Roberts Island Research Articles

Intensification revisited: assessing resource specialization at Crystal River (8CI1) and Roberts Island (8CI41), Florida

ABSTRACT Archaeologists typically associate resource intensification with population expansion, environmental change, and political strategizing. Many Late Woodland and Mississippian societies of the Southeast eschewed dietary diversity in favor of harvesting fewer types of resources that could meet the subsistence demands of incipient aggregation. Foods such as maize and shellfish can provide humans with predictable caloric yields and are amenable to control by individuals or corporate groups. However, some archaeologists have identified scenarios in which small-scale societies intensified resources in the absence of population growth and social inequality. Ritual economies can periodically place high demands on the materials used for gatherings and ceremonies. These events then may leave material residues of economic intensification, which archaeologists might easily mistake as evidence for population expansion or social evolution. We use diversity and equitability estimates of zooarchaeological deposits from Crystal River (8CI1) and Roberts Island (8CI41), Florida, to demonstrate that some Woodland period societies periodically intensified their use of resources amidst population decline and heightened ritual activity. We suggest that the inhabitants of the area harvested shellfish at increasingly high rates to provide the material basis for a series of ritual interventions that aimed to circumvent the effects of rapid social and ecological change.

Exploring Oyster (Crassostrea virginica) Habitat Collection via Oxygen Isotope Geochemistry and its Implications for Ritual and Mound Construction at Crystal River and Roberts Island, Florida

ABSTRACTResearch at Crystal River and Roberts Island Shell Mound Complex, on the western coast of Florida, USA, offers a quantitative assessment of the temporality of shell deposit construction, Native subsistence practices, and mobility patterns through stable oxygen isotope data from eastern oyster (C. virginica). The δ18Owater values of oysters vary synchronously with salinity, assuming relatively constant δ18Owater/salinity gradients since the time of occupation, allowing for an examination of shifts in oyster habitat exploitation over time. Our previous (Thompson et al. 2015) study indicated that midden accumulation occurred throughout the year, while oysters from mound deposits were collected in colder months. New data indicate that in addition to differential season of collection, habitat exploitation also varied. During early occupation at the site, oysters were collected primarily from lower saline habitats, while in later phases oysters were obtained from higher salinity waters; we relate this to a lower sea level and concomitant settlement shift seaward. Additionally, oyster from later mound contexts was collected from higher saline habitats relative to those in midden contexts; Native people may have targeted specific bioherms at certain times for the year for feasting-related mound construction.

Constructing community at civic-ceremonial centers: pottery-making practices at Crystal River and Roberts Island

ABSTRACTCrystal River (8CI1) and Roberts Island (8CI36, 8CI40, 8CI41) are neighboring mound complexes on Florida’s west-central Gulf Coast, with mainly sequential occupations during the Middle and Late Woodland periods, respectively. Previous work at Crystal River produced assemblages marked by a diversity of pastes and surface treatments, suggestive of distinct communities of practice. However, these excavations were unsystematic and poorly controlled, thus confounding understanding of temporal and spatial variation in practice. Recent excavations in domestic areas, combined with the analysis of older collections from mounds, support a finer-grained understanding of variation in ceramic production. Our analysis suggests that communities of practice persisted through time, although there is variation that corresponds well with changes in settlement.

Evidence for Stepped Pyramids of Shell in the Woodland Period of Eastern North America

AbstractAntiquarians of the nineteenth century referred to the largest monumental constructions in eastern North America as pyramids, but this usage faded among archaeologists by the mid-twentieth century. Pauketat (2007) has reintroduced the term pyramid to describe the larger, Mississippian-period (A.D. 1050 to 1550) mounds of the interior of the continent, recognizing recent studies that demonstrate the complexity of their construction. Such recognition is lacking for earlier mounds and for those constructed of shell. We describe the recent identification of stepped pyramids of shell from the Roberts Island Complex, located on the central Gulf Coast of Florida and dating to the terminal Late Woodland period, A.D. 800 to 1050, thus recognizing the sophistication of monument construction in an earlier time frame, using a different construction material, and taking an alternative form.

Assessing village life and monument construction (cal. AD 65–1070) along the central Gulf Coast of Florida through stable isotope geochemistry

Given recent debates and reinterpretation of large-scale shell deposits as monuments rather than midden in the emerging dialogue in world archaeology regarding shell sites, we need better quantitative assessments of the temporality of shell deposit formation. Here we couple Bayesian analysis of radiocarbon dates with the results from our study of shell isotopes from C. virginica from the Crystal River and the Roberts Island Shell Mound Complex, neighboring and temporally overlapping shell mound sites on the central Gulf Coast of Florida. Linking these two lines of data is a new methodological approach which provides a more detailed understanding of the temporality of traditions. The results indicate that midden accumulation occurred throughout the year in the later phases of occupation at both sites. In contrast, oyster from mound deposits appears to indicate season of collection predominately during the colder months of the year. This contrast between mound and midden season of collection suggests that episodes of feasting and associated monument construction may have been of relatively limited temporal duration.

Radiocarbon and Luminescence Age Determinations on Mounds at Crystal River and Roberts Island, Florida, USA

Optically stimulated luminescence (OSL) has been employed successfully to determine the ages of palaeosols from earthen mounds in the southeastern USA, providing archaeologists with a means of dating monument construction in the absence of carbonaceous materials and geologists with a setting for understanding factors that can affect the luminescence intensity (i.e., burial dose) of soils. However, OSL dating has not been adequately tested on mounds whose principal component is sand, shell, or a combination of these two, despite the fact that monuments composed of such materials are common to the coasts and interior coastal plains of the region. Radiocarbon dating of bone collagen and soil‐carbon and OSL dating of quartz grains extracted from the fill of mounds at the Crystal River and Roberts Island sites on the west‐central coast of Florida, USA are used to determine the timing and history of mound construction at the sites. Comparison of OSL and radiocarbon ages on materials from the same or closely related contexts provides insight into factors influencing age determinations in mound fill deposits, particularly the type of construction material (sand or shell) and the manner in which these were deposited. The results contribute to the understanding of the temporal context of platform mound construction in southeastern USA.

The temporality of shell-bearing landscapes at Crystal River, Florida

We employ a landscape perspective to the shell middens at Crystal River (8CI1) and Roberts Island (8CI41), Woodland period (ca. 1000 BC to AD 1000) mound centers on the west-central coast of Florida in the American Southeast. Specifically, we adopt Ingold’s (1993: 162) understanding of landscape as the physical incorporation of social life, with all of its complexities of temporality and movement. Mapping, geophysical survey, and coring were used to document the location and scale of the contemporary and ancient landscapes. We followed this with small scale excavations to understand the form and timing of midden deposition. We employ Bayesian chronological modeling of radiocarbon dates from our investigations in the middens at Crystal River and Roberts Island to identify the broader rhythms of human activities. To characterize finer rhythms of social life within these phases, we compare rates of midden accumulation and other quantitative and qualitative measures of the distributions of artifacts and sediments. Our results indicate that the shell-bearing landscape at Crystal River and Roberts Island incorporates activities that fall in four broad phases over the interval from around AD 150 to 1050. These phases are characterized by diverse activities and temporalities, including both repetitive, small-scale refuse disposal and temporally discrete, larger-scale depositional episodes. Consistent with recent work on shell midden variability, both the archaeological deposits and the activities they encapsulate blur the lines between midden and monument.

ASPARAGUS CULTIVAR EVALUATION IN THE SACRAMENTO-SAN JOAQUIN DELTA REGION OF CALIFORNIA

For the past three seasons, data on asparagus marketable yield, spear number and average spear weight have been collected from two furrow-irrigated cultivar trials in California’s Sacramento-San Joaquin Delta Region. In addition, the first full-harvest-season data was taken on a new, drip irrigated asparagus cultivar trial. Asparagus establishment was accomplished with the use of one-year-old crowns and 10-week-old seedlings. International Asparagus Cultivar Evaluation Trial II, established in 1995 on Roberts Island near Stockton, California, contains 18 replicated cultivars and 17 single replication observation cultivars. Highest replicated trial marketable yield was attained by UC 157F1 one-year-old crowns at 5,154 kg/ha (kilograms per hectare), followed by 10-week-old, seedling-planted cultivars UC 157F1 (4,462 kg/ha), UCR 8 (3,907 kg/ha) and UCR 7 (3,899 kg/ha). All of these cultivars exhibited excellent spear quality (tight tips at 23 cm or greater spear length, spear straightness, uniform color). Spear weight at 23 cm trimmed length ranged from 22.3 g to 32.4 g. Best observation cultivar marketable yield came from Golia (4,621 kg/ha), followed by 89P72 (4,453 kg/ha), Eros (4, 263 kg/ha), Tainan #2 (4,031 kg/ha) and DA 911 (4,004 kg/ha). UCR 115, DA 911, 89P72 and Golia provided the best spear quality. Spear weight varied from 21.2 g to 32.7 g. Another asparagus cultivar evaluation trial on Victoria Island, in a field using subsurface drip irrigation, was harvested for the first complete season in 2000. Twelve replicated and eleven observation cultivars make up the trial that was established with one-year-old asparagus crowns. Greatest replicated trial marketable yield occurred with Atlas (9,176 kg/ha), followed by UC 157F1 (6,943 kg/ha), Grande (6,553 kg/ha) and UCR 115 (6,451 kg/ha). These same cultivars produced excellent spear quality; spear weight ranged from 27.9 g to 42.5 g. Best marketable yield in the observation cultivars came from Cipres (9,718 kg/ha), UCR 69 (9,187 kg/ha), UCR 122 (6,905 kg/ha) and PLA-2232 (6,662 kg/ha). UCR 69, UCR 122, UCR 96 and UCR 64 provided the best spear quality; spear weight varied from 22.3 g to 43.8 g. A final asparagus cultivar evaluation trial, established on King Island with one-year-old asparagus crowns in 1997, contains 7 replicated and 6 observation cultivars. Plagued by disease (Fusarium) and insect (garden symphylan) problems, this trial has never produced well. Best replicated production came from UCR 5 (2,987 kg/ha), UCR 4 (2,854 kg/ha) and UCR 10 (2,394) kg/ha. UC 157F1, UCR 5, UCR 7 and UCR 4 gave the best spear quality, while spear weight ranged from 17.8 g to 24.6 g. The best observation cultivar in yield and quality was F145 x A1 (3,149 kg/ha).

Hydrocarbon Development in the Urban Sector--the Stockton Gas Field, San Joaquin County, California

The Stockton gas field is located within the city limits of Stockton, California, in San Joaquin County on the east side of the Sacramento Valley. The field was discovered in 1983 by the Chevron USA, Stockton 1, R/D 1 well from Starky formation sands of upper Cretaceous age below 6,500 ft vertical depth. The entrapment is caused by low amplitude closure on a northeast trending structure. Gas production is from three wells in an enclosed site with cumulative production of 5.1 billion ft{sup 3} as of January 1990. Exploration interest in this area was the result of (1) regional subsurface mapping that indicated local thinning in shallow formations and pinchout of deeper sands; (2) detail review of seismic data; (3) nearby significant gas fields, Roberts Island and McDonald Island, caused by northeast trending anticlinal closure; and (4) early presence of produced gas in the area dating back to 1854 when water wells were completed from depths of 1,000-2,000 ft to augment the city's water and fuel needs.

A Delta-Slope-Submarine Fan Model for Maestrichtian Part of Great Valley Sequence, Sacramento and San Joaquin Basins, California

Shoaling of the Great Valley fore-arc basin, which began in the northern Sacramento basin during the Campanian Stage, shifted southward in the Maestrichtian to the southern Sacramento and northern San Joaquin basins, where a regressive sequence of submarine fan, slope, and deltaic deposits aggregates 6,500 ft (2 km) in thickness. Water depth decreased to sea level as ponded submarine fans aggraded the basin floor, and slope and deltaic systems prograded basinward over them. The central part of the fore-arc basin was filled by the end of the Cretaceous; the southern part was not filled until the middle Eocene. Deltas prograded westward and southwestward from the Sierran magmatic arc. The Starkey delta system evolved from cuspate/lobate wave-dominated deltas with extensive upward-coarsening delta-front sand beds to highly lobate fluvial-dominated deltas with moderately developed delta-plain facies. This system was succeeded by elongate fluvial-dominated deltas with thick delta-plain/fluvial sequences in the Bethel Island fluvial-deltaic system. Thin, fine-grained, black, glauconitic shale beds were deposited during intervals of regional transgression and deltaic abandonment. Several progradational-transgressive cycles are recorded. The Roberts Island slope system comprises several thick shale sections deposited during progradational events. Benthic foraminifera grade upward from abundant, varied, lower bathyal species to sparse upper bathyal or neritic species. Clinoform beds that offlap from east to west are evident in electric-log cross sections and seismic profiles. Lenticular, channeled sand beds, transverse to the slope, appear to be incised slope channels that funneled sand from the delta front to submarine fans on the basin floor. Shelf-edge slumps can locally be recognized by flow folds, mudstone blocks, and anomalous thicknesses and bedding attitudes, particularly near growth faults. The Union Island submarine fan system consists of several fans; most are elongate parallel to the southeasterly trend of the basin. Fans consist of upper, middle, and lower segments. Thick-bedded, channelized, upward-fining, coarse-grained sand beds characterize the upper fan. Medium to thick-bedded, channelized, massive fine to medium-grained sand beds predominate on the midfan. Lower fan deposits are not extensive but consist of symmetrical or upward-coarsening sequences of shale and thin to medium-bedded fine to very fine sand. Channeling is less common on the lower fan. Most fans have multiple lobes, each lobe having been fed by a different delta. Gas is produced from deltaic and submarine fan deposits. Delta-front sand beds form very small reservoirs where growth faults provide subtle structural traps. Distributary-channel sand beds are highly productive, owing mainly to stratigraphic pinch-outs but locally perhaps to structural traps. Submarine fan traps have so far been located mainly in midfan deposits, either in the structurally high suprafan mounds or along lateral fan margins near the base of the slope where fan deposits onlap and pinch out against slope shale. Exploration in lower and upper fan deposits has barely begun.

V.—On the Innes Wilson Collection of Rocks and Minerals from the South Shetland Islands and Trinity Island

Mr J. Innes Wilson of Port Stanley, Falkland Islands, during the whaling-season of 1916–17 paid a visit to the South Shetland Islands and Palmer Archipelago. He collected specimens of rocks and minerals from Deception Island, Roberts Island, Trinity Island, and the coast bordering the Gerlache Channel, which were transmitted to the Colonial Office by the Governor, Sir William Douglas Young. It is with the kind permission of the Undersecretary of State for the Colonies that I am allowed to submit some account of Mr Innes Wilson's interesting and important collection.