Early Pliocene volcanic ash rests on a polar desert pavement
Oligocene to lower Miocene San Gregorio Formation, Baja California Sur, Mexico. Diatom Research, 1(2), 169-187. Lohman, K., and G. Andrews. 1968. Late Eocene non-marine diatoms from the Beaver Divide area, Fremont County, Wyoming. (U.S. Geological Survey Professional Paper 593-E.) Washington, D.C.: U.S. Government Printing Office. Scherer, R. 1989. Microfossil assemblages in "deforming till" from Upstream B, West Antarctica: Implications for ice-stream flow models. Antarctic Journal of the U.S., 24(5).
Scherer, R., D. Harwood, S. Ishman, and P. Webb, 1988. Micropaleontological analyses of sediments from Crary Ice Rise. Antarctic Journal of the U.S., 23(5), 34-36. Schrader, H., and J. Fenner. 1976. Norwegian Sea Cenozoic diatom biostratigraphy and taxonomy. initial Reports of the Deep Sea Drilling Project, 38, 921-1,099.
Webb, P., D. Harwood, B. McKelvey, J. Mercer, and L. Stott. 1984. Cenozoic marine sedimentation and ice volume variation on the East Antarctic craton. Geology, 12, 287-291.
Early Pliocene volcanic ash rests on a polar desert pavement DAVID R. MARCHANT
Department of Geological Sciences
and
Institute for Quaternary Studies University of Maine Orono, Maine 04469-0110 DANIEL
R. Lux
Department of Geological Sciences University of Maine Orono, Maine 04469-0110 CARL C. SWISHER, III
Berkeley Geochronology Laboratory Institute of Human Origins Berkeley, California 94720 GEORGE
H. DENTON
Department of Geological Sciences
and
Institute for Quaternary Studies University of Maine Orono, Maine 04469-0110
Reconstructions of Pliocene climate based on the ecology of marine diatoms and Nothofagus wood of assumed Pliocene age within the Sirius formation suggest extensive ice-sheet collapse accompanied by warm (2-5° C) marine seas in the interior of East Antarctica (Harwood 1986; Webb et al. 1986) and the growth of Nothofagus in the adjacent Transantarctic Mountains. We report here an alternative climate reconstruction based on isotopically dated volcanic deposits that overlie in situ polar desert pavements. One such ash deposit of early Pliocene age occurs in Arena Valley (77°51'S 161°E), Quartermain Mountains, Antarctica, and is described below. A 30-centimeter-thick, light-gray to pale white, in situ volcanic ash surface deposit overlies a well-developed desert pavement at 1,500 meters elevation in central Arena Valley (figures 1 and 2). The buried pavement is composed of an 58
Figure 1. Photograph of volcanic ash in west central Arena Valley. The vertical face of the ash has been cut back to expose the underlying buried desert pavement. A highly weathered colluvial deposit (devoid of volcanic material) underlies the ash.
interlocking mosaic of closely spaced ventifacts of Ferrar Dolerite and Beacon Heights Orthoquartzite. The ventifacts are commonly pitted and exhibit thick coatings of desert varnish. The ash is overlain by a second desert pavement which is identical in form, composition, and texture to the buried pavement (figure 1). The age of the ash was determined by conventional argon40/argon-39 methods on bulk ash samples and by the laserfusion argon-40/argon-39 technique of dating single crystals. Qualitative X-ray microprobe analyses indicated that the crystal fraction included anorthoclase, aegerine, subcalcic augite, and magnetite. Anorthoclase was isolated using heavy liquids and a Franz magnetic separator. Approximately 1,000 grams of ash yielded about 1.0 grams of "pure" anorthoclase. Conventional argon-40/argon-39 incremental release heating of bulk samples of anorthoclase using a Nuclide 6-60-SGA 1.25 mass spectrometer yielded a plateau age of 4.69 ± 0.10 million years, although the release spectrum was saddle-shaped. Lo Bello et al. (1987) showed that such a release spectrum from volcanic feldspars suggests xenocrystic contamination of the sample. Therefore, the age determined by bulk analysis of hand-picked feldspars probably represents only a maximum age of the ash. To obtain a more accurate age for the ash, individual anorthoclase crystals were dated using the argon-40/argon-39 laserfusion technique. Results indicated that the ash was composed of at least two distinct populations. The younger population ANTARCTIC JOURNAL
F 4
I-, ;
Figure 2. Scanning electron microscope image of glass shards within the ash. Scale bar is 0.1 millimeter.
was by far the largest and yielded an age of 4.474 ± 0.032 million years. This is our most accurate date for the eruption and subsequent deposition of the ash. The polar desert pavement preserved beneath the early Pliocene ash-fall deposit has the following paleoclimatic implications: • It suggests that a dry, polar climate existed in Arena Valley prior to 4.474 ± 0.032 million years. Nothofagus cannot exist under such a climate (Sakai 1981). • It demonstrates that Arena Valley was not filled with ice from local valley glaciers or the east antarctic ice sheet at this time. • Because Arena Valley is a tributary to Taylor Valley, (which opens to McMurdo Sound), the volcanic ash suggests that tectonic uplift of Arena Valley is restricted to less than 1,500 meters within the last 4.474 million years. • It suggests that the general morphology of Arena Valley antedates 4.474 ± 0.032 million years. • It demonstrates that the massive ice sheet overridings postulated by Denton et al. (1984) antedate 4.474 ± 0.032 million years. 1989 REVIEW
We thank David P. West, Jr., at the University of Maine Geochronology Lab for initial dating of the ash from bulk anorthoclase samples. W.C. McIntosh provided figure 2. Martin Yates assisted in the microprobe analyses of the crystal com ponent of the ash, and S.C. Wilson assisted in the field. This work was supported by National Science Foundation grant number DPP 861-3842.
References Denton, G.H., M.L. Prentice, D.E. Kellogg, and T.B. Kellogg. 1984. Late Tertiary history of the Antarctic Ice Sheet: Evidence from the Dry Valleys. Geology, 12, 263-267. Harwood, D.M. 1986. Recycled siliceous microfossils from the Sirius Formation. Antarctic Journal of the U.S., 21(5), 101-103. Lo Bello, Ph., C. Feraud, G.M. Hall, D. York, P Lavina, and M. Bernat. 1987. 40Ar/39 Ar step-heating and laser fusion dating of Quaternary pumice from Neschers Massif, Central France: The defeat of xenocrystic contamination. Chemical Geology (isotope geoscience section), 66, 61-71.
59
Sakai, A. 1981. Freezing resistance of trees of the south temperate zone, especially subalpine species of Australia. Ecological Society of America, 62, 563-570.
Webb, P. N., D. M. Harwood, B. McKelvey, M.G.C. Mabin, and J.H. Mercer. 1986. Late Cenozoic tectonic and glacial history of the Transantarctic Mountains. Antarctic Journal of the U.S., 21(5), 99-100.
The origin of isolated gravel ripples in the western Asgard Range, Antarctica
Measurements made across ripples document the strongly asymmetric cross sections (figure). The actual crest is not a line but a diffuse area up to 20 centimeters wide. Two indices used to describe the form and shape of ripples were calculated. Average values of the ripple index (RI L/H) and the Ripple Symmetry Index (RSI = SL/LL) also appear in table 1. Average values for symmetry indexes range from slightly asymmetric to very asymmetric. Perfectly symmetric ripples have values of RSI = 1. The asymmetry is the opposite of most wind and water current ripples which typically have values of RSI > 2 (Tanner 1967). The surfaces of the gravel ripples consist primarily of weathered sandstone. Individual clasts commonly have thick quartz weathering rinds and a desert varnish composed of a reddish silicious crust (Weed and Ackert 1986). Clasts of sandstone and dolerite up to 15 centimeters in diameter commonly occur; some clasts are much larger. Frost cracks commonly occur on the lee sides of ripples. A concentration of well-sorted gravel up to 2 centimeters in diameter occurs on the lee side of the ripple crest on many ripples. A poor to fairly well-developed slip face sometimes occurs within this material on the lee side edge. The pavement is crudely sorted. The largest clasts commonly occur on the stoss side of the ripples and typically rest directly on the bedrock at the toe of the ripple. Other clasts on the stoss side are commonly setting or leaning on one another with no matrix material between them. The average clast size on the surface of the lee side decreases toward the heel of the ripple (figure). Lithologic, shape and textural data on samples of gravel collected from ripple surfaces appear in table 2. Excavations through the ripples show that the surface pavement overlies a thin, sandy, pebbly diamicton. A layer of rotted bedrock up to several centimeters thick commonly occurs between the bedrock and the overlying diamicton. Bedrock structures such as worm tubes and bedding are sometimes preserved within this layer. The layer pinches out at the toe and heel of the ripple. The ripples are generally less than 30 centimeters thick at the crest. Although excavations were situated to avoid visible frost cracks, sand wedge structures (Berg and Black 1966) occurred in many excavations (figure). The pebbles and small cobbles within the diamicton are similar in lithology, size, and surface texture to those on the surface of the ripple. The weathered clasts are supported by a matrix of sand and fine gravel. Individual sand grains have a reddish stain. The coarse sand and fine gravel is composed largely of dolerite grus. Six samples were analyzed to determine the grain-size distribution of the matrix material. The results appear in table 3. The samples are gravelly muddy sands and gravelly sands. The frequency distributions are bimodal. The primary peak includes at 1.5 4 is inherited from the Beacon Sandstones (Barrett 1972). The secondary peak occurs at -0.5 . The samples are well sorted and slightly fineskewed. There is virtually no mud in the samples.
ROBERT P. ACKERT, JR.
Department of Geological Sciences University of Maine Orono, Maine 04469
Fields of isolated gravel ripples occur throughout the uplands of the Dry Valleys. The ripples were reported in Denton et al. (1984) as part of a system of features indicative of subglacial sheet flow of meltwater beneath an ice sheet which overrode the Transantarctic Mountains. As part of a project designed to test the Denton et al. (1984) hypothesis of ice sheet overriding, detailed field studies were made in Njord Valley (77°36'S 1617E) in the western Asgard Range. Njord Valley is an ice-free, north-facing, hanging valley which overlooks the Dais in upper Wright Valley. The valley is eroded into sandstones of the Beacon Super Group; surrounding heights are capped by Ferrar Dolerites (McKelvey and Webb 1962). Within the valley, a complete set of features reported by Denton et al. (1984) are preserved. Among the features studied were several well-developed fields of isolated gravel ripples. Fieldwork was conducted during the 1983-1986 field seasons. Preliminary results are presented here. The ripples are conspicuous features due to their size and to the contrast between the reddish, weathered sandstone gravel composing the bulk of the ripples and the light, unvarnished sandstone bedrock exposed between them. From a distance, the asymmetric cross section of the ripples is readily apparent. The ripples have short, steep slopes facing up-valley and longer, gentler slopes on the down-valley sides. For purposes of discussion, the up-valley side is assumed to be the stoss side and the down-valley side the lee side. Although the fields of gravel ripples occur in topographic lows such as troughs or basins on valley floors, individual ripples tend to occur on local topographic highs. The figure shows a schematic cross-section of a typical gravel ripple. Table 1 summarizes data which describe the size and form of the ripples. The largest ripples are up to 0.5 meters high, 8 meters wide, and 100 meters long. The ripple crests are generally perpendicular to the valley axis, sinuous, and in a few cases bifurcated. The distance between crests is relatively constant within a given area. As the sinuosity decreases, the wavelength decreases. Typically, the width of the area of exposed bedrock between the ripples is several times greater than the width of the ripples. 60
ANTARCTIC JOURNAL
Scherer, R., D. Harwood, S. Ishman, and P. Webb, 1988. Micropaleontological analyses of sediments from Crary Ice Rise. Antarctic Journal of the U.S., 23(5), 34-36. Schrader, H., and J. Fenner. 1976. Norwegian Sea Cenozoic diatom biostratigraphy and taxonomy. initial Reports of the Deep Sea Drilling Project, 38, 921-1,099.
Webb, P., D. Harwood, B. McKelvey, J. Mercer, and L. Stott. 1984. Cenozoic marine sedimentation and ice volume variation on the East Antarctic craton. Geology, 12, 287-291.
Early Pliocene volcanic ash rests on a polar desert pavement DAVID R. MARCHANT
Department of Geological Sciences
and
Institute for Quaternary Studies University of Maine Orono, Maine 04469-0110 DANIEL
R. Lux
Department of Geological Sciences University of Maine Orono, Maine 04469-0110 CARL C. SWISHER, III
Berkeley Geochronology Laboratory Institute of Human Origins Berkeley, California 94720 GEORGE
H. DENTON
Department of Geological Sciences
and
Institute for Quaternary Studies University of Maine Orono, Maine 04469-0110
Reconstructions of Pliocene climate based on the ecology of marine diatoms and Nothofagus wood of assumed Pliocene age within the Sirius formation suggest extensive ice-sheet collapse accompanied by warm (2-5° C) marine seas in the interior of East Antarctica (Harwood 1986; Webb et al. 1986) and the growth of Nothofagus in the adjacent Transantarctic Mountains. We report here an alternative climate reconstruction based on isotopically dated volcanic deposits that overlie in situ polar desert pavements. One such ash deposit of early Pliocene age occurs in Arena Valley (77°51'S 161°E), Quartermain Mountains, Antarctica, and is described below. A 30-centimeter-thick, light-gray to pale white, in situ volcanic ash surface deposit overlies a well-developed desert pavement at 1,500 meters elevation in central Arena Valley (figures 1 and 2). The buried pavement is composed of an 58
Figure 1. Photograph of volcanic ash in west central Arena Valley. The vertical face of the ash has been cut back to expose the underlying buried desert pavement. A highly weathered colluvial deposit (devoid of volcanic material) underlies the ash.
interlocking mosaic of closely spaced ventifacts of Ferrar Dolerite and Beacon Heights Orthoquartzite. The ventifacts are commonly pitted and exhibit thick coatings of desert varnish. The ash is overlain by a second desert pavement which is identical in form, composition, and texture to the buried pavement (figure 1). The age of the ash was determined by conventional argon40/argon-39 methods on bulk ash samples and by the laserfusion argon-40/argon-39 technique of dating single crystals. Qualitative X-ray microprobe analyses indicated that the crystal fraction included anorthoclase, aegerine, subcalcic augite, and magnetite. Anorthoclase was isolated using heavy liquids and a Franz magnetic separator. Approximately 1,000 grams of ash yielded about 1.0 grams of "pure" anorthoclase. Conventional argon-40/argon-39 incremental release heating of bulk samples of anorthoclase using a Nuclide 6-60-SGA 1.25 mass spectrometer yielded a plateau age of 4.69 ± 0.10 million years, although the release spectrum was saddle-shaped. Lo Bello et al. (1987) showed that such a release spectrum from volcanic feldspars suggests xenocrystic contamination of the sample. Therefore, the age determined by bulk analysis of hand-picked feldspars probably represents only a maximum age of the ash. To obtain a more accurate age for the ash, individual anorthoclase crystals were dated using the argon-40/argon-39 laserfusion technique. Results indicated that the ash was composed of at least two distinct populations. The younger population ANTARCTIC JOURNAL
F 4
I-, ;
Figure 2. Scanning electron microscope image of glass shards within the ash. Scale bar is 0.1 millimeter.
was by far the largest and yielded an age of 4.474 ± 0.032 million years. This is our most accurate date for the eruption and subsequent deposition of the ash. The polar desert pavement preserved beneath the early Pliocene ash-fall deposit has the following paleoclimatic implications: • It suggests that a dry, polar climate existed in Arena Valley prior to 4.474 ± 0.032 million years. Nothofagus cannot exist under such a climate (Sakai 1981). • It demonstrates that Arena Valley was not filled with ice from local valley glaciers or the east antarctic ice sheet at this time. • Because Arena Valley is a tributary to Taylor Valley, (which opens to McMurdo Sound), the volcanic ash suggests that tectonic uplift of Arena Valley is restricted to less than 1,500 meters within the last 4.474 million years. • It suggests that the general morphology of Arena Valley antedates 4.474 ± 0.032 million years. • It demonstrates that the massive ice sheet overridings postulated by Denton et al. (1984) antedate 4.474 ± 0.032 million years. 1989 REVIEW
We thank David P. West, Jr., at the University of Maine Geochronology Lab for initial dating of the ash from bulk anorthoclase samples. W.C. McIntosh provided figure 2. Martin Yates assisted in the microprobe analyses of the crystal com ponent of the ash, and S.C. Wilson assisted in the field. This work was supported by National Science Foundation grant number DPP 861-3842.
References Denton, G.H., M.L. Prentice, D.E. Kellogg, and T.B. Kellogg. 1984. Late Tertiary history of the Antarctic Ice Sheet: Evidence from the Dry Valleys. Geology, 12, 263-267. Harwood, D.M. 1986. Recycled siliceous microfossils from the Sirius Formation. Antarctic Journal of the U.S., 21(5), 101-103. Lo Bello, Ph., C. Feraud, G.M. Hall, D. York, P Lavina, and M. Bernat. 1987. 40Ar/39 Ar step-heating and laser fusion dating of Quaternary pumice from Neschers Massif, Central France: The defeat of xenocrystic contamination. Chemical Geology (isotope geoscience section), 66, 61-71.
59
Sakai, A. 1981. Freezing resistance of trees of the south temperate zone, especially subalpine species of Australia. Ecological Society of America, 62, 563-570.
Webb, P. N., D. M. Harwood, B. McKelvey, M.G.C. Mabin, and J.H. Mercer. 1986. Late Cenozoic tectonic and glacial history of the Transantarctic Mountains. Antarctic Journal of the U.S., 21(5), 99-100.
The origin of isolated gravel ripples in the western Asgard Range, Antarctica
Measurements made across ripples document the strongly asymmetric cross sections (figure). The actual crest is not a line but a diffuse area up to 20 centimeters wide. Two indices used to describe the form and shape of ripples were calculated. Average values of the ripple index (RI L/H) and the Ripple Symmetry Index (RSI = SL/LL) also appear in table 1. Average values for symmetry indexes range from slightly asymmetric to very asymmetric. Perfectly symmetric ripples have values of RSI = 1. The asymmetry is the opposite of most wind and water current ripples which typically have values of RSI > 2 (Tanner 1967). The surfaces of the gravel ripples consist primarily of weathered sandstone. Individual clasts commonly have thick quartz weathering rinds and a desert varnish composed of a reddish silicious crust (Weed and Ackert 1986). Clasts of sandstone and dolerite up to 15 centimeters in diameter commonly occur; some clasts are much larger. Frost cracks commonly occur on the lee sides of ripples. A concentration of well-sorted gravel up to 2 centimeters in diameter occurs on the lee side of the ripple crest on many ripples. A poor to fairly well-developed slip face sometimes occurs within this material on the lee side edge. The pavement is crudely sorted. The largest clasts commonly occur on the stoss side of the ripples and typically rest directly on the bedrock at the toe of the ripple. Other clasts on the stoss side are commonly setting or leaning on one another with no matrix material between them. The average clast size on the surface of the lee side decreases toward the heel of the ripple (figure). Lithologic, shape and textural data on samples of gravel collected from ripple surfaces appear in table 2. Excavations through the ripples show that the surface pavement overlies a thin, sandy, pebbly diamicton. A layer of rotted bedrock up to several centimeters thick commonly occurs between the bedrock and the overlying diamicton. Bedrock structures such as worm tubes and bedding are sometimes preserved within this layer. The layer pinches out at the toe and heel of the ripple. The ripples are generally less than 30 centimeters thick at the crest. Although excavations were situated to avoid visible frost cracks, sand wedge structures (Berg and Black 1966) occurred in many excavations (figure). The pebbles and small cobbles within the diamicton are similar in lithology, size, and surface texture to those on the surface of the ripple. The weathered clasts are supported by a matrix of sand and fine gravel. Individual sand grains have a reddish stain. The coarse sand and fine gravel is composed largely of dolerite grus. Six samples were analyzed to determine the grain-size distribution of the matrix material. The results appear in table 3. The samples are gravelly muddy sands and gravelly sands. The frequency distributions are bimodal. The primary peak includes at 1.5 4 is inherited from the Beacon Sandstones (Barrett 1972). The secondary peak occurs at -0.5 . The samples are well sorted and slightly fineskewed. There is virtually no mud in the samples.
ROBERT P. ACKERT, JR.
Department of Geological Sciences University of Maine Orono, Maine 04469
Fields of isolated gravel ripples occur throughout the uplands of the Dry Valleys. The ripples were reported in Denton et al. (1984) as part of a system of features indicative of subglacial sheet flow of meltwater beneath an ice sheet which overrode the Transantarctic Mountains. As part of a project designed to test the Denton et al. (1984) hypothesis of ice sheet overriding, detailed field studies were made in Njord Valley (77°36'S 1617E) in the western Asgard Range. Njord Valley is an ice-free, north-facing, hanging valley which overlooks the Dais in upper Wright Valley. The valley is eroded into sandstones of the Beacon Super Group; surrounding heights are capped by Ferrar Dolerites (McKelvey and Webb 1962). Within the valley, a complete set of features reported by Denton et al. (1984) are preserved. Among the features studied were several well-developed fields of isolated gravel ripples. Fieldwork was conducted during the 1983-1986 field seasons. Preliminary results are presented here. The ripples are conspicuous features due to their size and to the contrast between the reddish, weathered sandstone gravel composing the bulk of the ripples and the light, unvarnished sandstone bedrock exposed between them. From a distance, the asymmetric cross section of the ripples is readily apparent. The ripples have short, steep slopes facing up-valley and longer, gentler slopes on the down-valley sides. For purposes of discussion, the up-valley side is assumed to be the stoss side and the down-valley side the lee side. Although the fields of gravel ripples occur in topographic lows such as troughs or basins on valley floors, individual ripples tend to occur on local topographic highs. The figure shows a schematic cross-section of a typical gravel ripple. Table 1 summarizes data which describe the size and form of the ripples. The largest ripples are up to 0.5 meters high, 8 meters wide, and 100 meters long. The ripple crests are generally perpendicular to the valley axis, sinuous, and in a few cases bifurcated. The distance between crests is relatively constant within a given area. As the sinuosity decreases, the wavelength decreases. Typically, the width of the area of exposed bedrock between the ripples is several times greater than the width of the ripples. 60
ANTARCTIC JOURNAL
