Version classiqueVersion mobile
OpenEdition Books

Géoarchéologie des îles de la Méditerranée

Matthieu Ghilardi

Partie 4. Deltas, lagunes et marais : des interfaces propices à l'implantation des sociétés humaines / Deltas, lagoons, and marshes as suitable environments for human habitation

Holocene Fluvial Dynamics and Geoarchaeology on Mediterranean Islands

Brown Tony et Walsh Kevin


This paper presents a review and preliminary metadata analysis, of Holocene fluvial archives from Mediterranean islands by island group. For geomorphological reasons islands have not received as much attention from fluvial geomorphologists as the continental areas surrounding the Mediterranean. However, the studies that have been undertaken, suggest that they can reveal both climatic and cultural drivers of erosion and sedimentation from relatively small catchments. Although more research is needed, the preliminary analysis published here, suggests that during the mid-Holocene ca. 3000 BC to 800 BC, and the later Holocene, ca. 400 BC to 1600 AD, many islands experienced elevated rates of fluvial activity. Both periods, and particularly the earlier period are the result of climatic fluctuations combined with agriculture and cleared landscapes. At present the Little Ice Age signal appears to be less strong than it is on the continent although it has been recognised in Corsica and Crete. Due to their bounded nature and generally distinct cultural history islands do, however, offer excellent opportunities to research the non-linear relationships between culture, climatic change and fluvial response.

Texte intégral


1The vast majority of research studies of fluvial history, including both erosion and alluviation, have been on the continental regions bordering the Mediterranean Sea, rather than on Mediterranean Islands. The reasons are obvious, as, out of over 180 islands only a few are large enough to have well-developed fluvial systems with floodplains and depositional reaches, in addition to bedrock-dominated headwaters. Indeed most catchments can be regarded as headwaters of highland massifs surviving above the high sea levels of this, and previous, interglacial periods, and only a few islands have stream systems that achieve higher than first-order stream networks. This review and preliminary metadata analysis will compare published records from the larger Mediterranean Islands and attempt to relate them to the better-known studies of marginal continental rivers draining into the Mediterranean, and to climate records from the Mediterranean and North Atlantic. For the Holocene the record of human occupation of the islands, and human impact on island floras and faunas, is also an essential context for the interpretation of fluvial records. In practice the biotic record, including humans, and fluvial archives, are not independent, as in many cases evidence of the former comes from the later. Indeed much of the impetus for studying fluvial deposits on Mediterranean islands originated from a desire to test competing genetic theories of alluviation that emerged in the 1960s by Vita-Finzi (1969), van Andel et al. (1986), Rackham (2008) and others – a theme that will be returned to in the discussion. It must also be remembered that the Mediterranean environment produces a variant of the standard genetic model alluvial model (cf. Nanson and Croke, 1992; Brown, 1997) that is characterised by cut – and – fill sequences (Vita-Finzi, 1969; Macklin and Woodward, 2009).

The Western Mediterranean

2An increasing amount of data on fluvial dynamics is available for the Balearic Islands, Corsica and Sardinia although it is highly variable both in time and space. Of the four Balearic Islands (Menorca, Mallorca, Ibiza and Formentera) only Mallorca is large enough to contain second order fluvial networks, however, all channels are ephemeral and generally seasonal (e.g. the River Sóller).

3The Balearic islands (Figure 1) have important geomorphic linkages between karstification, neotec-tonics and fluvial network development (Gelabert et al., 2003; Segura et al., 2012; Ginés et al., 1995) and observable interactions between neotectonics, relative sea level and bedrock channels – as seen on Mallorca (Capó and Garciá, 2014). Mallorca is also known for Quaternary palaeosols (Wagner et al., 2014) developed on alluvial fans in the southwest of the island, and complex polygenetic sequences from the north- eastern coast (Rose et al., 1999). These sequences reveal high rates of energetic fluvial activity in MIS 2 due to rapid runoff and low vegetation cover. This response is typical of Mediterranean islands where small steep headwater catchments have created thick alluvial fan sequences that have normally been incised during the Holocene (Lewin et al., 1995; Rose and Meng, 1999). In contrast to research on long-term channel dynamics, far less research has been under-taken on Holocene alluviation and geoarchaeology in Mallorca, However, the island was first settled by humans ca. 6000 BC, and the earliest structures (talayots) were constructed around 2000 BC. Later, the Balearics became a key stop on Phoenician trade routes (Abulafia, 2002; Ninyerola et al., 2007). Pollen analyses have revealed vegetation disturbance dating back to the first inhabitants with deforestation, probably by the burning of oak, and again during the Roman period there is an increase in deforestation, overgrazing, and the spread of grassland and a loss of biodiversity (Carrion et al., 2010). Disturbance can be seen in both lowlands and highlands, but to different extents.

4However, recent work at a number of sites has suggested that the dominant effect on vegetation in both Menorca and Mallorca may have been climate change, especially Bond events reflecting the strong influence of the North Atlantic on these islands (Burjachs et al., 2016). As with most other alluvial systems discussed in this paper, there is clear evidence of incision over the last 100 to 200 years (pers. obs.) which is probably related to the combined effect of the Little Ice Age (LIA) and afforestation. In the 1700s AD, the Balearics were one of the most wooded areas in Spain with contemporary surveys recording 2.38 million holm oaks, 4.71 million pines and 100 000 other trees within Mallorca alone (Grove and Rackham, 2001). This is, however, far less extensive than the current tree coverage as Mallorca now has twice as many oaks and six times as many pines (Grove and Rackham, 2001). The increase in forest coverage across the Balearics is due to the decline in the use of the timber for shipbuilding and industry.

Figure 1: Map of the Mediterranean Islands and fluvial study locations discussed in this paper. 1: SE Spain (Schulte, 2002); 2: S Etruria (Brown, 1997; Brown and Ellis, 1996); 3: Tunisia (Faust et al., 2004); 4: Boetia (Bintliff, 2002); 5: Macedonia (Lespez, 2003); 6: Central Greece (van Andel et al., 1990); 7: Peloponese (Fuchs, 2007); 8: Jordan (Shuldenrein, 2007).

5The trade history of the Balearics is an important factor in its anthropogenic disturbance history, and trading history is an important factor in environ- mental change across Europe and particularly for small Mediterranean islands (Grove and Rackham, 2001). As yet this environmental history has not been quantitatively related to alluvial valleys, which do exist, especially in Mallorca.

6Corsica and Sardinia are two of the largest islands, and until c. 14 000-15 000 years ago were connected forming “Corsardinia” (Conchon, 1978), which at the last glacial maximum (LGM) had a landmass of approximately 35 000 km2 making it the largest island ever to have existed in the Mediterranean. Both islands have a number of rivers with basin and gorge topography, which includes reaches with extensive alluvial terraces of both Pleistocene and Holocene age. Corsica has been inhabited since at least the early Mesolithic (ca. 7000 BC, Patton, 1996; Llobera et al., 2010) and more recently it has been argued, based upon finds on the river terraces in the Aléria area, that there may have been Palaeolithic occupation (Asta et al., 2015). Corsica is particularly well known for its Chalcolithic Torrean (tower) Culture, which has architectural similarities to the Nuraghi in Sardinia and the Talayotic culture in the Balearics, although there is no strong evidence of direct connections between them (Patton, 1996). Unusually for a Mediterranean island, Corsica has a good record of environmental change from high mountain lakes; a consequence of glaciation. Cores from Lake Creno reveal a high frequency of fire events throughout the Holocene (Leys et al., 2013), as well as natural vegetation change; tree clearance for agriculture, timber and charcoal, and live- stock grazing especially by goats (Reille et al., 1999). Due to the high relief, Corsican valleys have well-developed fluvial systems, which on the eastern and southern sides of the island include both Pleistocene and Holocene terrace systems which are just starting to provide geoarchaeological data (Ghilardi et al., 2016a; Vella et al., 2016). In the north west of the island a geomorphological study has revealed 7 flood phases since the mid-18th century (Hewitt, 2002; Macklin and Woodward, 2009) and lichenometry has revealed an incision phase in canyons on Corsica (Gob et al., 2003).

7Despite its proximity of Corsica, Sardinia is not only topographically very different to Corsica, but also unusual amongst Mediterranean islands due to a relatively low relief for its size (Figure 2; Table 1). This difference is of geological origin, as Sardinia is a mixture of rock types including granites but also basalt, schists, sandstone and dolomitic limestone. As with Corsica, the date of the earliest human occupation is a matter of much debate, but there is worked stone and bone tools alongside butchered Prolagus sardus at Cordeddu Cave (ca. 11000 BC) and a human maxilla alongside butchered Megaloceros cazioti at Lenaittu Cave at 12600 BC, and possibly as old as 15000 BC (Hofmeijer et al., 1989). Tykot and Andrews (1992) identified “Clactonian type” stone tools at Sa Pedrosa-Pantalinu and Cordebbu cave, but these may well be late Upper Palaeolithic (post LGM) or even early Mesolithic. The occurrence on Sardinia of valuable mineral resources, including obsidian (Mount Arci), copper, tin, iron, and lead has driven human occupation and landscape change, especially in the lowlands. Sardinia is famous for its Nuragic Civilisation, which began in the Bronze Age around 1800 BC (Webster, 1991 and 1996). Th culture is characterised by the monumental “Nuraghe” struc- tures, 7 000 of which remain throughout the land- scape, generally located by natural springs. Recent studies of alluvial deposits near the Nuragic site of Sa Osa show coastal progradation due to increasing fluvial sedimentation, and interestingly both vine (Vitis) pollen and pips (Melis pers. com. 2016). Overall, Sardinia has experienced a complicated and highly patterned record of environmental change due to its varied geology and relief. This atypical history includes the Holocene development of climate and environments favourable to the malaria vector (Sallares, 2006) which resticted development. In AD 170, Pausanias labelled Sardinia’s climate as “unwholesome and pestilential”, and the Emperor Nero banish political enemies with the expectation of them dying of malaria. The disease continued to be a major issue for human development and expansion on the island until 1951 (Tognotti, 1998).

Figure 2: Plot of island area vs relative relief with a power function trend-line.

Table 1: Largest Mediterranean island statistics and geology where Koppen classification is in parentheses it denotes a minor classification ie for parts of the island(s).

The Central Mediterranean

8As defined here, this region includes Sicily, Malta, Gozo, Crete, and the Greek Island groups. These latter groups include the Ionian Islands, the NE and mid-Aegean Islands, the Cyclades, the Sporades Islands, the Saronic Gulf islands, and the Dodecanese Islands (Figure 1).

9Being the largest island in the Mediterranean, Sicily has larger river catchments and greater relative relief than any other island. The island is also separated from southern Italy by the Straits of Messina, which at their narrowest are only 3.1 km wide, although this is a deep channel with strong currents. Despite the relative dearth of alluvial research on Mediterranean islands, we should not forget that an early example of alluvial geoarchaeological work was that presented from Sicily by Judson (1963) that prefigured much of the later work by Vita-Finzi (1969). He articulated questions that are still important today i.e. whether changes in the landscape were caused by people or “by some non-human cause”, and that such changes may have had an impact on human exploitation of a given landscape (Judson, 1963). He dated the river terraces along the Gornalunga Valley in central Sicily, and suggested a post-325 BC date for the 8–10 m terraces based on the identification of a burial within the terrace deposits. Judson demonstrated that prior to the Greek period burial; the valley was 10 m deeper than it is today. More recently, Ayala and French (2005) have used erosion model- ling, specifially a version of the universal soil loss equation (USLE), in order to investigate Holocene alluviation in north-central Sicily. When combined with archaeological and literary data they argue that hill-slope stability thresholds were exceeded because of clearance undertaken for an increase in intensive pastoralism during the Roman Period. In contrast the alluvial sequence of the Torcicoda River in Central Sicily shows a major phase of alluviation began in the Neolithic period before ca. 5500 BP (Macklin and Woodward, 2009). With the aim of disentangling climatic drivers from human drivers of vegetation change in southern Sicily, Calò et al. (2012) used a nested-lake approach with pollen and ostracod proxies. Their conclusions are that despite hydrological changes, probably due to an increase in winter rain- fall, the evergreen forest persisted until its exploitation by Greek settlers (ca. 800 BC) who cleared the forest as a part of agricultural intensification. This involved the use of fire, and resulted in an expansion of the maquis, garrigue and grasslands, along with an expansion of the only palm native to the northern Mediterranean, Chamaerops humilis (Tinner et al., 2009). These results have been confirmed by research in western Sicily on the Chuddia and Belice catchments (Heinzel and Kolb, 2011). Infilled channels on alluvial fans show that the most marked period of change was in the Roman period when fans were re-activated and old land surfaces buried by gravels. We might expect the Sicilian record to correlate with records from Basilicate (ca. 300 km to the north) and this would appear to be the case with a pronounced increase in fluvial activity 2350-1850 BP (Piccareta et al., 2011). However, this record suffers from the radiocarbon probability density function problem in that the wetter periods appear to constitute 35% of the record; therefore, any correlation has a high probability of being coincidental.

10Malta and Gozo are both limestone plateaus with relatively little relative relief, and valleys formed via combined karstification and incision during low sea levels through the Pleistocene (Segura, 2012). There are no permanent rivers, and a notable feature of these islands’ geomorphology is a network of dry-valleys (Haslam, 1998) which contain colluvial deposits. It is not surprising therefore that relatively little alluvial research has been undertaken on these islands. However, their unique cultural history and associated built structures, particularly the elaborate Neolithic Temple complexes, has attracted much archaeological research, particularly into the question of the demise of the Tarxian culture and the possibility of abandonment of the islands prior to re-occupation in the Bronze Age. In one of the few locations where a sedimentary archive has been preserved, is in a ria (marine flooded-valley) off the north west coast where Marriner et al. (2012) have shown a fluvio-marine sequence spanning the last 7500 years. They report intensified human impact on the vegetation beginning ca. 5300 cal. BC, associated with the Tarxian culture, and rapid human-induced sedimentation throughout the Neolithic. This “cultural degradation” model is being examined by the current FRAGSUS project (Malone, 2015).

11The relatively small size of most of the Greek Islands has meant that only a few studies of fluvial and slope histories have taken place within in a few small catchments, and more often than not, this research is linked with archaeological projects. One of the most sophisticated studies has been the study of agriculturally terraced slopes on the island of Amorgos in the Cyclades by French and Whitelaw (1999). The Cyclades are of course particularly interesting, as the complex Bronze Age culture here was at least sustained by, if not built on, agriculture (Renfrew 1972). Based on soil micromorphology, French and Whitelaw (1999) argue that the sub-terrace red soil was a reworked palaeosol heavily disturbed by slope erosion processes which started with the clearance of the pre-Bronze age vegetation. Unfortunately the lack of lakes, or other sediment traps, from which fossil pollen could be derived has severely limited the record of Holocene vegetation change by both humans and climate change. However, more recently alluvial studies have been undertaken on Corfu in the Ionian islands by Berger and Guilaine (2009). A long sequence at Sidari has palaeosols and alluvial deposits spanning the Mesolithic through the Neolithic with three phases of elevated alluviation (Berger et al. 2015). This record can be compared to the Adriatic climate record with which it shares some similarities (Combourieu-Nebout et al., 2013).

12Despite the well-known archaeology of the North Eastern Aegean Islands (Thassos, Limnos, Lesvos, Chios, etc.), the Sporades (Skiathos, Skopelos, Skioros), the Saronic Gulf islands (Hydra, Aegina, Salamis, etc.), the Cyclades (Paros, Naxos, Milos, etc.) and the Dodecaneses (Rhodes, Samos, Cos, etc.), the only islands which may have significant alluvial potential are Rhodes and Euboea (off the east coast of the Greek mainland, Figure 1). Current archaeological studies at Eretria on Euboea are revealing that the earliest city in the 9th to the 7th centuries BC developed on a deltaic floodplain, and that the river required embanking due to frequent flooding of the city which was geomorphologically constrained (Ghilardi et al., 2016b). The obvious exception to this lack of fluvial studies in the Central Mediterranean is Crete. The size of the island, and its tectonic activity have both promoted studies of fluvial forms and dynamics. Maas et al. (1998) have using slack-water deposits derived a palaeohydrological record for the Omalos system which shows 6 major units, the first of which (U6) is LGM, followed by a prolonged phase of stability before renewed sedimentation in the Little Ice Age (LIA) and three smaller units in the last 200 years. What is most notable about this record is the stability and the high magnitude of the LIA events. In a more detailed study of the post LIA period Mass and Macklin (2002) identified twelve periods of increased flooding during the last 150 years which appear to correlate with negative or declining phases of the North Atlantic Oscillation (NAO). However, earlier work by Hempel (1982; 1984) has shown localised soil erosion on Crete between 700 BC and 200 BC. Further studies may well identify earlier Holocene alluviation as Crete is exceptional in being the only Mediterranean island to have produced an agriculturally-based culture that colonised large areas of the central Mediterranean and the surrounding coasts – the Minoan culture, that was based upon an intensive stock-based economy. So river valleys in Crete have high geoarchaeological potential with an example being the Istron valley in Mirabello Region which saw early development of Proto Palatial settle- ment (Schultz-Barrick, 2007).

The Eastern Mediterranean

13The only island in the eastern Mediterranean, Cyprus, is the third largest island in the Mediterranean; highly mountainous, and possesses the most arid climate (average 300-400 mm yr-1, although precipitation is far higher in the mountains). Over the last 40 years, more fluvial studies have been undertaken on Cyprus than any other Mediterranean island. This is due to its size, relief, and colonial history. The geomorphology of Cyprus is dominated by two mountain ranges one in the south (Troodos massif) and a smaller range in the north (Kyrenia range), with a narrow piedmont zone in-between. Of particular geoarchaeological importance are the mineral resources of the island. Copper deposits (after which the island is named) are associated with volcanic massive sulphide deposits. The Cypriot ore deposits, which belong to the Troodos Ophiolite complex, are some of the most well-studied in the world (Stos-gale et al., 1997). These deposits have attracted humans from the Neolithic/Chalcolithic onwards. In the 1960s, it was thought that the first established culture on Cyprus was the aceramic culture recorded at Khirokitia (Tatton-Brown, 1999; Knapp, 2010). Although aceramic, many artefacts found here resemble Jericho-type pottery, and are Neolithic in nature. However, more recent work in a cave at Akrotiri on the south coast of the island has revealed hearths and chipped stone implements, together with the bones of pygmy elephants and some fish (Knapp, 2010). Links from the Levantine mainland were inferred from the stone chips, indicating that some form of pre-agricultural migration and settlement from the mainland. With new dates from the rockshelter at Akrotiri, Cyprus is now thought to have been inhabited from around 12000 BC and these first settlers may well have been implicated in the extinction of the pygmy hippopotamus (Simmons, 2014). Although there were observations of Quaternary fluvial deposits in the 1960s (Ducloz, 1968), the first directed study of alluviation was that by Gomez (1987) on the Lower Vasilikos Valley. Although Gomez (1987) found that the pattern of an older and younger alluvial model approximated to the Vita-Finzi model (Vita-Finzi, 1969), the younger fill comprised two units, the older of which was mid- Holocene and corresponded to the aceramic Neolithic (ca. 5800-5250 BC). In a study of several valleys in eastern Cyprus Deckers (2005) has argued for waves of Medieval and post-medieval alluviation, which he associates with drier conditions in the Byzantine period, wetter conditions in the Frankish period contemporaneous with increased agriculture, mining and smelting, and lastly fluvial deposition during the Ottoman Period that correlates with the “Little Ice Age”. In a specifically geoarchaeological and heuristic approach to the environmental history of Cyprus Butzer and Harris (2007) examined sections from five river systems in the north-central Troodos. Their study revealed important elements of fluvial inheritance and also suggested that slope stability over the last 2700 years was “at least as sensitive to climatic anomalies as it was to land use practices” (Butzer and Harris, 2007, 1950). In a more recent study of the fluvial history of the Tremithos river in south central Cyprus, Ghilardi et al. (2015) have recognised a phase of alluviation in all the profiles they studied dating from ca. 5000 to ca. 2800 BC, which spans the Late Neolithic Sotira (ca. 4,800/4,000 BC) and Late Chalcolithic (ca. 2900-2500 BC). However, given a lack of Neolithic archaeology in the catchment Ghilardi et al. (2015) associate the phase with the regional 5.2 ka dry event known from the eastern Mediterranean (Zanchetta et al., 2014).


Island biogeography and fluvial processes

14If we are to address issues of human engagements with island fluvial/alluvial processes, we need to incorporate some discussion of island biogeographical theory (MacArthur and Wilson, 1967). Although island biogeographical models have limited utility in the assessment and interpretation of socio-ecological systems on islands (Patton, 1996; Rainbird, 1999), we do need to consider how alluvial processes on islands differ from continental processes. In this section, we consider some important island biogeographical precepts and briefly assess how these might intersect with the study of fluvial dynamics and human engagements with these processes.

15One of the fundamental descriptive measures employed in island biogeography characterises the size of an island along with its distance from a continental zone. One of the problems of such island biogeographical rankings is that they fail to consider explicitly the nature and reliability of water sources; for example, one element in a revised ranking statistic might include the number of rivers and their discharge values. In addition changes in relative sea level, especially around small islands, will have important consequences for fluvial and alluvial processes. The spatial extent of many continental river systems means that these will often adapt to changes in relative sea level, however, this may not be the case on islands, and therefore the adaptation of island systems to relative sea level change is an important area for investigation. Once again, such processes are linked with biogeographical ranking, especially the size of an island, and changes in sea level will have important consequences for river behaviour on small islands. On larger islands, such as Crete, the varied nature of relative sea change (uplift to the west, and downwarping to the east) will have led to complex responses in fluvial/alluvial processes. The precise correlation of these different processes is complex, but is an essential endeavour (Butzer and Harris, 2007).

16Part of the traditional island biogeographic narrative is the notion that island communities might have been less resilient (Evans, 1973). This idea is largely founded on distance-area statistics; i.e. the distance of an island from the closest continental land mass, and the area of the island – smaller, more distant islands possessing relatively impoverished ecologies. One such “impoverished” island that has to manage with no permanent rivers is Malta. Sedimentological and mala- cological studies (Fenech, 2007) suggest that phases of aridity, including that during the Neolithic, resulted in ecological stress. In island contexts such as this, we should consider socio-ecological responses, or cultural “negative feedback” mechanisms that can promote resilience (Butzer and Harris, 2007; Butzer, 2011). Conversely negative cultural (socio-political) feed- backs may be “peculiarly” insular e.g. intensive temple building, ritual control of water – and such initiatives may not mitigate the environmental problem leading in some cases to island abandonment (Antikythera and possibly Malta). The incredible density of prehistoric monuments on Malta may also represent an investment in ritual or ideological activity designed to mitigate problems, or control societies during periods of environmental stress (Walsh, 2014).

17From cultural and economic perspectives, we might also consider how the impact of people on island alluvial systems will differ from continental areas, as many Mediterranean islands were exploited for quite specific resources. At one level, focussed niche activities allow island communities to survive periods of environmental stress, but only if these islands are integrated into networks of trade and exchange that fulfil subsistence requirements. Although agriculture was practiced on some small islands from the Neolithic onwards, certain islands emerged as key nodes in the Mediterranean network (Horden and Purcell, 2000). Melos emerged as one important node due to obsidian extraction and trade (Renfrew and Wagstaff, 2005; Tykot, 2004; Tykot and Ammerman, 1997). Such an example might seem irrelevant vis a vis the study of alluvial histories, however, we need to consider how specific activities might leave signals in the alluvial sedimentary record. One alluvial record for Melos does suggest landscape degradation, but well after the Neolithic: pottery sherds covering the period ca. 600 BC to 500 AD were found in one alluvial unit, and late protohistoric population growth was suggested as the root cause (Davidson and Tasker, 1982).

Comparison with continental records and the North Atlantic and Mediterranean climate records

18Fluvial archives are the result of combined changes in land use and human activities in catchments, and until recently natural, climatic changes and associated vegetation response. There are several key continuous climate records that island history can be compared to, at both a regional and Mediterranean- wide scale. For the Eastern Mediterranean the δ18O speleothem-derived record from Soreq cave, and six long lake records have been used by Roberts et al. (2011) to produce a stacked isotope-climate record (Figure 3). This shows the well-known drying of the Mediterranean region in the mid-Holocene as well as wet and dry phases. The most pronounced dry phase being 5300-5000 BP, 4500-3900 BP and 3100-2800 BP, which are correlated by Roberts et al. (2011) with regional cultural changes, such as cultural changes in the early Bronze Age, and increased urbanism and irrigation or the “collapse” period at the end of the Bronze Age. However, all these cultural events are multi-causal and involve social changes, which can include migration and conflict, including for example the appearance and effects of the so-called “Sea Peoples” in the late Bronze Age (Kaniewski et al., 2010). This integrated record, along with the more conspicuous signals from the Central and Western Mediterranean, which include the 8.2 ka, 5.2 ka and 4.2 ka events, were driven by northern hemisphere and global climate. These processes are reflected in the ice-rafted debris (IRD) and oceanic sediment records (Bar-Matthews and Ayalon, 2011) and can be correlated westwards to Italy with subtle changes (Magny et al., 2012; Zanchetta et al., 2014), and again to the western Mediterranean, which is even more closely coupled to North Atlantic sea surface temperature records (Fletcher et al., 2012). However, there is also the North-South palaeoclimatic gradient that particularly reflects the movement north and south of westerly storm tracks (Magny et al., 2013). Of critical signifiance for both alluviation and past societies was the occurrence of winter cyclones, with Atlantic cyclones decreasing eastwards across the Mediterranean whilst for the Eastern and Northern Mediterranean cyclones are predominantly are gene-rated inside the Mediterranean (Zanchetta et al., 2014). It is significant that these records contrast with the integrated fluvial history of the Nile which is largely driven by rain- fall in the Intertropical Convergence Zone (ITCZ, White Nile) and the Northern Hemisphere summer monsoon (Blue Nile; Macklin et al ., 2015). We can therefore compare these relatively independent records with the island fluvial records to date (Figure 3). Apart from showing how poor the island record is, there are two broad periods of agreement between these records; the first is alluviation especially in the Eastern Mediterranean from the mid to late Holocene ca. 3000 BC to 800 BC, and alluviation in the Central Islands ca. 200 BC to 1600 AD. Nevertheless, within these broad periods there were periods of reduced fluvial activity, particularly at the decadal scale. Although there does appear to be some correspondence with climate records from the Mediterranean region and Northern Atlantic, but not the Nile, there is also a strong association with increasing human impact especially from the Bronze Age to the end of the Greco-Roman Period.

Figure 3: Summary table of Holocene alluviation phases. Black is strong evidence of increased alluviation, grey moderate evidence, blue is wetter periods, yellow is drier periods and * is multiple alluviating events on the decadal scale. From many sources including those in the text and Schulte et al. (2002) and Thorndycraft and Benito (2006).
Long record are GISP2 (NOAA, GISP2) and the Soreq cave record (Bar-Matthews and Ayalon, 2011).

Key geoarchaeological questions in island fluvial geoarchaeology

19The review presented above demonstrates that the number of detailed alluvial histories of Mediterranean islands is still limited; consequently, a detailed and meaningful correlation of island alluvial histories with their continental counterparts is difficult. However, some general themes and issues are highlighted by this overview. We can identify some common or shared processes, and we can make suggestions regarding possible strategies of future research in these all important, fragile environments. The Mediterranean islands are increa-singly producing data on alluviation history that can be compared to the coastal regions. At the scale of hundreds of thousands of years, climate-related changes in catchment hydrology and vegetation cover are clearly the primary control of large-scale (catchment wide) sedimentation across the Mediterranean region (Macklin et al., 2002). However, although at the millennial scale nearly all studies show increased alluviation in the mid-later Holocene this record is clearly diachronous (time transgressive), as would be expected due to the variation in human activities, island climates, and geologies. Indeed, it mirrors the diachrony from large continental areas such as the Balkans or Italy (Brown, 1997). Indeed the central Italian area is similar in size to the larger Mediterranean islands and shows how regional diachrony could exist on such islands caused by variations in relief and land use history. As Butzer has observed “climatic impulses may have favoured geomorphic activation, but the variable timing, and the unprecedented extent and scope of these slope and valley changes… remain inconceivable without a pre-eminently cultural impetus” (Butzer, 1980). On Cyprus there is substantial variation in the nature and timing of alluvial/fluvial events and environmental history requires multiple data sources and multiple readings (Butzer and Harris, 2007). Unsurprisingly, large island fluvial systems, such as that on Cyprus, are difficult to compare with those from smaller islands, therefore, as noted above, some kind of ranking may well facilitate the ways in which we characterise and assess island fluvial histories. For example, island size will usually limit stream orders, and the propensity for changing channel sinuosity might be related to island size.

20Indeed, in many ways the complexity of the alluvial record mirrors the complexity of attitudes in the Classical world to the natural environment, deforestation and alluviation (Hughes, 1983). Thucydides noted the great quantities of timber on Sicily, whereas timber supplies on Rhodes were exhausted, and on other islands, such as Cyprus, forests were protected by management (Hughes, 1983). If island forest cover varied during this period, so would any response to climate change, therefore, land use and climate can never be truly independent factors in our understanding of alluvial sequences. Moving beyond debates about the relative role of natural versus human factors in causing catchment erosion and alluviation there are several key archaeological questions where Mediterranean islands provide the ideal natural laboratories. One is the question of human mobility, and technological sophistication in the late Pleistocene and early Holocene. Mediterranean island archaeology has already demonstrated the uniqueness of many island cultures and economic practices. Insular cultural and economic processes often “niche” or highly specialised, such as that on Melos. As Butzer (2011) observes, we need to consider sociocultural responses to environmental variability, which means that we must consider the potential impact on fluvial/alluvial processes of these specific activities and practices; economic strategies that go beyond the typical agro-pastoral systems so often cited as the core, predictable impacts on landscape.

21The search for the date of the earliest human arrivals onto Mediterranean islands has quite unsurprisingly fascinated archaeologists and geologists alike. The high stakes in such studies also produces both debate and elements of controversy. Obvious problems include the normal first occurrence problem of absence of data not being evidence of absence. A second problem is the complexity of geomorphology rendering different islands emergent at different times and the the third is the possibility of visitation rather than habitation (Ferentinos et al., 2012). Table 2 summarises the current position and debate that suggests that at least some of the islands were inhabited in the Late Palaeolithic. The most significant claims for Palaeolithic inhabitations come from the islands that can never have been connected to the coast during the late Pleistocene, specifi ally Crete and Cyprus. There have been claims for early Palaeolithic (ca. 130000 BP) presence on Crete for many years (Runnels and Van Andel, 1993; Runnells, 2014) and recently credible Palaeolithic tools have been reported from the island of Gavdos 21 km to the south of Crete (Kopaka and Matzanas, 2009). In fact, although Crete lies 160 km from the Greek mainland today in the mid-Pleistocene the islands of Kythera and Antikythera formed a peninsular that reached to within 10 km of Crete (Sakellariou and Galanidou, 2015). It is also likely that the best potential source for further evidence of Pleistocene occupation of Crete will come from fluvial deposits. In relation to this debate, Rainbird (1999) has shrewdly observed that Mediterranean islands have been overemphasized as “bounded landscapes” rather than tips of a “broader seascape”.

Table 2: Current estimates of the earliest dates of human/hominin occupation of selected Mediterranean islands with ancillary data.
Updated from Mouillot et al. (2008), * estimate of distance at Pleistocene lowest RSL.


22This review has focussed on the larger Mediterranean islands partly because on many of the smaller islands, alluvial/fluvial activity is minimal, or non-existent, and subterranean hydrology of greater importance. For example, the geological structure of Naxos (one of the “large” small islands) means that there are many natural springs across the island (Dalongeville and Rénault-Miskovsky, 1993), as is also the case for Andros and Tenos (Broodbank, 2000), but not all small islands possess natural springs. On islands with large enough fluvial systems, an increasing number of studies are now using sediment-based dating techniques (e.g. OSL). Alluvial geoarchaeologies are either explicitly or implicitly underpinned by a notion of continuous diachrony – i.e. the sedimentary records that we study represent unbroken time series with phases of stability and instability. Alluvial, and by default, fluvial histories are comprised of ruptures; extreme floods that truncate deposits, or phases of aridity. For example, the floodplain at Lefkosia, Nicosia, Cyprus is characterised by an extensive series of flood deposits, with palaeosols representing hiatuses between flooding events/phases. In some cases, these ruptures are caused neither by climate change or human impact on the hydrological system as for example is the change from an aggrading river system to an entrenched or incising system caused by tectonic activity. These ruptures or sudden unpredictable events will test human communities to a greater extent than gradual climate change, especially on smaller islands with limited water resources (Newell et al., 2004). The remarkable diversity of Mediterranean islands in size, relief, geology, and geographic position within the Mediterranean is bound to cause a spatially complex response to all forcing factors. It is not possible, even using modelling to disentangle them as they are genetically related. However, the alluvial archives of Mediterranean islands are only just starting to be systematically researched and the bounded nature of island environments confers certain advantages that should be exploited. The most obvious is that population densities and cultural histories can be better defined and ultimately it is the linkage of such data with palaeoclimatic proxies and alluvial lithostratigraphy that will underlie an alluvial geoarchaeology of Mediterranean islands.



ABULAFIA D., A Mediterranean emporium: the Catalan kingdom of Mallorca, Cambridge, Cambridge University Press, 2002, 320 p.

ASTA V., ENEI F., FATUCCI M., BIDDITTU I., Preliminary results of archaeological surveys conducted between Aleria and Le Strette (Eastern Corsica). Abstract of paper given to GEOMEDISLANDS – The Geoarchaeology of Mediterranean Islands, Corsica, 30th June- 2nd July 2015.

ALCOVER J.A., «The first Mallorcans : Prehistoric colonisation in the Western Mediterranean», Journal of World Prehistory, 21, 2008, p. 19-84.

AYALA G., FRENCH C., «Erosion Modeling of Past Land-use Practises in the Fiume di Sotton di Troina River Valley, North-Central Sicily», Geoarchaeology: An International Journal, 20, 2005, p. 149-167.

BAR-MATTHEWS M., AYALON A., «Mid-Holocene climate variations revealed by high-resolution speleothem records from Soreq Cave, Israel and their correlation with cultural changes», The Holocene, 21, 1, 2011, p. 163-171.

BERGER J.F., GUILAINE J., «The 8200 cal. BP abrupt environmental change and the Neolithic transition: A Mediterranean perspective», Quaternary International, 200, 2009, p. 31-49.

BERGER J.F., COMBOURIEU-NEBOUT N., DELHON C., ERTLEN D., BARRA A., METALINOU G., GUILAINE J., PHILIPAKIS G., GUILBERT R. «Les enregistrements continentaux des changements climatique rapides de l’Holocene ancient/moyen sur L’île de Corfou (Grèce). Correspondance avec les archives marines de l’Adriatique et interactions avec le développement et la mobilité des derniers chasseurs et des premiers paysan néolithiques», Geomedislands - The geoarchaeology of Mediterranean Islands, Abstracts, 2015.

BINTLIFF J., «Time, process and catastrophism in the study of Mediterranean alluvial a history: A review», World Archaeology, 33, 2002, p. 417-435.

BROODBANK C., An Island Archaeology of the Early Cyclades, New York, Cambridge University Press, 2000, 436 p.

BROWN A.G., Alluvial Geoarchaeology, Cambridge, Cambridge University Press, 1997, 404 p.

BROWN A.G., ELLIS C., «People, climate and alluviation: theory, research design and new sedimentological and stratigraphic data from Etruria, Italy», Papers of the British School in Rome, 63, 1996, p. 45-73.

BURJACHS F., PÉREZ-OBIOL R., PICORNELL L., REVELLES J., SERVERA-VIVES G., YLLR.,»Changements environnementaux et colonisation humaine des Îles Baléares : effet sur la végétation», in Ghilardi M., Géoarchéologie des îles de Méditerranée, Paris, CNRS Éditions, 2016 (ce volume).

BUTZER K.W., «Holocene alluvial sequences: problems of dating and correlation», in Cullingford R.A., Davidson D.A., Lewin J., Timescales in Geomorphology, Chichester, John Wiley & Sons, 1980, p. 131-142.

BUTZER K.W., «Geoarchaeology, climate change, sustainability: a Mediterranean perspective», The Geolological Society of America Special Paper, 476, 2011, p. 1-14.

BUTZER K.W., HARRIS S.E., «Geoarchaeological approaches to the environmental history of Cyprus: explication and critical evaluation», Journal of Archaeological Science, 34, 2007, p. 1932-1952.

CALÒ C., HENNE P., CURRY B., MAGNY M., VESCOVA E., MANTIA T., PASTA S., VANNIÈRE B., TINNER W., «Spatio-temporal patterns of Holocene environmental change in southern Sicily», Palaeogeography, Plaeoclimatology, Palaeoecology, 323-325, 2012, p. 110-122.

CAPÓ A., GARCÍA C., «Análises e interpretación de los profiles longitudinales en la red fluvial del sur y sureste de Mallorca», Cuadernos de Investigacíon Geográfica, 40, 2014, p. 147-167.

CARRIÓN J. S., FERNÁNDEZ S., GONZÁLEZ-SAMPÉRIZ P., GIL-ROMERA G., BADAL E., CARRIÓN-MARCO Y., BURJACHS F., «Expected trends and surprises in the Lateglacial and Holocene vegetation history of the Iberian Peninsula and Balearic Islands», Review of Palaeobotany and Palynology, 162, 3, 2010, p. 458-475.

COMBOURIEU-NEBOUT N., PEYRON O., BOUTROUMAZEILLES V., GORING S., DORMOY I., JOANNIN S., SADORI L., SIANI G., MAGNY M., «Holocene vegetation and climate changes in the central Mediterranean inferred from a high-resolution marine pollen record (Adriatic Sea)», Climate of the Past, 9, 2013, p. 2023-2042.

CONCHON O., «Quaternary Science in Corscia (France)», Quaternary Research, 9, 1978, p. 41-53.

COSTA L.J., Corse Préhistorique: Peuplement d’une île et modes de vie des sociétés insulaires (IXe-IIe Millénaires av. J.-C.), Paris, Errance, 2004, 192 p.

DALONGEVILLE R., RÉNAULT-MISKOVSKY J., «Paysages de l’île de Naxos», in Dalongeville R., Rougemont G., Recherches dans les Cycaldes, Résultats des travaux de la RCP 583, Maison de l’Orient et de la Méditerranée, Lyon, Lyon, 1993, p. 21-32.

DAVIDSON D., TASKER C., «Geomorphological Evolution during the Late Holocene», in Renfrew C., Wagstaff M., An Island Polity, the Archaeology of Exploitation in Melos, Cambridge, Cambridge University Press, 1982, p. 82-94.

DECKERS K., «Post-Roman History of River Systems in Western Cyprus: Causes and Archaeological Implications», Journal of Mediterranean Archaeology, 18, 2, 2005, p. 155-181.

DUCLOZ C.,» Les formations quaternaires de la région de Klepini (Chypre) et leur place dans la chronologie du Quaternaire Méditerranéen», Archives Scientifiques, Genève, 20, 1968, p. 123-198.

EVANS J.D., «Islands as laboratories for the study of culture process», in Renfrew C., The Explanation of Culture Change: Models in Prehistory, London, Duckworth, 1973, p. 517-520.

FAUST D., ZIEHOFER C., BAENA ESCUDERO R., DEL OLMO D., «High resolution fluvial record of late Holocene geomorphic change in Northern Tunisia: Climate of human impact?», Quaternary Science Reviews, 23, 2004, p. 1757-1775.

FENECH K., Human-induced changes in the environment and landscape of the Maltese Islands from the Neolithic to the 15th Century AD. British Archaeological Reports International Series, 1682, 2007.

FER ENTINOS G., GK IONI M., GER AGA M., PAPATHEODOROU G., «Early seafaring activity in the southern Ionian Islands, Mediterranean Sea», Journal of Archaeological Science, 39, 2012, p. 2167-2176.

FLETCHER W.J., DEBRET M., SANCHEZ GOÑI M., «Mid-Holocene emergence of a low frequency millennial oscillation in western Mediterranean climate: Implications for past dynamics of the North Atlantic atmospheric westerlies», The Holocene, 23, 2012, p. 153-166.

FRENCH C.A.I., WHITELAW T.M.,»Site erosion, agricultural terracing and site formation processes at Markiani, Amorgos, Greece: The micromorphological prespective», Geoarchaeology, 14, 1999, p. 151-189.

FUCHS M., «An assessment of human versus climatic impacts on Holocene soil erosion in NE Peloponnese, Greece», Quaternary Research, 67, 2007, p. 349-356.

GELABERT B., FORNÓS J.J., GÓMEZ-PUJOL L., «Geomorphological characteristics and slope processes associated with different basins: Mallorca (Western Mediterranean)», Geomorphology, 52, 2003, p. 253-268.

GHILARDI M., CORDIER S., CAROZZA J-M, PSOMIADIS D., GUILAINEJ., ZOMENIZ., DEMORYF., DELANGHE- SABATIER D., VELLA M.A., BONY G., MORHANGE C., «The Holocene fluvial history of the Tremithos river (south central Cyprus) and its linkage to archaeological records», Environmental Archaeology, 20, 2015, p. 184-201.

GHILARDI M., ISTRIA D., CURRAS A., VELLA C., DUSSOUILLEZ P., CREST Y., COLLEU M., «Approche géoarchéologique de l’évolution des paysages dans la basse vallée du Sagone (Corse) depuis l’âge du Bronze jusqu’à l’époque génoise», in Ghilardi M., Géoarchéologie des îles de Méditerranée, Paris, CNRS Éditions, 2016 (ce volume).

GHILARDI M., MÜLLER-CELKA S., THEURILLAT T., FACHARD S., VACCHI M., «Évolution des paysages et histoire de l’occupation d’Érétrie (Eubée, Grèce) du Bronze ancien à l’époque romaine», in Ghilardi M., Géoarchéologie des îles de Méditerranée. Paris, CNRS Éditions, 2016b (ce volume).

GINÉS J., «Mallorca’s endokarst: The Speleogenethic mechanisms» in Ginés A., Ginés J., Karst and Caves in Mallorca, Edins, 203, 1995, p. 71-86.

GOB F., PETIT F., BRAVARD J.P., OZER A., GOB A., «Lichenometry application to historical and subrecent dynamics and sediment transport of a Corsican stream (Figarella River – France)», Quaternary Science Reviews, 22, 2003, p. 2111-2124.

GOMEZ B., «The Alluvial Terraces and Fills of the Lower Vasilikos Valley, in the Vicinity of Kalavasos, Cyprus», Transactions of the Institute of British Geographers, no12, 1987, p. 345-359.

GROVE A.T., RACKHAM O., The Nature of Mediterranean Europe: An Ecological History, Yale University Press, China, 2001, 384 p.

HASLAM S., The River Valleys in the Maltese Islands, Malta, Malat University Press, 1998.

HEINZEL C., KOLB M., «Holocene land use in western Sicily: a geoarchaeological perspective», in Wilson L., Human Interactions with the Biosphere: The Geoarchaeological Perspective. Geological Society of London, Special Publications, 352, 2011, p. 97-107.

HEMPEL L., «Jungquärtar Formungsprozessien Südgriechenland und auf Kreta», Forschunysbendcehst Landes Nordrhein- Westphalen, 3114, 1982, p. 1-80.

HEMPEL L., «Geookodynamik im Mittelmeerrauwm Wärhend des Jungquartärs: Beobachtungeznur Frage “Menschu und/ oder Klima” in Sfidgriechenland auf Kreta», Geoőkodynamik, 5, 1984, p. 99-104.

HEWITT S., Pleistocene river behaviour and sediment transferv in glaciated catchments of the Mediterranean. PhD Thesis, University of Leeds, 2002.

HOFMEIJER G.K., Late Pleistocene deer fossils from Corbeddu Cave: implications for human colonization of the island of Sardinia, Oxford, British Archaeological Reports, 1997.

HORDEN P., PURCELL N., The corrupting sea: a study of Mediterranean history, Oxford, Blackwell, 2000, 776 p.

HOFMEIJER G.K., ALDERLIESTEN C., VAN DER BORG K., HOUSTON C.M., DE JONG A.F., MARTINI F., SANGES M., SONDAAR P.Y., «Dating of the Upper Palaeolithic Lithic Industry of Sardinia», Radiocarbon, 31, 3, 1989, p. 986-991.

HUGHES D.J., «How the ancients viewed deforestation», Journal of Field Archaeology, 10, 1983, p. 437-443.

JUDSON S., «Erosion and deposition of Italian stream valleys during historic time», Science, 140, 1963, p. 898-899.

JUDSON S., «Stream Changes during Historic Time in East-Central Sicily», American Journal of Archaeology, 67, 3, 1963, p. 287-289.

KANIEWSKI D., VAN CAMPO E., GUIOT J., BUREL S., OTTO T., BAETEMAN C., «Environmental roots of the Late Bronze Age crisis», PLos ONE, 8, 2010, 8 p.

KNAPP A.B., «Cyprus’s Earliest Prehistory: Seafarers, Foragers and Settlers», Journal of World Prehistory, 23, 2010, p. 79- 120. KOPAKA K., MATZANAS C., «Palaeolithic industries from the island of Gavdos, near neighbour to Crete in Greece», Antiquity, 83, 2009, online version

LEIGHTON R., Sicily Before History, New York, Cornell University Press, 1999

LESPEZ L., «Geomorphic responses to long-term land use changes in Eastern Macedonia (Greece)», Catena, 51, 2003, p. 181-208.

LEWIN J., Macklin M.G., Woodward J.C., «Mediterranean Quaternary River Environments - some future research needs», in Lewin J., Macklin M.G. and Woodward J.C., Mediterranean Quaternary River Environments : Refereed Proceedings of an International Conference University of Cambridge/United Kingdom/28-29 September, Rotterdam, Aa Balkema, 1995.


BRADSHAW R.H.W., «A comparison of charcoal measurements for reconstruction of Mediterranean palaeo-fire frequency in the mountains of Corsica», Quaternary Research, 79, 2013, p. 337-349.

LLOBERA M., WILKINSON K.N., WEISS M.C., FLAMING R.J., MARINI N.A.F., MAZET S., «Into the Maquis: Methodological and Interpretational Challenges in Surveying La Balagne, Northwest Corsica», Journal of Mediterranean Archaeology, 23, 2, 2010, p. 169-196.

MACARTHUR R.H., WILSON E.O., Island Biogeography, Princeton, Princeton University Press, 1967.

MAAS G.S., MACKLIN M.G., KIRKBY M.J., «Late Pleistocene and Holocene river development in Mediterranean steepland environments, southwest Crete», in Benito G., Baker V.R., Gregory K.J., Palaeohydrology and Environmental Change, Chichester, Wiley, 1998, p. 153-166.

MAAS G.S., MACKLIN M.G., «The impact of recent climate change on flooding, sediment supply within a Mediterranean mountain catchment, southwestern Crete, Greece», Earth Surface Processes and Landforms, 27, 2002, p. 1087-1105.

MACKLIN M.G., WOODWARD J.C., «River systems and environmental change», in WOODWARD J.C., The Physical Geography of the Mediterranean, Oxford, Oxford University Press, 2009, p. 319-352.

MACKLIN M.G., FULLER I.C., LEWIN J. MAAS G.S., PASSMORE D.G., ROSE J., WOODWARD J.C., BLACK S., HAMLIN R.H.B., ROWAN J.S., «Correlation of fluvial sequences in the Mediterranean basin over the last 200 ka and their relationship to climate change», Quaternary Science Reviews, 21, 2002, p. 1633-1641.

MACKLIN M.G., TOONEN W.H.J., WOODWARD J.C., WILLIAMS M.A.J., FLAUX C., MARRINER N., NICOLL K., VERTSTRAETEN G., SPENCER N., WELSBY D., «A new model of river dynamics, hydroclimatic change and human settlement in the Nile valley derived from meta-analysis of the Holocene fluvial archive», Quaternary Science Reviews, n°130, 2015, p. 109-123.


«Contrasting patterns of precipitation seasonality during the Holocene in the south- and north-central Mediterranean», Journal of Quaternary Science, 27, 2012, p. 290-296.

MAGNY M., COMBOURIEU-NEBOUT N., DE BEAULIEU J.L., BOUT-ROUMAZEILLES V., COLOMBAROLI D., DESPRAT S., FRANCKE A., JOANNIN S., ORTU E., PEYRON O., REVEL M., SADORI L., SIANI G., SICRE M.A., SAMARTIN S., SIMONNEAU A., TINNER W., VANNIÈRE B., WAGNER B., ZANCHETTA G., ANSELMETTI F., BRUGIAPAGLIA E., CHAPRON E., DEBRET M., DESMET M., DIDIER J., ESSALLAMI L., GALOP D., GILLI A., HAAS J.N., KALLELN., MILLET L., STOCK A., TURON J.L., WIRTH S., «North–south palaeo-hydrological contrasts in the central Mediterranean during the Holocene: tentative synthesis and working hypotheses», Climate of the Past, no 9, 2013, p. 2043-2071.

MALONE C., FRAGSUS: Fragility and Sustainability in restricted island environments: Adaptation, Culture Change and Collapse in prehistory, Queen’s University Belfast, 2015. http://www/

MARRINER N., GAMBIN T., DJAMALI M., MORHANGE C., SPITERI M., «Geoarchaeology of the Burmarrad ria and early human impacts in western Malta», Palaeogeography, Plaeoclimatology, Palaeoecology, 339-341, 2012, p. 52-65.

MOUILLOT F., PARADIS G., ANDREI-RUIZ M.C., QUILICHINI A., «Corsica» in Vogiatis I.N., Pungetti G., Mannion M.M., Mediterranean island landscapes: natural and cultural approaches, New York, Springer, 2008, p. 359-378.

NANSON G.C., CROKE J.C., «A genetic classification of flood- plains», Geomorphology, 4, 1992, p. 459-486.

NEWELL W.L., STONE B., HARRISON R., «Holocene alluvium around Lefkosia (Nicosia), Cyprus: an archive of land-use, tectonic processes, and climate change», in Martin- Duque J.F., Brebbia C.A., Godfrey A.E., Diaz de Teran J.R., Geo-Environment: Monitoring and Remedation of the Geological Environment, Warehouse, 2004, p. 71-80.

NINYEROLA M., SAEZ L., PEREZ-OBIOL R., «Relating postglacial relict plants and Holocene vegetation dynamics in the Balearic Islands through field surveys, pollen analysis and GIS modelling», Plant Biosystems - An International Journal Dealing with all Aspects of Plant Biology: Official Journal of the Societa Botanica Italiana, 141, 2007, p. 292-304.

PATTON M., Islands in Time: Island Sociogeography and Mediterranean Prehistory, London, Routledge, 1996, 213 p.

PICCARRETA M., CALDERA M, CAPOLONGO D., BOEMZI F. «Holocene geomorphic activity in Basilicata, Southern Italy», Geomorphology, 128, 2011, 137-147.

RACKHAM O., «Holocene History of Mediterranean Island Landscapes», in Vogiatis I.N., Pungetti G., Mannion M.M., Mediterranean island landscapes: natural and cultural approaches, New York, Springer, 2008, p. 359-378.

RACKHAM O., MOODY J., Themaking of the Cretan Landscape, Manchester, Manchester University Press, 1996, 200 p.

RAINBIRD P., «Islands Out of Time: Towards a Critique of Island Archaeology», Journal of Mediterranean Archaeology, 12, 2, 1999, p. 216-234.

RAMIS D., ALCOVER J. A., COLL J., TRIAS M., «The Chronology of the First Settlement of the Balearic Islands», Journal of Mediterranean Archaeology, 15, 1, 2002, p. 3-24.

REILLE M., GAMISANS J., ANDRIEU-PONEL V., DE BEAULIEU J.L., «The Holocene at Lac de Creno, Corsica, France: a key site for the whole island», New Phytologist, 141, 1999, p. 291-307.

RENFREW C., The Emergence of Civilisation: Cyclades and the Aegean in the Third Millennium BC, London, Methuen Publishing Ltd, 1972, 650 p.

RENFREW C., Wagstaff M., An Island Polity, the Archaeology of Exploitation in Melos. Cambridge, Cambridge University Press, 2005, 376 p.

ROBERTS N., EASTWOOD W.J., KUZUCUOĞLU C., FIORENTINO G., CARACUTA V., «Climatic, vegetation and cultural change in the eastern Mediterranean during the mid-Holocene transition», The Holocene, 21, 2011, p. 147-162.

ROSE J., MENG X., «River activity in small catchments over the last 140 ka, North-east Mallorca, Spain», in Brown A.G., Quine T., Fluvial Processes and Environmental Change, Chichester, Wiley, 1999, p. 91-102.

ROSE J., MENG X., WATSON C., «Palaeoclimate and palaeoenvironmental responses in the western Mediterranean over the last 140 ka: evidence from Mallorca, Spain», Journal of the Geological Society of London, 56, 1999, p. 435-448.

RUNNELS C., «Early Palaeolithic on the Greek Islands?», Journal of Mediterranean Archaeology, 27, 2, 2014, p. 211-230.

RUNNELS C., VAN ANDEL T.H., «The Lower and Middle Palaeolithic of Thessaly, Greece», Journal of Field Archaeology, 20, 1993, p. 299-317.

SAKELLARIOU D., GALANIDOU N.»Pleistocene submerged landscapes and Palaeolithic archaeology in the tectonically active Aegean region», in Harff J., Bailey G., Lüth F., Geology and Archaeology: Submerged Landscapes of the Continental Shelf, London, Geological Society, Special Publications, 411, 2015, p. 145-178.

SALLARES R., «Role of environmental changes in the spread of malaria in Europe during the Holocene», Quaternary International, 150, 1, 2006, p. 21-27.

SCHULDENREIN J. A., «Reassessment of the Holocene stratigraphy of the Wadi Hasa Terrace and Hasa Formation, Jordan», Geoarchaeology, 22, 2007, p. 559-588.

SCHULTE L., «Climatic and human influence on river systems and glacier fluctuations in southeast Spain», Quaternary International, 93-94, 2002, p. 85-100.

SCHULTZ-BARRICK M., Social Organisation in the Mirabello Region of Crete: From Farm to Palace, PhD Thesis, University of Michigan, Anne Arbor. 2007, 371 p.

SEGURA F., «Cuando el mar invade las desembocaduras fluviales», Mètode, 74, 2012, p. 45-49.

SIMMONS A.H., «Of tiny hippos, large cows and early colonists in Cyprus», Journal of Mediterranean Archaeology, 11, 1998, p. 232-241.

SIMMONS A.H., Stone Age Sailors: Palaeolithic Seafaring in the Mediterranean, Walnut Creek, CA, Left Coast Press, 2014, 264 p.

SONDAAR P.Y., DeBoar P.L., Sanges M., Kotsakis T., Esu D., «First report of a Palaeolithic culture in Sardinia», in Waldren W.H., Chapman R.W., Lewthwaite J., Kennard R.C., Early Settlement in the Western Mediterranean Islands and Peripheral Areas, Oxford, British Archaeological Reports, no 229, 1984, p. 29-59.

STOS-GALE Z.A., MALIOTIS G., GALE N.H., ANNETTS N., «Lead isotope characteristics of the Cyprus copper ore deposits applied to provenance studies of copper oxide lingots», Archaeometry, 39, 1, 1997, p. 83-123.

TATTON-BROWN V., Ancient Cyprus, London, British Museum Press, 1997, 72 p.

THORNDYCRAFT V., BENITO G., «The Holocene fluvial chronology of Spain: Evidence from a newly compiled radio-carbon database», Quaternary Science Reviews, 25, 2006, p. 223-234.

TINNER W., JACQUELINE F.N., VAN LEEUWEN J.F.N., COLOMBAROLI C., VESCOVI E., VAN DER KNAAP W.O, HENNE P.D., PASTA S., D’ANGELO S., LA MANTIA T., «Holocene environmental and climatic changes at Gorgo Basso, a coastal lake in southern Sicily», Quaternary Science Reviews, 28, 2009, p. 1498-1510.

TOGNOTTI E., «Malaria in Sardinia», International Journal of Anthropology, 13, 1998, p. 237-242.

TRUMP D., CILIA D., «Malta: Prehistory and Temples», Malta Midsea Books, 2002.

TYKOT R.H., «Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean: New Results and Implications for Cultural Interaction», British Archaeological Reports, International Series, 1303, 2004, p. 25-36.

TYKOT R.H., AMMERMAN A.J., «New directions in central Mediterranean obsidian studies», Antiquity, 71, 274, 1997, p. 1000-1006.

TYKOT R.H., ANDREWS T.K., Sardinia in the Mediterranean: a footprint in the sea. Studies in Sardinian archaeology presented to Miriam S. Balmuth, Monograph in Mediterranean Archaeology, Sheffield, Adademic Press, 1992, 530 p.

VAN ANDEL T.H., RUNNELS C.N., POPE K.,»Five thousand years of land use and abuse in the southern Argolid, Greece», Hesperia, 55, 1986, p. 103-128.

VAN ANDEL T.H., ZANGGER E., DEMITRACK A., «Land Use and Soil Erosion in Prehistoric and Historical Greece», Journal of Field Archaeology, 17, 1990, p. 379-396.

VELLA C, COSTA K., ISTRIA D., DUSSOUILLEZ P., GHILARDI M., FLEURY J.T., DELANGHE D., DEMORY F., CIBECCHINI F., MOREAU J., JOUËT G., «Evolution du fleuve Golo autour du site antique et médiéval de Mariana (Corse, France)», in Ghilardi M., Géoarchéologie des îles de Méditerranée, Paris, CNRS Éditions, 2016 (ce volume).

VITA-FINZI C., «The Mediterranean Valleys: Geological Change in Historical Times», Cambridge, Cambridge University Press, 1969, 140 p.

WALSH K., «The Archaeology of Mediterranean Landscapes: Human-Environment Interaction from the Neolithic to the Roman Period», Cambridge, Cambridge University Press, 2014, 384 p.

WAGNER S., ECKMEIER E., SKOWRONEK A., GÜNSTER N., «Quaternary paleosols and sediments on the Balearic Islands as indicators of climate changes», Catena, 112, 2014, p. 112-124.

WEBSTER G., «Monuments, Mobilisation and Nuragic Organisation», Antiquity, 65, 1991, p. 840-856.

WEBSTER G., A Prehistory of Sardinia, Sheffield, Sheffield Academic Press, 1996, 224 p.

ZANCHETTA G., BAR-MATTHEWS M., DRYSDALE R.N., LIONELLO P., AYALON A., HELLSTROM J.C., ISOLA I., REGATTIERI E, «Coeval dry events in the central and eastern Mediterranean basin at 5.2 and 5.6 ka recorded in Corchia (Italy) and Soreq caves (Israel) speleothems», Global and Planetary Change, 122, 2014, p. 130-139.

Table des illustrations

Légende Figure 1: Map of the Mediterranean Islands and fluvial study locations discussed in this paper. 1: SE Spain (Schulte, 2002); 2: S Etruria (Brown, 1997; Brown and Ellis, 1996); 3: Tunisia (Faust et al., 2004); 4: Boetia (Bintliff, 2002); 5: Macedonia (Lespez, 2003); 6: Central Greece (van Andel et al., 1990); 7: Peloponese (Fuchs, 2007); 8: Jordan (Shuldenrein, 2007).
Fichier image/jpeg, 435k
Légende Figure 2: Plot of island area vs relative relief with a power function trend-line.
Fichier image/jpeg, 84k
Légende Table 1: Largest Mediterranean island statistics and geology where Koppen classification is in parentheses it denotes a minor classification ie for parts of the island(s).
Fichier image/jpeg, 222k
Légende Figure 3: Summary table of Holocene alluviation phases. Black is strong evidence of increased alluviation, grey moderate evidence, blue is wetter periods, yellow is drier periods and * is multiple alluviating events on the decadal scale. From many sources including those in the text and Schulte et al. (2002) and Thorndycraft and Benito (2006).Long record are GISP2 (NOAA, GISP2) and the Soreq cave record (Bar-Matthews and Ayalon, 2011).
Fichier image/jpeg, 308k
Légende Table 2: Current estimates of the earliest dates of human/hominin occupation of selected Mediterranean islands with ancillary data.Updated from Mouillot et al. (2008), * estimate of distance at Pleistocene lowest RSL.
Fichier image/jpeg, 231k


Palaeoenvironmental Laboratory University of Southampton (PLUS), University of Southampton, SO17 1BJ United Kingdom (

Department of Archaeology, King’s Manor, University of York, York YO1 7EP (

© CNRS Éditions, 2016

Conditions d’utilisation :