Measuring the past: A look into the history of relative sea-level variations on the coast of Alexandria

The ancient city of Alexandria was founded in 331 BC by Alexander the Great on the western edge of the Nile delta. This Egyptian-Hellenistic city was regarded as one of the main centres for Mediterranean trade and remains an important port city today. Archaeologists have been surveying and excavating the harbour for years in order to gain further insight on Alexandria’s extensive maritime past. Mahmoud-Bey (1872) and Goddio et al., (1998) discovered and mapped out Late Roman port structures found between 5-6 m below present sea-level (bpsl) in the eastern bay of Alexandria (known as the Magnus Portus, see Fig. 1 & 2). Similarly, thousands of stone blocks were found at ~6-8 m bpsl situated east of the island of Pharos near the fort of Qaît Bey (Frost, 1975; Empereur, 2000) (Fig. 3). These findings raised the debate on whether the submergence of these structures was a result of rising sea-level (SL) and/or caused by a combination of both Holocene SL rise and tectonic variation (e.g. seismicity and subsidence). A paper published in Egypt & the Levant by Goiran et al. 2019, explores this archaeological conundrum further by using interdisciplinary research methods to understand the implications caused by regional relative SL rise. The objective of this study was to determine when the city’s port structures submerged and whether the changes in SL were continuous or sequential. The findings of Goiran et al. 2019, will be summarized in this blog post.

Eustatic changes in SL are largely influenced by tectonic and glacial factors as well as sedimentation rates. It is difficult to precisely measure past SL; however, relative measurements of SL can be obtained using a combination of geomorphological, biological, and archaeological proxies. Through the correlation of these methods, a SL curve can be established to understand the evolution of Alexandria’s coast. Geoarchaeological underwater surveys were conducted via SCUBA techniques to find geomorphological, biological, and archaeological SL proxies to understand relative SL variations during the last millennia. Several onshore sediment cores were also used for the purpose of radiocarbon dating (¹⁴C) the proxies to determine the chrono-stratigraphy of this maritime façade.

Notches, an erosional geomorphological indicator, were used as a reference for the ancient SL in this study. These notches were first discovered in the Pleistocene sandstone substratum at a depth of ~6.5 m bpsl by Goddio et al., 1998. Other proxies included radiocarbon dating colonies of Cladocora caespitose (native Mediterranean coral colonies), macrofaunal indicators such as, marine bivalves (pelecypods) and biodepositions of Egagropili found in sediment core records. Egagropili are fibrous balls formed from the foliage of Posidonia seagrass and are often used to determine the proximity of the shoreline with an imprecision range related to the highest storm surges measured in Alexandria. Granulometric indicators such as, rounded fragments from pottery and rounded building materials preserved in the sediment, were also used to reveal the position of the ancient pebble beach shoreline. The rounding of the fragments is achieved through high energy transportation which, in the case of the eastern bay of Alexandria, is the surf zone (mediolittoral). Additionally, archaeological structures were also used as reference points to reconstruct the relative changes of SL in Alexandria.

A notable aspect of the discoveries of the Late Roman port structures by Goddio et al., (1998), is the visibility of the upper pavement of the quay which remains well preserved. This suggests that the quays were submerged rather than exposed to the high energy surf zone environment for a long period of time which would have led to severe abrasion of the upper pavement. As such, these harbour structures provide a chronological and bathymetric marker for determining the position of the SL. The ¹⁴C results of the aforementioned geomorphological, biological, and archaeological proxies used to produce an age-depth model (Fig. 4) to illustrate major changes in regional relative SL.

The age-depth model shows a concave curve indicating an abrupt relative rise in SL during the second part of the 1^st millennium AD and reveals three successive ancient SL stages (Fig. 4). The First Period dates to between the 4^th millennium BC and the 5^th c. BC, and corresponds to SL proxies found at a depth greater than 5.4 m. The Second Period begins during the 5^th c. BC continuing until the 8^th c. AD and was determined from SL proxies found at a depth between ~5.60 m and ~4.20 m. The Third Period dates to between the 8^th c. AD and the 13^th c. AD at a depth above 2.70 m.

During the 4^th millennium BC, relative SL was around 7 m bpsl. This was determined through sampling of in-situ Cladocora on underwater reefs found at depths between 7.50 m and 7 m. They correspond to a biological indicator on the sea bed resting on a hard substratum, and indicate that the SL in this period was above 7 m bpsl. In the same depth range, two erosional notches were discovered on the reefs in the eastern harbour by Goddio et al. (2006). Furthermore, during the Late Roman Period (~5^th-8^th c. AD), SL was between 5 m – 5.40 m bpsl. This was determined using marine shells found in cores obtained from the infralittoral zone near the tombolo. This confirms that SL was above the marine shells but does not indicate the precise position of the biological ancient SL.

Based on archaeological structures, the upper pavementof quays dating to the 5^th-6^th c. AD is well preserved and situated at 5 m bpsl. Another proxy is the pebble beach facies found at ~5.40 m bpsl. This facies is dated to the 5^th-6^th c. AD and indicates a SL that agrees with the level determined by the quays. Additionally, the mass of egagropili from the top of the beach at 4.60 m bpsl indicates that SL was situated between 4.60 m and 6.10 m bpsl. Based on these observations, relative SL during the 5th-6th c. AD would have been between 5 m and 5.4 m bpsl.

From the middle of the 8^th c. AD to the end of the 13^th c. AD, SL was around 1.75 m bpsl. The pebble beach suggests an ancient SL at about 2 m bpsl (±1.5 m bpsl) between the end of the 6^th c. AD and the middle of the 9^th c. AD with the highest concentration around the 8^th c. AD. It is important to note that samples corresponding to an infralittoral sea bed are positioned at 2.70 m bpsl which indicates that SL was above 2.70 m bpsl. Subsequently, using a combination of various proxies provides an estimation of a relative SL of about 1.75 m bpsl for the period between the middle of the 8^th c. AD and the end of the 13^th c. at latest. According to these results, the relative SL has risen to ~6.75 m during the past 5-6 millennia. It can be argued that these major changes in metre scale between centuries can be explained by a repeated seismic activity of the delta and a compaction of sediments, particularly clay-rich sediments. As such, it is possible to identify a sudden subsidence around the 2^nd half of the 8^th to 9^th c. AD.

Analyses using various SL proxies are necessary to estimate the changes of SL over long periods of time. What can be gathered through this study by Goiran et al. (2019) is that regional SL had risen in successive stages between centuries. This data has important implications for understanding SL rise from the 4^th millennium BC to the 1^st millennium AD but also for the future management and planning of Alexandria and other coastal cities as SL continues to gradually rise. Problems caused by this transgression include the salinization of coastal lagoons and deltaic arable land. Both urban and agricultural development plans should use the results from such studies that show a rising trend of relative SL over several millennia in order to prepare and anticipate the progression of soil salinization.

References

Empereur, J.Y. 2000. Underwater Archeological Investigations of the Ancient Pharos, in Hassan, M. H., Grimal, N., Nakashima, D., (Eds), Underwater Archaeology and Coastal Management : Focus on Alexandrina, Coastal Managements Sourcebooks 2, 54-59.

Frost H. 1975. The Pharos Site, Alexandria, Egypt. International Journal of Nautical Archaeology 4, 126–130

Goddio, F. Bernand, A. Bernand, E. Darwish, I. Kiss, Z., and Yoyote, J. 1998. Alexandrie, les quartiers royaux submergés, Periplus, Londres.

Goiran, J.-P., Vittori, C., Noirot, B., and Torab, M. 2018. Relative Sea Level Variations at Alexandria (Nile Delta, Egypt) over the Last Millennia: Archaeological Implications for the Ancient Harbour. Ägypten und Levante / Egypt and the Levant, Verlag der Österreichischen Akademie der Wissenschaften 28, 219-234.

Goiran, J.-P. 2001. Recherches géomorphologiques dans la region littorale d’Alexandrie en Egypte, Ph.D. thesis, Université de Provence, Marseille, France.

Mahmoud, Bey. 1872. Mémoires sur l’antique Alexandrie, Copenhague.

Stanley, J.-D. and Clemente, P. L. 2014. Clay Distributions, Grain Sizes, Sediment Thicknesses, and Compaction Rates to Interpret Subsidence in Egypt’s Northern Nile Delta. Journal of Coastal Research 30.1, 88-101.

Les auteurs

Jean-Philippe Goiran est Chargé de Recherche au CNRS. Géo-archéologue, il travaille sur l’histoire des ports antiques et les paléoenvironnements littoraux et deltaïques. UMR 5133 – Archéorient, Maison de l’Orient et de la Méditerranée, Lyon.

Winnie May Chan is a research scientist that specialises in paleoenvironmental reconstruction using sedimentological and micropaleontological (foraminifera and arcellaceans) methods. During her undergraduate and graduate studies (MSc) at McMaster University (Hamilton, Canada), she focused her research on reconstructing the potability of groundwater in the Yucatan Peninsula in response to Holocene climate change. Since the completion of her MSc, she has been assisting Dr. Jean-Philippe Goiran and his team at CNRS on understanding the geoarchaeological implications of ancient harbours in the Mediterranean.

---

---