Ostia & Portus



  • The coastal belt of Roma

Four recent articles form a breakthrough in the knowledge of the Holocene history of the coastal area of Rome (Di Rita et al. 2010, Bellotti et al. 2011, Di Rita et al. 2011, Milli et al. 2013 1*). Data were mainly collected from drillings. The viewpoints are sedimentological and paleobotanical.

Bellotti et al. 2011 provided a challenging and unitary model on the development of the strand plain of the coastal area of Rome over the last millennia. The several phases of migration of the Tiber are recorded by the different positions of the delta cusp. In the first phase, from ca. 3000 B.C. until the 8th – 7th century B.C., a cusp spread out over the entire area from Capo Due Rami to the place of the later imperial harbours, and unto the outer margin of the Stagno di Ostia. The transition to the second phase coincided with the sudden migration of the Tiber to the south, at first flowing into the Stagno Ostiensis and later breaking through the dune belt near future Ostia. With the opening of the channel of Trajan (early 2nd century A.D.) a new complex system of delta progradation developed, with two river branches active almost simultaneously (the third phase).


Main changes of the Tiber river mouth location during the strand-plain evolution (modified from Bellotti 2011; the transition between the phases 1 and 3 north of Trajan’s channel has been shifted slightly westward, following the data in Arnoldus-Huyzendveld 2005 *2 ).

Here follows an abstract of the rest of of the mentioned articles relevant to Ostia Antica and to salt extraction.

- In the pollen diagram of the Lagoon of Ostia, an environment with marshy reeds and sedges with stagnant pools is recorded from about 3900 to 2600 years ago, a period characterized by geological and ecological instability associated with the migration of the cusp of the Tiber delta. Ceramic materials found in the area of Ostia Antica witness human presence in the delta area during the XIII-X centuries b.C.

- Around 3000 years ago, the former lake of the Stagno Maccarese is transformed rapidly in a marsh. There are no traces of human occupation in this period. But we know that salt extraction was already carried out by the Etruscans from 2600 years ago on. The presence of salt works is reflected by the presence of halophitic vegetation remains in the cores.

- In the Republican era, the Maccarese area became an integral part of the Campus Salinarum Romanarum, the system of salt winning localized in the area of the Tiber delta, active until the fifteenth century. A.D.

- Around 2600 years ago the Stagno Ostiense is affected by a sudden environmental change through the input of sea water, which transforms the basin from lacustrine to brackish. This transformation coincides chronologically with the foundation of Ostia. We don’t know if the sudden change was natural or due to human influence. Anyway, the width of the dune belt was too small too accommodate a real city, as Ostia was later to become. The hypothesis is that, at the time, it was only an outpost.

- The cuspate delta advanced seaward very quickly, at an estimated progradation rate of 5 – 6 m/year, so that by the V-IV century BC it was almost fully developed. By then, the dune belt was wide enough to allow the expansion of the city of Ostia.

- In the III century AD, the growth of the delta was interrupted by an erosive process, which occurred during a warm period, characterized by a decrease of the Tiber floods.

The possibility of a temporary presence of the Tiber outlet in the area of the future imperial harbours, before its migration to the present course alongside Ostia, is now generally accepted, and has found confirmation in several stratigraphic drilling data. The phenomenon results also clearly from the study of the directions of the coastal barriers of Bellotti et al. 2011, fig. 3 (1*). In the cores of the ports area, the abandonment is supposed to correspond to the transition from river sediments to those distinctly marine or lagoonal, whereas in the Stagno Ostiense the end of the peat formation is considered an indicator of the Tiber arrival (Bellotti et al. 2011). Milli et al. 2013 consider the river shift as an instantaneous event during a major flood.

The time/period of the shift is only approximately known: based upon 14C datings of drilling cores in the basin of Claudius, Giraudi et al. 2007-9 (3*) propose its occurrence between the eighth and fifth century BC (780 – 400 cal. BC), Goiran et al. 2010 (3*) between 635 and 360 cal. BC, Bellotti et al. 2007 (3*) between 515 and 395 cal. BC. Based upon a drilling near Ostia, Bellotti et al. 2011 (1*) propose a date range after 810 – 540 cal. BC., and Arnoldus-Huyzendveld in preparation (*4), based upon three samples of the uppermost Tiber sediments from drillings in the Claudius basin near Monte Giulio: 765 – 525 BC, 335 – 210 BC and 820 – 590 cal. BC (*5). To these we have now added the data from drilling PO2 in the fluvial harbour of Ostia mentioned in Goiran et al. 2013 (10*): 837 – 734 cal. BC for the substrate layers under an evidently major influence of the river with respect to the sea.
If we accept an intrusion error of the second sample of the Monte Giulio group, this complex of data converges to a shifting of the Tiber outlet between the VIII and VI century BC, with average extreme values between ca. 750 and 500 years BC and the central value in the second half of the VII century.

14C datings relative to the shift of the Tiber outlet from the ports area to the South

But this kind of statistics may be too simple. So next I used a more reliable method for comparison of the data, by recalibrating the original 14C measures (11*) in the program Oxcal 4.2, with the use of the latest continental and marine curves (2013), and treating the group as an “delimited unordered phase”. This presumes that the datings belong to contemporaneous sediments, which is not necessarily completely true.
The probability peaks of the modelled curves result to be centered in the VIII century BC., with a “tail” in the VII century (see full colours in the graph).

On the base of the graphs reproduced here below (the best approximation for the moment) it seems resonable to conclude that the Tiber shift took place in the VIII century BC or slightly later, so at least a century before the founding of the first settlement of Ostia which, according to  tradition, occurred in the last quarter of the VII century BC (cfr. ostia intro).


Download the PDF file .

Results of the re-calibration of the various data with Oxcal 4.2 according to the latest continental and marine curves (2013), see legend. Full colours, modeled 2σ (=95.4%) range; lighter colours, original range. "A" is an index for the agreement between the posterior and prior distributions. The samples were considered to belong to a “delimited unordered phase.

For info on Ostia Antica, see the site www.ostia-antica.org and the Facebook page Ostia L.

  • The harbour basin of Claudius

The extension and orientation of the Claudian harbour basin has long been subject to debate (see Morelli 2005).

A series of 140 drillings executed between 2004 and 2007 has resolved this problem without any doubt : the basin is east-west oriented and sets out farther into the sea than ever suspected: the distance between Monte Giulio and the lighthouse island is more than 2 km. The various distances indicated by Antonio Labacco on a reconstructive map of the XVIth century turned out to be approximately correct (Morelli al. 2011; 6*).

Remains of structures were encountered in the drillings only from a depth of several meters on, being covered by dune sand.


Position of the drillings, in red squares with remains of structures.

The reconstruction shows two protruding moles and a lighthouse island, separated by two evident entrances. A third narrower entrance (probably only a channel) was demonstrated to exist between the northern pier and Monte Giulio (7*).


Reconstruction of the outline of the piers and lighthouse island and the entrances of the Claudian harbour, based upon the recent drilling data.

The same data have been overlain, as well as possible, on a photo mosaic of 1911, and on a stretched and overturned image of the fresco of A. Danti of 1582, which demonstrates first the reliability of this fresco and second the visibility, at the time, of the lighthouse relicts still in the sea, before their burial by the dune sands of the advancing coast.


The drilling data and interpreation overlain on a fotomosaic of 1911; red squares indicate structures encountered in the drillings.


In red lines the of the Claudian harbour overlain on a stretched and overturned image of the fresco of A. Danti of 1582.

According to contemporary writers, in the same period the remains of the lighthouse were still visible in the sea (Giuliani 1996 p. 41, 8*):

Biondo Flavio a questo proposito scrive (nel 1558): “di questa torre ne veggiamo insino ad hoggi una buona parte in pie, se non che ne sono stati tolti i marmi, dei quali ella era incrustata”. Ma è Pio II nei suoi Commentarii (1614; 9*) a darci la informazione più utile: “ancora rimangono vestigi della torre le quali si vedono là nel mare; tutti gli altri monumenti sono periti interamente” (There are still traces of this tower which can be seen from far out at sea. Everything else has perished utterly).


One of the 24 images of the lighthouse known from Ostia and Portus (mosaic in statio 46 on the Square of the Corporations, Ostia); from ostia-antica.org.

  • Monte Giulio

Arnoldus-Huyendveld A., Turi P, Morelli C., 2015: “Il paleoambiente di Monte Giulio e della parte nord-orientale del bacino portuale di Claudio.”, con Patrizia Turi e Cinzia Morelli;  http://www.fastionline.org/docs/FOLDER-it-2015-324.pdf.

This publication focuses on the site of Monte Giulio (Fiumicino, RM, Italy), the complex that makes up the internal (eastern) delimitation of the harbour basin of Claudius, investigated through a series of cores, trenches and excavations between 2007 and 2009. We recognized various natural and anthropical phases, from the construction of the structures on Monte Giulio in the Trajanic period until the end of the occupation of the area, dated not beyond the beginning of the fifth century. Hypotheses are put forward as to the extension and the limited depth of the interdunal lagoon located in pre-imperial times behind the coastline north of the future harbour of Trajan, and on Monte Giulio as a complex functional for small-sized boats, constructed on a natural dune ridge in this lagoon. Moreover, hypotheses are put forward on the general environmental conditions of the area in pre-Roman and Roman times, and on the probability that Claudius did not dredge the entire lagoon, but may have done so only locally through the excavation of channels in the bottom; he then cut through the coastal barrier to let the sea enter, and concentrated the maritime traffic on Portus through the creation of harbour structures along the original coast line. On the basis of the excavation data, the sea level in the second century AD results to have been more than one meter below the present  level, and rising continuously in the centuries thereafter.

Harbour of Claudius and the area of Monte Giulio on a Google Earth image: 1, Capitaneria; 2, thermal building; 3, Cistern; 4, ramp; 5, trench 3; 6, trench 1; 7) trench 2.

Interpretation of the actions of Claudius and the distribution of environments after the harbour construction and in first centuries A.D. A dotted area marks the zone of initial silting up; a brown line indicates the coast line (certain and alleged) directly prior to the imperial works; the text “conchiglie” refers to the barnacles attached to the pier dated by Goiran et al. 2009.


For info on the Claudian lighthouse, see ostia-antica.org/portus.

Article: Andrea Locatelli’s (1695 – 1741) painting of the salt pans of Ostia.

- Federico Di Rita, Alessandra Celant and Donatella Magri, 2010, Holocene environmental instability in the wetland north of the Tiber delta (Rome, Italy): sea-lake-man interactions, Journal of Paleolimnology, Volume 44, Number 1 (2010), 51-67, DOI: 10.1007/s10933-009-9385-9.
- P. Bellotti, G. Calderoni, F. Di Rita, M. D’Orefice, C. D’Amico, D. Esu, D. Magri, M. Preite Martinez, P. Tortora and P. Valeri . 2011, The Tiber river delta plain (central Italy): Coastal evolution and implications for the ancient Ostia Roman settlement , The Holocene 2011 21: 1105, originally published online 26 May 2011 DOI: 10.1177/0959683611400464.
- Federico Di Rita, Alessandra Celant, Cecilia Conati Barbaro, 2011, Interazioni tra clima,, ambiente e uomo nell’evoluzione olocenica del delta del Tevere: dati e paleobotanici e ritrovamenti archeologici; Soc. Geol. It., Vol. 18 (2011) – (DOI 10.3301/ROL.2011.60).
- Milli S., D’Ambrogi C., Bellotti P., Calderoni G, Carboni M.G., Celant A., Di Bella L., Di Rita F., Frezza V., Magri D., Pichezzi R.M., Ricci V., 2013, The transition from wave-dominated estuary to wave-dominated delta: The Late Quaternary stratigraphic architecture of Tiber River deltaic succession (Italy); Sedimentary Geology, Volumes 284 – 285, 159-180.
- Arnoldus-Huyzendveld A., 2005: The natural environment of the Agro Portuense. In: Portus, an archaeological survey of the port of imperial Rome, S. Keay, M. Millett, L. Paroli and K. Strutt (eds), the British School at Rome, pp. 14-30.
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, folder 80, www. fastionline.org.
Idem: Giraudi C., Tata C., Paroli L., 2009 – Late Holocene evolution of Tiber river delta and geoarchaeology of Claudius and Trajan Harbor, Rome. Geoarchaeology, Volume 24, Issue 3: pp. 371 – 382, Wiley Periodicals, Inc., DOI: 10.1002/gea.20270.

– Goiran Jean-Philippe, Hervé Tronchère, Ferréol Salomon, Pierre Carbonel, Hatem Djerbi, Carole Ognard, 2010 – Palaeoenvironmental reconstruction of the ancient harbors of Rome: Claudius and Trajan’s marine harbors on the Tiber delta, Quaternary International 216 (2010) pp. 3-13.
- Bellotti P., Calderoni G., Carboni M.G., Di Bella L., Tortora P., Valeri 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 fuer Geomorphologie 51,4, 505-534 (well 4).
Arnoldus-Huyzendveld A., 2014  – “Le dinamiche evolutive dell’ambiente costiero e del Tevere”; in Caravaggi L., Morelli C. “Paesaggi dell’aercheologia invisibile. Il caso del distretto Portuense”, Quodlibet, Macerata, pp. 72-89.
Datings by the Centre de Datation par le Radiocarbon, courtesy of the CNRS – UMR 5133 – Maision de l’Orient et de la Méditerranée, Lyon, France.
- Morelli, C., 2005 – The Claudian Harbour in Light of New Investigations. In S. Keay, M. Millett, L. Paroli and K. Strutt (eds.), Portus. An Archaeological Survey of the Port of Imperial Rome. Archaeological Monographs of the British School at Rome 15: 241-68. London, BSR.
- Morelli C., Marinucci A, Arnoldus-Huyzendveld A., 2011  – Il Porto di Claudio: nuove scoperte, in Portus and its Hinterland, recent archaeological research, Simon Keay & Lidia Paroli (eds), Archaeological Monographs of the British School at Rome, pp. 47-65.
- Goiran J.-Ph., Salomon F., Tronchere H., Carbonel P., Djerb H., Ognard C., 2011 – Caractéristiques sédimentaires du bassin portuaire de Claude: nouvelles données pour la localisation des ouvertures, In: Portus and its Hinterland, Archaeological Monographs of the British School at Rome, a cura di Simon Keay e Lidia Paroli, p. 31-45.
- Giuliani C.F. (1996) – Note sulla topografia di Portus; in: Manucci V. (a cura di), 1996, Il Parco Archeologico Naturalistico del Porto di Traiano; Ministero per i Beni Culturali Ambientali, Soprintendenza Archeologica di Ostia, pp. 29-44.
Testo originale in Inglese: http://www.ostia-antica.org/anctexts.htm, Later Texts, Pius II.
Goiran J-P, Ferréol S., Pleuger E., Vittori C., Mazzini I., Boetto G., Arnaud P., Pellegrino A., 2013, “Port antique d’Ostie”, Chronique des activités archéologiques de l’École française de Rome [En ligne], Italie centrale, mis en ligne le 19 décembre 2012, consulté le 13 mars 2013. URL: http://cefr.revues.org/299.
Direct confrontation of the raw data with the program  Oxcal is only possible by separating the samples according to the calibration curve applied (marine or continental).  I could infer the  curve applied from the calibrated data given in the publications.
Original 14C data: Giraudi et al. 2007-9 (charcoal; cal. cont.?), 2450 ± 40; Bellotti et al.2007 (well 4; shell; cal.cont.?), 2385 ± 35; Monte Giulio 2009-1, 2845 ± 35 (Pos.; cal. mar.); Monte Giulio 2009-2, 2265 ± 30 (wood; cal. cont.); Monte Giulio 2009-3, 2910 ± 40 (Pos.; cal. mar.); Goiran et al. 2010, 2785 ± 30 (cal. mar.); Bellotti et al. 2011, 2555 ± 50 (cal.cont.?); Goiran et al. 2013 (cal. mar.?), 2955 ± 25. Oxcal analysis confirmed MG2 as an “outlier” of the group.