From the analysis of the figure, it can be noticed how this 
deformation, with about radial symmetry, includes an area 
centred in Pozzuoli and bounded by Posillipo hill (Naples) 
at the east side and by the center of Baia at the west side. 
Towards the inland includes Solfatara, Astroni and 
Agnano craters up to the western suburb of the city of 
Naples. 
Thus, in fig.5 a 2% cycle (fringe) corresponds to a 
deformation of 2.8 cm, depending on the wavelength (C 
band=5.6 cm) of the radar signal. Really the deformation 
value deduced by the analysis of the interferogram must 
be lightly corrected, as it was deduced for the slant range 
dimension. 
The second interferogram has been generated in order to 
demonstrate the results obtained during the 1997 spirit 
leveling measurements, that is the measured 
deformations were quite little if compared with those ones 
of 1996; the consequence is that the interferogram 
obtained on a November 1995/February 1997 temporal 
interval should result substantially identical to the first one. 
The strong temporal decorrelation pointed out in the 
coherence map (being a two year period), associated with 
a larger baseline (about 25 m) which requires the 
necessity to remove from the interferogram the 
contribution in phase of the topography, have not allowed 
to obtain such an immediate result. 
3. SURVEYING METHODOLOGIES APPLIED IN THE 
ISLAND OF ISCHIA 
3.1 High precision spirit leveling 
The spirit leveling network of the island of Ischia (fig.6) is 
formed by 250 benchmarks with a mean distance of 300 
m, on about 90 km on seven loops, besides the coast line. 
The last spirit leveling measurements along the whole 
network (May 1997) have been carried out using 
autolevels equipped with plain-parallel lamina 
micrometers, and couples of rods with invar ribbon. 
Fig.6 Geodetic surveillance networks of the island of 
Ischia 
The measurements have been submitted to the least 
squares adjustment with the method of indirect 
observations, with a final value of the standard deviation 
per unit of weight ao= 1.6 mm. 
In fig.7 the height variations (reference period: 
penultimate measurements of October 1994/May 1997) 
are shown along the line in which these variations result 
the most significant of the whole network. Also in this case 
the dotted lines represent the limits of the confidence 
interval at 95% for the series of data being compared. 
Fig.7 Vertical displacements (10/94-5/97) along the 
Borbonica line (NW side) in the island of Ischia 
3.2 GPS 
In October 1996 on the island of Ischia a GPS network 
formed by 20 3D vertices has been installed. The 
measurement of the whole network has been carried out 
during the first months of 1997 by means of three different 
techniques (Static, Fast-static and RTK Real Time 
Kinematics), also with the purpose of comparing their 
precision levels in reduced dimensions areas like the 
island of Ischia. During 1998 the measurement of the 
whole network has been repeated, adding two new 3D 
vertices in Forio d’lschia (fig.6). 
For the 1997 measurement of the whole network (20 
vertices), the three above mentioned techniques have 
been used, according with the following standards: 
- Static Measurements, with a sampling time of 15 sec in 
continuous acquisition; 
- Fast-static Measurements, with a sampling time of 10 
sec and acquisition time from 5 to 15 min; 
- RTK Measurements, kinematic measurements on 15 
vertices of the network. 
For each of the three different techniques proper receivers 
have been used, while data processing has been carried 
out using GEOTRACER vers.2.24 software. In fig.8 are 
shown the residual values of the Helmert transformation 
between data obtained by the static technique and the 
RTK technique, in order to compare the precision levels of 
these two techniques in reduced dimensions areas. 
In July 1998 a new measurement of the whole network 
(22 vertices) has been carried out. The choice of these 
two new points has been suggested by the results of the 
spirit leveling measurements, indicating a vertical 
displacement in Forio d’lschia. 
The measurements have been performed in static mode, 
with a sampling time of 15 sec in continuous acquisition 
for two days; in this case, double frequency receivers 
have been employed.