My Account: Log In | Join | Renew
Table of Contents
Select All Chapters
Soils in alluvial valleys are of interest in pedo-archaeological studies in developing the environmental history of a site. Although complex soil formation processes exist in alluvial valleys, some general approaches appear to provide a basis for developing Quaternary history and providing information to archaeologists for excavation strategies. Our investigations of alluvial systems have focused on the following: (i) identification of Quaternary or older surfaces; (ii) study of stream migration patterns; (iii) detailed morphological descriptions of soils on typical landscape units, such as flood plains, flood chutes, levees, and terraces; (iv) dating by 14C, diagnostic artifacts, and soil morphology; (v) laboratory analyses, e.g., particle-size distribution and elemental analysis; and (vi) using a team approach in the interpretation phase. Although each alluvial system requires individual analyses and interpretation, several model systems developed may aid in understanding alluvial soil-landscape sequences.
Evaluation of individual soil horizons and sequences of soil horizons in archaeological studies is critical to the correct and meaningful interpretation of archaeological context. We focus on the evaluation of soils in the stratigraphic framework of an archaeological site and offer a guide to assist in the interpretation of context of cultural materials in specific master horizons. In the North American Stratigraphic Code, the formal pedostratigraphic unit, the geosol, by definition requires being overlain by a formally defined lithostratigraphic or similar material unit. This criterion can rarely be met in shallow, mid-to-late Holocene settings. In addition, no subdivisions of the geosol are recognized, a problem at the scale of archaeological excavation. Chronostratigraphic and pedostratigraphic units are often confused in concept. The main distinction between these two, critical to archaeology, is in their boundaries and in the subdivision into smaller units. Boundaries of chronostratigraphic units are synchronous and form isochrons, whereas boundaries of pedostratigraphic units are time-transgressive. Subdivision of chronostratigraphic units results in subunits that represent shorter periods of time than the larger unit and that lie in temporal succession with each other (i.e., they follow the Law of Superposition). When a pedostratigraphic unit is subdivided, logically into soil horizons, the individual horizons are not separate from the whole soil, or from each other in a temporal framework. Each soil horizon has a unique set of properties and processes, and is separated in space, but not in time from adjacent horizons. The distinction between subdivision of chronostratigraphic and pedostratigraphic units is a fundamental difference between soils and sediments. The guide we present is based on pedogenic and geo-morphic processes, both past and contemporaneous, occurring in specific master horizons.
Pedology is a key component of arid lands archaeological research, providing information about landscape evolution, the stratigraphic context of artifacts, and environmental change. As an example, a pedologic study was conducted for archaeologists at the Fort Bliss Military Installation, which covers slightly >445 000 hectares of the northern Chihuahuan desert in southern New Mexico and western Texas. Part one of the study consisted of (i) mapping geomorphic surfaces and (ii) mapping deflated areas. Maps of geomorphic surfaces reveal the age and evolution of the arid landscape, and where archaeological sites might be buried. Deflation maps reveal areas where soils are deflated, stable, or recently buried. Artifact visibility is highest in deflated areas, but the stratigraphic integrity is generally lost. Areas buried by eolian sands, on the other hand, are most likely to contain artifacts with stratigraphic integrity, but artifact visibility is low. Part two of the study focused on paleoenvironmental changes. The most useful paleoenvironmental information was obtained from erosion-sedimentation history, fossil pollen preserved in buried soils, and δ13C and δ18O signatures in soil carbonates. Erosion, pollen, and isotopes all revealed a major period of desertification beginning at the middle Holocene about the time Paleoindians gave way to people of the Archaic period.
Soil studies in archaeological contexts in Mesoamerica have focussed on land use and paleoenvironmental reconstruction. I examine the role of pedology in archaeology in general, and then examine three case studies from the Maya lowlands, all associated with wetlands. Two sites, Cobweb Swamp and Pulltrouser Swamp, are located in perennial wetlands, while the third site, Nakbe, is located in an area of seasonal wetlands. The pedostratigraphic record was severely confounded by geochemical, anthrogenic, and shrink-swell phenomena. Pedological inference was used in an effort to reconstruct the evolution of the wetlands and the impact of the Maya on them. Initial investigations reveal marked changes in the hydrology and chemistry of these wetlands through time, due at least in part to changes induced by the Maya, but natural effects also are seen as important.
Moist to wet and cold or temperate climates foster the development of Spodosols. Archaeologists working in areas where Spodosols dominate contend with many similar themes. Problematic topics regarding the interpretation of archaeological sites in a region where Spodosols occur often include context, chronology, paleo-environmental reconstruction, disturbance, archaeological material preservation, and especially anthropogenic geochemistry. Project planning, associated geoarchaeological or pedoarchaeological research, and sediment sampling may require protocols unique to this soil order. A soil order level approach to interdisciplinary research is illustrated at several New England localities, Eddington Bend, Michaud, Hedden, and Nahanada, where archaeological sites are associated with Spodosols. An analogous soil order level genetic concept should be applicable to many other taxonomic categories.
The oxidizable carbon ratio (OCR), expressed as the ratio of total C by loss-on-ignition to readily oxidizable C by wet oxidation, challenges the presumed biological stability of carbonized organic matter. Previous studies demonstrate that charcoal is biologically recycled at a slow, but measurable rate, and that the rate of biochemical degradation of the carbonized organic matter varies within the specific physical and environ-mental contexts of the sample. The OCR-dating procedure determines an age for the C sample by use of a systems formula designed to account for the biological influences of O2, moisture, temperature, C concentration and the media's (soil) reactivity. These variables are measured by soil texture and depth below the soil surface, the site-specific mean annual temperature and rainfall, percentage of total C, and the soil pH. Residual influences on this system are included through a statistically derived constant. Using data from locations in North America and East Africa, a strong correlation [r = 0.98, standard error (SE) = 0.03] is demonstrated between the age estimates obtained by the OCR-dating procedure and 14C radiocarbon age estimates. Outliers define how river inundation and poor sample preparations adversely affect the results. This procedure is expanded to measuring the OCR of relatively stable residual organic matter, and to date the age of buried soil horizons. The OCR-dating procedure accounts for the interdependent dynamics of climate, biota, parent material, and time, providing an empirical test of Jenny's model for pedogenesis.
Geochemical studies have been traditionally undertaken at complex, multicomponent archaeological sites where chemical techniques are supplementary to more standard analyses. This study proposes geochemical procedures that have diagnostic potential for isolating activity areas at more typical single component sites. Two examples are selected from functionally distinct and ecologically diverse settings: a village site in Piedmont Georgia and a hunter-gatherer station in the Nebraska Sand Hills. It is demonstrated that trace element analysis of anthropogenic soils can infer patterning in the site formation record irrespective of natural and soil forming environments. Phosphorous emerges as the most accurate barometer of anthropogenic enrichment of subsurface occupations. A new method of presenting and interpreting phosphate fractionation results is proposed that isolates activity areas in diverse archaeological contexts.
The Huntsville site is a Caddoan (1250–1400 A.D.), civic-ceremonial center in northwest Arkansas. A trench was excavated in Mound A at the site revealing numerous construction episodes and the remains of several charnel houses, structures used for temporary interment of the dead. Levels of inorganic phosphates in mound sediments confirmed that the mound was human-made and that the site could be distinguished from its natural surroundings. While specific activities associated with particular stratigraphic units in the mound could not be determined, general interpretations of mound stratigraphic unit and feature functions were possible. The analysis of inorganic phosphate levels also was useful for delineating activity areas and feature boundaries on a living surface within the mound. The analysis suggested that bony material was deposited on this surface that represented a structure floor. Removal of the bodies from the charnel structure was not complete; small bones were probably inadvertently left behind. It is likely that ritual cleaning of the floor of the structure redeposited the small bones near the structure walls.
Heavy accumulations of Pb were noted in soil samples gathered within the boundaries of Hadrian's Villa, near Tivoli, Italy. Lead quantities of surface samples extracted with a dilute HC1-HNO3 solution ranged from 13 to 6750 mg kg−1. Samples representative of extractable Pb ranges were sequentially extracted to determine which soil fractions contained Pb accumulations. Extractants and corresponding fractions included MgCl2 (exchangeable), Na−OOCCH3 (carbonate), NH2OH−HCl (oxide), H2O2−HNO3 (organic), and HF−HNO3−HCL−H3BO4 (residual). All analyses were run using an inductively-coupled argon plasma atomic emission spectrophotometer (ICAP-AES). Total Pb quantities summed across fractions ranged from 150 to 12 000 mg kg−1. In samples having low total Pb quantities (150–500 mg kg−1), most Pb was found in the residual fraction. Within samples with total Pb quantities from 800 to 1000 mg kg−1, oxide and organic fractions held the largest percentage of Pb. The majority of Pb in the highest fractions was nearly equally divided between the carbonate and oxide fractions.