JASREP-00585; No of Pages 11

Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Archaeological Science: Reports

journal homepage: www.elsevier.com/locate/jasrep

Sedimentation as geomorphological bias and indicator of agricultural

(un)sustainability in the study of the coastal plains of South and Central Italy
in antiquity
Peter Attema
Groningen Institute of Archaeology, University of Groningen, Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: Environmental research of ancient landscapes in the coastal plains, river valleys and uplands of the Mediterra-
Received 25 January 2016 nean shows how erosion and sedimentation studies play a significant role in the evaluation of the archaeological
Received in revised form 14 July 2016 record at the regional and local scales. As a rule, directors of landscape archaeological projects nowadays involve
Accepted 22 July 2016
physical geographers in order to study erosion and sedimentation as potentially influential post-depositional
Available online xxxx
processes that may expose or cover up archaeological remains (long) after regions or sites were abandoned.
Keywords:
This is a phenomenon in the literature known as geomorphological bias, i.e. bias caused by landscape taphonomic
Landscape archaeology processes. Key question here is to what extent archaeological settlement patterns are an artefact of landscape
Landscape taphonomy change, with deposition obscuring large parts of the ancient Mediterranean landscape. At the same time, it is im-
Agricultural sustainability portant for our knowledge of past societies to establish whether these landscape processes affected the sustain-
Greek and Roman colonization ability of the human environments of sites and regions already while they were settled, and how people adapted
Italy to environmental changes in accordance with the socio-political and socio-economic context. Sustainability is de-
fined in this paper as the capacity of a rural economy to endure in a given environmental and socio-economic set-
ting. A key question from this perspective is whether erosion and sedimentation studies can help explain why
some rural landscapes in the long run were economically more viable than others. Drawing on case studies
from landscape archaeological and excavation projects of the Groningen Institute of Archaeology, this paper ap-
proaches Mediterranean sedimentation history in South and Central Italy from the angles of geomorphological
bias and sustainability studies. The focus is on the coastal plains of the Sibaritide in South Italy and the Pontine
plain in Central Italy, both of which have been subject to profound landscape changes caused by sedimentation
starting at least in the Bronze Age, and caused by erosion in their hinterlands as the result of long term human
impact in combination with climatic changes, sea-level change and neotectonics. Although already settled in
pre- and protohistory, both coastal plains were targeted for the first time during phases of Greek and Roman col-
onization as areas of organized agricultural expansion (see Table 1 for a chronological overview of archaeological
periods). However, in both cases long term sustainable exploitation proved difficult due to a complex of environ-
mental, technological, socio-economic, and political factors.
© 2016 Published by Elsevier Ltd.

1. Introduction landscapes in the coastal plains, river valleys and uplands of Central
and South Italy shows how erosion and sedimentation studies play a
Since Vita Finzi's, 1969 seminal publication The Mediterranean Val- significant role in the evaluation of the archaeological record at the re-
leys: Geological Changes in Historical Times, in which he outlined basical- gional and local scales as they can elucidate geomorphological biases
ly two phases of alluviation, a post-glacial Older Fill and a Late or post- in data recovered from archaeological surface survey, and at the same
Roman Younger Fill, each comprising several erosion and sedimentation time yield information on the sustainability of the Mediterranean land-
cycles, Mediterranean landscapes have seen much geoarchaeological scapes of the past. As Roberts (1998, 191–192) states, historical soil ero-
work detailing Mediterranean alluviation history of the later pre- and sion in the Mediterranean will often have been a combined product of
protohistorical periods and Classical Antiquity (Bell and Walker, 1992, natural and cultural forces, notably of climate and agricultural impact,
190–192; Walsh 2014, 81–101). Geoarchaeological research of ancient and `thin or erodible soils, steep slopes, a vegetation vulnerable to fire,
and rainfall that can be intense and erosive make the Mediterranean
ecosystems sensible to uncontrolled human impact. Casana (2008), for
E-mail address: [email protected]. instance, has shown how a variable precipitation regime increase,

http://dx.doi.org/10.1016/j.jasrep.2016.07.024
2352-409X/© 2016 Published by Elsevier Ltd.

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
2 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

including extreme weather events, will have strong effects on the rate of lower parts of the landscape. Next, the issue of how far we may study
erosion in intensively exploited landscapes, such as the Hellenistic and sedimentary archives from a contemporary perspective will be dealt
Roman landscapes in the Northern Levant. From an archaeological per- with, in other words: how did erosion and sedimentation actually affect
spective, John Bintliff's 1977 publication Natural Environment and the sustainability of the human environments of sites and regions while
Human Settlement in Prehistoric Greece (1977) may be viewed as a they were settled? For this aspect, I will draw on the same two case
benchmark in establishing a firmer disciplinary relationship between studies, the Sibaritide and Pontine plain, defining sustainability in a
regional settlement history (as practiced by archaeologists) and the landscape archaeological context as the capacity of a rural economy to
study of the natural environment of the Mediterranean (as practiced endure in a given environmental and socio-economic setting. I empha-
by geographers and biologists). However, those earlier models were size the special position of the coastal plains in the Mediterranean envi-
still rather deterministic, as Bintliff himself readily remarked in a later ronment being both among the most fertile and most vulnerable
overview paper on landscape change in Classical Greece published in landscape units.
2000. In that paper, he concluded how between the end of the seventies
and the end of the nineties rapidly growing body of empirical evidence
would rather support more multi-causal interpretations ‘whilst also 2. Sedimentation as geomorphological bias in South and Central
raising important questions regarding the impact of these events on Italy
contemporary societies’ Bintliff (2000), 68; see also Bintliff (1992) and
Bintliff (2002). At the same time survey archaeologists working in the The geography of South and Central Italy is characterized by inland
Mediterranean became aware of the biases erosion and sedimentation mountainous areas, river valleys and wide coastal plains. In the latter,
represented for the interpretation of the regional archaeological record without exception, considerable sediments accumulated over time in
they were mapping. As van Leusen stated in his dissertation published their hinterlands (Brückner 1986; Abbott 1997; Abbott in Carter and
in 2000, geopedological research had become a required part of regional Prieto 2011; Attema and Sevink, forthcoming); posing serious problems
projects not only in order to map one of the most important factors in for our interpretation of regional settlement development. While early
past land use, but also to map geomorphological bias in the survey re- forms of urbanization in many of the South and Central Italian land-
sults (van Leusen 2000, chapter 8, 4). This paper focuses on these two scapes can now be traced back into the final phases of the Bronze Age
main themes that can be summarized as landscape change as a major (Pacciarelli 2000), we must realize that due to geomorphological bias
bias in regional settlement studies and the effect of landscape change we have but a partial view of rural settlement accompanying nucleated
in coastal plains on sustainable land use. (See Table 1.) settlement. This not only holds true for prehistoric phases, but also for
While many survey projects nowadays integrate environmental the landscapes of Classical Antiquity. For example, in the Metapontine
work, to date there are only few studies available that have integrated plain on the Adriatic coast, many of the farmsteads of the early Archaic
such work on a regional scale. Moreover, it is uncertain whether archae- period (around 600 BCE) may be buried below the alluvium of the
ologists, or ancient historians for that matter, who work with regional Basento and Bradano rivers and their tributaries. In the Basento river
settlement data always grasp the implications of erosion and sedimen- valley, they were only found when infrastructural works were carried
tation processes for a reliable interpretation of regional settlement pat- out below the current plough zone (Carter and Prieto 2011, 641–643).
terns across space and time. But neither are the economic problems and However, remains of farmsteads of the Classical and Hellenistic period,
opportunities that natural and man-induced landscape change brought whose owners did not seek locations along the river but founded their
to past societies always fully realized; in other words, how did land- farms over the plain between the main river valleys of the Basento
scape change affect the sustainability of settlement and land use in a and Bradano, do appear in abundance in the plough zone (Carter and
given region in antiquity, and how did humans react to perceived Prieto 2011, 677 ff.). Clearly the sedimentation regime of the
changes in their environment and how did people develop and apply Metapontine rivers has consequences for our chronological and spatial
environmental knowledge (Walsh 2014, 7–9)? understanding of the scale of ruralisation during the Archaic period con-
Below I will address the issues of geomorphological bias and sustain- nected with Greek colonization.
ability of Mediterranean landscapes outlined in the introduction by The Holocene sedimentation record is not only a problem for under-
discussing two case studies from fieldwork projects carried out by the standing the settlement archaeology of the Metapontino, but that of
Groningen Institute of Archaeology in Central and South Italy (Attema other coastal plains of South Italy as well (see e.g. Abbott and Valastro
et al. 2010, chapters 2 and 4 for introductions to these landscapes). 1995, Abbott 1997), and similar problems occur in the coastal land-
First I will present examples of sedimentation regimes that had caused scapes and river valleys in Central and northern Italy, as exemplified
the profound burial of ancient land surfaces, as such seriously impeding by the buried prehistoric landscape of the Po plain (Calabrese et al.
the reading of settlement development on the regional scale. Two cases 2010). The two case studies of the Sibaritide and Pontine plains are
will be presented; the Sibaritide coastal plain in northern Calabria and therefore not exceptional as to their dynamic character, but at the
the Pontine coastal plain in south Lazio (Fig. 1). In both cases hand same time serve to show the notable variability in timing, nature and
augerings, the cores of which were radiocarbon dated, have revealed impact of sedimentation regimes within regions across Italy, and by ex-
the sedimentary impact of erosion during the later Holocene on the tension the Mediterranean.
In the plain of the Sibaritide, sedimentation has been intense, and
here the entire prehistoric to (post) Roman landscape is buried under
Table 1 a many meters of alluvial sediment, the causes of which are debated
Periodisation used in this article (date ranges are in years BC/AD). in the geological literature (Vanzetti 2013, 24–28 for an overview). In
6500–3650 Neolithic
the Pontine plain the situation is different. Here the protohistoric settle-
3650–2250 Eneolithic (Copper Age) ment phases are only locally deeply buried, while as a rule the Roman
2250–1700 Early Bronze Age landscape is visible in the plough soil (Attema et al. 2010, 31–58). This
1700–1300 Middle Bronze Age signifies that sedimentation was reduced in post-Roman times, possibly
1300–925 Late Bronze Age
because of lack of sediment supply from an already eroded hinterland.
925–725 Early Iron Age
725–580 Late Iron Age The two case studies illustrate not only the problem of how to cover se-
580–480 Archaic period rious chronological and spatial gaps in our knowledge of regional settle-
480–30 Republican period ment development that are due to geomorphological bias but also the
30–476 Imperial period problem how to reconstruct evolving and contracting regional settle-
476–1000 Early Middle Ages
ment on the basis of partial data.

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx 3

Fig. 1. Map of Italy showing the location of the plain of Sybaris in South Italy and the Pontine plain in Central Italy. Also indicated are the Metapontine and Po plains.

2.1. The Sibaritide plain possible, although hydrological management would have been required
(Bellotti et al., 2009, 69). Sedimentation increased again during the later
The vast alluvial coastal plain of Sybaris is located on the Gulf of Ta- historical periods which, in combination with failing hydrological man-
ranto on the Adriatic Sea and is surrounded by the foothills of the Pollino agement, resulted in a deeply buried archaeological landscape.
mountains in the north-west and the foothills of the Sila and Mula The effects sedimentation in the plain of Sybaris have had on the vis-
mountains in the south-west (Attema et al. 2010, 82). Fig. 2 shows the ibility of the archaeological record is clear from even a quick glance on
extent of the plain with sites mentioned in the text and major rivers the archaeological map of the Sibaritide produced in the late 1960′s by
draining it. These rivers over time deposited thick layers of clay and De Rossi et al. (1969). (Fig. 3) While the foothills show a busy protohis-
clayey silts. Sedimentary deposits at places contain intermediary layers toric and classical landscape, the plain is suspiciously empty, except for
with organic materials suitable for radiocarbon dating. From published the excavated site of Sybaris that was founded by Greek settlers in the
dated sequences obtained in various investigations by coring, it can be period of the Greek migrations starting in the late 8th century BC. In-
safely deduced that sedimentation in the plain started well before the deed, Sybaris was only discovered after an intensive mechanical coring
foundation of the Greek colony of Sybaris (e.g. Cherubini et al. 1994; and geophysical campaign carried out in the 1960s by a research team
Cucci 2005; Bernasconi et al., 2010; Attema et al. 2010, 22 and fig. 1.6; of the University of Pennsylvania led by Froehlich Rainey in collabora-
Ferranti et al. 2011). A recently published, though highly tentative re- tion with Carlo M. Lerici of the Italian Lerici foundation (Rainey and
construction of the intensity of sedimentation across time on basis of Lerici 1967). The Greek colony was found to be located along the river
all available radiocarbon dates by Italian protohistorian Alessandro Van- Crati below a thick cover of alluvium. Palaeogeographical research indi-
zetti, would indicate that sedimentation had already started by around cates that the settlement bordered a lagoon that would have functioned
6000 BCE. This is interpreted by this author as the consequence of seri- as an inland harbor (tentative reconstructions in Bellotti et al. 2009,
ous human impact on the landscape in the hinterland of the plain due to Fig. 2). Considerable progradation of the coastline since the Iron Age, al-
the neolithisation process. This, however, seems unlikely given low though reconstructed in various ways, is responsible for the now more
population numbers (compared with later periods), the use of hoe cul- inland position of ancient Sybaris that at the time was certainly a coastal
tivation and the non-anthropogenic factors that should be taken into ac- city like most other Greek and Phoenician colonies planted around the
count (Vanzetti 2013, 27 and Fig. 4). Sedimentation in following phases Mediterranean and the Black Sea. Archaeological materials collected at
would, according to this author, have been less intense, and relatively considerable depths from the over 1500 Rainey/Lerici corings and radio-
stable periods occurred, as for instance during the Eneolithic and An- carbon dating of sediments, done as part of the various environmental
cient Bronze Age. During the periods of Greek colonization and subse- investigations referred to above, demonstrate continuous alluviation
quent Hellenistic and Roman periods, rural expansion in the plain was in the coastal plain of Sybaris following ancient occupation phases,

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
4 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

Fig. 2. The plain of Sybaris in South Italy with archaeological sites and rivers mentioned in the text (after Fig. 4.1 in Attema et al. 2010).

this renders archaeological surface survey a hopeless exercise. Fig. 4 period, settlement in the foothills was also lower, as we have observed
shows the dated deposits in hand augerings carried out by the Groning- in systematic archaeological surveys along part of the foothills of the
en Institute of Archaeology in the plain. At certain locations we found Sibaritide (van Leusen and Attema 2003).
that radiocarbon dates at a depth of over 7 m were not older than
2150 years BP. The regression line indicates an average sedimentation 2.2. The Pontine plain
speed of 0.5 cm/year over the period 2100 to 450 BP and suggests that
strong sedimentation stopped at about 450 BP. The number of dated The Pontine plain is located on the Tyrrhenian coast in Central Italy
augerings on the whole is still limited, as Roovers 2011 has stressed, and borders inland on the limestone massifs of the Lepini mountains
and it is not possible to differentiate sedimentation processes in more in the north and the Ausoni mountains in the east (Fig. 6). It comprises
detail over the plain, as is possible for the second case study presented a complex series of marine terraces on the coast and an inland lagoon,
below. However, a conclusion that can be made is that the sedimenta- still open during the Bronze Age (Feiken 2014, 259–280 and Fig. 9.2
tion rates will differ according to the proximity of the rivers (Roovers on the palaeography of the Pontine plain). The lagoon was gradually
2011, 21). The calculations on average sedimentation speed are in line filled in with lagoonal sediments turning it into a swamp as the marine
with similar estimates by Italian researchers (Cotecchia and Pagliarulo terraces impeded drainage of rivers originating in the hinterland and
1996; Cherubini et al. 1994). water coming from springs at the foot of the Lepini mountains
In the Rainey/Lerici corings, datable archaeological layers with iden- (Kamermans 1991, Sevink et al. 1984). The swamp, today drained, be-
tifiable Archaic, Classical and Hellenistic or Roman materials were gen- came historically known as the Pontine marshes. Colluviation and
erally found at depths between 3 and 6 m, far below the level reached by alluviation in the plain notably affected those parts of the plain adjacent
modern deep-ploughing. Mapping the archaeologically relevant to the foothills of the Lepini and Ausoni mountains, but were also caused
augerings in GIS in period maps, as illustrated in Fig. 5, gives an interest- by rivers originating in the nearby Alban hills that deposited sediments
ing but of course very partial insight in the rural development of settle- in the northwestern part of the Pontine plain (Sevink et al. 1984, Attema
ment in the plain. In the sixth century BC, the city of Sybaris would and Delvigne, 2000). In the northeastern part, the Amaseno river depos-
already have been quite large and ruralisation low, but by the 4th c. ited sediments burying parts of the ancient landscape, including that of
BC, after the refounding of the city, then called Thurioi, the city had the Roman period (van Joolen, 2003). Like in the Sibaritide, the land-
contracted, but ruralisation had increased (Roovers 2011). In the scape, as we perceive it today, is thus very different from that of antiq-
Rainey/Lerici corings in the plain only few Roman Republican or Impe- uity, and in the Pontine plain sedimentation has likewise been an
rial materials were found that can be related to the Roman colonia of important agent of environmental change. However in contrast to the
Copiae. The plain appears to have been largely abandoned during this plain of Sybaris, sedimentation in the Pontine plain has only buried

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx 5

Fig. 3. Archaeological sites mapped in the archaeological surveys of De Rossi et al. 1969 (after Fig. 1.7 in Attema et al. 2010).

parts of the archaeological landscape, and alluvial and colluvial deposits deposition could be reconstructed in the Amaseno basin dating between
generally cover only the pre-Roman phases; Roman and later artefacts 1600 and 1400 BCE and CE 600–700 potentially burying archaeology
pertaining to ancient settlements in large parts of the Pontine plain (al- (van Joolen, 2003, 68–84). In the plain below Sezze, radio carbon
though not everywhere) will appear at the surface after ploughing dated hand augerings likewise revealed a thick alluvial/colluvial sheet
while Roman built remains may still be visible on the surface (Attema that covered the former landscape, but this sedimentation started earli-
1993, Attema and van Leusen 2004, de Haas 2011). This means that, al- er. Peaty deposits to depths of up to 6 m were dated to 2500 to 2200 BCE
though to a lesser extent, the Pontine plain is influenced by a strong (Attema et al. 2010, 41–43; Feiken 2014, 196–200). The supply of sedi-
geomorphological bias as the archaeological surface record will high- ments needed for this gradual transformation may be attributed to soil
light Roman occupation phases but underrepresent previous phases. instability and consequent erosion in the Lepini and Ausoni mountains
Below I will briefly discuss two areas in the Pontine plain, the lower and in the Alban Hills. Palynological data from cores carried out in the
Amaseno valley and the Campi di Sezze respectively. These areas are plain point to increased human impact in all landscape units (van
geographically near to each other, but show different sedimentation Joolen, 2003), while the chronology of sedimentation phases would in
histories relevant to ancient occupation and land use potential (Fig. 7). general terms correlate with increased settlement pressure in the
Such investigations were carried out within the framework of the Pon- wider landscape around the plain from the Middle Bronze Age onwards
tine Region Project which, since the mid-1980s has studied the settle- (Alessandri 2013 for Bronze Age settlement evolution). Sedimentation
ment and land use history of the Pontine region from the Bronze Age in the plain may thus have been caused by deforestation and subse-
into the Medieval period (Attema et al., 2010, 31–58). On the basis of quent grazing on the slopes of the Lepini and Ausoni mountains and
the sedimentological information obtained from hand augerings and ra- the Alban Hills although the influence of climatic factors cannot be
diocarbon dating of peat layers, several phases of (fluvio)-colluvial ruled out.

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
6 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

consequence of inland erosion. In this sense our examples can be said

to be representative for the larger part of the coastal plains of the Italian
peninsula (Attema and Sevink forthcoming). However, sedimentation
histories appear to have quite different impacts on the visibility of the
archaeological record. While in the plain of Sybaris, sedimentation cov-
ered the entire sequence of past occupation, in the Pontine plain this
was more discontinuous and localised. Clearly sedimentation regimes
are dependent on local geological circumstances, climatic changes and
local episodes of human impact, and many more case studies from Med-
iterranean landscapes are available that reinforce this point (see e.g.
Walsh 2014). While we must thus acknowledge local diversity in
causes, effects and timing of sedimentation phases, there may be
supraregional spatial and chronological patterning too, and more re-
search should be done to correlate data on episodes of human impact
with climate change now that in various studies Holocene climate
changes are being described (e.g. Magny and Combourieu Nebout,
2013, Mensing et al. 2015). For instance, above we discussed the case
of Tratturo Caniò in the Pontine plain; the challenge however will be
to correlate this case with the many other observations done in the Pon-
tine plain in order to arrive at a detailed long term regional picture of the
interaction between man and his environment that includes Holocene
climatic changes. As of yet, correlating the substantial body of environ-
mental data from sedimentation studies and palynological research
from the Pontine Plain with data relating to climate change is hampered
by insufficient robustness of the available radiocarbon dates. The data
from sedimentation studies and individual pollen cores have not yet
been interconnected to carry out an integrated environmental study
Fig. 4. Sedimentation speed in the Sibaritide plain as derived from depth of radiocarbon
on the regional scale, though this is planned in current research of the
dated deposits in hand augerings (after Attema et al. 2010, Fig. 1.6). Pontine Region Project. From our case studies it is clear that systematic
research into the variable sedimentary regimes in order to map geomor-
phological bias is of the utmost importance for the evaluation of settle-
Nowadays, the Lepini and Ausoni mountains appear very eroded, ment patterns of any coastal landscape or river valley, and therefore
and sediment supply in the plain of Sezze may have stopped in some should be an integral component in every landscape archaeological pro-
areas during the Roman period. Excavations carried out by the Groning- ject. A second reason to study sedimentation and its origins, erosion, in
en Institute of Archaeology in the plain below Sezze at Tratturo Caniò re- archaeological survey projects is to obtain knowledge on the conse-
vealed a succession of occupational layers, the earliest of which was quences of erosion and sedimentation for contemporaneous society,
found just above an ash layer attributed to the Vesuvius Avellino erup- and how the phenomenon was perceived and dealt with in past
tion dated shortly before 2000 BCE (Feiken et al. 2012; Feiken 2014). societies.
This ash layer has been observed in multiple corings and test pits in
the Pontine plain (Sevink et al. 2011). Middle Bronze Age artefacts 3. Sedimentation as indicator of past sustainability of the Mediterra-
and ecofacts at Tratturo Caniò point to a settlement context indicating nean landscapes of antiquity
that by the beginning of the Middle Bronze Age the landscape bordering
on the former lagoon had become sufficiently dry for permanent human Ancient agro-ecosystems, as Abbas Farshad (2002, 190) calls them,
occupation, though concentrated at certain attractive locations, such as change “so gradually that they looked almost unchanged for centuries”.
relatively high and dry levees. Levels higher in the stratigraphy at However, we know from environmental studies, as exemplified above,
Tratturo Caniò attested to occupation well into the Roman Republican that erosion and sedimentation in certain landscapes were such that
period (Feiken et al., 2012; Feiken 2014; Attema forthcoming). Early they must have had serious impact on the sustainability of ancient agri-
Iron Age to Roman Republican phases showed a, possibly continuous, culture and subsistence strategies. Regional studies have shown that
cultic use of the site, culminating in the erection in the Republican peri- landscape change due to erosion and sedimentation in the Mediterra-
od of a small stone temple dedicated to the goddess Juno. In its sur- nean occurred on the scale of centuries, most often due to intensifica-
roundings intensive archaeological survey mapped a densely settled tion of land use in uplands and sub-mountainous zones, with climate
Roman countryside (Attema et al. 2014).The excavation at Tratturo change as an important factor too. Locally landscape change due to ero-
Caniò however made clear how the Roman landscape visible in the sion and sedimentation could however have had short term conse-
plough soil was partly a continuation of earlier occupation phases not quences and may have taken the form of local catastrophic events,
visible in the archaeological surface record due to alluvial and colluvial such as landslides, mudflows and lahars. An instance of a mudflow af-
deposition following the initial Bronze Age occupation phase. At the fecting the ancient landscape was mapped in detail in the Pontine
same time, it makes clear how people adapted to changing environmen- plain and can be cited as an example of a short term, potentially cata-
tal circumstances in the periods following the Bronze Age. strophic change (Attema 1993, 97–105). While Farshad (2002, 189) is
right that knowledge of past environmental changes helps us to under-
2.3. Local and regional variability in sedimentation regimes stand the dynamics, behavior, tolerance and resource potential of land-
scapes across time and space, and that this knowledge may be used for
The Sibaritide and Pontine plains show that both landscapes were present-day strategies of land evaluation, it also informs us on the actual
dynamic environments due to alluviation and colluviation as a sustainability of ancient landscapes and the ways farmers dealt with

Fig. 5. Approximate urban surfaces of Sybaris (6th c. BC), Thurioi and Copiae and intensity of rural occupation as deduced from the Rainey/Lerici corings (after Roovers 2011,
Figs. 9b, 10b, and 11b).

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx 7

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
8 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

Fig. 6. Map showing the location of the Pontine plain in Central Italy with archaeological sites and rivers mentioned in the text (after Fig. 1.2 in Attema et al. 2010).

environmental change, or in the words of Kevin Walsh how ‘people un- rural settlement was dependent on hydrological technology to drain
derstand and engage with their landscape and environment’. The latter and irrigate the land, and implementation of canals led to successful
scholar asserts that Mediterranean cultures do share certain forms of rural exploitation (Carter and Prieto 2011, 1027–1051). Maintenance
landscape-management strategies but acknowledges at the same time of hydrology was therefore imperative and dependent on a well-func-
that ‘these strategies are contingent upon historical, cultural, and eco- tioning socio-political organization. Joseph Coleman Carter has on the
nomic processes that vary across time and space’ (Walsh 2014, 7–8). basis of his archaeological surveys convincingly shown how fluctuations
Current cultural ecological approaches emphasize this relationship occurred in the density of rural settlement during the Hellenistic period
and are helpful in understanding the interaction between humans and in the Metapontino and how already in the Roman period the intensive
their environments. Below I will explore this approach further, for cultivation of the land was abandoned, due to a complex of political and
which I again turn to the coastal plains of South and Central Italy. socio-economic changes and subsequent environmental deterioration
Greek colonization on the shores of the Mediterranean was charac- due to lack of caring for the land.
terized by the founding of coastal settlements, many of which were The same scenario can be sketched for the ruralisation of the plain of
quick to develop into sizeable port towns (Osanna 1992). In South Sybaris where agriculture during the Hellenistic periods was wide-
Italy, Sybaris, situated on a slightly inland lagoon and river, is an emi- spread judging from the Rainey/Lerici corings with archaeological mate-
nent example of this. Clearly the population of these steadily growing rials found in them. Apparently the produce from the plain, including
cities had to be sustained by agricultural produce from the surrounding that of the surrounding foothills, was capable to sustain the inhabitants
land. While initially the food demand would have been met by citizens of Thurioi, the successor city of Sybaris, with enough surplus for trading
working the land but living in the colonies, soon ruralisation of the sur- to bring in the capital needed to invest in the upkeep and embellish-
rounding countryside would have been necessary to meet the food de- ment of the city. But also in the plain of Sybaris farming was not sustain-
mands of the expanding centre (Yoffee 2005, 60). This demand resulted able in the longer term due to complex socio-economic, political and
in growing numbers of permanent rural settlements, villages, hamlets, environmental factors. Already at the start of the Roman period, land
farmsteads or a combination of these, producing food surpluses. This was abandoned, while failing counter measures against the silting up
process must have started in the 6th century BC, as in the Metapontino of drainage canals led to the concurrent formation of extensive
with its Archaic farmsteads found below the alluvium along the Basento marshlands. Indeed, malaria ridden swamps were a common sight in
river, and in the Sibaritide with sporadic Archaic pottery from cores in the coastal plains of Italy from later antiquity onwards, a situation that
the plain of Sybaris. From the abundant archaeological surface record only ended with the wholesale reclamation programs under Mussolini
dating to the Hellenistic periods, it is clear that successful expansion of around the 1930's when many of these swamps (both in Italy and

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx 9

Fig. 7. Map showing locations of cores in the ‘Campi di Sezze’ area (white dots) and the lower Amaseno river valley (black dots with white outline) and sediment cover (after Fig. 2.4 in
Attema et al. 2010). The numerals 2–4 indicate various phases of sedimentation in the lower Amaseno valley.

elsewhere in Europe) were eliminated on a grand scale to be turned into plain and foothills (Attema et al. 2014). At the same time the landscape
agricultural land (Renes and Piastra 2011). Of the many projects, the along the lower Amaseno remained subject to environmental change
reclamation of the Pontine Marshes is archaeologically and historically for a longer period that extended well into the medieval period.
probably researched in most detail (Attema 1993, 28; Linoli 2005). As such, the Pontine plain is a pre-eminent example of an agricul-
Coastal plains like the plain of Sybaris and the Metapontino in antiquity, tural expansion zone in a vulnerable landscape. Like in the coastal
as well as in the recent past, were pre-eminently agricultural expansion plains of South Italy colonized by the Greeks, attempts were also
zones with the more stable agricultural agro-ecosystems situated in the made in the Pontine plain to reclaim the area for intensive cultiva-
foothills. In the hills around the plain of Sybaris, for instance, settlement tion, here slightly later, during the mid-Republican period, and in
continued after the period of Hellenistic expansion, be it much reduced this case the Romans (Walsh et al. 2014). While farming the Pontine
in terms of settlement numbers. plain initially may have been successful, as elsewhere, it was not sus-
Also in the Pontine plain sedimentation caused landscape instability tainable, and we have evidence that the situation in the graben dete-
in the plain at various points in time and with spatial variability. Sedi- riorated by the end of the Republican period. In addition, here the
mentological and ecological investigations below the level of the plain became covered in swamps, again most likely due to lack of
Avellino ashes that covered the Pontine plain suggest that the surround- maintenance of the drainage works that initially had made settle-
ings of the fresh water lagoon were already cultivated well before the ment and intensive agriculture possible, but now led to the forma-
deposition of the ash shortly after 2000 BCE. However, at a later stage, tion of marshes (Walsh et al. 2014).
at the end of the Bronze Age, alluvio-colluvial deposits would cover On the one hand, the above case studies show that landscape-man-
this particular part of the landscape, turning it into an unstable environ- agement strategies are indeed contingent upon historical, cultural, and
ment to live in, negatively affecting the Bronze Age wetland economy economic processes that vary across time and space, as Walsh asserts,
on the shores of the former lake. Judging from the archaeological evi- but also that we, through comparative research, may start to discern
dence, occupation of the area around Tratturo Caniò remained possible, patterning in sustainable and unsustainable landscapes and land-
although settlement may have been intermittent. However, by the scape-management strategies in ancient Mediterranean landscapes. To
Roman period, the area, then either constituting farmland of the study the phenomenon of episodic intensive cultivation cycles in the
Roman colony of Setia, or land designated as ager romanus, was widely Mediterranean coastal plains, we clearly need extensive detailed sedi-
available for intensive agriculture, and a flourishing local economy de- mentological research of past drainage systems in combination with ab-
veloped here focusing on the production of wine and olive oil in the solute dating of their initial construction, use history and their final

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
10 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

neglect as indicated by sedimentary infill. This will increase our under- References
standing of the technology, life-cycle and sustainability of reclamations
in antiquity as well as give insights into environmental knowledge that Abbott, J.T., 1997. Late Quaternary Alluviation and Soil Erosion in Southern Italy (PhD dis-
sertation) The University of Texas at Austin.
was available. This will allow us to bring episodes of intensive and often Abbott, J.T., Valastro, S., 1995. In: Lewin, J., Macklin, M.G., Woodward, J.C. (Eds.), The Ho-
centrally organized farming into relation with socio-political organiza- locene Alluvial Records of the Chorai of Metapontum, Basilicata, and Croton, Calabria,
tion. When Roman farming in the Pontine plain in late Antiquity ceased Italy, Mediterranean Quaternary River Environments: Referred Proceedings of an In-
ternational Conference University of Cambridge/United Kingdom/28–29 September.
altogether, it was, as in the Sibaritide, the foothills where settlement and AA Balkema, Rotterdam, pp. 195–205.
land use continued. This is another example of patterning: the foothills Alessandri, L., 2013. Latium vetus in the bronze age and early iron age/Il latium vetus
as relatively stable intermediate landscape units between the Mediter- nell'età del Bronzo e nella prima età del Ferro. BAR International Series 2565,
Oxford.
ranean coastal plains and the uplands. These landscape units seem to Attema, P.A.J., 1993. An archaeological survey in the Pontine region. A Contribution to the
have been the most resilient and sustainable parts of the Mediterranean Early Settlement History of South Lazio 900–100 BCE, Vol I Text. University of
landscape being suitable for polyculture and pastoral exploitation. The Groningen.
Attema P.A.J. Palus or ager, changing perceptions of the economic landscape of the Ponti-
point that Mediterranean coastal plains may have in common is that ne wetlands (South Lazio, Central Italy) (forthcoming).
they could not be sustainably exploited for larger scale commercial ara- Attema, P.A.J., Delvigne, J., 2000. Settlement dynamics and alluvial sedimentation in the
ble farming over longer periods, and that in antiquity this rather hap- Pontine region, Central Italy: a complex relationship. In: Vermeulen, F., De Dapper,
M. (Eds.), Geoarchaeology of the Landscapes of Classical Antiquity. BABesch Supple-
pened in relatively short (boom-bust) cycles due to the inherent
ment 5, Leiden: Stichting BABesch, pp. 35–47.
environmental instability of the coastal plains in combination with Attema, P.A.J., Sevink, J.J., 2016. In: de Angelis, F., Maiuro, M. (Eds.), Environmental Re-
changing socio-economic and geo-political contexts less conducive to search and Landscape Archaeology, Oxford Handbook of Pre-Roman Italy (1000–
their exploitation, among which we should certainly rank the waxing 49 BCE). Oxford University Press, Oxford (forthcoming).
Attema, P.A.J., van Leusen, P.M., 2004. The early roman colonization of South Lazio: a sur-
and waning of the coastal cities. To substantiate such a claim, more com- vey of three landscapes. In: Attema, P. (Ed.), Centralization, Early Urbanization and
parative work is needed. Colonization in First Millennium BC Italy and Greece, Part 1: Italy. Peeters, Leuven-
Paris-Dudley, MA, pp. 157–195.
Attema, P.A.J., Burgers, G.-J., van Leusen, P.M., 2010. Regional pathways to complexity,
4. Conclusions landscape and settlement dynamics in early Italy from the Bronze age to the Repub-
lican period. Amsterdam Archaeological Studies 15. Amsterdam University Press,
In this paper I have explored two themes that can be investigated Amsterdam.
Attema, P.A.J., de Haas, T.C.A., Tol, G.W., 2014. Villas and farmsteads in the ager Setinus
through sedimentation studies. (Sezze, Italy). Palaeohistoria 56, 177–244.
The first was geomorphological bias, i.e. bias caused by landscape Bell, M., Walker, J.C., 1992. Late Quaternary Environmental Change, Physical and Human
taphonomic processes, as a fundamental factor in the interpretation of Perspectives, Longman Scientific and Technical: Harlow.
Bellotti, P., Caputo, C., Dall'Aglio, P.L., Davoli, L., Ferrari, K., 2009. Insediamenti umani in un
local and regional settlement studies that are based on archaeological paesaggio in evoluzione: interazione uomo-ambiente nella Piana di Sibari (Calabria
surface data. In order to establish whether settlement data acquired in Ionica), Il Quaternario. Ital. J. Quat. Sci. 22 (1), 61–72.
archaeological surveys are representative, survey archaeologists need Bernasconi, M.P., Stanley, J.D., Caruso, C., 2010. Sybaris–Thuri–Copia deltaic settings in Ca-
labria, Italy: Molluscs, associated biogenic components, and Ecobiostratigraphy ap-
knowledge of the sedimentation history of a given study area. Of course
plied to archaeology. J. Coast. Res. 262, 377–390.
this also holds for erosion, the related counterpart of sedimentation. Re- Bintliff, J., 1977. Natural Environment and Human Settlement in Prehistoric Greece, Ox-
cent geoarchaeological surveys by Van Leusen within the framework of ford: BAR Supplementary Series 28.
the ‘Hidden Landscapes Project’ have gone into both aspects of sedi- Bintliff, J., 1992. Erosion in the Mediterranean lands: a reconsideration of pattern, process
and methodology. In: Boardman, J., Bell, M. (Eds.), Past and Present Soil Erosion.
mentation and erosion in the mountainous hinterland of the Sibaritide Oxbow Books, Oxford, pp. 125–131.
and that of the Pontine Region with a strong focus on the development Bintliff, J., 2000. Landscape change in classical Greece: a review. In: Vermeulen, F., de
of new methodologies of dealing with this bias factor. The results of Dapper, M. (Eds.), Geoarchaeology of the Landscapes of Classical AntiquityBabesch,
Supplement 5. Stichting Babesch, Leiden, pp. 49–70.
those recent studies indicate that more effort has to be put into this as- Bintliff, J., 2002. Time, Process and Catastrophism in the study of Mediterranean alluvial
pect of archaeological surveying (Feiken 2014). history: a review. World Archaeol. 33, 417–435.
The second theme I explored concerns sedimentation as a proxy for Brückner, H., 1986. Man's impact on the evolution of the physical environment in the
Mediterranean region in historical times. GeoJ. 13 (1), 7–17.
sustainability. Empirical evidence would indicate that the landscapes of Calabrese, L., Di Cocco, I., Centineo, M.C., 2010. Hydrographical and paleogeographic evo-
antiquity may be broadly categorized in two categories; stable land- lution of the southeastern Po Plain (Italy) during the last 4,000 years: an example of
scapes with corresponding stable agro-ecosystems where traditional integration between stratigraphy and archaeology. Geology of the Adriatic area.
GeoActa, Spec. Publ. 3, pp. 103–108
environmental knowledge and limited technological investment go Casana, J., 2008. Mediterranean valleys revisited: linking soil erosion, land use and climate
hand in hand with traditional Mediterranean farming and reclaimed variability in the Northern Levant. Geomorphology 101, 429–442.
landscapes where technology and centralized socio-political organiza- Cherubini, C., Cotecchia, V., Pagliarulo, R., 1994. Geological and geotechnical problems
connected with the disappearance of the ancient city of Sybaris. Sci. Technol. Cult.
tion are needed to sustain durable exploitation. Of course this is a far-
Herit. 3, 95–112.
reaching simplification of reality, since, if we define sustainability in a Cotecchia, V., Pagliarulo, R., 1996. State of the art in geological, hydrogeological and geo-
landscape archaeological context as the capacity of a rural economy to technical researches carried out on the archaeological site of Sybaris. Geol. Appl.
endure in a given environmental and socio-economic setting, we also Idrogeol. 31, 43–54.
Cucci, L., 2005. Geology versus myth: the holocene evolution of the Sybaris plain. Ann.
need to study, besides erosion and sedimentation, other factors, such Geophys. 48 (6), 1017–1033 (December 2005).
as climate, hydrographical conditions, crop choice, soil depletion and De Haas, T.C.A., 2011. Fields, Farms and Colonists, Intensive Field Survey and Early Roman
other forms of mismanagement as well as all kinds of socio-economic Colonization in the Pontine Region, Central Italy. Barkhuis & Groningen University
Library.
and political factors. De Rossi, G.M., Pala, L., Quilici, L., Quilici Gigli, S., 1969. Carta Archeologica della piana di
Cultural ecological approaches as advocated by Walsh (2014) fi- Sibari, extract from Atti e Memorie della Società Magna Grecia. Nuova Serie IX-X
nally may help archaeologists to understand the cycles of settlement (1968–1969), Roma, pp. 91–155.
Farshad, A., 2002. Land evaluation in archaeology, a way to enrich the reconstruction of
and land use in the regional landscapes they survey in terms of sus- past agrarian landscape: a case study of Iran. In: Attema, P.A.J., Burgers, G.-J., van
tainability, as defined above. Eventually, this may offer insight in Joolen, E., van Leusen, M., Mater, B., BAR International Series (Eds.), New Develop-
the interesting topic of past human perception of and responses to ments in Italian Landscape Archaeology. Vol. 1091, pp. 189–195.
Feiken, H., 2014. Geo-archaeological approaches to the hidden landscapes of Italy (PhD
environmental change. thesis) Rijksuniversiteit Groningen.
Feiken, H., van Leusen, P.M., Anastasia, C., Tol, G.W., 2012. Reconstructing a Bronze age
Acknowledgements hidden landscape: geoarchaeological research at Tratturo Caniò (Italy, 2009).
Palaeohistoria 53–54, 109–159.
Ferranti, L., Pagliarulo, R., Antonioli, F., Randisi, A., 2011. “Punishment for the Sinner”: ho-
The author likes to thank the editors and the two anonymous re- locene episodic subsidence and steady tectonic motion at ancient Sybaris (Calabria,
viewers for their helpful comments. southern Italy). Quat. Int. 232 (2011), 56–70.

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024
 P. Attema / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx 11

Kamermans, H., 1991. Faulted land: the geology of the Agro Pontino. In: Voorrips, A., S., L., Rasines, P., Wood, R., Walsma, A., 2011. Robust date for the Bronze age Avellino erup-
H., K. (Eds.), The Agro Pontino Survey Project, Studies in Prae- en Protohistorie 6, Am- tion (Somma-Vesuvius): 3945 ± 10 calBP (1995 ± 10 calBC). Quat. Sci. Rev. 3 0 (9–
sterdam, pp. 21–30. 10), 1035–1046.
Linoli, A., 2005. Twenty-six centuries of reclamation and agricultural improvement on the The chora of metaponto 3, archaeological Field survey Bradano to Basento. In: Carter, J.,
Pontine marshes. Integrated Land and Water Resources Management in History, Prieto, A. (Eds.), Institute of Classical Archaeology, Packard Humanities Institute vol.
Siegburg, pp. 27–55. 1–4. University of Texas Press, Austin.
Magny, M., Combourieu Nebout, N., 2013. Holocene changes in environment and climate van Joolen, E., 2003. Archaeological Land Evaluation, a Reconstruction of the Suitability of
in the central Mediterranean as reflected by lake and marine records. Clim. Past 9, Ancient Landscapes for Various Land Uses in Italy Focused on the First Millennium BC
1447–1454. (PhD thesis) Rijksuniversiteit Groningen.
Mensing, S.A., Tunno, I., Sagnotti, L., Florindo, F., Noble, P., Archer, C., Zimmerman, S., van Leusen, P.M., 2000. Pattern to Process, Methodological Investigations into the Forma-
Pavón-Carrasco, F.J., Cifani, G., Passigli, S., Piovesan, G., 2015. 2700 years of Mediterra- tion and Interpretation of Spatial Patterns in Archaeological Landscapes (PhD disser-
nean environmental change in central Italy: a synthesis of sedimentary and cultural tation) University of Groningen.
records to interpret past impacts of climate on society. Quat. Sci. Rev. 116, 72–94. van Leusen, P.M., Attema, P.A.J., 2003. Regional archaeological patterns in the Sibaritide:
Osanna, M., 1992. Chorai coloniali da Taranto a Locri: documentazione archeologica e preliminary results of the RPC field survey campaign. Palaeohistoria 43 (44),
ricostruzione storica. Istituto poligrafico e zecca dello stato, Libreria dello stato, Roma. 397–416.
Pacciarelli, M., 2000. Dal Villaggio alla Città, la svolta protourbana del 1000 a. C. nell'Italia Vanzetti, A., 2013. Sibari Protostorica. In: G., D., Masneri, T. (Eds.), Sibari, Archeologia,
tirrenica, Firenze. Storia, Metafora, Castrovillari, pp. 11–33.
Rainey, F.G., Lerici, C.M., 1967. The Search for Sybaris, 1960–1965, Roma. Vita Finzi, C., 1969. The Mediterranean Valleys: Geological Changes in Historical Times.
Renes, H., Piastra, S., 2011. Polders and politics: new agricultural landscapes in Italian and Cambridge University Press, Cambridge.
Dutch wetlands, 1920s to 1950s. Landscapes 1, 24–41. Walsh, K., 2014. The Archaeology of Mediterranean Landscapes, Human-Environment In-
Roberts, N., 1998. The Holocene, an Environmental History. Blackwell Publishers, Oxford. teraction from the Neolithic to the Roman Period. Cambridge University Press,
Roovers, T., 2011. Landscape of the Sibaritide, a Reconstruction Attempt for the Coastal Cambridge.
Plain in the 7th to 4th Century BC, Researchmaster Graduate Thesis. University of Walsh, K., Attema, P.A.J., de Haas, T.C.A., 2014. The Pontine marshes (Central Italy): a case
Groningen. study in wetland historical ecology. BABesch 89, 27–46.
Sevink, J., Remmelzwaal, A., Spaargaren, O.C., 1984. The Soils of Southern Lazio and Adja- Yoffee, N., 2005. Myths of the Archaic State, Evolution of the Earliest Cities, States, and
cent Campania. Universiteit van Amsterdam, Asmterdam. Civilizations. Cambridge University Press, Cambridge.
Sevink, J., van Bergen, M.J., van der Plicht, J., Feiken, H., Anastasia, C., Huizinga, A., Maroto,
J., Vaquero, M., Arrizabalaga, Á., Baena, J., Baquedano, E., Jordá, J., Julià, R., Montes, R.,

Please cite this article as: Attema, P., Sedimentation as geomorphological bias and indicator of agricultural (un)sustainability in the study of the
coastal plains of Sout..., Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.07.024