Page 413 - Caldera Volcanism Analysis, Modelling and Response

388 Claudia Troise et al.

observation allows us to assume that any change in the ratio between maximum
horizontal and vertical ground displacements is due to a change in source depth and
geometry.
In order to make a homogeneous comparison of the horizontal/vertical
displacement ratio, we must consider that the actual continuous GPS points are in
place only since January 2000, so that we can compare GPS data only with 2000
uplift. Figure 9b shows the average DH max /DV max ratios computed in periods with
an almost constant positive or negative vertical displacement rate, as recorded at
the RITE and ACAE stations, where vertical and horizontal displacements are
maximum. In the following, we assume for simplicity that the source location is just
below the point of maximum vertical uplift (RITE). If, for example, the real
location were more South in the Gulf (about 1 km, see Beauducel et al., 2004), the
real value of DH max /DV max could be smaller. It is noteworthy that, within the
errors, the average DH max /DV max ratio is very similar in the whole period, ranging
on average between 0.29 and 0.37, except in the period October 2004–November
2005, where the negligible amount of displacement makes the error so large that
the inferred ratios are meaningless.

4. Discussion and Conclusion

The new uplift at Campi Flegrei caldera is characterized by a slow
deformation rate, at least in the initial phase, but in the last few months the rate
has been increasing (peak in September 2006: about 1 cm/month; Troise et al.,
2007). At first glance, the slower initial rate seems to distinguish it from the episodes
of small uplift that have occurred since 1989 (Gaeta et al., 2003; Lanari et al., 2004).
However, as the previous mini-uplift episodes occurred in periods of general
subsidence trend, a first lower rate phase could have easily gone undetected, mainly
because no continuous GPS measurements existed before 2000. The shape of
vertical ground deformation matches well with that from the large uplift of
1982–1984 and the small uplift of 2000. Furthermore, the ratio among maximum
horizontal and maximum vertical displacement is equivalent, at the same
continuous GPS points, to that observed during the small uplift of 2000. These
observations suggest that, besides the apparent differences in initial deformation
rates and duration, the last uplift and the previous small uplift episodes share the
same source geometry and mechanism. A formal comparison with the 1982–1984
large uplift can only be made with the shape of vertical displacement, which appears
very similar. Horizontal to vertical ratios cannot really be compared because
continuous GPS are in place only since 2000, and horizontal data pertaining to the
last large uplift episode consist of EDM measurements, which are not directly
comparable. Battaglia et al. (2006) showed that the horizontal to vertical
displacement ratios are significantly smaller (about one half) during the large uplift
with respect to the 1990–1995 subsidence. Such data are interpreted as two
different source geometries and depths for the large uplift. The first part of the
uplift is interpreted to be due to overpressures within a penny-shaped crack which