Page 430 - Caldera Volcanism Analysis, Modelling and Response
P. 430

Hydrothermal Fluid Circulation and its Effect on Caldera Unrest      405


             gravity residuals and significant changes in the composition of gases discharged at
             Solfatara crater. A slow subsidence begun in 1985, periodically interrupted by
             minor uplift (few cm each), accompanied by significant changes in gas composition
             and occasional minor seismic activity (see Troise et al., 2008–this volume, and
             references therein). As mentioned above, several authors highlight the role of
             hydrothermal fluid circulation in governing the recent evolution of the Phlegrean
             Fields caldera (Bonafede, 1991; Gaeta et al., 1998; Orsi et al., 1999; De Natale
             et al., 1991, 2001; Castagnolo et al., 2001). Recent new analyses of deformation
             and gravity data confirm the role of hydrothermal fluids in generating uplift (at least
             partially) and subsidence observed since the last unrest crisis (Gottsmann et al.,
             2003, 2006; Battaglia et al., 2006).
                Physical modelling of heat and fluid flow is a useful tool to quantify the effects of
             hydrothermal fluid circulation. The simulations presented below describe fluid
             circulation within the shallow hydrothermal system that feeds surface discharges at
             the Solfatara crater (Todesco et al., 2003a, b, 2004; Chiodini et al., 2003; Todesco
             and Berrino, 2005). The role of new magmatic intrusions in the recent unrest crises
             is not explicitly accounted for, but it is represented in terms of variable magmatic
             degassing and by the emplacement of a deep source of hot fluids. Simulations were
             performed with the TOUGH2 geothermal simulator (Pruess, 1991; Pruess et al.,
             1999). The model describes the coupled heat and fluid flow through porous media
             for a multi-phase, multi-component system. Phase transitions (gas–liquid) and
             associated latent heat effects are fully accounted for. Water and carbon dioxide are the
             two fluid components considered in the model. Details on model formulation and
             solution techniques can be found in Pruess et al. (1999). Shallow hydrothermal
             circulation at Solfatara is simulated on a uniform, two-dimensional and axisymmetric
             domain (Figure 2a). A source of hot (3501C) water and carbon dioxide is placed at
             the bottom (near the symmetry axis) to represent magmatic degassing. Discharge of
             these hot fluids generates a wide two-phase plume, within which a shallow dry-gas
             region forms (Figure 2b, Todesco et al., 2003a). Existence and conditions of such
             single-phase gas region are in good agreement with geochemical data, postulating
             that fumaroles are fed by a super-heated vapour zone (Chiodini and Marini, 1998).
                Using these initial system conditions the model can be applied to study the
             recent evolution at Solfatara. Simulations were carried out under the assumption
             that observed compositional changes were driven by periods of increased magmatic
             degassing. Unrest crises are therefore simulated as periods of higher gas flow rate
             and CO content at the deep source (Chiodini et al., 2003). The model describes
                     2
             the fluid composition and properties throughout the simulation, allowing
             comparison of modelling results with available geochemical and geophysical data.
             The composition of the single-phase gas region is taken as representative of
             fumarolic gases, and is compared with observed gas composition. Appropriate
             number, timing and duration of each unrest period allow successful matching with
             the observed compositional variation (Figure 3). If magmatic degassing increases,
             pore pressure and fluid temperature are also expected to increase. Mechanical
             effects associated with such changes can be evaluated based on the calculated
             pressure and temperature distribution. Coupling between TOUGH2 and FLAC3D,
             a commercial code for rock mechanics (Itasca Consulting Group Inc., 1997), was
   425   426   427   428   429   430   431   432   433   434   435