No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
The Ryan/Harley site: Sedimentology of an inundated Paleoindian site in north Florida
Abstract
The Ryan/Harley site (Florida Master Site File Number: 8Je-1004) is a Middle Paleoindian habitation site containing Suwannee points. Based on stratigraphic correlation and diagnostic artifact seriation, Suwannee-age sites have been relatively dated from ˜10,900 14C yr B.P. to ˜10,500 14C yr B.P. Clovis-like traits on the Suwannee points and other stone tools from the Ryan/Harley site suggest it dates to the earlier end of the Suwannee timeframe. The currently inundated site is partially buried beneath a sediment column located in a swamp forest and partially exposed in a side channel section of the Wacissa River, Jefferson County, Florida. Research done prior to this analysis determined that the artifact assemblage appeared to be unsorted and was contained in a midden-like unit. Our purpose here is to assess the issue of site integrity further. Unconsolidated sediment samples collected from the artifact-bearing horizon and from horizons immediately above and below the artifact horizon were analyzed using granulometric techniques. Arithmetic probability plots of the grain-size distributions show that the sediments were transported and deposited by fluvial processes. Thus, the Suwannee points and associated artifacts, and faunal remains appear to have accumulated during a time of subaerial exposure perhaps after a regional water-table decline, and have remained largely or essentially intact, with little or no postdepositional reworking. The artifacts and faunal remains recovered from the artifact- bearing horizon at Ryan/Harley are distributed randomly, showing no sign of sorting. In the fossil suite, two articulated white-tailed deer (Odocoileus virginianus) vertebra were recovered in situ. The unsorted nature of artifacts and articulated faunal remains that are contained within the fluvially deposited sediments suggests the Suwannee point level of the Ryan/Harley site has remained undisturbed since original deposition. © 2006 Wiley Periodicals, Inc.
... ephemeral encampment along a riverbank or levee (Balsillie et al., 2006;Dunbar et al., 2006). The archaeological assemblage from Ryan-Harley contains debitage as well as unifacial and bifacial stone tools, including Suwannee points. ...
... Ryan-Harley contains six stratigraphic units (Figure 2; Balsillie et al., 2006). The stratigraphy described here is based on field observations from profiles exposed in 2015 and 2017, four ground-penetrating transects and six cores. ...
... At this time, extensive eolian activity deposited widespread dunes on inland fluvial and marine terraces (such as the Pamlico Terrace in Figure 3) across the southeastern United States that correlate with the expansion of continental ice sheets, the shrinking of the Gulf of Mexico, and the onset of overall arid conditions (Ivester & Leigh, 2003;Ivester et al., 2001;Leigh, 1998;Leigh, 2008;Markewich et al., 2015;Otvos & Price, 2001). (Balsillie et al., 2006). This determination was likely a result of the Gaussian analyses used during the previous study, which assume a roughly normal distribution of sediment sizes; a condition not met by eolian sediments, which are inherently skewed toward fine clasts. ...
Ryan‐Harley (8JE1004) in the Wacissa River of North Florida is one of few sites in the Eastern United States that contains an unfluted lanceolate point component (Suwannee) in a buried context. Extinct and extirpated faunal remains found within the Suwannee artifact layer at Ryan‐Harley have been cited as evidence of a Pleistocene faunal refugium in the Southeastern United States. In 2015, geoarchaeological excavations were undertaken to determine if the Ryan‐Harley Suwannee assemblage is culturally discrete and associated with Pleistocene fauna. Geoarchaeological data demonstrate that the archaeological component was exposed until ~8,000 calbp on a relict eolian landscape before burial. No radiocarbon ages were obtained on either extinct or extant faunal material from the site, and the excavated portion of the geologic stratum containing the artifact assemblage is devoid of other datable organics. Therefore, no unequivocal conclusion can be reached about the association of the extinct faunal material with the Suwannee artifacts. However, multiple lines of evidence fail to reject the hypothesis that the Ryan‐Harley lithic assemblage represents a single occupation of Suwannee point makers. Multiple lines of evidence suggest that the Suwannee occupation of Ryan‐Harley most likely occurred during the Younger Dryas (~12,900–11,700 calbp).
... Archaeological excavations were undertaken at the Ryan-Harley site in 2015 and 2017 to address this limitation. The Ryan-Harley site in northern Florida is the only known professionally excavated site with an intact Suwannee artifact horizon (Balsillie, Means, and Dunbar 2006;Dunbar et al. 2006;Smith 2020). Geoarchaeological studies were conducted prior to excavation, which support the hypothesis that Ryan-Harley contains a discrete Suwannee component (Smith 2020). ...
... At these sites, the contiguity makes it difficult to discern which curated tools are associated with the late Paleoindian period and which tools are associated with the Early Archaic period. The Ryan Harley site is the only known Suwannee site where points and curated tools have been recovered from sealed contexts (Balsillie, Means, and Dunbar 2006;Dunbar et al. 2005; Smith forthcoming). To date, there are no radiometric ages from the site, but a publication describing the assemblage is forthcoming (M. ...
In this article, we review evidence for the initial presence, later expansion, and subsequent settling-in of first Floridians during times when climate change and sea level rise decreased the amount of habitable land. We present projectile-point and formal-tool sequences and estimated chronologies that describe Florida’s: (1) pre-Clovis presence (exploration); (2) Clovis presence focused on river channels, springs, chert resources, and possibly megafauna (colonization); (3) continuation and proliferation of Clovis-related, but post-megafauna late Paleoindian lanceolate point makers that remained focused on river channels, springs, and chert (expansion); (4) transition to side- and corner-notched points and a plethora of formal tools, along with significant population increase and landscape use occurring away from waterways (settlement); and (5) possible population decline or abandonment, or both, by 10,000 calendar years ago or soon thereafter.
... j The occupation zone contains Suwannee artifacts (∼10,900 to ∼10,500 14 C years BP), but, based on "Clovis-like traits," it is considered early Suwannee (Balsillie, Means, and Dunbar 2006). k This isolated, small (1.84-mm-diameter, 0.61-mm thick, 0.38-mm-diameter hole) disc was discovered inadvertently, adhering to sand grains that in turn were stuck to flakes collected from this large surface site exposed in a dune blowout. ...
Beads are the only direct indicator of body adornment from Upper Paleolithic and Paleoindian contexts, although not common. Geoarchaeological coring at the Mockingbird Gap (Clovis) site, New Mexico, resulted in the recovery of a small tubular bead of Paleoindian age. The bead was found in alluvial sand 9.20 m below the surface of Chupadera Draw, an intermittent drainage that borders the site. The ornament is made of calcium carbonate and is the only known Paleoindian bead of this material in North America. All known tubular beads of this age are bone. Other beads are made of shell or teeth. The bead probably was made by using a rotary drill to modify a segment of a rhizolith, which is a cemented calcium carbonate cylinder representing root molds. Beads could easily be made from these cylinders although they would be relatively fragile, likely accounting for their rarity. The bead has a minimum age of ∼10,550 14C years BP. This suggests a Folsom association. Scattered Folsom sites are known within 1 km of the site. The Clovis occupation is more extensive and intensive, however, and within 120 m of the findspot.
Archaeological data have demonstrated that modern Florida was occupied by at least 14,550 years ago, but evidence of Late Pleistocene and Early Holocene peoples (ca. 14,600–8,000 cal B.P.) is limited to far inland and upland settings, as more than half of Florida’s peninsula was drowned between ca. 21,000–6,000 cal B.P. Rising aquifer levels of the Late Pleistocene allowed some interior sites to preserve within forming river channels, especially some springfed sinkholes that became the Aucilla River of northwest Florida. Terrestrial sites are poorly preserved in comparison, containing stone tools in mixed and/or undateable stratigraphy. Geospatial analysis of the 92 early sites in the Aucilla basin demonstrates that the underwater sites are crucial to provide a more robust understanding of early people, as the earliest sites are found only underwater, and the preponderance of the multicomponent sites also are inundated.
The karstic Aucilla River of Northwest Florida is renowned for its well‐preserved late Pleistocene cultural material and Rancholabrean fauna. Much of this material was recovered by avocational SCUBA divers from displaced contexts in mid‐channel sinkholes, but underwater excavations into sediment banks on sink margins have demonstrated that faunal material and early artifacts can be recovered in situ from inundated terrestrial strata that contain dateable organics and microfossils useful for paleoenvironmental reconstructions. Underwater environmental data document the transition of Aucilla River localities from isolated spring‐fed ponds into an inter‐connected fluvial system due to rising sea levels and climate amelioration with several major periods of sinkhole infilling during the late Quaternary. Late Pleistocene components on land are largely absent or are undateable; these components, located in a subtropical cypress swamp in shallow clay‐rich soils, tend to be less well‐preserved, but still contain a rich record of human use on the landscape. The karstic Aucilla River of Northwest Florida is renowned for its well‐preserved late Pleistocene cultural material and Rancholabrean fauna. Underwater excavations into sediment banks on sink margins have demonstrated that faunal material and early artifacts can be recovered in situ from inundated terrestrial strata that contain dateable organics and microfossils useful for paleoenvironmental reconstructions. Underwater environmental data document the transition of Aucilla River localities from isolated spring‐fed ponds into an inter‐connected fluvial system due to rising sea levels and climate amelioration with several major periods of sinkhole infilling during the late Quaternary.
These essays cast new light on Paleoindians, the first settlers of North America. Recent research strongly suggests that big-game hunting was but one of the subsistence strategies the first humans in the New World employed and that they also relied on foraging and fishing. Written in an accessible, engaging style, these essays examine how migratory waterfowl routes may represent one impetus for human migration into the Americas, analyze settlement and subsistence in the major regions of the United States, and reinvestigate mammoth and bison bone beds in the western Plains and the Rocky Mountains to illuminate the unique nature of Paleoindian hunting in that region. The first study of Paleoindian subsistence on a continental scale, this collection posits regional models of subsistence and mobility that take into account the constraints and opportunities for resource exploitation within each region: Research on the Gault site in Texas reveals new subsistence strategies there, while data from the Shawnee-Minisink site in Pennsylvania connects seed collecting with fishing in that region, and plant remains from Dust Cave in Alabama provide important information about subsistence. With research ranging from fauna and lithic data from Paleoindian campsites in Florida that illuminate subsistence technologies and late megamammals to an analysis of plant remains from the eastern United States that results in a revised scheme of environmental changes, this volume serves as an important sourcebook and guide to the latest research on the first humans in North America.
Despite the discovery of Paleoindian sites east of the Mississippi River, the stratigraphic and temporal establishment of the Early Paleoindian Clovis and Middle Paleoindian Folsom sequence west of the Mississippi River has been a standard by which other Paleoindian sites in the Americas have been judged. This is particularly true of Clovis sites in the Desert Southwest because they are viewed as evidence of the founding human populations, at least regionally, and because western archaeologists were first to establish a regional chronostratigraphy and geoclimatic model of the Southwestern US (Bryan and Albritton, 1943; Bryan, 1950; Antevs, 1954, 1962a,b; Haynes, 1971, 1982, 1984, 1990, 1991, 1993; Haynes et al., 1999). Following Kirk Bryan and Ernst Antevs, C. Vance Haynes has dedicated much of his professional career fine-tuning the geoclimatic evidence gathered from Paleoindian sites in the Southwest. These researchers were the first to develop a regionally based chronology that placed stratigraphy and environmental data in time. The resulting chronostratigraphy and geoclimatic data have played a vital role in the interpretation of the Southwestern Paleoindian cultural development. In short, it is the strength of the Southwestern approach that has been lacking elsewhere in the Americas. However, it would be inappropriate to adopt the regionalized data from the Southwest in order to impose it in a different geographic region; for example the Southeast US. The patterns of climate, habitat, and human occupation should be evaluated on their own merits keeping in mind the likelihood of distinct regional differences. It is from this perspective that the Paleoindian components at the Page-Ladson site are discussed.
Comprehensive, high-resolution, composite sea-level curves for the U.S. Gulf of Mexico since the last glacial maximum have been developed based on all available radiocarbon and calibrated absolute age-data. They are based on sea-level elevation indicators that, on the average, were measured once every 60 years for the past 20,000 years. The data sets consist primarily of geological sea-level indicators (some are archaeological). Published sea-level histories of the Gulf of Mexico exhibit significant variability. While there is error associated with the 14C age dating methodology, the bulk of error is undoubtedly associated with the indicator material chosen to represent sea-level elevation. It is the latter that must be judicially treated. Such error has, perhaps, been inflated to such an argumentative and defeatist extent among researchers that comprehensive compilation and analysis of sea-level data for the Gulf, until now, has been avoided.
The objective of this investigation was to analyze all of the available sea-level data for the northern Gulf of Mexico, and to assess associated error and select data using three data editing procedures (one geological, the other two statistical) in order to identify a sea-level curve attaining an accuracy of least equivocal status. (1) We selected data for the Gulf of Mexico exhibiting tectonic and/or crustal stability, which yielded 353 radiocarbon-dated sea-level indicator data points. (2) We addressed the problem of identifying “spurious” sea-level data outliers that can be justifiably excused from inclusion in analytical procedures. This is not, in fact, a problem isolated to Gulf of Mexico data, but is normally the case for most data sets as can be easily verified by inspecting the comprehensive world-wide national and regional sea- level compilation of Pirazzoli (1991). Utilizing the eustatic data of Siddall et al, (2003), a statistically-based method has been proposed that might be considered by other researchers as a useful tool for post-initial editing of sea-level data. We found that only a few spurious data points can significantly affect analytical outcomes (only 12 spurious outliers were identified, or but 3.4% of the Gulf of Mexico sea-level indicators). (3) Once spurious “error” was eliminated, a sufficient amount of data remained (341 dated sea-level indicators) for which there was some considerable associated variability. We treated these remaining assembled data using a seven- point floating averaging method. By smoothing some of the noise, the moving average method mitigated the degree of probable associated variability, while allowing longer-term probable trends to remain; on the average seven dated points encompassed a period of 400 years with each floating point average representing a 60-year period.
In addition, we investigated the controversial subject of sea-level history younger than about 6,000 years (mid- to late-Holocene), and identified two “younger data sets” based on sampling location bias. One younger data set can be defined by sea-level indicators collected seaward of the present shoreline (younger data set A), the other by sea-level indicators collected landward of the present shoreline (younger data set B). Location relative to current sea level was assessed based on physical location in conjunction with elevation of sampling. By definition, sea-level indicators sampled seaward of current sea-level do not define high- stands. In addition, a detailed treatment of littoral processes associated with physiographic features (beach ridges, cheniers, and storm ridges) has been presented, indicating favor in the case of younger data set B. Both younger data sets are presented for scientific scrutiny.
As a consequence, the comprehensive compilation of northern Gulf of Mexico sea-level analytical results has significance beyond the local region. Gulf of Mexico data compare favorably with a recent late Quaternary sea-level data set from the Red Sea (Siddall et al., 2003), a high-resolution index of eustatic sea-level. Given its geologic stability throughout the late Quaternary (in terms of data selected) and its relatively low-energy environments, the northern Gulf of Mexico might be expected to have experienced near-eustatic sea-level
conditions, and therefore offers a detailed record of global sea-level. In particular, the persistent evidence of mid- to late-Holocene high-stands in the Gulf of Mexico may be among the best global verifications of such events.
Two-dimensional plotting tools can be of invaluable assistance in analytical scientific pursuits, and have been widely used in the analysis and interpretation of sedimentologic data. We consider, in this work, the use of arithmetic probability paper (APP). Most statistical computer applications do not allow for the generation of APP plots, because of apparent intractable nonlinearity of the percentile (or probability) axis of the plot. We have solved this problem by identifying an equation(s) for determining plotting positions of Gaussian percentiles (or probabilities), so that APP plots can easily be computer generated. An EXCEL example is presented, and a programmed, simple-to-use EXCEL application template is hereby made publicly available, whereby a complete granulometric analysis including data listing, moment measure calculations, and frequency and cumulative APP plots, is automatically produced.
The graphic portrayal of quantitative information has deep roots. These roots reach into histories of thematic cartography, statistical graphics, and data visualization, which are intertwined with each other. They also connect with the rise of statistical thinking up through the 19th century, and developments in technology into the 20th century. From above ground, we can see the current fruit; we must look below to see its pedigree and germination. There certainly have been many new things in the world of visualization; but unless you know its history, everything might seem novel.
ROOTH1 proposed that the Younger Dryas cold episode, which chilled the North Atlantic region from 11,000 to 10,000 yr BP, was initiated by a diversion of meltwater from the Mississippi drainage to the St Lawrence drainage system. The link between these events is postulated to be a turnoff, during the Younger Dryas cold episode, of the North Atlantic's conveyor-belt circulation system which currently supplies an enormous amount of heat to the atmosphere over the North Atlantic region2. This turnoff is attributed to a reduction in surface-water salinity, and hence also in density, of the waters in the region where North Atlantic Deep Water (NADW) now forms. Here we present oxygen isotope and accelerator radiocarbon measurements on planktonic foraminifera from Orca Basin core EN32-PC4 which reveal a significant reduction in meltwater flow through the Mississippi River to the Gulf of Mexico from about 11,200 to 10,000 radiocarbon years ago. This finding is consistent with the record for Lake Agassiz which indicates that the meltwater from the southwestern margin of the Laurentide Ice Sheet was diverted to the northern Atlantic Ocean through the St Lawrence valley during the interval from ~11,000 to 10,000 years before present (yr BP).
The history of meltwater flow from the Laurentide Ice Sheet to the Gulf of Mexico during the last deglaciation, which holds possible implications for the cause of the Younger Dryas cold episode, is not well understood. We propose a new chronology based on using the percentage of reworked calcareous nannofossils in Orca Basin sediments as a proxy for erosion. The period of greatest meltwater flow to the gulf was between 12.7 and 12.1 ka (during the Bølling warm interval), and flow remained high until the beginning of the Younger Dryas cold episode at 11.3 ka; this corresponds to meltwater pulse IA. A sharp meltwater decrease at 12.2 ka may represent the Older Dryas glacial readvance. Little or no meltwater is inferred to have flowed to the gulf from ˜10 to 9 ka, which is the time of the second major meltwater pulse, IB. Therefore, meltwater must have been permanently diverted away from the Gulf of Mexico at the beginning of the Younger Dryas.
Principles and terminology for classification of the Quaternary are discussed, including lithostratigraphy, biostratigraphy. morphostratigraphy, climatostratigraphy and chronostratigraphy. The main conclusion is a proposal for a common chronostratigraphical classification of the Quaternary in Norden (and partly continental NW Europe). The Quaternary is subdivided into the Pleistocene and the Holocene Series. The Pleistocene is further subdivided into several provisional stages (Weichselian, Eemian, etc.), based on the sequence of glacials/interglacials. but with the boundaries preferably defined by stratotypes. The Late Weichselian and the Flandrian (Holocene) are subdivided into chronozoncs (Bolling, Older Dryas, Allerød, Younger Dryas, Preboreal, Boreal, Atlantic, Subboreal, Subatlantic) with the boundaries dcfined in conventional radiocarbon years.
Much of the world's surface, even under the oceans, is covered in thick deposits of sedimentary particles - gravel, sand, silt and clay. The nature of the deposits and their formation is very much dependent on the distribution of particles of different sizes. However, different instruments measure different attributes of a particle's size, based on how fast a particle settles in water, or the surface area of a particle, or its length. This book provides information on the how and why of particle size analysis in terms of understanding these sediment deposits.
Much of the world's surface, even under the oceans, is covered in thick deposits of sedimentary particles - gravel, sand, silt and clay. The nature of the deposits and their formation is very much dependent on the distribution of particles of different sizes. However, different instruments measure different attributes of a particle's size, based on how fast a particle settles in water, or the surface area of a particle, or its length. This book provides information on the how and why of particle size analysis in terms of understanding these sediment deposits.
Three methods of using suite statistics are given here: 1) the variability diagram, based on whole sample means and standard deviations, 2) the tail-of-fines diagram, based on fine-tail percentages, and 3) the segment-analysis triangle, based on the probability plots. Each of these methods has a hydrodynamic basis and a useful historical basis. The combination of all three provides for much better interpretation than can be obtained from individual samples. -from Author
In former works TANNER (1998) discussed seven red flags, and BALSILLIE and TANNER (2000) listed an additional six red flags, where by "red flags" it was meant common uncertainties or errors in coastal applications. Though the list is still by no means complete, an additional seven items are discussed here. They are: (14) Granulometric statistics-an evaluation problem. (15) Graphic measures versus moment measures. (16) The mean equals the median? (17) Composite versus suite statistics. (18) Sieving versus settling. (19) Particle size classification scales. (20) Carbonate granulometry.
In a former treatment Tanner (1998) listed seven red flags. By 'red flags' it was meant common uncertainties or errors in coastal work. It was not thought at that time that the list was complete, but in fact selected from a longer list compiled over the years. We here present six additional items as follows: (8) Wave data description and definition. (9) Is the wave period really conserved? (10) The significant wave height - putting it in perspective. (11) Where or when does shore-breaking occur? (12) Reassessment of wave energy content. (13) Misuse of tidal datum reference planes. Again, this does not constitute an exhaustive list. Moreover, editorials by their very nature are generalized. Certain issues are probably well deserved of greater detail and justification; some for which future in depth treatments are underway. But, one must start somewhere.
Briefly reviews selected localities in the US that have provided an opportunity for rigorous stratigraphic radiocarbon control, showing large mammal extinctions were completed by 10 500 BP. There is stratigraphic coincidence of the first visible evidence of hunters (Clovis culture) and the last skeletons of late Pleistocene megafauna. Considerable stress may also have been caused to large mammals by the loss of streams and watering holes in the early part of the interglacial. -T.M.Kennard
Suite and composite methodologies, two statistically valid approaches for producing statistical descriptive measures, are investigated for sample groups representing a probability distribution where, in addition, each sample is a probability distribution. Suite and composite means (first moment measures) are always equivalent. Composite standard deviations (second moment measures) are always larger than suite standard deviations. Suite and composite values for higher moment measures have more complex relationships. Very seldom, however, are they equivalent, and they normally yield statistically significant but different results. Multiple samples are preferable to single samples (including composites) because they permit the investigator to examine sample-to-sample variability. These and other relationships for suite and composite probability distribution analyses are investigated and reported using granulometric data.
Submicrometer variations in the median diameter of siliciclastic mud in
Orca Basin, northern Gulf of Mexico, sensitively monitor Laurentide ice
sheet runoff to the gulf during early stages of the last deglaciation.
Grain-size data document a prominent pulse of very fine-grained sediment
to the northern Gulf from 12.6 to 12.0 ka, corresponding with the
well-known sea-surface meltwater spike in δ18O records.
Meltwater megafloods at this time eroded and transported large volumes
of clay to the Gulf via the Mississippi River system. The nonquartzose
megaflood sediments appear to have been derived from clay-rich
midcontinental terrace deposits rather than from glacial outwash
supplied by the Laurentide ice margin. Sediment grain-size data, in
combination with other terrigenous proxies (clay minerals and reworked
calcareous nannofossils), indicate substantial variability in deglacial
meltwater discharge and associated sediment load of meltwater runoff.
This variability reflects changes in continental plumbing consistent
with terrestrial records of North American deglaciation.
230Th/234U and 14C dating of Barbados
corals has extended the calibration of 14C years B.P. to
calendar years B.P. beyond the 9200 year tree ring series (Bard et al.,
1990). This now permits the conversion of 14C chronozones,
which delimit major climate shifts in western Europe, to calendar years.
The Younger Dryas chronozone, defined as 11,000 to 10,000 14C
years B.P., corresponds to 13,000 to 11,700 calendar years B.P. This
calibration affects the interpretation of an intensely studied example
of the "Younger Dryas climate event," the δ18O anomaly
between 1785 and 1793 m in Dye 3 ice core. The end of the
δ18O anomaly in Dye 3 ice core has been dated by
measurements of 14C in air bubbles (Andree et al., 1984,
1986) and by annual layer counting (Hammer et al., 1986). The older
14C dates fall out of the range of the tree ring calibration
series but can now be calibrated to calendar years using the Barbados
230Th/234U calibration. The
14Ccorrected age for the end of the
δ18O event is 10,300 ± 400 calendar years B.P.
compared to the annual layer counting age of 10,720 ± 150 years
B.P. Thus, the "Younger Dryas" event in the Dye 3 ice core ends in the
Preboreal chronozone (11,700 to 10,000 calendar years B.P.) and is not
correlative with the end of the Younger Dryas event identified in pollen
records marking European vegetation changes. The end of the Dye 3
δ18O event is, however, correlative with the end of
meltwater pulse IB (Fairbanks, 1989), marking a period of intense
deglaciation with meltwater discharge rates exceeding 13,000
km³/yr.
The remit of the INTIMATE2 project of the INQUA Palaeoclimate Commission is to synthesise marine, terrestrial and ice-core data for the North Atlantic region during the Last Glacial–Interglacial Transition (LGIT: ca 13–1014C kyr BP; ca 15–11.5 ice-core kyr BP). A major problem, however, is the difficulty of effecting correlations at a temporal resolution that are adequate for defining ‘leads’ and ‘lags’ between the polar ice, terrestrial, marine, and atmospheric realms. The limitations of the dating and correlation methods currently employed are summarised, and new quality assurance protocols are proposed. These include recommendations on the contextual information that should accompany radiocarbon dates, procedures for radiocarbon calibration, the use of an event-stratigraphic approach in inter-regional correlations, and the more widespread use of time-parallel marker horizons (based on tephra layers, oxygen isotope stratigraphy, palaeomagnetic stratigraphy, and radiocarbon ‘wiggle-matching’) to underpin the geochronology and correlation of events during the LGIT. These protocols will be adopted by the INTIMATE project in future international, collaborative research and are recommended to other groups working on this important time period.
Four objectives of quantitative study of sediments are outlined, each of which requires specially collected samples. The corresponding techniques of collection are here called engineering sampling, descriptive sampling, environmental sampling, and correlation sampling, according to the purpose of the samples. The steps in the systematic collection of samples for the first three purposes are outlined. Certain principles underlying efficient correlation sampling are considered. There are two techniques for environmental sampling depending on the kind of measurements to be made. Criteria are given for determining whether or not composite samples may be used. When composite samples are unsuitable, samples from individual layers are required. In the latter case, the sedimentation-unit concept furnishes a practical field tool for the rational selection of the layer or combination of layers to be sampled as a unit. An example of the use of the sedimentation unit, showing how the boundaries of the individual units are determined, is given in detail. Six general steps for the determination of a sedimentation unit are given. Devices for obtaining the samples are described.
Parameters of sedimentary size distributions are most reproducible in repeated measurements by the same investigator using the same methods. In such determinations, means should be reproducible to within 0.05 phi and standard deviations to within 0.05 phi by careful work. Separate measurements by different techniques and investigators should yield means reproducible to within 0.25 phi and standard deviations to within 0.2 phi. Skewnesses and kurtoses exhibit somewhat larger percentage deviations than do those of mean and standard deviation. The reproducibility of the various parameters is satisfactory for all current geologic investigations providing the measurements are made with reasonable care.
Otto's (1938) concept of the sedimentation unit and Apfel's (1938) definition of a phase are considered; both authors select velocity as the diagnostic parameter, and they relate the bedded structure of the unit to chance variations in velocity of the transporting current. Flume experiments on deposition of artificial units show that: A well-defined bedded structure can develop under steady-state conditions of flow (velocity) and sediment transport; velocity fluctuations represent but one of several important processes responsible for sorting and segregation of grain sizes into a bedded structure; a microchange in regimen may cause deposition of a group of laminae rather than a single lamina. The sedimentation unit must be interpreted, therefore, within wide tolerance limits.
The method of first and second differences permits, in certain instances, the detachment of major components from zig-zag curves. Because of the overlap between components, and other factors, the method is commonly subjective in part and therefore not rigorous. Three applications are shown: two dealing with modern shoreline marsh samples and one dealing with an Eocene sand. Recombination of the detached components provides curves which, generally, deviate from the original less than two percent. After components have been obtained, it is still not possible to identify the site of deposition of the sample. Additional applications are known from geomorphology, stratigraphy, river hydrology, and other branches of earth science. Because the results in geomorphic problems can be verified i the field, it is thought that the analysis of sediment samples is also meaningful.
Distinction is made between continuous and discontinuous variables in frequency distributions. Histograms, frequency curves, and cumulative curves are discussed in relation to each other. On the basis of these relations an attempt is made to explain why histograms vary with the grade scale used in mechanical analysis, while cumulative curves do not. The disadvantages of histograms may be avoided by a simple graphic method for obtaining a continuous frequency curve independent of the grade scale. This curve is obtained by graphic differentiation of the cumulative curve, and is expressed in terms of the Wentworth descriptive grade scale. The analytic scale for obtaining the data is, however, shown to he independent of the descriptive scale units. It is suggested that in analyzing sediments, as many points as necessary be determined along the cumulative curve, to reduce errors of smoothing to a minimum.
Isotopic, micropaleontologic and pollen analyses of deep-sea cores from the Bay of Biscay and the northeastern Atlantic Ocean show that the deglacial warming of this oceanic area was closely correlated with the paleoclimatic evolution of the adjacent European continent. Temperatures were at least as warm as those of today in the Bay of Biscay between 13 300 and 11 000 B.P. coinciding with the combined Bølling/Allerød warm continental events. A major spread of polar water occurred between 11 000 and 10 000 B.P. During this event which coincides with the Younger Dryas continental cold event, marine temperatures were almost as low as those of the last glacial maximum. The final deglacial warming of the norteastern Atlatntic Ocean occurred during the following 3000 yr.
Indian mounds, near Careyville, Florida, about 2.0 m high, are located on hillsides and hilltops 10 to 20 m above the floodplain of the nearest river (Choctawhatchee). Each mound is composed largely of quartz sand, with a scattering of artefacts and stream pebbles (not in layers), but with no visible bedding.Probability plots showed 25 Gaussian distributions, 18 having the ‘dune hump’, three having the ‘surf break’ and nine being doubly-truncated or having other patterns of unknown or uncertain origin. The surf breaks probably were inherited from pre-Pleistocene marine terraces in the area. The pebbles and the sand were not introduced by the same agency. The sand probability plots, taken as a set, indicate an eolian origin. The rough symmetry of the mounds, and the lack of cross-bedding, argue against a migrating dune origin.On a variability plot (showing the variability of the means versus the variability of the standard deviations), one suite of samples fell clearly within the ‘dune’ number field, a second suite in the overlap area between ‘dune’ and ‘beach’, and a third suite, taken immediately adjacent to a creek bed, plotted in the overlap area between ‘beach’ and ‘coastal plain stream’.The pebbles, of common Southern Appalachian types, are attributed to the activities of the inhabitants, perhaps children. The sand is thought to have been carried by the wind, perhaps from nearby river sand bars, or from areas burned either by lightning-set wildfires or as part of “slash-and-burn” agriculture. The mounds are thought to represent clearings (for huts), and hence good trapping devices for wind-borne sand.
The Preboreal oscillation (PBO) has been attributed to increased meltwater, but the source of the meltwater and causative mechanism of the PBO has remained elusive. Here we attribute the source to a massive meltwater discharge event from an abrupt drainage of glacial Lake Agassiz, Canada, via the Mackenzie River into the Arctic Ocean. A maximum volume of 21,000 km3 was discharged over a 1.5-3 yr period with a peakdischarge of 0.500 Sverdrups (Sv), equivalent to a 6 m rise in the Arctic Ocean (or 0.062 m rise in global sea level). The flood occurred at about 11,335 cal yr BP, and was followed by a B0.042 Sv flow until 10,750 cal yr BP when the southern outlet of Lake Agassiz reopened and diverted drainage to the Mississippi River system. We estimate that only 2-4% of the flood water would have frozen into sea ice within the Beaufort region, but coupled with increased river ice production during winter, and thicker pack ice growth throughout the Arctic Ocean, a thicker, longer lasting and more extensive packice may have been flushed through Fram Strait. The thicker and more extensive packice, and freshened sea surface, may have triggered the PBO by increasing albedo, and generating a low salinity anomaly upon melting in the North Atlantic, thus decreasing the formation of North Atlantic Deep Water. r 2002 Elsevier Science Ltd. All rights reserved.
Use of the phi notation for describing particle-size distributions of sediments is briefly reviewed, beginning with its introduction by Krumbein (1934). Its uses and implications in dimensional analysis are discussed. The author agrees with McManus (1963) in suggesting more careful use of phi as a number and as a symbol.
THE cause of the Younger Dryas cold event, which interrupted the last deglaciation, is still a matter of debate1. A prevalent hypothesis, proposed by Broecker et al.2 is that the abrupt climate change was driven by a decrease in the rate of North Atlantic Deep Water production, triggered by a sudden dilution of North Atlantic surface water in response to the diversion of Laurentide icesheet melt water from the Mississippi drainage system to that of the St Lawrence river. Here we investigate the feasibility of this triggering mechanism by reconstructing sea-surface temperature, salinity and sea-ice cover records for the outlet of the Gulf of St Lawrence into the North Atlantic Ocean. These reconstructions—based on dinoflagellate-cyst assemblages3,4 in sediment cores from the region—show reduced meltwater runoff, low temperatures and extensive sea-ice cover during the Younger Dryas, dated here between 10,800 and 10,300 BP. Meltwater pulses did occur before and after the Younger Dryas event: as early as 11,700 BP, during the development of the Champlain Sea in the St Lawrence Lowland, and afterwards, until 10,100 BP. At the resolution of our salinity proxy (0.7‰), the meltwater pulses preceding the Younger Dryas did not affect sea-surface salinity off the shelf break. These constraint on the meltwater outflow through the St Lawrence drainage system do not support the triggering mechanism of the Broecker et al. hypothesis2, unless the North Atlantic thermohaline circulation is more sensitive to small salinity changes than most models suggest23.
Oxygen isotopic analyses of individual Orbulina universa from Orca Basin
core EN32-PC6 document the presence of low salinity surface waters in
the northern Gulf of Mexico over the past 16 kyr. Isotopic data from an
interval immediately following the Younger Dryas Event indicate the
rapid decrease in δ18O values at the conclusion of the
Younger Dryas was due to a year-round return of meltwater to the Gulf of
Mexico. Data indicate periodic or seasonal low-salinity waters existed
over the region of the Orca Basin prior to the initiation of the
meltwater spike. Estimates suggest O. universa grew its shell in
salinities at least 4.5 ‰ below ambient. Since O. universa may
have calcified deep in the mixed layer during periods of low salinity,
surface salinities could have been even lower. Comparison of the average
of individual O. universa oxygen isotopic values with data from multiple
shell samples of white Gs. ruber from the same core samples demonstrates
that the two species record similar values during the late Holocene. In
contrast, O. universa records lower oxygen isotopic values during the
late glacial/deglacial intervals, possibly due to differences in
seasonal distribution or shell ontogeny between the two species.
An enigmatic circular pit uncovered during archaeological excavations at the Clovis type site, Blackwater Draw, New Mexico, in 1964 has been reexposed and posited as a water well excavated by Clovis people around 11,500 B.C. The prehistoric well, the oldest in the New World, was probably a dry hole. Other Clovis wells may exist in the area. The excavation of wells near where there had been surface water shortly before adds to the evidence for drought during the Pleistocene-Holocene transition. © 1999 John Wiley & Sons, Inc.
This paper comprises a review of the many graphical and mathematical techniques that have been proposed for the statistical summary of grain-size data. Satellitic problems, such as laboratory techniques, choice of size scales, and interpretation, are only considered briefly.
Grain-size distributions of sand samples by means of a Benthos Rapid Sediment Analyzer were determined, using SCHLEE'S (1966) fall times. The results were compared with analysis of the same samples by sieving. The Rapid Sediment Analyzer (RSA) consistently overestimated the mean diameter of fine samples and underestimated the diameter of coarse samples relative to sieving. Since we used a wider settling column and a smaller sample weight than did Schlee, we infer that the differences were due to weaker grain interaction effects in our analyzer. Thus, by using Schlee's fall times, we overcompensated for these effects. We conclude that the “new grain parameter” of fall velocity (SENGUPTA and VEENSTRA, 1968) is indeed valuable. MIDDLETON'S (1967) psi transform is more convenient than the analogous phi transform in that it yields Middle-ton sand classes of 0 to 5 psi, as compared to Wentworth-Krumbein sand classes of -1 to 4 phi. However, use of fall velocity does not completely obviate the need to determine grain diameter, since this is the most expedient criterion with which to correct fall velocity to standard fall velocity.
Orbitally induced increase in northern summer insolation after growth of a large ice sheet triggered deglaciation and associated global warming. Ice-albedo, sea-level, and greenhouse-gas feedbacks, together with tropical warming from weak-ening winds in response to polar amplification of warming, caused regional-to-global (near-) synchronization of deglaciation. Effects were larger at orbital rather than millennial frequencies because ice sheets and carbon dioxide vary slowly. Ice-sheet–linked changes in freshwater delivery to the North Atlantic, and possibly free oscillations in the climate system, forced millennial climate oscillations associated with changes in North Atlantic deep water (NADW) flow. The North Atlantic typically operates in one of three modes: modern, glacial, and Heinrich. Deglaciation occurred from a glacial-mode ocean that, in com-parison to modern, had shallower depth of penetration of NADW formed further south, causing strong northern cooling and the widespread cold, dry, and windy conditions associated with the glacial maximum and the cold phases of the mil-lennial Dansgaard-Oeschger oscillations. The glacial mode was punctuated by meltwater-forced Heinrich conditions that caused only small additional cooling at high northern latitudes, but greatly reduced the formation of NADW and triggered an oceanic "seesaw" that warmed some high-latitude southern regions centered in the South Atlantic.
Some important papers have apparently gone unnoticed by most sedimentologists, as shown by their absence from bibliographies of recent texts. These papers concern sample size, permissible number of splits, sieving time, and sieve-vs-settling tube comparisons. These papers were published where sedimentologists would not ordinarily see them, but should be required reading for students.