Mineralogia e geoquímica de terra preta arqueológica para identificação de padrão ocupacional pré-histórico no vale do Baixo rio Amazonas (Juruti, Pará)

Abstract

In the Amazon basin, the soils of a large number of sites have been modified by ancient pottery-making peoples. These soils, known as Indian Dark Earth or Archeological Dark Earth (ADE), generally cover relatively small areas, but in some cases, may extend to dozens of hectares. Some of these ADE sites are surrounded by soils known as Terra Mulata (TMA). The ADEs are dark in color, and have relatively high levels of Ca, Mg, P, Mn, Zn, Cu and organic C, ceramic fragments (CF), and charcoal. By contrast, while the TMA are relatively dark, they lack ceramic fragments, and have lower levels of Ca, Mg, P, Mn, Zn, Cu, and organic C than the ADE, even though these levels are generally higher than those found in adjacent soils or Latosol (Oxisols). Along the right bank of the lower Amazon River, at Juruti, in the Brazilian state of Pará, ample areas of ADE were found surrounded by equally extensive tracts of TMA the during archeological rescue operation that preceded the construction of the ALCOA bauxite extraction and processing plant. These locations, known to be archeological sites rich in ceramic fragments and organic matter, were investigated with the aim of understanding the possible inter-relationships between the ADE and TMA soils and the specific role of the prehistoric occupation of these soils, and the determination of the pattern of occupation, and the time scale of the occupation of the Amazon basin. Samples of soil and CF were obtained from the two archeological sites with ADE and TMA selected for this study. The soil and CF samples were collected from pedological profiles within a standardized 60 m x 120 m grid, representative of the ADE, TMA and adjacent soils. The mineralogy of the samples was then analyzed using XRD, SEM-EDS and optical microscopy, and their chemistry (including major and trace elements) was determined using ICP-OES and ICP-MS. In addition to these chemical and mineralogical analyses, the CF were examined archeologically with a binocular lens, and dated using thermoluminescence. The fragments of charcoal were dated using the radiocarbon (C14) method using the AMS technique. The results show that the soils are made up essentially of quartz and kaolinite, with illite + muscovite, goethite + hematite, and anatase in smaller proportions. These minerals reflect the Al2O3-Fe2O3-TiO2 association that has been interpreted as the signature of soils derived from the ferrous-aluminous crust of lateritic profiles, equivalent to the adjacent soils. While ADE and TMA are relatively similar in their mineralogical and chemical composition, they are partly different in the content of each mineral and in their concentrations of the chemical elements. On the other hand, apatite, Al phosphate, and cristobalite are practically exclusive to the ADE, being found only rarely in the TMA, and entirely absent from the Latosol. The apatite and Al phosphate represent the principal sources of the high levels of Ca and P. The ADE and TMA soils are also characterized by the high levels of organic matter, and can be identified by the P2O5-CaO-MgO-Na2O-K2O-Zn-Cu-Mn-Ba-Sr-Li-Ni association, which can be interpreted as an indicator of human activity. The geochemical maps of these elements and the distribution of the CF permitted the delimitation of the ADE zones as representing areas of permanent human occupation, defined by the higher concentrations. Seven patches of darker, enriched soil were grouped in two principal nuclei, separated by corridors of soils equivalent to adjacent area (Al2O3-Fe2O3-TiO2-Cr-Y-V-Zr geochemical association). These patches, characterized by an abundance of ceramic fragments, were interpreted as being the sites of ancient settlements. The TMA lies in a wide, continuous belt parallel to the river, between the ADE and the adjacent soil, and is interpreted as an area of campsites or the locations of agricultural activities. The stylistic characteristics and the chronological order of these fragments indicate that the occupation of the area is represented primarily by two distinct ceramic phases, the older Pocó phase, and the more recent Konduri phase. The thermoluminescence and C14 dating indicate that these pottery-making peoples probably occupied the region between 140 B.C. and the 13th century, with the ADE being formed subsequently. The ceramic fragments represent the remains of both ceremonial vessels and everyday artifacts. Their mineralogical and chemical composition is similar to that of the clay used by the region’s present-day potters. The differences are restricted to the high levels of phosphorus (between 1% and 4% P2O5) found in the fragments in the form of amorphous Al and Fe phosphates, which are found at crustal levels in the present-day clay. This reinforces the conclusion that their origin was related to the use of the ceramic vessels, especially for cooking food. From the mineralogical and chemical similarities between the CF of distinct cultures and ages, it is possible to infer that the raw material used by the ancient settlers of this region were invariably sourced from the same deposits or geological formations, beginning more than 2,000 years ago, and which are still used by the region’s present inhabitants. Overall, then, the evidence indicates that the deposits of ADE and TMA soils in the area of Juruti, which extend for more than 350 hectares, are a consequence of extensive and intense prehistoric human activities, which imposed significant transformations on the original local soils. The hot and humid climatic conditions, and the dense forest cover were fundamental to the formation of these soils.