New Insights on the Seismic Activity of Ostuni (Apulia Region, Southern Italy) Offshore

Abstract

On 23 March 2018, an event of magnitude ML 3.9 occurred about 10 km from the town of Ostuni, in the Adriatic offshore. It was the most energetic earthquake in South–Central Apulia ever recorded instrumentally. On 13 February 2019, in the same area, a second ML 3.3 event was recorded. The analysis of the 2018 event shows that the ambiguity of the solution of the fault plane reported by INGV (Istituto Nazionale di Geofisica e Vulcanologia) on the Italian National Earthquake Centre website can be solved considering existing seismic profiles, exploration well logs and the Quaternary activity of faults in the epicentral area. A seismogenic source was identified in the rupture of a small portion of a 40 km length structure with strike NW-SE, dipping at a high angle toward the south. In this work, we have relocated the recent earthquakes by using the seismic stations managed by the University of Bari (UniBa), one of which is quite close to the event’s epicenter (about 20 km), together with data coming from the RSN (Rete Sismica Nazionale). Furthermore, we have determined the focal mechanism of some events, with implications on stress field of the area. Our results show right-lateral transtensional kinematics of the seismogenic faults along approximately E-W striking planes, with a tension, T, with a trend of about 60° (NE-SW direction) and a plunge of 20°. Finally, we have tried to correlate the location of the four best constrained earthquakes with their seismogenic structures.

Full text

Appl. Sci. 2025, 15, 784 https://doi.org/10.3390/app15020784
Article
New Insights on the Seismic Activity of Ostuni
(Apulia Region, Southern Italy) Offshore
Pierpaolo Pierri 1, Marilena Filippucci 1,2,*, Vincenzo Del Gaudio 1, Andrea Tallarico 1,2, Nicola Venisti 1
and Vincenzo Festa 1
1 Department of Earth and Geo-Environmental Science, University of Bari “Aldo Moro” (UniBa),
70126 Bari, Italy; pierpaolo.pierri@uniba.it (P.P.); vincenzo.delgaudio@uniba.it (V.D.G.);
andrea.tallarico@uniba.it (A.T.); nicola.venisti@uniba.it (N.V.); vincenzo.festa@uniba.it (V.F.)
2 National Institute of Geophysics and Volcanology (INGV), 00143 Roma, Italy
* Correspondence: marilena.filippucci@uniba.it
Abstract: On 23 March 2018, an event of magnitude ML 3.9 occurred about 10 km from the
town of Ostuni, in the Adriatic offshore. It was the most energetic earthquake in South–
Central Apulia ever recorded instrumentally. On 13 February 2019, in the same area, a
second ML 3.3 event was recorded. The analysis of the 2018 event shows that the ambigu-
ity of the solution of the fault plane reported by INGV (Istituto Nazionale di Geofisica e
Vulcanologia) on the Italian National Earthquake Centre website can be solved consider-
ing existing seismic profiles, exploration well logs and the Quaternary activity of faults in
the epicentral area. A seismogenic source was identified in the rupture of a small portion
of a 40 km length structure with strike NW-SE, dipping at a high angle toward the south.
In this work, we have relocated the recent earthquakes by using the seismic stations man-
aged by the University of Bari (UniBa), one of which is quite close to the event’s epicenter
(about 20 km), together with data coming from the RSN (Rete Sismica Nazionale). Fur-
thermore, we have determined the focal mechanism of some events, with implications on
stress field of the area. Our results show right-lateral transtensional kinematics of the seis-
mogenic faults along approximately E-W striking planes, with a tension, T, with a trend
of about 60° (NE-SW direction) and a plunge of 20°. Finally, we have tried to correlate the
location of the four best constrained earthquakes with their seismogenic structures.
Keywords: earthquake; seismogenic fault; Apulia Murge; Adriatic offshore; OTRIONS;
focal mechanism; stress field
1. Introduction
The Apulia region, in Southern Italy, is part of the Adria plate, a microplate whose
tectonic evolution is dominated by the collision with two major plates, Eurasia and Africa.
Two subduction zones in opposite directions are recognized: on the western side, the sub-
duction is considered still active beneath the Northern Apennine and the Calabrian arc
[1]; and on the eastern side, beneath the Dinarides, the subduction is considered extinct,
while it is considered still active in the Hellenic arc [2]. These complex tectonics involve
the collision toward the northwest with the Eurasia plate, where Adria is the upper plate,
involving a counterclockwise rotation in the Western Alps [3], whose Euler pole of rota-
tion is still debated (refer to the discussion in Le Breton et al. [4]). In Figure 1, we show a
schematic framework of the Adria plate and surroundings, according to the seismotec-
tonic model proposed by Meletti et al. [5].
Academic Editor: Felix Borleanu
Received: 31 October 2024
Revised: 8 January
2025
Accepted: 10 January 2025
Published:
14 January 2025
Citation:
Pierri, P.; Filippucci, M.;
Del Gaudio, V.; Tallarico, A.; Venisti,
N.; Festa, V. New Insights on the
Seismic Activity of Ostuni (Apulia
Region, Southern Italy) Offshore.
Appl. Sci.
2025, 15, 784. https://
doi.org/10.3390/app15020784
Copyright:
© 2025 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and con-
ditions of the Creative Commons At-
tribution (CC BY) license (https://cre-
ativecommons.org/licenses/by/4.0/).

Appl. Sci. 2025, 15, 784 2 of 21
All the boundaries of the Adria plate are characterized by an intense seismic activity
that can also be very deep (as Wadati–Benioff zone in the Hellenic and Calabrian arcs)
[1,6]. The inner part of Adria, where Apulia is located, is characterized by modest recent
seismic activity, if compared to the adjacent areas of the Apennine or of Albania and
Greece; however, it has been the site of strong earthquakes in the past. The Apulia region
has different characteristics of seismicity moving from the north to the south. The histori-
cal seismicity of Northern Apulia was analyzed in detail by using data of the CPTI Work-
ing Group [7]. They report 22 events with magnitude Mw > 5.2 [8], demonstrating that the
seismogenic structures of Northern Apulia can generate damaging earthquakes. Central
Apulia is characterized by low and sporadic seismicity due to a tensional stress regime,
possibly related to both Apennine and Northern Apulia seismogenic activity [9]. How-
ever, it suffers from a lack of seismological knowledge as a consequence of relatively poor
spatial and temporal coverage of seismic monitoring.
Figure 1. Structural sketch of Italy and surrounding areas (modified from Meletti et al. [5]). The
black arrow indicates the slip vectors of Africa vs. Europe and of Adria vs. Europe obtained from
geodetic data. Adria RP is the Adria rotation pole. The red square delimits the area analyzed in this
paper.
Seismic monitoring in Italy is managed by the National Institute of Geophysics and
Volcanology (Istituto Nazionale di Geofisica e Vulcanologia—INGV) through the Na-
tional Seismic Network (“Rete Sismica Nazionale”—RSN), whose code, within the Inter-
national Federation of Digital Seismograph Networks—FDSN—is IV. At present, this net-
work has only three seismic stations (NOCI, MESG and SCTE) operational in Southern
Apulia, after the closure of two other stations (LCI and BRT) in 2006. In recent years, the

Appl. Sci. 2025, 15, 784 3 of 21
seismic monitoring of this region has been supported by five seismic stations (TAR1,
MASS, FASA, CGL1 and PE1) managed by the Seismologic Observatory of the University
of Bari (OSUB). The recordings of this network, in conjunction with those of the RSN,
allowed for the detection of several low energy earthquakes, such as the ML 2.8 event of
May 5th, 2012, which occurred near Ostuni and was felt by many inhabitants.
To improve the seismic monitoring of the Apulian territory, in 2013, thanks to the
European project INTERREG, and in collaboration with INGV, the University of Bari
(UniBa) installed a local seismic network covering the entire territory of Apulia, the OTRI-
ONS network (FDSN code OT). The OT network also incorporated some stations already
operating in Southern Apulia, previously managed by OSUB (Figure 2). The details on the
operation of the OT network from 2013 to 2019 are described in Filippucci et al. [10]. Since
2021, other stations have been added, and in 2024, some sensors have been replaced (for
further details, see the page “https://www.fdsn.org/networks/detail/OT/ (accessed on 30
October 2024)”). From May 2019, the OT registrations are available also on EIDA (Euro-
pean Integrated Data Archive) INGV node and can be used to detect earthquakes on Ital-
ian territory. As a result, the OT network has improved the detection of earthquakes
throughout the territory of Apulia and Southern Italy, as demonstrated by the earth-
quakes listed in the ONT (National Earthquake Observatory) section of the INGV website.
The most energetic event ever recorded by seismic networks in Central–Southern Pu-
glia occurred on March 23rd, 2018, with Mw = 3.7 and ML = 3.9; this earthquake occurred
in the Southeast Murge Adriatic offshore, at the transition between the Apulian Foreland
and the Dinarides–Albanides foredeep domains, about 10 km from the town of Ostuni,
one of the most touristy and populated municipalities in the province of Brindisi. This
event was analyzed in a previous work by Festa et al. [11] to retrieve the geometry of the
seismogenic structure responsible for this earthquake, but the authors used the infor-
mation as downloaded from ONT, before the OT network was integrated into the RSN.
Figure 2. Seismic stations used in this study: in red, those belonging to seismic network managed
by UniBa (OSUB/OTRIONS); in blue, those belonging to other seismic networks (IV, IX, GE, MN,
CR, AC and TV).

Appl. Sci. 2025, 15, 784 4 of 21
Due to the lack of seismicity and of seismic monitoring, no information is available
on the tectonic stress regime for the Apulia foreland area in the Italian Present-Day Stress
Indicators database [12] (IPSI, https://ipsi.rm.ingv.it/ (accessed on 30 October 2024)). Fur-
thermore, no potential seismogenic faults capable of producing significant permanent tec-
tonic deformation at the surface (capable faults) are reported for this area in the Database
of Individual Seismogenic Sources [13] and in the database of active capable faults of the
Italian territory [14].
The aim of this paper is to integrate all the seismic data available on the ONT with
the OT registrations and to obtain a more robust catalog of earthquakes that occurred in
Central–Southern Apulia during the last 25 years. We also computed the focal mecha-
nisms of some available events and stress field to improve the knowledge of the seismo-
genic structures in Central–Southern Apulia. Finally, we have correlated the location of
the four best constrained earthquakes with their seismogenic structures.
2. Structural Setting
In the framework of the Adria plate geodynamics, the Southern Adriatic Sea area
represents the Oligocene–Quaternary foreland basin of the Dinarides–Albanides–Hellen-
ides orogen’s portion [15–18] (Figure 3). During the orogenic growth of the Dinarides–
Albanides–Hellenides, the Mesozoic–Eocene Adriatic Basin was gradually involved to the
west in the foreland basin, whose tectonic subsidence came to affect the southeastern part
of the adjacent Apulia Platform as well [18]. To the west, the Apulia Platform progres-
sively subsided, eastward, in the Neogene–Quaternary foredeep domain of the Southern
Apennines [19,20]. Therefore, a remnant of the Apulia Platform dominates the Apulian
Foreland, i.e., the Plio–Pleistocene foreland shared by Apennines and Dinarides–Alba-
nides–Hellenides [19,21] (Figure 3).
The uplift since the Middle Pleistocene of the Apulian Foreland occurred in relation
to a NW-SE striking regional, gentle buckle fold of the Adria plate, which occurred due to
the difficult eastward roll-back of the continental lithosphere during Apennines subduc-
tion [20].
In its upper part, the Apulian Foreland is chiefly represented by a sedimentary cover
lying above a Variscan crystalline basement [19] (Figure 3). From the bottom to the top,
the sedimentary cover consists of Permo–Triassic continental deposits belonging to the
Verrucano Fm (up to ca. 1000 m thick), Upper Triassic limestones/dolostones and anhy-
drites of the Burano Fm (up to ca. 2500 m in thickness), and Lower Jurassic limestones of
the Calcare Massiccio Fm (up to ca. 1000 m thick); moreover, the Middle Jurassic–Upper
Cretaceous inner platform carbonates, belonging to the Apulia Platform, are widely ex-
posed in the Apulian Foreland with a thickness of ~4 km [19,22,23].
The Apulia Platform–basin transition and adjacent Adriatic Basin (Figure 3) are tes-
tified by marginal and pelagic carbonates (both cropping out and drilled), respectively
[24,25]. Similar deposits, moreover, occupied during Upper Cretaceous narrow intra-plat-
form basins governed by extensional faults [25–27].

Appl. Sci. 2025, 15, 784 5 of 21
Figure 3. Schematic structural map of the region around the Southern Adriatic Sea (modified after
[28]); M-G = Mattinata–Gondola fault; HELLEN. = Hellenides; the solid black line encloses the
Murge area and the Northern Salento.
The central sector of the emerged Apulian Foreland is represented by the Murge (Fig-
ure 4), a morpho-structural high where the exposed carbonates of the Apulia Platform
have been grouped in the “Calcare di Bari” Fm (Lower Cretaceous) and the overlying
“Calcare di Altamura” Fm (Upper Cretaceous) [22] (Figure 4). Thin Plio–Pleistocene sed-
imentary bodies made of calcarenites, which in turn belong to the “Calcarenite di
Gravina” Fm, unconformably rest on the Cretaceous carbonates, and they crop out in
some inner places and especially on the flanks of the Murge high [22], the latter controlled
by normal faults [29–31] (Figure 4). Normal faults striking from NW-SE to W-E, toward
the Adriatic Sea coastline, and dipping from the NE to N, respectively, characterize the
Quaternary tectonics of the Murge area [30,31], and, together with associate faults dipping
in the opposite direction, gave rise to narrow grabens [30]. According to Festa [32], these
tectonic structures composed a system of faults (deformation zones, DZs) with normal
and right transtensional kinematics that were active during the deposition of the lime-
stones of the “Calcare di Altamura” Fm (Figure 4).
Oligocene–Quaternary deposits unconformably overlie the Mesozoic–Eocene plat-
form and basin carbonates in the Adriatic Sea offshore Murge [25], and they exhibit a
thickness increasing toward the inner Albanides foreland basin [15,17,25,26,33]. Dominant
extensional faults affected the platform–basin transition in the southeastern Murge [34]
and its Adriatic Sea offshore, where, moreover, they were active during Neogene as well
[26]. Here, the Monte Giove, a narrow E-W submarine relief, and its adjacent Rosaria Mare
intra-platform basin [25,26] (Figure 4) were formed as a result of faults activity up to the
Quaternary [28]. In this respect, the southern fault bordering Monte Giove relief, at the
transition with Rosaria Mare basin, exhibits evidence of present-day activity, representing
the possible seismogenic source of the 23 March 2018 event [11].

Appl. Sci. 2025, 15, 784 6 of 21
Figure 4. Structural sketch map of the Murge area and the Northern Salento (modified after [32]);
the fault striking from WNW (near the town of Matera) to ESE (south of the town of Brindisi) bor-
ders the Murge (to the north) from the Salento (to the south); the Monte Giove submarine relief and
Rosaria Mare basin (modified after [11,28]) are also indicated; red filled squares represent epicentral
relocation of seismic events for which the focal mechanism was determined in the present paper.
3. Analysis of the Historical and Instrumental Seismicity
The area of Central and Southern Apulia is historically characterized by low seismic-
ity [35] and is classified as a low-seismic-hazard area. In Figure 5, the historically docu-
mented events that occurred within 35 km from Ostuni and reported by different cata-
logues are plotted on the map as circles representing the focal volumes, according to Bath
and Duda formula [36].
In this area, the largest earthquake and the only one with a significant damage po-
tential is that which occurred on 26 October 1826, with Mw = 5.22 and macroseismic in-
tensity VI-VII MCS (Mercalli–Cancani–Sieberg), near Grottaglie (see red circle, data from
Parametric Catalogue of Italian Earthquakes, CPTI15 v. 4.0 [37,38]).
This event is not reported by the Catalogue of Strong Italian Earthquakes
(CFTI5MED) [39,40], which has selected only a subset of the events historically docu-
mented; however, it includes four smaller earthquakes with magnitudes between 3.2 and
3.7) (grey circles in Figure 5), all based on “a single location” and therefore considered of
uncertain reliability.
References to a set of additional events are found in the PFG catalogue [41], compiled
within the Geodynamic Finalized Project. This set consists of 16 events (yellow circles in
Figure 5), often with unreliable locations. Only four of them have intensities that could
have caused damage (I ≥ VI MCS): one of these events is that of 1826, reported by the
CPTI15; two others, of magnitude 4.1, that occurred in 1833 were attributed identical lo-
cations; and the fourth, which occurred on 26 February 1947, was placed in the offshore
Adriatic Sea, but its location, based on few instrumental observations, was very uncertain,
and the CPTI15 catalogue reports it with a completely different location (in the Tyrrhenian
Sea).
With regard to instrumental seismicity, the map in Figure 5 shows, with blue circles,
the earthquakes that occurred around the town of Ostuni between 1981 and 1999 (before

Appl. Sci. 2025, 15, 784 7 of 21
the network was digitized; source, Italian Seismicity Catalog—CSI catalogue [42]). Only
17 earthquakes are present in the analyzed area, with Mmax = 3.2.
According to the historical record of earthquakes felt in Ostuni, reported by the Da-
tabase Macrosismico Italiano (DBMI15) [43], only one event caused slight damage (VII
MCS), due to the strong Salento earthquake that occurred on 20 February 1743. The other
11 earthquakes were only perceived in Ostuni, with a maximum of V MCS on the occasion
of the disastrous Irpinia earthquake of 23 November 1980.
Figure 5. Seismic events located within 35 km from Ostuni (red dashed circle), identified by (i) the
CPTI15 version 4.0 catalogue (red circles, [37,38]), (ii) the CFTI5MED (grey circles, [39,40]), (iii) the
PFG catalogue (yellow circles, [41]) and (iv) the CSI catalogue (blue circles, [42]). Circles represent
focal volumes according to Bath and Duda formula [36]. The geographical position of main localities
mentioned in the paper is also shown as black squares.
4. Data Selection and Hypocenter Re-Location
We extracted from the ONT web-catalog “https://terremoti.ingv.it/ (accessed on 30
October 2024)” a dataset consisting of earthquakes that occurred within a radius of 35 km
from the town of Ostuni (lat. 40.7332 N–long. 17.5786 E). The extraction covers a period
from January 2000 to September 2024, which coincides with the era of the seismic network
digitization in Italy. There are a total of 32 events in the ONT web-catalog, and they are
listed in Table 1 and mapped in Figure 6. The magnitude ranges from 2.0 to 3.9, with an
average value 
� = 2.4.
Table 1. List of earthquakes located by INGV between 2000 and September 2024 within 35 km from
Ostuni (40.7332 N–17.5786 E), with M ≥ 2.0. ID = identification number. Date is expressed in Year-
Month- Day. Time is the earthquake origin time (UTC). Depth (km) is the earthquake depth (fixed
if marked with *). RMS (s) is the root mean square of the travel time residuals; M is the “preferred”
magnitude taken from INGV Seismic Bulletin (ML, Md or Mw).

Appl. Sci. 2025, 15, 784 8 of 21
Of these 32 events, seismograms were available for a review of picking only for the
RSN recordings from 2008 onward, while for those up to 2007, we used the time picks
provided by INGV. When possible, the picking procedure of P and S waves was carried
out manually, by visual inspection. These data were integrated with the recordings ac-
quired by the OSUB and OT networks and stored in their laboratories. So, the dataset was
collected as follows:
• From January 2000 to December 2007: time picks of P and S waves were downloaded
from the INGV web-service “https://terremoti.ingv.it/?timezone (accessed 15 April
2024)” and integrated by OSUB recordings;
• From January 2008 to March 2013: recordings were downloaded from the INGV
web-service “https://terremoti.ingv.it/?timezone (accessed 15 April 2024)” and inte-
grated by OSUB recordings;
• From April 2013 to April 2019: recordings of the IV network were downloaded from
the INGV web-service “https://terremoti.ingv.it/?timezone (accessed 15 April 2024)”
and integrated by OT network recordings;
• From May 2019 to today: recordings of the IV and OT networks were downloaded
from the INGV web-service “https://terremoti.ingv.it/?timezone (accessed 15 April
2024)”.
ID
Date
Time
Lat (N)
Long (E)
Depth
RMS
M
Location
1
2000-08-26
10:27:15.02
40.9940
17.7750
5 *
0.30
2.7
Costa Adriatica Brindisina (Brindisi)
2
2001-05-15
23:00:46.90
40.8080
17.3400
10.8
0.20
2.3
3 km SW Fasano (BR)
3
2001-07-15
11:17:29.88
40.6650
17.5400
5 *
0.60
3.0
3 km NE Ceglie Messapica (BR)
4
2002-06-06
02:18:22.59
40.7780
17.3150
10 *
0.20
2.4
3 km NW Locorotondo (BA)
5
2006-05-16
09:04:23.51
40.5440
17.3050
6.1
0.10
2.1
3 km SW Montemesola (TA)
6
2006-08-18
12:01:54.48
40.5250
17.3680
10 *
0.45
2.0
3 km NW Monteiasi (TA)
7
2006-08-23
12:44:37.74
40.5480
17.3640
10 *
0.33
2.0
3 km SE Montemesola (TA)
8
2007-06-04
13:06:28.40
40.5690
17.3540
10 *
0.44
2.0
2 km E Montemesola (TA)
9
2007-07-20
12:16:17.20
40.5420
17.3670
10 *
0.27
2.0
4 km SE Montemesola (TA)
10
2008-05-11
23:03:14.39
40.8210
17.6980
1 *
0.84
2.6
Costa Adriatica Brindisina (Brindisi)
11
2009-05-12
09:46:10.49
40.5560
17.2740
6.9
0.08
2.3
5 km W Montemesola (TA)
12
2009-06-16
09:27:08.09
40.5680
17.2480
7.1
0.39
2.0
4 km E Statte (TA)
13
2009-08-23
10:15:13.53
40.8050
17.8270
4 *
0.65
2.0
Costa Adriatica Brindisina (Brindisi)
14
2009-09-07
13:25:53.31
40.7910
17.6660
10 *
0.47
2.1
Costa Adriatica Brindisina (Brindisi)
15
2010-07-07
09:01:56.74
40.5620
17.2610
6.3
0.40
2.3
5 km E Statte (TA)
16
2011-05-13
06:21:29.61
40.7480
17.5160
7.9
0.41
2.3
6 km W Ostuni (BR)
17
2012-05-05
12:44:02.94
40.5393
17.5415
5.1
0.44
2.8
4 km W Francavilla Fontana (BR)
18
2012-06-13
09:11:41.29
40.5513
17.2473
5 *
0.32
2.0
4 km E Statte (TA)
19
2012-12-22
19:31:28.31
40.9992
17.3572
10 *
0.32
2.2
Costa Adriatica Barese (Bari)
20
2013-08-11
06:37:09.27
40.7378
17.4150
5.0
0.34
2.2
1 km W Cisternino (BR)
21
2015-06-25
14:37:47.55
40.5142
17.5865
2.8
0.22
2.5
2 km S Francavilla Fontana (BR)
22
2015-10-28
18:53:31.69
40.7808
17.4195
4.9
0.34
2.8
5 km N Cisternino (BR)
23
2018-03-23
23:31:56.81
40.8003
17.6938
29.7
0.27
3.9
Costa Adriatica Brindisina (Brindisi)
24
2019-02-13
21:56:44.06
40.8253
17.8143
29.8
0.48
3.2
Costa Adriatica Brindisina (Brindisi)
25
2019-05-20
00:57:24.29
40.7938
17.6608
4 *
0.42
2.5
Costa Adriatica Brindisina (Brindisi)
26
2019-10-09
22:36:49.75
40.9382
17.3863
6.8
0.25
2.2
Costa Adriatica Barese (Bari)
27
2020-01-16
05:00:47.66
40.7682
17.2002
25.8
0.37
2.3
4 km SW Alberobello (BA)
28
2021-02-11
19:36:08.18
40.6893
17.2183
29.9
0.29
2.6
9 km N Crispiano (TA)
29
2021-04-21
06:10:18.37
40.7557
17.9547
25.5
0.51
2.9
Costa Adriatica Brindisina (Brindisi)
30
2021-05-31
22:41:20.32
40.8958
17.3127
22.7
0.37
2.5
6 km S Monopoli (BA)
31
2021-05-31
23:11:36.74
40.8923
17.3028
21.1
0.32
2.3
7 km S Monopoli (BA)
32
2022-04-09
13:45:22.67
40.9078
17.7398
27.2
0.32
2.6
Costa Adriatica Brindisina (Brindisi)

Appl. Sci. 2025, 15, 784 9 of 21
In the Supplementary Materials, we release the txt file containing all the arrival times
of the events analyzed in this paper.
The 32 events were relocated with the HYPOELLIPSE code [44], using, in addition to
the previously available data, the new recordings and the revised phases derived from the
re-picking. Seven different velocity models were tested: those commonly adopted by the
INGV, CSTI, AK135 and PREM; and the models specifically proposed for this region by
Calcagnile and Panza [45], Costa et al. [46] and Venisti et al. [47]. In addition, different
values of the / ratio were tested.
Figure 6. Circles (with ID number) and squares indicate the epicenters determined by INGV and
resulting from the re-locations, respectively, of the 32 events examined in this work. In red, the
events occurred offshore in the Adriatic Sea; in light blue, the events occurred on land; and in blue,
the events occurred near Taranto, some of which are likely quarry blast (according to [48]). Black
squares indicate municipalities.
The results of all of these relocations indicate that the best model for this area in terms
of travel-time residuals and hypocenter location errors is that of Calcagnile and Panza
[45], with / = 1.78. The parameters of relocation of the 32 events are reported in
Table 2 and plotted on the map in Figure 6. The velocity model is shown in Table 3.
Some of the events in Table 2 had already been localized by Pierri et al. [48] (ID: 10,
13, 16, 17 and 19), who analyzed the seismicity in the “Penisola Salentina” seismic district,
an area much larger than that analyzed in this paper. Regarding the seismicity around the
city of Taranto, some earthquakes (for example the one with ID 9) are most likely quarry
blast (according to Pierri et al. [48]).
We observe a reduction in the hypocentral errors (generally less than 5 km) and in
the minimum epicentral distance (Dmin), which is reflected in the lower value of the RMS
(on average, from 0.37 s to 0.26 s) with respect to the ONT catalog; instead, we do not
observe a significant improvement in the azimuthal gap since the azimuthal coverage of
the network has not changed. The variations in the epicentral location are almost always
less than 5 km: only for events 2 and 3, the variations are greater than 15 km, but for these
events, no repickings were carried out.

Appl. Sci. 2025, 15, 784 10 of 21
An analysis of the relocation parameters shows a clear improvement from the first 19
events up to 2012 to the other 13 events that have occurred since 2013: the number of
phases recorded by the UniBa and INGV stations (on average, equal to 6 and 18, respec-
tively) has increased from 4 to 10 and from 12 to 27; the number of stations used in total
(on average equal to 15) has increased from 9 to 25; the azimuthal gap (on average equal
to 212°) has decreased from 245° to 163°; and the minimum distance (on average equal to
27 km) has decreased from 29 to 24 km.
Table 2. List of relocated earthquakes: Nd1 and Nd2 represent the number of used phases (P and S)
of the UniBa and INGV stations; Ns is the number of used stations; Dmin (km) is the minimum
epicentral distance; Gap (°) is the azimuthal gap; SEH1, SEH2 and SEZ are the horizontal and verti-
cal 68% confidence limits of the error ellipsoid; refer to Table 1 caption for other parameters.
Table 3. Velocity model [45] used in seismic event relocation by the Hypoellipse code [44]: Vp and
Vs are P-wave and S-wave velocities (Vp/Vs = 1.78); D is the depth of the bottom of each layer.
Layer
Vp (km/s)
Vs (km/s)
D (km)
1
4.00
2.25
2.0
2
6.10
3.43
19.0
3
6.80
3.83
33.0
4
8.10
4.55
90.0
5
8.20
4.61
∞
ID
Date
Time
Lat (N)
Long (E)
Depth
Nd1
Nd2
Ns
Dmin
Gap
RMS
SEH1
SEH2
SEZ
1
2000-08-26
10:27:15.03
41.0843
17.8188
14.8
6
10
9
61.1
233
0.20
1.2
3.3
1.8
2
2001-05-15
23:00:46.60
40.9548
17.3603
18.0
4
8
6
17.7
204
0.31
5.7
14.4
5.3
3
2001-07-15
11:17:30.00
40.7872
17.5989
17.6
6
26
21
15.4
192
0.34
3.3
7.0
2.4
4
2002-06-06
02:18:22.03
40.8274
17.3086
21.4
6
8
7
9.6
156
0.30
2.8
4.3
4.9
5
2006-05-16
09:04:24.13
40.5327
17.3060
10.8
0
6
3
35.0
325
0.05
1.6
3.4
9.1
6
2006-08-18
12:01:56.63
40.5324
17.2058
6.3
3
6
5
30.9
246
0.16
1.5
3.4
29.4
7
2006-08-23
12:44:39.08
40.5491
17.2446
16.6
2
6
4
30.7
305
0.14
3.1
6.7
10.8
8
2007-06-04
13:06:29.96
40.5358
17.2276
5.0
4
6
5
31.3
226
0.15
1.5
1.9
35.5
9
2007-07-20
12:16:17.82
40.5355
17.3327
18.9
0
8
4
36.1
295
0.05
0.8
3.1
1.0
10
2008-05-11
23:03:13.17
40.8425
17.8297
8.0
5
19
17
58.1
303
0.39
2.0
2.5
7.4
11
2009-05-12
09:46:11.06
40.5801
17.2830
16.6
0
6
3
29.7
318
0.01
0.2
0.3
0.4
12
2009-06-16
09:27:08.74
40.5995
17.2473
14.2
0
16
8
26.0
286
0.18
1.5
3.2
4.2
13
2009-08-23
10:15:13.61
40.8210
17.8711
4.5
0
6
3
26.0
241
0.23
4.5
35.8
99.0
14
2009-09-07
13:25:53.69
40.8197
17.6766
5.9
0
6
3
29.5
244
0.24
3.5
14. 5
69.3
15
2010-07-07
09:01:57.57
40.6108
17.2518
16.0
0
9
5
25.3
287
0.19
2.5
5.1
3.3
16
2011-05-13
06:21:29.82
40.7722
17.5286
18.7
15
19
18
13.8
178
0.38
0.8
1.2
0.8
17
2012-05-05
12:44:03.92
40.5325
17.5418
7.3
14
38
29
14.7
99
0.43
0.5
1.1
3.3
18
2012-06-13
09:11:41.70
40.5688
17.3045
6.7
0
7
4
31.8
313
0.25
17.9
25.8
99.0
19
2012-12-22
19:31:28.24
41.0038
17.3650
23.6
12
8
10
20.9
212
0.27
0.6
1.0
1.8
20
2013-08-11
06:37:09.95
40.7680
17.4156
10.0
10
12
12
13.6
168
0.15
0.5
1.0
3.1
21
2015-06-25
14:37:48.44
40.5037
17.5967
5.9
0
16
9
23.5
121
0.14
0.5
1.4
14.0
22
2015-10-28
18:53:32.53
40.7853
17.4401
5.0
0
52
34
31.7
169
0.60
1.7
4.6
4.1
23
2018-03-23
23:31:57.42
40.8134
17.7035
24.2
23
60
61
26.1
99
0.51
1.2
1.8
1.4
24
2019-02-13
21:56:45.22
40.8480
17.7432
13.6
6
14
15
30.1
212
0.28
1.9
3.1
8.9
25
2019-05-20
00:57:25.39
40.7948
17.6619
6.7
4
27
24
27.9
200
0.38
0.9
1.8
4.4
26
2019-10-09
22:36:50.81
40.9528
17.3850
7.5
18
29
24
32.6
185
0.57
1.6
2.5
3.4
27
2020-01-16
05:00:48.28
40.7592
17.1693
23.3
7
21
16
9.5
140
0.14
0.6
0.9
1. 7
28
2021-02-11
19:36:08.83
40.6762
17.1937
27.2
10
23
21
6.4
124
0.18
0.6
1.0
1.3
29
2021-04-21
06:10:17.71
40.8621
17.9819
17.3
8
14
14
32.3
215
0.26
1.2
4.6
2.2
30
2021-05-31
22:41:20.97
40.9011
17.3076
16.4
16
41
44
24.0
119
0.37
3.6
5.9
3.7
31
2021-05-31
23:11:37.39
40.8874
17.2925
16.5
16
25
27
22.1
165
0.28
2.9
5.2
3.6
32
2022-04-09
13:45:23.62
40.8761
17.7062
25.8
11
18
20
31.3
196
0.32
1.8
3.3
4.7

Appl. Sci. 2025, 15, 784 11 of 21
5. Focal Mechanisms
Information on the orientation of the seismogenic structures can be derived from the
determination of focal mechanisms. These are also necessary to calculate the stress field
to which the area is subjected, since it is well inside the Adria microplate, which is highly
deformed at the edges, not too far from the area under study (as discussed in Section 1
and shown in Figure 1).
We computed focal mechanisms by using P-wave polarities and the FPFIT code [49].
Since the magnitude of the earthquakes selected for this study is generally low, it is not
easy to correctly distinguish polarities from the signal noise in a sufficient number to de-
termine the focal mechanisms. P-wave polarities were determined manually and picked
on the seismograms of the OSUB and OT network for the entire period examined and on
those of the INGV network only for the events from 2008 onward.
The velocity model used for the computation of take-off angles is the same as that
used for location [45]. Fault plane solutions were considered well constrained only if de-
rived using a minimum of 10 clearly readable polarities homogeneously distributed on
the focal sphere. This quality criterion has led us to accept the focal mechanism solution
for only 7 earthquakes (ID 10, 16, 17, 23, 24, 25 and 29) out of the initial 32. Three of these
earthquakes (ID 10, 16 and 17) were already analyzed by Pierri et al. [48].
For each of these events, we have chosen, from the multiple solutions proposed by
FPFIT, the one that minimized the distance of the nodal planes from the mismatching
polarity data points. To better constrain the FPFIT inversion, we assigned a weight to the
polarity from 0 to 2 (following the picking weighting used by the SAC v. 101.6a software).
Assigning weights to the polarity data used in the FPFIT inversion resulted in a reduction
in the misfit, F. The best fit solution of each event was determined by minimizing the
residuals between the observed and theoretical amplitudes, exploring a search grid of val-
ues of strike, φ; dip, δ; and rake, λ, spaced at 5°.
The results are shown in Table 4, where we labelled with the subscripts 1 and 2 the
two nodal planes of the double-couple solution in terms of φ, δ and λ; we also reported
the P and T axes orientation as the plunge (inclination measured downward relative to
the horizontal plane) and trend (azimuth measured in the direction of the plunge) angles.
The P and T axes represent the directions of the maximum, σ1, and the minimum, σ3,
principal stress axes, respectively. The fault type (FT in Table 4) can be identified, as pro-
posed by Frepoli et al. [50], by plotting the combination of T and P plunges: based on the
position of the point representing the fault plane solution on this graph, the fault type can
be defined according to the diagram legend, as shown in Figure 7.
In Table 4, Npol is the number of polarities matching with the focal solution com-
pared to the total number of polarities, Ntot, available for each event. The quality of the
solution is expressed by the quality factors, Qf and Qp. Qf gives information about the
solution misfit, F, of the polarity data and assumes values that depend on F; Qp reflects
the solution uniqueness in terms of the 90% confidence region for the three angular pa-
rameter uncertainties, ∆φ, ∆δ and ∆λ. Qf and Qp range from class A to class C for decreas-
ing quality, according to Table 5.
Table 4. List of fault plane solutions of the events. For each event, the table reports the identification
number (ID); the angles of strike (φ), dip (δ) and rake (λ) of the 2 nodal planes; the trend and plunge
angles of the P and T axes; the number of polarities, Npol, that match compared to the total; the
quality factors (Qf and Qp); and the fault type (FT: NS, normal/strike–slip; SS, strike–slip; U, un-
known; and NO, normal). The fault plane solution obtained for the composite mechanism is also
reported.
ID
φ
1
δ
1
λ
1
φ
2
δ
2
λ
2
Trend P
Plunge P
Trend T
Plunge T
N
pol
/N
tot
Qf
Qp
FT

Appl. Sci. 2025, 15, 784 12 of 21
10
100
35
−160
353
79
−57
297
46
58
26
10/11
C
C
NS
16
95
55
−180
5
90
−35
314
24
56
24
11/13
C
B
SS
17
80
65
−150
336
63
−28
299
38
208
1
10/10
A
A
NS
23
117
84
125
215
35
10
179
31
59
41
44/48
B
B
U
24
80
80
−140
342
51
−13
309
35
205
19
10/10
A
B
NS
25
281
31
−107
120
60
−80
55
73
203
14
10/10
A
A
NO
29
127
31
−161
20
80
−60
321
47
86
29
11/13
C
A
NS
joint
70
30
−160
323
80
−62
262
47
30
29
84/115
C
B
NS
We can observe that four of the selected focal mechanisms (ID = 10, 17, 24 and 29)
represent normal/strike–slip faults, whereas only one (ID = 16) is a pure strike–slip mech-
anism and one (ID = 25) is a pure normal fault.
The focal mechanism of the strongest event (ID = 23) has an uncertain character, as it
is of type U; the mechanism found is quite similar to that obtained by INGV using the
Time-Domain Moment Tensor (TDMT) technique [51] on eight stations (“https://terre-
moti.ingv.it/event/18504011 (accessed on 15 April 2024)”), also classified as type U.
For most events (see Figure 8), the best solution has a pressure axis (P) with a trend
of about 300° and a plunge of about 40°, whereas the tension (T) axis has a trend of about
60° (NE-SW direction) and a plunge of 20°. These events reveal normal/strike–slip faulting
mechanisms along approximately E-W striking planes, and, in particular, the best con-
strained mechanism of the 5 May 2012 earthquake (ID = 17) is the great representative of
this kind of solution.
A composite fault plane solution was also obtained by combining the 115 P-onset
polarities of these seven events; in Figure 8, the beach-ball is shown with black and white
quadrants; the joint solution of the normal/strike–slip type shows, on average, a trend and
plunge of T and P axes similar to those obtained for the individual focal mechanisms.
Analysis of the T-axis orientations of normal and normal/strike–slip solutions sug-
gests a widespread NE-SW extensional regime.
Table 5. Criteria for the assignment of the quality factors, Qf and Qp, to the fault plane solution (as
described in [49]), based on the value of the misfit function, F (first panel), and the uncertainties
affecting the parameters ΔSTR, ΔDIP and ΔRAK (second panel), respectively.
F
Qf
Δ
STR
, Δ
DIP
, Δ
RAK
Qp
F < 0.025
A
< 20°
A
0.025 < F < 0.1
B
20° to 40°
B
F > 0.1
C
> 40°
C

Appl. Sci. 2025, 15, 784 13 of 21
Figure 7. Diagram of T-axis plunge vs. P-axis plunge for the 7 fault plane solutions determined in
this study. In this diagram the field of values near the vertices represent strike–slip (SS), reverse (RE)
and normal (NO) solutions. RS and NS are oblique-type mechanisms, whereas the solutions in the
U field are defined as unknown (modified by [50]). ID = identification number (see Table 1).
Figure 8. Focal mechanisms of single events (grey/white beach balls) computed in this study and
relative epicentral relocation (red square); for each beach-ball date, magnitude of event, quality

Appl. Sci. 2025, 15, 784 14 of 21
factors and polarities (open circles for distensive and little crosses for compressive polarities) and
stress axes (grey triangle for P-axis and white triangle for T-axis) are shown. Composite focal mech-
anism solution is shown as black/white beach ball (black triangle for P-axis and white triangle for
T-axis). Black squares indicate municipalities.
6. Stress Regime
As we discussed in the Introduction, information on stress field orientation and seis-
mogenic sources for Southern Apulia is lacking in the Italian catalogs, and an effort is
needed to fill in this gap. A crucial parameter that provides information about regional
tectonics and deformation mechanisms is the stress axis orientation, which can be re-
trieved using different techniques based on focal mechanisms.
To obtain an estimate of the orientation of the stress tensor, we performed the inver-
sion of stress field orientation by applying the FMSI (Focal Mechanism Stress Inversion)
code developed by Gephart and Forsyth [52,53]. This inversion method can retrieve four
of the six independent components of the stress tensor, commonly represented by the di-
rections of the three principal stress axes (σ1, σ2 and σ3) and a dimensionless parameter
R = (σ2 − σ1)/(σ3 − σ1), which constrains the shape of the stress ellipsoid and ranges be-
tween 0 and 1. The angular difference between the shear stress on the fault plane, com-
puted by the stress tensor inversion, and the observed slip direction on the same fault
plane, obtained by focal mechanisms, measures the discrepancy (or misfit) between the
data and the model.
The dataset consists of the trend and plunge angles of the T and P axes, as obtained
by FPFIT inversion. To better constrain the stress values obtained by the inversion, we
adopted a weighting scheme given by the weight, W, which takes into account the quality
of the focal mechanism solution, described by Qf and Qp, and the event magnitude (taken
as reported in Table 1). The Qf and Qp of each focal mechanism solution (Table 4) are
converted from letters to numbers, according to Table 6. Earthquake magnitude is in-
cluded in the value of W on the hypothesis that the regional stress should be better repre-
sented by main earthquakes since small earthquakes may represent stress accommodation
near the seismogenic source of the main event. We assigned W values that decrease with
the quality factors of the focal mechanism inversion and increase with the earthquake
magnitude. We then assigned a weight, W, to each fault plane solution in Table 4 as a
function of the total quality factor, Qt = Qf + Qp + M, according to Table 7.
Table 6. Weighting criteria assigned to the fault plane solution in the FMSI inversion.
Qf = A, Qp = A
Qf = B, Qp = B
Qf = C, Qp = C
3
2
1
Table 7. Relative weight as a function of the total quality factor (Qt).
Qt
Weight W
Qt < 4.5
3.0
Qt ≥ 4.5
4.0
Qt ≥ 5.5
5.0
Qt ≥ 6.5
6.0
The acceptability and homogeneity of the stress inversion solution was evaluated
following the procedure of Lu et al. [54], based on the simultaneous satisfaction of two
criteria. The first requires that the 95% confidence intervals of σ1 and σ3 do not overlap
for the solution to be acceptable. The second accounts for the degree of heterogeneity of

Appl. Sci. 2025, 15, 784 15 of 21
the investigated medium, requiring that the misfit angle be under a certain threshold (mis-
fit < 6°) to consider a solution to be homogeneous.
The result is shown in Figure 9. The 95% confidence interval of the solution is very
narrow, and the misfit = 2.4° indicates homogeneity of the medium. The stress ratio of R
= 0.5 indicates that σ2 has a value exactly intermediate, and the misfit = 2.4° indicates a
high degree of homogeneity of the solution. Following the notation of the stress regime
assignment for earthquake focal mechanism data [55] in World Stress Map, we have the
following:
• σ3, corresponding to the minimum horizontal stress, Shmin, is sub-horizontal and
oriented as the trend of T axes, 36° N;
• σ1, corresponding to the maximum horizontal stress, Shmax, is approximately sub-
horizontal and oriented normal to Shmin;
• σ2, corresponding to the vertical stress or pure lithostatic pressure, Sv, is quite verti-
cal.
In Figure 9, the double-couple focal mechanism corresponding to the mean stress
tensor solution is also shown; between the two nodal planes, having a strike approxi-
mately N-S and E-W, the hypothetical fault plane is most likely the one having strike = 77°
(dip = 53°; and rake = −175), according to Tropeano et al. [31], Gambini and Tozzi [56], and
ZS9 [57].
This result is remarkable and reflects the general dominance of a right-lateral strike–
slip regime with approximately N-S/E-W nodal planes.
Figure 9. Results of the FMSI inversion for the considered event group. From the left: stereonet plot
with 95% confidence limits for the principal stress axes σ1 (in fuchsia), σ2 (in green) and σ3 (in light
blue); plunge/trend angles of σ1, σ2, σ3, R value and misfit angle of solution; double-couple focal
mechanism corresponding to the mean stress tensor solution (black triangle for P-axis and white
triangle for T-axis).
7. Discussion and Conclusions
In the present paper, the seismicity of southeastern Murge was reconsidered and an-
alyzed. Thanks to the registration of the OSUB and OTRIONS seismic networks, new re-
cordings from stations close to the epicenters were retrieved, improving the quality of the
relocations in terms of minimum epicentral distance and location uncertainty. The seismic
activity is sporadic, as evidenced by the very low number of significant events recorded
(only 32 earthquakes with M ≥ 2.0 in 25 years).
The collected earthquakes, distributed over a horizontal distance of ~50 km, with a
range of focal depths from 4.5 km to 27 km, indicate that the whole Earth crust is involved
in this sporadic seismic activity.
The seven focal mechanisms, although obtained using a limited number of P-wave
polarities, appear quite well constrained and homogeneous, as can be observed in Figure
8. All the focal solutions have a common nodal plane in the approximately E-W direction,
N
σ1 = 21°/297°
σ2 = 59°/168°
σ3 = 22°/36°
R = 0.5
Misfit = 2.4°

Appl. Sci. 2025, 15, 784 16 of 21
as shown by the φ1 values, which range between 80° N and 127° N (see Table 4). This is
the same direction along which earthquake epicenters tend to align (Figure 6), neglecting
the earthquakes in the neighboring of Taranto City, so, between the two nodal planes of
the mechanism solution, the plane 1 should correspond to the actual fault plane. It should
be noted that this E-W fault striking fits the seismogenic zonation of Italy [57] very well.
The stress regime inferred in this study indicates that seismicity, even if it is of low
frequency and magnitude, occurs according to right-lateral transtensional kinematics of
the seismogenic faults. The homogeneity of the focal plane solutions provided a high-
quality stress-inversion solution. The orientation of Shmin, according to the direction of
the trend of the σ3 axis (36° N), agrees with the results of the Shmin in adjacent areas (as
shown in Figure 10, modified from Mariucci and Montone [12]), indicating that this region
is subject to a regional stress regime that controls the tectonics of the Adria plate.
In an attempt to correlate earthquakes and their seismogenic structures, we selected
the seismic events with the best constrained location (20120505, 20180323, 20190213 and
20190520), recorded by a more recent and denser network of seismographs, which pro-
vided locations with a total quality factor A or B. In agreement with Festa et al. [11], the
location of the seismic event 20180323 is consistent with its association to the western
branch of the fault bounding the Monte Giove structural high and the adjacent Rosaria
Mare basin (Figure 11a). Moreover, the eastern branch of this fault seems to have been the
seismogenic source for the seismic event 20190213 (Figure 11a). A good consistency is also
observed between the eastern branch of the fault belonging to the Central Deformation
Zone (Figure 4) and the location of the seismic event 20190520. Finally, the epicenter of
the seismic event 20120505 is close to the fault striking from WNW (near the town of Ma-
tera) to ESE (south of the town of Brindisi), bounding the southern Murge (Figure 11a).
In the southeastern Murge area, the ca. NE-SW regional elongation has occurred
since Middle Pleistocene [58]. Accordingly, such a feature of the regional strain field is
coherent with the obtained stress field (Figure 9), which could determine, in the south-
eastern Murge, the reactivation of ca. E-W striking ancient faults (at least of the Late Cre-
taceous) with a right horizontal component of simple shear (Figure 11a).
Such a stress field is, moreover, tectonically coherent with the outer NW-SE striking
lithospheric buckle fold of the Adria plate, which occurred during Apennines subduction
and determined the uplift of the Apulian Foreland since the Middle Pleistocene (Figure
11b) [20].
Since the area is characterized by very sporadic events, the possibility of underesti-
mating the resulting hazard and thus increasing the probability of facing unexpected
earthquakes is significant. This makes uncertain the prediction of the magnitude of major
local earthquakes that can be expected, their peak ground acceleration and their recur-
rence interval. In this regard, it is useful to introduce the concept of Maximum Credible
Earthquake (MCE), which is based on the seismic history and seismotectonics of the area
and is a basic ingredient in the Neo-Deterministic Seismic Hazard Assessment (NDSHA)
to calculate seismic hazard [59].
With this approach, the Maximum Credible Earthquake, representing the largest
physically possible scenario event at a given site, can be represented by a value, Mdesign,
equal to the maximum magnitude, Mmax, observed or estimated in the study area, plus a
multiple of its overall standard deviation, γEMσM [60]. To adopt a conservative approach,
it is currently wise to set γEMσM as 0.7 (Panza–Rugarli law), according to [61], so that Mdesign
= Mmax + γEMσM = Mmax + 0.7 [60]. For the study area, this would result in Mdesign = 3.9 + 0.7
= 4.6.
This magnitude value suggests looking at the geometry of the seismogenic structure
at a wider scale than what has been considered so far, since the most common

Appl. Sci. 2025, 15, 784 17 of 21
relationships between the magnitude value vs. length of the fault [62,63] show that to M
~4 corresponds to a length of the fault of about 1 km.
For this purpose, it may be wise to increase the density of the seismic network and
thereby improve seismic monitoring with a view toward reliable seismic risk assessment.
Figure 10. Map extracted from the IPSI database (modified from [12]). Colored segments refer to
minimum horizontal stress (Shmin) orientations; see legend for color explanation. The result of this
study is also superimposed.
Figure 11. (a) Structural sketch map showing the attempt to correlate the major faults with the seis-
mic events 20120505, 20180323, 20190213 and 20190520 in the southeastern Murge area (legend for
the faults as in Figure 4). (b) Schematic block diagram showing the Apennines subduction in corre-
spondence of the Apulian Foreland, adjusted for the Murge area (modified after [20]); the red arrows
indicate the regional elongation (as in Figure 11a) on the outer lithosphere buckle fold.
Supplementary Materials: The following supporting information can be downloaded at:
www.mdpi.com/10.3390/app15020784/s1.
Author Contributions: Conceptualization, P.P., M.F. and V.F.; methodology, P.P.; validation,
V.D.G. and A.T.; formal analysis, P.P., M.F. and N.V.; investigation, P.P. and V.F.; resources, P.P.,
V.D.G. and A.T.; data curation, P.P., M.F. and N.V.; writing—original draft preparation, P.P., M.F.
and V.F.; writing—review and editing P.P., M.F., V.D.G., N.V., V.F.; visualization, P.P., M.F., N.V.
and V.F.; funding acquisition, V.F. All authors have read and agreed to the published version of the
manuscript.
Funding: This study was carried out within the:

Appl. Sci. 2025, 15, 784 18 of 21
1. RETURN Extended Partnership and received funding from the European Union—
NextGenerationEU (National Recovery and Resilience Plan—NRRP, Mission 4, Compo-
nent 2, Investment 1.3- D.D. 1243 2/8/2022, PE0000005);
2. Progetto “GeoSciences: un’infrastruttura di ricerca per la Rete Italiana dei Servizi
Geologici—GeoSciences IR” (codice identificativo domanda: IR0000037); CUP:
I53C22000800006. Piano Nazionale di Ripresa e Resilienza, PNRR, Missione 4,
Componente 2, Investimento 3.1, “Fondo per la realizzazione di un sistema integrato di
infrastrutture di ricerca e innovazione” finanziato dall’Unione Europea—Next Generation
EU.
Data Availability Statement: Data of UniBa (from the seismic networks OSUB and OTRIONS) de-
scribed in Section 4, for the period preceding 2019, are available upon request.
Acknowledgments: Some figures were obtained by employing the GMT freeware package by Wes-
sel and Smith [64] and subsequent versions by Google Earth Pro, Google, Inc., California (accessed
on 15 April 2024). We kindly thank the four anonymous reviewers who improved the paper with
their useful comments.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1. Faccenna, C.; Becker, T.W. Topographic expressions of mantle dynamics in the Mediterranean. Earth. Sci. Rev. 2020, 209, 103327.
https://doi.org/10.1016/j.earscirev.2020.103327.
2. Handy, M.R.; Ustaszewski, K.; Kissling, E. Reconstructing the Alps-Carpathians-Dinarides as a key to understanding switches
in subduction polarity, slab gaps and surface motion. Int. J. Earth Sci. 2015, 104, 1–26. https://doi.org/10.1007/s00531-014-1060-3.
3. Serpelloni, E.; Anzidei, M.; Baldi, P.; Casula, G.; Galvani, A. Crustal velocity and strain-rate fields in Italy and surrounding
regions: New results from the analysis of permanent and non-permanent GPS networks. Geophys. J. Int. 2005, 161, 861–880.
https://doi.org/10.1111/j.1365-246X.2005.02618.x.
4. Le Breton, E.; Handy, M.R.; Molli, G.; Ustaszewski, K. Post-20 Ma motion of the Adriatic plate: New constraints from surround-
ing Orogens and implications for crust-mantle decoupling. Tectonics 2017, 36, 3135–3154. https://doi.org/10.1002/2016TC004443.
5. Meletti, C.; Patacca, E.; Scandone, P. Construction of a seismotectonic model: The case of Italy. Pure Appl. Geophys. 2000, 157,
11–35. https://doi.org/10.1007/PL00001089.
6. Ninivaggi, T.; Selvaggi, G.; Mazza, S.; Filippucci, M.; Tursi, F.; Czuba, W. Nature and origin of an undetected seismic phase in
waveforms from Southern Tyrrhenian (Italy) intermediate-depth and deep earthquakes: First evidence for the phase-A in the
subducted uppermost lithospheric mantle? Tectonophysics 2023, 860, 229919. https://doi.org/10.1016/j.tecto.2023.229919.
7. CPTI Working Group (ING; GNDT; SGA; SSN). Catalogo Parametrico dei Terremoti Italiani; Editrice Compositori: Bologna, Italy,
2004; p. 88. Available online: https://emidius.mi.ingv.it/CPTI/home.html (accessed on 15 April 2024).).
8. Del Gaudio, V.; Pierri, P.; Frepoli, A.; Calcagnile, G.; Venisti, N.; Cimini, G.B. A critical revision of the seismicity of Northern
Apulia (Adriatic microplate—Southern Italy) and implications for the identification of seismogenic structures. Tectonophysics
2007, 436, 9–35. https://doi.org/10.1016/j.tecto.2007.02.013.
9. Del Gaudio, V.; Pierri, P.; Calcagnile, G.; Venisti, N. Characteristics of the low energy seismicity of central Apulia (southern
Italy) and hazard implications. J. Seismol. 2005, 9, 39–59. https://doi.org/10.1007/s10950-005-1593-9.
10. Filippucci, M.; Miccolis, S.; Castagnozzi, A.; Cecere, G.; de Lorenzo, S.; Donvito, G.; Falco, L.; Michele, M.; Nicotri, S.; Romeo,
A.; et al. Seismicity of the Gargano promontory (Southern Italy) after 7 years of local seismic network operation: Data release of
waveforms from 2013 to 2018. Data Brief 2021, 35, 106783. https://doi.org/10.1016/j.dib.2021.106783.
11. Festa, V.; De Giosa, F.; Del Gaudio, V.; Moretti, M.; Pierri, P. In search of the seismogenic source of the March 23rd 2018 earth-
quake (Mw 3.7) near Brindisi (Puglia, Southern Italy). Geol. Croat. 2019, 72, 137–144. https://doi.org/10.4154/gc.2019.10.
12. Mariucci, M.T.; Montone, P. IPSI 1.6, Database of Italian Present-Day Stress Indicators; Istituto Nazionale di Geofisica e
Vulcanologia (INGV): Roma, Italy, 2024. https://doi.org/10.13127/IPSI.1.6.
13. DISS Working Group. Database of Individual Seismogenic Sources (DISS), Version 3.3.0: A Compilation of Potential Sources for Earth-
quakes Larger than M 5.5 in Italy and Surrounding Areas; Istituto Nazionale di Geofisica e Vulcanologia (INGV): Roma, Italy, 2021.
https://doi.org/10.13127/diss3.3.0.

Appl. Sci. 2025, 15, 784 19 of 21
14. ITHACA Working Group. ITHACA (ITaly HAzard from CApable Faulting), a Database of Active Capable Faults of the Italian Territory;
Version December 2019; ISPRA Geological Survey of Italy: Roma, Italy, 2019. Available online https://sgi.isprambiente.it/itha-
caweb/Default.aspx# (accessed on 15 April 2024).
15. de Alteriis, G.; Aiello, G. Stratigraphy and tectonics offshore of Puglia (Italy, southern Adriatic Sea). Mar. Geol. 1993, 113, 233–
253. https://doi.org/10.1016/0025-3227(93)90020-V.
16. Bertotti, G.; Picotti, V.; Chilovi, C.; Fantoni, R.; Merlini, S.; Mosconi, A. Neogene to quaternary sedimentary basins in the south
adriatic (central mediterranean): Foredeeps and lithospheric buckling. Tectonics 2001, 20, 771–787.
https://doi.org/10.1029/2001TC900012.
17. Fantoni, R.; Franciosi, R. Tectono-sedimentary setting of Po Plain and Adriatic foreland. Rend. Lincei-Sci. Fis. 2010, 21, 197–209.
https://doi.org/10.1007/s12210-010-0102-4.
18. Cicala, M.; Festa, V.; Sabato, L.; Tropeano, M.; Doglioni, C. Interference between Apennines and Hellenides foreland basins
around the Apulian swell (Italy and Greece). Mar. Pet. Geol. 2021, 133, 105300. https://doi.org/10.1016/j.marpetgeo.2021.105300.
19. Ricchetti, G.; Ciaranfi, N.; Luperto Sinni, E.; Mongelli, F.; Pieri, P. Geodinamica ed evoluzione sedimentaria e tettonica
dell’Avampaese Apulo. Mem. Soc. Geol. Ital. 1988, 41, 57–82.
20. Doglioni, C.; Mongelli, F.; Pieri, P. The Puglia uplift (SE Italy): An anomaly in the foreland of the Apenninic subduction due to
buckling of a thick continental lithosphere. Tectonics 1994, 13, 1309–1321. https://doi.org/10.1029/94TC01501.
21. Selli, R. Il Paleogene nel quadro della geologia dell’Italia meridionale. Mem. Soc. Geol. Ital. 1962, 3, 737–790.
22. Ciaranfi, N.; Pieri, P.; Ricchetti, G. Note alla Carta Geologica delle Murge e del Salento (Puglia centromeridionale). Mem. Soc.
Geol. Ital. 1988, 41, 449–460.
23. Spalluto, L.; Pieri, P.; Ricchetti, G. Le facies carbonatiche di piattaforma interna del Promontorio del Gargano: Implicazioni
paleoambientali e correlazioni con la coeva successione delle Murge. Boll. Soc. Geol. Ital. 2005, 124, 675–690.
24. Borgomano, J.R.F. The Upper Cretaceous carbonates of the Gargano-Murge region, southern Italy: A model of platform-to-
basin transition. AAPG Bull. 2000, 84, 1561–1588. https://doi.org/10.1306/8626BF01-173B-11D7-8645000102C1865D.
25. Nicolai, C.; Gambini, R. Structural architecture of the Adria Platform-and-Basin System. Boll. Soc. Geol. Ital. 2007, 7, 21–37.
26. de’ Dominicis, A.; Mazzoldi, G. Interpretazione geologico-strutturale del margine orientale della Piattaforma Apula. Mem. Soc.
Geol. Ital. 1987, 38, 163–176.
27. Mastrogiacomo, G.; Moretti, M.; Owen, G.; Spalluto, L. Tectonic triggering of slump sheets in the Upper Cretaceous carbonate
succession of the Porto Selvaggio area (Salento peninsula, southern Italy): Synsedimentary tectonics in the Apulian Carbonate
Platform. Sediment. Geol. 2012, 269–270, 15–27. https://doi.org/10.1016/j.sedgeo.2012.05.001.
28. Cicala, M.; De Giosa, F.; Festa, V.; Lisco, S.; Moretti, M. The northern fault of the onshore-offshore Monte Giove relief in the
southern Adriatic Sea, Italy: Implications for tectonic reactivation in the Apulian Foreland. Geol. Q. 2023, 67, 11.
29. Martinis, B. Sulla tettonica delle Murge Nord-Occidentali. Acc. Naz. Lincei Rend. Sc. Fis. Mat. Nat. 1961, 31, 299–305.
30. Iannone, A.; Pieri, P. Caratteri neotettonici delle Murge. Geol. Appl. Idrogeol. 1982, 17, 147–159.
31. Tropeano, M.; Pieri, P.; Moretti, M.; Festa, V.; Calcagnile, G.; Del Gaudio, V.; Pierri, P. Quaternary tectonics and seismotectonic
features of the Murge area (Apulian foreland, SE Italy). Il Quat. 1997, 10, 543–548.
32. Festa, V. Cretaceous structural features of the Murge area (Apulian Foreland, Southern Italy). Eclogae Geol. Helv. 2003, 96, 11–
22. https://doi.org/10.5169/seals-169003.
33. de Alteriis, G. Different foreland basins in Italy: Examples from the central and southern Adriatic Sea. Tectonophysics 1995, 252,
349–373. https://doi.org/10.1016/0040-1951(95)00155-7.
34. Pieri, P.; Laviano, A. Tettonica e sedimentazione nei depositi senoniani delle Murge sud-orientali (Ostuni). Boll. Soc. Geol. Ital.
1989, 108, 351–356.
35. Del Gaudio, V.; Festa, V.; Ripa, R.R.; Iurilli, V.; Pierri, P.; Calcagnile, G.; Moretti, M.; Pieri, P.; Tropeano, M. Evidence of Apulian
crustal structures related to low energy seismicity (Murge—Southern Italy). Ann. Geophys. 2001, 44, 1049–1066.
https://doi.org/10.4401/ag-3555.
36. Bath, M.; Duda, S.J. Earthquake volume, fault plane area, seismic energy, strain, deformation and related quantities. Ann. Geofis.
1964, 17, 353–368.
37. Rovida, A.; Locati, M.; Camassi, R.; Lolli, B.; Gasperini, P. The Italian earthquake catalogue CPTI15. Bull. Earthq. Eng. 2020, 18,
2953–2984. https://doi.org/10.1007/s10518-020-00818-y.
38. Rovida, A.; Locati, M.; Camassi, R.; Lolli, B.; Gasperini, P.; Antonucci, A. Catalogo Parametrico dei Terremoti Italiani (CPTI15),
Versione 4.0 [Data Set]; Istituto Nazionale di Geofisica e Vulcanologia (INGV): Roma, Italy, 2022.
https://doi.org/10.13127/cpti/cpti15.4.

Appl. Sci. 2025, 15, 784 20 of 21
39. Guidoboni, E.; Ferrari, G.; Mariotti, D.; Comastri, A.; Tarabusi, G.; Sgattoni, G.; Valensise, G. CFTI5Med, Catalogo dei Forti
Terremoti in Italia (461 a.C.-1997) e Nell’area Mediterranea (760 a.C.-1500); Istituto Nazionale di Geofisica e Vulcanologia (INGV):
Roma, Italy, 2018. https://doi.org/10.6092/ingv.it-cfti5.
40. Guidoboni, E.; Ferrari, G.; Tarabusi, G.; Sgattoni, G.; Comastri, A.; Mariotti, D.; Ciuccarelli, C.; Bianchi, M.G.; Valensise, G.
CFTI5Med, the new release of the catalogue of strong earthquakes in Italy and in the Mediterranean area. Sci. Data 2019, 6, 80.
https://doi.org/10.1038/s41597-019-0091-9.
41. Postpischl, D. Catalogo dei Terremoti Italiani Dall’anno 1000 al 1980; C.N.R.—Progetto Finalizzato Geodinamica—Quaderni della
Ricerca Scientifica: Bologna, Italy, 1985; Volume 114 2B, 239p.
42. Castello, B.; Di Stefano, R. Catalogo della Sismicità Italiana (CSI): Fasi (Version 2) [Data Set]; Istituto Nazionale di Geofisica e
Vulcanologia (INGV): Roma, Italy, 2022. https://doi.org/10.13127/csi.2.0/phs.
43. Locati, M.; Camassi, R.; Rovida, A.; Ercolani, E.; Bernardini, F.; Castelli, V.; Caracciolo, C.H.; Tertulliani, A.; Rossi, A.; Azzaro,
R.; et al. Database Macrosismico Italiano (DBMI15), Versione 4.0 [Data Set]; Istituto Nazionale di Geofisica e Vulcanologia (INGV):
Roma, Italy, 2022. https://doi.org/10.13127/dbmi/dbmi15.4.
44. Lahr, J.C. HYPOELLIPSE/Version 2.0: A Computer Program for Determining Local Earthquakes Hypocentral Parameters, Magnitude,
and First-Motion Pattern; Open-File Rep. n. 89-116; U.S. Geological Survey: Reston, VA, USA, 1989; 95p.
https://doi.org/10.3133/ofr89116.
45. Calcagnile, G.; Panza, G.F. The main characteristics of the lithosphere-asthenosphere system in Italy and surroundings regions.
Pure Appl. Geophys. 1980, 119, 865–879. https://doi.org/10.1007/BF01131263.
46. Costa, G.; Panza, G.F.; Suhadolc, P.; Vaccari, F. Zoning of the Italian territory in terms of expected peak ground acceleration
derived from complete synthetic seismograms. J. Appl. Geophys. 1993, 30, 149–160.
47. Venisti, N.; Calcagnile, G.; Pontevivo, A. ; Panza, G.F.; Tomographic Study of the Adriatic Plate. Pure Appl. Geophys. 2005, 162,
311–329. https://doi.org/10.1007/s00024-004-2602-6.
48. Pierri, P.; de Lorenzo, S.; Calcagnile, G. Analysis of the low-energy seismic activity in the Southern Apulia (Italy). Open J. Earthq.
Res. 2013, 2, 91–105. https://doi.org/10.4236/ojer.2013.24010.
49. Reasenberg, P.; Oppenheimer, D. FPFIT, FPPLOT and FPPAGE: FORTRAN Computer Programs for Calculating and Displaying
Earthquake Fault Plane Solutions; Open File Rep. n. 85-739; U.S. Geological Survey: Reston, VA, USA, 1985; 109p.
https://doi.org/10.3133/ofr85739.
50. Frepoli, A.; Cimini, G.B.; De Gori, P.; De Luca, G.; Marchetti, A.; Monna, S.; Montuori, C.; Pagliuca, N.M. Seismic sequences and
swarms in the Latium-Abruzzo-Molise Apennines (central Italy): New observations and analysis from a dense monitoring of
the recent activity. Tectonophysics 2017, 712–713, 312–329. https://doi.org/10.1016/j.tecto.2017.05.026.
51. Scognamiglio, L.; Tinti, E.; Michelini, A. Real-time determination of seismic moment tensor for Italian region. Bull. Seism. Soc.
Am. 2009, 99, 2223–2242. https://doi.org/10.1785/0120080104.
52. Gephart, J.W. FMSI: A Fortran program for inverting fault/slickenside and earthquake focal mechanism data to obtain the re-
gional stress tensor. Comput. Geosci. 1990, 16, 953–989.
53. Gephart, J.W.; Forsyth, D.W. An improved method for determining the regional stress tensor using earthquake focal mechanism
data: Application to the San Fernando earthquake sequence. J. Geophys. Res. 1984, 89, 9305–9320.
54. Lu, Z.; Wyss, M.; Pulpan, H. Details of stress directions in the Alaska subduction zone from fault plane solutions. J. Geophys.
Res. 1997, 102, 5385–5402. https://doi.org/10.1029/96JB03666.
55. Heidbach, O.; Rajabi, M.; Reiter, K.; Ziegler, M. World Stress Map 2016 [Dataset]; GFZ Data Services: Potsdam, Germany, 2016.
https://doi.org/10.5880/WSM.2016.002.
56. Gambini, R.; Tozzi, M. Tertiary geodynamic evolution of Southern Adria Microplate. Terra Nova 1996, 8, 593–602.
https://doi.org/10.1111/j.1365-3121.1996.tb00789.x.
57. Meletti, C.; Valensise, G. Zonazione Sismogenetica ZS9—App. 2 al Rapporto Conclusivo. Gruppo di Lavoro per la Redazione della Mappa
di Pericolosità Sismica Prevista Dall’ordinanza PCM 3274 del 20 marzo 2003; Istituto Nazionale di Geofisica e Vulcanologia: Roma,
Italy, 2004; 38p.
58. Di Bucci, D.; Caputo, R.; Mastronuzzi, G.; Fracassi, U.; Selleri, G.; Sansò, P. Quantitative analysis of extensional joints in the
southern Adriatic foreland (Italy), and the active tectonics of the Apulia region. J. Geodyn. 2011, 51, 141–155.
https://doi.org/10.1016/j.jog.2010.01.012.
59. Panza, G.F.; Bela, J. NDSHA: A new paradigm for reliable seismic hazard assessment. Eng. Geol. 2020, 275, 105403.
https://doi.org/10.1016/j.enggeo.2019.105403.

Appl. Sci. 2025, 15, 784 21 of 21
60. Rugarli, P.; Vaccari, F.; Panza, G.F. Seismogenic nodes as a viable alternative to seismogenic zones and observed seismicity for
the definition of seismic hazard at regional scale. Vietnam. J. Earth Sci. 2019, 41, 289–304. https://doi.org/10.15625/0866-
7187/41/4/14233.
61. Wen, Z.P.; Wang, G.X. Earthquakes and Sustainable Infrastructure-Neodeterministic (NDSHA) Approach Guarantees Preven-
tion Rather Than Cure. Earthq. Sci. 2024, 37, 493–496. https://doi.org/10.1016/j.eqs.2024.06.010.
62. Leonard, M. Earthquake Fault Scaling: Self-Consistent Relating of Rupture Length, Width, Average Displacement, and Moment
Release. Bull. Seism. Soc. Am. 2010, 100, 1971–1988. https://doi.org/10.1785/0120090189.
63. Trippetta, F.; Petricca, P.; Billi, A.; Collettini, C.; Cuffaro, M.; Lombardi, A.M.; Scrocca, D.; Ventura, G.; Morgante, A.; Doglioni,
C. From mapped faults to fault-length earthquake magnitude (FLEM): A test on Italy with methodological implications. Solid
Earth 2019, 10, 1555–1579. https://doi.org/10.5194/se-10-1555-2019.
64. Wessel, P.; Smith, W.H.F. New, Improved Version of Generic Mapping Tools Released. Eos Trans. Am. Geophys. Union 1998, 79,
579.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual au-
thor(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.