9
Original article Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece) § Investigation palynologique des changements pale ´oenvironnementaux holoce `nes de la plaine co ˆtie `re du Marathon (Attica, Gre `ce) Katerina Kouli a, * , Maria Triantaphyllou a , Kosmas Pavlopoulos b , Theodora Tsourou a , Panagiotis Karkanas c , Michael D. Dermitzakis a a Department of Historical Geology-Palaeontology, University of Athens, Panepistimiopolis, 15784 Greece b Faculty of Geography, Harokopio University, 70 El Venizelou Avenue, 17671 Athens, Greece c Ephoreia of Palaeoanthropology-Speleology, 34b Arditou street, 11636 Athens, Greece Received 23 October 2007; accepted 26 July 2008 Available online 9 October 2008 Abstract The identification of Middle-Late Holocene palaeoenvironmental conditions of the Marathon coastal plain gained great interest in the last decades due to its high environmental and archaeological importance. Palynological analysis of samples from two boreholes and two trenches along a transect in the marshy area of the Marathon coastal plain, enabled the tracing of the vegetation and the main environmental changes for the last 6000 cal BP. Pollen data suggest a human disturbed environment with Pinus, Quercus, Juniperus and Ericaceae, while a general trend towards Mediterranean vegetation patterns is observed during the last 3000 cal BP. Pollen grains from aquatic and hydrophilous plants, dinoflagellate cysts, algal remains and other palynomorphs were used in order to determine the local depositional environment and its evolution through time. # 2008 Elsevier Masson SAS. All rights reserved. Résumé L’identification des conditions paléoenvironnementales de l’Holocène moyen-supérieur de la plaine côtière de Marathon a gagné un grand intérêt durant les dernières décennies, de part son importance environnementale et archéologique. Les analyses palynologiques des échantillons provenant de deux forages et deux fossés situés dans le secteur marécageux de la plaine côtière de Marathon, ont permis le traçage de la végétation et des changements environnementaux principaux durant les derniers 6000 cal BP. Les analyses polliniques suggèrent l’influence humaine sur une végétation dominée par Pinus, Quercus, Juniperus et les Ericaceae. Une tendance générale vers une végétation méditerranéenne est plus visible pendant les derniers 3000 cal BP. Les grains de pollen des plantes aquatiques et hydrophiles, kystes de dinoflagellés, résidus d’algues et autres palynomorphes ont été utilisés pour déterminer les caractéristiques du milieu de dépôt local et son évolution dans le temps. # 2008 Elsevier Masson SAS. All rights reserved. Keywords: Pollen analysis; Vegetation history; Palaeoecology; Holocene; Greece Mots clés : Analyse pollinique ; Histoire de la végétation ; Paléoécologie ; Holocène ; Grèce 1. Introduction The Marathon coastal plain has been continuously inhabited since the Neolithic (Pantelidou-Gofas, 1997) and has a great historical as well as archaeological significance. During the 6th century BC, the four flourishing communities of the area were Geobios 42 (2009) 4351 § Corresponding editor: Serge Legendre. * Corresponding author. E-mail address: [email protected] (K. Kouli). 0016-6995/$ see front matter # 2008 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.geobios.2008.07.004

Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

Embed Size (px)

Citation preview

Page 1: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

Original article

Palynological investigation of Holocene palaeoenvironmental changesin the coastal plain of Marathon (Attica, Greece)§

Investigation palynologique des changements paleoenvironnementaux

holocenes de la plaine cotiere du Marathon (Attica, Grece)

Katerina Kouli a,*, Maria Triantaphyllou a, Kosmas Pavlopoulos b, Theodora Tsourou a,Panagiotis Karkanas c, Michael D. Dermitzakis a

a Department of Historical Geology-Palaeontology, University of Athens, Panepistimiopolis, 15784 Greeceb Faculty of Geography, Harokopio University, 70 El Venizelou Avenue, 17671 Athens, Greece

c Ephoreia of Palaeoanthropology-Speleology, 34b Arditou street, 11636 Athens, Greece

Received 23 October 2007; accepted 26 July 2008

Available online 9 October 2008

Abstract

The identification of Middle-Late Holocene palaeoenvironmental conditions of the Marathon coastal plain gained great interest in the lastdecades due to its high environmental and archaeological importance. Palynological analysis of samples from two boreholes and two trenchesalong a transect in the marshy area of the Marathon coastal plain, enabled the tracing of the vegetation and the main environmental changes for thelast �6000 cal BP. Pollen data suggest a human disturbed environment with Pinus, Quercus, Juniperus and Ericaceae, while a general trendtowards Mediterranean vegetation patterns is observed during the last �3000 cal BP. Pollen grains from aquatic and hydrophilous plants,dinoflagellate cysts, algal remains and other palynomorphs were used in order to determine the local depositional environment and its evolutionthrough time.# 2008 Elsevier Masson SAS. All rights reserved.

Résumé

L’identification des conditions paléoenvironnementales de l’Holocène moyen-supérieur de la plaine côtière de Marathon a gagné un grandintérêt durant les dernières décennies, de part son importance environnementale et archéologique. Les analyses palynologiques des échantillonsprovenant de deux forages et deux fossés situés dans le secteur marécageux de la plaine côtière de Marathon, ont permis le traçage de la végétationet des changements environnementaux principaux durant les derniers �6000 cal BP. Les analyses polliniques suggèrent l’influence humaine surune végétation dominée par Pinus, Quercus, Juniperus et les Ericaceae. Une tendance générale vers une végétation méditerranéenne est plus visiblependant les derniers �3000 cal BP. Les grains de pollen des plantes aquatiques et hydrophiles, kystes de dinoflagellés, résidus d’algues et autrespalynomorphes ont été utilisés pour déterminer les caractéristiques du milieu de dépôt local et son évolution dans le temps.# 2008 Elsevier Masson SAS. All rights reserved.

Keywords: Pollen analysis; Vegetation history; Palaeoecology; Holocene; Greece

Mots clés : Analyse pollinique ; Histoire de la végétation ; Paléoécologie ; Holocène ; Grèce

Geobios 42 (2009) 43–51

§ Corresponding editor: Serge Legendre.* Corresponding author.

E-mail address: [email protected] (K. Kouli).

0016-6995/$ – see front matter # 2008 Elsevier Masson SAS. All rights reserveddoi:10.1016/j.geobios.2008.07.004

1. Introduction

The Marathon coastal plain has been continuously inhabitedsince the Neolithic (Pantelidou-Gofas, 1997) and has a greathistorical as well as archaeological significance. During the 6thcentury BC, the four flourishing communities of the area were

.

Page 2: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

K. Kouli et al. / Geobios 42 (2009) 43–5144

integrated into Tetrapolis, one of the twelve districts (deme)into which Attica was divided before the time of Theseus(Petrakos, 1995). The area is famous for the ancient battle of490 BC between the Athenians and the Persians and, based onthis, was proclaimed a national park. For the 2004 OlympicGames in Athens, a rowing center was constructed at this place.Given the great environmental and archaeological importanceof the area, the investigation of its palaeoenvironmentalconditions during the Holocene becomes very interesting.

Sedimentological, micromorphological and micropalaeon-tological analysis together with radiocarbon dating of thesediments determined the depositional environments and thesea level changes recorded in the area for the last�6,000 cal BP(Pavlopoulos et al., 2003; Triantaphyllou et al., 2003).

However, relatively little is known about Holocenepalaeovegetation of this area. The only published pollen recordin Attica comes from the archaeological deposits of the cave ofKitsos in Lavrio (Renault-Miskovsky, 1981) since all otherpollen sites are found at a distance of several kilometers inneibour basins, like Kopais (Turner and Greig, 1975; Allen,1990), Megaris (Jahns, 2003) and in Peloponnese (Aliki:Kontopoulos and Avramidis, 2003; Lerna: Jahns, 1993; Nemea:Atherden et al., 1993; Koiladha: Bottema, 1990).

This study investigates the potential of palynodata inrecording the local microenvironmental changes in the coastalplain of Marathon, comparing them with the existingmicropalaeontological and sedimentological results (Pavlo-poulos et al., 2003, 2006; Triantaphyllou et al., 2003) and aimsto contribute to our knowledge on Holocene vegetation ofAttica.

2. Site setting

The NE-SW elongated Marathon coastal plain is located inthe NE Attica, eastern Greece (Fig. 1). The broader areaconsists of ‘‘NE Attica’’ geotectonic units, that represent a

Fig. 1. Topographic map of the Marathon coastal plain showing the locations ofcores 6 and 7 and trenches 4 and 10.Carte topographique de la plaine côtière de Marathon avec la localisation desforages 6 et 7 et des fosses 4 et 10.

‘‘relatively autochtonous’’ metamorphic sequence (Lozios,1993). The Holocene is represented by various types of alluvialdeposits, formed mainly by the Inois River and other smalltorrents. The area has been affected by two main fault systems,the older one having a NE-SW direction and the younger onehaving a NW-SE direction (Lozios, 1993). An apparent coastalstability has been evidenced at least for the last 5000 yr (Kraft,1972; Pavlopoulos et al., 2006).

The plain is divided into two parts by the Inois Riverchannel. In the southwest of the deltaic fan of the Inois River(Kainourgio Rema), there was a marshy area (Vreksisa) thatwas drained some decades ago. The large marsh of Marathon(also known as the Great Marsh) extends to the east, separatedfrom the sea by a barrier beach with low sand dunes (Baeteman,1985). The Marathon plain resembles the typical coastal plainsof Greece in morphology (Kraft et al., 1975, 1977; Fouache,1999). The plain coastline is almost straight, with the exceptionof a ledge formed near the recent river mouth, not affected bytides (tide amplitude less than 20 cm). No significant changeshave been observed on the coastline limit between 1889 and1938. The area near the mouth of Kenourio Rema wasretreating at a rate of about 2 m/yr in the 1950’s and 1960’s,slowing down to 1 m/yr in the last two decades (Maroukianet al., 1993).

The climate is typical Mediterranean with warm, drysummers and mild, humid winters. Mean annual precipitationfor the meteorological station of Marathon is 567 mm and meanair humidity is 59–64%. Monthly air temperature rangesbetween 27 and 10 8C with a mean annual value of 18 8C.About 50 cloudy days each year are recorded for the area andthe mean sunlight is 2920 h per year. Sea surface temperaturefluctuates between 10.3 and 26.7 8C. The history of the fertileland of Marathon coastal plain goes back to the Neolithic, whenthe first important human inhabitations in Nea Makri(Pantelidou-Gofas, 1991, 1995), Vreksisa, Plasi, Kato Souliand in Panos cave of Oenoe is recorded (Petrakos, 1995;Pantelidou-Gofas, 1997). The older so far known Bronze Agecemetery has been discovered in the area (Pantelidou-Gofas,2005). During the Geometric times, four inhabitation centersexisted in the plain of Marathon: Marathon, Provalinthus,Tricorythus and Oinoe. In the 6th century BC, they wereincorporated into Tetrapolis (Petrakos, 1995). The area isfamous from the ancient battle of 490 BC between theAthenians and the Persians in which the heavily outnumberedAthenian army defeated the Persians. A burial mound (tymbos)for the dead Athenian soldiers was erected near the battlefield.Tymbos nowadays is marked by a marble memorial stele andsurrounded by a park.

Three main villages, Kato Souli, Marathon and Nea Makri,exist in the area today. In the area of the drained Great Marsh,many summer houses are built during the last decades and arowing center was constructed for the 2004 Olympic Games inAthens. The plain is widely cultivated, being one of the mainvegetable producer areas in Attica. The vegetation of themarshy area consists mainly of Phragmites, Typha latifolia andJuncus maritime. Along the streams some Platanus orientalis,Populus alba and Salix alba are found. Parallel to the seashore,

Page 3: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

K. Kouli et al. / Geobios 42 (2009) 43–51 45

a natural forest of Pinus pinea exists were Pinus halepensis,Juniperus phoenica, Quercus ilex and Quercus coccifera arealso found. The hilly rocky areas around the plain are coveredwith maquis vegetation mainly represented by Juniperusphoenicea, Pistacia lentiscus, P. terebinthus, Ceratonia siliqua,Olea europaea, Ephedra foemina, Quercus coccifera, Rhamnusalaternus, Calicotome villosa and Prunus webbii. Furtherinland and in the fields’ margins, small traces of the naturalwoodland vegetation are evidences consisting of rare trees ofQuercus ilex, Quercus pubescens, Phillerea latifolia andMyrtus communis.

3. Depositional environment and age assessment

Baeteman (1985) conducted a systematic drill-hole study forthe area and presented detailed information on the stratigraphicsequence of the plain.

Triantaphyllou et al. (2003) investigated the microfaunarecovered from the deposits and determined three distinctbiofacies based on foraminifera and ostracoda assemblages: theshallow mesohaline-oligohaline biofacies, indicative of lowmarch environments approximately at the mean sea level(Petrucci et al., 1983), the shallow oligohaline-fresh watersbiofacies indicative of high-middle marsh environments (Scottet al., 1979) suggesting an approximate elevation of 20 cmabove mean sea level and finally the mesohaline-oligohaline tooligohaline-freshwater biofacies characteristic of an inter-mediate mesohaline-oligohaline to oligohaline-freshwaterlagoonal environment.

Pavlopoulos et al. (2003, 2006) determined the sequence ofdepositional environments and the sea level changes recordedin the area for about the last 6000 yr, using micromorphologicaland micropalaeontological methods in addition to AMSradiocarbon datings (Table 1). Three sedimentary units, namely

Table 1Radiocarbon dating for the Marathon coastal plain deposits (from Pavlopoulos etDatations radiocarbones des depots sedimentaires de la plaine cotiere de Maratho

Laboratory code Trench/Borehole Noa

Unit Absolutealtitude (m)b

Material

Hv 8546a 6–7a C +0.10 Peat 5GX-27909 (AMS) 1 C +1.20 Wood fragmGX-27908 (AMS) 1 C +1.00 Wood fragmHv 8547a 6–7a C �0.10 Peat 4GX-27915 4 A-/B �1.80 Shells (brackHv 8533a 6a A +0.30 to �0.50 Carbonate mHv 8551a 37a A �0.80 to �1.20 Carbonate mHv 8548a 36a A �1.95 to �1.85 Carbonate mGX-27910 6 A �1.47 Peat 3, topGX-27911 6 A �1.65 Peat 3, middGX-27914 7 A �2.20 Peat 2, topHv 8552a 37a A �2.10 to 2.20 Peat 1, meanGX-27913 6 A �2.13 Peat 1, topGX-27912 (AMS) 6 A �2.13 Peat 1, woodHv 8549a 36a A �2.15 to �2.25 Peat 1, middHv 8550a 36a A �2.35 �2.45 Peat 1, base

a Radiocarbon dates from Baeteman (1985).b Absolute altitudes are corrected based on new map data of present work (withc Corrected for reservoir effect.

A, B and C, were recognized after grouping the sedimentaryfacies characteristics (Fig. 2).

Sedimentary unit A consists of fossiliferous bioturbatedlagoonal mud predominating in the central part of theembayment, with peloidal charophytic mud prevailing in thelower parts of the unit. Three peat layers are included in unit A.The lower one (peat 1), dated at approximately 5500 cal BP(Pavlopoulos et al., 2006), starts as laminated algal peat in thelower horizons and changes to a reed swamp peat. The othertwo peats found in unit A (peat 2 dated to about 4700 cal BP andpeat 3 dated to about 3800 cal BP; Pavlopoulos et al., 2006) arecomposed of almost pure plant remains. Depositionalenvironment of unit A is a typical mesohaline-oligohalinelagoon with an embayment depth never exceeding a few meters(Triantaphyllou et al., 2003; Pavlopoulos et al., 2006).

Unit B consists of mixed carbonate and siliciclastic mudaccompanied by high amounts of fragmented plant material. Atthe boundary of units A and B, the presence of mollusc shellsprovided a radiocarbon age of about 3500 cal BP, while thepresence of wood fragments in the upper horizons ofsedimentary Unit B provided a radiocarbon age of about2500 cal BP (Pavlopoulos et al., 2006), close to the time that theMarathon Battle took place (490 BC). Deposits of Unit Brepresent a freshwater to oligohaline-mesohaline mixture ofwindblown or riverside silts and clays and palustrine lime mud.Communication with the sea was frequent but not perennial(Triantaphyllou et al., 2003).

Deposits of the uppermost sedimentary unit C vary locally,representing a mixture of fluviatile and palustrine environments(Pavlopoulos et al., 2006). In the area of the former nearshoreenvironment, the deposits consist of well-sorted weaklycemented sands and silty sands, while, towards the centre ofthe embayment, palustrine mud is sometimes interrupted bylayers of well sorted sands. Palustrine deposits include two thin

al., 2006)n (d’apres Pavlopoulos et al., 2006).

14C Age(yr BP)

d13C PDB(%)

Calibrated Age Age(BC)

1360 � 40 n.r. 1309–1194 641–756 ADent 2400 � 30 �25.2 2465–2351 516–402ents 2410 � 40 �27.6 2693–2351 744–402

2480 � 60 n.r. 2711–2467 762–518ish) 3570 � 180 �4.5 3684–3243c 1735–1294c

ud 4020 � 60 n.r. 4115–3945c 2166–1996c

ud 3985 � 65 n.r. 4070–3894c 2121–1945c

ud 3550 � 80 n.r. 3529–3348c 1570–1339c

3560 � 60 �27.4 3959–3725 2010–1776le 3540 � 70 �28.6 3898–3701 1949–1752

4160 � 60 �27.2 4824–4588 2875–26394575 � 60 n.r. 5445–5053 3496–31044770 � 60 �27.4 5590–5334 3541–3385

fragment 5080 � 40 �24.1 5894–5751 3945–3802le to top 4869 � 75 n.r. 5707–5486 3758–3557

4570 � 105 n.r. 5449–5048 3500–3099

a � 20 cm uncertainty).

Page 4: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

Fig. 2. Trenches and cores logs and position of the analyzed samples (from Pavlopoulos et al., 2006, modified). (1) artificial fill, (2) topsoil, (3) sand, (4) sand withgravels and pebbles, (5) silt, (6) clay, (7) silty sand, (8) sandy silt, (9) clayey silt, (10) sandy clay, (11) silty clay, (12) sandy clay with subrounded gravels,(13) palustrine mud, (14) lagoonal mud, (15) peloidal algal mud, (16) mudcracs, (17) rootlets, (18) charophytes, (19) pellets, (20) fossils, (21) bioturbation, (22) peat,(23) algal peat, (24) hard horizon. Peat ages are in cal BP. * Age according to Baeteman (1985).Logs des forages et fosses et position des échantillons analysés (modifiée d’après Pavlopoulos et al., 2006). (1) remplissage artificiel, (2) sol, (3) sable, (4) sable avecgraviers et galets, (5) limon, (6) argile, (7) sable limoneux, (8) limon sableux, (9) limon argileux, (10) argile sableuse, (11) argile limoneuse, (12) argile sableuse avecgraviers, (13) boue palustre, (14) boue lacustre, (15) boue algaire pelloïdale, (16) fente de dessiccation, (17) petites racines, (18) charophytes, (19) grains,(20) fossiles, (21) bioturbation, (22) tourbe, (23) tourbe algaire, (24) tourbe. Âge en cal BP. * Âge en accord avec Baeteman (1985).

K. Kouli et al. / Geobios 42 (2009) 43–5146

Page 5: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

K. Kouli et al. / Geobios 42 (2009) 43–51 47

peat layers (peat 4 and 5) dated to about 2590 cal BP and1250 cal BP respectively (Baeteman, 1985). In the central partof the embayment, an isolated, frequently exposed wetland hasbeen recognized (Triantaphyllou et al., 2003; Pavlopouloset al., 2006), mainly influenced by fresh water inputs.

The whole sedimentary sequence of Marathon coastal plainis a typical progradation sequence with several minor cycles.The sea level rise indicated by the several peat formations(Pavlopoulos et al., 2006) has been estimated to be lower thanthe one predicted by the glaciohydroisostatic model (Lambeck,1995, 1996) and the data from other Greek areas (Kraft et al.,1975, 1977; van Andel and Lianos, 1984; Kambouroglou,1989) that are considered relatively stable. Hence, an averagetectonic uplift of the area has been suggested at a rate of about0.4–0.5 mm/yr, which almost counterbalances the predictedrate of relative sea-level rise of about 0.6–0.7 mm/yr for the last2000 yr (Pavlopoulos et al., 2003, 2006). This explains also therelative geomorphological stability since at least the Classicaltimes suggested by the historical documents (Kraft, 1972).

4. Material and pollen analytical methods

The material used in the present study comes from 4 (2boreholes and 2 trenches) out of the 6 profiles realized for thestudy of Pavlopoulos et al. (2003, 2006) along a transect in themarshy area of the Marathon coastal plain (Fig. 2). A total of 20samples were available for palynological analysis, out of eachabout 1 cm3 was chemically treated with HCl (37%), HF(40%), acetolysed and sieved using a 10 mm sieve. Residueswere mounted in silicon oil.

Pollen and spores were identified using the Moore et al.(1991) key and the Reille (1992–1998) pollen floras, whileother non-pollen palynomorph identification was based on thevan Geel et al. (1989, 2003) studies. The differentiation ofevergreen and deciduous Quercus was not always possible, asmany pollen grains were crumbled, so a composite histogram

Fig. 3. Percentage palynological diagram of core 7.Diagramme pollinique en pourcentages du forage 7.

was plotted, within which the positively identified evergreenQuercus is indicated.

Pollen percentage calculations were based on regionalpollen. Aquatic pollen and Cyperaceae were excluded frompollen sum. Palynomorph concentrations were calculatedbased on their comparison to the introduced Lycopodiumspores and expressed as grains/cm3 of dry sediment. Pollenpercentage histograms were constructed using the programsTILIA and TGVIEW (Grimm, 1992). The results ofpalynomorph analysis for each profile are presented aspercentage pollen diagrams (Figs. 3–6) in correlation withthe recognized sedimentary units for the Marathon coastalplain (Pavlopoulos et al., 2003, 2006). Pollen and spores fromaquatic and hydrophilous plants, dinoflagellate cysts, algal andfungal remains as well as other palynomorphs were used todetermine various depositional environments of the area andtheir evolution through time.

5. Results

The main pollen contributors in the pollen flora are Pinus;Quercus, Poaceae, Asteraceae, Juniperus and Ericaceae. Thehighest pollen concentrations (62,000 grains/cm3) wasrecorded in trench 10 at 0.95 m and the lowest (500 grains/cm3) in trench 4 at 1.40 m). The few available pollen spectrarecovered from each profile do not allow for interpretations interms of vegetation evolution but can only provide punctualimages of the landscape during short time intervals.

Pollen spectra of core 7 (Fig. 3) derive from sedimentaryUnit A, with the exception of the upper spectrum coming fromsedimentary Unit B. Pollen concentration values rangebetween 10,700 grains/cm3 (at 3.20 m) and 3500 grains/cm3

(at 3.10 m). Core 7 is characterized by high arboreal pollenpercentages. Pinus and Quercus prevail in the pollen spectrawith percentages around 30 and 15% respectively, while thepresence of Ericaceae (10.2%), Juniperus (5.8%), Carpinus/

Page 6: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

Fig. 4. Percentage palynological diagram of core 6.Diagramme pollinique en pourcentages du forage 6.

Fig. 5. Percentage palynological diagram of trench 4.Diagramme pollinique en pourcentages de la fosse 4.

Fig. 6. Percentage palynological diagram of trench 10.Diagramme pollinique en pourcentages de la fosse 10.

K. Kouli et al. / Geobios 42 (2009) 43–5148

Ostrya type (2.2%) and Sorbus is noteworthy. Herb vegetationis mainly represented by Poaceae (11.4%), Ranunculus acris(7.4%), Asteroideae (2.9%) and Cistaceae.

The general image of the pollen flora recovered from core 6(Fig. 4) exhibits some similarities with core 7. Unfortunatelyonly three levels, two from sedimentary Unit A and one fromsedimentary Unit B, were available for pollen analysis. Pollenconcentrations are about 4800 grains/cm3.

In trench 4 (Fig. 5), the analyzed levels draw from allrecognized sedimentary Units. Pollen concentrations are lowerthan in core 7, ranging between 8700 grains/cm3 (at 0.76 m)and 500 grains/cm3 (at 1.40 m). Pinus is the dominant arborealpollen contributor, especially in sedimentary Unit C, where itspercentages exceed the 50% of the pollen sum. Ericaceae(3.5%), Juniperus (1.3%), Quercus and Pistacia comprise therest of the arboreal taxa, while herb vegetation is mainly

Page 7: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

K. Kouli et al. / Geobios 42 (2009) 43–51 49

represented by Poaceae (8.7%), Asteroideae (6.2%), Cichor-ioideae, Cistaceae and Sanquisorba. Cichorioideae abundanceappears higher (about 33.4%) in top two spectra of sedimentaryUnit C.

Spectra from trench 10 (Fig. 6) exhibit large fluctuations inpollen concentrations ranging between 62,000 grains/cm3 (at1.00 m) and 2350 grains/cm3 (at 2.30 m). In accordance withtrench 4, Pinus is the dominant pollen contributor, while othertree taxa like Carpinus/Ostrya type, and Juniperus exhibit lowabundances. Quercus pollen has not been recorded. At the toplevels, Chenopodiaceae and Plantago lanceolata display theirhighest abundances in the Marathon coastal plain pollen spectra(15.1 and 8%, respectively).

Cerealia type pollen was encountered in spectra of allprofiles, though it appears less abundant in spectra ofsedimentary Unit C.

The fluctuations of the curves of aquatics – Cyperaceae andSparganium emersum – are significant, due to the localconditions of each profile. The record of freshwater algae(Spirogyra, Botryococcus) is perpetual, while the marinedinoflagellates (Operculodinium centrocarpum and Spiniferitesspp.) are only found up to the middle part of the profiles.Finally, the presence of several fungal and animal remainscompletes the image of the pollen spectra.

6. Discussion – palaeoenvironmental interpretation

We focus the discussion on the Marathon pollen profiles ontwo main aspects: the delineation of the palaeovegetation of thearea and the determination of the local depositional environ-ments related to their evolution through time.

The pollen flora recovered, despite the punctual character ofthe available data, comprises representatives from differentphytogeographic zones and provides an image of the landscapefor the last �6000 cal BP.

Mediterranean vegetation is fairly represented by theoccurrence of pollen of evergreen Quercus, Juniperus, Pistaceaand by the great variability of herbaceous taxa – e.g.Sanguisorba, Helianthemum, etc. – , while deciduous Quercusand Carpinus/Ostrya-type are the main representatives of thethermophilous deciduous mixed woodlands that was coveringthe slopes around the coastal plain. The presence of Abies andFagus pollen is indicative of the altitudinal conifer forests andlocal beech communities that flourish even today at ParnithaMountain, about 30 km to the west.

The continuous presence of human in the area since theNeolithic (before 6000 BC) (Pantelidou-Gofas, 1995; Petrakos,1995) and the consequent exploitation of the environment isevidenced in the Marathon coastal plain record. All pollendiagrams are featured by the presence of human indicatorspecies such us Cerealia type, Ranunculus acris, Plantagolanceolata type, Coprophilous sordariaceae spores (van Geelet al., 2003) and Puccinia teleutospores (van Geel et al., 1980-1981; Carrion and van Geel, 1999), providing clear evidenceabout cereal cultivation and stock breeding activities in thearea. Moreover, the presence of taxa like Ericaceae andJuniperus must have been favored by grazing. Their

expansion – together with Pinus abundance – on clearedground is considered to characterize the human disturbedvegetation in Greece (Bottema, 1974; Bottema and Woldring,1990; Jahns, 1993). Moreover, the presence of Olea pollen inspectra leaves no doubt about the cultivation of olives in thearea, as its pollen is generally considered under-represented inpollen diagrams from Greece (Jahns, 1993).

The upper part of the sequence (top of sedimentary unit Band sedimentary unit C; �3000 cal BP to present) ischaracterized by very high abundances of Pinus in all profiles.The expansion of Pinus pollen after 3000 cal BP has beenrecorded in several pollen diagrams of southern Greece likeKoiladha (Bottema, 1990), lake Lerna (Jahns, 1993) andKleonai (Atherden et al., 1993) and has been attributed to thespread of pine woods on coastal areas, where they still flourishtoday (e.g., the wood of Pinus pinea covers most of the presentcoastline of Marathon bay). The instability of hydrologicalconditions (shallow waters, periodic sea and fresh waterflooding; Pavlopoulos et al., 2003, 2006; Triantaphyllou et al.,2003) seems to control the expansion of the aquatic andhydrophilous vegetation in the area throughout �6000 cal BP.The studied time span is featured by peaks in the Sparganiumemersum curve most probably related to the water-tablechanges. The presence of a shallow aquatic environment issupported by the continuous record of spores of Spirogyra andother Zygnemataceae algae that characterize shallow andstagnant waters (van Geel et al., 1980-1981).

Overall, the distribution of aquatic and hydrophilous pollen,algal spores, dinoflagellate cysts, fungal spores and insect andother invertebrate remains in pollen spectra of the coastal plainof Marathon, indicates an ongoing variation of the depositionalenvironment both in time and space. A Q-mode cluster analysiswas performed on the complete dataset in order to define areasof similar hydrological conditions (Fig. 7).

Cluster I is represented by samples T8/7, T8/5, T10/1,T6/F1, T8/6, T6/P1, T10/3 and T10/4. The vast majority ofthese samples belong to sedimentary unit C. All clusteredsamples are related to a very shallow depositional environment(SDE facies) as they are characterized by a high abundance ofSpirogyra and other Zygnemataceae spores and type 128.Zygnemataceae are among the most common freshwater algaeand indicate shallow and stagnant waters (van Geel et al., 1980-1981), while type 128 is indicative of shallow and eutrophicwater (van Geel et al., 1983).

Clusters IIa and IIb are represented by samples T7/1, T10/5,T8/4, T6/P2 and T7/4 and samples T7/9, T8/3, T8/1, T7/2, T7/5,T7/7 and T7/3, respectively. They are associated with arelatively deeper depositional environment (DDE facies),since they are characterized by a lower frequency of Spirogyraand the presence of Botryococcus, Pediastrum and dinoflagellatecysts. More specifically cluster IIa is characterized by thepresence of dinoflagellate cysts. The small number of theOperculodinium centrocarpum and Spiniferites spp. cystsindicate that, although there was a periodic sea water input inthe lagoon, there was no regular connection with the sea.Dinoflagellate cysts have been used as salinity indicators byseveral authors (e.g. Wall et al., 1973; Mudie et al., 2001; Head

Page 8: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

Fig. 7. Cluster analysis leading to the description of different depositional environments.Analyse de groupement conduisant à la description des divers environnements de dépôt.

K. Kouli et al. / Geobios 42 (2009) 43–5150

et al., 2005). In order to establish a salinity signal in cystdistributions, Dale (1996) compared the late Quaternarydata from the Black Sea with recent observations fromNorwegian fjords and the Baltic Sea; he suggested that cystassemblages that are restricted to the cosmopolitan speciesO. centrocarpum and Spiniferites spp. are indicative of salinitieslower than 7–3%.

Cluster IIb is characterized by the total absence of indicationof sea water input and the abundance of local vegetation, likeSparganium emersum and Cyperaceae.

Both depositional environments recorded by palynomorphanalysis in the coastal plain of Marathon (Fig. 7) appear tocorrelate directly with the biofacies that have been recognizedby the micropalaeontological study in the area (Triantaphyllouet al., 2003). In particular, the shallow environments (SDE facies)that characterize a very shallow depositional environmentcorrelates with the oligohaline-fresh water biofacies recordedin sedimentary unit C (2500 cal BP to present). The DDE faciescorrelates with the mesohaline-oligohaline biofacies recorded insedimentary Units A and B (before 5800 to 3500 cal BP and3500–2500 cal BP, respectively). The palynomorph analysisprovides further evidences for the Marathon coastal plain being amarshy environment with periodical sea influence between�6000 cal BP to 2500 cal BP. During the last 2500 years, thewetland became shallower and partially desiccated resulting tothe restriction of the high-middle marsh conditions to thesouthern part of the plain.

7. Conclusions

The environmental history of the Marathon coastal plain forthe Middle-Late Holocene has been recorded in the palyno-morphs found in the studied deposits. The palaeovegetation ofthe area was found to be influenced by humans during the wholeperiod, with the cereal cultivation and the stock breeding beingthe main anthropic activities. Vegetation patterns show thegeneral trend towards Mediterranean environments for the last�3000 cal BP approximately, in accordance with previousstudies in southern Greece (Atherden et al., 1993; Jahns, 1993).Mesohaline-oligohaline biofacies (MO) that have beenrecognized by the micropalaeontological analysis (Trianta-phyllou et al., 2003) correlate with relatively deeper environ-

ments (DDE facies), while the oligohaline-fresh water (OFW)biofacies correlate with very SDE facies. The palaeodata of theMarathon coastal plain show its continuous shift towards thelandward area that has been attributed to a slowing down of thesea level rise.

In conclusion, the determination of the local depositionalenvironments and their evolution through time by the joined useof micropalaeontological, sedimentological and palynologicalresults appears to be successfully applicable in small coastalbasins.

Acknowledgments

We wish to thank Dr. M. Papanikolaou for her helpfuldiscussion on dinoflagellate cysts and both Dr Alexandra vander Geer and Dr. H. Koufosotiri for their comments and helpwith the manuscript. An anonymous reviewer is thanked forconstructive comments that improved the manuscript.

References

Allen, H., 1990. A postglacial record from the Kopais basin, Greece. In: Bottema,S., Entjes-Nieborg, G., van Zeist, W. (Eds.),Man’s role in the Shaping of theEastern Mediterranean landscape. Balkema, Rotterdam, pp. 173–181.

Atherden, M., Hall, J., Wright, J.C., 1993. A pollen diagram from the northeastPeloponnese Greece: implications for vegetation history and archaeology.The Holocene 3, 351–356.

Baeteman, C., 1985. Late Holocene geology of the Marathon plain (Greece).Journal of Coastal Research 1, 173–185.

Bottema, S., 1974. Late Quaternary vegetation history of Northwestern Greece.Ph.D. thesis, Rijksuniversiteit te Groningen (unpublished).

Bottema, S., 1990. Holocene environment of the Southern Argolid: a pollen corefrom Kiladha Bay. In: Wilkinson, T.J., Duhon, S.T. (Eds.), Excavations inFranchthi cave 6. Indiana University Press, Bloomington, pp. 117–138.

Bottema, S., Woldring, H., 1990. Anthropogenic indicators in the pollen recordof the Eastern Mediterranean. In: Bottema, S., Entjes-Nieborg, G., vanZeist, W. (Eds.), Man’s role in the Shaping of the Eastern Mediterraneanlandscape. Balkema, Rotterdam, pp. 231–264.

Carrion, J.S., van Geel, B., 1999. Fine-resolution Upper Weichselian andHolocene palynological record from Navarrés (Valencia, Spain) and adiscussion about factors of Mediterranean forest succession. Review ofPalaeobotany and Palynology 106, 209–236.

Dale, B., 1996. Dinoflagellate cyst ecology: modeling and geological applica-tions. In: Jansonius, J., McGregor, D.C. (Eds.), Palynology: principles andapplications 3. American Association of Stratigraphic Palynologists Foun-dation, Texas, pp. 1249–1275.

Page 9: Palynological investigation of Holocene palaeoenvironmental changes in the coastal plain of Marathon (Attica, Greece)

K. Kouli et al. / Geobios 42 (2009) 43–51 51

Fouache, E., 1999. L’alluvionnement historique en Grèce Occidentale et auPéloponnèse. Bulletin de Correspondance Hellénique, École Françaised’Athènes, Supplément 35. Éditions De Boccard, Paris.

Grimm, E., 1992. Tilia and Tilia-graph: Pollen spreadsheet and graphicsprograms. Programs and Abstracts, 8th International PalynologicalCongress, Aix-en-Provence, 6-12 September 1992, pp. 56.

Head, M.J., Seidenkrantz, M.-S., Janczyk-Kopikowa, Z., Marks, L., Gibbard,P.L., 2005. Last Interglacial (Eemian) hydrographic conditions in thesoutheastern Baltic Sea, NE Europe, based on dinoflagellate cysts. Qua-ternary International 130, 3–30.

Jahns, S., 1993. On the Holocene vegetation history of the Argive Plain(Peloponnese, southern Greece). Vegetation History and Archaeobotany2, 187–203.

Jahns, S., 2003. A late Holocene pollen diagram from the Megaris, Greece,giving evidence for cultivation of Ceratonia siliqua L. during the last 2000years. Vegetation History and Archaeobotany 12, 127–130.

Kambouroglou, E., 1989. Eretria–paleogeographic and geomorphological evo-lution during the holocene. Ph.D. Thesis, University of Athens, Munici-pality of Eretria editions.(in Greek).

Kontopoulos, N., Avramidis, P., 2003. A late holocene record of environmentalchanges from the Aliki lagoon, Egion, North Peloponnesus, Greece. Qua-ternary International 111, 75–90.

Kraft, J.C., 1972. A reconnaissance of the geology of the sandy coastal areas ofeastern Greece and the Peloponnese–with special speculations on Middle-Late Helladic paleogeography (3000–4000 years before present). Universityof Delaware, Technical Report 9, Newark.

Kraft, J.C., Aschenbrenner, S.E., Rapp, G.J., 1977. Paleogeographic recon-struction of coastal Aegean archaeological sites. Science 195, 941–947.

Kraft, J.C., Rapp, G.J., Aschenbrenner, S.E., 1975. Late Holocene paleogeo-graphy of the coastal plain of the Gulf of Messenia, Greece, and itsrelationships to archaeological settings and coastal change. GeologicalSociety of America Bulletin 86, 1191–1208.

Lambeck, K., 1995. Late pleistocene and Holocene sea level change in Greeceand southwestern Turkey: a separation of eustatic, isostatic and tectoniccontributions. Geophysical Journal International 115, 960–990.

Lambeck, K., 1996. Sea level change and shore evolution in Aegean Greecesince Upper palaeolithic time. Antiquity 70, 588–611.

Lozios, S., 1993. Tectonic Analysis of Northeastern Attica MetamorphicFormations. Ph.D Thesis, University of Athens, Athens (in Greek). (unpu-blished).

Maroukian, H., Zamani, A., Pavlopoulos, K., 1993. Coastal retreat in the plainof Marathon (East Attica), Greece: cause and effects. Geologica Balcanica23, 67–71.

Moore, P.D., Webb, J.A., Collinson, M.E., 1991. Pollen analysis. BlackwellScience, Oxford.

Mudie, P.J., Aksu, A.E., Yasar, D., 2001. Late Quaternary dinoflagellate cystsfrom the Black, Marmara and Aegean seas: variations in assemblages,morphology and paleosalinity. Marine Micropaleontology 43, 155–178.

Pantelidou-Gofas, M., 1991. [The Neolithic Nea Makri: the building materials].The Archaeological Society at Athens Library Series 119, Athens. (in Greek).

Pantelidou-Gofas, M., 1995. [The Neolithic Nea Makri: the pottery]. TheArchaeological Society at Athens Library Series 153, Athens. (in Greek).

Pantelidou-Gofas, M., 1997. The Neolithic Attica. The Archaeological Societyat Athens Library Series 167, Athens.

Pantelidou-Gofas, M., 2005. [Tsepi Marathonos: the protohelladic cemetery].The Archaeological Society at Athens Library Series 235, Athens. (in Greek).

Pavlopoulos, K., Karkanas, P., Triantaphyllou, M., Karymbalis, E., 2003.Climate and sea level changes recorded during late Holocene in the coastalplain of Marathon, Greece. In: Fouache, E. (Ed.), The Mediterranean Worldenvironment and history. Elsevier, Paris, pp. 453–465.

Pavlopoulos, K., Karkanas, P., Triantaphyllou, M., Karymbalis, E., Tsourou,Th., Palyvos, N., 2006. Palaeoenvironmental evolution of the coastal plainof Marathon, Greece, during the late Holocene: depositional environment,climate and sea-level changes. Journal of Coastal Research 22, 424–438.

Petrakos, B., 1995. Marathon: an Archaeological Guide. The ArchaeologicalSociety at Athens Library Series 167, Athens.

Petrucci, F., Medioli, F.S., Cavazzini, R., Scott, D.B., 1983. Evaluation of theusefulness of foraminifera as sea level indicators in the Venice lagoon (N.Italy). Acta Naturalia de l’Ateneo Parmense 19, 63–77.

Reille, M., 1992–1998. Pollen et spores d’Europe et d’Afrique du Nord (1992).Suppléments 1 (1995), 2 (1998). Laboratoire de Botanique Historique etPalynologie, Marseille.

Renault-Miskovsky, J., 1981. Analyse pollinique des sediments de la grotte deKitsos (Lavrion-Greece). In: Lambert, N. (Ed.), La grotte préhistoriquede Kitsos (Attique). Édition ADPF, École Française d’Athènes, Paris, pp.633–655.

Scott, D.B., Piper, D.J.W., Panagos, A.G., 1979. Recent salt marsh and intertidalmudflat foraminifera from the western coast of Greece. Rivista Italiana diPaleontologia e Stratigrafia 85, 243–266.

Triantaphyllou, M.V., Pavlopoulos, K., Tsourou, T., Dermitzakis, M.D., 2003.Brackish marsh benthic microfauna and palaeoenvironmental changesduring the last 6000 years on the coastal plain of Marathon (SE Greece).Rivista Italiana di Paleontologia e Stratigrafia 109, 539–547.

Turner, J., Greig, J.R.A., 1975. Some holocene pollen diagrams from Greece.Review of Palaeobotany and Palynology 20, 171–204.

van Andel, T.H., Lianos, N., 1984. High resolution seismic reflection profiles forthe reconstruction of postglacial transgressive shorelines: an example fromGreece. Quaternary Research 22, 31–45.

van Geel, B., Bohncke, S.J.P., Dee, H., 1980-1981. A palaeoecological study ofan upper Late Glacial and Holocene sequence from ‘‘De Borchert’’, TheNetherlands. Review of Palaeobotany and Palynology 31, 367–448.

van Geel, B., Buurman, J., Brinkkemper, O., Schelvis, J., Aptroot, A., vanReenen, G., Hakbijl, T., 2003. Environmental reconstruction of a RomanPeriod settlement site in Uitgeest (The Netherlands), with a special referenceto coprophilous fungi. Journal of Archeological Sciences 30, 833–873.

van Geel, B., Coope, G.R., van der Hammen, T., 1989. Palaeoecology andstratigraphy of the Late Glacial type section at Usselo (The Netherlands).Review of Palaeobotany and Palynology 60, 25–129.

van Geel, B., Hallewas, D.P., Pals, J.P., 1983. A Late Holocene deposit under theWestfriese Zeedijk near Enkhuizen (Prov. of N-Holland, The Netherlands):palaeoecological and Archaeological aspects. Review of Palaeobotany andPalynology 38, 269–335.

Wall, D., Dale, B., Harada, K., 1973. Description of new fossil dinoflagellatesfrom the late Quaternary of the Black Sea. Micropaleontology 19, 18–31.