Conceptual origins and geomorphic evolution of the temple of Amun-Ra at Karnak (Luxor, Egypt)

1 Conceptual origins and geomorphic evolution of the temple of Amun-Ra at Karnak (Luxor, Egypt) Benjamin T. Pennington1, Angus Graham2,*, Aurélia Masson-Berghoff3,4, Marie Millet5, Jan Peeters6, Willem H. J. Toonen7,8, Timotheus G. Winkels9, Luke H. Sollars10, Virginia. L. Emery11, Kristian D. Strutt12, Dominic S. Barker12 1 Geography & Environmental Science, University of Southampton, UK 2 Department of Archaeology, Ancient History and Conservation, Uppsala Universitet, Sweden 3 Department of Greece & Rome, The British Museum, London, UK 4 Department of Egypt & Sudan, The British Museum, London, UK 5 Department of Egyptian Antiquities, Musée du Louvre, Paris, France 6 Jebel Barkal Archaeological Project, Kelsey Museum of Archaeology, University of Michigan, Ann Arbor, USA 7 Faculty of Science, Earth & Climate, Vrije Universiteit Amsterdam, The Netherlands 8 Egyptology Unit, Katholieke Universiteit Leuven, Belgium 9 Water and Environment Group, WSP, Nieuwegein, The Netherlands 10 Independent Researcher, Salisbury, UK 11 National Park Service, Fort Union National Monument, Watrous, USA 12 Department of Archaeology, University of Southampton, UK *Corresponding author (email: angus.gra[email protected]) Abstract Despite almost a century and a half of excavation, the dynamic landscape into which the temple complex of Karnak was embedded is not well understood. Presenting the results of the first comprehensive geoarchaeological survey of the area, the authors show that Karnak was built upon a fluvial terrace segment surrounded by river channels in an island configuration potentially recalling the ‘primeval mound’ of Egyptian creation myths. Permanent occupation of the site became possible after 2520 BC ±420 years, likely 2 during the Old Kingdom. Subsequent landscape changes were dramatic, with the occupants of the island responding both opportunistically and proactively. Introduction Five hundred metres east of the present course of the River Nile near Luxor (Egypt) stands one of the ancient world’s largest temple complexes—Karnak—at the ancient Egyptian religious capital of Thebes (Barguet 1962). Occupied for some three thousand years, religious areas were dedicated to three main deities: Amun-Ra, Montu and Mut (Figure 1), with additional deities also incorporated into the built architecture. Karnak’s principal domain is the approximately 30ha temple precinct of Amun-Ra (Figure 2). Archaeological investigations have been ongoing at the site for approximately 150 years (Legrain 1929; Barguet 1962; Arnaudiès & Laroze 2007; Gabolde 2018), yet the dynamic riverine landscape within which the temple was conceived, built and extended has not been understood in detail and the age of earliest occupation continues to be debated. Most researchers favour a First Intermediate Period (c. 2152–1980 BC; see also online supplementary material (OSM) Table S1) origin for Karnak based on in-situ excavated archaeological remains and a written reference to a temple of a ‘Ra-Amun’ from the reign of Intef I, II or III (traditionally Intef II c. 2066–2017 BC; Postel 2004: 72–73; Hornung et al. 2006: 491; Ullmann 2007: 4; Millet 2008: 308–9; Charloux & Mensan 2011: 231; Larché 2020: 121). These earliest excavated remains are found in the eastern part of the complex and south-east of the Sacred Lake (Figure 2: E4–6) and variously consist of mudbrick walls (including silos at E4) associated with ceramic material from the First Intermediate Period/early Eleventh Dynasty (Charloux et al. 2021: 928, 931; Gabolde 2018: 154–66; Millet 2007, 2008). Further north, near the Ptah Temple (Figure 2), a hearth is dated to the mid-Eleventh Dynasty (Eleventh Dynasty = c. 2080–1940 BC; Hornung et al. 2006: 491) (Charloux et al. 2021: 24, fig. 4). Other authors suggest Predynastic occupation (3900–3100 BC) at Karnak, based on material from old excavations and objects ex situ (Legrain 1906a: 21; Franchet 1917: 87, 99; Gabolde 2018: 135–36, 167). 3 Figure 1. Location of study area: a) within Egypt; b) locally. Archaeological features: 1 Amun-Ra temple complex, Karnak; 2 Montu temple complex, North Karnak; 3 Mut temple complex; 4 Kom elAhmar; 5 Avenue of Sphinxes; 6 Luxor Temple; 7 Numerous temples and necropoleis (not all shown). The core transects outside the Karnak area are published (Toonen et al. 2018, 2019; Peeters et al. 2024). Recent attempts to reconstruct the palaeoenvironmental setting of early Karnak (Gabolde 2018; Charloux et al. 2021) have primarily relied on details obtained from archaeological excavations and limited coring (Millet 2007: pl. 39; Bunbury et al. 2008; Ghilardi & Boraik 2011). These attempts have often tried to locate ancient river channels by identifying palaeosurfaces sloping down to presumed channel depressions, and by crosscomparing pottery levels found in different excavations. They bring together ideas that an unidentified river channel, of unknown date and size, likely existed east of the site (Redford 1988: 37, fig. 8), and that a second channel was probably present during the Middle Kingdom (c. 1980–1760 BC) west of the north-south axis of the Amun-Ra temple—an axis defined by the monumental gateways of the seventh to tenth pylons (Figure 2) (Legrain 1906b: 137–61; Bunbury et al. 2008: 364, 367). This second channel may have subsequently shifted westward (Bunbury et al. 2008: 364, 367; Ghilardi & Boraik 2011; Boraik et al. 2017: 130–35). The site was thus at some point possibly situated on an island (Egli 1959: 40–43; Millet 2008: 324–30; Graham 2010), whose northern limit may have been at North Karnak (Figure 2). A channel could have existed further north of this point, as an early Middle Kingdom surface slopes downwards to the north-west (Jacquet 1983: 95), and ceramics potentially dating from the Middle Kingdom 4 onwards are found at relatively deep levels, suggesting a topographic depression (Bunbury et al. 2008: 365; Graham 2010: 134; Boraik et al. 2017: 123). South of the Amun-Ra temple, Late Period ceramics (664–332 BC) at depths of up to 12m may suggest another channel depression, constraining the southern limit of the settlement (Lauffray 1968: 339). Figure 2. Location of coring sites and transects. a) Temple of Amun-Ra, Karnak. Pylons (monumental gateways) are indicated with Roman numerals; archaeological excavations as follows: E1 Karnak North (Jacquet 1983: 80, 95–96, 2001: 13–14); E2 Ptah Temple (Charloux et al. 2018, 2021: 924–26); E3 MK Court (Carlotti et al. 2010; Charloux & Mensan 2011; Larché 2020); E4 Osirian Catacombs (Charloux et al. 2021: 928); E5 East Karnak (Redford et al. 1991); E6 SE Sacred Lake (Millet 2007, 2008; Masson-Berghoff 2021); E7 Opet Temple (Charloux et al. 2012: 255); E8 Xth pylon court (Azim 1980). The Chevrier Drain is modern. b) hand auger in use at AS040; c) percussion corer in use at PC026; d) extracting drilled sediments at PC027. 5 Although informative, these recent reconstructions remain incomplete, drawn as they are from data that are fragmentary, spatially restricted to excavation localities—limiting the scope for palaeoenvironmental conclusions—and often collected as a byproduct of other research objectives. Few dedicated palaeogeographic surveys of the area have been conducted, and fewer than 10 deep (>5m) sediment cores have been published in any detail (Lauffray 1968, 1969; Bunbury et al. 2008; Ghilardi & Boraik 2011; Charloux et al. 2021). The geoarchaeological survey conducted by Bunbury and colleagues (2008) provides a notable exception, as part of a research programme that preceded the present study. The preliminary reconstruction presented therein was based primarily on 15 sediment cores with an average depth of 4.9m set in locations limited by the handaugering method. As detailed below, the present programme builds on this earlier work with a dataset more than four times larger, derived from cores strategically placed to align with the research objectives. Coring covers both the Karnak site and its surroundings, and the data collected are synthesised with archaeological information (Gabolde 2018; Charloux et al. 2021), luminescence dating as well as an updated understanding of local floodplain dynamics (Toonen et al. 2018; Peeters et al. 2024). As a result, the reconstructions presented here are the most comprehensive interpretations currently available. We present information from 61 sediment cores (average depth: 6.4m, maximum depth: 11.65m) (Figure 2, Table S2), most of which were archaeologically dated through analysis of the tens of thousands of ceramic fragments they contained (Table S3). By positioning these cores at relatively even intervals, broadly along northern (‘N’) and southern (‘S’) west–east transects through the site (Figure 3), two cross-sections of subsurface deposits were created. These cross-sections were then interpreted using standard geoarchaeological methods (Miall 1996; Brown 1997; Toonen et al. 2018). Isochrons, based on typologically dated ceramic fragments, reveal the palaeotopography (Figure 3) and shed light on the dynamic nature of the environment (Figure 4). 6 The work is palaeogeographically contextualised (Figure 1) using lithological and chronostratigraphic information from approximately 150 cores similarly drilled across the wider local area (Toonen et al. 2018, 2019; Peeters et al. 2024), and a novel set of 48 optically stimulated luminescence (OSL) ages from the vicinity of Karnak (Peeters et al. 2024). Methods Sediments were retrieved using an Eijkelkamp hand auger (ASnnn) and/or a Cobra TT percussion corer (PCnnn) (Figure 2b–d) and their basic sedimentological characteristics were analysed in the field (see OSM). Recovered sediments were wet-sieved in approximately 100mm intervals, and fractions were manually sorted into ceramic and non-ceramic material. Ceramic fragments were assigned an age within a Karnak-specific typology (see OSM). Through this, archaeological chronostratigraphies were ascertained for most cores in as much detail as possible (Table S3). Results Our survey reveals that the entire Karnak zone is ultimately founded on sandy deposits (Figure 3: Units T, A, B). Based on their sedimentary characteristics, particularly their fine to medium sand grade, well-sorted nature and fining-upwards sequences, these deposits are interpreted as river channel sediments, though not all were laid down simultaneously. Further east, and also at isolated localities within the temple precinct, silts prevail (Units C & D). These record infills of abandoned river channels (Unit C) or floodplain sedimentation (Unit D), as fine material is dropped from suspension in lowenergy conditions. Atop these sands and silts typically lies several metres of cultural material: heterogeneous deposits comprising the ‘archaeology’ of the site (Unit E), as well as windblown cover (Unit F). 7 Figure 3. Simplified transects through Karnak: a) transect N; b) transect S. Cores/excavations indicated to aid location (not all shown). Figures S1–3 provide further detail. Elevation: masl, metres above sea level. Karnak’s river terrace foundation Although all the basal deposits are sands, a difference in age is inferred between the lowermost sands in the central/eastern zone of Karnak (Unit T: from the Middle Kingdom Court to East Karnak/east of the Sacred Lake), and the sands at either side (Units A & B) (Figure 3). Unit T contains no in-situ ceramic fragments except at its surface, but Units A and B (and all others) usually contain tens to hundreds of ceramic 8 fragments per vertical metre of cored sediment. This suggests that Unit T was deposited pre-occupation, but Units A and B were deposited contemporaneously with occupation. These units are also lithologically distinct. Further detail is provided by ceramic typology. The oldest ceramics atop Unit T date from sometime between the Sixth and early Eleventh dynasties: c. 2305–1980 BC (core AS012 and excavations E4–6) (Figure 2, Table S3), placing them within the First Intermediate Period or possibly the late Old Kingdom (Old Kingdom = c. 2591–2152 BC). They are of a similar age to the oldest ceramics within Units A and B (early Eleventh Dynasty: c. 2080–1980 BC; Hornung et al. 2006: 491), in cores AS143 and PC014. This also suggests Unit T is older than Units A and B. This age difference is reflective of the fact that Unit T, at an elevation of approximately 72m above sea level (masl), is an old river terrace—a preserved remnant of an earlier riverplain—a conclusion supported by 29 more cores drilled to the east of Karnak (Peeters et al. 2024) (Figure 1). The relatively large grain size (150–350µm) of Unit T sands suggests deposition by fast-flowing water. Later, this alluvial plain was eroded on both its eastern and western sides by incising river channels, leaving a segment of high ground—a terrace (island)—in the east/south-east of the present site that then became occupied. The courses of the river channels that carved out this island are marked by Units A and B. Unit A is interpreted as lower channel bar sediments: sands laid down near the margins of the channels (Miall 1996), which generally fine upwards into the upper bar and levee sediments of Unit B. After deposition of Unit T ceased, the terrace top lay above the normal level of inundation; since Unit B deposits dating to the early Middle Kingdom are nowhere situated atop this terrace. The 29 cores drilled further east (Peeters et al. 2024) show that the terrace existed for several kilometres in this direction, and they also provide an OSL age for the terminal phase of deposition at this level of 4.54±0.42 ka (2520 BC ±420 yr). The fluvial erosion on either side that formed the terrace/island would thus have taken place after this date. 9 The age of this river terrace is important because it places a temporal constraint upon earliest occupation at Karnak. When this area was being actively deposited by fastflowing water, at and prior to 2520 BC ±420 years, it was unsuitable for permanent occupation/construction, although non-permanent activities may have occurred there during annual stages of low flow. After this date, incision resulted in the formation of the terrace segment/island, upon which occupation could have been initiated. This chronology corroborates the ceramic information: the earliest ceramics date from sometime between c. 2305–1980 BC, within the OSL 68.27% (1σ) confidence interval. The coring results cannot precisely delimit the northern or southern margins of the island. Five cores further south, at the temple of Mut, provide little information, primarily encountering Unit E only (AS024–028: Figure S3). However, the narrowness of the terrace observed in transect S (Figure 3b) suggests proximity to its southern tip, corroborating previous ideas (see Introduction). The northern limit may be at North Karnak, based on earlier cores and excavation data (Bunbury et al. 2008). The island may, therefore, have extended from North Karnak to near the southern enclosure wall, an area of approximately 10ha (Figure 4a). A small channel crossed the terrace at the rear of the extant Amun-Ra temple, shown by Unit C in PC030 and PC061(Figure 3a); the ceramic records of these cores suggest that the channel gradually silted up through the Middle Kingdom and Second Intermediate Period (Second Intermediate Period = c. 1759–1539 BC) (Table S3). It also appears that another, smaller terrace segment lay in the south-west corner of the site, upon which the Opet Temple was constructed (Figures 3b & 4a), since early Middle Kingdom archaeology is found atop Unit T (AS030) in this location (Charloux et al. 2012). The deposits here (Unit T) are lithostratigraphically correlated with the main terrace segment. They were deposited concurrently and were originally continuous across the area. However, by the early Middle Kingdom (if not before) the areas were interposed by an incised channel (Units A, B, C in AS033, AS143, PC060, AS011, hosting early Middle Kingdom ceramics), thus placing early occupation on individual islands (Figure 4a). 16 abated, the mound upon which Karnak was built would have appeared to rise from the receding water. The illusion of Karnak rising from the inundation would also have been enhanced by a further aspect of the site’s geological configuration. A surface of early Middle Kingdom ceramics lying atop bar deposits in PC026, AS029 and PC048 (transect N) and PC045 and PC073 (transect S) (Figures 3 & 4a) indicate that an area of land 1m lower than the terrace top (at approximately 71masl) surrounded the terrace in the early Middle Kingdom. During flood conditions this level would have been inundated, since early Middle Kingdom levee deposits of Unit B lie directly above, while the top of the terrace remained dry. As the flood receded, the island could have doubled in area (Figure 4a) as the bar deposits ‘came out’ of the water, thus completing the impression of the upward movement of the mound. Seasonal activity would have been possible on the lower levels. The occupants of the island were also opportunistic in adapting to changes in the local fluvial environment. The post-New Kingdom westward expansion of the temple (Bunbury et al. 2008: 351) demonstrates the progressive utilisation of new land, with the construction of the seventh to tenth and first to third pylons as the western channels silted up (Figure 4b–d). In some instances, changes to the landscape were made proactively, indicated by the dumping of desert sand into a channel just south of what was shortly to become the Hypostyle Hall (PC060). Given that a large river channel originally lay to the east of the site, it would be instructive to consider further archaeological surveys in that area. Early archaeological features may be focused there, dating from before the eastern channel moved away and the western channel became a larger, more clearly-defined feature in the landscape (Figure 4a–e). A change of temple orientation from east to west based upon architectural evidence has even been proposed, with the earliest temple facing east (Larché 2007: 481– 83), though this remains contentious (Gabolde 2018: 225–26). 17 Conclusion New data from 61 sediment cores has allowed for a more detailed palaeogeographic reconstruction of evolving landscape settings at Karnak. Earliest possible occupation at the site is constrained probably to the Old Kingdom, and activity there demonstrates a coupling between the natural environment and the religious, functional, and constructional aspects of the temple. Understanding the evolution of Karnak—from a small island to one of the defining institutions of Ancient Egypt—is thus only possible with advancing knowledge of its ever-changing environment. Acknowledgments We thank the Egyptian Ministry of Tourism and Antiquities, all at the Centre FrancoÉgyptien d'Étude des Temples de Karnak (CFEETK), and at Chicago House (Luxor / University of Chicago), the Farouk family, and our local team members. The research was carried out under the auspices of the Egypt Exploration Society (London). Funding Statement The work was supported by the Knut och Alice Wallenbergs Stiftelse (KAW 2013.0163) and Uppsala Universitet (HUMSAM 2014/17) to A.G. as Wallenberg Academy Fellow 2014 –20, together with a small grant from M och S Wångstedts Stiftelse (A.G.). Data availability statement The summary geological data supporting the findings of this study are available in Zenodo at https://doi.org/10.5281/zenodo.11581016. Technical details of the sedimentary interpretation as well as detailed sedimentary figures, core metadata and ceramic chronostratigraphies are provided in the Online Supplementary Material. Competing Interest Declaration Competing interests: The authors declare none. The views expressed in the article do not necessarily represent the views of the NPS or the government of the United States. 18 References ARNAUDIÈS, A. & E. LAROZE. 2007. Localisation des interventions archéologiques dans le temple de Karnak, 1967-2004. Cahiers de Karnak 12: 91–103. ARNOLD, D. 1976. Gräber des Alten und Mittleren Reiches in El-Tarif. Mainz am Rhein: Deutsches Archologisches Institut. Abteilung Kairo. AZIM, M. 1980. La fouille de la cour du Xe pylône. Rapport préliminaire. Cahiers de Karnak 6: 160–61. BARGUET, P. 1962. Le temple d’Amon-Rê à Karnak: Essai d’exégèse. Cairo: Institut Français d’Archéologie Orientale. BICKEL, S. 1994. La cosmogonie égyptienne avant le Nouvel Empire. Fribourg, Switzerland: Orbis Biblicus et Orientalis, 134. BORAIK, M., L. GABOLDE & A. GRAHAM. 2017. Karnak’s quaysides: evolution of the embankments from the eighteenth dynasty to the Graeco-Roman period, in H. Willems & J.-M. Dahms (ed.) The Nile: Natural and cultural landscape in Egypt.: 97–144. Bielefeld: Transcript Verlag. BROWN, A.G. 1997. Alluvial Geoarchaeology: Floodplain archaeology and environmental change. Cambridge: Cambridge University Press. BUNBURY, J.M., A. GRAHAM & M.A. HUNTER. 2008. Stratigraphic landscape analysis: charting the Holocene movements of the Nile at Karnak through ancient Egyptian time. Geoarchaeology 23: 351–73. CARLOTTI, J.-F., E. CZERNY & L. GABOLDE. 2010. Sondage autour de la plate-forme en grès de la “cour du Moyen Empire” à Karnak. Cahiers de Karnak 13: 111–93. CHARLOUX, G. & R. MENSAN. 2011. Karnak avant la XVIIIe dynastie. Paris: Soleb. CHARLOUX, G., R. ANGEVIN, S. MARCHAND, H. MONCHOT, J. ROBERSON & H. VIRENQUE. 2012. Le parvis du temple d’Opet à Karnak: exploration archéologique, 2006-2007. Cairo: Institut Français d’Archéologie Orientale. CHARLOUX, G. et al. 2018. Le temple ‘primitif’ de Ptah à Karnak. Bulletin de l’Institut Français d’Archéologie Orientale 117: 125–29. CHARLOUX, G., M.A. ABADY MAHMOUD, A.M.S. ELNASSEH & S. MARCHAND. 2021. The shifting Nile and the origins and development of ancient Karnak. Antiquity 95: 919–39. EGLI, E. 1959. Geschichte des Stdtebaues: Erster Band: Die alte Welt. Erlenbach: Eugen Rentsch. FRANCHET, L. 1917. Travaux effectués en Égypte. Rapport sur une mission en Crète et en Égypte (1912-1913), dans Nouvelles archives des missions scientifiques 22: 83–99. GABOLDE, L. 2018. Karnak, Amon-Rê: la genèse d’un temple, la naissance d’un dieu (avec des rapports de Sylvie Marchand et de Samuel Guérin). Cairo: Institut Français d’Archéologie Orientale. GHILARDI, M. & M. BORAIK. 2011. Reconstructing the Holocene depositional environments in the western part of ancient Karnak temples complex (Egypt): a geoarchaeological approach. Journal of Archaeological Science 38: 3204–16. GRAHAM, A. 2010. Islands in the Nile: a geoarchaeological approach to settlement locations in the Egyptian Nile Valley and the case of Karnak, in M. Bietak, E. Czerny & I. ForstnerMüller (ed.) Cities and Urbanism in Ancient Egypt: 125–43. Vienna: Österreichischen Akademie der Wissenschaften. 19 HASSAN, F.A. 1997. The dynamics of a riverine civilization: a geoarchaeological perspective on the Nile Valley, Egypt. World Archaeology 29: 51–74. HORNUNG, E., R. KRAUSS & D.A. WARBURTON. 2006. Ancient Egyptian Chronology. Leiden: Brill. JACQUET, J. 1983. Karnak-Nord V (Fouilles de l’Institut français d’archéologie orientale 31). Cairo: Institut Français d’Archéologie Orientale. —. 2001. Karnak-Nord IX (Fouilles de l’Institut français d’archéologie orientale 44). Cairo: Institut Français d’Archéologie Orientale. LARCHÉ, F. 2007. Les nouvelles observations sur les monuments du Moyen et du Nouvel Empire dans la zone centrale du temple d’Amon à Karnak. Cahiers de Karnak 12: 407– 592. —. 2020. Chronologie des vestiges découverts sous le temple de Karnak. Turnhout: Brepols. —. 2023. Sous le sol de la Salle Hypostyle et de son voisinage à Karnak, in F. Colin (eds.) Audelà de Karnak : recueil d'études offert à Claude Traunecker, Cahiers de l'ENIM (CENIM) 35: 357–404. Paris: Khéops. LAUFFRAY, J. 1968. Nouvelles découvertes aux temples de Karnak. Comptes-rendus des séances de l’Académie des Inscriptions et Belles-Lettres 112: 337–51. —. 1969. Rapport sur les travaux de Karnak. Activités du Centre franco-égyptien en 19671968. Cahiers de Karnak 2: 111–35. LEGRAIN, G. 1906a. Excavations and explorations, in F.L. Griffith (ed.) Archaeological report: comprising the work of the Egypt Exploration Fund and the progress of egyptology during the year 1905-1906: 21–23. London: Egypt Exploration Fund. —. 1906b. Nouveaux renseignements sur les dernières découvertes faites à Karnak (15 novembre 1904 - 25 juillet 1905). Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes et assyriennes 28: 137–61. —. 1929. Les temples de Karnak. Brussels: Vromant & Co. MASSON-BERGHOFF, A. 2021. Le quartier des prêtres dans le temple d’Amon à Karnak. Leuven: Peeters. MIALL, A.D. 1996. The Geology of Fluvial Deposits. New York: Springer-Verlag. MILLET, M. 2007. Architecture civile antérieure au Nouvel Empire: rapport préliminaire des fouilles archéologiques à l’est du lac Sacré 2001–2003. Cahiers de Karnak 12: 681–743. —. 2008. Installations antérieures au Nouvel Empire au sud-est du Lac sacré du temple d’Amon de Karnak. PhD Thesis; École doctorale Mondes anciens et médiévaux, Sorbonne Université. PEETERS, J. et al. 2024. Shift away from Nile incision at Luxor ~4,000 years ago impacted ancient Egyptian landscapes. Nature Geoscience 17: 645–53. PENNINGTON, B.T., P. WILSON, F. STURT & A.G. BROWN. 2020. Landscape change in the Nile Delta during the fourth millennium BC: a new perspective on the Egyptian Predynastic and Protodynastic periods. World Archaeology 52: 550–65. POPIELSKA-GRZYBOWSKA, J. 2016. Contexts of the appearance of water in the Pyramid Texts: an introduction. Études et Travaux 29: 157–67. POSTEL, L. 2004. Protocoles des souverains égyptiens et dogme monarchique au début du Moyen Empire. Turnhout: Brepols, Fondation égyptologique Reine Élisabeth. REDFORD, D.B. 1988. Interim Report on the 20th Campaign (17th Season) of the Excavations at East Karnak. Journal of the Society for the Study of Egyptian Antiquities 18: 36–47. 20 —. 1994. The Excavations of Kom el-Ahmar and Environs. Toronto: Akhenaten Temple Project. REDFORD, D.B., S.E. OREL, S. REDFORD & S. SHUBERT. 1991. East Karnak Excavations, 1987-1991. Journal of the American Research Center in Egypt 28: 75–106. TOONEN, W.H.J. et al. 2018. Holocene fluvial history of the Nile floodplain at ancient Thebes (Luxor, Egypt) and its relation with cultural dynamics and basin-wide hydroclimatic variability. Geoarchaeology 33: 273–90. TOONEN, W.H.J. et al. 2019. Amenhotep III’s Mansion of Millions of Years in Thebes (Luxor, Egypt); submergence of an elevated mound by river floods and Nile sediments. Journal of Archaeological Science: Reports 25: 195–205. TOONEN, W.H.J., K. CORTEBEECK, S. HENDRICKX, B. BADER, J. PEETERS & H. WILLEMS. 2022. The hydro-geomorphological setting of the Old Kingdom town of al-Ashmūnayn in the Egyptian Nile Valley. Geoarchaeology 37: 267–83. ULLMANN, M. 2007. Thebes: Origins of a ritual landscape, in P.F. Dorman & B.M. Bryan (ed.) Sacred space and sacred function in Ancient Thebes: 3–27. Chicago: Oriental Institute of the University of Chicago. 21 [ONLINE SUPPLEMENTARY MATERIAL] Conceptual origins and geomorphic evolution of the temple of Amun-Ra at Karnak (Luxor, Egypt) Benjamin T. Pennington1, Angus Graham2,*, Aurélia Masson-Berghoff3,4, Marie Millet5, Jan Peeters6, Willem H. J. Toonen7,8, Timotheus G. Winkels9, Luke H. Sollars10, Virginia. L. Emery11, Kristian D. Strutt12, Dominic S. Barker12 1 Geography & Environmental Science, University of Southampton, UK 2 Department of Archaeology, Ancient History and Conservation, Uppsala Universitet, Sweden 3 Department of Greece & Rome, The British Museum, London, UK 4 Department of Egypt & Sudan, The British Museum, London, UK 5 Department of Egyptian Antiquities, Musée du Louvre, Paris, France 6 Jebel Barkal Archaeological Project, Kelsey Museum of Archaeology, University of Michigan, Ann Arbor, USA 7 Faculty of Science, Earth & Climate, Vrije Universiteit Amsterdam, The Netherlands 8 Egyptology Unit, Katholieke Universiteit Leuven, Belgium 9 Water and Environment Group, WSP, Nieuwegein, The Netherlands 10 Independent Researcher, Salisbury, UK 11 National Park Service, Fort Union National Monument, Watrous, USA 12 Department of Archaeology, University of Southampton, UK *Corresponding author (email: angus.gra[email protected]) Extended Methodology Sediments were retrieved using an Eijkelkamp hand auger (cores coded ASnnn), or Cobra TT percussion corer (cores coded PCnnn), before being analysed with a hand lens in the field in terms of their basic sedimentary characteristics (grain size, sedimentary texture (Ditzler et al. 2017), sorting, rounding, mineralogy, Munsell colour, organic 22 material, inclusions). Summary supporting geological information are provided on Zenodo as outlined in the Data Availability Statement. The cores and the transects along which they were placed were then interpreted using standard sedimentological, geomorphological and geoarchaeological methods (Miall 1996; Brown 1997). The two transects (N and S) were primarily drilled in 2017–2018 (Table S2) using the percussion corer, which was able to penetrate archaeological deposits effectively. Prior to this date (see Table S2), earlier investigations with the hand auger were often limited to excavations within the temple complex and were less spatially systematic. This was due to the fact that it was often not possible to hand-auger through the sometimes dense anthropogenic debris, which often included (pebble-sized) ceramic and stone fragments. Core locations were primarily surveyed using either RTK-DGPS or total station using the Survey of Egypt datum (Table S2). The sediments recovered were divided in c.10cm intervals, wet-sieved through 2mm and 4mm mesh, and each sieved fraction sorted manually into ceramic and non-ceramic material. For the majority of cores, within each sieved fraction of each sample, each item of non-ceramic material was identified (rhizoconcretions, bone, quartz, mudbrick etc.) and counted and overall abrasion and rounding was recorded. All sieved non-ceramic and ceramic material is stored at Karnak in the storerooms of the Ministry of Tourism and Antiquities, for possible future re-study. After the ceramics recovered from the cores had been sieved and isolated, they were studied from each core in sequential order to provide full archaeological stratigraphies for the majority of the cores (Graham et al. 2012: 30–32). The ceramic material was studied with no knowledge of the sedimentary data or interpretation to avoid bias. All sherds from each sieved fraction within each 10cm sample in these cores were analysed in terms of total number of sherds, total weight, size range, fragment type (rim, base, handle, body), decoration, technique (handmade, slow wheel, quick wheel), abrasion, rounding, hardness, material (Nile silt, marl clay, oasis, Aswan, import), fabric and period. The fabric system used is a local system based upon analysis of material originating from stratified settlement contexts from the First Intermediate Period through to the Roman Period in Karnak, excavated between 2001–2007 (Masson 2007, 2009, 2011, 2012, 2015, 2016; Millet 2007, 2008; David et al. 2016; Masson-Berghoff 2021): some of these 23 fabrics correspond to the well-known Vienna system (Arnold & Bourriau 1993; Bourriau et al. 2000). Each diagnostic sherd was drawn, photographed, and recorded individually. Through this process, a chronology for each core was established (Table S3). In numerous instances, changes in the ceramic assemblage corresponded to independentlyobserved changes in the hosting sediments, lending weight to both interpretations. Using both artefactual and ‘articlast’ (Rapp & Hill 2006: 29, 51–52) information from the sherds, it was also possible to identify when downhole contamination during the coring process had occurred, ruling out the use of certain material from the chronology of the remaining assemblage. Priority for comprehensive ceramic determination as described above was given to deep (>c.5m) cores that yielded plentiful, well stratified ceramics in significant locations. Five cores which terminated at shallow depths did not have their ceramics studied, while 15 for which it would not necessarily have been more productive to look at exhaustively were analysed via a ‘bulk assessment’ (Table S3, Figures S1–S3 indicate which ones). The ceramics from these cores were not systematically examined with a fresh break and a lens. Instead, priority was given to diagnostics (forms and specific fabrics) or sherds that retained surface treatments (slip, decoration), although general observations on the amount of material and the state of preservation were also noted for each sample; some of these cores also did not have their non-ceramic clasts individually studied. Nonetheless, the large number of ceramics present in many of these cores – including numerous diagnostics – still allows for a very high degree of confidence in their chronological interpretation: a sequence was noted only when there were clear shifts and none of the material was contradictory. This research programme from which this work originates – the Theban Harbours and Waterscapes Survey (THaWS) – has been directed since 2010 by Angus Graham. The permit granted by the Ministry of Tourism and Antiquities to the Egypt Exploration Society covers an area of approximately 70 km2 within the Nile Valley from at-Tarif in the north to south of Birket Habu on both the east and west banks of the floodplain and includes the Temple of Karnak complex. A precursor research programme, the Karnak 24 Landand Waterscapes Survey (KLaWS) was co-directed by Judith Bunbury (University of Cambridge, UK) and Angus Graham (at the time University College London, UK) from 2002 until 2009 (Bunbury et al. 2008). This research programme worked principally within the area of the Karnak complex. Sedimentary interpretation – technical details The deposits encountered in the sediment cores can be divided into eight main sedimentary units (Figures S1–3). They display many similarities with other deposits within the Theban floodplain region (Toonen et al. 2018, 2019; Peeters et al. 2024), and in the vicinity of other archaeological sites in Egypt situated in proximal fluvial contexts (Tronchère et al. 2009; Pennington & Thomas 2016; Hassan et al. 2017; Pennington 2019; Toonen et al. 2022). At the base of the sequence lies Unit T (terrace), localised both within central/eastern Karnak and also in the vicinity of AS030. This unit usually consists of very well sorted, micaceous fine to medium sands (150–350µm), although near the top there are also interbedded coarse silts. Texturally the unit comprises loams to sands (Ditzler et al. 2017). The deposits include short (1–2m) fining upward sequences and usually have a Munsell colour in the vicinity of 10YR 4/2. Small calcareous rhizoliths (Klappa 1980), other small concretions (manganese and iron suspected), and black mottling (reduction haloes around organic matter as it decomposes) are frequent, especially within the finer upper parts of the unit; some plant remains and other organic materials are also present in these upper sections. The relatively large grain size of the sediments suggests deposition by reasonably fast-moving water, while the concretions suggest minor soil formation processes within a setting that was subsequently subaerially exposed. The unit contains no convincing in situ ceramic fragments. While 17 tiny (mm-scale), undatable, abraded fragments were recovered from the unit (all of which would likely have been missed in a standard archaeological investigation of these layers), they are all thought to result from downhole contamination from the coring process. When a core segment was retrieved for study (every ~10cm with the hand auger / 1m with the 25 percussion corer), sometimes a very small amount of surface sediment fell down the hole, despite best efforts. At Karnak, where the modern-day surface (Unit F, below) contains plentiful minute abraded fragments, on occasion a few mm-scale sherds would have unavoidably fallen in too. They are not considered to have been originally hosted within Unit T. Such tiny numbers of miniscule fragments are of course present as 'background noise' within the other units. However, these other units also contain tens of thousands of in situ fragments (as 'signal'), so the noise is lost. In Unit, where a signal is absent, such noise is of course noticeable in Table S3. Some aspects of Unit A (bar deposits) are similar to Unit T, in that Unit A also comprises sandy deposits with fining upwards sequences. However, the sequences are longer, Unit A is less variable in grain size and it contains very little (if any) organic material or rhizoliths, nor any black mottling. Instead, it contains frequent ceramic material, often smoothed by water erosion. It is also texturally more homogeneous (sandy loams to sands), often a little finer (125–250µm), a little less micaceous, and has a slightly lighter colour (commonly 10YR 4/3). The unit comprises a textbook example of fluviatile bar sediments deposited in an ancient branch of the Nile. The fining-up sequences are typical, while the lack of rhizoliths and organic matter are to be expected in a fluvial channel where deposition is taking place within flowing water and there is no subsequent subaerial exposure. The unit is very similar to other such deposits encountered elsewhere in Egypt (Garzanti et al. 2015; Pennington & Thomas 2016; Pennington 2019; Toonen et al. 2022). Unit A often fines upward into Unit B (upper bar/levee), which usually comprises very dark brown (often 10YR 3/2), well-sorted, micaceous very fine sands (loams to sandy loams) with frequent black mottling, rhizoliths, ceramics and other anthropogenic material. The finer grain size (30–150µm) reflects a lower energy of deposition than the lower bar deposits of Unit A, and the presence of rhizoliths reflects minor soil formation processes within a periodically subaerially exposed environment, as would be expected on a levee or upper bar. 32 Table S1. Chronological periodisation. Calendar dates follow dynasty numbers, and are approximate prior to 664 BC. Dynasty numbers 9, 10, 14 and 15 are not applicable for the Theban region. Time period Abbreviation Dynasty Age Neolithic Period - - 5500 – 3900 BC (Tassie 2014) Predynastic Period - - 3900 – 3100 BC (Dee et al. 2013; Tassie 2014) Early Dynastic Period ED D1 – D2 3100 – 2592 BC (Hornung et al. 2006; Dee et al. 2013) Old Kingdom OK D3 – D6 2592 – 2152 BC (Hornung et al. 2006) First Intermediate Period FIP D7 – Early D11 2152– 1980 BC (Hornung et al. 2006) Middle Kingdom MK Late D11 – D13 1980 – 1760 BC (Hornung et al. 2006) Second Intermediate Period SIP D16 – D17 1759 – 1539 BC (Hornung et al. 2006) New Kingdom NK D18 – D20 1539– 1077 BC (Hornung et al. 2006) Third Intermediate Period TIP D21 – D25 1076 – 664 BC (Hornung et al. 2006) Late Period LP D26 – D31 664 – 332 BC (Hornung et al. 2006; Payraudeau 2020) Macedonian and Ptolemaic Periods PP - 332 – 30 BC (Hornung et al. 2006) Roman and Byzantine Periods RP - 30 BC – AD 642 (Lloyd 2010) 33 Table S2. Locations of cores drilled at Karnak. Easting and Northing are in UTM 36N; Elevation refers to the top of the core with respect to the Survey of Egypt datum. Core Easting Northing Elevation (masl) Total Depth (m) Date AS001*,† 466078.75 2844874.52 76.41 0.29 Spring 2002 AS002*,† 466083.13 2844872.10 76.79 10.95 Spring 2002 AS003*,† 466106.18 2844940.01 76.55 5.51 Spring 2002 AS004*,† 466102.92 2844938.89 76.66 3.70 Spring 2004 AS006*,† 466106.42 2844939.77 76.67 7.80 Spring 2004 AS007*,† 466115.41 2844946.12 76.05 2.88 Spring 2004 AS008*,† 466111.91 2844948.47 75.91 7.32 Spring 2004 AS010† 465767.72 2844529.70 72.88 5.48 Spring 2004 AS011† 465597.96 2844386.59 74.19 8.64 Spring 2004 AS012†,‡ 465807.64 2844221.76 75.30 7.20 Spring 2004 AS013† 465414.57 2844710.93 76.11 2.24 Spring 2004 AS014† 465489.19 2844675.48 71.97 2.34 Spring 2004 AS015† 465487.54 2844673.29 71.98 1.91 Spring 2004 AS016† 465487.79 2844666.41 72.40 3.79 Spring 2004 AS017†,‡ 465792.87 2844492.80 72.84 3.68 Spring 2004 AS024* 465414.96 2843764.80 73.35 1.18 Spring 2006 AS025* 465409.45 2843767.48 73.12 4.42 Spring 2007 AS026* 465434.28 2843758.77 73.26 2.81 Spring 2007 AS027* 465349.07 2843822.63 71.91 6.46 Spring 2007 AS028* 465587.37 2844001.45 76.85 4.76 Spring 2007 AS029‡ 465748.63§ 2844507.74 71.74 3.99 Spring 2007 AS030‡ 465423.70 2844348.16 71.44 2.84 Spring 2007 AS031 465445.98 2844547.16 71.17 2.37 Spring 2007 AS032 465860.90§ 2844807.75 79.27 3.87 Spring 2007 AS033 465514.36 2844310.23 77.39 9.20 Spring 2008 AS034 465528.58 2844727.42 71.97 4.30 Spring 2008 AS039‡ 465486.45 2844653.72 70.85 3.85 Spring 2012 AS040 465733.39§ 2844617.71 74.71 1.66 Spring 2012 AS113‡,¶ 466045.30 2844460.98 74.70 7.00 Autumn 2017 AS114¶ 466568.98 2844391.83 74.89 7.60 Autumn 2017 AS143 465583.45 2844266.02 75.70 10.88 Spring 2018 AS165 465829.09 2844465.60 72.76 6.20 Spring 2018 * Core not shown on Figures 2–4. AS001–008 lie further north and are presented elsewhere (Bunbury et al. 2008); AS024–028 are further south: their records are shown on Figure S3. † Core previously published (Bunbury et al. 2008), although record and ceramic stratigraphy reexamined in the current work, where applicable. ‡ Core drilled within archaeological excavation. Ceramics from the excavations lying directly above the top of AS012 and AS030 were also studied by the project. § The XY locations of cores AS029, AS032, AS040 are correct to approximately 75cm, 1m and 4m respectively (with associated vertical errors). All other core locations were surveyed by total station (AS001–AS039) or RTK-DGPS (others) with a negligible XYZ error. ¶ Basic core record included in a previously published dataset (Peeters et al. 2024). 34 AS167 466468.58 2844370.78 74.22 3.10 Autumn 2018 AS169¶ 466483.01 2844360.38 74.84 9.20 Autumn 2018 PC013 465812.93 2844178.72 77.39 6.00 Autumn 2016 PC014 465798.77 2844178.67 76.61 6.90 Autumn 2016 PC025¶ 465707.04 2844628.59 74.81 7.00 Autumn 2017 PC026‡,¶ 465748.90 2844570.53 73.31 4.20 Autumn 2017 PC027¶ 465834.94 2844558.14 74.89 9.80 Autumn 2017 PC028 465903.11 2844527.63 74.76 1.20 Autumn 2017 PC029¶ 465981.76 2844473.09 75.22 8.20 Autumn 2017 PC030¶ 465894.29 2844512.92 74.41 11.20 Autumn 2017 PC031¶ 466164.74 2844387.35 76.10 11.20 Autumn 2017 PC032¶ 466343.62 2844397.21 76.70 10.00 Autumn 2017 PC033 465572.53 2844723.51 77.14 2.20 Autumn 2017 PC034¶ 465572.23 2844668.68 74.37 10.70 Autumn 2017 PC042¶ 465479.98 2844726.80 75.82 7.50 Autumn 2017 PC043¶ 465447.01 2844744.85 76.41 10.00 Autumn 2017 PC044¶ 465309.58 2844791.32 77.05 5.00 Autumn 2017 PC045 465686.94 2844214.24 76.65 9.80 Spring 2018 PC046 465739.27 2844194.58 76.94 9.90 Spring 2018 PC047¶ 465661.14 2844654.99 75.01 10.10 Spring 2018 PC048¶ 466129.62 2844407.31 77.91 11.00 Spring 2018 PC049¶ 466251.71 2844411.59 77.31 11.65 Spring 2018 PC060 465633.48 2844519.57 74.24 8.60 Spring 2018 PC061 465888.01 2844480.71 73.85 9.85 Spring 2018 PC062 465522.61 2844555.75 74.68 10.00 Autumn 2018 PC063 465806.83 2844515.42 73.68 5.40 Autumn 2018 PC071 465973.79 2844104.66 77.06 9.00 Spring 2019 PC072 465919.02 2844103.08 77.63 9.00 Spring 2019 PC073 465853.04 2844135.68 77.50 7.00 Spring 2019 35 Table S3. Core chronostratigraphies from ceramic data (all cores on Figures 2–4). Dashes (–) indicate that the material may encompass the range given, or date from any point within the time period. B.A. indicates "Bulk Assessment" (see extended methodology). Core Year studied Sherds (number) Elevation (masl) Archaeological period AS010* 2005 99 72.42–72.88 MK–LP 69.24–72.12 MK–NK 68.00–69.24 Mainly MK 67.40–68.00 No ceramics AS011 2005 & 2009 3073 71.15–73.55 NK–PP 69.83–71.15 MK–LP 66.91–69.83 Early MK–MK (mid-Dynasty 12) 65.55–66.91 No dating possible AS012 2005 141 72.61–75.30 MK 72.43–72.61 Late OK–FIP (Dynasty 6 – early Dynasty 11) 68.1–72.29 Very few sherds, thought to be downhole contaminants AS013 Not studied† - - - AS014 Not studied - - - AS015 Not studied - - - AS016* 2005 376 68.61–71.49 MK–RP AS017*,‡ 2005 14 72.46–72.84 NK–RP 72.29–72.46 SIP–TIP 71.89–72.29 No dating possible 69.16–71.89 No ceramics AS029 2009 247 70.56–71.74 Early MK 67.75–70.56 Various small ceramics, no dating possible§ AS030 2009 2 - No ceramics except two downhole contaminants AS031 2009 119 68.80–71.17 TIP–PP AS032 2007 1106 77.42–79.27 NK–PP 76.43–77.42 NK–LP 75.40–76.43 NK AS033 2009 1869 75.52–77.39 RP 74.40–75.14 NK–RP 73.61–74.40 NK 73.31–73.61 MK–SIP 72.57–73.31 MK * Ceramics from cores AS010, AS016 and AS017 were studied by Sally-Ann Ashton and Irmgard Hein, without the insights of some of the local excavations. † AS013–015 all terminated at very shallow depths due to impenetrable debris. ‡ Only large sherds looked at in this core. § Sherds heavily abraded. 36 71.51–72.57 MK, probably Dynasty 12 68.31–71.51 Early MK 68.19–68.31 No dating possible AS034 2009 613 67.67–71.97 NK–PP AS039 2012 485 68.94–70.85 NK 68.45–68.94 MK–NK 67.00–68.45 MK–early NK AS040 2012 187 73.41–74.71 TIP–PP/RP? 73.05–73.41 NK AS113 2019 113 72.60–74.70 MK–NK / MK–SIP 72.48–72.60 Early MK? 71.01–72.48 5x tiny sherds, possibly contaminants 67.70–71.01 No ceramics AS114 2018 B.A. 68.49–74.89 No dating possible 67.29–68.49 NK–PP AS143 2018 2076 73.40–75.70 NK–RP 73.00–73.40 NK 72.70–73.00 SIP–NK 72.10–72.70 (MK–)SIP 71.25–72.10 MK 69.28–71.25 Early MK 68.55–69.28 FIP (Dynasty 11) 65.65–68.55 Very few sherds; no dating possible 64.82–65.65 No ceramics AS165 2019 318 70.96–72.76 MK–NK¶ 70.16–70.96 Early MK–MK 66.56–70.16 No ceramics AS167 Not studied# - - - AS169 2019 B.A. 74.24–74.84 No dating possible 69.49–74.24 NK–PP? 66.04–69.49 No dating possible PC013 2017 1150 76.79–77.39 Mixed 75.79–76.79 SIP–NK 75.39–75.79 MK–NK 74.29–75.39 MK–SIP 73.89–74.29 MK 71.99–73.89 Few sherds; MK inferred 71.39–71.99 Early MK PC014 2017 1682 76.11–76.61 Mixed 76.01–76.11 SIP–NK 75.01–76.01 MK–SIP 72.51–75.01 MK ¶ Some sherds have concretions. # Terminated at a very shallow depth; AS169 regarded as representative of the area. 37 71.01–72.51 Early MK 69.71–71.01 FIP (Dynasty 11?) PC025 2018 B.A. 74.61–74.81 RP 74.31–74.61 MK–RP 72.61–74.31 TIP–PP (mostly Dynasty 25–26) 71.51–72.61 MK–NK 67.81–71.51 SIP–NK PC026 2019 471 72.91–73.31 No dating possible 72.51–72.91 Mixed MK–NK? 70.56–72.51 MK–SIP 69.11–70.56 No ceramics PC027 2018 B.A. 65.09–74.89 Mixed NK–RP PC028 Not studied|| - - - PC029 2019 1035 72.55–75.29 MK–NK 72.12–72.55 SIP–NK 67.02–72.12 No ceramics PC030 2018 B.A. 73.91–74.41 RP 73.51–73.91 No dating possible 73.31–73.51 TIP–PP 72.41–73.31 MK–NK? 64.71–72.41 Early MK–MK? Mixture of material 64.31–64.71 MK 63.21–64.31 Early MK? PC031 2018 1594 75.80–76.10 Mixed 74.70–75.80 SIP–NK 73.10–74.70 Downhole contamination 69.50–73.10 Early MK plus downhole contamination 64.90–69.50 Sporadic sherds throughout; no dating possible PC032 2018 B.A. 66.70–76.70 Mixed NK–RP?** PC033 2019 B.A. 75.94–77.14 RP 74.94–75.94 No ceramics PC034 2018 B.A. 73.77–74.37 PP–RP 72.57–73.77 LP–PP 72.07–72.57 Sporadic sherds throughout; no dating possible 70.57–72.07 MK–NK 63.67–70.57 Sporadic sherds throughout; no dating possible†† PC042 2019 B.A. 70.72–75.82 PP–RP 69.82–70.72 MK–RP 69.62–69.82 RP 68.32–69.62 NK–RP?‡‡ || Terminated at a very shallow depth. ** In lower part of core, sherds are very rolled and blackened by water. †† Possibly all downhole contamination. ‡‡ The sherds are few, very abraded, and difficult to date. 38 PC043 2018 B.A. 75.81–76.41 RP–Modern 75.11–75.81 RP? 66.42–75.11 Sporadic sherds throughout; no dating possible PC044 2019 B.A. 75.85–77.05 No ceramics 75.35–75.85 RP 72.05–75.35 No ceramics PC045 2019 2335 76.15–76.65 NK–RP 75.55–76.15 NK 73.65–75.55 SIP–NK 73.35–73.65 SIP? 72.65–73.35 MK–SIP 71.35–72.65 MK (below 72.35m is Dynasty 12) 71.25–71.35 No dating possible 68.25–71.25 Early MK? 67.45–68.25 No dating possible 66.85–67.45 No ceramics PC046 2018 B.A. 76.44–76.94 Mixed RP–Modern? 75.94–76.44 RP 74.74–75.94 MK–RP? 74.04–74.74 MK–NK 73.84–74.04 Early MK 67.04–73.84 Very few sherds, thought to be downhole contaminants PC047 2019 2457 71.81–75.01 TIP–LP 71.01–71.81 NK–TIP 69.61-71.01 NK 66.31–69.61 SIP–NK (Dynasty 17/Early Dynasty 18)§§ 64.91–66.31 No ceramics PC048 2018 794 76.61–77.91 NK–RP/Mixed 75.51–76.61 NK–PP 73.71–75.51 NK/Mixed 73.31–73.71 NK 72.81–73.31 MK (early MK?) 72.21–72.81 Very few mm-scale sherds 71.41–72.21 MK 70.51–71.41 Early MK? 68.91–70.51 No ceramics 68.51–68.91 MK?/Possibly downhole contamination 66.91–68.51 No ceramics PC049 2018 1411 73.01–77.31 Mixed 72.81–73.01 NK–TIP 68.71–72.81 MK (Dynasty 12)¶¶ 65.66–68.71 Early MK §§ Sherds damaged by water. ¶¶ Sherds abraded in lower part of this sequence. 39 PC060 2018 665 73.64–74.24 RP 72.49–73.64 SIP–NK (late Dynasty 17/early Dynasty 18) 67.44–72.49 No dating possible 65.64–67.44 SIP–NK (late Dynasty 17/early Dynasty 18) PC061 2018 1670 73.25–73.85 NK 71.65–73.25 SIP 70.35–70.75 SIP 68.05–70.25 MK–SIP 66.35–68.05 early MK? 64.75–66.35 early MK? (fewer sherds) 64.00–64.75 No ceramics PC062 2018 B.A. 73.98–74.68 LP–PP 73.68–73.98 MK–NK? 70.18–73.68 Downhole contamination 64.68–70.18 No ceramics PC063 2018 25 72.58–73.74 RP?/Mixed? 69.48–72.58 Very few mm-scale sherds, thought to be contamination 68.28–69.48 No ceramics PC071 2019 B.A. 75.93–77.06 TIP–PP 74.56–75.93 No dating possible 72.86–74.56 NK–TIP 72.36–72.86 NK 69.06–72.36 SIP–NK 68.06–69.06 Few small sherds; no dating possible PC072 2019 B.A. 76.93–77.63 Mixed? TIP–PP? 73.83–76.93 Mixed? Modern? 69.87–73.83 NK? Mixed? Modern? 68.63–69.87 MK?/NK? PC073 2019 3454 77.20–77.50 No dating possible 75.60–77.20 MK–NK 74.90–75.60 MK–NK (Dynasty 17?) 73.90–74.90 MK–NK (Dynasty 13?) 73.10–73.90 MK–NK (Dynasty 12?) 72.10–73.10 MK–SIP 71.20–72.10 Early MK## 70.70–72.10 No dating possible 70.50–70.70 No ceramics ## Sherds damaged by water. 40 References ARNOLD, D. & J. BOURRIAU. 1993. An Introduction to Ancient Egyptian Pottery. Mainz am Rhein: Deutsches Archäologisches Institut Abteilung Kairoa. BORAIK, M., L. GABOLDE & A. GRAHAM. 2017. Karnak’s quaysides: evolution of the embankments from the eighteenth dynasty to the Graeco-Roman period, in H. Willems & J.-M. Dahms (ed.) The Nile: Natural and cultural landscape in Egypt.: 97–144. Bielefeld: Transcript Verlag. BOURRIAU, J., P.T. NICHOLSON & P. ROSE. 2000. Pottery, in P.T. Nicholson & I. Shaw (ed.) Ancient Egyptian Materials and Technology: 121–47. Cambridge: Cambridge University Press. BRIDGE, J.S. 2003. Rivers and Floodplains: Forms, Processes, and Sedimentary Record. Oxford: Blackwell Science. BROWN, A.G. 1997. Alluvial Geoarchaeology: Floodplain archaeology and environmental change. Cambridge: Cambridge University Press. BUNBURY, J.M., A. GRAHAM & M.A. HUNTER. 2008. Stratigraphic landscape analysis: charting the Holocene movements of the Nile at Karnak through ancient Egyptian time. Geoarchaeology 23: 351–73. CHARLOUX, G., R. ANGEVIN, S. MARCHAND, H. MONCHOT, J. ROBERSON & H. VIRENQUE. 2012. Le parvis du temple d’Opet à Karnak: exploration archéologique, 2006-2007. Cairo: Institut Français d’Archéologie Orientale. CHARLOUX, G., M.A. ABADY MAHMOUD, A.M.S. ELNASSEH & S. MARCHAND. 2021. The shifting Nile and the origins and development of ancient Karnak. Antiquity 95: 919–39. DAVID, R., Z. BARAHONA MENDIETA, R. DAVID, G. LEÇUYOT, A. MARANGOU, S. MARCHAND, A. MASSON-BERGHOFF, G. SCHREIBER & A. SIMONY. 2016. Theban ceramics in Hellenistic context - fabrics classification, in R. David (ed.) Céramiques ptolémaïques de la région thébaine, Cahiers de la Céramique Égyptienne 10: 11–22. Cairo: Institut Français d’Archéologie Orientale. DEE, M., D. WENGROW, A. SHORTLAND, A. STEVENSON, F. BROCK, L.G. FLINK & C.B. RAMSEY. 2013. An absolute chronology for early Egypt using radiocarbon dating and Bayesian statistical modelling. Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences 469: 1–10. DITZLER, C., K. SCHEFFE & H.C. MONGER (ed.). 2017. Soil survey manual (United States Department of Agriculture Handbook 18). Washington, D.C.: Government Printing Office. FOURIER, B. 1812. Description de l’Égypte: Antiquités, planches, vol III. Paris: Imprimerie Impériale. FRENCH, C.A.I. 1991. An analysis of the sediment at East Karnak. Journal of the Society for the Study of Egyptian Antiquities 11: 263–78. GARZANTI, E., S. ANDÒ, M. PADOAN, G. VEZZOLI & A. EL-KAMMAR. 2015. The modern Nile sediment system: processes and products. Quaternary Science Reviews 130: 9–56. GHILARDI, M. & M. BORAIK. 2011. Reconstructing the Holocene depositional environments in the western part of ancient Karnak temples complex 41 (Egypt): a geoarchaeological approach. Journal of Archaeological Science 38: 3204–16. GRAHAM, A., K.D. STRUTT, M.A. HUNTER, S. JONES, A. MASSON, M. MILLET & B.T. PENNINGTON. 2012. Theban harbours and waterscapes survey, 2012. Journal of Egyptian Archaeology 98: 27–42. HASSAN, F.A., M.A. HAMDAN, R.J. FLOWER, N.A. SHALLALY & E. EBRAHEM. 2017. Holocene alluvial history and archaeological significance of the Nile floodplain in the Saqqara-Memphis region, Egypt. Quaternary Science Reviews 176: 51–70. HORNUNG, E., R. KRAUSS & D.A. WARBURTON. 2006. Ancient Egyptian Chronology. Leiden: Brill. KLAPPA, C.F. 1980. Rhizoliths in terrestrial carbonates: classification, recognition, genesis and significance. Sedimentology 27: 613–29. LAUFFRAY, J. 1968. Nouvelles découvertes aux temples de Karnak. Comptesrendus des séances de l’Académie des Inscriptions et Belles-Lettres 112: 337–51. —. 1969. Rapport sur les travaux de Karnak. Activités du Centre francoégyptien en 1967-1968. Cahiers de Karnak 2: 111–35. LLOYD, A.B. 2010. A Companion to Ancient Egypt. Chichester: WileyBlackwell. MASSON, A. 2007. Le quartier des prêtres du temple de Karnak: Rapport préliminaire de la fouille de la Maison VII. Cahiers de Karnak 12: 593– 655. —. 2009. Vivre à la porte du sacré, les maisons des prêtres dans le sanctuaire d’Amon. Les dossiers d’archéologie 16: 48–55. —. 2011. Persian and Ptolemaic ceramics from Karnak: change and continuity. Cahier de la céramique égyptienne 9: 269–310. —. 2012. Domestic and cultic vessels from the quarter of priests in Karnak: the fine line between the profane and the sacred, in B. Bader & M. Ownby (ed.) Functional Aspects of Egyptian Ceramics within their Archaeological Context: Proceedings of a Conference held at the McDonald Institute for Archaeological Research, Cambridge, July 24th– July 25th, 2009. Leuven: Orientalia Lovaniensia Analecta. —. 2015. Toward a new interpretation of the fire at North-Karnak? A study of the ceramic from the building NKF35. Cahiers de Karnak 15: 189–213. —. 2016. Ptolemaic ceramics of the Theban region between tradition, imitation and innovation: the Priests’ Quarter as a study case, in R. David (ed.) Céramiques ptolémaïques de la région thébaine, Cahiers de la Céramique Égyptienne 10: 149–64. Cairo: Institut Français d’Archéologie Orientale. MASSON-BERGHOFF, A. 2021. Le quartier des prêtres dans le temple d’Amon à Karnak. Leuven: Peeters. MIALL, A.D. 1996. The Geology of Fluvial Deposits. New York: SpringerVerlag. MILLET, M. 2007. Architecture civile antérieure au Nouvel Empire: rapport préliminaire des fouilles archéologiques à l’est du lac Sacré 2001–2003. Cahiers de Karnak 12: 681–743. —. 2008. Installations antérieures au Nouvel Empire au sud-est du Lac sacré du temple d’Amon de Karnak. PhD Thesis; École doctorale Mondes anciens et médiévaux, Sorbonne Université.