We present a detailed geoarchaeological study of landscape processes that affected prehistoric formation and modern preservation of archaeological sites in three areas of the Colorado River corridor in Grand Canyon, Arizona, USA. The methods used in this case study can be applied to any locality containing unaltered, non-pedogenic sediments and, thus, are particularly relevant to geoarchaeology in arid regions. Resolving the interaction of fluvial, aeolian, and local runoff processes in an arid-land river corridor is important because the archaeological record in arid lands tends to be concentrated along river corridors. This study uses sedimentary structures and particle-size distributions to interpret landscape processes; these methods are commonplace in sedimentology but prove also to be valuable, though less utilized, in geoarchaeology and geomorphology. In this bedrock canyon, the proportion of fluvial sediment generally decreases with distance away from the river as aeolian, slope-wash, colluvial, and debris-flow sediments become more dominant. We describe a new facies consisting of 'flood couplets' that include a lower, fine-grained fluvial component and an upper, coarser, unit that reflects subaerial reworking at the land surface between flood events. Grain-size distributions of strata that lack original sedimentary structures are useful within this river corridor to distinguish aeolian deposits from finer-grained fluvial deposits that pre-date the influence of the upstream Glen Canyon Dam on the Colorado River. Identification of past geomorphic settings is critical for understanding the history and preservation of archaeologically significant areas, and for determining the sensitivity of archaeological sites to dam operations. Most archaeological sites in the areas studied were formed on fluvial deposits, with aeolian deposition acting as an important preservation agent during the past millennium. Therefore, the absence of sediment-rich floods in this regulated river, which formerly deposited large fluvial sandbars from which aeolian sediment was derived, has substantially altered processes by which the prehistoric, inhabited landscape formed, and has also reduced the preservation potential of many significant cultural sites.

This report analyzes various depositional environments in three archaeologically significant areas of the Colorado River corridor in Grand Canyon. Archaeological features are built on and buried by fluvial, aeolian, and locally derived sediment, representing a complex interaction between geologic and cultural history. These analyses provide a basis for determining the potential influence of Glen Canyon Dam operations on selected archaeological sites and thus for guiding dam operations in order to facilitate preservation of cultural resources. This report presents initial results of a joint effort between geologists and archaeologists to evaluate the significance of various depositional processes and environments in the prehistoric formation and modern preservation of archaeological sites along the Colorado River corridor in Grand Canyon National Park. Stratigraphic investigations of the Palisades, Lower Comanche, and Arroyo Grande areas of Grand Canyon yield detailed information regarding the sedimentary history at these locations. Reconstruction of past depositional settings is critical to a thorough understanding of the geomorphic and stratigraphic evolution of these three archaeologically significant areas. This examination of past sedimentary environments allows the relative significance of fluvial, aeolian, debris-fan, and slope-wash sedimentary deposits to be identified at each site. In general the proportion of fluvial sediment (number and thickness of flood deposits) is shown to decrease away from the river, and locally derived sediment becomes more significant. Flood sequences often occur as 'couplets' that contain a fluvial deposit overlain by an interflood unit that reflects reworking of fluvial sediment at the land surface by wind and local runoff. Archaeological features are built on and buried by sediment of various depositional environments, implying a complex interaction between geologic and cultural history. Such field analysis, which combines geological and archaeological information and techniques, can provide a basis for future determination of the effects of Glen Canyon Dam operations on selected areas of the river corridor. This knowledge is essential to the development of preservation strategies for cultural resources in Grand Canyon.

Cross-bedded oolitic grainstones in the Ste. Genevieve Limestone (Mississippian) of the Illinois basin have generally been considered to be shallow marine. However, fine- to medium-grained cross-bedded grainstones of mixed clast type in the Ste. Genevieve of Harrison County, southern Indiana, are here interpreted to be of eolian dune origin on the basis of small-scale sedimentary structures, particularly climbing-wind-ripple structures. In addition, subaerial exposure of surfaces at the tops and bases of the eolian units is indicated by pedogenic features such as in-situ breccias and rhizoliths. Associated skeletal and oolitic grainstones of marine origin are distinguished from the eolian grainstones by the presence of pebble-sized fossils. The presence of several interva s of eolian deposits in the Ste. Genevieve is probably a result of eustatic sea level fluctuations.

The Merced Formation consists of approximately 2,000 m of shallow marine and coastal nonmarine sediment of late Cenozoic age that accumulated in a structural trough south of the city of San Francisco. About 1,750 m of these deposits crop out in a well exposed tilted sequence in sea cliffs on the north side of the San Andreas fault. The part of the section north of the fault appears to be of Pleistocene age. Depositional facies within the Merced Formation can be identified with a high degree of confidence based on biota, sedimentary structures, and textural character. In vertical succession, the facies define alternating episodes of transgression and regression. The individual transgressive/regressive cycles almost certainly reflect the eustatic sea-level fluctuations that occurred during Pleistocene time. The general pattern of transgression and regression can be fairly well matched in much of the section to a Pleistocene sea-level curve determined from oxygen-isotope data. If eustatic fluctuations can be subtracted from the stratigraphic record, tectonic and sedimentologic influences become evident. The accumulation of such a thick succession of shallow-marine and coastal deposits implies long-term subsidence at an average rate of 1 m/103 yr. Evidence suggests that the rate of subsidence may have been greater in the section deposited prior to emplacement of the Rockland ash (at 0.4 Ma) than in the section deposited subsequent to the ash. A pronounced shift to predominantly nonmarine deposits about 290 m below the top of the section may be due in part to diminished subsidence rates. However, the fact that this change coincides with a change in provenance from local Franciscan sources to the Sacramento-San Joaquin River systems indicates that the shift to nonmarine facies more likely resulted from increased rates of sedimentation.

AbstractThe Merced Formation and consanguineous superjacent strata are well exposed in seacliffs that extend 4.3 mi (7 km) south from Lake Merced to the trace of the San Andreas fault (Fig. 1). The exposures can be reached by walking south along the beach from oceanfront public parking areas west of Lake Merced or via several well-defined trails that lead to the beach from Fort Funston (Fig. 1), a part of the Golden Gate Recreational Area, which provides public parking. The southern part of the exposure can be reached by walking north from a public parking area at Mussel Rock (Fig.1), which can be reached via an access road from Westline Drive in northern Pacifica (Edgemar). The central part of the exposure is presently accessible via a road to the former Thornton Beach State Park at the western end of Alemany Boulevard. This road, however, is now badly disrupted by landslides and affords uncertain future accessibility.The character of the exposure changes as a consequence of local landsliding and seasonal variation in the level of beach sand. The deposits are best viewed following winter storms when the erosion of beach sand provides for fresh, wave-washed exposure. The observer of these deposits should continually be alert forrock falls and the possibility of being isolated by arising tide.

Dunes that are morphologically of linear type, many of which are probably of longitudinal type in a morphodynamic sense, are common in modern deserts, but their deposits are rarely identified in aeolian sandstones. One reason for non-recognition of such dunes is that they can migrate laterally when they are not exactly parallel to the long-term sand-transport direction, thereby depositing cross-strata that have unimodal cross-bed dip directions and consequently resemble deposits of transverse dunes. Dune-parallel components of sand transport can be recognized in ancient aeolian sands by examining compound cross- bedding formed by small dunes that migrated across the lee slopes of large dunes and documenting that the small dunes migrated with a component in a preferred along-crest direction over the large dunes.

Abstract This field-trip guidebook discusses the Merced Formation (of Pliocene and Pleistocene age) in its type section (Lawson, 1893) and associated Pleistocene beds in the same sea-cliff outcrops. The outcrops extend from Mussel Rock on the south to Fort Funston on the north (Fig. 1A). This section is notable for its thickness and excellence of exposure and for the wide variety of shallow marine and coastal depositional environments represented. Although the section contains a potentially important record of changes in relative sea level, the record has not yet been fully interpreted because of a dearth of precisely dated horizons. On this field trip we will proceed stratigraphically upward through the section, starting at Woods Gulch, proceeding northward past Thornton Beach State Park (presently closed because of landsliding and gullying), and ending at the northwest corner of the old Fort Funston military reservation, now part of Golden Gate National Seashore (Fig. 1A). We will not see the lower part of the section between Mussel Rock and Woods Gulch. Although the exposures are often excellent where newly eroded by ocean waves, they tend to disappear rapidly because of landsliding and sand deposition on the beach. Because of the rapid changes, all the features described in this guidebook cannot be expected to be seen at any one time. The exposures are usually best at the base of the sea cliffs in winter, when much of the beach sand has been removed by erosion. The Merced Formation and associated Pleistocene beds crop out in a belt that trends northwest-southeast for a distance of 25 km across the northern San Francisco Peninsula (Fig. IB).

This chapter examines the interpretation of cyclic crossbedding with an example from the Navajo sandstone. Cyclic crossbedding of the fluctuating-flow type can be produced by periodic fluctuations in flow direction, flow velocity, depth of flow in aqueous environments, or other parameters affecting sediment transport rates and depositional mechanisms. The cycles may or may not be bounded by erosional surfaces. In the absence of any erosional surfaces, the cycles are concordant and are made evident by variations in texture or sedimentary structure. Fluctuating-flow cycles bounded by surfaces of erosion generally imply major changes in flow direction or, in aqueous flows, changes in water depth. Erosion surfaces produced by reversals of the normal-to-bedform component of flow tend to be best developed on the upper lee slope and to become nonerosional hiatal surfaces downward. Superimposed-bedform cycles differ considerably, depending on whether the superimposed bedforms exist on both the stoss and lee slopes of the main bedform or whether they exist only on the stoss slope, moving up it until they reach the crest, whereupon their sand avalanches down the slipface of the main bedform.

This chapter discusses the reconstruction of bedform assemblages from compound crossbedding. Sedimentologists have observed that small bedforms are commonly superimposed on large bedforms, and they have realized that many complicated cosets of crossbeds in eolian and subaqueous sandstones can be readily explained by the migration of small bedforms on the lee surfaces of other large bedforms. The critical factor that determines whether the crossbedding deposited by a bedform is simple or compound is the absence of or the occurrence of intermittent erosion on the lee slope. It is found that where no erosion occurs, beds deposited on the lee slope are not truncated, bounding surfaces are not generated within the set, and crossbedding is simple. In contrast, where parts of a lee slope occasionally undergo erosion, layers on the lee slope are truncated, bounding surfaces are produced within the set, and the resulting crossbedding is compound. The key to reconstructing bedform geometry from compound crossbedding is visualizing the structures generated when hypothetical lee slopes are translated through space and later exposed in variably oriented outcrop planes.

Most of the the basic types of stratification that occur in modern dune sands have been identified in Pennsylvanian to Jurassic crossbedded sandstones of the western interior United States that are generally considered eolian The most common types are sandflow cross stratification formed by the avalanching of sand down slipfaces subcritically climbing translatent stratification formed by wind ripples and grainfall lamination formed by the settling out of grains in zones of flow separation A common type of stratification formed by poorly understood processes on damp or ponded interdune flats is characterized by irregular small scale waviness Identification of the basic types of stratification in a sand body has several uses Eolian and subaqueous sands may be distinguished by certain types of stratification especially by the types occurring in climbing ripple structures More sophisticated paleocurrent determinations can be made when thetype of stratification is known Knowledge of thetype of stratification can be used to estimate the initial porosity such porosity estimates help in the analysis of sand compaction and pre lithification deformation The occurrence and distribution of the basic types of stratification are useful clues in interpreting dune size dune morphology dune orientation relative to the wind direction and thereasons for the typically low dip angles of eolian cross strata in the lower parts of sets

Where dunes migrate during deposition, they move upward (climb) with respect to the generalized depositional surface. Sediment deposited on each lee slope and not eroded during passage of a following trough is left behind as a cross-stratified End_Page 982------------------------------ bed. Because sediment is thus transferred from dunes to underlying strata, bed forms must decrease in cross-sectional area or in number, or both, unless sediment lost from dunes during deposition is replaced with sediment transported from outside the depositional area. Using equations that relate the amount of sediment lost by dunes to the amount gained by sets of cross-strata, we calculate that the dunes which deposited the De Chelly (Permian), Navajo (Triassic? and Jurassic), and Entrada (Jurassic) Sandstones had mean heights with lower and upper limits of 16 and 450 m, respectively. Although these calculated dune heights are surprisingly large, two kinds of field observations support the hypothesis that the dunes that deposited 10-m thick sets of cross-strata common in eolian sandstones may have been 100 m high or higher--comparable in size to large modern dunes or draas. First, many sets of eolian cross-strata are primarily bottomset beds; they rarely contain dune-crest deposits or convex-upward cross-strata deposited on lupper lee slopes, and s me sets contain sand-flow toes that pinch out near the top of the set. Second, the sand-flow layers in these eolian sandstones are thicker and laterally more extensive than in modern 10-m high dunes. The hypothesis that large dunes deposit relatively thin sets of cross-strata explains the absence in the geologic record of sets of eolian cross-strata comparable in thickness to the height of large modern dunes. End_of_Article - Last_Page 983------------