Geoarchaeological studies in Nara,Japan: The integrated findings

In this article, the authors summarize 20 years of scholarship and two field projects on geoarchaeology in Nara, Japan, carried out by researchers from 12 different institutions in 4 countries. The research goal was to test an aerial photographic reconstruction of surface landforms in the Nara Basin with subsurface data. Project A was conducted at Asawa; it tested, through geological coring, whether a suspected swampy backmarsh in the eastern basin existed and whether it would yield data on the transition to wet rice agriculture in the mid‐1st millennium B.C. Project B was conducted at Miwa; it tested, through geological coring and subsequent excavation, the nature of upland agricultural terrace formation in the southeastern basin and whether the suspected existence of a 4th‐century palace site could be confirmed. Two layers of carbonaceous clay at Asawa were dated to the Early (˜5000–3500 B.C.) and Final (˜1000‐300 B.C.) phases of the Jomon period. Pollen data revealed the establishment of an evergreen oak forest from 5,000 years ago and anthropogenic changes in forest cover from 2,000 years ago. Phytoliths from rice, millets, reeds, and bamboo were recovered in layers postdating the Final Jomon carbonaceous clay. A fault scarp with anthropogenic modification of the terraces was identified at Miwa. It was discovered that an incised stream valley had been infilled in the Medieval period at the same time surface layers were razed; the front of the terrace was extended in the premodern period. Remains were recovered from the Middle Yayoi (˜100 B.C.–A.D. 100) and the Medieval (˜A.D. 1185–1603) periods. However, as the terracing involved razing the early historic levels, no data were recovered on the alleged 4th‐century palace site. The significance of these findings lies in the identification of (a) a swampy backmarsh at Asawa, where initial agricultural efforts in growing wet rice in the basin may have occurred, confirming the aerial photographic reconstruction; and (b) hillside terracing activities at Miwa, from the Medieval period onwards, which have radically changed the configuration of the natural topography. © 2005 Wiley Periodicals, Inc.     

Lead isotopic analysis within a multiproxy approach to trace pottery sources. The example of White Slip II sherds from Late Bronze Age sites in Cyprus and Syria

Lead isotope analyses were carried out on fragments of White Slip II ware, a Late Bronze Age Cypriote pottery ware, and on raw materials possibly used for their production. Sherds originate from three Late Bronze Age sites (Hala Sultan Tekke and Sanidha in Cyprus and Minet el-Beida in Syria) and clays come from the surroundings of Sanidha, a production site for White Slip ware. X-ray fluorescence (XRF) and a Principal Component Analysis (PCA) are combined with Pb isotope analyses to further investigate the effectiveness of the latter method within a multiproxy approach for pottery provenance study. The pottery sherds from the three sites are compared between themselves and with potential raw material. Additional X-ray diffraction (XRD) and analyses using a scanning electron microscope (SEM) equipped with an energy dispersive X-ray detection (EDX) facility were performed on selected sherds and clays. This work confirms that the clay source used for pottery production in Sanidha derives from local weathered gabbro. It also shows that different origins can be proposed for White Slip II ware sherds from Hala Sultan Tekke and Minet el-Beida and that clays were prepared prior to White Slip II ware production. It finally confirms the effectiveness of Pb isotopes in tracing pottery provenance not only by comparing sherd assemblages but also by comparing sherds to potential raw materials.     

Provenance of sand temper in Hohokam Ceramics,Arizona

Sand temper compositions of regionally distinct Hohokam pottery sherds were determined by detailed point counts using sedimentary petrographic methods. Different temper compositions from different sites were compared with maps of sand composition zones within the geographic range of the pottery in order to establish the probable provenance of each sherd. A number of probable instances of intraregional pottery exchange were identified. The larger number of sand sources in undecorated vs. decorated pottery at each site suggests that undecorated pottery was made in more places than decorated vessels, and may have been traded differently as well. Petrographic analysis of temper is a useful method for studying exchange of homogeneous pottery in geologically diverse areas, and for investigating prehistoric Hohokam interaction in the Tucson and Red Rock Basins on a scale not possible with traditional archaeological techniques.     

Using Lead Isotopes to Determine Pottery Provenance in Cyprus: Clay Source Signatures and Comparison with Late Bronze Age Cypriote Pottery

This paper presents the lead isotopic composition of potential clay sources for pottery production, collected in the four major geological zones of Cyprus (Troodos Ophiolite, Circum Troodos Sedimentary Succession, Mamonia Terrane, Kyrenia Terrane) and evaluates its usefulness in Cypriote pottery provenance studies. The clay isotopic signatures from the four zones are compared to each other and to the isotopic composition of various utilitarian pottery wares from three Late Bronze Age sites, respectively located in southern Cyprus (Alassa‐Pano Mandilaris), east Cyprus (Enkomi), and southeast Cyprus (Hala Sultan Tekke). It also explores the potential of this method to better discriminate between potential raw materials used for the production of Base‐ring ware, one of the most characteristic fine pottery of Late Bronze Age Cyprus, which was widely spread in the Eastern Mediterranean (Courtois, 1981; Vaughan, 1991, 1994). Results show that three main lead isotopic fields can be distinguished among the Cypriote clay sources and the comparison of Plain sherds with the clay sources allows discrimination between local products and imports. They also clearly indicate that all the Base‐ring sherds analyzed in this study were made of the clays from the Kathikas Formation that crops out in only limited parts of southwest Cyprus.     

Sourcing Olive Jars Using U‐Pb Ages of Detrital Zircons: A Study of 16th Century Olive Jars Recovered from the Solomon Islands

We present U‐Pb ages of zircons extracted from olive jars recovered from two sites associated with Alvaro de Mendaña y Neyra's colonising expedition to the Solomon Islands, c. 1595–1596 A.D. The olive jars were previously associated with Panamanian and Peruvian origins based on petrological and geochemical studies. To further define provenance, 143 zircons were extracted from five olive jar sherds, analyzed and dated. The resultant U‐Pb ages range from the Archaean to the Cenozoic (2977.2 ± 29.0–3.2 ± 4.0 Ma), but the dominance of Cretaceous and Palaeogene ages (∼ 90% of the total age population is between ∼ 145 and 23 Ma) supports a Peruvian origin based on comparative geology, with the Coastal Batholith of Peru a prime candidate area of ceramic production. These results are significant for the characterization of 16th Century Peruvian‐made pottery and our understanding of its production and trade.     

A comparative electron microprobe study of “Aeginetan” wares with potential raw material sources from Aegina,Methana, and Poros,Greece

Qualitative stylistic evidence from ceramic vessels and limited petrographic analysis suggested that a distinctive group of ceramics with visible inclusions of biotite (Gold Mica Fabric) was produced on the island of Aegina, Greece, during the Middle Helladic and Late Helladic I periods. To quantitatively evaluate this provenance, we sampled all potential source rocks on Aegina, Methana, and Poros. Electron microprobe analysis of amphibole in these samples revealed that each of these volcanic centers has its own unique mineralogical signature. Comparative analyses of amphibole in Zerner's original stylistic “Gold Mica Fabric” type sample with the reference samples reveal that two sherds are Aeginetan. Three additional sherds from this sample may have a non‐Aeginetan provenance, probably from a back‐arc setting outside the Saronic Gulf. These results suggest that the hypothesis of a single source production site for Aeginetan Ware should be reexamined. © 2002 Wiley Periodicals, Inc.     

Geochemical study of pottery sherds from an archaeological site near Mošnje (Slovenia)

This study is a geochemical analytical approach to the characterization of pottery samples from an archaeological site near Mo?nje (Slovenia). Inductively coupled plasma–mass spectrometry and statistical analyses were used to determine detailed geochemical properties of the pottery sherds and to identify potentially individual groups among the samples studied. The geochemical results indicated the existence of four major groups of the pottery sherds: the first and second groups are assembled from eight samples, all generally characterized by their high CaO and TOT/C content; the third group comprised the samples with the highest SiO₂ concentrations; and the fourth group is represented by the samples K3, K5 and K13. The principal component analysis and cluster analysis validated the existing groups and revealed a high degree of chemical similarity between these groups. The geochemical and statistical data confirmed the archaeologists’ hypothesis and interpretation of a similar origin/alteration of source material/probable local ceramic production for the majority of the pottery sherds; the imported origin of samples K3 and K13 was recognized, while sample K5 had been intentionally imported as a sample for comparative purposes.     

Preliminary chemical and micromorphological observations on Urartu (800–600 B.C.) ceramics,eastern Turkey

Anatolia has long been a major pottery production center of the ancient world, dating back 7000 cal yr B.P. The Early Iron Age Urartu Kingdom (800–600 B.C.) of eastern Anatolia is known for the production of high‐quality pottery, but little is known regarding firing technology and manufacture of these ceramics. Here we present a preliminary study of Urartu ceramic micromorphology and chemistry and suggest that the Urartus had good knowledge of local geology and intentionally used chemical fluxes (Pb, Rb, and Li) to attain desired firing temperatures. The sophisticated production of Urartu ceramics is comparable to later high‐quality Roman pottery (terra sigillata) procured from the same area. © 2010 Wiley Periodicals, Inc.     

Provenance and microprobe assays of phyllite‐tempered ceramics from the uplands of central Arizona

The ceramics in use across a broad upland zone of central Arizona during the early Classic period (ca. A.D. 1100–1300) were characterized by a lack of mineralogical variability; nearly all of the clay containers were tempered with one rock type, phyllite. Consequently, nearly all of the upland pottery is assigned to a single pottery type, Wingfield Plain. This compositional uniformity has frustrated ceramic provenance studies, and, as a result, little has been learned previously about the organization of ceramic production and exchange in the upland territory. There are, however, considerable and interpretable chemical differences in the phyllite‐tempered wares, as shown with microanalyses of the temper fragments and pottery clay fractions with an electron microprobe. The chemical patterning is useful for investigating issues pertaining to the upland zone, including the organization of ceramic manufacture, community arrangements, and pottery transactions during a time of prevalent hostilities in central Arizona. © 2008 Wiley Periodicals, Inc.     

Quantitative WD‐XRF calibration for small ceramic samples and their source material

A wavelength‐dispersive X‐ray fluorescence (WD‐XRF) calibration is developed for small powdered samples (300mg) with the purpose of analyzing ceramic artifacts that might be available only in limited quantity. This is compared to a conventional calibration using a larger sample mass (2g). The comparison of elemental intensities obtained in both calibrations shows that the decrease in analyzed sample mass results in a linear decrease in measured intensity for the analyzed elements. This indicates that the small‐ and large‐sample calibrations are comparable. Moreover, the elemental contents of four ceramic sherds and two potential clay sources fall well within the range of the certified reference materials that are the basis of the calibration curves. The advantage with the analytical method presented here is that it is rapid and requires only a small amount of sample that can easily be re‐used for further analyses. This method has great potential in ceramic provenance studies. © 2011 Wiley Periodicals, Inc.     

The source of raw materials for Roman pottery from Leptiminus,Tunisia

Leptiminus, a Roman port city on the west coast of Tunisia, North Africa, exported olive oil and a garum fish paste to Rome. Excavations have uncovered many facilities including kilns and a potter's workshop, indicating an extensive ceramic industry. The vessels, manufactured at Leptiminus, included African red‐slip fineware, coarseware, and amphorae. A petrographic study of pottery sherds showed them to contain very similar temper, rounded aeolian sand grains and limestone, but varying in proportions to produce different textures. An investigation of the source of raw materials for the pottery found three distinct types of clay within a 50 km radius of Leptiminus: grey Miocene, brown Pliocene, and green Late Pliocene clay. Statistical analysis of trace element compositions, using induced neutron activation, of clays and sherds showed that the Pliocene brown clay was used to create all types of pottery. The specific clay horizon, used in Roman times, has apparently been removed by quarrying. However, a lateral continuation of this bed was found 7 m beneath the Leptiminus site 290. © 2002 Wiley Periodicals, Inc.     

Soil parent materials and the pottery of Roman Galilee: A comparative study

The paper presents a comparative micromorphological analysis of the range of soil materials used to make the pottery of the hilly Galilee during the Roman period, and the ceramic products made from these materials. The four soil units that served as raw material for most of the pottery made in this period and region are examined along with pottery derived from each of them. For each soil unit, the soil characteristics and processes are described, followed by a presentation of the micromorphological characteristics of the soil material and those of the pottery made from that material. The contribution of the aeolian dust component to the soil materials is discussed as well as the identification of the tempering materials (nonplastics or other soil materials) added to the pottery paste. The study demonstrates the close correlation in microfabric between the pottery and original soil materials, sheds light on the raw material selection and modification practices of the potters of Roman Galilee, and has significant implications for provenance studies, using chemical analysis, on the pottery of this period and region. © 2002 Wiley Periodicals, Inc.     

The management of arable land from prehistory to the present: Case studies from the Northern Isles of Scotland

The arable soils from two multiperiod settlements were analyzed to identify changes in agricultural methods over time. The settlement middens were also analyzed to determine whether potential fertilizers were discarded unused. Results suggest that in the Neolithic period (˜4000–2000 B.C. in the UK) the arable soils at Tofts Ness, Orkney, and Old Scatness, Shetland, were created by flattening and cultivating the settlements' midden heaps in situ. The arable area at Tofts Ness was expanded in the Bronze Age (˜2000–700 B.C. in the UK), and the new land was improved by the addition of ash, nightsoil, and domestic waste. Cultivation continued briefly after the fields were buried in windblown sand in the Late Bronze Age or Early Iron Age, but by the Early Iron Age cultivation ceased and organic‐rich material was allowed to accumulate within the settlement. By contrast, at Old Scatness, arable production was increased in the Iron Age (˜700 B.C.–A.D. 550 in Scotland) by the intensive use of animal manures. The results indicate that during the lifespan of the two settlements the arable soils were fertilized to increase production, which was intensified over time. © 2006 Wiley Periodicals, Inc.     

Evidence for a single clay/temper source for the manufacture of Middle and Late Helladic Aeginetan pottery from Asine,Greece

In an effort to further characterize the Middle and Late Helladic pottery industry on Aegina, we have analyzed amphibole in 23 sherds imported to the coastal settlement of Asine. The sherds derive from vessels of different classes and shapes and range in age from MH I‐II to LH IIIB‐IIIC Early. All sherds come from vessels that carry manufacturing marks, and their amphiboles have compositions that are incompatible with those of Methana, Poros, and Melos. Twenty of the sherds have amphiboles that are identical in composition and overlap a narrow range of amphibole compositions found in specific lava flows on the northern portion of Aegina. Given that the dacites across Aegina contain amphiboles with a wide range in compositions, we suggest that the narrow range of amphibole compositions in the sherds indicates that they were derived from either a specific clay source on the island, located in a stream system southeast of the prehistoric settlement at Kolonna, or that the potters used a specific temper source along the same stream system. Multiple clay or temper sources would have produced sherds with a broader range of amphibole compositions reflecting the diversity of amphibole compositions found on Aegina. One sherd has amphibole compositions indicative of an additional Aeginetan component that is not found in the other sherds. Two sherds have amphiboles with compositions that do not match any known reference amphiboles for Aegina, Methana, Poros, or Melos. These may have been derived from still unsampled dacites on Aegina or have been manufactured from materials located outside the Saronic Gulf. © 2004 Wiley Periodicals, Inc.     

A magnetic study of Etruscan bucchero pottery: An application of rock magnetism to archaeometry

We have applied techniques used in rock magnetism to the study of possible temporal changes in provenance and firing conditions of a collection of Etruscan bucchero pottery fragments representing the interval between 800 B.C. to 400 B.C. To our knowledge, this is the first time that such procedures have been applied to ancient ceramic wares. Our preliminary magnetic measurements demonstrate that this approach can provide useful archaeological information, as exemplified by the following results: (1) Bucchero ceramic styles sottile, spesso, and possibly pesante, contained magnetic particles of similar grain sizes, which suggested a common provenance and similar firing conditions for these pot typologies. (2) Ceramic styles buccheroid impasto and grey bucchero each had magnetic characteristics different from those in (1). However, too few samples of these were available for any definite conclusions. (3) The occurrence of magnetite, and not hematite, in all samples indicated that firing conditions were effectively reducing. (4) Variations in the amount of magnetite in different samples indicated that slight inhomogeneities must have existed in the iron oxide/hydroxide content of the local clay source. Most magnetic techniques are fast, easy, and generally nondestructive. Some magnetic properties, such as initial susceptibility, saturation remanent magnetization, and coercivity of remanence, can even be measured in the field. At the least, magnetic measurements can be used initially to discriminate among different groups of potsherds. Selected samples can then be analyzed with more elaborate methods of archaeometry.     

Petrographic and geochemical characterization of Archaic‐Hellenistic tableware production at Solunto,Sicily

A selected assortment of Archaic‐Hellenistic tableware samples from Solunto, a Phoenician‐Punic site located 20 km east of Palermo (Sicily), has been subjected to thin‐section petrography and chemical analysis (XRF). In this settlement several ceramic kilns remained operative over a long time period (7th to 3rd century B.C.). The main goal of this analytical study is to distinguish the ceramics manufactured locally from regional and off‐island imports. Analytical results were matched to similar data concerning local natural clay sources and to coeval tableware productions from other sites in the same area. The ceramic pastes used by the ancient craftsmen of Solunto in the case of this class of pottery could be differentiated clearly by their petrochemical characteristics. We conclude that ceramics were locally produced far beyond satisfying just internal consumption needs, indicating interaction of Solunto with neighboring Greek colonies, indigenous people, and Phoenician‐Punic colonies of Sicily. © 2009 Wiley Periodicals, Inc.     

Thin section petrography of northern Lesser Antilles ceramics

Thin sections from 44 sherds representing eight prehistoric sites on four islands (Barbuda, Montserrat, Anguilla, St. Martin) in the northern Lesser Antilles (West Indies) were examined using a petrographic microscope. Point counting distinguished three temper associations: (1) exclusively volcanic,(2) volcanic and carbonate, and (3) volcanic, carbonate, and grog. Exclusively volcanic or dominantly volcanic (with low carbonate) temper associations occur in all Saladoid and many post-Saladoid sherds, with plagioclase feld-spar and volcanic rock fragments being most abundant. Thin sections with significant carbonate content or with grog were restricted to post-Saladoid sherds from Barbuda and to a lesser extent Anguilla. the presence of volcanic grains in sherds from the limestone islands of Barbuda and Anguilla indicates that volcanic islands in the region served as sources of pottery (or volcanic temper) for these two islands.     

Determination of the initial firing conditions of Ayyubid/Mamluk painted pottery excavated from northern Jordan

A combination of different scientific techniques has been utilized to estimate the initial firing temperature and kiln atmosphere of a collection of Islamic painted sherds excavated from an archaeological site in northern Jordan. Scanning electron microscope investigation of samples refired at different temperatures proves that the pottery was fired at around 750°C in an oxidizing atmosphere. Similar results were obtained by the thermal analysis study using thermogravometric, differential thermogravemetric and differential thermal analysis and by the temperature-minerological scale analysis. The obtained results shed light on the manufacturing techniques adopted by the Islamic potter of the period. © 1999 John Wiley & Sons, Inc.     

Geoarchaeological evidence of a Chola‐period tsunami from an ancient port at Kaveripattinam on the southeastern coast of India

Literature of the ancient Chola Dynasty (A.D. 9th–11th centuries) of South India and recent archaeological excavations allude to a sea flood that crippled the ancient port at Kaveripattinam, a trading hub for Southeast Asia, and probably affected the entire South Indian coast, analogous to the 2004 Indian Ocean tsunami impact. We present sedimentary evidence from an archaeological site to validate the textual references to this early medieval event. A sandy layer showing bed forms representing high‐energy conditions, possibly generated by a seaborne wave, was identified at the Kaveripattinam coast of Tamil Nadu, South India. Its sedimentary characteristics include hummocky cross‐stratification, convolute lamination with heavy minerals, rip‐up clasts, an erosional contact with the underlying mud bed, and a landward thinning geometry. Admixed with 1000‐year‐old Chola period artifacts, it provided an optically stimulated luminescence age of 1091 ± 66 yr and a thermoluminescence age of 993 ± 73 yr for the embedded pottery sherds. The dates of these proxies converge around 1000 yr B.P., correlative of an ancient tsunami reported from elsewhere along the Indian Ocean coasts. © 2011 Wiley Periodicals, Inc.     

It all comes out in the wash: Actualistic petrofacies modeling of temper provenance,Tonto Basin,Arizona, USA

Prehistoric potters frequently tempered their pottery with sand, the provenance of which can be established petrographically. In the Tonto Basin of central Arizona, the bedrock geology is highly variable, giving rise to geographically unique sands. Zones of sands with similar compositions can be modeled at an archaeologically relevant scale. Here we use the actualistic petrofacies concept, the Gazzi–Dickinson point‐counting technique, and multivariate statistics to create a petrofacies model, then apply it to sand‐tempered utilitarian pottery recovered from three Tonto Basin project areas. Data analysis reveals strong temporal and spatial ceramic production and consumption patterns. Production of pottery for exchange was established in at least one petrofacies ca. A.D. 600–950, with increasing specialization through time. By ca. A.D. 1150, corrugated wares had been added, and specialized production by ware was established in two petrofacies. Provenance evidence suggests different regional affiliations for groups in different parts of the basin. © 2001 John Wiley & Sons, Inc.