segall, 2004; Gottsmann and Battaglia,
2008–this volume). Even though such changes are also caused by ground
displacement, in some cases they can be ascribed to the motion of aqueous fluids
(Berrino et al, 1992; Gottsmann and Rymer, 2002; Gottsmann et al., 2003;
Todesco and Berrino, 2005).
To some extent, hydrothermal fluids can also drive ground deformation.
Coupling of thermal gradients, fluid flows and mechanical deformation of rocks
have been widely recognised, and complex thermo–hydro–mechanical (THM)
interactions are known to occur in many geological contexts and applications
(Tsang, 1999). Prolonged fluid extraction at the Wairakei geothermal field,
New Zealand, was shown to have caused up to 14 m of ground subsidence, over
almost 40 years of production (1950–1997) (Allis, 2000). Localised subsidence as a
consequence of fluid production was also recorded in Long Valley near the Casa
Diablo power plants (Howle et al., 2003). Similarly, ground uplift may follow
pore pressure increase and rock thermal expansion, associated with the circulation