GeoSat ReSeArch IMS-FORTH

Transcription

1 PRELIMINARY REPORT ON GEOPHYSICAL SURVEY AT PHERAI APOSTOLOS SARRIS, CARMEN CUENCA-GARCÍA, JAMIESON C. DONATI, TUNA KALAYCI, MEROPI MANATAKI, FRANÇOIS-XAVIER SIMON Lab, Institute for Mediterranean Studies, Foundation for Research and Technology Hellas (IMS/FORTH) Abstract The Laboratory of Geophysical, Satellite Remote Sensing and Archaeoenvironment of the Institute for Mediterranean Studies (FORTH) conducted an integrated geophysical survey at the Greek settlement of Pherai in Thessaly during March The purpose of the fieldwork was to identify near-surface features of archaeological interest relevant to the organization of urban space and its use and development over time. Today the modern town of Velestino largely covers the ancient settlement and restricts the scope of archaeological exploration. Geophysical fieldwork focused primarily on open fields north and northwest of the town. It was successful in identifying an extensive network of streets, city-blocks, and what appear to be buried buildings. This new and valuable information provides insights into the wider urban dynamics of Pherai and shows that the city, at least in part, was likely a planned settlement at some point in its history. Introduction In March 2014, the Laboratory of Geophysical, Satellite Remote Sensing and Archaeoenvironment of the Institute for Mediterranean Studies (FORTH) explored the structure and urban development of the Greek city of Pherai (modern Velestino) in Thessaly through an intensive geophysical fieldwork campaign. The survey was carried out in areas bordering the northern and northeastern edges of the town (Figure 1). In classical antiquity, this zone comprised the northern limits of the Greek settlement. A large prehistoric mound, Magoula Bakalis, is the most prominent feature of the local landscape and acted as a small acropolis along with the neighboring hill of Agios Athanasios. Our chief target in carrying out the survey was to identify any subsurface features of archaeological interest that relate to the prehistoric and historic settlement. For the latter, this includes the arrangement of streets and city-blocks and the physical composition of buildings. Since the urban setting of Pherai largely remains buried under Velestino, our work was an important step toward a better appreciation of the city s ancient built environment. History and Archaeological Background seemingly abandoned in the Early Roman Imperial period (1st century CE) (Béquignon 1937; Dougleri Intzesiloglou 1990; 1994). It was not settled again until the post-byzantine town, called Velestino, was established in the 13th century CE. Prehistoric remains consist of the Neolithic tell, Magoula Bakalis, and Bronze Age graves scattered on the plateau and sloping terrain east of the tell. An Early Iron Age (EIA) settlement was located on the plateau as well, if the location of contemporary burials, terrace walls, and surface pottery serve as an indication (Morgan 2003, 92-5). Smaller clusters of burials have been found elsewhere, most notably some impressive tholos tombs to the northeast (Arachoviti 1994). By the end of the 8th century BCE, an important cult of En(n)odia and Zeus was established northeast of the plateau, but it was not until the 6th century BCE that a stone temple was constructed (Østby 1992; 1994). The establishment of the cult coincided with an expansion of the settlement, as indicated by the distribution of surface pottery and excavated remains of the same period. Overall, the nature of the EIA settlement, with dispersed bands of villages eventually coalescing into a core habitation zone, resembles in many ways the better known examples of early Argos and Corinth (Donati 2014). Excavations carried out by the 13th Ephorate of Prehistoric and Classical Antiquities within and around Velestino demonstrate that Pherai was occupied more or less continuously from the Late Neolithic period (ca BCE) until the site was Throughout the Archaic and Classical periods, Pherai developed into one of the most prominent Thessalian cities. Its apex occurred during the 4th century BCE, when the city was ruled by a tyranny, most notably Jason of Pherai (Sprawski 1999). At this time, 1 of 16

2 fortification walls were built to protect the acropolis and regions of the city to the south and east. The best preserved wall sections are the towers and gate found on the Agios Athanasios hill (Kakovoyiannis 1977). The urban boundaries of the rest of the city remain uncertain, since the modern town limits the scale of exploratory excavations. Estimates of the size of the classical city range from ha (Hansen and Nielsen 2003, 705). Other known archaeological features are residential buildings, a Hellenistic stoa in the northeastern part of the city that may mark the location of the agora, and the famous Hypereia spring in the center of the modern town. Satellite Remote Sensing Before fieldwork, a preliminary remote sensing campaign was conducted using high-resolution multispectral Quickbird (15 June 2009) and GeoEye- 1 (4 May 2010) satellite imagery and historical aerial photographs from the 1960s and 1970s. Spectral filters and band ratios applied to the satellite imagery proved to be valuable in identifying a network of organized streets, especially in the plateau east of Magoula Bakalis. Here, several slanting north-south streets appear as parallel linear anomalies spaced roughly equidistant from one another (Figures 2-4). To be more specific, we found evidence for at least 12 north-south streets spaced approximately 30 m from one another. There is a distinct convergence of the central cluster of streets toward a common location to the north. It is plausible that a gate once existed in the now destroyed northern fortification walls and the roads were therefore provided access. An east-west section of the buried fortification wall appears just to the east as a linear anomaly. At least one east-west street was detected in the satellite imagery at the southern edge of the plateau; however, we did not find an extension of the north-south roads further south. Perhaps the sloping terrain limited the extension of an urban street system here. Anomalies in the fields below the plateau may tentatively be interpreted as a continuation of the same diagonal street system. It is noteworthy that the anomalies on the plateau east of Magoula Bakalis are not organized in a strict orthogonal manner in the classical and Hellenistic traditions, but, instead, they have a diagonal arrangement analogous to planned street systems of the Archaic period, most notably in the Greek colonial foundations in Sicily such as Megara Hyblaia (Figure 5). The implication is that this region of Pherai may have been planned with a regular arrangement of streets at an early period in the city s history. Of course, only future fieldwork 2 of 16 and ground truthing could confirm or reject our initial theory on the chronology. Techniques, Equipment, and Parameters of Geophysical Exploration The geophysical survey was implemented in part to confirm and complement the findings of satellite remote sensing (in particular the roads), while uncovering additional urban features at a higher spatial resolution. The results were quite successful, especially in clarifying the location of archaeological features and the general layout of the city on the plateau east of Magoula Bakalis. Near-surface geophysical methods are able to detect and map buried archaeological features by measuring the physical properties of the Earth s surface in a non-destructive manner. Since each geophysical technique collects information of different physical properties, it is beneficial to use different methods together. This integration may complement the information extracted from the datasets of each technique and enhance the interpretation of the results. Localized environmental conditions (e.g., matrix composition of soil, depth of features, soil saturation) may impact the detection capabilities of each method. Therefore, the use of an integrated approach in geophysical prospection maximizes the ability to detect buried features. At Pherai, the scope of geophysical exploration was extensive and the methods of extracting archaeological data were diverse. New generations of multi-component geophysical equipment used by a team of six people enabled us to survey more surface area per day (~3.5 hectares) compared to conventional equipment. Survey was applied on a systematic basis over the course of three days using three separate methodologies (magnetometry, ground-penetrating radar, and Slingram electromagnetics). In total, we covered ~10 hectares at Pherai and out of this ~6.5 hectares was new coverage without overlap from the different methodologies (Figure 6). Magnetometry Magnetic measurements deal with anomalies of the geomagnetic field, caused by contrasts of magnetization produced by magnetic rocks or soils rich in magnetic oxides. The magnetization of bodies contains shares of inductive and remnant magnetization (Kaufman, Hansen, and Kleinberg 2008). Areas with enhanced magnetic susceptibility (with respect to the surrounding soil) are represented

3 as positive anomalies, while areas with smaller percentage content in iron oxides (i.e. smaller values of the magnetic susceptibility) are represented as negative magnetic anomalies. Both kinds of anomalies are interesting in the interpretation process of magnetic data. At Pherai, the measurement of the magnetic field was carried out by a multi-sensor SENSYS Magnetometer MX system, controlled by a team of three operators (Figure 7). The SENSYS consists of an array of eight fluxgate gradiometers accompanied by a GPS rover station, which continuously communicates with a base station GPS. This setup achieves high positioning accuracy in the motion (<1 cm) of the magnetometer and the rapid assessment of the archaeological site. The wide frame (~4 m) of the instrument fitted with multiple sensors greatly increases coverage area. In use at Pherai, 5-8 times larger area was covered than conventional single-sensor magnetometers. Ground-Penetrating Radar Ground-penetrating radar (GPR) uses transmission and reflection of high frequency electromagnetic energy to characterise the subsurface. A transmitting antenna transfers the energy into the ground which propagates as EM waves. If the electromagnetic waves travelling through one material meet another having different electromagnetic properties (dielectric permittivity), the waves will reflect or scatter from the boundary between the materials. Part of the transmitted energy is reflected back to the surface and captured by a receiving antenna. The time for this energy to be transmitted and returned back to the surface is measured and can be used to estimate the depth of the reflectors by knowing the velocity of wave propagation in medium. The depth estimation capacity of GPR as well as the high resolution of the results and its 3D data visualization capacities are clear advantages for archaeological investigations. At Pherai, we used a single channel Sensors and Software NOGGIN Plus-Smart Cart system fitted with a 250 MHz shielded antenna (Figure 8). Slingram Electromagnetics 3 of 16 Slingram electromagnetic (EM) methods use transmitting alternating primary fields which induce eddy currents in underground bodies with dissimilar electric conductivity. Changes of out-of-phase components allow conclusions regarding the positions of high or low conductive bodies, and inphase components allow conclusions regarding the magnetic susceptibility. Compared to the magnetometer and GPR, an EM device is very lightweight and mobile. Only a single person is required to operate the system in the field, walking in alternating straight lines m apart. We used two systems at Pherai: (1) GEM-2 multi-frequency electro-magnetometer, and (2) CMD multi-coils spacing electro-magnetometer (Figure 9). One major advantage, despite the lower resolution, is that EM techniques can provide both electrical conductivity and magnetic susceptibility data without ground contact and with a well-known depth of investigation (unlike magnetics). The EM is also advantageous because it is well-adapted in high conductivity environments in contrast to the GPR. Results of Geophysical Survey 1. Magoula Bakalis The density of archaeological material on the Neolithic tell from such a long period of occupation presents challenges for geophysical prospection and interpretation. -Today, Magoula Bakalis is almost completely flat on top and covers an area of approximately 0.5 ha. There are few above surface remains apart from a scattering of (late?) chamber tombs exposed in old excavation trenches and an early 20th century sarcophagus of the eponymous owner (Bakalis) of the land. A combination of all three methodologies were used to survey the tell; however, the best results came from magnetics (Figures 10-11). Here, we identified a number of negative anomalies with square and rectilinear forms. One explanation for these features is that they are buried chamber tombs, similar to the few that one can see on the site today. Yet we cannot exclude the possibility that they are structures related to the prehistoric and/or historic settlement. A large feature (ca. 25 m x 17 m) appears on the western side of the tell with an east-west orientation. Its size is distinct from the smaller rectilinear features. Rather than a tomb, the feature is more likely a monumental structure (temple?) that was positioned on a prominent position on the tell. 2. Streets The geophysical survey was successful in mapping the ancient street system on the plateau east of Magoula Bakalis. Magnetics and electromagnetics both confirmed the evidence from satellite remote sensing and expanded to include zones where the satellite data was ambiguous or lacking. Streets appeared as magnetic and conductive anomalies in comparison to the surrounding soil matrix. They raise attention because of their uniform orientation and

4 spacing sequence. Sections of six north-south streets were identified in the northern region of the plateau from magnetics (Figures 12-13). All terminate at an incomplete, yet distinct anomaly that either marks the location of the fortification walls or a ring road inside the walls. The steep terrain limited our ability to map this feature in more detail. In the southern half of the plateau, we identified nine north-south streets and one east-west street with magnetics. The electromagnetic results from the same region are comparable to the magnetics, and at least seven north-south streets and the same east-west street were mapped (Figures 14-15). The EM revealed in greater detail the widths of the streets, which range from 6-7 m. All streets found with magnetics and electromagnetics correspond to most of the anomalies identified in satellite remote sensing (Figure 16). The spacing of the north-south streets (ca. 30 m) and their slanted orientations are characteristics that we anticipated from the satellite data. The geophysical results also confirm the convergence of the central cluster of streets toward a common location to the north. The single east-west street in the southwestern region of the plateau was the only clear example of a street bisecting the north-south streets. Other anomalies in the southern region could plausibly be from east-west streets, but the data so far is inconclusive. It is not possible at this time to comment on the dimensions of city-blocks at Pherai. However, from what we observe, it seems that the north-south streets were extended features with few crossroads. This may indicate that city-blocks on the plateau were organized per strigas (i.e. elongated blocks, rather than squares), which is usually (but not always) representative of early Greek street systems (Greco and Torelli 1983). Although satellite remote sensing identified potential streets from the region below the plateau (see Figure 4), the geophysical survey found no similar evidence for a continuation of the street system (Figures 17-18). The general lack of data here from magnetics, GPR, and EM may be an indication that architectural features are deeper (and therefore harder to map) or heavily eroded. 3. Buildings (residential?) buildings. It would be beneficial to explore the same region using ground-penetrating radar, which we did not employ here, in the hopes of having better clarity in the architectural details. Preliminary Conclusions Geophysical survey at Pherai was successful in revealing the organization of urban space and the probable location of buildings and tombs. Many of the anomalies that we identified appear to define a subsurface system of organized streets. The frequency, ordered arrangement, and metrology of these anomalies are too coincidental to presume otherwise. Many are spaced at regular intervals from one another. A repeated pattern that emerges is a system of slanting north-south anomalies spaced 30 m from one another. While geophysical prospection has definitively shown that the region of Pherai east of Magoula Bakalis was planned with regular streets, it is challenging to establish the chronology of the street system without ground truthing. Ongoing rescue excavations by the 13th Ephorate of Prehistoric and Classical Antiquities may prove decisive. Bibliography Arachoviti, P Θολοτός προτογεωμετρικός τάφος στην περιοχή των Φερών, in Θεσσαλία: δεκαπέντε χρόνια αρχαιολογικής έρευνας, , Athens: Ministry of Culture. Béquignon, Y Recherches archéologiques à Phères de Thessalie. Paris: Société d'édition: Les belles lettres. Donati, J.C The City in the Greek and Roman World, in Oxford Handbook of Greek and Roman Art and Architecture, ed. C. Marconi, New York: Cambridge University Press. Dougleri Intzesiloglou, A Φεραί, Αρχαιολογία 34: Dougleri Intzesiloglou, A Οι νεότερες αρχαιολογικές έρευνες στην περιοχή των αρχαίων Φερών, in Θεσσαλία: δεκαπέντε χρόνια αρχαιολογικής έρευνας, , Athens: Ministry of Culture. In addition to the features on Magoula Bakalis (see above), many anomalies were mapped on the plateau as clusters of subsurface material between the roads. Although poorly defined in magnetics and electromagnetics, these are likely the remains of Greco, E., and M. Torelli Storia dell urbanistica: Il mondo greco. Rome: Laterza. 4 of 16

5 Hansen, M.H., T.E. Nielsen, eds An Inventory of Archaic and Classical Poleis. Oxford: Oxford University Press. Hoepfner, W Die Griechische Agora im Überblick, in Die griechische Agora, eds. W. Hoepfner, and L. Lehman, 1-28, Mainz: P. von Zabern. Kakovoyiannis, E Ανασκαφικές έρευνες στις Φέρες της Θεσσαλίας το 1977, Athens Annals of Archaeology 10: Kaufman, A.A., R.O. Hansen, and R.L. Kleinberg Principles of the Magnetic Methods in Geophysics. Amsterdam: Elsevier. Morgan, C Early Greek States Beyond the Polis. New York: Routledge. Østby, E Der dorische Tempel von Pherai, Opuscula Atheniensia 19.8: Østby, E A Reconsideration of the Classical Temple at Pherai, in Θεσσαλία: δεκαπέντε χρόνια αρχαιολογικής έρευνας, , Athens: Ministry of Culture. Sprawski, S Jason of Pherae: A Study on the History of Thessaly in Years B.C. Krakow: Jagiellonian University Press. 5 of 16

6 h eo S IM a t S- R FO es e R A TH rc G Figure 1. The ancient city of Pherai and modern Velestino from a high-resolution Quickbird satellite image taken on 15 June of 16

7 Figure 2. Decorrelation Stretch applied to the 15 June 2009 Quickbird image in the northern region of Velestino. The linear anomalies in the northern half of the field likely mark the location of buried streets Figure 3. Modified Simple Ratio (MSR) applied to the 4 May 2010 GeoEye-1 image in the same region as Figure 2. Linear anomalies appear this time in the southern half of the field 7 of 16

8 Figure 4. Linear anomalies identified through satellite remote sensing at Pherai 8 of 16

9 Figure 5. The Archaic street system around the agora at Megara Hyblaia (from Hoepfner 2006) 9 of 16

10 Figure 6. The coverage extent of the March 2014 geophysical survey at Pherai showing the different methodologies 10 of 16

11 Figure 7. 8 sensor SENSYS Magnetometer MX operated by Carmen Cuenca-García, Jamieson C. Donati, and Tuna Kalaycı. Photo by Meropi Manataki Figure 8. Noggin GPR operated by Meropi Manataki. Photo by Apostolos Sarris 11 of 16

12 Figure 9. CMD electromagnetometer operated by François-Xavier Simon. Photo by Meropi Manataki Figure 10. Magnetic results from Magoula Bakalis Figure 11. Interpretation of the magnetic results from Magoula Bakalis 12 of 16

13 Figure 12. Magnetic results from the plateau east of Magoula Bakalis Figure 13. Interpretation of the magnetic results from the plateau east of Magoula Bakalis. Red lines indicate the likely location of streets 13 of 16

14 Figure 14. EM results from the plateau east of Magoula Bakalis Figure 15. Interpretation of the EM results from the plateau east of Magoula Bakalis. Red lines indicate the likely location of streets 14 of 16

15 Figure 16. Anomalies identified through both geophysical prospection and satellite remote sensing on the plateau east of Magoula Bakalis 15 of 16

16 Figure 17. Magnetic results from the region south of Magoula Bakalis Figure 18. Anomalies identified through both geophysical prospection and satellite remote sensing from the region south of Magoula Bakalis 16 of 16