Page 411 - Caldera Volcanism Analysis, Modelling and Response

386 Claudia Troise et al.

horizontal to vertical displacements. In order to compare the present uplift episode
with previous ones, and in particular to constrain the source depth, two elements
are considered:
1. the shape of vertical displacement pattern, which is well constrained by precision
leveling;
2. the ratio between horizontal and vertical displacements (DH max /DV max ), which
can be obtained from GPS continuous measurements and tentatively compared
with previous uplift and subsidence episodes.

In order to monitor DH max /DV max with a limited number of GPS points, it is
useful to determine theoretically the range in distance from the maximum elevation
point at which the maximum horizontal displacement can be recorded as a function
of source depth and shape. Accordingly, we have used the FEM procedure to plot
horizontal displacements as a function of distance, normalized to the maximum
vertical displacement, for the same source shapes and depths as Figure 8a. Resulting
curves are shown in Figure 8b. We note that the range in which maximum
horizontal displacements are expected is very narrow, and that this corresponds
to an average distance from the maximum elevation point of 2.170.2 km. This
stability is partially due to the effect of ring faults, and clearly shows that for
sources with constant axis location the point of maximum measured horizontal
displacement is expected to vary over a very narrow distance for any depth or
source shape.
The shape of the vertical displacement pattern can be computed by normalizing
the leveling data with respect to the maximum, and by comparing different episodes
among them. We have accordingly normalized the leveling data along the line
Naples–Pozzuoli–Baia–Miseno with respect to the maximum of the 1982–1985
unrest (Pozzuoli harbor), and compared the May 2004–March 2006 and May
2004–September 2006 to 1982–1985 and 1999–2000 previous uplift episodes. As
the four episodes refer to very different displacement amounts (1.768 m for
1982–1985 to 0.021 m for 2004–March 2006) the confidence limits in the
normalized plots are also very different, and must be considered. Figure 9a shows
the normalized vertical displacements of the four periods, with relative confidence
intervals of 1s. Confidence intervals for the 1982–1985 data are generally negligible
in the plot (given the large amount of displacement) except at the borders
where the uplift is very small, and they have been neglected in the plot accordingly.
It is clear that the shape appears constant for all the episodes, although 2004–2006
data show a higher scatter at some localized zones (explained by larger relative
errors). In particular, at some points between Naples and Pozzuoli, and at most of
the points between Baia and Miseno, May 2004–March 2006 relative uplift seems
slightly but somewhat systematically larger than in 1982–1985. However, when
considering the last data (May 2004–September 2006) uplift at the same points is
very slightly lower, but much closer to 1982–1985 data. No measurements are
available along the Baia–Miseno line for 1999–2000 uplift, which at the other
benchmarks seem very consistent with the 1982–1985 period.
However, the similarity of normalized vertical displacements is not necessarily
indicative of equal source mechanisms, due to the effect of bordering caldera-collapse