Biotic provinces in modern Amundsen Sea sediments: Implications for ...
Biotic provinces in modern Amundsen Sea sediments: Implications for glacial history D.E. KELLOGG and T.B. KELLOGG
PcEsNSL4N: Institute for Quaternary Studies
and Department of Geological Sciences University of Maine Orono, Maine 04469
1
EASTERN
ICE NSULA 5HEL PEN coscRove
In January 1985, the U.S. Coast Guard icebreaker Glacier penetrated the heavy pack ice of the outer Amundsen Sea. We took advantage of this unprecedented opportunity to collect 20 piston cores from the Amundsen Sea continental shelf and Pine Island Bay (Kellogg, Kellogg, and Hughes 1985). These new cores supplement existing coverage of the outer continental shelf of the Amundsen Sea (Anderson and Myers 1981) with material from the eastern side of the Amundsen embayment (figure 1). Cores were collected for the purpose of microfossil and sedimentary analyses that would provide data bearing on the glacial history of the region. Microfossil distributions in core-top samples define three regions or provinces in sediments of the eastern Amundsen Sea (Kellogg and Kellogg in preparation). These provinces are defined best by foraminiferal and diatom distributions, and less clearly by radiolarian data (figure 1). Outer shelf province. The presence of abundant well-preserved calcareous benthic foraminifera suggests that bottom waters on the outer shelf are well oxygenated and that dissolution is not significant here at typical depths between 400 and 463 meters. Low marine-diatom abundances were recorded in core-top sediments from this area. While consistent with our findings further west on the outer shelf (Kellogg, Kellogg, and Anderson 1982), these findings are inconsistent with the high diatom abundances we found in ice and water samples from this area. Diatomaceous material was observed staining the bottom of numerous ice floes in this region in January of 1985. Possibly our core tops represent older material outcropping at the surface. If this hypothesis is correct, the presence of the stratigraphicindicator diatom species Nitzschia curta suggests that deposition ceased during the Quaternary. Another alternative is that an ice shelf, or multi-year sea ice, thicker than ice found in the region today, covered the region until very recently, preventing deposition of the photosynthetic diatoms. The ice-shelf hypothesis appears unlikely because no source for shelf ice, at a time when the eastern margin province was apparently the site of a seasonal polynya (discussed below), has been identified and because there are no known shallows or pinning points for an ice shelf in this area. Our preferred interpretation is that currents may be sufficiently strong on the outer shelf to prevent deposition of diatoms. Eastern margin province. Abundances of marine diatoms in eastern margin province sediments are much higher than in adjacent regions to the north and south. These high abundances suggest that surface waters in this area were highly productive during the time represented by deposition of the top 1-2 centimeters of sediment used in our analyses. Satellite data suggest that this area is relatively ice-free during austral sum154
ICEBERG
Co
ToNGur
Is
Is
0 is-Is I
LQ
Figure 1. Map of the Amundsen Sea showing core locations, localities mentioned in the text, bathymetry, and boundaries of biotic provinces. Bathymetry (contours in meters) for the shelf break region is based on data from Anderson and Myers (1981) and Kellogg (unpublished), and on U.S. Naval Oceanographic Office (1971): contours dashed to indicate uncertainties in locations of soundings. Bathymetry south of 720s shows only 500-meter contour, separating major troughs and shallows (dashed were estimated). Boundaries of biotic provinces (heavy dashed lines) are uncertain beyond area where cores were collected. ("km" denotes "kilometer.")
mers in some years (Zwally et al. 1983). We conclude that the Amundsen Sea polynya has been a persistent feature of the eastern margin area throughout the period of deposition of our core-top samples. Our analyses of the cores from this region shows that the uppermost layer is a soft, diatomaceous, sandy mud which averages 4-6 centimeters in thickness. This layer probably represents Recent sedimentation at a high deposition rate, resulting from the proximity of numerous glaciers. If sedimentation rates are high (greater than 0.005 centimeter per year), the uppermost diatom-bearing layer may result from deposition related to the presence of the polyna during only the last 1,000 to 2,000 years. The presence of predominantly arenaceous benthic foraminifera in the eastern margin province results from the fact that most of our cores were taken from deep (perhaps glacially scoured) troughs in water depths exceeding 500 meters. At the site of grab B175, where the depth was only 293 meters, calcareous benthic foraminifera predominated. We conclude that the calcium carbonate compensation depth in this area must lie between 300 and 500 meters. Pine Island Bay province. Pine Island Bay core tops (and cores) have very low microfossil abundances, usually consisting of only a few specimens per sample, yet we recovered over 2 meters of sediment at each location. We believe this nearly microfossil-barren sediment was deposited beneath a formerly more-extensive Pine Island Glacier. We know this is true for core DF85-107 because tricamera aerial photography in 1966 (AnonyANTARCTIC JOURNAL
January 1985 105
- - - - January 1973
PINE ISLAND 106
1966
SAY -
1050 10 20 4/ 1
km
106 0 W
102 0 W
Aq
98° W
Figure 2. Map of Pine Island Bay region showing changing positions of the calving margin of Pine Island Glacier. The 1966 position, from Anonymous (1968a), is based on tricamera aerial photography; the 1973 position, from Anonymous (1975) is based on Earth Resources Technology Satellite imagery; the 1985 position was established by radar bearings from U.C. Coast Guard icebreaker Glacier controlled by satellite navigation. ("km" denotes "kilometer:')
mous 1968a) and Earth Resources Technology Satellite imagery in 1973 (Anonymous 1975) both show the calving front at positions west of DF85-107 (figure 2). Using these marginal positions, we calculated a retreat rate of approximately 0.8 kilometers per year for the floating terminus of Pine Island Glacier. If retreat continued at this rate for more than the last few decades, the calving front of Pine Island Glacier would have been located west of DF85-108 less than 100 years ago. An alternative to our hypothesis of gradual calving-margin retreat is illustrated in figure 3. We speculate that "Thwaites Iceberg Tongue" is actually a huge segment of Pine Island Glacier that formerly occupied Pine Island Bay. The calving margin for this iceberg must have been close to the present terminus of Pine Island Glacier. Prevailing easterly winds (Zwally et al. 1983; figures 2-6) drifted the iceberg westward until it grounded northwest of Thwaites Glacier, the terminus of which must have been further south than it is now. A subsequent advance of Thwaites GLacier brought the two ice masses into contact, perhaps contributing to the observed northward movement and counterclockwise rotation of "Thwaites Iceberg Tongue," and giving this grounded iceberg the appearance of having calved from Thwaites Glacier [compare the positions of Thwaites Iceberg Tongue shown by Anonymous (1968b) and Anonymous (1975); note also the separation between Thwaites Glacier and the iceberg in Anonymous (1968b); figure 31. Whichever hypothesis is correct, microfossil abundances in Pine Island Bay core-top samples are consistent with the extremely low productivity expected beneath an ice shelf but not beneath sea ice. The sediments of Pine Island Bay support this interpretation, consisting of inorganic silt- and clay-sized particles, which we interpret as glacially derived rock flour, with occasional drop stones. The few microfossils encountered were 1986 REVIEW
probably introduced by episodic marine currents penetrating beneath the ice. Because diatoms are quite abundant in sea-ice and surface-water samples from Pine Island Bay at present, the very low diatom abundances inPine Island Bay core tops suggest that removal of the former shelf ice from this region was rapid and very recent, whether occurring by gradual calvingmargin recession or as one large block. We thank Captain W. Hewell and the officers and crew of U.S. Coast Guard icebreaker Glacier. Terence Hughes, and John Anderson and his students, who assisted with our coring project in January of 1985. Dennis Cassidy assisted with sampling the cores. David Thompson and Stephanie Staples assisted with sample preparation. This work was supported by National Science Foundation grant DPP 80-20000. References Anderson, J.B., and N.C. Myers. 1981. USCGC Glacier Deep Freeze 81 expedition to the Amundsen Sea and Bransfield Strait Antarctic Journal of the U.S., 16(5), 118-119. Anonymous. 1968a. Antarctic Sketch Map: Thurston Island—Jones Mountains. (U.S. Geologic Survey map.) Washington, D.C.: U.S. Government Printing Office. Anonymous. 1968b. Antarctica Sketch Map: Bakutis Coast—Marie Byrd Land. (U.S. Geologic Survey map.) Washington, D.C.: U.S. Government Printing Office. Anonymous. 1975. Thurston Island—Thwaites Area, Antarctica, 1972-1974. (Experimental printing). (Geological Survey map.) Washington, D.C.: U.S. Government Printing Office. Kellogg, D.E., and T.B. Kellogg. In preparation. Microfossil distributions in modern Amundsen Sea sediments. Palaeigeography Palaeoclunatology, Palaeoecology.
155
Kellogg, T.B., D.E. Kellogg, and J.B. Anderson. 1982. Preliminary results of microfossil analyses of Amundsen Sea sediment cores. Antarctic Journal of the U.S., 17(5), 125-126. Kellogg, TB., D.E. Kellogg, and T.J. Hughes. 1985. Amundsen Sea sediment coring. Antarctic Journal of the U.S., 20(5), 79-81. U.S. Naval Oceanographic Office. 1971. Thwaites Ice Tongue to Thurston
Island. Chart No. 29200 (corrected to 1975). Zwally, H.J., J.C. Comiso, C.L. Parkinson, W.J. Campbell, F.D. Carsey, and P. Gloersen. 1983. Antarctic Sea Ice, 1973-1976: Satellite passivemicrowave observations. (National Aeronautics and Space Administra-
tion Scientific and Technical Information Branch, SP-459.) Washington, D.C.: U.S. Government Printing Office.
H iOV
ABBor CE SHEL
04
I
ISLAND
430S
COSGROVE ICE SHELF •I4NSUL4 KINGPENINSULA
7WWAITES !ICEBERG TONGUE
I
74'S
/
C,
t4J
I'
-_
HUDSON
I I S.
•
4-'
MO UNTAINS
1
S...
-4
-75S
Lu
0 ?0 40 40 SO '00 L' . _a L._..&
L
Figure 3. Map illustrating the hypothesis that "Thwaites Iceberg Tongue" is actually an iceberg calved from Pine Island Glacier. Solid outline of Thwaites Iceberg Tongue and Thwaites Glacier is the December 1972 to January 1973 position from Earth Resources Technology Satellite imagery (Anonymous 1975); dotted outline is the 1966 position from tricamera aerial photography (Anonymous 1968b); dashed outline in Pine Island Bay is the hypothetical position approximately 100 years ago. Note that an advance of Thwaites Glacier since 1966 has caused northward movement and counterclockwise rotation of Thwaites Iceberg Tongue. ("km" denotes "kilometer.")
156
ANTARCTIC JOURNAL
PcEsNSL4N: Institute for Quaternary Studies
and Department of Geological Sciences University of Maine Orono, Maine 04469
1
EASTERN
ICE NSULA 5HEL PEN coscRove
In January 1985, the U.S. Coast Guard icebreaker Glacier penetrated the heavy pack ice of the outer Amundsen Sea. We took advantage of this unprecedented opportunity to collect 20 piston cores from the Amundsen Sea continental shelf and Pine Island Bay (Kellogg, Kellogg, and Hughes 1985). These new cores supplement existing coverage of the outer continental shelf of the Amundsen Sea (Anderson and Myers 1981) with material from the eastern side of the Amundsen embayment (figure 1). Cores were collected for the purpose of microfossil and sedimentary analyses that would provide data bearing on the glacial history of the region. Microfossil distributions in core-top samples define three regions or provinces in sediments of the eastern Amundsen Sea (Kellogg and Kellogg in preparation). These provinces are defined best by foraminiferal and diatom distributions, and less clearly by radiolarian data (figure 1). Outer shelf province. The presence of abundant well-preserved calcareous benthic foraminifera suggests that bottom waters on the outer shelf are well oxygenated and that dissolution is not significant here at typical depths between 400 and 463 meters. Low marine-diatom abundances were recorded in core-top sediments from this area. While consistent with our findings further west on the outer shelf (Kellogg, Kellogg, and Anderson 1982), these findings are inconsistent with the high diatom abundances we found in ice and water samples from this area. Diatomaceous material was observed staining the bottom of numerous ice floes in this region in January of 1985. Possibly our core tops represent older material outcropping at the surface. If this hypothesis is correct, the presence of the stratigraphicindicator diatom species Nitzschia curta suggests that deposition ceased during the Quaternary. Another alternative is that an ice shelf, or multi-year sea ice, thicker than ice found in the region today, covered the region until very recently, preventing deposition of the photosynthetic diatoms. The ice-shelf hypothesis appears unlikely because no source for shelf ice, at a time when the eastern margin province was apparently the site of a seasonal polynya (discussed below), has been identified and because there are no known shallows or pinning points for an ice shelf in this area. Our preferred interpretation is that currents may be sufficiently strong on the outer shelf to prevent deposition of diatoms. Eastern margin province. Abundances of marine diatoms in eastern margin province sediments are much higher than in adjacent regions to the north and south. These high abundances suggest that surface waters in this area were highly productive during the time represented by deposition of the top 1-2 centimeters of sediment used in our analyses. Satellite data suggest that this area is relatively ice-free during austral sum154
ICEBERG
Co
ToNGur
Is
Is
0 is-Is I
LQ
Figure 1. Map of the Amundsen Sea showing core locations, localities mentioned in the text, bathymetry, and boundaries of biotic provinces. Bathymetry (contours in meters) for the shelf break region is based on data from Anderson and Myers (1981) and Kellogg (unpublished), and on U.S. Naval Oceanographic Office (1971): contours dashed to indicate uncertainties in locations of soundings. Bathymetry south of 720s shows only 500-meter contour, separating major troughs and shallows (dashed were estimated). Boundaries of biotic provinces (heavy dashed lines) are uncertain beyond area where cores were collected. ("km" denotes "kilometer.")
mers in some years (Zwally et al. 1983). We conclude that the Amundsen Sea polynya has been a persistent feature of the eastern margin area throughout the period of deposition of our core-top samples. Our analyses of the cores from this region shows that the uppermost layer is a soft, diatomaceous, sandy mud which averages 4-6 centimeters in thickness. This layer probably represents Recent sedimentation at a high deposition rate, resulting from the proximity of numerous glaciers. If sedimentation rates are high (greater than 0.005 centimeter per year), the uppermost diatom-bearing layer may result from deposition related to the presence of the polyna during only the last 1,000 to 2,000 years. The presence of predominantly arenaceous benthic foraminifera in the eastern margin province results from the fact that most of our cores were taken from deep (perhaps glacially scoured) troughs in water depths exceeding 500 meters. At the site of grab B175, where the depth was only 293 meters, calcareous benthic foraminifera predominated. We conclude that the calcium carbonate compensation depth in this area must lie between 300 and 500 meters. Pine Island Bay province. Pine Island Bay core tops (and cores) have very low microfossil abundances, usually consisting of only a few specimens per sample, yet we recovered over 2 meters of sediment at each location. We believe this nearly microfossil-barren sediment was deposited beneath a formerly more-extensive Pine Island Glacier. We know this is true for core DF85-107 because tricamera aerial photography in 1966 (AnonyANTARCTIC JOURNAL
January 1985 105
- - - - January 1973
PINE ISLAND 106
1966
SAY -
1050 10 20 4/ 1
km
106 0 W
102 0 W
Aq
98° W
Figure 2. Map of Pine Island Bay region showing changing positions of the calving margin of Pine Island Glacier. The 1966 position, from Anonymous (1968a), is based on tricamera aerial photography; the 1973 position, from Anonymous (1975) is based on Earth Resources Technology Satellite imagery; the 1985 position was established by radar bearings from U.C. Coast Guard icebreaker Glacier controlled by satellite navigation. ("km" denotes "kilometer:')
mous 1968a) and Earth Resources Technology Satellite imagery in 1973 (Anonymous 1975) both show the calving front at positions west of DF85-107 (figure 2). Using these marginal positions, we calculated a retreat rate of approximately 0.8 kilometers per year for the floating terminus of Pine Island Glacier. If retreat continued at this rate for more than the last few decades, the calving front of Pine Island Glacier would have been located west of DF85-108 less than 100 years ago. An alternative to our hypothesis of gradual calving-margin retreat is illustrated in figure 3. We speculate that "Thwaites Iceberg Tongue" is actually a huge segment of Pine Island Glacier that formerly occupied Pine Island Bay. The calving margin for this iceberg must have been close to the present terminus of Pine Island Glacier. Prevailing easterly winds (Zwally et al. 1983; figures 2-6) drifted the iceberg westward until it grounded northwest of Thwaites Glacier, the terminus of which must have been further south than it is now. A subsequent advance of Thwaites GLacier brought the two ice masses into contact, perhaps contributing to the observed northward movement and counterclockwise rotation of "Thwaites Iceberg Tongue," and giving this grounded iceberg the appearance of having calved from Thwaites Glacier [compare the positions of Thwaites Iceberg Tongue shown by Anonymous (1968b) and Anonymous (1975); note also the separation between Thwaites Glacier and the iceberg in Anonymous (1968b); figure 31. Whichever hypothesis is correct, microfossil abundances in Pine Island Bay core-top samples are consistent with the extremely low productivity expected beneath an ice shelf but not beneath sea ice. The sediments of Pine Island Bay support this interpretation, consisting of inorganic silt- and clay-sized particles, which we interpret as glacially derived rock flour, with occasional drop stones. The few microfossils encountered were 1986 REVIEW
probably introduced by episodic marine currents penetrating beneath the ice. Because diatoms are quite abundant in sea-ice and surface-water samples from Pine Island Bay at present, the very low diatom abundances inPine Island Bay core tops suggest that removal of the former shelf ice from this region was rapid and very recent, whether occurring by gradual calvingmargin recession or as one large block. We thank Captain W. Hewell and the officers and crew of U.S. Coast Guard icebreaker Glacier. Terence Hughes, and John Anderson and his students, who assisted with our coring project in January of 1985. Dennis Cassidy assisted with sampling the cores. David Thompson and Stephanie Staples assisted with sample preparation. This work was supported by National Science Foundation grant DPP 80-20000. References Anderson, J.B., and N.C. Myers. 1981. USCGC Glacier Deep Freeze 81 expedition to the Amundsen Sea and Bransfield Strait Antarctic Journal of the U.S., 16(5), 118-119. Anonymous. 1968a. Antarctic Sketch Map: Thurston Island—Jones Mountains. (U.S. Geologic Survey map.) Washington, D.C.: U.S. Government Printing Office. Anonymous. 1968b. Antarctica Sketch Map: Bakutis Coast—Marie Byrd Land. (U.S. Geologic Survey map.) Washington, D.C.: U.S. Government Printing Office. Anonymous. 1975. Thurston Island—Thwaites Area, Antarctica, 1972-1974. (Experimental printing). (Geological Survey map.) Washington, D.C.: U.S. Government Printing Office. Kellogg, D.E., and T.B. Kellogg. In preparation. Microfossil distributions in modern Amundsen Sea sediments. Palaeigeography Palaeoclunatology, Palaeoecology.
155
Kellogg, T.B., D.E. Kellogg, and J.B. Anderson. 1982. Preliminary results of microfossil analyses of Amundsen Sea sediment cores. Antarctic Journal of the U.S., 17(5), 125-126. Kellogg, TB., D.E. Kellogg, and T.J. Hughes. 1985. Amundsen Sea sediment coring. Antarctic Journal of the U.S., 20(5), 79-81. U.S. Naval Oceanographic Office. 1971. Thwaites Ice Tongue to Thurston
Island. Chart No. 29200 (corrected to 1975). Zwally, H.J., J.C. Comiso, C.L. Parkinson, W.J. Campbell, F.D. Carsey, and P. Gloersen. 1983. Antarctic Sea Ice, 1973-1976: Satellite passivemicrowave observations. (National Aeronautics and Space Administra-
tion Scientific and Technical Information Branch, SP-459.) Washington, D.C.: U.S. Government Printing Office.
H iOV
ABBor CE SHEL
04
I
ISLAND
430S
COSGROVE ICE SHELF •I4NSUL4 KINGPENINSULA
7WWAITES !ICEBERG TONGUE
I
74'S
/
C,
t4J
I'
-_
HUDSON
I I S.
•
4-'
MO UNTAINS
1
S...
-4
-75S
Lu
0 ?0 40 40 SO '00 L' . _a L._..&
L
Figure 3. Map illustrating the hypothesis that "Thwaites Iceberg Tongue" is actually an iceberg calved from Pine Island Glacier. Solid outline of Thwaites Iceberg Tongue and Thwaites Glacier is the December 1972 to January 1973 position from Earth Resources Technology Satellite imagery (Anonymous 1975); dotted outline is the 1966 position from tricamera aerial photography (Anonymous 1968b); dashed outline in Pine Island Bay is the hypothetical position approximately 100 years ago. Note that an advance of Thwaites Glacier since 1966 has caused northward movement and counterclockwise rotation of Thwaites Iceberg Tongue. ("km" denotes "kilometer.")
156
ANTARCTIC JOURNAL
