The upper two sample intervals (A and B) record similar environmental conditions, and the lower two intervals (C and D) record different sets of environmental conditions.
Interval A covers the period from ~2020 to 1380 cal. BP, and interval B was deposited between ~3390 and 2740 cal. BP (Fig. F2) (Domack et al., 2000). The increased MS values throughout the upper portion of Hole 1098C are most likely due to increased input of terrestrial materials to the basin resulting from glacial erosion of high magnetite-bearing bedrock (Brachfeld, 1999; Brachfeld and Banerjee, 2000) (Fig. F2). In all samples from intervals A and B, the abundance of the diatom Fragilariopsis curta (sea-ice indicator) is greater than Fragilariopsis kerguelensis (open-water indicator) (Fig. F3). These data suggest the surface water environment of the late Holocene was characterized by a significant increase in sea-ice and ice-edge production. Uniform composition of the agglutinated foraminifer fauna throughout the late Holocene suggests that the Palmer Deep Basin was continually influenced by modified UCDW (Hofmann and Klink, 1998) and there were no major changes in the bottom-water mass.
The low values of MS within intervals A and B are interpreted to represent an increased influx of diatoms (Leventer et al., 1996; Shipboard Scientific Party, 1999). In all cases, laminated sediments, because of rapid deposition of diatom blooms, characterize the susceptibility lows. The rapid deposition of diatoms results in good preservation of more fragile diatoms but can affect the benthos in two ways. First, the diatom decay uses oxygen and results in lowered bottom-water oxygen. Second, the decay of the organic matter lowers the pH, which results in CaCO3 dissolution. Syndepositional calcareous carbonate dissolution due to reduced bottom-water pH is indicated during the MS lows. However, the presence of mobile epifaunal agglutinated foraminifers (Kaminski et al., 1995) in the susceptibility lows indicates oxygen was present and any anoxia that might have developed would have been short lived (Fig. F4).
The extensive laminations throughout Hole 1098C may themselves be a cause for decreased bioturbation. Laminations of mat-forming diatoms have been shown to suppress benthic activity regardless of bottom-water oxygen levels (Kemp and Baldauf, 1993; Boden and Backman, 1996; Pike and Kemp, 1999). These authors show that the macrobenthos were prevented from bioturbation through the intermeshed diatom frustules, which resulted in the increased preservation of the laminations even in well-oxygenated bottom waters.
Within intervals A and B, sediments with relatively higher MS are bioturbated and massive but weak laminations are present. The diatom assemblages of the susceptibility highs indicate a more diverse oceanic assemblage, suggesting a well-mixed ocean with average surface productivity and the reduced influence of meltwater. However, it must be remembered that even the "average productivity" of the Palmer Deep is still very high.
Less meltwater and surface stratification would result in fewer diatom blooms. The slight decrease in productivity allows for more bottom-water oxygen, enabling more vigorous macrobenthos bioturbation and the destruction of diatom frustules, which results in "poorer" diatom preservation. Poorer diatom preservation also results from the slower settling rate of the nonmat-forming diatoms. The reduced formation and deposition of diatom mats also created less CaCO3 dissolution of calcareous benthic foraminifers in the massive sediments. Therefore, surface-water processes can be used to explain the greater numbers of calcareous foraminifers along with the agglutinated species in the susceptibility highs (Fig. F6).
Interval C (14.42-15.81 mbsf) comes from the section of Hole 1098C that has low-amplitude MS fluctuations and the most strongly laminated sediments of the middle Holocene (~5920-5430 cal. BP) (Fig. F2) (Domack et al., 2001). Bioturbated sediments and relatively higher MS values are rare in interval C, and overall there is a very weak correlation between laminated sediments and MS (Fig. F6). In this interval of Hole 1098C, the decreased MS (Fig. F2) results from a low magnetite-bearing sediment source (Brachfeld, 1999; Brachfeld and Banerjee, 2000), as well as an overall dilution of the signal by an increased mass accumulation rate (MAR) during the Holocene Climatic Optimum (Domack et al., 2001).
Both the diatom (Table T2) and benthic foraminifer ABFAR (Table T1) data indicate the highest productivity rates of Hole 1098C, which explains the high MAR of this section at Site 1098 (Domack et al., 2001). The diatom assemblage records a relatively higher abundance of F. kerguelensis and lower values of F. curta (Fig. F3), which implies a lengthier season of open water and less sea-ice/meltwater influence. Therefore, during intervals C and D, the surface water was not as influenced by sea ice as during the late Holocene (intervals A and B).
Agglutinated benthic foraminifers do not record any assemblage changes, which suggests that environmental conditions were unchanging and that the same bottom-water mass (modified UCDW) likely influenced this site as during the late Holocene (Fig. F4). There is only a weak correlation between the dissolution of calcareous foraminifers and laminated sediments in interval C (Fig. F5), which is probably due to low carbonate values throughout this part of the hole. The dissolution of foraminifers in interval C is believed to be a result of decreased pH due to the decay of abundant organic matter deposited in this interval of increased productivity (Tables T1, T2) and was probably both syn-and postdepositional. Similar agglutinated faunas are found in other Antarctic diatom oozes (Schröder-Adams, 1990; Ishman and Domack, 1994; Harloff and Mackenson, 1997).
Interval D (24.74-24.01 mbsf) similarly records the low-amplitude MS fluctuations in Hole 1098C during the early Holocene (~8880-8590 cal. BP) (Domack et al., 2001). In general, there is a correlation between laminated sediments, decreased calcareous foraminifers, and low MS that may be syndepositional (Fig. F6).
However, the biota of interval D records significant changes in both bottom and surface waters. Changes in the benthic foraminifer assemblage, including an increase in the number of minor calcareous species and reduced numbers of both P. eltaninae (agglutinated) and B. pseudopunctata (calcareous) in this interval, suggest a change in bottom-water conditions (Table T7; Fig. F4). In addition, a shift in the oxygen isotopes of benthic foraminifers suggests that a different bottom-water mass was influencing this site during interval D (Fig. F6).
Changes in the surface water are indicated by an increase in the number of planktonic foraminifers, implying an influx of more open-oceanic water (Table T8). The diatom assemblage records more F. kerguelensis, which implies more open-water and less sea-ice influence during this interval (Fig. F3). In addition, interval D contains an increased number of the asymmetric "more northerly" form of the diatom E. antarctica (Fryxell, 1989; Kaczmarska et al., 1993) (Table T3). All of this evidence points to the intrusion of more northerly open-ocean surface and bottom CDW and a reduced meltwater influence on the Antarctic Peninsula shelf during the early Holocene.