Preservation of primary productivity signal in antarctic fjord sediments ...
Total Organic Carbon
Preservation of primary productivity signal in antarctic fjord sediments: Andvord Bay, Antarctica EUGENE W. DOMACK AND Titcy A. MAsHI0TrA1
Geology Department Hamilton College Clinton, New York 13323 M. I. VENKATSEN
Institute of Geophysics and Planetary Physics University of California, Los Angeles Los Angeles, California 90024-1567
0.8 1.0 1.2 1.4 1.6 1.8
100
200
300
400
500
This study is an interdisciplinary investigation into the paleoproductivity of an antarctic fjord based on sedimentologic and biogeochemical studies of a 9-meter-long piston core. We collected core 22 from Andvord Bay (6449.625' S 62'39.001' W)in early 1988 as part of R/V Polar Duke cruise III. The chronology of core 22 extends back for approximately the last 3,000 years as based upon a set of five radiocarbon ages that range in age from 2025 ± 60 to 4480 ± 75 (table 1). These data result in a linear accumulation rate of 0.305 centimeters per year and interval rates of 0.23 and 0.52 centimeters per year for the upper and lower portions of the core respectively. Total organic carbon (TOC) analyses were conducted at 10centimeter sample intervals, and the smoothed results (three sample running average) are illustrated in figure 1. Of note is the pronounced cyclicity in the TOC content, with maxima alternating with minima approximately every 270 years. Also of note is the generally high TOC (greater than 1.0 percent) below 550 centimeters depth in the core. Since the organic carbon and foraminifera calcite ages are in agreement, it is apparent the organic carbon is derived from autochthonous marine carbon, primarily phytoplankton material, and that there is very little reworking of particulate organic carbon within this system. Because of this cycles of TOC content within the core may reflect direct variations in the productivity of the overlying water column within Andvord Bay. To evaluate the source of the organic carbon we carried out chemical characterization of the lipid fraction from the sediments. We quantitated three sterol biomarkers by gas chromatography (GC) after confirmation with GC/mass spectrometer (MS), using established procedures of Venkatesan et al. (1987). They were brassicasterol (a Phaeocyctis marker, Smith et al. 1989), cholesta-5, 22E-dienol (a diatom marker, Nichols et al. 1986), and dinosterol (a dinoflagellate marker, Volkman 1986). We selected 10 samples at various depths in order to evaluate the variation of these biomarkers with variations in the TOC content of the sediment. The results in nanograms per gram of dry sediment are illustrated in figure 2. From the distribution profiles of these 'Present address: Department of Geology, University of California Santa Barbara, California 93106
66
600
700
800
900 12
18
Weight Percent Biogenic Silica Figure 1. Downcore trend of total organic carbon (solid dots) biogenic silica (open dots) for core PD88 22. Note correspon& of both compositional parameters.
sterols it is quite evident that the preserved productivity islinke to dominantly diatoms and phaeocystis with some contributio from dinoflagellates. It is also evident that the levels of all thre sterols vary in tandem with the TOC content of the sediment. Thi tends to support the idea that fluctuations in TOC are a reflectio of fluctuations in paleoproductivity rather than fluctuations i meltwater-derived sediment input. If diatoms are a major component of the preserved sediment, then there should be goo agreement between the TOC contents and the preserved biogeni silica contents. We took samples every 20 centimeters and analyzed them for! their biogenic silica content, using standard methods. The results ranged from 12 to 21 percent biogenic silica and are illustrated ir figure 1. The pattern of biogenic silica content is strikingly similar to the pattern of TOC contents (figure 1) and is characterized b peaks and troughs in the biogenic silica content that parallel th variation in TOC. There is generally higher biogenic silica belowl 550 centimeters in the core, that is before about 2,400 years ago.
ANTARCTIC JOURNA!.
nglg dry sediment %TOC
0
0.0 0.5 1.0 1.5 2.0 -1-
Cholesta5,22Edien-38-ol
Dinosterol
% Biogenic Silica
0 .500 1000
0 200 400 600
0 5 10 15 20 25
Brassicasterol 0 500 1000
IIJ
200 . 400-1 EE j 6001
4
800-1
I1
1000
rID]
II
Figure 2. Downcore variation in three biomarker compounds in nanogram; per gram of dried sediment. Also shown for comparison is the total organic carbon content of selected intervals. Note correspondence in organic carbon and abundance of key biomarker compounds. Radiocarbon ages of samples from core 22 Sample number Depth(cm) 14 C Age, years B.P. Carbon source AA-4751 66 2025 ±60 organic matter AA-5210 472-510 3750±65 foraminifera AA-4752 487 3825±65 organic matter AA-5209 778-798 4480±75 foraminif era AA-4753 792 4415±60 organic matter
The close agreement between the biogenic silica, TOC, and biomarker abundances suggests that core 22 has a preserved record of paleoproductivity that is both striking in its apparent cyclicity and its time resolution. However, there are other factors that would influence the preserved biogenic content of the sediment besides fluctuations in primary productivity. These include sedimentation rates, mixing in the surface mixed layer (usually the upper 10 centimeters), and bottom-water temperatures, all of which could be variable, and downcore dissolution. If sedimentation rates were variable, then there should be significant offset of the downcore trends of carbon-14 age and fluctuations in the background levels of ice-rafted debris. Neither of these two conditions are found as the carbon-14 ages suggest a uniform rate of sedimentation with slightly greater rates below 500 centimeters in depth. The distribution of ice-rafted material is uniform throughout the core, which suggests a lack of current erosion or intervals of increased fine-grain sedimentation. Variable rates of biological mixing can influence the preservation of organic car-
1992 REVIEW
bon, but there is no basis to expect changes in preserved biogenic silica. Downcore dissolution is not apparent from an examination of the diatom frustules (Mashiotta 1992) and is certainly not significant because of the increase in biogenic silica downcore. Hence, the most reasonable interpretation of the data would suggest the fluctuations in primary productivity responsible for the changes in TOC below 550 centimeters are a reflection of greater sedimentation rates and that this in turn is driven by higher primary productivity in the period before 2,400 years ago. This work was supported by National Science Foundation grant DPP 86-13565, DPP 89-15977 and DPP 88-16292. The authors thank F. Saneghi for technical assistance, E. Ruth for CC! MS analysis, and D. Cassidy for samples of core 22. References Mashiotta, T. A. 1992. Biogenic sedimentation in Andvord Bay, Antarctica: A 3000 year record of paleoproductivity. B.A. thesis, Hamilton College, Clinton, New York, 86 pp. Nichols, P. D., A. C. Palmisano, G. A. Smith, and D. C. White. 1986. Lipids of the Antarctic sea ice diatom Nitchia cylindrus. Phytochemistry, 25: 1,649-1,653. Smith, C. A., P. D. Nichols, and D. C. White. 1989. Triacyiglycerol fatty acid and sterol composition of sediment microorganisms from McMurdo Sound, Antarctica. Polar Biology, 9:273-279. Venkatesan, M. I., E. Ruth, S. Steinberg, and I. R. Kaplan. 1987. Organic geochemistry of sediments from the continental margin off southern New England, USA-Part II. Lipids. Marine Chemistry, 21:267-299. Volkman, J . K. 1986. A review of sterol markers for marine and terrigenous organic matter. Organic Geochemistry, 9:83-100.
67
Preservation of primary productivity signal in antarctic fjord sediments: Andvord Bay, Antarctica EUGENE W. DOMACK AND Titcy A. MAsHI0TrA1
Geology Department Hamilton College Clinton, New York 13323 M. I. VENKATSEN
Institute of Geophysics and Planetary Physics University of California, Los Angeles Los Angeles, California 90024-1567
0.8 1.0 1.2 1.4 1.6 1.8
100
200
300
400
500
This study is an interdisciplinary investigation into the paleoproductivity of an antarctic fjord based on sedimentologic and biogeochemical studies of a 9-meter-long piston core. We collected core 22 from Andvord Bay (6449.625' S 62'39.001' W)in early 1988 as part of R/V Polar Duke cruise III. The chronology of core 22 extends back for approximately the last 3,000 years as based upon a set of five radiocarbon ages that range in age from 2025 ± 60 to 4480 ± 75 (table 1). These data result in a linear accumulation rate of 0.305 centimeters per year and interval rates of 0.23 and 0.52 centimeters per year for the upper and lower portions of the core respectively. Total organic carbon (TOC) analyses were conducted at 10centimeter sample intervals, and the smoothed results (three sample running average) are illustrated in figure 1. Of note is the pronounced cyclicity in the TOC content, with maxima alternating with minima approximately every 270 years. Also of note is the generally high TOC (greater than 1.0 percent) below 550 centimeters depth in the core. Since the organic carbon and foraminifera calcite ages are in agreement, it is apparent the organic carbon is derived from autochthonous marine carbon, primarily phytoplankton material, and that there is very little reworking of particulate organic carbon within this system. Because of this cycles of TOC content within the core may reflect direct variations in the productivity of the overlying water column within Andvord Bay. To evaluate the source of the organic carbon we carried out chemical characterization of the lipid fraction from the sediments. We quantitated three sterol biomarkers by gas chromatography (GC) after confirmation with GC/mass spectrometer (MS), using established procedures of Venkatesan et al. (1987). They were brassicasterol (a Phaeocyctis marker, Smith et al. 1989), cholesta-5, 22E-dienol (a diatom marker, Nichols et al. 1986), and dinosterol (a dinoflagellate marker, Volkman 1986). We selected 10 samples at various depths in order to evaluate the variation of these biomarkers with variations in the TOC content of the sediment. The results in nanograms per gram of dry sediment are illustrated in figure 2. From the distribution profiles of these 'Present address: Department of Geology, University of California Santa Barbara, California 93106
66
600
700
800
900 12
18
Weight Percent Biogenic Silica Figure 1. Downcore trend of total organic carbon (solid dots) biogenic silica (open dots) for core PD88 22. Note correspon& of both compositional parameters.
sterols it is quite evident that the preserved productivity islinke to dominantly diatoms and phaeocystis with some contributio from dinoflagellates. It is also evident that the levels of all thre sterols vary in tandem with the TOC content of the sediment. Thi tends to support the idea that fluctuations in TOC are a reflectio of fluctuations in paleoproductivity rather than fluctuations i meltwater-derived sediment input. If diatoms are a major component of the preserved sediment, then there should be goo agreement between the TOC contents and the preserved biogeni silica contents. We took samples every 20 centimeters and analyzed them for! their biogenic silica content, using standard methods. The results ranged from 12 to 21 percent biogenic silica and are illustrated ir figure 1. The pattern of biogenic silica content is strikingly similar to the pattern of TOC contents (figure 1) and is characterized b peaks and troughs in the biogenic silica content that parallel th variation in TOC. There is generally higher biogenic silica belowl 550 centimeters in the core, that is before about 2,400 years ago.
ANTARCTIC JOURNA!.
nglg dry sediment %TOC
0
0.0 0.5 1.0 1.5 2.0 -1-
Cholesta5,22Edien-38-ol
Dinosterol
% Biogenic Silica
0 .500 1000
0 200 400 600
0 5 10 15 20 25
Brassicasterol 0 500 1000
IIJ
200 . 400-1 EE j 6001
4
800-1
I1
1000
rID]
II
Figure 2. Downcore variation in three biomarker compounds in nanogram; per gram of dried sediment. Also shown for comparison is the total organic carbon content of selected intervals. Note correspondence in organic carbon and abundance of key biomarker compounds. Radiocarbon ages of samples from core 22 Sample number Depth(cm) 14 C Age, years B.P. Carbon source AA-4751 66 2025 ±60 organic matter AA-5210 472-510 3750±65 foraminifera AA-4752 487 3825±65 organic matter AA-5209 778-798 4480±75 foraminif era AA-4753 792 4415±60 organic matter
The close agreement between the biogenic silica, TOC, and biomarker abundances suggests that core 22 has a preserved record of paleoproductivity that is both striking in its apparent cyclicity and its time resolution. However, there are other factors that would influence the preserved biogenic content of the sediment besides fluctuations in primary productivity. These include sedimentation rates, mixing in the surface mixed layer (usually the upper 10 centimeters), and bottom-water temperatures, all of which could be variable, and downcore dissolution. If sedimentation rates were variable, then there should be significant offset of the downcore trends of carbon-14 age and fluctuations in the background levels of ice-rafted debris. Neither of these two conditions are found as the carbon-14 ages suggest a uniform rate of sedimentation with slightly greater rates below 500 centimeters in depth. The distribution of ice-rafted material is uniform throughout the core, which suggests a lack of current erosion or intervals of increased fine-grain sedimentation. Variable rates of biological mixing can influence the preservation of organic car-
1992 REVIEW
bon, but there is no basis to expect changes in preserved biogenic silica. Downcore dissolution is not apparent from an examination of the diatom frustules (Mashiotta 1992) and is certainly not significant because of the increase in biogenic silica downcore. Hence, the most reasonable interpretation of the data would suggest the fluctuations in primary productivity responsible for the changes in TOC below 550 centimeters are a reflection of greater sedimentation rates and that this in turn is driven by higher primary productivity in the period before 2,400 years ago. This work was supported by National Science Foundation grant DPP 86-13565, DPP 89-15977 and DPP 88-16292. The authors thank F. Saneghi for technical assistance, E. Ruth for CC! MS analysis, and D. Cassidy for samples of core 22. References Mashiotta, T. A. 1992. Biogenic sedimentation in Andvord Bay, Antarctica: A 3000 year record of paleoproductivity. B.A. thesis, Hamilton College, Clinton, New York, 86 pp. Nichols, P. D., A. C. Palmisano, G. A. Smith, and D. C. White. 1986. Lipids of the Antarctic sea ice diatom Nitchia cylindrus. Phytochemistry, 25: 1,649-1,653. Smith, C. A., P. D. Nichols, and D. C. White. 1989. Triacyiglycerol fatty acid and sterol composition of sediment microorganisms from McMurdo Sound, Antarctica. Polar Biology, 9:273-279. Venkatesan, M. I., E. Ruth, S. Steinberg, and I. R. Kaplan. 1987. Organic geochemistry of sediments from the continental margin off southern New England, USA-Part II. Lipids. Marine Chemistry, 21:267-299. Volkman, J . K. 1986. A review of sterol markers for marine and terrigenous organic matter. Organic Geochemistry, 9:83-100.
67
