1With a length of ~6500 km, the Nile is the world’s longest river, extending across northeast Africa through 35° of latitude to the Mediterranean coast of Egypt (Woodward et al., 2007; fig. 1). Charting the lateral migration of the Nile River in Upper Egypt (fig. 1 and fig. 2) in relation to human occupation is a recent topic and few studies adopting a geoarchaeological approach have been undertaken in the ancient Theban area. In Egypt, the deltaic area has been widely explored for geoarchaeological purposes (Goiran, 2001; Stanley et al., 2003; Tristant, en préparation). In contrast, there is a lack of studies combining landscape evolution and human occupation in Upper Egypt, more exactly in the Theban valley where important archaeological sites include Coptos, Dendera, Karnak, Luxor, Medamud and Tod temples (fig. 2), and the Valley of the Kings. At Karnak, the world’s largest temples complex built from ca. 2000 BC to Roman times, recent work combining a study of the archaeological material and the interpretation of digital data highlighted the important lateral mobility of the Nile River during the Dynastic period (ca. 3000 BC) until AD 300-400 (Hillier et al., 2007). However, these studies did not use chronostratigraphy to precisely reconstruct the palaeogeographic evolution of the Nile River around Ancient Karnak during the Dynastic period. The aims of the present study are to reconstruct the evolution of the Nile River and its ephemeral tributaries (wadis) and to closely examine the recent Holocene sediments deposited by this river in the vicinity of Karnak and Coptos. Using chronostratigraphy, our work aims to better understand the period when the Nile River flowed in front of the Karnak temples.
Fig. 1 – Location map of the Nile catchment and of the study area.
Fig. 1 – Carte de localisation du basin-versant du Nil et du secteur d’études.
After Woodward et al., 2007.
D'après Woodward et al., 2007.
Fig. 2 – Geological map of the Theban plateau area, Upper Egypt.
Fig. 2 – Carte géologique du plateau thébain, Haute Égypte.
The black circles represent the modern cities and the red squares indicate the location of pharaonic sites mentioned within this paper. 1: Nile alluvial sediments (recent Holocene); 2: Neonile deposits (Holocene); 3: Prenile deposits (Pleistocene); 4 : wadi deposits; 5 : fanglomerates; 6 : sandstones, siltstones and claystones (Pliocene); 7: thinly bedded outer shelf chalk and chalky limestone (Eocene); 8: greenish-grey open marine shale (Palaeocene to Eocene); 9: dark grey marine shale oscillating with calcareous intercalations (Cretaceous-Maastrichtian); 10: glauconitic sandstone and grey shale, oyster beds dominating around the Nile valley (Cretaceous-Campanian); 11: shale, siltstone and flaggy sandstone, containing freshwater gastropods (Cretaceous-Campanian); 12: main fault line; 13: hidden fault line (data source: The Egyptian General Company Petroleum, 1987, map of Qena, scale 1:500000).
Les cercles noirs indiquent la position des villes modernes et les carrés rouges révèlent l’emplacement des sites pharaoniques mentionnés dans cet article. 1 : alluvions du Nil (Holocène récent) ; 2 : dépôts du Néo-Nil (Holocène) ; 3 : dépôts du Pré-Nil (Pléistocène) ; 4 : dépôts de wadi ; 5 : cônes de wadi indurés ; 6 : grès (Pliocène) ; 7 : marnes finement litées et calcaires marneux (Eocène) ; 8 : schistes argileux gris-vert d’origine marine (Paléocène à Eocène) ; 9 : schistes argileux gris foncé d’origine marine à intercalations calcaires (Crétacé-Maastrichtien) ; 10 : grès à glauconite et schistes argileux gris, à lits d’huitres abondant dans la vallée du Nil (Crétacé-Campanien) ; 11 : schistes argileux et grès contenant des gastéropodes d’eau douce (Crétacé-Campanien) ; 12 : principale ligne de faille ; 13 : ligne de faille masquée (données : The Egyptian General Company Petroleum, 1987, carte de Qena, échelle 1/500000).
2The Modern Nile catchment has a total area of ~2.9 Mkm2, covering ten countries, but almost all of this area is located in the upper reaches of the river. The source of the principal Nile affluent, the White Nile, is in the Great Lakes region of equatorial Africa. This is joined by two other tributaries, the Blue Nile at Khartoum in central Sudan and the Atbara at Atbara in northern Sudan, both of which drain the Ethiopian highlands (Zaki, 2007). Further north, in Egypt, where the Nile River flows through the Sahara Desert, it has no significant tributaries. In this region, tributary wadis typically reach no further than a few 10 km from the main Nile channel (Zaki, 2007). Some of these intermittent streams join the Nile in the Theban valley. For instance, wadis Hammamat and El Medamud have formed large detritic fans on the right bank of the Nile River near Qena and of Luxor (fig. 2). These Pleistocene fans are partly buried under the Holocene Nile deposits and most of the time only the apex is visible. From a geological point of view, Middle and Upper Egypt consist of an extended sedimentary plateau of Eocene age, characterised by low relief topography that dips towards the west (Said, 1990; Badawy et al., 2006) and the predominance of carbonated rocks such as Eocene chalky limestones and Pliocene sandstones/siltstones (fig. 2). The plateau reaches 300 m asl with a sharp scarp facing the Nile valley on its eastern side (Badawy et al., 2006). Carved into the African plateau ca. 5-8 Ma ago and then mostly re-filled with sediment, the ~10 km wide Nile valley is cliff-bounded and flat-bottomed in most of Upper Egypt (Butzer, 1980; Hillier et al., 2007). R. Said (1981) reveals that the Nile valley lies along a seismically active belt. The evolution of the modern drainage network and its fluvial geomorphology reflect both long-term tectonic and volcanic processes and associated changes in erosion and sedimentation, in addition to sea-level changes (Said, 1980, 1981). One of the geological features of the drainage basin of the Nile River is the predominance of volcanic rocks containing an important proportion of magnetic minerals in its upper part (plateau of Ethiopia) and in contrast Upper Egypt is characterised by a local drainage (wadi streams) on carbonated rocks (low magnetic minerals content).
3The Thebes area was an important religious and economic centre during part of the Dynastic period and several places such as Karnak, Luxor, Dendera, Medamud, Tod and Coptos played a key role in the emergence of a powerful state from ca. 2000 BC until the end of Roman times.
4Construction of the Karnak temples complex started ca. 2000 BC (during the 12th Dynasty) and was completed during Roman times (Legrain, 1903; Barguet, 2006). It comprises a main temple dedicated to Amun Ra and a vast conglomeration of divine temples dedicated to different gods from Egyptian mythology, chapels, pylons, and other buildings. During the dynastic period, the spatial distribution follows a E-W direction. Another significant edifice is the quay built in front of the first pylon, at the western margin of the temple complex. Recent excavations by the Egyptian Supreme Council of Antiquities (under the supervision of Dr. Mansour Boraik) revealed the presence of this large 1000-m long (the exact length is still unknown because excavations are ongoing around Luxor temple, situated 3 km to the south) and approximately 6 m to 8 m thick structure. Based on Egyptological and archaeological studies, the quay is hypothesized to date from the first quarter of the 10th c. BC, during the 22nd Dynasty (Lauffray, 1975). An important archaeological find is the ramp of the Pharaoh Taharqa (25th Dynasty, 690-664 BC) superimposed on the quay and potentially dating from the 25th Dynasty (fig. 3 B and C). The use and the environmental contexts of the construction of these large buildings is still unclear: was the quay constructed to protect the Karnak temples complex from violent flood events or was the Taharqa ramp intended to facilitate direct access to the river? The presence of the Nile River in front of the first pylon has never been proven and our paper investigates its position in the western part of the Karnak temples.
5The archaeological site of Coptos, today excavated under the direction of Prof. Laure Pantalacci on behalf of Lyon University/IFAO, is situated 35 km north of the Karnak temples. Occupation is attested during Predynastic and Dynastic Times: it was a major centre for pottery making during Roman times and was the starting point of the road to the Red Sea through the eastern desert (Collectif, 2000). The presence of temples dedicated to the gods Min and Isis built during early Dynastic Times highlights the importance of the site in the Theban valley (Petrie, 1896; Reinach, 1910; Weil, 1911; Traunecker, 1992; Collectif, 2000). It is obvious that its position along the Nile River and on the road to the Red Sea helps to explain the long-term existence of Coptos. However, there is a paucity of palaeoenvironmental data pertaining to the fluvial landscape history.
Fig. 3 – Map and sketch of the most western part of the Karnak archaeological site with sampling places and a picture of the quay.
Fig. 3 – Carte et plan de la partie occidentale du site archéologique de Karnak avec les différents sites de prélèvement et une photographie du quai.
A: The most western part of the site, based on a Digital globe image (resolution is 6.25 m) and integrated in a GIS. Geographic coordinates are expressed in WGS 84 geoid model and the red square indicates fig. 3B. B: Sketch of the most western part of the Karnak temples complex with the location of the stratigraphic sections (SP1, SP2 and SP3 and the deep well (DW). C: Picture of the quay located in front of the first Pylon. The Taharqa ramp is superimposed on the quay. 1: Roman well; 2: Chapel of Akoris (394-380 BC, 29th Dynasty).
A : Image Digital Globe (résolution de 6,25 m) de la partie occidentale du complexe cultuel de Karnak. Le système de coordonnées géographiques est exprimé dans le WGS 84, et le carré rouge localise la fig. 3B. B : Plan de la partie occidentale du complexe des temples de Karnak indiquant l’emplacement des coupes stratigraphiques (SP1, SP2 et SP3) et du forage (DW). C : Photographie du quai situé au pied du premier Pylône. La rampe de Taharqa est construite sur le quai. 1 : puits romain ; 2 : chapelle d’Akoris (394-380 av. J.-C., XXIXe Dynastie).
6Three stratigraphic profiles (SP1, SP2 and SP3; fig. 3), situated along the quay and in front of the Taharqa ramp, were sampled for sediment grain-size analysis, magnetic susceptibility measurements and radiocarbon dating. In addition, a 25-m well (DW; fig. 3 B and C) in front of the Taharqa ramp, was also studied. All sediments were analysed at the American Research Centre in Egypt laboratory (ARCE, Karnak). Finally, a series of auger drills were carried out using an Eijkelkamp hand-auger in Coptos (5 boreholes until 5.5-m deep; fig. 4). The sampling was carried out in Cairo in the laboratory of IFAO with permission of the SCA. In addition, Digital Globe and satellite images with Shuttle Radar Topography Mission (SRTM) data were integrated into a Geographic Information System (GIS) and helped to interpret the palaeoenvironmental results.
Fig. 4 – Location map of the Coptos site and location of the hand augering.
Fig. 4 – Carte de localisation du site de Coptos et des prélèvements à la tarière manuelle.
The black star indicates the position of the main archaeological structures.
L’étoile noire indique l’emplacement des principales structures archéologiques.
7Grain-size analyses of 300 bulk samples were performed using standard sieving and sedimentation techniques (Folk, 1974). The sediments collected were hand sieved with different mesh sizes from 20 µm to 2000 µm. Ultrasound was employed for the finest particles. The mean and modal grain sizes, including standard deviation (σ) and skewness (SKI), were calculated according to R.L. Folk and W.C. Ward (1957). These parametres were plotted against each other to define depositional environments. The medium sand fractions were separated and examined using a Leica DM2500P polarising microscope (PM) at the ARCE laboratory in Karnak. Surface features of the quartz sand grains provide information about transportation processes and depositional history (Krinsley and Doornkamp, 1973; Whalley and Krinsley, 1974; Higgs, 1979; Williams and Morgan, 1993).
8They were performed using a Bartington Instruments MS2B meter with a resolution of 10-5 SI units at the American Research Centre in Egypt laboratory. The sediment cores were sampled at a ~5-cm interval, except at levels including reworked material, yielding 300 samples in total. These samples were placed in 10 cm3 plastic boxes, dried and weighed. In addition to the low field magnetic susceptibility, usually measured at the 465 Hz frequency, measurements were also taken at the 4650 Hz frequency. The magnetic susceptibility values were divided by the density of the dried samples in order to derive specific susceptibilities (χ). Magnetic susceptibility is used as an indicator of the concentration of magnetic particles and can help to identify different sediment source areas in fluvial deposits (Ghilardi et al., 2008) characterised by magnetic minerals concentration. For our study, there is a possibility that Nile sediments deposited at Karnak and Coptos show a different signal from the local ephemeral streams draining the Theban plateau. The size of the ferromagnetic particles can also influence magnetic susceptibility values. Magnetic susceptibility measurements performed at two frequencies are used to detect the ultrafine (< 0.03 µm) superparamagnetic particles, which are produced by bacteria or chemical processes during soil formation (Dearing et al., 1996). The contribution of fine-grained viscous/superparamagnetic particles is given by the frequency dependant susceptibility (χfd).
9For the Karnak temples complex, radiocarbon dating was performed at the IFAO laboratory (with permission from the SCA) on material collected from the stratigraphic profiles (SP2 and SP3) and the deep well (fig. 5 and fig. 6). Tab. 1 summarises the information concerning the type of sample and the exact elevation, ranging from 56.09 to 72.136 m asl. 14C ages were subsequently calibrated using OxCal version 4.1 (Bronk Ramsey, 1995, 2001; Reimer et al., 2009).
10A multispectral Digital Globe image dated from 07/06/2006 with a resolution of 6.25 m was acquired for the Karnak-Luxor area. The document was integrated into a GIS and georeferenced in relation to the WGS 84 geodetic datum. In order to have a complete view of the Theban valley, a LANDSAT ETM+ image dated from 2000 was also included in the GIS (resolution 30 m). In addition, high-resolution topographic data derived from SRTM surveys were added to the GIS and superimposed on the satellite imagery in order to obtain altimetric information of the different landforms identified on the Digital Globe and LANDSAT ETM+ False Colour Composition (FCC). Spectral bands 2, 3 and 5 were combined in order to reveal the contrast between wet areas (swamps, irrigated zones, etc.) and rocky areas (desert, wadi fans, valleys, etc). Previous work has highlighted the accuracy of this method (Ghilardi et al., 2008; Ghilardi and Desruelles, 2009) and in the Eastern Sahara the use of radar images applied to archaeology has already yielded major results (Robinson et al., 2006).
Fig. 5 – Stratigraphic sections SP1, SP2 and SP3 studied in front of the Taharqa ramp.
Fig. 5 – Profils stratigraphiques SP1, SP2 et SP3 étudiés en face de la rampe de Taharqa.
See fig. 3 for exact location; geographic coordinates are expressed in a local projection datum. 1: pottery sherds mixed with mud bricks; 2: fragments of sandstones coming from quarrying activity and quay construction; 3: charcoal, ash layer; 4: Nile silts with local clay intercalations; 5: fine sands; 6: medium to coarse sands locally bedded; 7: coarse grey sands, thinly bedded; 8: well sorted sands, bedded and alternating with silt deposits (beds of 0.5 cm approx.); 9: sample (100 g) carried out for sedimentological analyses (dry sieving method).
Voir fig. 3 pour localisation précise ; les coordonnées géographiques sont exprimées dans un système local de projection. 1 : tessons de poteries mélangés avec des briques de terre crue ; 2 : blocs et fragments de grès provenant de la rénovation/destruction du quai (période ptolémaïque ?) ; 3 : charbons, niveau de cendres ; 4 : limons du Nil avec intercalations épisodiques d’argiles ; 5 : sables fins ; 6 : sables moyens à grossiers localement lités ; 7 : sables grossiers gris finement lités ; 8 : sables bien triés, lités et alternant avec des dépôts limoneux (lits d’environ 0,5 cm d’épaisseur) ; 9 : échantillon (100 g) prélevé pour analyses sédimentologiques (granulométrie par tamisage à sec).
Fig. 6 – Core profile of the Deep Well (DW) drilled in Karnak.
Fig. 6 – Log stratigraphique du Deep Well (DW) réalisé à Karnak.
Elevations are expressed asl. Grain-size analyses have been realised using sieving method. A: Picture of sands, from Unit B, realised under polarizing microscope (ARCE laboratory). B: Picture of wood fragment (Acacia sp.) worked by human.
Les altitudes sont exprimées par rapport au niveau moyen de la mer. Les analyses granulométriques ont été réalisées au moyen de la méthode de tamisage à sec. A : Photographie, réalisée à l’aide du microscope polarisant (laboratoire de l’ARCE), de sables provenant de l’unité B. B : Photographie d’un morceau de bois (Acacia sp.) façonné par l’Homme.
Tab. 1 – Radiocarbon dating results (Karnak site).
Tab. 1 – Résultats des datations par le radiocarbone (site de Karnak).
11The late deposition of the Nile sediments during the Roman times along the quay, in front of the Taharqa ramp, is attested by the chronostratigraphy derived from the stratigraphic profiles SP2 and SP3 (fig. 5). SP3 is characterised by a continuous accumulation of fine to medium sands (the mean size is comprised between 105 µm and 160 µm) showing magnetic susceptibility signals between 350 x 10-8 m3/kg to 650 x 10-8 m3/kg (where samples 16 and 17 show the highest values, respectively 450 x 10-8 m3/kg and 650 x 10-8 m3/kg). This sediment sequence is ca. 1-m thick and no archaeological artifact was found, a radiocarbon dating performed on a charcoal layer (tab. 1 and fig. 5) yielded an age of cal. AD 126-262 (1806±74 BP) and can be attributed to the Early Roman period. Another dating performed on an ash layer derived from SP2 (tab. 1 and fig. 5) indicates an age of cal. AD 208-415 (1736±110 BP) consistent with Late Roman times. The ash layer identified at 72.136 m asl is located above the last Nile River deposits; mainly thin clay layers intercalated with silt overlying the sandy sediments identified in SP3 (see above), the sequence is approximately 0.5 m thick. SP1 records similar sediments as SP2: the lower part is composed of medium grey sands showing high magnetic signals of between 170 x 10-8 m3/kg to 390 x 10-8 m3/kg (samples 8 to 12; fig. 5) and the mean grain size reveals values ranging from 130 µm to 160 µm. The lack of pottery sherds and human artefacts, allied with very well-sorted sands, attest to a high-energy fluvial environment. Gradually, due to a decrease in the river’s transport capacity, silts and clays are deposited with intercalations of thin sandstone lenses. In the upper part of SP1, human occupation is attested by the presence of mud bricks, probably belonging to houses dating from late Roman times, which overlay the last fluvial deposits from the Nile River. The study of the three stratigraphic profiles SP1, SP2 and SP3 allowed us to identify the transition between the fluvial activity of the Nile River and the human occupation during Roman times. The radiocarbon dating indicates a phase of important sediment aggradation during Roman times where peaks of high magnetic susceptibility have been recorded. The necessity to obtain further data concerning the relationship between Nile flow and the quay at Karnak forced us to go deeper.
12The deep well (DW) reveals the longest sedimentary sequence ever observed in Upper Egypt (fig. 6). It records different depositional environments and can be described as follows:
13- From ~ +48 m to ~ +50 m asl, Unit A is composed of angular and well-rounded pebbles, with some coarse sands and gravels. The stones show different sizes and some of them are 20 cm long with sharp edges that indicate short-term fluvial transport. In contrast, flat pebbles of 5-10 cm in diameter are found. An autochthonous origin must be envisaged and the local wadi streams are the only possibility to explain this deposition. No human artifacts and organic material were identified, suggesting that the deposit is older than Holocene, probably Pleistocene.
14- From ~ +50 m up to ~ +52 m asl, Unit B is a mixture composed of coarse yellow to white sands and well rounded pebbles (mainly limestone fragments). Magnetic susceptibility signals of the coarse sands show low values, comprised between 21 x 10-8 m3/kg and 38 x 10-8 m3/kg (χfd) is less than 1% and confirms the detrital origin of the signal. Grain size distribution reveals that the mean size is between 400 µm to 500 µm and shows a multimodal distribution, consistent with a poorly sorted sample. A wadi stream transport together with an aeolian origin is suggested by surface features of quartz grains, observed under a polarising microscope (fig. 6A).
15- From ~ +52 m up to ~ +55 m asl, Unit C comprises yellow to light grey medium to coarse sands. The magnetic susceptibility measurements reveal increasing values from the lower part (38 x 10-8 m3/kg) to the upper part of the sequence (130 x 10-8 m3/kg) with important values oscillating ca. 80 x 10-8 m3/kg (χfd) is less than 1% and indicates a low contribution of the superparamagnetic minerals. The grain-size distribution indicates well-sorted sands with a unimodal distribution at ca. 300-350 µm. The depositional environment is consistent with a high-energy fluvial system where local wadi sediments (coarse material identified from ~ +50 m up to ~ +52 m asl) were reworked and transported away by an important flooding activity, probably from the Nile River. Indeed, the increasing signal of (χ) can be explained by a higher contribution of magnetic minerals from the Nile catchment (e.g., the volcanic plateau of Ethiopia). Recent research on the Nile delta (Tristant, 2004; Tristant and De Dapper, 2009; Tristant et al., 2011; Tristant, in press) shows similar deposition where Predynastic settlements have been built upon. Unit C can be interpreted as a sandy levee (i.e., “gezira”) composed of aeolian sands and affected by episodic reworking during Nile floods.
16- Unit D is composed of dark grey medium sands and is found from ~ +55 m asl up to the surface (~ +73 m asl). Human artifacts are attested including intact pottery, pottery sherds, red bricks, etc. At ~ +56 m asl, medium grey sands (mean size between 175 µm and 275 µm) showing high magnetic susceptibility values (χ is comprised between 160 x 10-8 m3/kg and 260 x 10-8 m3/kg) are mixed together with pottery and wood fragments (Acacia sp.) of large dimensions (fig. 6B). This unit is composed of well-sorted fluvial sediments (ranging from clay to medium sands) deposited by the Nile River. The high magnetic susceptibility signals (up to 470 x 10-8 m3/kg) attest to detrital deposition since the χfd is less than 2%. The geology of the drainage basin of the Nile River is characterised by volcanic rocks in its most upper part (Woodward et al., 2007) and their erosion, under tropical to equatorial climate controls. Radiocarbon dating performed on one of these wood fragments reveals an age of 1494-1402 cal. BC (3158±47 BP). This artifact can be attributed to the early New Kingdom/18th Dynasty (1550-1292 BC), under the reign of Thutmose III (1479-1424 BC). Unfortunately, it was not possible to identify the function of this piece of wood. The fluvial sequence of the Nile River sediments (thickness is ~18 m) and its recent age of deposition (~3500 BP) shows a higher magnetic susceptibility signal and a finer grain-size distribution (from 10 µm to 180 µm) compared to the gezira and the wadi deposits described above.
17- The upper part of the sequence, close to the surface (+73 m asl) is affected by intense human activities and mud bricks are clearly identified by the recent archaeological excavations. The date of these remains, located between the tribune and Taharqa ramp (fig. 3) can be reasonably placed at the end of the Roman times, based on the study of material (presence of pottery found in situ and walls built with mud bricks found in the upper part of SP1; fig. 5 and fig. 6).
18To summarize, we can assume that Nile River sediments (~18-m thick) overlie gezira and wadi deposits. The period of accumulation for the fluvial material between the ramp of Taharqa and the tribune is ~3500 BP and can be placed from the early times of the 18th Dynasty (mid-15th c. BC) to the end of the Roman period (ca. 4th-5th c. AD). Sediments of the Nile River are generally composed of alternating beds of fine to medium grey/brown sands and silts, showing high magnetic susceptibility values. In contrast, wadi fan sediments reveal the presence of heterogeneous material, ranging from coarse yellow sands (with a low magnetic susceptibility signal) to gravels and pebbles. Gezira formations are composed of a mixture of aeolian and flood sands from the Nile River where the magnetic susceptibility values are higher compared to the wadi fan sediments.
19Based on the material study of the hand augers from Coptos (fig. 7), the results can be described as follows:
20- The first unit is only identified in cores 0906 and 0907 (situated ~50 m to the NNE of Min and Isis temples) and is composed of a mixture of coarse yellow sands and small gravels showing a very low magnetic susceptibility values (from 30 x 0-8 m3/kg to 35 x 10-8 m3/kg). No archaeological artefacts were found. The thickness of this layer is ~0.3 m although we were unable to estimate the total thickness because the corer would not go any deeper. The quartz microscopy shows features typical of aeolian transport (fig. 7A) and fits well with the surface features observed in the upper part of the wadi sequence derived from the Karnak deep well (unit B of DW; fig. 6A): well-rounded material with shock crescents are mainly observed. This unit was deposited in a wadi context with important aeolian dynamics and occasional water body transport, where local accumulation of sands created levees.
21- The second unit is found in cores 0906, 0907 and 0908 and is composed of a mixture of medium yellow sands and of fine grey sands. For cores 0906 and 0907, this layer overlies the aforementioned wadi sediments and for core 0908, it is situated above a layer composed of very fine and dark deposits (high organic matter content). It is 1-1.5-m thick and is composed of well-sorted brown to yellow sands. The mean size is from 120 µm to 200 µm and shows unimodal distribution and the magnetic susceptibility signals are comprised between 50 x 10-8 m3/kg and 130 x 10-8 m3/kg with a significant population around 110-120 x 10-8 m3/kg. The increase of (χ) is attributed to a detrital origin because the value is changing as a function of the grain size distribution. This observation is well attested in fluvial contexts and shows similar features with an increase of the magnetic susceptibility signal related to an increase in particle size (Ghilardi et al., 2008). This second unit shows similar features with Unit C derived from the Karnak Deep Well (DW) and can be interpreted as a gezira environment where Nile River sediments rework aeolian/wadi deposits from local sources with the far-field fractions originating from the Nile catchment.
22- The third unit is composed of alternating beds of homogeneous sandy layers (0.3-m thick) and of clay/silts strata. The important magnetic susceptibility values and the general well-sorted fine material identified indicate a similar environment to Unit D from Karnak DW: Nile sediment accumulation. The relatively low thickness of this unit seems to reveal that the main course of the river is situated at a certain distance from the Coptos archaeological site. The absence of organic material for radiocarbon dating and presence of only small pottery sherds makes it impossible to precisely date this fluvial layer.
23- The last unit is 1-m thick and composed of fine grey silts. It corresponds to the modern soil, strongly affected by human activities (agriculture, house building, etc.) during the 20th c. AD.
Fig. 7 – Cross section including the hand augering 0902, 0901, 0906, 0907, 0908 for Coptos site.
Fig. 7 – Section stratigraphique incluant les prélèvements à la tarière manuelle 0902, 0901, 0906, 0907, 0908 réalisés sur le site de Coptos.
A: Picture of the gezira sands taken with polarizing microscope (IFAO laboratory). Geographic coordinates are expressed in WGS 84 geoid model, elevation are expressed above mean sea level. 1: Modern soils disturbed by archaeological excavations; 2: Nile silts and clays; 3: Nile floods; 4: gezira deposits; 5: gezira deposits mixed with wadi sediments; 6: rich organic Nile sediments (very high χ signals).
A : Photographie des sables de gezira réalisée à l’aide du microscope polarisant (laboratoire de l’IFAO). Les coordonnées géographiques sont exprimées dans le système de projection WGS 84, les altitudes sont exprimées par rapport au niveau moyen de la mer. 1 : sols récents remaniés lors des travaux archéologiques ; 2 : argiles et limons du Nil ; 3 : dépôts de crue du Nil ; 4 : dépôts de gezira ; 5 : dépôts de gezira mélangés à des sédiments de wadi ; 6 : sédiments du Nil riches en matière organique (fortes valeurs pour χ).
SRTM data reveals the local surface topography around Ancient Coptos. Two profiles AB and A’B were established from the apex of the wadi Hammamat fan (fig. 8) to Coptos (AB) and the middle part of the valley (A’B). The interface between the wadi fan and the Nile valley is ca. 79 m asl, gradually dipping towards Coptos. The 'valleys' with surface topography values situated ca. 71 m correspond to a Modern irrigation channel. We can clearly observe that the site of Coptos is partly located on a mound ca. 73-74-m high. The satellite image interpretation helps to accurately measure the width of the Nile River valley in the vicinity of Coptos (9.21 km) and Karnak (ca. 8.24 km). Nevertheless, due to important land reclamation and irrigation works, it is almost impossible to identify former Nile River courses. In particular, it would be speculative to interpret modern oxbows or ancient channels as landforms created during the Dynastic period. Indeed, recent papers (Hillier et al., 2007) propose a palaeogeographic reconstruction of the lateral mobility of the Nile River along the recent Holocene based upon historical maps (dated from the 19th c.) and SRTM data. The difficulty is to precisely date the different landforms from spatial information, palaeoenvironmental data are necessary to confirm the age of the oxbows, etc.
Fig. 8 – LANDSAT ETM+ image.
Fig. 8 – Image satellite LANDSAT ETM+.
FCC (spectral bands 2, 3, 5) realised with Er Mapper software version 7.0. SRTM data were superimposed in order to obtain the elevation of the Upper Egypt.
Combinaison spectrale (bandes spectrales 2, 3, 5) réalisée à l’aide du logiciel Er Mapper version 7.0. Les données SRTM ont été superposées afin d’obtenir le cadre altimétrique de la Haute Égypte.
24Based on the palaeoenvironmental results described in this paper, it is possible to reconstruct the landscape evolution linked to human occupation in the vicinity of Karnak and Coptos. Before the Dynastic period (before 3000 BC), we demonstrate that the Nile River did not flow around the Pharaonic temple complex. Indeed, aeolian dynamics combined with intermittent hydrological activity (wadi streams) were dominant: within the Nile valley, large detrital fans were formed comprising coarse material from local sources. The age of these structures is uncertain and the lack of human artifacts suggests that we probably reached the pre- to early Holocene sub-surface. The Nile River was probably more deeply incised between the different fans than at present where an important accumulation of alluvial deposits covered a large part of theses formations during and after the Dynastic period (a period of ~3500 years). Gradually, the influence of the Nile floods increased throughout Upper Egypt and alluvial material was deposited on a larger width: wadi fans and aeolian sediments have been washed away by occasional major high flood events and mixed with alluvial material. Sand levees formed at the interface between the desert area and the riverbed channel. The development of small mounds, similar to the ones observed in the delta (Tristant, 2004; Tristant and De Dapper, 2009; Tristant et al., 2011; Tristant, in press) probably facilitated the installation of the first settlers from the Sahara Desert, when arid and warm conditions affected the whole area between the mid-6th and the 4th millennium BC (Lario et al., 1997; Midant-Reynes, 2003; Kuper and Kröpelin, 2006; Kröpelin et al., 2008). Coptos archaeological site attests to a possible occupation before Predynastic period, although more archaeological data are needed to confirm this assumption. The sandy levees, formed by a mixture of Nile sediments and aeolian deposits, are called gezira in Arabic and have been archaeologically (Tristant, in press) and geologically (Dufton and Branton, 2010) investigated in the delta area. However, little research has been undertaken in Upper Egypt to better understand the settlement history of the Pharaonic sites on these gezira, at the interface between wadi fans and the Nile River. Due to an intense phase of aridity since the mid-Holocene and an increase in deforestation in the Nile basin, landscapes changed drastically from the end of the Predynastic period (Woodward et al., 2007). The gezira gradually turned into protected islands during Nile floods and Pharaonic temples were then well preserved. During very high water levels, the monuments were partially inundated by fluvial waters. The Nile’s lateral migration is difficult to understand in the absence of chronostratigraphic sequences from the different Pharaonic sites mentioned in the sources. Our preliminary study of the Nile’s evolution during the recent Holocene has revealed a novel feature: from the end of the Middle Kingdom/beginning of the New Kingdom up to the end of the Roman period, the main course of the river was flowed past the quay, built at the beginning of the 1st millennium BC. Since the end of the Roman period, the main course shifted 450 m to the west with possible rapid changes of main direction. The construction of the Aswan Dam in the late 1960s affected the hydrological regime of the Nile River: this allowed a reduction of the floods and the development of irrigation throughout the valley.
25Based on the sedimentological results two important periods of high Nile flow are recorded. The first is dated to the beginning of the New Kingdom (18th Dynasty) and can be placed ~1450 BC and the second major event recorded in the sediments is from Late Roman times (~AD 150-300). Research dealing with Nile flood events in Upper Egypt are rare and most of the time are very general (Krom et al., 2002) and concern later periods (e.g., Medieval Times; Hassan, 1981, 2007). Moreover, the Theban area does not record any significant palaeoenvironmental data to understand the relationships between climate and hydrological dynamics. However, regional studies have highlighted that 4200 cal. BP reflects a progressive increase in sediment flux from the Ethiopian Highlands as Nile flows declined following a decrease in monsoon intensity and a reduction in vegetation cover in the Blue Nile headwaters (Woodward et al., 2007). The central part of the record (after 4200 cal. BP) shows fluctuating Nile flows around 1500 cal. BP when the record may be partly influenced by enhanced sediment yields from Ethiopia following deforestation (Woodward et al., 2007). D.J. Stanley et al. (2003) reveals that the New Kingdom is a period of abundant Nile floods and this fits well with our results based on the chronostratigraphy of the Karnak Deep Well. For Roman times, a comparative study with Mediterranean areas (Martin-Puertas et al., 2009) can be established for the flood-dominated contexts. Indeed, the Late Roman period (AD 150-350) is considered to be a period of humid climate conditions and of important aggradation throughout the Mediterranean; this is well recorded in Spain (Martin-Puertas et al., 2009) and in France on the Rhône River (Arnaud-Fassetta, 2000; Bruneton et al., 2001; Arnaud-Fassetta, 2002; Salvador et al., 2004).
26For the first time, the landscape evolution over the last millennia around the Karnak temples complex and the Coptos archaeological sites has been studied using a geoarchaeological approach. The combination of the archaeological data, provided by the different excavation programmes, and the palaeoenvironmental results allows us to reconstruct the environmental changes, including the position of the Nile River during the different phases of occupation of both sites. The first chronostratigraphic sequence, based on a series of three 14C datings performed on material sampled around Karnak’s ancient quay, helped to estimate the period of accumulation of Nile sediments. Approximately 18 m of fluvial material, composed of sediments ranging from clay to medium sands, accumulated in less than 2000 a, from ~1450 BC to ~AD 350, in the vicinity of the quay (between the tribune and the Taharqa ramp). An average sedimentation rate of 1 m/century during the Dynastic period (from the 18th Dynasty until Roman times) can be inferred. The first phase of construction of the Karnak temples, estimated ca. 2000 BC (under the reign of 12th Dynasty) coincides with a position at a certain distance from the Nile River: the site has been installed on a sandy levee system, called gezira. This formation is composed of a mixture of aeolian (originating from the wadi fans) and fluvial sands (transported by the Nile River) and creates a small mound at the interface between the Nile floodplain and the wadi fans (desert area). Both Coptos and Karnak show similar situations for the strategy of implantation of the temples during early Dynastic times. Similar observations have been made in the Nile delta to explain the location of Predynastic sites (Tristant and De Dapper, 2009; Tristant et al., 2011; Tristant, in press). The material from the gezira formations was reworked by wadi streams (aeolian and fluvio-torrential dynamics) which created large detrital fans within the present Nile valley. However, due to the important sediment accumulation of the Nile River, they have been gradually buried by fluvial deposits as shown by the SRTM data in the vicinity of Ancient Coptos. It is evident that the general sea-level rise observed in the Mediterranean for the last 20000 years (Pirazzoli, 1996) had an influence on the base level of the river profile of the Nile River. No studies have addressed this topic for the Upper Egypt and must be studied in order to characterize the width and the flow of the Nile River course during the early to mid-Holocene. Based on our research, the idea that the Karnak temple complex was built on a fluvial island must be abandoned. For the other Pharaonic temples such as Luxor, Medamud, etc., further palaeoenvironmental research must be conducted in order to better understand the position of the Nile River during the early stages of their occupation. The palaeoclimatic context of Upper Egypt is still unknown and the consequences of past environmental changes must be linked with the different Holocene phases of humidity/dryness that occurred in this rich archaeological area.
This article is part of the project Projet Exploratoire Premier Soutien (PEPS) titled 'Étude géoarchéologique des anciens ports nilotiques de Haute Égypte : Coptos, Karnak, Médamoud et Tôd'. It is funded by the Centre National de la Recherche Scientifique (CNRS, department of Human and Social Sciences, INSHS) and directed by Matthieu Ghilardi. The authors are grateful to the Egyptian Supreme Council of Antiquities (SCA) and its General Secretary Dr. Zahi Hawass for his contribution to significantly develop the geoarchaeological study of the Karnak Temples and for having authorized the transportation of the sediment/wood/charcoal samples from the archaeological sites of Karnak and of Coptos to Cairo (IFAO Radiocarbon dating laboratory). Warm thanks are given to Dr. Ibrahim Soliman (SCA) Director of the Karnak temples, and to the inspectors of the SCA Salah El Masekh, Mohamed Ali Hatem, Tayeb Guarib for fruitful discussions and precious help. Christophe Thiers, Director of the Centre Franco-Egyptien d’Étude des Temples de Karnak (CNRS, USR 3172) is acknowledged for the financial support provided and for his interest in the palaeoenvironmental reconstruction of the Nile River during 2009. His kindness and his hospitality were also well appreciated. Marie-Françoise Courel, former scientific Director of the Department for Human and Social Sciences from the CNRS is acknowledged for having promoted the geoarchaeological approach of the archaeological sites in Upper Egypt and in particular in Ancient Karnak. Professor Laure Pantalacci (former Director of the IFAO and Professor of Egyptology at the University of Lyon) is thanked for her kind invitation to work at Coptos during autumn 2007, 2008 and 2009. The radiocarbon dating laboratory of the IFAO and its team composed by Michel Wuttmann (Director) Mohammed Mahran and Nagui Sabri are warmly acknowledged for having dated the three samples from the Karnak site. The authors thank the American Research Centre in Egypt (branch of Luxor) and in particular its Director John Shearman, Elsa Bourguignon and Ed Johnson for use of the laboratory facilities (access to the polarizing microscope) and for providing the possibility to analyse the sediments derived from the deep wells drilled in front of the First Pylon of the Karnak temples (water table lowering project). The authors address their acknowledgements to Nick Marriner (CNRS, UMR 6635 CEREGE) for editing the manuscript and for improving the English. Sincere thanks to the Editor-in-Chief, Gilles Arnaud-Fassetta, as well as to two anonymous referees for remarks and helpful comments on the paper.