NEW DATA ON THE CHRONOLOGY OF THE VALE DO FORNO SEDIMENTARY SEQUENCE (LOWER TAGUS RIVER TERRACE STAIRCASE) AND ITS RELEVANCE AS FLUVIAL ARCHIVE OF THE MIDDLE PLEISTOCENE IN WESTERN IBERIA

NEW DATA ON THE CHRONOLOGY OF THE VALE DO FORNO SEDIMENTARY SEQUENCE (LOWER TAGUS RIVER TERRACE STAIRCASE) AND ITS RELEVANCE AS FLUVIAL ARCHIVE OF THE MIDDLE PLEISTOCENE IN WESTERN IBERIA Pedro P. Cunha 1, António A. Martins 2, Jan-Pieter Buylaert 3,4, Andrew S. Murray 4, Luis Raposo 5, Paolo Mozzi 6, Martin Stokes 7 1 MARE - Marine and Environmental Sciences Centre, Department of Earth Sciences, University of Coimbra, Portugal: pcunha@dct.uc.pt 2 MARE - Marine and Environmental Sciences Centre, Dep. Geociências, University of Évora, Portugal; aam@uevora.pt 3 Centre for Nuclear Technologies, Technical University of Denmark, Risø Campus, Denmark; jabu@dtu.dk 4 Nordic Laboratory for Luminescence Dating, Aarhus University, Risø DTU, Denmark; anmu@dtu.dk 5 Museu Nacional de Arqueologia, Lisboa, Portugal; 3raposos@sapo.pt 6 Department of Geosciences, University of Padova, Italy; paolo.mozzi@unipd.it 7 School of Geography, Earth and Environmental Sciences, University of Plymouth, UK; m.stokes@plymouth.ac.uk The stratigraphic units that record the evolution of the Tagus River in Portugal (study area between Vila Velha de Ródão and Porto Alto villages; Fig. 1) have different sedimentary characteristics and lithic industries (Cunha et al., 2012): - a culminant sedimentary unit (the ancestral Tagus, before the drainage network entrenchment) – SLD13 (+142 to 262 m above river bed – a.r.b.; with probable age ca. 3,6 to 1,8 Ma), without artefacts; - T1 terrace (+84 to 180 m; ca. 1000? to 900 ka), without artefacts; - T2 terrace (+57 to 150 m; top deposits with a probable age ca. 600 ka), without artefacts; - T3 terrace (+43 to 113 m; ca. 460 to 360? ka), without artefacts; - T4 terrace (+26 to 55 m; ca. 335 a 155 ka), Lower Paleolithic (Acheulian) at basal and middle levels but early Middle Paleolithic at top levels; - T5 terrace (+5 to 34 m; 135 to 73 ka), Middle Paleolithic (Mousterian; Levallois technique); - T6 terrace (+3 to 14 m; 62 to 32 ka), late Middle Paleolithic (late Mousterian); - Carregueira Sands (aeolian sands) and colluvium (+3 a ca. 100 m; 32 to 12 ka), Upper Paleolithic to Epipaleolithic; - alluvial plain (+0 to 8 m; ca. 12 ka to present), Mesolithic and more recent industries. The differences in elevation (a.r.b.) of the several terrace staircases results from differential uplift due to active faults. Longitudinal correlation with the terrace levels indicates that a graded profile ca. 200 km long was achieved during terrace formation periods and a strong control by sea base level was determinant for terrace formation. The Neogene sedimentary units constituted the main source of sediments for the fluvial terraces (Fig. 2). Geomorphological mapping, coupled with lithostratigraphy, sedimentology and luminescence dating (quartz-OSL and K-feldspar post-IRIR290) were used in this study focused on the T4 terrace, which comprises a Lower Gravels (LG) unit and an Upper Sand (US) unit. The thick, coarse and dominantly massive gravels of the LG unit indicate deposition by a coarse bed-load braided river, with strong sediment supply, high gradient and fluvial competence, during conditions of rapidly rising sea level. Luminescence dating only provided minimum ages but it is probable that the LG unit corresponds to the earlier part of the MIS9 (ca. 335 to 325 ka), immediately postdating the incision promoted by the very low sea level (reaching ca. -140 m) during MIS10 (362 to 337 ka), a period of relatively cold climate conditions with weak vegetation cover on slopes and low sea level. Fig. 1. Main Portuguese reaches in which the Tagus River can be divided (Lower Tagus Basin): I – from the Spanish border to Arneiro (a general E–W trend, mainly consisting of polygonal segments); II – from Arneiro to Gavião (NE–SW); III – from Gavião to Arripiado (E–W); IV – from Arripiado to Vila Franca de Xira (NNE-SSW); V – from Vila Franca de Xira to the Atlantic shoreline. The faults considered to be the limit of the referred fluvial sectors are: F1 – Ponsul-Arneiro fault (WSW-ENE); F2 – Gavião fault (NW-SE); F3 – Ortiga fault (NW-SE); F4 – Vila Nova da Barquinha fault (W-E); F5 – Arripiado-Chamusca fault (NNE-SSW). 1 – estuary; 2 – terraces; 3 – faults; 4 – Tagus main channel. The main Iberian drainage basins are also represented (inset). The lower and middle parts of the US unit, comprising an alternation of clayish silts with paleosols and minor sands to the east (flood-plain deposits) and sand deposits to the west (channel belt), have a probable age of ca. 325 to 200 ka. This points to formation during MIS9 to MIS7, under conditions of high to medium sea levels and warm to mild conditions. The upper part of the US unit, dominated by sand facies and with OSL ages of ca. 200 to 154 ka, correlates with the early part of the MIS6. During this period, progradation resulted from climate deterioration and relative depletion of vegetation that promoted enhanced sediment production in the catchment, coupled with initiation of sea-level lowering that increased the longitudinal slope. The Vale do Forno and Vale da Atela archaeological sites (Alpiarça, central Portugal) document the earliest human occupation in the Lower Tagus River, well established in geomorphological and environmental terms, within the Middle Pleistocene. The Lower Palaeolithic sites were found on the T4 terrace (+26 m, a.r.b.). The oldest artefacts previously found in the LG unit, display crude bifacial forms that can be attributed to the Acheulian, with a probable age of ca. 335 to 325 ka. The T4 US unit has archaeological sites stratigraphically documenting successive phases of an evolved Acheulian, that probably date ca. 325 to 300 ka. Notably, these Lower Palaeolithic artisans were able to produce tools with different sophistication levels, simply by applying different strategies: more elaborated reduction sequences in case of bifaces and simple reduction sequences to obtain cleavers. Fig. 2. . Simplified geologic map of the Lower Tagus Cenozoic basin, adapted from the Carta Geológica de Portugal, 1/500000, 1992). The study area (comprising the Vale do Forno and Vale de Atela sites) is located on the more upstream sector of the Lower Tagus River reach IV, between Arripiado and Chamusca villages. 1 – alluvium (Holocene); 2 – terraces (Pleistocene); 3 – sands, silts and gravels (Paleogene to Pliocene); 4 – Sintra Massif (Cretaceous); 5 – limestones, marls, silts and sandstones (Mesozoic); 6 – quartzites (Ordovician); 7 – basement (Proterozoic to Palaeozoic); 8 – main fault. The main Portuguese reaches of the Tagus River are identified (I to V). The VF3 site (Milharós), containing a Final Acheulian industry, with fine and elaborated bifaces) found in a stratigraphic level located between the T4 terrace deposits and a colluvium associated with Late Pleistocene aeolian sands (32 to 12 ka), has an age younger than ca. 154 ka but much older than 32 ka. In the study area, the sedimentary units of the T4 terrace seem to record the river response to sea-level changes and climatically-driven fluctuations in sediment supply. REFERENCES Cunha P. P., Almeida N. A. C., Aubry T., Martins A. A., Murray A. S., Buylaert J.-P., Sohbati R., Raposo L., Rocha L., 2012, Records of human occupation from Pleistocene river terrace and aeolian sediments in the Arneiro depression (Lower Tejo River, central eastern Portugal). Geomorphology, vol. 165-166, pp. 78-90.

Characterization of medieval glass artefacts from Miranduolo site, Chiusdino, Italy

This archaeovitreological study deals with artefacts of Miranduolo site, Tuscany region (Italy), dated 1250-1350 AD. The Miranduolo site is a medieval hill-village dated from 7th to 14th century. The information obtained reveal that Miranduolo was under control of noble families, which displayed the social, economic and political power. It is marked by controlling the farmers and metal workers on the site, as well as having control over agricultural surpluses. No in situ glass workshop has been recovered, implying that the glass artefacts were imported. One aim of this work is application of SEM-EDS to visualize textural characteristics and thickness of the pristine glass and corrosion layers. Preliminary qualification and semi-quantification of major and minor chemical elements will provide the data on the glass group present and fluxes employed. The data obtained will be integrated with the one obtained by more sensitive techniques such as PIXE/PIGE and LA-ICP-MS. Twenty cross-sections of transparent glasses (colorless, azure, and different hues of yellow and) have been analyzed by VP-SEM. All the analyzed glasses display a homogenous matrix. Only four samples (MD 24, MD 139, MD 143, MD 259) show corrosion layers of various thickness with 2.25μm, 136-500 μm, 26.8 μm and 17.01 μm. EDS linescan analyses indicate strong depletion in the corrosion layers of Na and K, while Ca depletes to a minor extent. In general, both glass composition and the burial conditions were favorable for preservation. Samples can be classified as mainly plant ash Na-Ca-Si glasses made with both unpurified and purified Levantine ash. Only sample MD 243 is made from Barilla plant ash. Sample MD 139 cannot be classified into main compositional groups as K2O is 1.33 wt% and MgO 5.92 wt%. In 8 samples MnO content is lower than 0.8 wt%, meaning that in these samples MnO is naturally present. In other 12 samples, MnO above 0.8 wt% indicates deliberate addition as a decolorant agent to intentionally obtain different hues or the amount added was not successful in making the glass transparent. The results considering fluxes are compatible with archaeovitreological study from contemporary primary glass workshops in Tuscany. For determining the provenance of silica sources, further analysis with more sensitive techniques has to be carried out; Resumo: Este estudo “arqueovitreologia” lida com artefatos do local Miranduolo, região da Toscana (Itália), datados de 1250-1350 AD. O sitio de Miranduolo é uma colina vila medieval datada do séc.VII ao séc.XIV. As informações obtidas revelam que Miranduolo estava sob o controle de famílias nobres, que exibiu o poder social, económico e político. É marcado por controlar os agricultores e trabalhadores do metal no sitio, bem como ter controlo sobre os excedentes agrícolas. Não há na oficina de vidro in situ foi recuperado, o que implica que os artefactos de vidro foram importados. Um dos objetivos deste trabalho é a aplicação de SEM-EDS para visualizar características de textura e espessura das camadas de corrosão do vidro também como da áreas originais. qualificação preliminar e semi-quantificação de maiores e menores elementos químicos irá fornecer os dados sobre o grupo presente vidro e fluxos empregado. Os dados obtidos são integrados com os dados obtidos por meio de técnicas mais sensíveis, como PIXE / PIGE e LA-ICP-MS. Vinte secções transversais de vidros transparentes (incolor, azul celeste, e diferentes tons de amarelo) foram analisados por VP-SEM. Todos os vidros analisados exibir uma matriz homogénea. Apenas quatro amostras (MD 24, MD 139, MD 143, MD 259) mostram camadas de corrosão de várias espessuras com 2.25μm, 136-500μm, 26,8μm e 17,01μm. Análises Linescan EDS indicam forte esgotamento nas camadas de corrosão de Na e K, enquanto Ca esgota, em menor grau. Em geral, tanto a composição de vidro e as condições de depósito foram favoráveis para a preservação. As amostras podem ser classificados como vidros principalmente Na-Ca-Si feitas com cinzas de plantas do tipo levantino, não purificada e purificada. Apenas a amostra MD 243 é feita a partir de cinzas vegetais tipo “Barilla”. A amostra MD 139 não pode ser classificada em grupos principais de composição porque K2O é 1,33% em peso e MgO 5,92% em peso. Em 8 amostras, o teor de MnO é menor do que 0,8% em peso, o que significa que nestas amostras MnO está naturalmente presente. Em outras 12 amostras, MnO acima de 0,8% em peso indica adição intencional como um agente de colorante para obter intencionalmente diferentes matizes ou o valor acrescentado não foi bem sucedido em fazer o vidro transparente. Os resultados, considerando os fluxos são compatíveis com o estudo “arqueovitreologico” com as principais oficinas de vidro contemporâneos na Toscana. Para determinar a origem das fontes de sílica, uma análise mais aprofundada com técnicas mais sensíveis tem de ser levada a cabo.