Stress pattern at Campi Flegrei from focal mechanisms of the 1982 1984 earthquakes (southern Italy

Journal qf Volcanology and Geothermal Research, 48 ( 1991 ) 127-137 127 Elsevier Science Publishers B.V., Amsterdam

Stress pattern at Campi Flegrei from focal mechanisms of the 1982-1984 earthquakes (Southern Italy )

Agostino Zuppetta and Alessandra Sava Dipartimento di Geofisica e Vulcanologia dell'Universitgt di Napoli, "Federico II", Largo S. Marcellino 10, 80138 Napoli, Italy

(Received June 16, 1990; revised version accepted November 5, 1990)

ABSTRACT

Zuppetta, A. and Sava, A., 1991. Stress pattern at Campi Flegrei from focal mechanisms of the 1982-1984 earthquakes (Southern Italy). In: G. Luongo and R. Scandone (Editors), Campi Flegrei. J. Volcanol. Geotherm. Res., 48:127-137.

Campi Flegrei is a Holocene volcanic area in the Campanian Plain (Southern Italy ) within the Apennine Chain, a neogenic thrust belt built up since the Middle Miocene. The volcanic complex consists of a c. 12-km-diameter caldera containing several monogenetic volcanoes, the youngest of which (Monte Nuovo) erupted in 1538. Since at least Roman times, the area has also been affected by slow vertical movements (bradiseismic activity). Between 1982 and 1985, this slow motion was interrupted by a period of more rapid displacement which caused a maximum uplift of 180 cm in the town of Pozzuoli. To model the local stress field associated with the uplift, the Angelier inversion technique has been applied using the focal mechanisms of forty-nine earthquakes which occurred between April 1982 and December 1984. The results show that doming coupled with a regional extensional strain can account for the seismic phenomena that affected the area.

Introduction

The classical analysis of fault plane solutions has been widely applied to evaluate local stress patterns from a given population of earthquakes.

The method has been recently used by some authors to determine the stress field at Campi Flegrei, a volcanic field west of Naples (South- ern Italy), during the 1984-1985 bradiseismic crisis.

The focal solutions of 15 seismic events during 1983 have been used by Gaudiosi and Ian- naccone (1984) to model the stress pattern at Campi Flegrei. From the spatial distribution of the P and T axes they concluded that, in the Solfatara area, "the least compressive stresses are predominantly horizontal and maximum compressive stresses (P axes) are nearly vertical", whereas " the maximum compressive

stress is horizontal in Pozzuoli bay", and that no regional components seem to dominate during the bradiseismic crisis.

Coppa et al. (1985) extended the focal mechanism analysis to 21 earthquakes which occurred during 1983-1984. They distin- guished a bimodal distribution of the P and T axes: seismic events with normal fault solutions on land and events with reverse or normal fault solutions in the Gulf of Pozzuoli. It is widely recognized, however, that the P and T axes represent the moment tensor and not the stress axes (Mckenzie, 1969).

Aster and Meyer ( 1988 ), who reconstructed a three-dimensional velocity structure for the Campi Flegrei caldera using 228 microearth- quakes recorded from August 1983 through May 1984, indicate a zone of concentrated se- ismicity north of the point of highest measured uplift and suggest a ring-fault structure to ex-

0377-0273/91/$03,50 © 1991 Elsevier Science Publishers B.V. All rights reserved.

128 A. ZUPPETTA AND A. SAVA

Cuma

/ I I I I S I I I I S I S I I I I I I I S I I I I I I I I I '

. . . . . . . / . , : ' " ' " " ' " " " ' " " " " ' " " ' "

/ , # : . . . . . . . . . . . . . . . . . . . . . . . . .

, . , . - , ~ , , . , ; , ~ ) ; , / - , ~ , ; , , % - , . . .K , , ; , ; , ; , ; , , . . . . . . . . . . . . . . . . . . . . . . : ' r 2 ~ " ) , ? l ~ , , , , , , ' , ' , , , x , ' , ' , ' , x ~ x , ' , ' • ' , ' , x , x , ' , ' , ' , x , x , ' , ' , ' , ' , ' , X l l ~ , , ' , ' , ' , ' , ' , ' , ",

, ' , ' , ' , ' , ' , ' . ' . ' , ~ . \ \ \ \ \ \ \ \ \ ' ~ ' ¢-~er, oa'_,,,Yl~"R~llillll Pienura Plain

m,ri~, terr ,¢e) ' , l l l l l l l l l l l l l l l l l l l l lP" Ag~no

Baia Pozzuol i Aooademia

Sar~ta Teresa

" s ' s ' l ' s ' 1 " ~ ' s ~ s ~ I ~ , , , x x , , , • • ' l ~ l x • • • i x • • s • • • • s

s l • • J s • s

I J

% % % % % % %

Fondi di Baia

Monte di Procida Porto Hiseno Nisida

Posi11|po

~ Products older than NYT (> 12ka)

I Neapolitan Yellow/ Tuff(12 ka)

~ Volcanoes of yello~ t u f f ( " 12-8 k i )

~ Pyro¢ltsti© fall deposits (<'8 Ka)

m]] ]~ Base surge deposits (< "a ka) Pyroclastic flow

~ deposits (Honte Nu~ vo. Honte Spina)

~ LaYa domes

Fig. 1. Schematic volcano-tectonic map of Campi Flegrei.

STRESS PATTERN AT CAMPI FLEGREI FROM FOCAL MECHANISMS OF THE 1982-1984 EARTHQUAKES 129

plain the inward elliptical hypocenter pattern. The aim of this paper is to evaluate the ori-

entation of the principal stress axes at Campi Flegrei from the fault plane solutions of earthquakes occurring during the 1982-1985 bradiseismic crisis using the method proposed by Angelier ( 1979a, 1984, 1989).

Regional geology

Campi Flegrei lies within the Southern Apennine chain, a Neogene thrust belt built up since the Middle Miocene by continental col- lision between the Corsica-Sardinia micro- plate and the African plate (Fig. 1 ).

The belt may be divided into three main tectonic complexes:

( 1 ) The uppermost complex consists of al- lochthonous terrains [ophiolites, black shales, metasiltites, metacarbonates of the Liguride Complex and the varicolored shales of the Si- cilide Complex of Ogniben (1969) ], derived from the deformation of the Neo-Tethys during Oligo-Miocene times.

(2) The intermediate complex comprises thrust sheets of the Campano-Lucana plat- form, a Meso-Cenozoic inner carbonate plat- form outcropping along the border of the Tyr- rhenian Sea.

( 3 ) The lowest complex rests upon the Plio- Quaternary sediments of the Bradanic trough and the underlying carbonate rocks of the Apulian foreland. It is composed of tectonic units produced during deformation of the Meso-Cenozoic terrains of the Lagonegro basin (Scandone, 1979) and of the Langhian-Tor- tonian turbidite sediments belonging to the Ir- pinian basin ( Cocco et al., 1972).

Since the Upper Pliocene, regional extension has broken the Southern Apennine chain into several blocks, leading to horst and graben development and formation of the Campanian Plain (Upper Pliocene), whose structure is controlled by two major normal fault systems trending N130 ° and N60 °. As shown by seismic profiling (Colantoni et al., 1972; De

Bonitatibus et al., 1970; Finetti and Morelli, 1974; Pescatore et al., 1984) these fault systems have had a major influence on the location, within the Plain, of the Quaternary volcanic fields of Roccamonfina, Campi Flegrei and Vesuvius.

The structural evolution of the Tyrrhenian area (Scandone, 1979) and the occurrence of the recent seismic activity demonstrate that an extensional tectonic regime is still active in the Southern Apennines. For example Del Pezzo et al. (1983) have shown that the 23 Novem- ber 1980 Irpinian earthquake has a normal fault focal solution with a sub-horizontal Taxes trending north-northeast, in agreement with regional seismic studies (Gasparini et al., 1982).

Volcanic activity of Campi Flegrei

The oldest volcanic products outcropping in Campi Flegrei range in age from about 47,000 to 34,000 years B.P. Subsequent activity in the area has been dominated by the emplacement of two major ignimbrite formations, the Cam- panian Ignimbrite (c3.) and the Neapolitan Yellow Tuff (N.Y.T.), about 34,000 and 12,000 years ago respectively (De Lorenzo, 1904; Rittman et al., 1950; Di Girolamo, 1970; Bar- beri et al., 1978; Capaldi et al., 1985; Rolandi, 1988).

The location of the vent area for the c.I. eruption, which involved some 80 km 3 D.R.E. (dense rock equivalent) of trachytic magma (Thunell et al., 1979), remains a matter of de- bate. The two principal hypotheses are:

(a) that the ignimbrite was erupted from an arcuate fracture in the Campanian Plain, to the north of the present Campi Flegrei caldera (Di Girolamo, 1970; Barberi et al., 1978; Di Giro- lamo et al., 1984; Rolandi et al., 1988);

(b) that the c.t. was erupted from one or more ring fractures close to the borders of the current Campi Flegrei caldera (Rosi et al., 1983).

Whichever the interpretation, the lack of C.l. outcrops in Campi Flegrei suggests that large-

130 A. Z U P P E T T A A N D A. SAVA

\ 52'

Ouarto ~ 5

I0

2 0

'40

"60

Napoli

• 4 0 " 50' M Nuovo

P. [ p i l a f f i e

Baia Pozzuoli •

0

P. il l Pennalm

Nisida N

I 2 km 4 7 ' 14" 05" C. H i s e n o 10 ' t , . •

I I

Fig. 2. Campi Flegrei uplift 1982-1985. Arrow shows Pozzuoli, where maximum uplift occurred (ca. 180 cm). The epicenters of the forty-nine analyzed earthquakes are also shown. Open circle refers to earthquakes with principal stress directions compatible with an extensional regime trending N23 ° (family S~ ); filled circle to seismic events which indicate a N 13 ° compressional tectonics ($2); cross indicates seismic events with a strike-slip stress axes distribution ($3 and $4).

scale subsidence occurred in the area after this ignimbrite had been emplaced. The resulting collapse structure was significantly modified about 22,000 years later by the Neapolitan Yellow Tuff eruption, which expelled of the or- der of 20-30 km 3 (D.R.E.) oftrachytic magma (Lirer et al., 1987) and lead to the formation of the present Campi Flegrei caldera (Rolandi, 1988).

Activity since the N.Y.T. eruption has been characterized by much smaller events producing a collection of monogenetic cones across the caldera floor. Most of these eruptions appear to have been concentrated within two main pe- riods, 12000-8000 years B.P. and 4500-3700 years B.P. (Rosiet al., 1983).

Between these two phases a significant uplift of the central part of the caldera took place, producing the raised marine terrace La Starza, which currently stands at 35 m a.s.1. (Rosi et al., 1983; Cinque et al., 1985 ).

The last eruption in Campi Flegrei occurred

in September 1538, producing the scoria cone ofM. Nuovo about 3 km west ofPozzuoli (Di Vito et al., 1987). Since then, the area has been affected only by slow oscillations in ground level (bradiseismic motion) near the centre of the caldera, a phenomenon which can be doc- umented at least back to Roman times, based on observations of the columns at Serapeo, a Roman market place near the modern port of Pozzuoli (Parascandola, 1947 ).

Since the early 1970's, these slow ground movements have been punctuated by two pe- riods of relatively rapid uplift. The first occurred without significant seismic activity between 1970 and 1972, resulting in a general uplift of the Pozzuoli district by c. 1.70 m (Corrado et al., 1976). Bianchi et al. (1984), modelling the surface ground deformations with a finite-elements method, use the work hypotesis that the overall trend of vertical dis- placements observed at Campi Flegrei is due to an interplay between local phenomena and

STRESS PATTERN AT CAMPI FLEGREI FROM FOCAL.MECHANISMS OF THE 1982-1984 EARTHQUAKES 13 1

the response of the volcanic system to the regional stress field. The uplift ended in 1972 and was followed by periodic ground oscillations of c. 10-15 cm yr -1 until 1982, when a new ground upheaval until December 1984 raised Pozzuoli by c. 1.8m.

Ground deformation studies made by Ber- rino et a1.(1984) showed that the zone affected during the second phase of uplift had an almost circular symmetry, with a radius of about 6 km, centred on Pozzuoli (Fig. 2 ). Un- like the 1970-1972 event, the vertical movements were accompanied by significant seismic activity, whose intensity increased until March-April 1984, initiating earthquakes of magnitude 4 on the Richter scale.

In January 1985, the seismic activity ab- ruptly diminished coinciding with a gradual deflation of the ground at about 1 cm month- 1.

Tectonic analysis

To model the local stress pattern associated with the 1982-1985 bradiseismic crisis in Campi Flegrei we made a tectonic analysis using focal mechanism solutions of forty-nine earthquakes (Table 1) from Gaudiosi and Iannaccone (1984) and Caccavale ( 1984/85, unpublished thesis). The data set is homoge- neous, since the authors both use similar ho- mogeneous half-space velocity models (see comment in Aster and Meyer, 1988 ).

Figure 3 shows the distribution of the P and T axes plotted on the lower-hemisphere Schmidt net. The data consist largely of seismic events with dip-slip faulting solutions; a subordinate set of earthquakes with predomi- nant strike-slip solutions is also present.

As Mckenzie (1969) has pointed out, however, the P and T axes represent the moment tensor and not the stress axes: if they were as- sumed to be stress axes, they would give no information about the shape of stress ellipsoid.

To determine the mean stress axes, we used a reduced stress tensor (progr. TENSOR, option INVD) and a numerical iterative technique

Table 1

Campi Flegrei earthquakes, February 1983-December 1984

Date n"

1 10. 02.

2 10. 02.

3 15. 05.

4 22. 08.

5 22. 08.

6 05. 09.

7 12. 09.

8 19. 09.

9 26. 09.

1 0 01. 10.

1 04. 10.

12 06. 10.

13 03. 11.

14 10. 11.

15 15. 11.

16 03. 12.

17 O7. 12.

18 07. 12.

1 9 15. 12.

2 0 16. 12.

21 16. 12.

2 2 20, 12.

2 3 24. 12.

2 4 30. 12.

2 5 30. 12.

2 6 30. 12.

2 7 31. 12.

2 8 05. 01.

2 9 10. 01,

3 0 19. 01.

3 23. 01.

3 2 26. 01.

3 3 29. 01.

3 4 31. 01.

3 5 14. 02.

3 6 21. O2.

3 7 22. 02.

3 8 25. 02.

3 9 02. O3.

4 0 09, 03.

41 10. 03.

4 2 ! 12. 03.

431 12. 03.

4 4 ; 14. 03.

4 5 03. 04.

4 6 09. 04,

4 7 15. O4.

4 8 28. 09.

4 9 20. 12.

Hour Lat i tude N Longi tude E

83 1 5 : 3 7 40 - 49 .58

83 t 5 : 4 0 40 - 49 .96

83 1 8 : 1 1 40 - 49 .65

83 0 7 : 3 1 40 - 49 .88

83 0 7 : 3 1 40 - 49 .60

83 1 5 : 3 6 40 - 48 .00

83 0 0 : 3 1 40 - 48 .03

83 0 9 : 1 9 40 - 49 .56

83 1 4 : 0 4 40 - 4 9 5 0

83 2 3 : 5 9 40 . 49.51

83 0 8 : 0 9 40 - 49 .56

83 2 0 : 3 1 40 - 49 .32

83 2 3 : 4 3 40 - 48 .00

83 1 9 : 5 6 40 - 49 .44

83 1 1 : 4 9 40 - 49 .38

83 0 0 : 4 8 4 0 - 4 9 . 5 6

83 i 0 4 : 5 0 40 - 49 .92

83 0 5 : 0 1 40 - 49 .68

83 1 1 : 5 5 40 - 49 .32

83 0 0 : 3 1 40 - 49 ,38

83 1 3 : 4 4 40 - 47 .94

83 2 1 : 2 0 4 0 - 49 .43

83 1 3 : 2 6 40 - 49 .56

83 2 3 : 1 6 40 - 49 .38

83 2 3 : 1 7 40 - 49 .38

83 2 3 : 1 9 40 - 49 .27

83 2 2 : 2 0 40 - 49 .79

84 0 0 : 1 5 40 - 50.06

84 2 0 : 3 5 1 40 - 47 ,87 I

84 1 7 : 5 4 ! 40 - 48 .04

14 - 7 .29

14 - 6 .97

14 - 8.91

14 - 7 .97

14 - 7.91

14 - 6 .32

14 - 6 .66

14 - 7 .25

14 - 8.41

14 - 7 .33

14 - 7 .92

14 - 7 .62

14 - 7 .02

14 - 8 .46

14 - 8 .76

14 - 7 .68

! 4 - 6 .54

14 - 6 .72

14 - 8 .10

14 - 8 ,40

14 - 6 .84

14 - 8 .65

14 - 8 .34

14 - 8 .22

14 - 8 .16

14 - 8 .96

14 - 7 .90

14 - 8 .34

14 - 6 .73

14 - 6 .99

- 49 .42 14 - 8 .87

- 49 .87 14 - 8 .72

- 48 .46 14 - 6 .46

- 49 .87 14 - 7 .83

- 49 ,74 14 - 6 .46

- 49 .70 14 - 6 .74

- 49 .63 14 - 8 .53

- 49.71 14 - 8 .83

- 49 .89 14 - 8 .59

- 49 .67 14 - 8 .08

- 49 ,29 14 - 7 ,90

- 48 ,06 14 - 6 .66

- 49 .29 14 - 8 .43

- 49 .23 14 - 8 .84

84 0 6 : 4 2 40

84 2 1 : 2 1 40

84 0 0 : 5 1 40

84 0 7 : 4 6 40

84 0 1 : 2 7 40

84 0 0 : 2 9 40

84 0 2 : 3 0 40

84 0 5 : 1 7 40

84 0 6 : 2 3 40

84 1 9 : 1 1 40

84 0 5 : 3 8 40

84 0 0 : 0 2 40

84 0 7 : 1 0 40

84 1 2 : 0 3 40

8 4 1 6 : 3 4 40

84 1 0 : 2 8 l 40 - 49 .68 I

84 0 4 : 5 9 40 - 49 ,27

84 0 6 : 2 8 4 0 - 5 0 . 2 0

84 2 1 : 2 0 40 - 49 .20

- 49 .87 14 - 8 .07

14 - 7 .54

14 - 7 .94

14 - 6 .43

14 - 8 .23

Z (k in) MAG Refer .

2 O 1 .7

i 2 . 2 1 , 7

2 .5 i 3 . 4 I i

2 . 6 2 . 8

2 . 0 2 . 8

; 3 .4 2 . 6

39 2 6 i: 1.7 3 . 0

1 .6 3 . 2

1 .5 1 .6

2 . 6 3 .8 I #

2 3 2 .2 #

i 2 . 7 2 . 6 # i

I 2 . 3 2 . 0 #

2 .4 3 . 3 #

2 7 : 2 . 3 ; #

2 . 6 1 .4 #

2 . 8 1 ,5 #

2 O 2 . 5 #

: 2 . 3 2 . 7 #

2 . 8 ! 3.1 ~ #

1 .7 3 . 8 ~ "

2 . 7 2 . 5 #

2 . 2 2 . 3 #

2.O 2 . 7 #

0 . 2 3 . 7

2 . 8 2 . 5

2 . 9 2 . 5

2 . 0 2 . 6

2 . 6 2 . 4

1 .8 3 , 4

1 .9 2 . 6

2 . 6 2 , 5

1 .5 3 . 8

2 . 6 2 . 7

2 . 0 2 . 8

1 .6 3 . 7

2 . 2 3 . 2

2 . 0 2 . 5

2 . 0 3 . 9

1 .5 2 . 5

4 . 8 2 . 8

0 .9 2 . 5

.9 4 . 0

.9 3 . 5

Earthquakes used in analysis. Symbols refer to source of data: ( * ) Caccava l e , G., unpublished thesis; ( # ) G a u d i o s i , G. and Iannnaccone, G. 1984.

132

Table 2

Individual results of stress tensor determinat ions for the S, events

Event n* NP NPS NPD SM s k- T k

1 a 33 47E norm. 32

1 b 103 70N norm. 34

3 a 60 40S norm. 1

3 b 106 60N norm. 14

4 a 68 64S norm. 20

4 b 102 30N norm. 0

5 a 76 60S norm. 9

5 b 120 39N norm. 11

8 a 28 67E norm. 1

8 b 72 30N norm. 15

9 it 54 40S norm. 13

9 b 114 67N norm. 20

10 a 124 60N norm. 8

10 b 80 39S norm. 4

11 a 76 50S norm. 27

11 b 85 40N norm, 12

12 a 112 44S norm. 12 I

12 b 1191 46N norm. 15

14 a 14 56E norm. 5

14 b 42 38W norm. 5

16 a 135 58W norm. 9

16 b 136 32E norm. 31 20 a 102 41S norm. 26

20 b 158 64E norm. 18

23 a 98 40S norm, 19

23 b 148 62E norm. 10

24 a 8 ! 49W norm. 15 i

24 b 179 42E norm. 6 I

25 a 60 ] 65S norm. 28

25 b 77 26N norm. 17

27 a 100 61N norm. 41

27 b 148 40W norm. 36

32 a 91 60N norm, 34

32 b 134 39S norm. 39

34 a 98 70S norm. 8

34 b 159 38E norm. 38

37:l ~ 10 50E norm. 28

37 ; ~ 60 53N norm. 29

38 a 120 69S norm. 29

38 b 164 28E norm. 22

40 a 18 30E I norm. 2

40 b 62 67N :norm. 27

41 a 68 38S norm. 2

41 b 110 6 O N norm, 10

43 a 66 40S norm. 2

43 b 114 60N norm. 12

44 a 27 40E norm. 1

44 b 69 58N norm. 21

46 a 89 70N norm. 0

46 ~ b 78 2OS norm. 34

47 a 72 50S norm. 0

47 b 118 50N norm. 13

Results of calculation for each event in S~. Event number refers to Table 1. N.P. = nodal plane; N.P.S. = nodal plane strike; N. P. D. = nodal plane dip; S.M. = sense of relative motion; ( sk, zk) = angle between nodal slip and theoretical shear stress (in degrees ).

A. ZUPPETTA AND A. SAVA

+ P axes N = 98

a T a x e s

Fig. 3. D i s t r i bu t ion o f the P and T axes p lo t ted on the lower hemisphere Schmidt net (see text for exp lana t ion ).

(progr. TENSOR, option R4DT), both devel- oped byAngelier (1979a, 1984, 1989) for ana- lyzing fault populations. This method assumes that the shear stress on the fault plane is parallel to the slip direction (Bott, 1959 ) and that the stress field is uniform for the data set (Carey and Brunier, 1974). The essential fea- ture of Angelier's method is the minimization of a function F which increases with the angle (Sk, Xk) between the actual slip and the com- puted shear on each fault. The simplest expres- sion of Fis:

F= ~ (sk, t~) 2 (1) k = l

For mathematical convenience and reasonable runtime, other function are implied, such a s "

F = ~ sin2[ (sk, Irk)/2] (2) k = l

where k is the number of faults, s the unit slip vector and lr is the unit shear stress resolved on the fault plane. The formula adopted in this

STRESS PATTERN AT CAMPI FLEGREI FROM FOCAL MECHANISMS OF THE 1982-1984 EARTHQUAKES

N N

(a) (b)

133

N N

!

(c) (d)

Fig. 4. Orientations of the calculated principal stress axes (a~, a2, a3) for each family of seismic events. All diagrams are lower-hemisphere Schmidt projections: (a) S~, (b) $2, (c) $3, (d) $4.

paper (R4DT) is a little bit different, but the results are very similar to those obtained with this function. The other formula takes into account the shear stress magnitude (details in Angelier, 1984).

Angelier (1984) has suggested that where two principal stresses are equal (or almost equal) earthquake focal mechanisms can be analyzed by treating both nodal planes as though they were fault planes. This assump- tion, however, has no physical basis (Michael, 1987 ): it is acceptable only if the stress field is uniaxial, a reasonable condition since calculated values of q~, the ratio between principal

stress differences [ ( a z - a3) / (am - a3) ], com- monly lie between 0 and 0.4. Testing the An- gelier's inversion technique on earthquakes on central Crete, Michael ( 1987 ) stated that, although the method does not provide meaning- ful information about confidence regions, it yields good information about the orientations of the principal stress axes and about the shape of the stress ellipsoid.

The seismic events were separeted using the procedure described in Angelier and Manous- sis (1980) and applied by Angelier (1984) to the Yuli earthquakes. Since Angelier argued that using earthquakes the analysis is valid for

134 A. Z U P P E T T A A N D A. SAVA

Table 3

Individual events

Event n ° NP

results of stress tensor determinations for the $2

NPS NPO SM Sk-'C k

6 a 56 40S rev, 14

6 b 103 60N rev. 25

21 a 74 51N rev. 17

21 b 133 58S rev. 5

26 a 75 40N rev. 15

26 b 120 60S rev. 4

28 a 96 55S ray. 5

28 b 128 40N rev. 2

29 a 72 39N rev. 5 29 b 110 58S ray. 7

31 a 100 50S roy. 27

31 b 147 50E rev. 12

33 a 85 69N ray. 12

33 b 129 28S rev. 18

36 a 70 48N ray. 13

36 b 126 58S rev. 0

48 a 55 30S rev. 15

48 {b 97 67N rev. 28 i

49 a 98 68S rev. 16 49 I b 159 40E rev. 4

See Table 2 for explanations.

Table 4

Individual events

Event n °

results of stress tensor determinations for the S 3

NP NPS NPD SM s k- z k

15 a 41 76W dex. 6

15 b 108 31S sen. 19

17 a 115 89S sen. 6

17 b 24 28E dex. 8

18 a 80 80N dex. 9

18 b 143 21W sen. 14

l g a 0 20E dex. 30

19 b 81 87N sen. 30

22 a 62 61N norm. 3

22 b 102 40S norm. 2g

35 a 58 70S dex. 0

35 b 170 43W sen. 1

39 a 74 69S inv. 4

39 b 25 30W inv. 3

42 a 5 70E inv. 13

42 b 124 40S inv, 34

45 a 38 30E dax. 27

45 b 152 77W sen. 17


(sk, ~rk) angles between 0 ° and 45 °, we have accepted all calculated values.

Table 5

Individual results of stress tensor determinations for the $4 events

Event n* NP NPS MID SM Sk-Z k

2 a 9 70W sen. 18

2 b 116 50N dex. 25

7 a 32 60E inv. 7

7 b 98 50N inv. 26

13 a 44 76W sen. 2

13 b 116 38S dex. 1

30 a 32 80E inv. 11

30 b g2 50N inv. 26


Table 6

Principal stress axes orientation in S,, $2, $3 and $4

F .O, 02 03 ~ Sk-~" k

I 71 /261 1 6 / 1 1 6 1 1 / 2 3 .20 17

II 6 / 1 3 1 6 / 1 0 5 7 3 / 2 6 5 . I0 12

III 81136 54•37 34•234 .70 14

IV 0 / 2 7 1 441161 46 /1 .90 14

Orientation of the principal stress axes for each family of seismic events. • values and mean (sk, Tk) angles (in degrees) are also reported.

2 .

~A. 40 . 8~ 4.14~ - ~J-2 ~ LW~'

Fig. 5. 3D surface fitt ing of hypocenters of events with normal fault solutions.

The lower-hemisphere equal-area diagrams of Figure 4 summarize the determined local stress tensors. The calculations made using di- rect inversion (INVD) and systematic iterative exploration (R4DT) on both A and B nodal planes of the forty-nine Phlegrean earthquakes show a polymodal distribution composed of four families of seismic events with different


orientations of the principal stress axes. Twenty-six earthquakes belong to a family

S, with the principal stress directions compatible with an extensional regime trending N23 ° (Fig. 4a), while the orientation calculated for the ten events of the $2 family indicates a N 13 ° compressional tectonics almost parallel with the St extension direction (Fig. 4b). In both cases the stress ellipsoids are almost uniaxial since tp equals 0.2 for St and 0.1 for $2 (Fig. 4a,b and Tables 2 and 3); the angles between the theoretical shear stress and "real" slip directions lie between 1 ° and 43 ° with a mean value of 17 ° for family St and of 12 ° for family 82.

Families $3 and $4 show stress axis distributions coherent with a strike-slip faulting trending obliquely to the Sl and $2 axes: the stress ellipsoids are still uniaxial ( ~ 1 in both cases ) and the (sk, lr~) angles range between 0 ° and 34 °, with a mean of 14 ° for both $3 and $4 (Fig. 4c,d and Tables 4 and 5 ). The al, tr2, t73 orientations for each family are reported in Table 6 together with ~ values and mean (sk, zk) angles, although family $4 seems not significant due to the scarce number of data.

Conclusions

The vertical movements which occurred in Campi Flegrei during 1982-1985 produced a quasi-circular dome 180 cm high at its apex (Berrino et al., 1984 ). Since recent seismic activity, such as the 23 November 1980 Irpinian earthquake ( M = 6.9), indicates that a NNE- SSW extensional tectonic regime is still active in the Southern Apennine chain (Del Pezzo et al., 1983), it seems reasonable to assume that the overall deformation pattern and related seismic activity in Campi Flegrei was a function of doming coupled with regional extensional strain.

The experimental and analytical studies of Withjack and Scheiner ( 1982 ) suggest that regional strain, either compressional or extensional, significantly affects the fault pattern

produced at shallow depth by doming. In the particular case of circular doming in the pres- ence of regional extension, normal faults on the crest of the circular structure trend perpendic- ular to the regional extensional direction, whereas many normal faults on the dome flanks trend obliquely to it. Strike-slip faults trending 60 ° from the regional extension direction de- velop near the periphery of the dome, while reverse faults form along its boundaries and be- come increasingly oriented normal to the extension direction as the regional strain increases.

Since the results of Withjack and Scheiner study successfully applied in the field on many domes in different areas, an a t tempt was made to verify if the fault pattern predicted by the model is also coherent with the results of our tectonic analysis of doming in Campi Flegrei.

The areal distribution of the Phlegrean earthquake epicenters (Fig. 2) shows that all the data points lie in a narrow area between the Solfatara Crater and Punta di Pennata. Twenty-six earthquakes with normal fault solutions tend to concentrate in the central part of the area along a WNW-ESE direction, while the seismic events with reverse and strike-slip fault solutions are distributed along a more external part of the dome. In addition, the results of the tectonic analysis (Fig. 4) show that deformation pattern in Campi Flegrei has been characterized by coexisting extensional and compressional tectonics in a NNE-SSW direction (S] and $2), together with obliquely trending strike-slip tectonic regimes ($3).

The relationship between stresses St and $2 is a simple swap between a, and tr3, with a2 re- maining constant, that could be also inter- preted as phenomena of elastic rebound and aftershock within a dominating NNE-SSW extension; stress $3 could be related to faulting along the external southwestern boundary of the dome in agreement with Aster and Meyer (1988).

The three-dimensional surface in Figure 5, fitted by a quadratic smoothing function

136 A. ZUPPETTA AND A. SAVA

through the hypocenters, suggests that the sources of the earthquakes with normal fault solutions are distributed along a WNW-ESE system of "quasi-conjugate" fractures comparable to that recognized by Cosentino et al. (1984) in the Solfatara area. Such a system is also consistent with the faulting pattern found at Gerolomini during excavation of the new tunnel of the Cumana railway (De Riso et al., 1988).

In the Campanian Apennine chain, the conjugate system correlates with a set of regional lineaments trending N 96 °- 105 ° mapped from Landsat satellite images (Sava et al., in prep. ) and corresponds to a family of a quasi-conjugate normal faults which show attitudes of the principal stress axes consistent with the al and a3 directions determined from S I earthquake focal mechanisms (Fig. 4a).

In conclusion, we infer that a NNE extensional tectonics (N 12 ° ) coupled with doming due to the vertical movements can account for the seismic pattern during the 1982-1985 Phlegrean crisis and that the spatial distribution of the normal fault related seismic activity was at least partly controlled by a pre-existing regional, quasi-conjugate, fracture system. The complex stress pattern we found at Campi Fle- grei is qualitatively comparable to the stress distributions calculated by Aster and Meyer ( 1988 ) using a different velocity model to de- rive fault plane solutions.

Acknowledgments

We thank our colleagues prof. Gennaro Cor- rado and prof. Giuseppe Luongo who made possible to access the Vesuvian Observatory seismic data. The work was greatly improved by the suggestions made by prof. Jacques An- gelier and by an anonimous colleague who re- viewed the manuscript. A special thank goes to our friend dr. Chris Kilbourne for critical reading and discussions. This work was funded under M.P.I. grants 40% and 60%.

References

Angelier, J., 1979a. Determination of the mean principal directions of stresses for a given fault population. Tec- tonophysics, 56: T 17-T26.

Angelier, J., 1984. Tectonic analysis of fault slip data sets. J. Geophys. Res., 89: 5953-5848.

Angelier, J., 1989. From orientation to magnitude in pa- leostress determinations using fault slip data. J. Struct. Geol., 11: 37-50.

Angelier, J. and Manoussis, S., 1980. Classification auto- matique et distinction des phases superpos6es en tec- tonique de failles. C. R. Acad. Sci., Paris, S6r. D, 290: 651-654.

Aster, R.C. and Meyer, R.P., 1988. Three-dimensional velocity structure and hypocenter distribution in the Campi Flegrei caldera, Italy. Tectonophysics, 149:195- 218.

Barberi, F., Innocenti, F., Lirer, L., Munno, R., Pescatore, T. and Santacroce, R., 1978. The Campanian Ignim- brite: A major prehistoric eruption in the Neapolitan area (Italy). Bull. Volcanol., 41: 1-22.

Berrino, G., Corrado, G., Luongo, G. and Toro, B., 1984. Ground deformation and gravity changes accompa- nying the 1982 Pozzuoli Uplift. Bull. Volcanol., 47: 187-200.

Bianchi, R., Coradini, A., Federico, C., Giberti, G., Sar- toffs, G. and Scandone, R., 1984. Modelling of Surface Ground Deformations in the Phlegraean Fields Vol- canic Area, Italy. Bull. Volcanol., 47 (2): 321-330.

Bott, M.H.P., 1959. The mechanisms of oblique slip faulting. Geol. Mag., 96:109-117.

Caccavale, G., 1984/85. Attivith sismica e meccanismi focali ai Campi Flegrei. Unpublished Thesis, Univer- sity of Napoli.

Capaldi, G., Civetta, L. and Gillot, P., 1985. Geochron- ology of Plio-Pleistocene rocks from Southern Italy. Rend. Soc. Mineral. Petrol., 40: 25-44.

Carey, E. and Brunier, B., 1974. Analyse th6orique et num6rique dun ModUle m6canique ~l~mentaire appli- qu6 h l'6tude d'une population de failles. C.R. Acad. Sci., Paris, S6r. D, 279: 891-894.

Cinque, A., Rolandi, G. and Zamparelli, V., 1985. L'es- tensione dei depositi marini olocenici nei Campi Fle- grei in relazione alia vulcano-tettonica. Boll. Soc. Geol. Ital., 104: 327-348.

Cocco, E., Cravero, E., Ortolani, F., Pescatore, T., Russo, M., Sgrosso, I. and Torre, M., 1972. Les Facies sedi- mentaires miocenes du Bassin Irpinien (Italie meridionale ). Atti Acc. Pontaniana, 21 : 1-13.

Colantoni, P., Del Monte, M., Fabbri, A., Gallignani, P., Selli, R. and Tomadin, L., 1972. Ricerche geologiche nel Golfo di Pozzuoli. Quad. Ric. Sci., 83: 26-76.

Coppa, U., Lo Bascio, A., Luongo, G. and Ricciardi, G. P., 1985. Meccanismi focali e campi di sforzi nell'area


flegrea. Atti 4 ° Convegno Gruppo Naz. Geof. Terra Solida, pp. 763-778.

Corrado, G., Guerra, I., Lo Bascio, A., Luongo, G. and Rampoldi, A., 1976/77. Inflation and microearth- quake activity of Phlegrean Fields. Italy. Bull. Vol- canol., 40: 169-188.

Cosentino, D., De Rita, D., Funiciello, R., Parotto, M., Salvini, F. and Vittori, E., 1984. Fracture System in Phlegraean Fields (Naples, Southern Italy). Bull. Vol- canol., 47: 247-257.

De Bonitatibus, A., Latmiral, G., Latmiral, G., Mirabile, L., Palumbo, A., Sarpi, E. and Scalera, A., 1970. Ri- lievi sismici a rifrazione:strutturali, ecograf- ici (fumarole), e batimetrici nel Golfo di Pozzuoli. Boll. Soc. Nat. Napoli, 79:97-113.

De Lorenzo, G., 1904. The history of volcanic action in the Phlegraean Fields. Q. J. Geol. Soc., 60:296-315.

Del Pezzo, E., Iannaccone, G., Martini, M. and Scarpa, R., 1983. The 23 November 1980 southern Italy earthquake. Bull. Seismol. Soc. Am., 73: 187-200.

De Riso, R., Ducci, D. and Rippa, F., 1988. Osservazioni geologiche lungo il margine meridionale della cupola del M. Olibano. Conv. Regione Campania-Univ. Na- poli. In: Bradisismo e Fenomeni Connessi, 4:184-199.

Di Girolamo, P., 1970. Differenziazione gravitative e curve isochimiche nella Ignimbrite Campana (Tufo Grigio Campano). Rend. Soc. It. Mineral. Petrol., 26/ 2: 547.

Di Girolamo, P., Ghiara, M.R., Lirer, L., Munno, R., Ro- landi, G. and Stanzione, D., 1984. Vulcanologia e Pe- trologia dei Campi Flegrei. Boll. Soc. Geol. Ital., 103: 349-413.

Di Vito, M., Lirer, L., Mastrolorenzo, G., Rolandi, G. and Scandone R., 1985. Volcanological Map of Campi Fle- grei. Min. Prot. Civ./Univ. Napoli, Napoli, Italy.

Di Vito, M., Lirer, L., Mastrolorenzo, G. and Rolandi, G., 1987. The 1538 Monte Nuo4o eruption (Campi Flegrei, Italy). Bull. Volcanol., 49:608-615.

Finetti, I. and Morelli, C., 1974. Esplorazione sismica a riflessione nei golfi di Napoli e Pozzuoli. Boll. Geofis. Teor. Appl., 16: 175-222.

Gaudiosi, G. and Iannaccone, G., 1984. A preliminary study of stress pattern at Phlegrean Fields as inferred from focal mechanism. Bull. Volcanol., 47:225-231.

Gasparini, C., lannaccone, G., Scandone, P. and Scarpa, R., 1982. Seismotectonics of the Calabrian arc. Tec- tonophysics, 84: 267-286.

Lirer, L., Luongo, G. and Scandone, R., 1987. On the volcanological evolution of Campi Flegrei. Eos, 68: 226- 227, 229, 231,233-234.

Mckenzie, D.P., 1969. The relation between fault plane solutions for earthquakes and the directions of the principal stresses. Bull. Seismol. Soc. Am., 59:591-601.

Michael, A.J., 1987. Use of focal mechanisms to determine stress: a control study. J. Geophys. Res., 92: 357- 368.

Ogniben, L., 1973. Conclusioni sullo stato attuale delle conoscenze nella geologia dell'Appennino. Acc. Naz. Lincei, 183: 369-445.

Parascandola, A., 1947. I fenomeni bradisismici del Ser- apeo di Pozzuoli. Genovese, Napoli.

Pescatore, T., Diplomatico, G., Senatore, M. R., Tramu- toli, M. and Mirabile, L., 1984. Contributo allo studio del Golfo di Pozzuoli:aspetti stratigrafici e strutturali. Mem. Soc. Geol. ltal., 27: 133-149.

Rittman, A., Vighi, L., Falini, F., Ventriglia, V. and Ni- cotera, P., 1950. Rilievo geologico dei Campi Flegrei. Boll. Soc. Geol. Ital., 69:117-362.

Rolandi, G., 1988. Le Ignimbriti della Piana Campana. Atti 74 ° Congr. Soc. Geol. Ital., vol.B: 350-352.

Rosi, M., Sbrana, A. and Principe, C., 1983. The Phle- graean Fields: Structural evolution, volcanic history and eruptive mechanisms. In: M.F. Sheridan and F. Barberi (Editors), Explosive Volcanism. J. Volcanol. Geotherm. Res., 27: 273-288.

Scandone, P., 1979. Origin of the Tyrrhenian sea and Cal- abrian Arc. Boll. Soc. Geol. Ital., 98: 27-34.

Thunell, R., Federman, A., Spark, S. and Williams, D., 1979. The age, origin and volcanological significance of the Y-5 ash layer in the Mediterranean. Quat. Res., 12: 241.

Westaway, R. and Jackson, J., 1984. Surface faulting in the southern Italian Campania-Basilicata earthquake of 23 November 1980. Nature, 312: 436-438.

Withjack, M.O. and Scheiner, C., 1982. Fault patterns associated with domes--An experimental and analytical study. Am. Assoc. Pet. Geol. Bull., 66: 302-316.

Stress pattern at Campi Flegrei from focal mechanisms of the 1982 1984 earthquakes (southern Italy

Documents