Navigation – Plan du site

The Canale di Comunicazione Traverso in Portus: the Roman sea harbour under river influence (Tiber delta, Italy)

Etude du Canale di Comunicazione Traverso à Portus : le port maritime antique de Rome sous influence fluviale (delta du Tibre, Italie)
Ferréol Salomon, Hugo Delile, Jean-Philippe Goiran, Jean-Paul Bravard et Simon Keay
p. 75-90

Résumés

Portus était le port maritime de Rome à l’époque impériale. En 42 apr. J.-C., le site d’implantation de ce port fut choisi à environ 3 km au nord d’Ostie, le long du littoral Tyrrhénien, mais en marge du Tibre. Si aujourd’hui on connaît bien Portus et sa façade maritime, on connaît moins bien Portus et son aspect fluvial. Selon nos connaissances actuelles, le Canale di Comunicazione Traverso est le seul canal attesté permettant la liaison entre le fleuve et les bassins portuaires. Comment les ingénieurs antiques ont-ils concilié l’existence d’une voie de navigation continue entre le système portuaire et le Tibre et la protection des bassins vis-à-vis des apports sédimentaires fluviaux ? Ces aménagements furent-ils efficaces ? Les objectifs de ce travail sont l’analyse précise des modalités de colmatage du Canale Traverso de manière à établir son rôle dans la sédimentation des bassins de Portus et la définition du fonctionnement et de l’usage de ce canal. Cette étude se base sur une relecture des données archéologiques et principalement sur des analyses granulométriques, et sur l’interprétation du diagramme de Passega. Une image assez complète des processus de dépôt en milieu portuaire est donnée à l’entrée du bassin de Trajan (TR-XIV). Ce diagramme de Passega est comparé à celui du Canale Traverso (CT-1) soumis davantage à l’influence du Tibre. Les résultats obtenus postulent en faveur d’un canal plutôt bien protégé de l’influence fluviale avec l’incursion épisodique des crues du Tibre. Cet article est également l’occasion d’étudier les rythmes de la sédimentation à Portus afin de définir l’usage de ce canal et les modalités de son entretien.

Haut de page

Notes de la rédaction

Article soumis le 20 janvier 2011, accepté le 7 octobre 2011.

Texte intégral

We are grateful to the Ecole Française de Rome, the British School at Rome, the University of Southampton, and to the CNRS (which granted the Homere project), for their financial and logistic support. In particular, we would like to thank Dr Michel Gras (Head of the Ecole Française) and Dr Yann Rivière (Director of the Classics department). We are very grateful to Lidia Paroli and Angelo Pellegrino of the Soprintendenza per i Beni Spacial Archeologici di Roma. Many thanks to Vincent Gaertner for his assistance in laboratory analysis conducted in the OMEAA laboratory at the University Lumière (Lyon 2). Many thanks to Olivia Saucier, Kieran Stanley and Kristian Strutt for reading the proofs of this article. Thanks also to the four reviewers and their helpful comments on this paper. We are also grateful to the Artemis programme for providing AMS radiocarbon dates. Thanks to Christine Oberlin for fruitful discussions on radiocarbon results. Finally, Daniele D’Ottavio and Marco Gizzi are acknowledged for their coring fieldwork.

Introduction

1Portus was the seaport of Rome during the Roman Imperial period. The construction of the harbour began in AD 42 under the reign of the Emperor Claudius. The harbour was located about 3 km north of Ostia, along the Tyrrhenian Coast, but on the margin of the Tiber. A few hundred metres east of Portus, the natural Tiber channel curved to the south near Ostia, where it flowed into the sea (fig. 1). At the beginning of the 2nd c. AD, Emperor Trajan complemented the Claudian harbour with a second basin (Juvenal, Saturae, XII, 75-78). To date, many sedimentological and palaeoenvironmental studies have been conducted at Portus, its basins and its maritime entrances (Arnoldus-Huyzendveld, 2005; Giraudi et al., 2007; Goiran et al., 2007, 2008; Bellotti et al., 2009; Giraudi et al., 2009; Goiran et al., 2010), but much less is known about the fluvial features at Portus (Salomon et al., 2010). Inscriptions made during the reigns of Claudius (CIL XIV 85 = ILS 207) and Trajan (CIL XIV 88 = CIL VI 964 = ILS 5797a) were found at Portus. They record the digging of canals linking the Tiber River to the sea. Today, a unique canal, the Fiumicino, is visible and still in use on the southern outskirts of Portus which was probably the Roman Fossa Traiana (Fea, 1924 a and b). This canal is shown on maps of Portus drawn during the Renaissance: for example in a fresco, painted by Ignazio Danti (1582; Galleria delle carte Geografiche of the Vatican Museum). The Canale di Comunicazione Traverso was a NNW-SSE channel almost perpendicular to the Fiumicino, which linked this canal to the southern access of the Trajanic basin (fig. 1). It was the only waterway that existed between the sea basins and the fluvial system (Keay et al., 2005). M. Reddé (1986) thinks that the rapid silting up of Portus’ basins was probably due to fluvial sedimentary input. How did Roman engineers preserve a continuous waterway from Portus’ basins to the Tiber River without accelerating the rate of siltation in the harbour? Or on the other hand, was this canal able to carry sediment away from the basins thanks to the increase in outflow? Sediment cores, allowed the analysis of the characteristics of the Canale Traverso in order to (i) establish its role in the sedimentation of Portus’ basins and (ii) to define the function and the operation of this canal.

Fig. 1 – Location map of Portus, the ancient sea harbour of Rome on the margins of the Tiber.
Fig. 1 – Localisation générale de Portus, l’ancien port maritime de Rome en marge du Tibre.

Fig. 1 – Location map of Portus, the ancient sea harbour of Rome on the margins of the Tiber. Fig. 1 – Localisation générale de Portus, l’ancien port maritime de Rome en marge du Tibre.

Setting

Geology and geography

2The Tiber River, 405 km long from the source to the Tyrrhenian Sea, originates in the Apennine Mountains (Mount Fumaiolo, 1268 m) and drains a 17375 km² basin. The mean annual discharge at the Ripetta gauging station (Rome, 43 km upstream the mouth) is 225 m3/s. The maximum historic discharge exceeds 3000 m3/s and the minimum discharge 60 m3/s. The mean suspended load was 7.2x106 t/a at the Ripetta station (1934-1950), before the impacts of large dams on sediment fluxes occurred (Iadanza and Napolitani, 2006).

3The Tiber delta has been studied since the 19th c. (Moro, 1871; Ponzi, 1875; Amenduni, 1884; Bocci, 1892), but knowledge on it has improved in the second half of the 20th c. (Dragone et al., 1967). Since the 1980’s, deep-coring studies have been completed in order to investigate Holocene stratigraphy (Belluomini et al., 1986; Belfiore et al., 1987; Bellotti et al., 1989, 1994, 1995; Milli, 1997; Amorosi and Milli, 2001; Bellotti et al., 2007). The Tiber delta results from two main influences, a combination of sea and fluvial processes. The rapid sea-level rise from 18000 BP to 7000 BP induced the submersion of the Pleistocene delta. The pace of sea-level rise reduced from 7000 BP to 5000 BP. Consequently, the Tiber delta adopted the lobate form in a progradational tendancy (Bellotti et al., 2007). The major migrations of the Tiber channel have been divided into three major phases: (i) During the Late Pleistocene, the palaeo-Tiber flowed into the northern part of the delta (Bellotti et al., 2007); (ii) The second stream way switched to the middle part of the present delta ca. 7000 BP (Bellotti et al., 2007); (iii) During the 9th-8th c. BC, the Tiber mouth moved to the south near Ostia Antica (Giraudi et al., 2009; Bellotti et al., 2011).

4Over the past twenty years, a lot of research has been conducted on the late Holocene period aiming to locate the successive Tiber palaeochannels (Segre, 1986; Arnoldus-Huyzendveld and Paroli, 1995; Arnoldus- Huyzendveld and Pellegrino, 2000), lagoons (Giraudi, 2002; Di Rita et al., 2009; Bellotti et al., 2011) and the coastline variations (Giraudi, 2004; Rendell et al., 2007; Bicket et al., 2009) more precisely. In central Italy, more intense fluvial activity seems to have occurred during Roman Imperial times and reduced during the Late Antiquity and the early Middle Ages (Benvenuti et al., 2006; Bicket et al., 2009). Some of those past hydrological events have been monitored in Rome and have been mentioned by ancient authors (Le Gall, 1953; Bersani and Bencivenga, 2001).

5The archaeological area of Portus has been the subject of many geoarchaeological and palaeoenvironmental studies. Studies focused on the environments that existed prior to the building of the harbour and on ancient harbour basins (Goiran and Morhange, 2003; Marriner et al., 2010). Pre-harbour aspects have been approached mainly through the study of the migration of the Tiber River in this area of the delta (Segre, 1986, Giraudi et al., 2007; Goiran et al., 2007; Giraudi et al., 2009; Bellotti et al., 2011) and through the position of the shoreline during historical times (Arnoldus-Huyzendveld, 2005). The harbour itself was studied through the reconstruction of basins’ shape, and through detailed sediment analysis (Arnoldus-Huyzendveld, 2005; Giraudi et al., 2007; Goiran et al., 2008; Bellotti et al., 2009; Giraudi et al., 2009; Goiran et al., 2010; Sadori et al., 2010). J.-P. Goiran et al. (2010) identified different sediment depositions in the two basins of Portus. The Claudian basin was a poorly protected basin that opened out to the sea; its sediment infill was mostly composed of sands as it was open to the influences of the sea wind and swell. The Trajanic basin was better protected; as a result, its sediment infill was mostly composed of silt and clay (Goiran et al., 2010). In with the continuity of these results, the present work focuses on the origin of sediments in the Canale Traverso and Trajan basin.

Archaeology and geoarchaeology

6The geoarchaeological work carried out at Portus would not have been possible without previous historical and archaeological research (Lugli and Filibeck, 1935; Lanciani, 1868; Testaguzza, 1970; Keay et al., 2005). The available data concerning the geographical units are the following:

7- The Fiumicino. In the Late Antiquity, the mouth of the Fiumicino channel was about 2000 m from the Tiber (Procopius, De Bello Gothico, 1, 26, 7-13). The western part of the deltaic system has been attributed to the rapid progradation of the coast over the last five centuries (Le Gall, 1953; Giraudi, 2004; Bersani and Moretti, 2008; fig. 1). This canal would have remained navigable until AD 1118 and perhaps even AD 1461 (Coccia, 1993, 2001; Paroli, 2004, 2005). Unfortunately, sedimentary archives have disappeared due to successive operations of dredging. For instance, in AD 1612, the canal was restored to its navigable state when Pope Paul V ordered for it to be dredged (Chiumenti and Bilancia, 1979; Paroli, 2005); also the restoration of the canal banks was implemented in the early 20th c. (Gatti, 1911). In the first monographs on the archaeology of Portus, the Fiumicino canal holds a prominent position. In several studies published in the early 19th c., C. Fea (1824 a and b) focused on this channel and identified the Fiumicino canal with the Roman Fossa Traiana, an ancient canal of Portus cited by Pliny the Younger. For S. Keay et al. (2005), the Fiumicino canal dates back to the Claudian period and was later incorporated into the new Trajanic complex.

8- The area of the Canale di Comunicazione Traverso. The Canale di Comunicazione Traverso is a 300 m long and up to 25 m wide channel. This canal is located in an area that was densely occupied after the construction of the harbour from the mid-1st c. AD to the late Middle Ages (Paroli, 2004, 2005). The Canale Traverso was probably surrounded by warehouses and storage areas as early as the mid-1st c. AD (Lugli and Filibeck, 1935; Keay et al., 2005). The recent excavation of the Basilica Portuense provided extensive data on the evolution of the urban sector. Three chronological phases were identified: (i) the port area which was occupied by administrative and commercial activities, along with warehouses, (ii) at the end of the 3rd and early 4th c. AD, the area became residential, then (iii) between the end of the 4th c. AD and the second half of the 5th c. AD, the basilica was built (Paroli, 2005, p. 258). The archaeological age of the Canale Traverso is not currently ascertained by any specific architectural studies. However, based on a study of the visible apparatus (Testaguzza, 1970), the Darsena-Canale Traverso-Fossa Traiana complex is thought to have been built at the same time. Thus, the discovery of a brick from Neronian times in one of the Darsena docks dates the Canale Traverso back to the 1st c. AD (Keay et al., 2005; Verduchi 2005).

A brief synthesis of palaeoenvironmental analysis

9Three recent publications (Goiran et al., 2010; Sadori et al., 2010; Mazzini et al., 2011) develop palaeoenvironmental reconstructions from macrofauna, microfauna (ostracods), and pollen analysis. Ostracod assemblages are related to water salinity (Carbonel, 1980). They are the best indicators to reconstruct marine and fluvial influences on a statistical base. Fig. 3 is a synthesis of J.-P. Goiran et al. (2010, in press), and I. Mazzini et al. (2011). Results concerning the ostracods groups were simplified into three palaeobiotopes: freshwater, brackish and marine waters. Each circle corresponds to a harbour basin unit defined by respective authors. Fig. 3 clearly indicates the contrast between the Claudian basin, mostly influenced by marine environment (cores CL-7, CL-2 and TR-IV), and the Trajanic channel area, characterised by a brackish palaeobiotope with temporary freshwater influence. Surprisingly enough, the sediment of Canale Traverso contains fewer freshwater ostracods (PTS-5) than the Trajanic access channel (TR-XIX, PTS-13) considering the basal layers of the harbour and of the canal during their activity. Could these results allow other freshwater sources to be considered? Indeed, near the TR-XIX core, thermae have been found (Lugli and Filibeck, 1935; fig. 2). The freshwater/brackish signals can then be due to inland water discharge from an aqueduct (Keay et al., 2005; Bedello-Tata and Bukowiecki, 2006) and not only from the Tiber River through the Canale Traverso.

Fig. 2 – Geomagnetic survey results (Keay et al., 2005) and archaeological data between the Claudius and the Trajan harbour: synthesis of successive cores location near the Canale Traverso.
Fig. 2 – Résultats des prospections géomagnétiques (Keay et al., 2005) et données archéologiques situées entre les bassins portuaires de Claude et de Trajan : synthèse des carottages réalisés à proximité du Canale Traverso.

Fig. 2 – Geomagnetic survey results (Keay et al., 2005) and archaeological data between the Claudius and the Trajan harbour: synthesis of successive cores location near the Canale Traverso. Fig. 2 – Résultats des prospections géomagnétiques (Keay et al., 2005) et données archéologiques situées entre les bassins portuaires de Claude et de Trajan : synthèse des carottages réalisés à proximité du Canale Traverso.

1: studied cores in this paper; 2: cores in J.-P. Goiran et al., 2010; 3: cores in C. Giraudi et al., 2009, L. Sadori et al., 2010, I. Mazzini et al., 2011; 4: bridges? (Lanciani, 1866; Keay et al., 2005); 5: streets ? (Keay et al., 2005).
1 : carottages étudiés dans cet article ; 2 : carottages présentés dans J.-P. Goiran et al., 2010 ; 3 : carottages présentés dans C. Giraudi et al., 2009, L. Sadori et al., 2010, I. Mazzini et al., 2011 ; 4 : ponts ? (Lanciani, 1866 ; Keay et al., 2005) ; 5 : routes ? (Keay et al., 2005).

Fig. 3 – Marine to freshwater palaeoenvironments in the Portus basin obtained by ostracods determination – A synthesis (Goiran et al., 2010; Mazzini et al., 2011).
Fig. 3 – Synthèse des données paléoenvironnementales obtenues pour les bassins de Portus par détermination de l’ostracofaune (Goiran et al., 2010 ; Mazzini et al., 2011).

Fig. 3 – Marine to freshwater palaeoenvironments in the Portus basin obtained by ostracods determination – A synthesis (Goiran et al., 2010; Mazzini et al., 2011). Fig. 3 – Synthèse des données paléoenvironnementales obtenues pour les bassins de Portus par détermination de l’ostracofaune (Goiran et al., 2010 ; Mazzini et al., 2011).

1: cores locations with portuary units (Goiran et al., 2009; Mazzini et al., 2011; Goiran et al., in press); all cores are represented according to the 3rd-5th c. sea level (Goiran et al., 2009). Only the portuary units data are shown; 2: harbour bottom; 3: ostracod groups with 3a: freshwater species, 3b: brackish species, and 3c: marine species.
1 : localisation des carottages avec leurs unités portuaires (Goiran et al., 2009; Mazzini et al., 2011; Goiran et al., sous presse) ; tous les carottages sont représentés en référence au niveau marin des IIIè-Vè apr. J.-C. (Goiran et al., 2009) ; 2 : fond des bassins portuaires ; 3 : ostracodes ; 3a : individus d’eau douce ; 3b : individus d’eau saumâtre ; 3c : individus caractéristiques du milieu marin.

Methods

10While recent articles have focused on macrofaunal bioindicators (shells), ostracods and pollen (Giraudi et al., 2009; Goiran et al., 2010; Sadori et al., 2010; Mazzini et al., 2011), this paper focuses on granulometric data and possible river influence on the canal. Core locations are presented on the fig. 2 with results of the geomagnetic survey obtained by fluxgate radiometer and published in S. Keay et al. (2005).

11Cores presented were collected using a mechanical rotary core barrel during several field excursions from 2004 to 2009. The objective of coring TR-XIV and CT-1 was to accurately characterise the sedimentary sequences of the access channel to the Trajanic basin and the Canale Traverso. Thus, the cores are around 10-m deep and include the complete harbour and canal sedimentation sequence. The cores were analysed in the OMEAA laboratory in Lyon.

12Magnetic susceptibility was measured three times every centimetre using a Bartington MS2E1 (Dearing, 1999). The value of susceptibility is recorded in CGS (centimeter, gram, second system) unit (corresponding to SI = CGS value * 0.4; Dearing, 1999). This non-destructive method efficiently separates different sedimentary units (Dearing, 1999), as attested in fig. 4 with a histogram depicting the CT-1 core. One colour rectangle corresponds to each unit. The x-axis represents the magnetic susceptibility in each 5 CGS. The occurrences of MS values in this range of 5 CGS are shown on the y-axis (fig. 4). Magnetic susceptibility offers a range of values with fairly consistent distributions for low values, but very large distributions for high values. Corresponding to the magnetic susceptibility, densitometric analysis was undertaken using an X-ray-scanner. Based on this high-resolution analysis, we sampled every 3 cm to 10 cm. Then, depending on sedimentary facies, we analysed a sample every 10 cm to 40 cm.

13Then, to get an overview of the texture of each of these samples, we sieved 30 g of the total fraction to distinguish between silt-clay (<63 µm), sand (63 µm-2 mm) and coarse fraction (>2 mm). Finally, on the sediment fraction <1.6 mm, accurate granulometrical analysis was undertaken using a Malvern Mastersizer 2000. Trask sorting index and the median grain have been used to describe the grain size distribution (Folk and Ward, 1957; Rivière, 1977). The interpretation of granulometric curves was based on the CM diagram, also called the Passega image (Passega, 1957; Bravard, 1983; Arnaud-Fassetta, 1998; Bravard and Peiry, 1999; Arnaud-Fassetta et al., 2003). This diagram uses the median (D50) and the coarsest percentile (D99) to determine depositional and transport processes.

14During this study, we used the biological mean sea level as determined on the quay of Portus (Goiran et al., 2009). It corresponds to the upper biological limit of barnacles dated at 2115±30 BP, calibrated AD 230-450. This biological, Roman sea level (R.s.l.) stands 0.8 m below the current biological sea level (Goiran et al., 2009). The studied cores are related to those current and R.s.l. This article will deal especially with the period from the 1st to the 7th c. AD. During this period, the sea level may have changed a few centimetres above and below the sea level of the 3rd-5th c. AD (Goiran et al., 2009). We do not exclude the possibility of limited compaction of the harbour sedimentary sequence, but without consequences for our interpretations.

15In order to establish an age-depth model, five radiocarbon dates were obtained on core CT-1. We focused on dating wood fragments, charcoal and seeds using the AMS dating technique. Calib 6 software was used for calibrating dates (tab. 1) with the continental curve (Reimer et al., 2004). A margin of error of 2 sigma was retained.

Fig. 4 – Magnetic susceptibility results for core CT-1 - Relevance of a high-resolution dataset (each centimetre) to determine sedimentary layers.
Fig. 4 – Résultats de susceptibilité magnétique pour le carottage CT-1.

Fig. 4 – Magnetic susceptibility results for core CT-1 - Relevance of a high-resolution dataset (each centimetre) to determine sedimentary layers. Fig. 4 – Résultats de susceptibilité magnétique pour le carottage CT-1.

Bars are grouped by sedimentary units and correspond to the occurrence of MS data (y-axis) included in a range of 5 CGS (x-axis).
Ce graphique valide la pertinence des données de haute résolution (chaque centimètre) pour définir des unités sédimentaires. Les barres sont identifiées selon les unités sédimentaires et correspondent à l’occurrence des valeurs de SM (en ordonnée) incluses dans un intervalle de 5 CGS (en abscisse).

Tab. 1 – Radiocarbon dating results.
Tab. 1 – Résultats des datations par le radiocarbone.

Samples

Laboratory samples

Sample description

Activity (in %)

Radiocarbon dating (in years BP)

Age calibated (Reimer et al., 2004) - 2ơ

CT1-A (344)

Lyon-6869

Charcoal

83.84 ± 0.25

1415 ± 30

600 to 660

CT1-C (554)

Lyon-7081

Wood

4.00 ± 0.26

1830 ± 30

90 to 245

CT 1-E (737)

Lyon-6894

Charcoal

78.54 ± 0.23

1940 ± 30

5 to 125

CT 1-G (792)

Lyon-6895

Seed

78.76 ± 0.23

1920 ± 30

25 to 130

CT1-B (1158)

Lyon-6870

Charcoal

60.55 ± 0.21

4030 ± 30

-2620 to -2475

Results

The detailed description of cores CT-1 and TR-XIV is presented below:

CT-1: Canale Traverso (fig. 5 and fig. 7)

16This sedimentary core focuses on the granulometric description. For more information about microfauna and pollen content of core PTS-5 (fig.2), we refer the reader to L. Sadori et al. (2010) and I. Mazzini et al. (2011).

17- Unit A is composed of yellow sterile layered sands (82%). The silt and clay fraction corresponds to 18% of the total dry sample. The sediment is moderately well sorted with a mean Trask index of 2. The median grain size of this unit is the highest of the whole sedimentary sequence (197 µm), but it rises progressively from the bottom of the core around 100 µm (graded suspension at a depth of 1183 cm, type 1) to 354 µm (medium sand) at the top (dominance of graded suspension and rolling, type 2). A layer at the bottom of the sequence is composed mostly of silts and clays (65%) with a median grain size around 13 µm. This sandy-silt layer has a coarse fraction with plant remains and seeds. A piece of charcoal has been dated to 4030±30 BP (2620-2475 BC). The global magnetic susceptibility value (MS) of this unit is around 156 CGS, ranging between 34 and 378 CGS.

18- Units B and C are composed of dark grey mud with shells and Posidonia, which is a marker of the marine influence. 95% of unit B is composed of silt and clay, a value reduced to 84% in unit C. The average median grain size for most of sequence B and C ranges between 10 µm and 13 µm. These units are characterised by a poor sorting index (mean of 3.5). The depositional processes point mostly to mixed graded and uniform suspensions (type 3 on the CM diagram; fig. 7). However, in units B-C, sample ‘741’ is a mix of rolled particles and graded/uniform suspension (type 4). The organic content of units B and C is high (10% of the dry weighted sediment). The coarse faction (>2 mm) is composed of shell fragments, clasts (rolled and not rolled), plant remains, charcoal, wood fragments and Posidonia. Seeds and charcoal have been radiocarbon dated 1920±30 BP (AD 25-130; Lyon-6895) and 1940±30 BP (AD 5-125; Lyon-6894) respectively, at the bottom of unit B. In the middle of unit C, a fragment of wood has been dated 1838±30 BP (AD 90-245; Lyon-7081). The mean MS is higher in unit B (38 CGS), than in unit C (9 CGS).

19- The sedimentary facies of unit D is a grey sandy-silt with shells and artefacts. Compared to unit C, a larger proportion of the dry weight sediment is composed of sand (31%) and coarse fraction larger than 2 mm (19%). The median grain size is very fine sand (67 µm). The sediment of unit D is very poorly sorted (mean of 5.6). Depositional processes are dominated by a mix of rolling and graded suspension (type 4); a mix of graded and uniform suspension (type 3) is represented in the lower part (sample ‘433’, type 3). We can observe a progressive decrease in the percentage of the organic material from 8.5% at the bottom to 5.6% at the top. Fragments of shells, clasts, wood fragments, vegetation remains, a lot of potshards and some Posidonia are present at the bottom of unit D. A later date has been obtained from a piece of charcoal: 1415±30 BP (AD 600-660; Lyon-6869). The MS value increases significantly to a mean of 512 CGS, ranging between 192 CGS and 842 CGS.

20- Units E and F (-68 cm to 180 cm in reference to the current sea level) are yellow sandy silt to laminated silt. The fine fraction increases from 67% (unit E) to 85% (unit F). The sorting index is low but slightly higher than in the lower units (4 to 3.6). The organic matter content decreases (5.14% to 3.2%). The depositional processes are mixed (types 3 and 4), even if some layers belong to the uniform suspension (type 5). The MS values are around 600 CGS.

TR-XIV: Access channel to the Trajanic basin (fig. 6)

21- Unit A is yellow sterile layered sand similar to unit A at the bottom of the CT-1 core. Sand is dominant, deposited as graded suspension (type 1), with a median grain size between 90 µm and 120 µm. The organic content is low (1.3%). The mean magnetic susceptibility value is 46 CGS.

22- The sedimentary environment changes abruptly since units B and D are composed of mud and sandy-mud with shells and Posidonia with 85% of fine sediments (<63 µm). They are characterised by a negative magnetic susceptibility around -25 CGS. The average median grain size is 13 µm. Compared to unit A, the organic matter content is multiplied by 6 with a maximum of 12%. The environment was relatively quiet with a mix of uniform and graded suspension processes (type 3). Between these two units, a sandy layer (70% of sand in unit C) was deposited by mixed rolling and graded suspension processes (type 4). The magnetic susceptibility increases to 30 CGS, in positive values. Sparse potshards have been collected in unit D.

23- Units E and F are quite similar to units B and D, but with a smaller proportion of fine sediments (78%) and a greater proportion of organic matter (up to 22%). Shells and Posidonia are more present. The dominant depositional processes are a mix of uniform and graded suspension. The major difference between the sedimentary units are given by the values ​​of magnetic susceptibility. Unit F has similar negative value, around -5 CGS, contrasting with unit E characterised by positive values with a mean of 7 CGS.

24- Units G and H are mostly composed of fine sediment (73%) but, between these two layers, the sandy content is similar to the fine fraction. A high frequency of shells has been observed in this intermediate level. In unit G the sandy content was deposited by mixed rolling and graded suspension processes increased. The magnetic susceptibility is positive with a mean value of 7 CGS. The sand content decreases in unit H which displayed a quieter environment with uniform and graded suspension mixing processes and with a low magnetic susceptibility (mean of -3 CGS).

25- The two upper layers (units I and J) are characterised by layers of silt and clay deposited by alternated uniform suspension and decantation. More than 99% of the total fraction is composed of fine fraction. In unit I, the magnetic susceptibility increases from 10/20 CGS at the bottom to 60 CGS at the top. Finally, those values in unit J are highly variable.

Fig. 5 – Canale Traverso, analysis of the core CT-1.
Fig. 5 – Le Canale Traverso, analyse de la carotte CT-1.

Fig. 5 – Canale Traverso, analysis of the core CT-1. Fig. 5 – Le Canale Traverso, analyse de la carotte CT-1.

1a: Posidonia; 1b: shells; 1c: pot shards; 2a: coarse sediments; 2b: sand; 3c: silt and clay; C/M diagram interpretation. 3: pure processes: 3a: rolling; 3b: rolling and graded suspension; 3c: graded suspension; 3d: uniform suspension; 3e: decantation; 4: mixed processes: 4a: mixed rolling/graded suspension + uniform suspension; 4b: mixed graded + uniform suspension.
1a : Posidonie ; 1b : coquille ; 1c : céramique ; 2a : fraction grossière ; 2b : sables ; 3c : limons et argiles ; Interprétation du diagramme C/M. 3 : processus purs : 3a: roulement ; 3b : roulement et suspension graduée ; 3c : suspension graduée ; 3d : suspension uniforme ; 3e : décantation ; 4 : processus mixtes : 4a : roulement/suspension graduée + suspension uniforme mixés ; 4b : suspension graduée et uniforme mixées.

Fig. 6 – The access channel to the Trajanic harbour, analysis of core TR-XIV.
Fig. 6 – Le chenal d’accès au port de Trajan, analyse de la carotte TR-XIV.

Fig. 6 – The access channel to the Trajanic harbour, analysis of core TR-XIV. Fig. 6 – Le chenal d’accès au port de Trajan, analyse de la carotte TR-XIV.

1a: Posidonia; 1b: shells; 1c: pot shards; 2a: coarse sediments; 2b: sand; 3c: silt and clay; C/M diagram interpretation. 3: pure processes: 3a: rolling; 3b: rolling and graded suspension; 3c: graded suspension; 3d: uniform suspension; 3e: decantation; 4: mixed processes: 4a: mixed rolling/graded suspension + uniform suspension; 4b: mixed graded + uniform suspension.
1a : Posidonie ; 1b : coquille ; 1c : céramique ; 2a : fraction grossière ; 2b : sables ; 3c : limons et argiles ; Interprétation du diagramme C/M. 3 : processus purs : 3a: roulement ; 3b : roulement et suspension graduée ; 3c : suspension graduée ; 3d : suspension uniforme ; 3e : décantation ; 4 : processus mixtes : 4a : roulement/suspension graduée + suspension uniforme mixés ; 4b : suspension graduée et uniforme mixées.

Fig. 7 – Cross-sections from the Canale Traverso to the access channel to Trajanic harbour, and CM diagram of the CT-1 and TR-XIV cores.
Fig. 7 – Transect longitudinal du Canale Traverso au chenal d’accès au basin de Trajan. Présentation des diagrammes CM pour les carottages CT-1 et TR-XIV.

Fig. 7 – Cross-sections from the Canale Traverso to the access channel to Trajanic harbour, and CM diagram of the CT-1 and TR-XIV cores. Fig. 7 – Transect longitudinal du Canale Traverso au chenal d’accès au basin de Trajan. Présentation des diagrammes CM pour les carottages CT-1 et TR-XIV.

1a: Posidonia; 1b: shells; 1c: pot shards; 1d: canal or harbour basin bottom; 2a: coarse sediments; 2b: sand; 3c: silt and clay; C/M diagram interpretation. 3: pure processes: 3a: rolling; 3b: rolling and graded suspension; 3c: graded suspension; 3d: uniform suspension; 3e: decantation; 4: mixed processes: 4a: mixed rolling/graded suspension + uniform suspension; 4b: mixed graded + uniform suspension.
1a : Posidonie ; 1b : coquille ; 1c : céramique ; 1d : fond du canal ou du port ; 2a : fraction grossière ; 2b : sables ; 3c : limons et argiles ; Interprétation du diagramme C/M. 3 : processus purs : 3a: roulement ; 3b : roulement et suspension graduée ; 3c : suspension graduée ; 3d : suspension uniforme ; 3e : décantation ; 4 : processus mixtes : 4a : roulement/suspension graduée + suspension uniforme mixés ; 4b : suspension graduée et uniforme mixées.

Discussion

26The analysis of the two studied cores allows us to determine three sedimentary phases: the pre-harbour environments, the harbour and canal in activity, and the final infill.

Pre-harbour deposits

27In the CT-1 and TR-XIV cores, basal units A correspond to pre-harbour deposits characterised by yellow sterile layered sands. In support of this interpretation, a piece of charcoal sampled 10 m below current sea level has been dated 4030±30 BP (2620-2475 BC). The sediment facies is similar to all the Portus’ basal units (Goiran et al., 2010, in press). On the Passega diagram, sediments plot parallel to the straight line of perfect sorting C (D99) = M (D50). Depositional processes correspond to graded suspension (type 1) and rolling (type 2; fig. 7).

The Canale Traverso geometry and location as evidence of Roman planning

28This channel is relatively narrow compared to other channels of the Portus complex. After re-analysing the results of a geomagnetic survey (Keay et al., 2005; fig. 2), it seems pretty clear that the channel widens towards the Fossa Traiana, thus facilitating the movement of ships entering this channel. On the side of the docks, the canal opens into an approximately 40-m wide space, opened to the north. Strong brick walls, over 3-m high (4 m above the Roman sea level) are still visible today. In an upstream-downstream perspective (from the Tiber channel to the Canale Traverso), it can be hypothesised that there is a progressive reduction of the hydraulic cross section and of the flow along the derivation canals (fig. 1). The natural Tiber channel was 100 m to 120 m wide in the 1st c. AD, which is confirmed by L. Bertacchi (1960) in the former Tiber channel or Fiume Morto near Ostia. The ancient Fossa Traiana was around 50-m wide (Testaguzza, 1970). Finally, the Canale Traverso was 25-m wide (see magnetical survey in S. Keay et al., 2005). In the successive reaches of the canals (Fossa Traiana, Canale Traverso), widths were divided twice into two. We cannot precisely locate the palaeochannel of the Tiber near ancient Portus, but in a logical upstream and downstream continuity, the water in the Canale Traverso would have flowed perpendicular or in the opposite direction of flow of the palaeo-Tiber. In this deltaic plain, the flow of water in the Canale Traverso should be small. This geometric data is interesting because it may demonstrate a clear awareness of the problem arising when planning the position of the port and of the river system by the ancient architects and engineers responsible for the construction of Portus. We are now able to see how through a succession of canals and thanks to their optimal orientation, Romans tried to reduce the carrying capacity of the channel entering into the harbour basins, while ensuring the presence of a waterway between the river and the harbour. However, one question remains unanswered: was this configuration effective against sedimentation?

Harbour and canal sedimentation occurred in a common quiet brackish environment

29The katolimenic limit corresponds to the bottom of a harbour sedimentary sequence (Goiran et al., 2010), between pre-harbour deposit (the yellow sterile layered sand) and the harbour mud, characteristic of a quiet environment protected from the sea’s influence. In the TR-XIV core, the first potshards were found in unit D. Thus at this location, the channel access to the Trajanic harbour is 7 (unit B) or 6.5 (unit D) meters deep below the R.s.l. From 7 m to 3.30 m below R.s.l. (unit B to H), the sedimentary sequence of the TR-XIV core is mostly composed of muds with shells and Posidonia fibers. The harbour deposits of the TR-XIV core correspond to the other cores described in this area of Portus and deposited under high marine influence (cores TR-XIX, TR-XX in Goiran et al., 2010). The environmental conditions under which deposition occurred were relatively quiet, with a mix of graded and uniform suspension processes. Some levels display energetic hydrodynamic condition with sediment originating from a mix of graded and rolling processes. These sandy units occur mostly in units C, E and G. The Canale Traverso is only 5.5 m below R.s.l. (CT-1). It is an unusual fluvial canal because it has conditions similar to those of a harbour basin. Indeed, units B and C are composed of mud with shells and Posidonia, corresponding to the first phase of deposition. The depositional processes are primarily represented by mixed uniform and graded suspension, similar to those of core TR-XIV. The presence of Posidonia validates the hypothesis of the marine influence in the canal, succeeded by a brackish environment with a freshwater input (Mazzini et al., 2011). Most significantly, the fluvial influence is deduced from the presence of Alnus pollen which were found trapped inside the sediments, and probably transported and deposited there by the Tiber (Sadori et al, 2010). The bottom of this canal sediment has been dated from the 1st c. to the beginning of the 2nd c. (seed-Lyon-6895/charcoal-Lyon-6894). Unit C is dated from the end of the 1st c. AD to the middle of the 3rd c. AD (wood-Lyon-7081). It is impossible to specifically date the construction of the canal between Claudius and Hadrian reigns. Whatever the answer, the basal unit B and probably unit C correspond to the first canal deposits.

Last sedimentation of Portus waterways

30The closely related evolution, as described in the first metre of the canal’s sediments infill (CT-1) and the access channel to the Trajanic harbour (TR-XIV), are not characteristic of the upper units. Unit D of the Canale Traverso demonstrates an increase in hydrodynamism, shown by a coarser median grain size (10-13 µm to 67 µm). The mixed processes of graded suspension and rolling are dominant. This layer has not been found in the TR-XIV core which displays constant conditions from unit B to unit H, with mixed uniform and graded suspension. In unit D of the CT-1 core, one meter below the former R.s.l., a piece of charcoal has been dated to the 7th c. AD (charcoal-Lyon-6869). Unfortunately, no dating was possible in the access channel above -3 m (R.s.l). However, on the fresco painted by Ignazio Danti in 1582, the southern end of the Canale Traverso, close to the Fiumicino canal, is plugged by sediments, and stagnant water is visible in the access channel which is both disconnected from the sea and the fluvial system. The upper units of the CT-1 core can be interpreted as the deposition of this upstream sediment blockage. In the same way, the upper units I and J of the TR-XIV core correspond to swamp and floodplain deposits, showing the alternation of uniform suspension and decantation processes. These deposits correspond to the upper part of the PTS-5 and 13 cores (Mazzini et al., 2011; fig. 3), with a high frequency of freshwater ostracods groups.

A canal and a harbour silted up with sediment transported by Tiber River floods?

31When operational, the Canale Traverso was a quiet environment similar to the environments present at the entrance of the Trajanic harbour. Another fundamental question is whether mixed rolling and graded suspension results from a marine origin (Claudius basin) or from a fluvial origin. In the CM diagram, the mean coarsest grain-size percentile (D99) present in type 4 indicates the maximum transport capacity of particles. This mean D99 grain of the CT-1 core is coarser (1379 µm) than the mean D99 of the TR-XIV core (1102 µm), with rolled particles. Therefore, it is hypothesised that there is an influence of the Tiber River’s flow (floods in particular) in the Canale Traverso (fig. 7). In this area, future work will ascertain whether fine sediments originate mainly from the Tiber River through the Canale Traverso. A preliminary answer is that the sedimentation rate in the bottom units of the Canale Traverso is higher than in the access channel to the Trajanic harbour (fig. 8; Goiran et al., 2010). So far, it is risky to attribute the presence of flood deposits in the Canale Traverso to a regional hydroclimatic crisis context. Indeed this canal is a specific sediment trap mainly controlled by anthropic influences. Besides, at a certain period, a canal lock may have been installed between the Fossa Traiana and the Canale Traverso (Testaguzza, 1970).

Water depth and boat draughts

32Fig. 8 synthesises all the published radiocarbon data concerning the access channel to Trajan’s basin in relation to the R.s.l. (Goiran et al., 2009, 2010). These dates are confronted by the altitudinal variations of the bottom of the harbour, called 'mesolimenic limit' (Goiran et al., 2010), and also to different Roman boat draughts (Boetto, 2010). To compute the sedimentation rate, we used the mean radiocarbon date. First of all, the Canale Traverso was not deep when it was excavated (5.5 m), while the access channel to the Trajanic basin was 6.8 m (TR-XX), 7 m (TR-XIV) to 8 m (TR-XI) deep (Goiran et al., 2010; fig. 8). Thus, a slope from the Canale Traverso to the Claudian basin may be deduced from these data. Thus, when it excavated, the canal was probably designed to accommodate vessels with a shallow draughts (fig. 8). This result is synonymous with the relatively limited width of the canal (25 m; Boetto, 2006, 2010). Using the radiocarbon dates, the sedimentation rate may be divided into two phases. In the TR-XI core from -8 m to -6 m (R.s.l.), the bottom of the access channel aggraded by approximately 1.1 cm/a. The two first metres of TR-XX (from 6.8 m to 4.3 m below R.s.l.) aggraded faster than TR-XI, at a rate of 1.6 cm/a. Finally, the CT-1 core displays a fast accretion rate (2.6 cm/a) for the bottom 2.3 m. It can be deduced that in the access channel to the Trajanic basin, the closer to the Canale Traverso, the faster the accretion of the bottom of the channel. This observation can also apply to the river sediment supply which originated from the Tiber River. The second aggradational phase is more differentiated. The channel bottoms of TR-XI and TR-XX aggraded at a higher rate from 1.6 cm/a (TR-XX -6 m to -3 m R.s.l.) to 4.2 cm/a (TR-XI -4.3 m to -3 m R.s.l.). An inversion of dates occurs in the TR-XI core and could be interpreted as a phase of dredging (Marriner and Morhange, 2006). Unfortunately, there is no date for the last 3 m (R.s.l.) inside the access channel to Trajan’s basin. After a first phase of rapid siltation in the canal (2.6 cm/a), the second phase displays a slower rate of sedimentation with coarser sediments (0.45 cm/a). Theoretically, the canal would have filled with sediments becoming finer at the top of the sequence. In reality, sediment is coarser than below, with a reduction in the rate of siltation. The canal may then have been dredged at the base of Unit D (the abrupt change in magnetic susceptibility could be reconciled with that location). If the depth of the canal is maintained by dredging at about -2.3 m below the Roman sea level, small boats (70 t), like those found in the northern part of Claudius harbour (Boetto, 2006), could have navigated along this canal before the 7th c. AD. At the beginning of the 5th c. AD, Philostrogius (Ecclesiastical History, XII, 3 in Le Gall, 1953) described three basins in Portus, probably the Claudius and Trajanic basins, along with the darsena (Le Gall, 1953). Also Philostrogius specified that the Fossa Traiana was the middle of the port. He may have seen the Canale Traverso, originating from the Fossa Traiana, flowing into the access channel between the Claudius and Trajanic basins on one side and the darsena on the other side.

Fig. 8 – Harbour infill, water depth and ships draught data. Comparisons of these parametres by chronology.
Fig. 8 – Compilations des données disponibles à Portus pour comprendre chronologiquement la relation entre le remblaiement du port, la lame d’eau disponible et le tirant d’eau des bateaux.

Fig. 8 – Harbour infill, water depth and ships draught data. Comparisons of these parametres by chronology. Fig. 8 – Compilations des données disponibles à Portus pour comprendre chronologiquement la relation entre le remblaiement du port, la lame d’eau disponible et le tirant d’eau des bateaux.

Conclusions

33The relationship between ancient harbour basins and the fluvial system of the Tiber delta is a complex and crucial question. This paper takes advantage of the plurality of studies published on Portus in the recent years. In a new perspective, the history, harbour and naval archaeology, fluvial and ancient harbour geoarchaeology, were combined with new palaeoenvironmental studies in order to give a renewed description of Portus basins. Environmental and depositional conditions are quite similar at the entrance of the Trajanic basin and in the Canale Traverso. Granulometrical analyses provided new CM diagrams (Passega) about the conditions of sediment deposition in the harbour, with a wider range of processes in the access channel to Trajanic harbour (TR-XIV). The Canale Traverso is prone to higher energy flows coming from the Tiber (CT-1). This approach allows at better understanding of sediment budgets and functioning in the Portus harbour. Obviously, the basins silted up inexorably, despite the initial project of reducing sediment input from the Tiber. The sediment budget depended on preventive actions (e.g., geometry and configuration of the canals sheltered from the Tiber flow direction; or maybe canal lock in the Canale Traverso according to Testaguzza, 1970), on adaptive actions (e.g., construction of internal moles; Lugli and Filibeck, 1935; Keay et al., 2005) and curative actions (dredging). Future research may provide a more accurate vision, in particular the modelling of the waterflow in Portus and in the canals. Finally, the Canale Traverso is the starting point for future research on the canals directly connected to the Tiber, like the Canale Romano and the Northern Canal (Keay et al., 2005; Salomon et al., 2010).

Haut de page

Bibliographie

Amenduni G. (1884) – Sulle opere di bonificazione della plaga litoranea dell’Agro Romano che comprende le paludi e gli stagni di Ostia, Porto, Maccarese e delle terre vallive di Stracciacappa, Baccano, Pantano e Lago dei Tartari. Relazione del progetto generale, 15, 7, 36 p.

Amorosi A., Milli S. (2001) – Late Quaternary depositional architecture of Po and Tevere river deltas (Italy) and worldwide comparison with coeval deltaic successions. Sedimentary Geology 144, 357-375.

Arnaud-Fassetta G. (1998) – Dynamiques fluviales holocènes dans le delta du Rhône. Ph.D thesis, University of Provence (Aix-Marseille 1). Atelier National de Reproduction des Thèses, Villeneuve d’Ascq, 358 p.

Arnaud-Fassetta G., Carre M.-B., Marocco R., Maselli-Scotti F., Pugliese N., Zaccaria C., Bandelli A., Bresson V., Manzoni G., Montenegro M.E., Morhange C., Pipan M., Prizzon A., Siché I., (2003) – The site of Aquileia (northeastern Italy): example of fluvial geoarchaeology in a Mediterranean coastal plain. In Arnaud-Fassetta G., Provansal M. (Eds.) Deltas 2003. Géomorphologie: relief, processus, environnement 4, 223-241.

Arnoldus-Huyzendveld A. (2005) – The natural environment of the Agro Portuense. In Keay S., Millett M., Paroli L., Strutt K. (Eds.) Portus, an archaeological survey of the port of imperial Rome. The British School at Rome, Archaeological Monographs of the British School at Rome 15, London, 14-30.

Arnoldus-Huyzendveld A., Paroli L. (1995) – Alcune considerazioni sullo sviluppo storico dell’ansa dell Tevere presso Ostia e sul porto-canale. Archeologia Laziale 12, 383-392.

Arnoldus-Huyzendveld A., Pellegrino A. (2000) – Traces of historical landscapes preserved in the coastal area of Rome. Memorie Descrittive della Carta Geologica d’Italia 54, 221-222.

Bedello Tata M., Bukowiecki E. (2006) – Le acque e gli acquedotti nel territorio ostiense e portuense. Ritrovamenti e ricerche recenti. Mélanges de l’Ecole Française de Rome 118, 463-526.

Belfiore A., Bellotti P., Carboni M.G., Chiari R., Evangelista S., Tortora P., Valeri, P. (1987) – Il delta del Tevere: le facies sedimentarie della conoide sommersa. Un’analisi statistica dei caratteri tessiturali, microfaunistici e mineralogici. Bolletino della Società Geologica Italiana 106, 425-445.

Bellotti P., Carboni M.G., Milli S., Tortora P., Valeri P. (1989) – La piana deltizia del Fiume Tevere: analisi di facies ed ipotesi evolutiva dell’ultimo 'low stand' glaciale all’attuale. Giornale di Geologia 51, 71-91.

Bellotti P., Chiocci F.L., Milli S., Tortora P., Valeri P. (1994) – Sequence stratigraphy and depositional setting of the Tiber delta: integration of high-resolution seismics, well logs, and archeological data. Journal of Sedimentary Research-Section B-Stratigraphy and Global Studies 64, 416-432.

Bellotti P., Milli S., Tortora P., Valeri P. (1995) – Physical stratigraphy and sedimentology of the Late Pleistocene-Holocene Tiber Delta depositional sequence. Sedimentology 42, 617-634.

Bellotti P., Calderoni G., Carboni M.G., Di Bella L., Tortora P., Veleri P., Zernitskaya V. (2007) – Late Quaternary landscape evolution of the Tiber River delta plain (Central Italy): new evidence from pollen data, biostratigraphy and 14C dating. Zeitschrift für Geomorphologie 5, 505-534.

Bellotti P., Mattei M., Tortora P., Valeri P. (2009) – Geoarchaeological investigations in the area of the imperial harbours of Rome. Méditerranée, 112, 51-58.

Bellotti P., Calderoni G., Di Rita F., D’Orefice M., D’Amico C., Esu D., Magri D., Preite Martinez M., Tortora P., Valeri P. (2011) – The Tiber River Delta Plain (Central Italy): Coastal evolution and implications for the Ancient Ostia roman settlement. Holocene in press.

Belluomini G., Iuzzolini P., Manfra L., Mortari R., Zalaffi M. (1986) – Evoluzione recente del delta del Tevere. Geologica Romana 25, 213-234.

Benvenuti M., Mariotti-Lippi M., Pallecchi P., Sagri M. (2006) – Late-Holocene catastrophic floods in the terminal Arno River (Pisa, Central Italy) from the story of a Roman riverine harbour. The Holocene 16, 863-876.

Bersani P., Benvivenga M. (2001) – Le piene del Tevere a Roma dal V secolo a.C. all’ano 2000. Servizio idrografico e mareografico nazionale, Dipartimento per i Servizi Tecnici Nazionali, Roma, 100 p.

Bersani P., Moretti D. (2008) – Evoluzione storica della linea di costa in prossimità della foce del Tevere. L’Acqua 5, 77-88.

Bertacchi L. (1960) – Elementi per una revisione della topografia ostiense. Atti della Accademia nazionale dei Lincei 8, 8-32.

Bicket A.R., Rendell H.M., Claridge A., Rose P., Andrews J., Brown F.S.J. (2009) – A multiscale geoarchaeological approach from the Laurentine shore (Castelporziano, Lazio, Italy). In Ghilardi M., Fouache E., Chiverrell R. (Eds.) Geoarchaeology: Human-environment Connectivity. Géomorphologie: relief, processus, environnement 4, 257-270.

Bocci D. (1892) – Il delta tiberino. Tipografia del Genio civile, Roma, 33 p.

Boetto G. (2006) – Les navires de Fiumicino (Italie) : architecture, matériaux, types et fonctions. Contribution à l’étude du système portuaire de Rome à l’époque impériale. Ph.D thesis, University of Provence (Aix-Marseille 1), 267 p.

Boetto G. (2010) – Le port vu de la mer: l’apport de l’archéologie navale à l’étude des ports antiques. Bolletino di Archeologia On line, special issue: XVII International Congress of Classical Archaeology, Roma 22-26 September 2008.

Bravard J.-P. (1983) – Les sédiments fins des plaines d’inondation dans la vallée du Haut-Rhône. Revue de géographie alpine, 71, 363-379.

Bravard J.-P., Peiry J.-L. (1999) – The CM image as a tool for the classification of alluvial units and floodplains along the river continuum. In Marriott S., Alenxander J., Hey R. (Eds.) Floodplains: Interdisciplinary Approaches. Geological Society, London, 259-268.

Carbonel P. (1980) – Les ostracodes et leur intérêt dans la définition des écosystèmes estuariens et de plateforme continentale. Essais d’application à des domaines anciens. Mémoires de l’Institut de géologie du Bassin d’Aquitaine, 11, Bordeaux, 350 p.

Chiumenti L., Bilancia F. (1979) – La campagna romana antica, medioevale e moderna. Edizione redatta sulla base degli appunti lasciati da Giuseppe e Francesco Tomassetti VI. Vie Nomentana e Salaria, Portuense, Tiburtina (Arte e archeologia. Studi e documenti 16-17). L.S. Olschki Editore, Florence, 616 p.

Coccia S. (1993) – Il 'Portus Romae' fra tarda antichità ed altomedioevo. In Storia economica di Roma nell’Altomedioevo alla luce dei recenti scavi archeologici. Firenze: Atti del Seminario di Roma. Atti del Seminario (Roma 2-3 aprile 1992). Florence, All’Insegna del Giglio, 183-188.

Coccia S. (2001) – Il recinto fortificato dell’episcopio di Porto come epilogo di una crisi urbana. In Cancellieri S. (Ed.) L’Episcopio di Porto presso Fiumicino. Metodo e prassi nel restauro architettonico. Ministero per i Beni e le Attività Culturali - Soprintenza per i Beni Architettonici e il Paesaggio, Rome, 15-35.

Dearing J. (1999) – Environmental magnetic susceptibility. Using the Bartington MS2 System 32, 54 p.

Di Rita F., Celant A., Magri D. (2009) – Holocene environmental instability in the wetland north of the Tiber delta (Rome, Italy): sea-lake-man interactions. Journal of Paleolimnology 44, 51-67.

Dragone F., Maino A., Malatesta A., Segre A. (1967) – Note illustrative del Foglio 149 Cerveteri della Carta Geologica d’Italia. Servizio Geologico d’Italia 4, 1-93.

Fea C. (1824a) – Alcune osservazioni sopra gli antichi porti di Ostia ora Fiumicino. L. Contedini, Rome.

Fea C. (1824b) – La Fossa Traiana confermata al Sig. Cav. Ludovico Linotte. L. Contedini, Rome.

Folk R.L., Ward W.C. (1957) – Brazos river bar: a study in the significance of grain size parameters. Journal of Sedimentary Petrology 27, 3-26.

Gassend J. (1982) – Le navire antique du Lacydon, Marseille. Musée d’Histoire de Marseille, Marseille, 149 p.

Gatti E. (1911) – Fiumicino. Avanzi di antiche fabbriche scoperte nell’Isola Sacra, presso S. Ippolito. Notizie degli Scavi di Antichità, 410-416.

Giraudi C. (2002) – Evoluzione ambientale tardo-olocenica nell’area comprente il sto eneolitico di Maccarese (Fiumicino). In Mafredini A. (Ed.) Le dune, il lago, il mare. Una communità di villaggio dell’Età del Rame a Maccarese. Istituto italiano di preistoria e protostoria, Firenze, 25-35.

Giraudi C. (2004) – Evoluzione tardo-olocenica del delta del Tevere. Il Quaternario-Italian Journal of Quaternary Sciences 17, 477-482.

Giraudi C., Tata C., Paroli L. (2007) – Carotaggi e studi geologici a Portus: il delta del Tevere dai tempi di Ostia Tiberina alla costruzione dei porti di Claudio e Traiano. The Journal of Fasti Online.

Giraudi C., Tata C., Paroli L. (2009) – Late Holocene evolution of Tiber river delta and geoarchaeology of Claudius and Trajan Harbor, Rome. Geoarchaeology 24, 371-382.

Goiran J.-P., Morhange C. (2003) – Géoarchéologie des ports antiques de Méditerranée : problématiques et études de cas. Topoï 11, 645-667.

Goiran J.-P., Ognard C., Tronchère H., Canterot X., Cluze J.A. (2007) – Recent geo-archeological findings of Portus, the ancient Harbour of Rome. In Cinque A. (Ed.) Abstracts of International Congress on: People/environment relationships from Mesolithic to Middle Ages: recent Geo-Archaeological findings in Southern Italy. Salerno, 4-7 September 2007. Abstract volume, 30-31.

Goiran J.-P., Tronchère H., Carbonel P., Salomon F., Djerbi H., Ognard C., Lucas G., Colalelli U. (2008) – Portus, la question de la localisation des ouvertures du port de Claude : approche géomorphologique. Mélanges de l’Ecole Française de Rome, 121, 217-228.

Goiran J.-P., Tronchère H., Collalelli U., Salomon F., Djerbi H. (2009) – Découverte d’un niveau marin biologique sur les quais de Portus: le port antique de Rome. Méditerranée, 112, 59-67.

Goiran J.-P, Tronchère H., Salomon F., Carbonel P., Djerbi H., Ognard C. (2010) – Palaeoenvironmental reconstruction of the ancient harbors of Rome: Claudius and Trajan’s marine harbors on the Tiber delta. Quaternary International 216, 3-13.

Goiran J.-P, Salomon F., Tronchère H., Djerbi H., Carbonel P., Ognard C., Oberlin C. (in press) – Géoarchéologie des ports de Claude et de Trajan, Portus, delta du Tibre. Mélanges de l’Ecole Française de Rome, 123, 155-234.

Iadanza C., Napolitani F. (2006) – Sediment transport time series in the Tiber River. Physics and Chemistry of the Earth 31, 1212-1227.

Juvenal – Satires/Saturae. Tome 2, Livres IX-XXVIII. French translation by P. de Labriolle et F. Villeneuve, Les Belles Lettres, Paris, 42, 1979.

Keay S., Millett M., Paroli L. (2005) – Portus: An archaeological survey of the Portus of Imperial Rome. British School at Rome, Archaeological Monographs of the British School at Rome 15, London, 360 p.

Lanciani R. (1868) – Ricerche topografiche sulla città di Porto. Annali dell’Instituto di Corrispondenza. Archeologica 40, 144-195.

Le Gall J. (1953) – Le Tibre, fleuve de Rome dans l’antiquité. Presses universitaires de France, Paris, 368 p.

Lugli G., Filibeck G. (1935) – Il porto di Roma imperiale e l’agro portuense. Officine dell’Istituto Italiano d’Arti Grafiche.

Marriner N., Morhange C. (2006) – Geoarchaeological evidence for dredging in Tyre’s ancient harbour, Levant. Quaternary Research 65, 164-171.

Marriner N., Morhange C., Goiran J.-P. (2010) – Coastal and ancient harbor geoarchaeology. Geology Today 26, 21-27.

Mazzini I., Faranda C., Giardini M., Giraudi C., Sadori L. (2011) – Late Holocene palaeoenvironmental evolution of the Roman harbour of Portus, Italy. Journal of Paleolimnogy 46, 243-256.

Milli S. (1997) – Depositional setting and high-frequency sequence stratigraphy of the Middle-Upper Pleistocene to Holocene deposits of the Roman Basin. Geologica romana 33, 99-136.

Moro G. (1871) – Lo stagno di Ostia. Monografia geologica ed idraulica, Firenze, 43 p.

Paroli L. (2004) – Il porto di Roma nella tarda antichità. In Gallina Zevi A., Turchetti R. (Eds.) Le strutture dei porti de degli approdi antichi (ANSER II). Rubbettino Editore, Soveria Mannelli, 247-266.

Paroli L. (2005) – The Basilica Portuense. In Keay S., Millet M., Paroli L., Strutt K. (Eds) Portus. An Archaeological Survey of th Port of Imperial Rome. BSR Archaeological Monographs 15, London, 258-268.

Passega R. (1957) – Texture as characteristic of clastic deposition. Bulletin of American Association of Petroleum Geologists 4, 1952-1984.

Pomey P., Tchernia A. (1978) – Le tonnage maximum des navires de commerce romains. Archeonautica, 2, 233-251.

Pomey P., Rieth, E. (2005) – L’archéologie navale. Editions Errance, Paris, 224 p.

Ponzi G. (1875) – Il delta del Tevere: storia naturale del Tevere. Bollettino della Società Geografica Italiana 12, 1-22.

Procopius – De Bello Gothico. Edited by Dewing H.B. 7 volumes. Loeb Classical Library, 1914-1940. Greek text and English translation.

Reddé M. (1986) – Mare Nostrum : les infrastructures, le dispositif et l’histoire de la marine militaire sous l’Empire Romain. BEFAR, 260, Ecole Française de Rome, 736 p.

Reimer P.J., Baillie M.G.L., Bard E., Bayliss A., Beck J.W., Bertrand C., Blackwell P.G., Buck C.E., Burr G., Cutler K.G., Damon P.E., Edwards R.L., Fairbanks R.G., Friedrich M., Guilderson T.P., Hogg A.G., Hughen K.A., Kromer B., McCormac F.G., Manning S., Bronk Ramsey C., Reimer R.W., Remmele S., Southon J.R., Stuiver M., Talamo S., Taylor F.W., van der Plicht J., Weyhenmeyer C.E. (2004) – IntCal04 Terrestrial Radiocarbon Age Calibrations, 0-26 cal kyr BP. Radiocarbon 46, 1029-1058.

Rendell H.M., Claridge A.J., Clarke, M.L. (2007) – Late Holocene Mediterranean coastal change along the Tiber delta and Roman occupation of the Laurentine shore, central Italy. Quaternary Geochronology 2, 83-88.

Rivière A. (1977) – Méthodes granulométriques, techniques et interprétations. Masson, Paris, 170 p.

Roman R. (1997) – Etude architecturale comparative de sept navires de commerce gréco-romains et byzantins. PhD thesis, University of Provence (Aix-Marseille 1), 259 p.

Sadori L., Giardini M., Giraudi C., Mazzini I. (2010) – The plant landscape of the imperial harbour of Rome. Journal of Archaeological Science 37, 3294-3305.

Salomon F., Goiran J.-P., Bravard J.-P., Millett M., Strutt K., Kay S., Earl G., Paroli L., Keay S. (2010) – Delta du Tibre - Campagne de carottage 2009 - Géoarchéologie des canaux de Portus : l’exemple du Canale Romano. Mélanges de l’Ecole Française de Rome, 122, 263-267.

Segre A.G. (1986) - Considerazioni sul Tevere e sull’Aniene nel Quaternario. In Il Tevere e le altre vie d’acqua del Lazio antico. Archeologia Laziale 7, 9-17.

Testaguzza O. (1970) – Portus: illustrazione dei Porti di Claudio e Traiano. Julia Editrice, Rome, 249 p.

Verduchi P. (2005) – Some thoughts on the infrastructure of the port of Rome. In Keay S., Millet M., Paroli L., Strutt K. (Eds.) Portus. An Archaeological Survey of th Port of Imperial Rome. The British School at Rome, Archaeological Monographs of the British School at Rome 15, London, 248-257.

Haut de page

Annexe

Version française abrégée

Le cœur de cette étude est un transect de carottages réalisés entre le Canale di Comunicazione Traverso et le chenal d’accès au bassin de Trajan dans le Portus de Rome (CT-1 et TR-XIV ; fig. 1 et fig. 2). Diverses analyses ont été effectuées sur les carottes (fig. 5 et fig. 6) : granulométrie, taux de matière organique, susceptibilité magnétique (fig. 4), densité apparente, interprétation du diagramme CM de Passega (1957 ; fig. 7). Ce travail concerne en particulier les données granulométriques et les questions de dynamique sédimentaire en milieu fluvio-marin portuaire. Dans ce secteur de Portus, trois publications récentes (Goiran et al., 2010 ; Sadori et al., 2010 ; Mazzini et al., 2011) développent l’aspect plus proprement paléoenvironemental avec des analyses macro- et microfaunistiques et la détermination des pollens présents. Une synthèse de ces résultats, plus spécifiquement orientée sur l’ostracofaune (fig. 3), confirme une très nette différence entre le bassin du port de Claude, où dominent les eaux salées, et le chenal d’accès au bassin de Trajan marqué par la rencontre des eaux salées et douces. Mais avant tout, pourquoi tant d’études récentes ont-elles concerné ce petit canal ? Le Canale Traverso est le seul lien entre le complexe portuaire de Portus et le système fluvial du Tibre. L’enjeu pour les ingénieurs romains était donc de 1) réaliser une voie navigable continue entre les bassins portuaires et le Tibre tout en 2) évitant l’envasement accéléré de ces bassins par les apports fluviaux (Reddé, 1986). Le Canale Traverso aurait été pensé pour satisfaire ces deux objectifs (Le Gall, 1953). Le présent travail se situe ainsi à l’interface des études de géoarchéologie fluviale et portuaire.

Le Canale Traverso a probablement été creusé dès le Ier s. apr. J.-C. (Keay et al., 2005). Le plan dressé par les archéologues (Lugli et Filibeck, 1935 ; Keay et al., 2005) révèle son implantation assez particulière (fig. 2) puisqu’il est en connexion secondaire avec le Tibre : le Canale Traverso (sud-nord) s’embranche en effet sur le canal de Fiumicino (est-ouest) à 1500 m du Tibre actuel. Ce tronçon amont du Fiumicino correspond à la probable Fossa Traiana antique qui se jetait dans la mer au sud de Portus (fig. 1). Cette position particulière s’accompagne aussi d’une réduction graduelle de la largeur des canaux successifs (Fossa Traiana, Canale Traverso). Le Tibre, aujourd’hui large de 100-120 m dans le delta était d’une largeur comparable dans l’Antiquité (Le Gall, 1953 ; Bertacchi, 1960). Grâce à la découverte des berges antiques de la Fossa Traiana sur plusieurs tronçons mis au jour au XXe s., la largeur de ce canal antique est estimée à environ 50 m (Testaguzza, 1970). Enfin, cette largeur se réduit encore pour atteindre 25 m dans le Canale Traverso. Chaque embranchement de canal s’accompagne ainsi d’une diminution de moitié de la largeur, réduisant ainsi l’écoulement des eaux fluviales vers les bassins portuaires. Ces données géométriques sont très intéressantes puisqu’elles témoignent sans équivoque de la conscience du problème de l’articulation entre le système portuaire et le système fluvial par les architectes et ingénieurs romains responsables de la construction de Portus. Mais cette configuration fut-elle efficace ?

La carotte CT-1 (fig. 5 et fig. 7), réalisée dans le Canale Traverso, indique une profondeur de 5,5 m sous le niveau marin antique (calage en référence à J.-P. Goiran et al., 2009). La lame d’eau antique du chenal d’accès au bassin de Trajan se fait plus profonde en direction du bassin de Claude (8 m dans la carotte TR-XI ; Goiran et al., 2010). Dès l’origine, le Canale Traverso semble ainsi ne pas avoir été conçu pour recevoir les plus gros navires antiques (Pomey et Tchernia, 1978 ; Boetto, 2010 ; fig. 8). Dans un premier temps, le canal se remblaie très rapidement (2,6 cm/a ; fig. 8). L’enregistrement sédimentaire entre -5,5 et -2,3 m sous le niveau marin antique est composé principalement de vases déposées par suspension uniforme mélangée à des sables déposés par suspension graduée (unités B et C, fig. 5 et fig. 7). Le Canale Traverso en fonctionnement offre ainsi un environnement de dépôt calme soumis aux influences marines et fluviales, proche des conditions existant à l’entrée du bassin de Trajan (fig. 6 et fig. 7). L’étude comparée des diagrammes CM des carottes CT-1 et TR-XIV (chenal d’accès au bassin de Trajan ; fig. 7) montre une taille moyenne du D99 des processus du type 4 (fig. 7), plus élevée dans le Canale Traverso. Ce gradient granulométrique atteste ainsi l’apport de sédiments roulés (les plus grossiers) dans le Canale Traverso en provenance non pas du port de Claude, mais du fleuve. La texture restant plutôt fine, ces apports grossiers sont très probablement le résultat d’événements hydrologiques ponctuels comme les crues. Dans le dépôt supérieur du Canale Traverso (de -2,3 m au 0 du niveau marin antique ; CT-1 ; fig. 5 et fig. 8), la présence plus nette de sédiments fluviaux est marquée par une fraction sableuse plus importante. Les conditions de dépôt se produisent plutôt par un mélange de roulement et de suspension graduée et uniforme. Bien que plus grossier, ce remblaiement est en revanche beaucoup plus lent que dans l’unité sous-jacente : 0,45 cm/a (fig. 8). La datation radiocarbone réalisée à environ 1 m sous le niveau marin antique donne la date la plus récente obtenue dans ce secteur de Portus et correspond à la fin d’utilisation du Canale Traverso : 1415±15 BP, soit 600-660 apr. J.-C. Ce taux de remblaiement très lent pourrait être expliqué par un (ou plusieurs ?) curage (s) réalisé (s) vers -2,3 m sous le niveau marin antique. Avec une telle lame d’eau, les petits navires de 70 t du type de ceux retrouvés au nord de Portus (Boetto, 2006, 2010) pouvaient emprunter le Canale di Comunicazione Traverso au plus tard jusqu’au VIè-VIIè s. apr. J.-C.

Haut de page

Table des illustrations

Titre Fig. 1 – Location map of Portus, the ancient sea harbour of Rome on the margins of the Tiber. Fig. 1 – Localisation générale de Portus, l’ancien port maritime de Rome en marge du Tibre.
URL http://geomorphologie.revues.org/docannexe/image/9754/img-1.png
Fichier image/png, 608k
Titre Fig. 2 – Geomagnetic survey results (Keay et al., 2005) and archaeological data between the Claudius and the Trajan harbour: synthesis of successive cores location near the Canale Traverso. Fig. 2 – Résultats des prospections géomagnétiques (Keay et al., 2005) et données archéologiques situées entre les bassins portuaires de Claude et de Trajan : synthèse des carottages réalisés à proximité du Canale Traverso.
Légende 1: studied cores in this paper; 2: cores in J.-P. Goiran et al., 2010; 3: cores in C. Giraudi et al., 2009, L. Sadori et al., 2010, I. Mazzini et al., 2011; 4: bridges? (Lanciani, 1866; Keay et al., 2005); 5: streets ? (Keay et al., 2005).1 : carottages étudiés dans cet article ; 2 : carottages présentés dans J.-P. Goiran et al., 2010 ; 3 : carottages présentés dans C. Giraudi et al., 2009, L. Sadori et al., 2010, I. Mazzini et al., 2011 ; 4 : ponts ? (Lanciani, 1866 ; Keay et al., 2005) ; 5 : routes ? (Keay et al., 2005).
URL http://geomorphologie.revues.org/docannexe/image/9754/img-2.png
Fichier image/png, 733k
Titre Fig. 3 – Marine to freshwater palaeoenvironments in the Portus basin obtained by ostracods determination – A synthesis (Goiran et al., 2010; Mazzini et al., 2011). Fig. 3 – Synthèse des données paléoenvironnementales obtenues pour les bassins de Portus par détermination de l’ostracofaune (Goiran et al., 2010 ; Mazzini et al., 2011).
Légende 1: cores locations with portuary units (Goiran et al., 2009; Mazzini et al., 2011; Goiran et al., in press); all cores are represented according to the 3rd-5th c. sea level (Goiran et al., 2009). Only the portuary units data are shown; 2: harbour bottom; 3: ostracod groups with 3a: freshwater species, 3b: brackish species, and 3c: marine species.1 : localisation des carottages avec leurs unités portuaires (Goiran et al., 2009; Mazzini et al., 2011; Goiran et al., sous presse) ; tous les carottages sont représentés en référence au niveau marin des IIIè-Vè apr. J.-C. (Goiran et al., 2009) ; 2 : fond des bassins portuaires ; 3 : ostracodes ; 3a : individus d’eau douce ; 3b : individus d’eau saumâtre ; 3c : individus caractéristiques du milieu marin.
URL http://geomorphologie.revues.org/docannexe/image/9754/img-3.png
Fichier image/png, 725k
Titre Fig. 4 – Magnetic susceptibility results for core CT-1 - Relevance of a high-resolution dataset (each centimetre) to determine sedimentary layers. Fig. 4 – Résultats de susceptibilité magnétique pour le carottage CT-1.
Légende Bars are grouped by sedimentary units and correspond to the occurrence of MS data (y-axis) included in a range of 5 CGS (x-axis).Ce graphique valide la pertinence des données de haute résolution (chaque centimètre) pour définir des unités sédimentaires. Les barres sont identifiées selon les unités sédimentaires et correspondent à l’occurrence des valeurs de SM (en ordonnée) incluses dans un intervalle de 5 CGS (en abscisse).
URL http://geomorphologie.revues.org/docannexe/image/9754/img-4.png
Fichier image/png, 37k
Titre Fig. 5 – Canale Traverso, analysis of the core CT-1. Fig. 5 – Le Canale Traverso, analyse de la carotte CT-1.
Légende 1a: Posidonia; 1b: shells; 1c: pot shards; 2a: coarse sediments; 2b: sand; 3c: silt and clay; C/M diagram interpretation. 3: pure processes: 3a: rolling; 3b: rolling and graded suspension; 3c: graded suspension; 3d: uniform suspension; 3e: decantation; 4: mixed processes: 4a: mixed rolling/graded suspension + uniform suspension; 4b: mixed graded + uniform suspension. 1a : Posidonie ; 1b : coquille ; 1c : céramique ; 2a : fraction grossière ; 2b : sables ; 3c : limons et argiles ; Interprétation du diagramme C/M. 3 : processus purs : 3a: roulement ; 3b : roulement et suspension graduée ; 3c : suspension graduée ; 3d : suspension uniforme ; 3e : décantation ; 4 : processus mixtes : 4a : roulement/suspension graduée + suspension uniforme mixés ; 4b : suspension graduée et uniforme mixées.
URL http://geomorphologie.revues.org/docannexe/image/9754/img-5.png
Fichier image/png, 215k
Titre Fig. 6 – The access channel to the Trajanic harbour, analysis of core TR-XIV. Fig. 6 – Le chenal d’accès au port de Trajan, analyse de la carotte TR-XIV.
Légende 1a: Posidonia; 1b: shells; 1c: pot shards; 2a: coarse sediments; 2b: sand; 3c: silt and clay; C/M diagram interpretation. 3: pure processes: 3a: rolling; 3b: rolling and graded suspension; 3c: graded suspension; 3d: uniform suspension; 3e: decantation; 4: mixed processes: 4a: mixed rolling/graded suspension + uniform suspension; 4b: mixed graded + uniform suspension.1a : Posidonie ; 1b : coquille ; 1c : céramique ; 2a : fraction grossière ; 2b : sables ; 3c : limons et argiles ; Interprétation du diagramme C/M. 3 : processus purs : 3a: roulement ; 3b : roulement et suspension graduée ; 3c : suspension graduée ; 3d : suspension uniforme ; 3e : décantation ; 4 : processus mixtes : 4a : roulement/suspension graduée + suspension uniforme mixés ; 4b : suspension graduée et uniforme mixées.
URL http://geomorphologie.revues.org/docannexe/image/9754/img-6.png
Fichier image/png, 179k
Titre Fig. 7 – Cross-sections from the Canale Traverso to the access channel to Trajanic harbour, and CM diagram of the CT-1 and TR-XIV cores. Fig. 7 – Transect longitudinal du Canale Traverso au chenal d’accès au basin de Trajan. Présentation des diagrammes CM pour les carottages CT-1 et TR-XIV.
Légende 1a: Posidonia; 1b: shells; 1c: pot shards; 1d: canal or harbour basin bottom; 2a: coarse sediments; 2b: sand; 3c: silt and clay; C/M diagram interpretation. 3: pure processes: 3a: rolling; 3b: rolling and graded suspension; 3c: graded suspension; 3d: uniform suspension; 3e: decantation; 4: mixed processes: 4a: mixed rolling/graded suspension + uniform suspension; 4b: mixed graded + uniform suspension.1a : Posidonie ; 1b : coquille ; 1c : céramique ; 1d : fond du canal ou du port ; 2a : fraction grossière ; 2b : sables ; 3c : limons et argiles ; Interprétation du diagramme C/M. 3 : processus purs : 3a: roulement ; 3b : roulement et suspension graduée ; 3c : suspension graduée ; 3d : suspension uniforme ; 3e : décantation ; 4 : processus mixtes : 4a : roulement/suspension graduée + suspension uniforme mixés ; 4b : suspension graduée et uniforme mixées.
URL http://geomorphologie.revues.org/docannexe/image/9754/img-7.png
Fichier image/png, 484k
Titre Fig. 8 – Harbour infill, water depth and ships draught data. Comparisons of these parametres by chronology. Fig. 8 – Compilations des données disponibles à Portus pour comprendre chronologiquement la relation entre le remblaiement du port, la lame d’eau disponible et le tirant d’eau des bateaux.
URL http://geomorphologie.revues.org/docannexe/image/9754/img-8.png
Fichier image/png, 235k
Haut de page

Pour citer cet article

Référence papier

Ferréol Salomon, Hugo Delile, Jean-Philippe Goiran, Jean-Paul Bravard et Simon Keay, « The Canale di Comunicazione Traverso in Portus: the Roman sea harbour under river influence (Tiber delta, Italy) », Géomorphologie : relief, processus, environnement, vol. 18 - n° 1 | 2012, 75-90.

Référence électronique

Ferréol Salomon, Hugo Delile, Jean-Philippe Goiran, Jean-Paul Bravard et Simon Keay, « The Canale di Comunicazione Traverso in Portus: the Roman sea harbour under river influence (Tiber delta, Italy) », Géomorphologie : relief, processus, environnement [En ligne], vol. 18 - n° 1 | 2012, mis en ligne le 01 avril 2014, consulté le 28 mars 2017. URL : http://geomorphologie.revues.org/9754 ; DOI : 10.4000/geomorphologie.9754

Haut de page

Auteurs

Ferréol Salomon

Université Lumière (Lyon 2) - CNRS-UMR 5600 - 5, avenue Pierre Mendès-France - 69676 Bron Cedex - France (ferreol.salomon@univ-lyon2.fr).

Hugo Delile

Université Lumière (Lyon 2) - CNRS-UMR 5600 - 5, avenue Pierre Mendès-France - 69676 Bron Cedex - France (hdelile@gmail.com).

Jean-Philippe Goiran

CNRS-UMR 5133 - Maison de l’Orient et de la Méditerranée - 7, rue Raulin - 69007 Lyon - France (jean-philippe.goiran@mom.fr).

Articles du même auteur

Jean-Paul Bravard

Université Lumière (Lyon 2) - CNRS-UMR 5600 - 5, avenue Pierre Mendès-France - 69676 Bron Cedex - France (jean-paul.bravard@univ-lyon2.fr).

Articles du même auteur

Simon Keay

University of Southampton - School of Humanities, Archaeology - Avenue Campus - Southampton, SO 17 1 BF - Great Britain (sjkl@soton.ac.uk).

Haut de page

Droits d’auteur

© Groupe français de géomorphologie

Haut de page
  • Logo CNRS - Institut des sciences humaines et sociales
  • Logo Groupe français de géomorphologie
  • Revues.org