(Table 2) Salinity, pH, alkalinity and major ion concentrations in ANDRILL core AND-2A pore water

The compositional record of the AND-2A drillcore is examined using petrological, sedimentological, volcanological and geochemical analysis of clasts, sediments and pore waters. Preliminary investigations of basement clasts (granitoids and metasediments) indicate both local and distal sources corresponding to variable ice-volume and ice-flow directions. Low abundance of sedimentary clasts (e.g., arkose, litharenite) suggests reduced contributions from sedimentary covers while intraclasts (e.g., diamictite, conglomerate) attest to intrabasinal reworking. Volcanic material includes pyroclasts (e.g., pumice, scoria), sediments and lava. Primary and reworked tephra layers occur within the Early Miocene interval (1093 to 640 metres below sea floor mbsf). The compositions of volcanic clasts reveal a diversity of alkaline types derived from the McMurdo Volcanic Group. Finer-grained sediments (e.g., sandstone, siltstone) show increases in biogenic silica and volcanic glass from 230 to 780 mbsf and higher proportions of terrigenous material c. 350 to 750 mbsf and below 970 mbsf. Basement clast assemblages suggest a dominant provenance from the Skelton Glacier - Darwin Glacier area and from the Ferrar Glacier - Koettlitz Glacier area. Provenance of sand grains is consistent with clast sources. Thirteen Geochemical Units are established based on compositional trends derived from continuous XRF scanning. High values of Fe and Ti indicate terrigenous and volcanic sources, whereas high Ca values signify either biogenic or diagenic sources. Highly alkaline and saline pore waters were produced by chemical exchange with glass at moderately elevated temperatures.

Element abundances, loss on ignition, total analyzed abundances, CIA values, and Al-oxide/Ti-oxide ratios of sediment core CRP-2A (Table1)

The CRP-2/2A core, drilled in western McMurdo Sound in October and November 1998, penetrated 624 m of Quaternary. Pliocene, lower Miocene, and Oligocene glacigenic sediments. The palaeoclimatic record of CRP-2/2A is examined using major element analyses of bulk core samples of fine grained sediments (mudstones and siltstones) and the Chemical Index of Alteration (CIA) of Nesbitt & Young (1982). The CIA is calculated from the relative abundances of AI, K, Ca, and Na oxides, and its magnitude increases as the effects of chemical weathering increase. However, changes in sediment provenance can also affect the CIA, and provenance changes are recorded by shifts in the Al2O3/TiO2 ratios and the Nb contents of these CRP-2/2A mudstones. Relatively low CIA values (40-50) occur throughout the CRP-2/2A sequence, whereas the Al2O3/TiO2 ratio decreases upsection. The major provenance change is an abrupt onset of McMurdo Volcanic Group detritus at ~300 mbsf and is best characterized by a rapid increase in Nb content in the sediments. This provenance shift is not evident in the CIA record, suggesting that a contribution from the Ferrar Dolerite to the older sediments was replaced by an input of McMurdo Volcanic Group material in the younger sediments. If this is true, then the relatively uniform CIA values indicate relatively consistent palaeoweathering intensities throughout the Oligocene and early Miocene in the areas that supplied sediment to CRP-2/2A.

Color reflectance and XRF drill core scanner data of ANDRILL core AND-2A

The compositional record of the AND-2A drillcore is examined using petrological, sedimentological, volcanological and geochemical analysis of clasts, sediments and pore waters. Preliminary investigations of basement clasts (granitoids and metasediments) indicate both local and distal sources corresponding to variable ice-volume and ice-flow directions. Low abundance of sedimentary clasts (e.g., arkose, litharenite) suggests reduced contributions from sedimentary covers while intraclasts (e.g., diamictite, conglomerate) attest to intrabasinal reworking. Volcanic material includes pyroclasts (e.g., pumice, scoria), sediments and lava. Primary and reworked tephra layers occur within the Early Miocene interval (1093 to 640 metres below sea floor mbsf). The compositions of volcanic clasts reveal a diversity of alkaline types derived from the McMurdo Volcanic Group. Finer-grained sediments (e.g., sandstone, siltstone) show increases in biogenic silica and volcanic glass from 230 to 780 mbsf and higher proportions of terrigenous material c. 350 to 750 mbsf and below 970 mbsf. Basement clast assemblages suggest a dominant provenance from the Skelton Glacier - Darwin Glacier area and from the Ferrar Glacier - Koettlitz Glacier area. Provenance of sand grains is consistent with clast sources. Thirteen Geochemical Units are established based on compositional trends derived from continuous XRF scanning. High values of Fe and Ti indicate terrigenous and volcanic sources, whereas high Ca values signify either biogenic or diagenic sources. Highly alkaline and saline pore waters were produced by chemical exchange with glass at moderately elevated temperatures.

Summary of 40Ar/39Ar results from sediment core CRP-2A (Table 1)

40Ar/39Ar analyses of tephra and clasts of volcanic rock provide age constraints for upper parts of the CRP-2A core. Single-crystal laser-fusion analyses of anorthoclase phenocrysts from three tephra-bearing layers yielded the most precise age constraints for CRP-2A. The dated tephra layers are: 1) a 2.7-m-thick interval of pumice and ash layers between 111.5 and 114.2 meters below sea floor (mbsf) (weighted mean age = 21.44 ± 0.05 Ma, +2.2); 2) a concentration of pumice near 193.4 mbsf (23.98 ± 0.13 Ma): and 3) a concentration of pumice near 280 mbsf (24,22 ± 0.03 Ma) (all ages are calibrated relative to Fish Canyon Tuff sanidine at 27.84 Ma). The 111 to 114 mbsf tephra is almost entirely non-reworked, and the 193 mbsf and 280 mbsf tephra concentrations are interpreted as being reworked and redeposited soon after eruption. All three of the tephra ages are therefore considered to be equivalent to depositional ages. The variation in precision of these three age determinations is largely a function of phenocryst size and abundance. The accuracy of these ages is equal to the accuracy of the current calibration of the 40Ar/39Ar methode (about ± 1 %). 40Ar/39Ar results from volcanic clasts provide three additional maximum age constraints for the CRP-2A core. Single-crystal laser-fusion of sanidine phenocrysts from a rhyolitic clast from 294 mbsf yielded a precise maximum depositional age of 24.98 ± 0.08 Ma, and plateau ages of groundmass concentrates from basaltic clasts near 36.02 mbsf and 125.92 mbsf yielded maximum depositional ages of 19.18 ± 0.12 Ma, and 22.56 ± 0.14 Ma, respectively. The 40Ar/39Ar data, in association with biostratigraphic, paleomagnetic, and isotopic age constraints for CRP-2A, confirm interpretation for rapid sedimentation rates in the 36 to 280 mbsf interval, particularly in the 193 to 280 mbsf interval where they support interpretations for sedimentation cycles spanning 100 k.y. intervals. In addition to the 19 to 25 Ma ages measured from thephra layers and clasts, provenance-related ages ranging from 150 to 450 Ma were determined from clasts and individual detrital or xenocrystic crystals from CRP-2A.