Imp.Milella&


SEA LEVEL DURING 4TH – 2ND CENTURY B.P. IN EGNAZIA (ITALY)
FROM ARCHAEOLOGICAL AND HYDROGEOLOGICAL DATA 

Maurilio Milella 1, Cosimo Pignatelli 1, Mario Donnaloia 2 & Giuseppe Mastronuzzi 3
1 PhD School of Geomorphology and Environmental Dynamic, University of Bari

2 Sismological Observatory, University of Bari
3 Department of Geology and Geophisics, University of Bari

ABSTRACT: Milella M., Pignatelli C., Donnaloia M. & Mastronuzzi G., Sea Level During 4th–2nd Century B.P. In Egnazia (Italy) From
Archaeological And Hydrogeological Data (IT ISSN 0394-3356, 2007).
This paper aims to estimate the mean sea level position during the 4th–2nd cent. B.C. along the Adriatic coast of Apulia using
archaeological and geomorphological data from the archaeological site of Egnazia. The detailed topographic survey of wells, cisterns
and tombs reported from Giornali di Scavo permitted us to reconstruct the probable groundwater piezometric surface in that period.
Two different scenarios have been considered. The first one considers the active state of wells: we added the value of 35 cm to the
elevation of the well–bottom required in order to take water by means of jars. Furthermore, we subtracted the same value to the eleva-
tion of the tomb / cistern–bottom because these structures couldn’t come into contact with groundwater. The second one considers
the non–activity of wells: the dating of the wells, based on their replenishment, performed on ceramic fragments, indicates the time
they were abandoned and the consequent transformation into dump. In this way only the values of well–bottoms for reconstructing
the piezometric surface trend are used. The piezometric surface trend corresponding to the two hypothesis was reconstructed with
ArcMap 8.3 GIS software through Kriging interpolating method. Sea level during 4th–2nd cent. B.C. stands at least -1,2 m present sea
level. 

RIASSUNTO: Milella M., Pignatelli C., Donnaloia M. & Mastronuzzi G., Livello del Mare nel IV–II Secolo a.C. da Dati Archeologici e
Idrogeologici nell’area di Egnazia (Italia) (IT ISSN 0394-3356, 2007).
sec. a.C. lungo la costa adriatica della Puglia nei dintorni della città di Egnazia sulla base della ricostruzione del probabile andamento
della superficie piezometrica della falda idrica di quel periodo. Per questo studio sono stati utilizzati: i – rilievo topografico di dettaglio
dell’area di Egnazia; ii – la profondità dei pozzi e delle cisterne desunte dai Giornali di Scavo. E’ stato ricostruito il possibile andamento
della superficie piezometrica della falda idrica dell’epoca in relazione a due differenti ipotesi di lavoro. La prima considera lo stato di
esercizio dei pozzi: alle quote dei fondo–pozzi è stato aggiunto il valore di 35 cm, necessario a garantire il prelievo dell’acqua tramite
vasi, mentre alle quote del fondo–tombe e cisterne è stato sottratto lo stesso valore di 35 cm, supponendo che tali strutture non
dovessero essere a contatto con acque di falda. La seconda ipotesi presuppone l’interruzione dell’esercizio dei pozzi: infatti la datazio-
ne dei pozzi basata sul loro riempimento, costituito soprattutto da frammenti ceramici, indica il momento del loro abbandono e la con-
seguente trasformazione in “discariche”. In tal modo, per la ricostruzione dell’andamento della superficie della falda sono stati conside-
rati esclusivamente i valori dei fondo–pozzi. Attraverso il software ArcMap 8.3 GIS è stato ricostruito, mediante il metodo di interpola-
zione del Kriging, l’andamento delle superfici piezometriche corrispondenti alle due ipotesi. Si è potuto così ipotizzare che la posizione
del livello del mare nel IV–II secolo a.C. fosse almeno 1,2 m al di sotto della sua posizione attuale.

Keywords: Sea level, Holocene, Archaeology, Hydrogeology, Puglia, Italy.

Parole chiave: Livello del mare, Olocene, Archeologia, Idrogeologia, Puglia, Italia

Il Quaternario
Italian Journal of Quaternary Sciences
19(2), 2006 - 253-260

INTRODUCTION

In the Mediterranean basin, indications of sea
level stands during historical times can be obtained
from geomorphological, biological and archaeological
sources. Particularly, in microtidal areas geomorpholo-
gical data provide a valid tool to indicate the biological
mean sea level within a few centimetres (PASKOFF &
SA N L A V I L L E, 1983; DA L O N G E V I L L E, 1987; LA B O R E L &
LABOREL DEGUEN, 1994; PIRAZZOLI, 1996) but it is difficult
to date them because of the frequent absence of bio-
genic encrustations useful for radiometric analyses
(AURIEMMA et al., 2004).

Archaeological sites are a source of valuable data
for reconstructing sea level changes. Here radiometric
data can be compared and confirmed by the archaeo-
logical chronology (EROL & PIRAZZOLI, 1992; STIROS &
PIRAZZOLI, 1995; ANTONIOLI & LEONI, 1998; FLEMING &

WEBB, 1986; MOURHANGE et al., 2001, SIVAN et al., 2001;
AURIEMMA et al., 2005).

In the past, along the Apulia coast there was an
extensive colonisation due to humans especially in the
areas characterised by stability of the landscape and
the availability of fresh water. In particular, in the
archaeological site of Egnazia, located along the
Adriatic coast of Apulia (Fig. 1), the more ancient
human settlement goes back to Bronze age (16th cent.
B.C.). Its coast location, like several Apulian protostoric
settlements, made maritime activities easier because
the jagged coastal area is suitable for docking and
sheltering vessels which carried wares from transadria-
tic areas (CREMONESI, 1979; SCIARRA BARDARO–ANDREASSI,
1982; AURIEMMA et al., 2005). Recent studies permit us
to guess the Egnazia settlement originated in the
Acropoli: an antropic–made low hill which spreads
seaward between two small inlets, reached out into the



254 M. Milella et al.

hinterland occupying the area where the later Roman
age houses were found (DELL’AGLIO, 1982). The Egnazia
economy was based mainly on ship commerce
between the two sides of the Adriatic Sea: in fact, the
town was an important merchant harbour and the
wares were transferred by a track road network to the
inner centres of Apulia and Lucania by the way of the
Traiana Road. So Egnazia represented an important
crossroad and meeting point of the several aged cultu-
res (DONVITO, 1988).

It is possible to divide the urban space into three
bands quite parallel to coastline (Diceglie, 1981a, tav.
I–V): the first one, the closest to coastal area, is related
to maritime spaces and comprise caves, some messa-
pica tombs and the ancient harbour; the second one, is
located between the first and the Traiana Road, inclu-
ding the Acropoli and public, religious and administrati-
ve buildings; the last one, situated to the south of the
Traiana Road, including all civil buildings. Beyond this
third band, about 600 m inward from Traiana Road,
there is the Western Necropoli area, used from the half
of 4th cent. B.C. up to 4th cent. A.D. (Fig. 1).

The guidelines at the base of this work was origi-
nated by the assumption of the existence of sea level
and land indicators (A U R I E M M A et al ., 2004).
Archaeological sea level indicators, such as harbour
structures, fishery and so on, permit us to recognise
with good approximation past sea level stands
(ANTONIOLI & LEONI, 1998). Land indicators such as
tombs, cisterns etc. at present below sea level indicate
a minimum altitude under which sea level stood. The
investigated hypothesis is that wells, cisterns and
tombs in the coastal archaeological area of Egnatia are
useful to reconstruct the piezometric surface and to put
in relation the sea level with the functioning of the wells.
Thus, the aims of this paper is to verify the availability
of this method along rocky coasts and to define the
error limit of the sea level identifications.

REGIONAL SETTINGS

The Apulia region represents the emerged part of
the foreland domain of both the Apenninic and Dinaric
orogens. The Apulian foreland is slightly deformed and
is affected by NW–SE and NE–SW trending faults that
subdivided it into five structural domains: Gargano
area, Tavoliere basin, Murge upland, Taranto–Brindisi
plain and Salento peninsula. The uplift rate of the
southernmost Adriatic part of Apulia, based on generic
stratigraphic and morphological evidence, ranges
between 0,15 and 0,19 mm/yr in the area around Bari
and about -0,02 mm/yr along the coastal area near
Brindisi (BORDONI & VALENSISE, 1998; FERRANTI et al.,
2006).

The archaeological area of Egnazia is located on
the Adriatic coast of Apulia between Monopoli and
Fasano. Here, at the feet of the Murge Plateau (locally
about 400 m high) the coastal landscape is characteri-
zed by a sequence of sub–horizontal surfaces sloping
towards sea from an elevation of about 120 m. These
surfaces should be produced by marine abrasion pha-
ses which occurred since the Middle Pleistocene as a
consequence of the superimposition of regional uplift
and the glacio–eustatic sea level changes. The mono-
tony of this landscape is broken by a relict drainage
network characterized by deep valleys, locally named
gravine or lame, generally parallel to each other and
perpendicular to coastline, that were shaped for sap-
ping processes (MASTRONUZZI & SANSÒ, 2002a). The gen-
tly sloping rocky coast is marked by the presence of lit-
tle inlets which guest pocket beaches; the present sea
level is shaping a poorly developed wave cut platform
locally alternated to a discontinuos notch.

The stratigraphy of the Egnazia area is characteri-
zed at the base by the presence of the Calcare di Bari
unit (Cretaceous), recognizable by drilling only, that
represents the local basement, marked by karstic pro-

Fig. 1 - Egnazia town: the different urban spaces are evidenced with bold lines (from CINQUEPALMI & COCCHIARO, 2000, mod., 4, fig. 3).

Città di Egnazia: in rilievo sono evidenziati i differenti spazi urbani (da CINQUEPALMI & COCCHIARO, 2000, modificata, 4, fig.3).



cesses. Biocalcareous sandstone related to the
Calcarenite di Gravina unit (Lower Pleistocene)
(D’ALESSANDRO & IANNONE, 1984) overlies in transgres-
sion the limestone and outcrops extensively along the
cost. Therefore, this lithological sequence is characteri-
zed by a high permeability, due to fractures and karst. It
causes the lack of a fluvial network due to the rapid
rainwater infiltration: this feeds a wide deep aquifer that
rests on seawater intruded from the nearby coastal
area and its interface follows Ghyben–Herzberg princi-
ple. At present, in the inner part of Murge plateau the
piezometric surface reaches a maximum altitude of
about +200 m m.s.l. sloping about 0,2÷0,8 % towards
the coastline (COTECCHIA, 1977).

BIBLIOGRAPHIC AND FIELD DATA PROCESSING

To reconstruct the piezometric surface and the
probable mean sea level during the 4th–2nd cent. B.C.
along the Egnazia coastal area the following data were
used: i – archaeological and hydrogeological biblio-
graphic data; ii – new detailed topographic survey of
the entire Egnazia area and the depth of wells and
cisterns (Fig. 2) inferred by Digging Journals (Giornali di
Scavo).

The piezometric level of the Egnazia coastal area
was reconstructed by means of bibliographic data
(TANZARELLA, 1998; MAGGIORE et al., 1999; MAGGIORE &
PAGLIARULO, 2004).

In 1997, the piezometric level in the
Archaeological Park area was about +0,12 m present
b.s.l. whereas it was at +2,5 m present b.s.l. about
5200 m landward (TANZARELLA, 1998). Moreover, the
bibliographic check allowed us to obtain data on the
typology and the depth of the archaeological structu-
res, wells and cisterns, used as indicators for this
study. Furthermore, the dating of the same indicators
carried out by means of ceramic replenishment were
taken on. Wells and cisterns have a good distribution in
the entire area of the ancient Egnazia, inside the
Archaeological Park area and outside, along the coast
and inside the western Necropoli. A well represents a
vertical hole dug by humans (Fig. 3) with a depth of
about 30 ÷ 40 cm below piezometric surface useful for
collecting water by jars (Sivan et al., 2001). Instead,
cisterns were used only to keep rainwater and
groundwater withdrawn through wells: so their bottoms
did not reach the piezometric surface (COCCHIARO, 1982;
TAGLIENTE, 2005).

During archaeological surveys, 70 structures for
water supplies were localized (Fig. 4): 52 inside the pre-
sent Archaeological Park area and 18 along coastal
area. Unluckily, only 10 structures out of the first group
and 8 structures out of the second one, in particular
wells and cisterns to the south of the Acropoli, show
enough data from Digging Journals about their depth.
In particular, wells located inside the Archaeological
Park area surrounding Traiana Road, have a mean
depth of about 3 m; while the ones situated near coa-
stline have a mean depth of about 1,8 m. Among the 10
structures of Archaeological Park 1 dates back to 9th
cent. B.C., 4 to 4th cent. B.C., 2 to 3rd cent. B.C., 2 to
2nd cent. B.C. and 1 to 1st cent. A.D.; all the 8 structu-
res of coastal area to the south of Acropoli date back to

255Sea level during 4th - 2nd century B.P. ...

Fig. 2 - Detail of a cistern entrance in the Archaeological Park
area (CINQUEPALMI & COCCHIARO, 2000, 16, fig. 37).

Particolare dell’imboccatura di una cisterna nell’area del Parco
Archeologico (da CINQUEPALMI & COCCHIARO, 2000, 16, fig. 37).

Fig. 3 - Well Scheme: in the vertical sections 4 cavity for each
side, named “pedarole”, can be distinguished. In the lower
part a vertical sub-rectangular cavity opens (from Digging
Journals).
Schema di un Pozzo: nelle sezioni verticali si possono distin-
guere 4 cavità per ciascun lato, chiamate “pedarole”. Nella
parte più bassa si apre una cavità subrettangolare (dai Giornali
di Scavo).



256

2nd cent. B.C. (TAGLIENTE,
2005). Only the structures
dating back to 4th–2nd
cent. B.C. have been consi-
dered for this study.

Detailed topographic
survey was carried out by
total station (Teodolite
“WILD T2” with a
Geodimeter “AGA 14A”); it
let us get spatial coordina-
tes related to Gauss Boaga
system datum ED 50 (X, Y
and Z) of all the wells,
cisterns and tombs inside
Egnazia area and depth of
all the tombs inside western
Necropoli. Z–elevations
were related to the present
biological sea level (= b.s.l.)
(LABOREL & LABOREL DEGUEN,
1994) thanks to the presen-
ce of living concretions of
Lithophyllum lichenoides
and Dendropoma sp. along
the coastal inlets near the
Acropoli; this occurrence
allows us to evaluate in +/-
0,10 m the error bar of
shown data. So, in the area
of the Archaeological Park
near Traiana Road the mean
elevation of well heads is
about +4,8 +/- 0,10 m pre-
sent b.s.l., while along the
coastline the mean eleva-
tion of well heads does not
go over +1,5 +/- 0,10 m
present b.s.l..

The spatial coordina-
tes of all the surveyed
points (wells, cisterns and
tombs) elaborated using
software (Surfer 8 and
ArcMap 8.3 GIS – Kriging
method) permit us to draw a
map where topography and
the piezometric surface of
Egnazia area are shown
(Figg. 4, 5). The piezometric
surface was reconstructed
according to two hypothesi-
zed scenario.

The A scenario consi-
ders the active state of
wells: we add the value of
35 cm to the elevation of
the well–bottom required to
take water by means of jars
(SIVAN et al., 2001); on the
contrary, we subtracted the
same value to the elevation
of the tomb and cistern bot-
tom because these structu-
res could not come into

Fig. 4 - Topography restitution of Egnazia town, by means of Surfer 8 software, with contour lines.
ox : Archaeological Points; __: Contour Lines; _: Coastline; AB: Hydrogeological section track.
Ricostruzione, ottenuta con il software Surfer 8, dell’andamento plano–altimetrico dell’area di
Egnazia mediante curve di livello.
ox : Punti Archeologici; __: Curve di Livello; _: Linea di Costa; AB: Traccia della Sezione
Idrogeologica.

Fig. 5 - Piezometric surface restitution of Egnazia town, by means of ArcMap 8.3 GIS software, for
B scenario.
ox : Archaeological Points; __: Contour Lines; _: Coastline; AB: Hydrogeological section track.
Ricostruzione, ottenuta con il software ArcMap 8.3 GIS, dell’andamento della superficie piezometri-
ca nell’area di Egnazia nel caso dello scenario B.
ox : Punti Archeologici; __: Curve di Livello; _: Linea di Costa; AB: Traccia della Sezione
Idrogeologica.

M. Milella et al.



257

contact with groundwater. The use of this value allows
us to obtain the best extrapolation of the piezometric
surface trend so that its slope is coherent with the one
of the uptodate groundwater.

The B scenario consi-
ders the non–active state of
wells: the dating of wells
based on their replenish-
ment, obtained by means of
ceramic fragments, indica-
tes the time they were
abandoned and consequent
transformation into dump.
In this way only the values
of well bottoms for recon-
structing the piezometric
surface trend were used
(Fig. 5).

According to these
hypothesis, either real
hydrogeological sections or
theoretical best–fitted on 2°
degree equations extrapola-
tions, were obtained (Figg.
6a,b).

DISCUSSION

AURIEMMA et al. (2004;
2005) reconstructed a curve
of relative sea level changes
in ancient time using geo-
morphological, archaeologi-
cal and radiometric data. In
particular, in the inlet north
to the Acropoli submerged
piers in opus caementicium
of the Roman age are pre-
sent. These structures and
the building techniques
indicate that sea level
stands at about -2,5 m pre-
sent b.s.l. during the 1st
cent. B.C..

Hydrogeological sec-
tion, “A–B”, made up for
both the work scenario rea-
ches out about 800 m from
the south limit of western
Necropoli to the present
coastline in front of Acropoli
(Fig. 4). Some uncertainties
derive from the historical
and bibliographical sources:
depth of wells and cisterns,
taken from Digging
Journals, have not been
ascertained because of
their replenishment and / or
covering protection fol-
lowing the archaeological
digging surveys.

In particular, in the
Archaeological Park, diffe-

rences of depth have been recognized:
- the bottom of one cistern is about 0,5 m deeper than

the bottom of very close coeval wells;
- bottoms of wells of coeval age, up to 30 m close,

Fig. 6a - Real hydrogeological section and theoretical best – fitted on 2° degree equation extrapola-
tion for A scenario. The rectangle represents the present Archaeological Park area. 1 = Past
Piezometric Level; 2 = Present Piezometric Level; 3 = Past Extrapolation Level based on “Y” 2°
degree equation.

Sezione idrogeologica reale e relativa estrapolazione teorica ottenuta mediante equazione di 2°
grado, nel caso dello scenario A. Il rettangolo corrisponde all’area del Parco Archeologico. 1 =
Livello Piezometrico passato; 2 = Livello Piezometrico presente; 3 = Estrapolazione del Livello
Piezometrico passato sulla base dell’equazione di 2° grado “Y”.

Fig. 6b - Real hydrogeological section and theoretical best – fitted on 2° degree equation extrapo-
lation for B scenario. The rectangle represents the present Archaeological Park area. 1 = Past
Piezometric Level; 2 = Present Piezometric Level; 3 = Past Extrapolation Level based on “Y” 2°
degree equation.

Sezione idrogeologica reale e relativa estrapolazione teorica ottenuta mediante equazione di 2°
grado, nel caso dello scenario B. Il rettangolo corrisponde all’area del Parco Archeologico. 1 =
Livello Piezometrico passato; 2 = Livello Piezometrico presente; 3 = Estrapolazione del Livello
Piezometrico passato sulla base dell’equazione di 2° grado “Y”.

Sea level during 4th - 2nd century B.P. ...



258

show differences of depth of about 0,8÷1,0 m.
To overcome this discrepancy the extrapolated

piezometric level, for both scenarios, has been consi-
dered (curve 3 in Figg. 6a,b). The scenario A (active
state of wells) suggest a piezometric level between
about +1,15 m present b.s.l. along Traiana Road and
about -0,8 m present b.s.l. on coastline. The scenario B
(non–active state of wells) indicate a piezometric level
between +0,75 m present b.s.l. in the Archaeological
Park and -1,2 m present b.s.l. along the coast (Tab. 1).

Assuming the position of sea-level stand at -0,8
or at –1,2 m during 4th–2nd cent. B.C. in both scenario
the piezometric level of +1,95 m past b.s.l. along
Traiana Road results. This value is too high and incon-
gruous for the topographic and hydrogeologic features
of Egnazia area (Tab.1). In fact the Archaeological Park
is located only 250 m landward and in the last 30 years
the piezometric level went down about 15% due to
human activity (POLEMICO & DRAGONE, 1999). Since the
climatic state during 4th–2nd cent. B.C., it was not very
different in comparison with the present time
(MASTRONUZZI & SANSÒ, 2002b) the past piezometric
level could not be much different in comparison with
the present trend too.

The present piezometric surface in the Egnazia
area has a mean slope of 0,048%: it is quite steady up
to 5200 m landward. In the curve 3 of figg. 6a,b it is
possible to detect two sketches characterized by diffe-
rent slope. Sketch 1, from western Necropoli to Traiana

Road, has a mean slope of about 0,076% and 0,073%,
respectively for A and B scenario. Sketch 2, from
Traiana Road to coastline, has a mean slope of about
0,77% and 0,76%, respectively for A and B scenario.
Only the values obtained for the Sketch 1 can be com-
pared with the present piezometric mean slope. Using
the mean slope of Sketch 1 the values of +0,19 m and
+0,18 m, respectively for A and B scenario, result:
these values should be the piezometric level in the
Archaeological Park referred to every stand of past sea
level. The piezometric levels calculated and the present
one of +0,12 m b.s.l. (TANZARELLA, 1998) can be brought
into comparison. Thereby, the piezometric surface
during 4th–2nd cent. B.C. along Traiana Road had to
stand necessarily about 1,6 m below the wells–bottom
(Tab. 2).

The Archaeological Park is only 250 m far from
the coastline: according to the present hydrogeological
features the piezometric surface near Traiana Road did
not have to overcome the elevation of about +0,2 m
past sea level. Mean elevation of the wells–bottom is
about 1,6 m above this piezometric surface.
Consequently, the wells of this area could not operate
during 4th–2nd cent. B.C., according to the B scenario.
Thus, the reconstruction of the piezometric surface
allowed us to estimate that sea level during 4th–2nd
cent. B.C. had to stand at least -1,2 m present b.s.l.
(Fig. 7).

In any case, the available data and the results of

Archaeological Park area Coastline

Past Piezometric Present Past Piezometric 
Level related to

Past Piezometric Surface
Piezometric Surface related to

Cisterns Related to Related to
Surface Present b.s.l.

Wells / Tombs Past Sea Present
related to = Past Sea Level 

Bottom Bottom Level b.s.l.
Present b.s.l. Position (+/- 0,10 m)

A Hypothesis:
Activity

+0,35 m -0,35 m +1,95 m +1,15 m +0,12 m -0,8 m

B Hypothesis:
Non – Activity

0 m -0,35 m +1,95 m +0,75 m +0,12 m -1,2 m

Sketch 1: A – 19 Sketch 2: 19 – B Archaeological Park area

Present Past Piezometric Past Piezometric Past
Piezometric Surface mean Surface mean Present Piezometric

Surface mean Slope from Slope from Piezometric Level
Slope along Necropoli to@ Traiana Road to Level calculated by
AB Section Traiana Road coastline Sketch 1 Slope

A Hypothesis:
Activity

0,048% 0,076% 0,77% +0,12 m +0,19 m

B Hypothesis:
Non – Activity

0,048% 0,073% 0,76% +0,12 m +0,18 m

Tab. 1 - Comparison among extracted levels for past and recent piezometric surface along different areas of Egnazia town.

Confronto tra i livelli ricavati per le superfici piezometriche passata e attuale in differenti zone della città di Egnazia.

Tab. 2. Comparison among extracted slopes and relative levels for present and past piezometric surface along different sketches of
AB section in Egnazia area.

Confronto tra i valori delle pendenze medie e dei relativi livelli ricavati per le superfici piezometriche presente e passata in corrisponde-
za di differenti tratti della sezione AB nell’area di Egnazia.

M. Milella et al.



elaboration indicate a relative sea level change lower
than supposed by A U R I E M M A et al. (2004; 2005).
Unluckily, they do not permit us to recognize the eusta-
tic and/or tectonic contribution although the entire area
where Egnazia is located, is considered by some
Authors in slow downlift (MASTRONUZZI & SANSÒ, 2002a;
LAMBECK et al., 2004). Moreover, this reconstruction fits
in well with some recent surveys that, in coastal areas,
individuate the bottom of Messapic tombs up to -1,1 m
present b.s.l.. Therefore taking into account that some
eastern Necropoli coastal tombs and Egnazia Roman
harbour structures stay currently below the present
b.s.l. (AURIEMMA et al., 2004) it is unlikely that Messapic
people used to bury dead people at the seaside, where
storms and spray could reach the tombs.

CONCLUSION

The aim of this work was to estimate the relative
position of the sea level during 4th–2nd cent. B.C. by
means of archaeological data from Egnazia. News and
bibliographic data permit us to imagine the trend of the
piezometric surface related to 4th–2nd cent. B.C.; two
hydrogeological sections respectively based on two dif-
ferent scenarios have been implemented. According to
the suggestions derived by Sivan et al. (2004), the first
one considers the active state of wells adding the value
of 35 cm to the elevation of the well–bottom and sub-
tracting the same value to the elevation of the tomb /
cistern–bottom. The second one considers the state of
abandoning of wells.

Data and discussion shown in these pages indica-
te that wells, cisterns and tombs dug in coastal areas
could be used in the reconstruction of piezometric sur-
face and related sea level with prudence. In fact, pot-
tery remains inside do not indicate necessarily the age
of their use but more probably their use as a dump.
Moreover, when using bibliographic data, their accu-

racy needs to be verified, in particular in the evaluation
of deep values. As a consequence the error limit of this
method on rocky coast should be bigger than the 0,10
m supposed in previous pages or the 0,10÷0,20 m sup-
posed by Sivan et al. (2001) for sandy coasts.

In summary, the shown data and the reconstruc-
tions obtained indicate that: i – the wells were dug in
times before the 4th–2nd cent. B.C.; ii – the wells were
abandoned and filled at least at the beginning of the 4th
century; iii – the piezometric level during the 4th–2nd
cent. B.C. ran at about 1,6 m below the wells-bottom;
iv – the sea level during the 4th–2nd cent. B.C. was at
least -1.2 m present one.

The discrepancy between this last data and that
indicated by the presence of the submerged harbour
structure is only apparent; in fact, the value -1,2 m is a
limit lower than the sea could stand and it is heavily
conditioned by the approximation of data registered
and derived by Digging Journals.

ACKNOWLEDGEMENTS

We would like to thank anonymous reviewers for
constructive criticism and useful suggestions, Dott.
Rossella Pagliarulo from CNR – Bari and Dott. Vittoria
Tagliente for the helpful discussions.

This work was financially supported by National
Project ARCHEOMAR by Ministero dei Beni e delle
Attività Culturali (Project Leaders: Dott. L. Fozzati; Dott.
C. Mocchegiani Carparo; Leader of Puglia Unit: G.
Mastronuzzi).

This is an Italian contribution to the IGCP Project
n.495 – International Geological Correlation Programme
“Quaternary Land-Ocean Interactions: Driving
Mechanisms and Coastal Responses ” dell’UNESCO –
IUGS (Project Leader: Dr. Anthony Long, University of
Durham, UK).

259

Fig. 7. - - - : Curve of sea level
change during the last 3000 years
in Southern Apulia reconstructed
from Auriemma et al. (2004) data;
… : Curve of sea level change
during the last 3000 years accor-
ding to the past piezometric sur-
face in Egnazia area. ▲ : Sea
level during 4th - 2nd cent. B.P.
deducted by Piezometric Level in
Archaeological Elements (Wells,
Cisterns and Tombs).
Curva della variazione del livello
del mare durante gli ultimi 3000
anni lungo la Puglia meridionale
ricostruita in base agli indicatori di
Auriemma et al. (2004); … :
Curva della variazione del livello
del mare durante gli ultimi 3000
anni secondo la ricostruzione
della superficie piezometrica del
passato nell’area di Egnazia. ▲ :
Livello del mare durate il VI - II
sec. a.C. dedotto dalla Superficie
Piezometrica ricavata dagli
Elementi Archeologici (Pozzi,
Cisterne e Tombe).

Sea level during 4th - 2nd century B.P. ...



REFERENCES

ANDREASSI G., COCCHIARO A & MARUCA A. (2000) - Egnazia
Sommersa: dalla terra al mare. Ministero per i
Beni e le Attività Culturali, Soprintendenza
Archeologica della Puglia. C.D.S. Valenzano, Bari.

ANTONIOLI F. & LEONI G. (1998) - Siti archeologici som-
mersi e loro utilizzazione quali indicatori per lo stu-
dio delle variazioni recenti del livello del mare. Il
Quaternario, 11 (1), 53 – 66.

AURIEMMA R., MASTRONUZZI G. & SANSÒ P. (2004) - Middle
to Late Holocene relative sea-level changes recor-
ded on the coast of Apulia (Italy) .
Geomorphologie: relief, processus, environne-
ment, 1, 19 – 34.

AURIEMMA R., MASTRONUZZI G., SANSÒ P. & ZONGOLO F.
(2005) - The harbour of the mansio ad speluncas
(Brindisi, Italy): a key to the lecture of sea level
changes in the past 3500 years?. In Maritime
Heritage and Modern Ports, Wit Press 2005, 5 –
14.

BORDONI P. & VALENSISE G. (1998) - Deformation of 125
ka marine terrace in Italy: tectonic implications. In
Stewart I.S., Vita-Finzi C. (eds.) Coastal Tectonics.
Geological Society Special Publications, London,
146, 71 – 100.

CALDARA M. & PENNETTA L. (1995) - Lineamenti del clima
nell’età del Bronzo lungo la fascia adriatica puglie-
se. In: Atti del Seminario di Studi: L’Età del
Bronzo lungo il versante adriatico pugliese, Bar
1995, Taras XV, 2, 21 – 37.

CINQUEPALMI A. & COCCHIARO A. (2000) - Egnazia nel
tempo. Dal villaggio protostorico al borgo medioe-
vale. Valenzano 2000.

COCCHIARO A. (1982) - I pozzi, le vasche, le cisterne. In:
Mare d’Egnazia, Fasano 1982, 93 – 106.

COCCHIARO A. (1997) - Fasano (Brindisi), Egnazia. In:
Taras, XVII, 1, 56 – 58.

COTECCHIA V. (1977) - Studi e Ricerche sulle acquee sot-
terranee e sull’intrusione marina in Puglia
(Penisola Salentina). Quad. Ist. Ric. Sulle Acque,
20, 1 – 34.

CREMONESI G. (1979) - L’età del Bronzo nella Puglia
Meridionale. In: AA. VV. La Puglia dal Paleolitico al
Tardo – Romano, Milano 1979, 179 – 192.

DALONGEVILLE R. (1987) - Le beach-rock. Travaux de la
Maison de l’Orient Méditerranéen, Lyon, 8, 197
pp.

D’ALESSANDRO A. & IANNONE A. (1984) - Prime considera-
zioni sedimentologiche e paleoecologiche su alcu-
ne sezioni della Calcarenite di Gravina
(Pleistocene) nei pressi di Monopoli. In Studi di
Geologia e Geofisica. 27, 1 – 16.

DELL’AGLIO A. (1982) - I segni dell’insediamento proto-
storico. In: Mare d’Egnazia, Fasano 1982, 29 – 36.

DICEGLIE S. (1981) - Gnathia. Forma della città delineata
mediante la prospettiva archeologica con 5 tavole
allegate. Fasano, 1981.

DONVITO A. (1988) - Egnazia. Dalle origini alla riscoperta
archeologica. Fasano, 1998.

EROL O. & PIRAZZOLI P. A. (1992) - Seleucia Pieria: an
ancient harbour submitted to two successive
uplift. The International Journal of Nautical
Archaeology, 21 (4), 317 – 327.

FLEMMING N.C. & WEBB C. O. (1986) - Tectonic and

Eustatic Coastal Changes During the last 10,000
Years Derived From Archaeological Data .
Zeitschrift für Geomorphologie N.F., Suppl. Bd.,
62, 1 – 29.

LABOREL J. & LABOREL - DEGUEN F. (1994) - Biological
indicators of relative sea – level variations and of
co – seismic displacements in the Mediterranean
region. Journal of Coastal Research 1994, 10, pp.
395 – 415.

LAMBECK K., ANTONIOLI F., PURCELL A. & SILENZI S. (2004) -
Sea–level change along the Italian coast for the
past 10,000 yr. Quaternary Science Reviews, 23,
pp. 1567 – 1598.

MASTRONUZZI G. & SANSÒ P. (2002a) - Pleistocene sea-
level changes, sapping processes and develop-
ment of valley networks in the Apulia region
(southern Italy). Geomorphology, 46, 19–34.

MASTRONUZZI G. & SANSÒ P. (2002b) - Holocene coastal
dune development and environmental changes in
Apulia (southern Italy). Sedimentary Geology, 150,
139 – 152.

MORHANGE C., LABOREL J. & HESNARD A. (2001) - Changes
of relative sea level during the past 5000 years in
the ancient harbour of Marseilles, Southern
France. Palaeogeography, Palaeoclimatology and
Palaeoecology, 166, 319 – 329.

PASKOFF R & SANLAVILLE P. (1983) - Le côte de la Tunisie,
variation du niveau marin depuis le Tyrrhènien.
Maison de l’Orient Méditerranéen, Lyon, 192 pp.

PIRAZZOLI P. A. (1996) - Sea-Level Changes. J. Wiley &
Sons, Chichester, 211 pp.

POLEMIO M. & DRAGONE V. (1999) - Serie storiche piezo-
metriche delle unità idrogeologiche pugliesi: regi-
me piezometrico, effetti climatici ed antropici.
Pubbl. GNDCI n. 2015, Quaderni di Geologia
Applicata, Pitagora Editrice, Bologna, 1999, 4,
153 – 162.

SCIARRA BARDARO B. & ANDREASSI G. (1982) - Il Porto. In:
Mare d’Egnazia, Fasano 1982, 107 – 118.

SIVAN D., WDOWINSKI S., LAMBECK K., GALILI E. & RABAN A.
(2001) - Holocene sea-level changes along the
Mediterranean coast of Israel, based on archaeo-
logical observations and numerical model. In:
Palaeogeography, Palaeoclimatology, Palaeoe-
cology, 167, 101 – 117.

STIROS S. C. & PIRAZZOLI P. A. (1995) - Paleoseismic stu-
dies in Greece: a review. Quaternary International,
25, 57 – 63.

TAGLIENTE V. (2005) - Problemi di Approvvigionamento
Idrico ad Egnazia. Tesi di Specializzazione in
Archeologia e Storia dell’Arte Romana, a.a.
2003/2004.

TANZARELLA V. (1998) - Idrogeologia della fascia costiera
adriatica compresa tra l’abitato di Fasano (BR) ed
il mare. Tesi di Laurea in Scienze Geologiche,
Università degli Studi di Bari, a.a. 1997/1998.

260

Ms. ricevuto il 20 febbraio 2006
Testo definitivo ricevuto il 31 ottobre 2006

Ms. received: February 20, 2006
Final text received:  October 31, 2006

M. Milella et al.