Geochemical Indicators of Weathering, Cenoxoic Palaeocliinates, and Provenance in Fine-Grained Sediinents from CRP-3,
Victoria Land Basin, Antarctica
L.A. KRISSEK':;:
&P.R. KYI,$
'Department of Geological Sciences and Byrd Polar Research Ccntcr. Tlic Ohio State University. Colurnbus. OH 43210 - U.S.A ¥'Departmen of Earth & Environmental Science. New Mexico Iiistihite of Mining and Technology (N.M. Tecli).
Socorro. N.M. 878Ol-4706 - U.S.A.
Received 28 October 2000: accc/~ted in revisedform 22 February 2001
Abstract - The CRP-3 core. drilled in western McMurdo Sound in October and November 1999, penetrated 823 m of lower Oligocene (to possibly upper Eocene) glacially influenced sediments. The palaeoclimatic record of CRP-3 is examined using major element analyses of bulk core samples of fine grained sediments (iniidstones and siltstones) and the Chemical Index of Alteration (CIA) of Nesbitt &
Young (1982). The CIA is calculated from the molar abundances of Al. 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, so provenance changes have been evaluated by examining the Al,O/riO., ratios and the Sr and Nb contents of the CRP-3 mudstones.
Relatively low Nb contents (<20 ppm) and relatively high Al.,O/TiO, ratios (generally 21.5) indicate that little or no McMurdo Volcanic Group detritus was incorporated in these sediments. Instead, these sediments appear to be mixtures of Beacon Supergroup. Ferrar Dolerite. and basement (Granite Harbour I n t r ~ ~ s i v e s ) components; increased Ss contents suggest that the importance of basement-derived material is greatest above -200 metres below seafloor (mbsf), whereas Beacon Supergroup andlor Ferrar Dolerite components dominate below that level.
Below -400 mbsf. the CIA profile is irregular and shows no correlation to the ALOJTiO., record: as a result.
the effects of weathering and provenance controls have not yet been interpreted for this part of the CIA profile. Above -400 mbsf. however. four "cycles" of CIA increaseldecrease (each spanning 50-100 m of stratigraphic thickness) are superimposed on a general decrease in CIA values upcore, and Al~Oi/TiO, ratios exhibit a matching pattern of "cycles"'. This correlation indicates that short-term provenance changes occurred during deposition of the upper half of CRP-3. and affected its CIA record. However. the Al,O-/TiO, profile does not show any long-term trend above -400 mbsf, so that the general decrease in CIA values above that level does not appear to be provenance-related. Instead, the general upcore decrease in the importance of chemical weathering above -400 mbsf may record an episode of climatic deterioration prior to -31 Ma.
INTRODUCTION
T h e C a p e R o b e r t s P r o j e c t is a multinational cooperative drilling project, designed to use sediment cores as the basis for reconstructing the tectonic and climatic histories of the western side of McMurdo Sound and the adjacent portion of East Antarctica for the period from 3 0 M a to approximately 100 M a , T h e scientific rationale for this work, as well as the technical and logistical details of the project, have b e e n presented by B a r r e t t & Davey (1992).
International Steering C o m m i t t e e (1994), Barrett (1997). and Cape Roberts Science Team (2000).
O n e goal of t h e C a p e R o b e r t s P r o j e c t is to reconstruct the palaeoclimatic history of the western R o s s Sea region. s o b i o l o g i c a l a n d i n o r g a n i c indicators of palaeoclimate are being examined by a number of the post-drilling studies discussed in this
volume. The objective of this paper is to discuss the record provided by one such indicator, the Chemical Index of Alteration (CIA) of Nesbitt & Young (1982).
T h e C I A is c a l c u l a t e d f r o m t h e major element geochemistry of bulk sediment samples, and was originally proposed as a means to quantify the extent to w h i c h s e d i m e n t s have e x p e r i e n c e d chemical weathering. B e c a u s e t h e CIA can b e affected by c h a n g e s in the p r o v e n a n c e of t h e s e d i m e n t , independent of changes in weathering intensity, the A12031Ti0, ratio and the Sr and Nb contents are also considered as i n d e p e n d e n t r e c o r d s of s e d i m e n t provenance. A similar approach was used by Krissek
& Kyle ( 1 9 9 8 . 2 0 0 0 ) to e x a m i n e the records of palaeoweathering and sediment provenance at CRP- 11CIROS-l and CRP-2/2A, respectively.
In this paper we present an extensive set of major and trace e l e m e n t a n a l y s e s of 8 5 s a m p l e s of
*Corresponding author (krissek@mps.ohio-state.edu)
m ~ ~ d s t o n e s and siltstones. The major element analyses are used to examine tlie stratigraphic record of C1A values in samples from CRP-3. Only minor reference is made to the trace element analyses to examine the sediment provenance; detailed examination of these data will be made in subsequent publications. T h e CIA values suggest that chemical weathering effects were relatively low during the early Oligocene (late l o c e n e ? ) , as would be e x p e c t e d in a glacially i n f l u e n c e d environment. C h a n g e s in sediment provenance during the early Oligocene/late Eocenc(?) also affected the major element geochemistry and, therefore, the CIAs of these sediments. As a result, more definitive interpretations of the palaeoweathering history contained in CRP-3 will b e available only a f t e r detailed mixing m o d e l s t u d i e s have b e e n performed to remove tlie effects of changing sediment provenance.
BACKGROUND OF GEOCHEMICAL INDICATORS USED
The CIA is calculated as
CIA = [A&O,/(Al&
+
CaO* +Na,O+
K20)] x 100 where the elemental abundances a r e expressed as molar proportions, and C a O * represents t h e C a O contained only in the silicate fraction. The CIA is generally used to provide an indication of the relative abundances of "unweathered" material and chemical weathering products; the "unweathered" materials of particular interest a r e t h e f e l d s p a r s , which a r e common and contain relatively mobile Ca, Na, and K, w h e r e a s t h e c h e m i c a l w e a t h e r i n g products o f particular interest are the Al-rich clays. However, the CIA of a sample can also b e affected by the grain s i z e of t h e s a m p l e a n d by t h e p r o v e n a n c e of t h e sediment, as discussed in more detail below.The CIA of a sediment increases as the extent of c h e m i c a l w e a t h e r i n g i n c r e a s e s , f r o m values o f approximately 5 0 f o r "unweathered" feldspar-rich r o c k s to values n e a r 1 0 0 f o r h i g h l y w e a t h e r e d . kaolinite- or gibbsite-rich sediments. CIA values for
"average" shales, dominated by illite, range from 7 0 to 75 (Young & Nesbitt, 1998). The CIA value for a s e d i m e n t a l s o t e n d s to i n c r e a s e a s g r a i n s i z e decreases, because clay minerals are preferentially enriched in the finest grain sizes. As a result, the CIA was originally proposed for use with true shales o r
"lutites" (Nesbitt & Young, 1982). I n a s e q u e n c e where true shales are rare, such as the section cored at CRP-3, care must be taken to consider the potential effect of grain size variations on stratigraphic trends in the CIA. T h e provenance effect i s particularly i m p o r t a n t if s e d i m e n t p r o v e n a n c e c h a n g e d significantly during d e p o s i t i o n of a stratigraphic sequence, and if any of the potential sediment sources
lias a n unusual geoclicniical c o m p o s i t i o n . Siirli :I
provenance cffcct must be considered f o r ('RI' ,3 because potential source rocks include t w o b a s i c igiicons units, the McMiirdo Volcanic Group and thc Ferrar Dolerite, whose bulk geochcniis~ries produce C I A values lower than t h e C l A s of i i n w c i i ~ l i c ~ ~ ~ c d feldspar.
Die AhOi/TiO, ratio of a sediment can serve as ii
preliminary indicator of that sediment's source rock composition (Nesbitt, 1979; Young & Nesbitt, IOOS) for two reasons: 1) t h e ratio varies m a r k e d l y i n primary igneous rocks, from approximate1 y 10 I'or basalts and gabbros to approximately 47 for granitrs ( L e M a i t r e , 1 9 7 6 ) , a n d 2) AI and Ti a r c bolli considered to be relatively immobile u n d e r most weathering regimes. Trace element abundances can a l s o serve as v a l u a b l e indicators of s c d i n i e n t provenance because trace elements are also rclativdy immobile d u r i n g w e a t h e r i n g , and b e c a u s e triice element abundances can vary significantly bctwecn two igneous or metamorphic bodies with relatively similar major element compositions (e.g., two granites can have s i g n i f i c a n t l y different trace c l e m e n t compositions). In this study, concentrations of the trace element Nb are used to evaluate t h e relative importance of input f r o m t h e McMurdo Voletinic G r o u p , a potential s o u r c e rock with elevated NI) contents, and S1- contents are used to evaluate the relative inputs of "basement" (Granite H a r b o u r Intrusives) vs. the Ferrar Dolerite and the Beacon Supergroup.
MATERIALS AND METHODS
In this s t u d y 8 5 f i n e g r a i n e d s a m p l e s were a n a l y z e d f o r m a j o r a n d t r a c e e l e m e n t s b y x-ray fluorescence (XRF; Tab. 1). Sixty-five of the analyzed samples were bulk c o r e collected from t h e finest grained lithologies (mudstones and siltstones) between 22 to 789 msbf; in relatively thick intervals of fine- grained sediments, samples were taken approximately every 5 m. Twenty of the samples were splits of the
<63pm size fraction that remained after foraminifera processing; each of these is indicated by an asterisk in table 1.
T h e X R F a n a l y s e s w e r e m a d e at New Mexico Tech, using procedures similar to those described by H a l l e t t a n d K y l e ( 1 9 9 3 ) w i t h s o m e m i n o r modifications. Major elements and S and Cl were analyzed on all samples using glass disks formed by fusing 1 gram of sample with 6 grams of a lithium borate flux (35.3% lithium tetraborate, 64.7% lithium metaborate) in a 95%Pt/5%Au crucible at 1100 ¡C Trace elements were determined on 6 3 samples by XRF (V, CS, Ni, Cu, Zn, Ga. As, Rb, Ss, Y, Zr, Nb, MO, Ba, Pb, T h , U) using pressed powder samples (Norrish & Chappell, 1977). The XRF was calibrated using a wide variety of well-analyzed rock standards.
Tab. I - l-'.lcinrutal abundances. loss-on-ignition (i.01). total analy/.o.l ahuntianccs. C I A v;ilins, ;m\ Al-oxicle/'l'i-oxitic ratios for samples from CRP 3. "'"" indicates sample o t 4 3 HITI fisiiction scp;~r;i~c(f clm'in: j~roccssing tor tnicrof'ossils, IVIit,ior element ab~inclcinccs arc weight
'/c of the oxide: trace clement abundances are i n ppm.
Several rock standards were used to m o n i t o r t h e analytical precision, and proficiency tests administered by t h e International Association of G e o a n a l y s t s provide data on the analytical accuracy.
The primary goal of this paper is to examine the CIA index, so no attempt is made here to examine all of the major and trace element data from CRP-3. All the data are listed in table 1 so as to b e useful to o t h e r investigators. T h e A120i, TiO, N b , a n d S r analyses are used h e r e in a prelimicary effort t o
evaluate the effect of sediment provenance changes on the CIA record from CRP-3.
In calculating CIA values, the CaO* abundance used is only the C a O c o n t a i n e d in t h e silicate fraction. For the vast majority of CRP-3 samples, the C a O content of the silicate fraction is assumed to e q u a l the C a O c o n t e n t of t h e bulk s a m p l e ; i.e., biogenic and diagenetic carbonates and biogenic apatite are assumed to contribute little or no CaO to the bulk s a m p l e . T h e C a O c o n t r i b u t i o n s from
hiogenic and diagenetic carbonates for these samples :ire j u d g e d to be low because: 1 ) measured C a O contents for the majority of samples arc consistent with the CaO contents of most potential source rocks (< 5 wt. %; Roser & Pyne, 1989). and 2 ) "loss-on- ignition" (LOI) values for the ma.jority of samples also are relatively low (Tab. I). The exceptions are four samples with C a O contents greater than 5%:
CIAs for these samples have not been included in this study because the elevated CaO contents suggest a contribution from non-silicate sources. Carbonate a n a l y s e s w e r e not conducted d u r i n g initial characterization of the CRP-3 cores. but diagenetic c a r b o n a t e s w e r e identified during visual c o r e d e s c r i p t i o n , particularly below 234 mbsf ( C a p e Roberts Science Team, 2000). Post-drilling analysis of carbonate contents confirms this depth distribution.
but a l s o s h o w s that c a r b o n a t e c o n t e n t s vary significantly over short stratigraphic intervals (Dietrich and others, this volume). In addition, the diagenetic carbonates were concentrated in sandstone lithologies, which were not sampled for this study. s o the low L 0 1 values a n d s o u r c e - c o n s i s t e n t C a O c o n t e n t s remain the best indicators of low CaO contributions from carbonates to most samples in this dataset. In s o m e other studies ( e . g . , F e d o et al., 1995). C a O contents also have been corrected for a contribution from biogenic apatite by assuming that all PTO, is present as biogenic apatite. Such a correction has not been made for the CRP-3 samples because the P 7 0 , is uniformly low (<0.2 wt % throughout CRP-3), and because the McMurdo Volcanic Group is a potential source rock and is usually enriched in P.,0,.
B e c a u s e a detailed a g e - d e p t h m o d e l i s n o t presently available for CRP-3, all data are presented as stratigraphic profiles vs. subbottom depth.
DATA AND RESULTS
The complete data set, the calculated CIAs, and the calculated Al,OJTiO, ratios for the 8 5 samples f r o m C R P - 3 a r e p r e s e n t e d in t a b l e 1 . T h e stratigraphic profiles of the calculated C I A s , t h e c a l c u l a t e d A 1 2 0 J T i 0 , r a t i o s , and t h e N b a n d S1- c o n t e n t s a r e s h o w n in f i g u r e s 1 , 2 , 3 , a n d 4 , respectively.
CIA values in C R P - 3 (Fig. 1) generally r a n g e b e t w e e n 4 5 a n d 6 0 , and t h e C I A p r o f i l e c a n b e subdivided into two parts: 1) a lower part (below -400 mbsf), where CIAs are relatively high at 750- 800 mbsf, low at -600 mbsf, and increase irregularly toward 400 n ~ b s f , and 2) an upper part (above -400 mbsf), where an overall upsection decrease in C I A values is overprinted by four smaller, quasi-cyclic c h a n g e s , e a c h o c c u r r i n g over 5 0 - 1 0 0 m o f stratigraphic thickness.
On first examination, the low CIA values that are distributed t h r o u d ~ o u t the CRP-3 mofile s u r e s t that
this site c o n s i s t e ~ i i l y received sediment t h ; i l liinl undergone little o r no chemical weathering chiri~i!:, (lie early Oligoccnc (and late Eocene?). This i n t c i ~ p i v ~ r i l input of unwcathered o r weakly weathered material is i n agreement with thc importance of glacij~,enic or glacially inf'lucnccil lithofacies througlnnit thc Ccnozoic portion of CRP-3. However, C I A ~ i i l n e s close to. or less 111:in. 50 occur t h r o u g h o ~ ~ t ('RP .1;
since 50 is the CIA value cited f o r ~ ~ n w c i i t l i e i ~ c d feldspar by Ncsbitt & Young (1982). the presence of values less than 50 i n CRP-3 indicates that primary phases with original K/A1, CaIAl, or NaIAI ratios higher than those found i n feldspars must b e prcseni.
Based on t h e s e low C l A s , and r e i n f o r c e d h y conclusions drawn from studies of CRP-l a n d ('RI'- 2/2A (Krissck & Kyle. 1998, 2000), the possible effects of provenance changes on the CIA profile for CRP-3 must be considered before the CIA profile is interpreted solely as an indicator of palaeowcathei.ing.
For example, the overall upsection decrease i n ClAs above -400 mbsf in CRP-3 could be produced either by a decrease in the amount of weathering or by ;m
CRP3
Smoothed
CIA40
w c , m50 _, LE* LF 70
l l
Fig. 1 - CIA profile for CRP-3. smoothed with a 5-point moving average and plotted vs. subbottom depth. Labeled arrows indicate CIA values calculated for potential source rock types from data of Roser & Pyne (1989): M V G = McMurclo Volcanic Group. FD = Fei-rar Dolerite. L F = Lashley Formation. LB = Lower Beacon S u p e r g r o u p s e d i m e n t s . a n d B S = c r y s t a l l i n e b a s e m e n t , N o t e irregular CIA profile below -400 mbsf. short-term fluctuations ("quasi-cycles") above -400 mbsf, and overall decrease i n CIA
L- -U values above -400 mbsf.
(icocliemical Indicators of Weathering, Ccnor.oic Palaeoclima~es, and Provenance in Fine-Grained Sediments 565 increase i n the relative importance of material whose
low "app;isent CIA" is independent of its weathering history. I n a similar manner, the four CIAL'quasi- cycles" above 400 mbsf may also record provenance changes, palaeoweathering changes, or a combination of these two.
Roser & Pyne (1989) summarized the representative geochemical compositions of six source r o c k types thought to have supplied sediment to CIROS- I : because CRP-3 and CIROS- 1 are located only 70 km apart and are in similar geologic settings, similar source rock types can be expected to have supplied sediment to CRP-3. As a result, the source r o c k compositions summarized by Roser & Pyne (1989) are used here to examine the potential effects of provenance changes on the CIA record at CRP-3.
The six potential source rock types are: 1) basement rocks (e.g., granitoids of ~ e r r a r Valley), 2) lower Beacon Supergroup sediments (Weller Coal Measures through the Windy Gully Sandstone), 3) Ferrar Dolerite, 4) McMurdo Volcanic Group basanites, 5) McMurdo Volcanic Group trachybasalts to trachytes, and 6) Lashly Formation sediments (upper Beacon Supergroup sediments). Selected major element oxide abundances for these six source rock types are listed in table 2, together with the resulting C I A s and Al2O3lTiO., ratios, and their Sr and Nb contents; these values are also shown and labeled on figures 1, 2, 3, and 4.
The relative importance of these sediment sources t o deposition at CRP-3 did vary during the early Oligocenellate E o c e n e ( ? ) , as indicated by the stratigraphic profile of A120JTi02 ratios (Fig. 2). The A120/Ti02 ratios at CRP-3 show little or no trend o v e r the entire stratigraphic interval, instead fluctuating repeatedly between
-
18 and -23. These variations are best-developed as three or four "quasi- cycles" above -400 mbsf. Comparing the A120JTi02 values of the potential source rock types (listed in table 2 and shown in figure 2) to the CRP-3 profile indicates that these sediments must contain little or no M c M u r d o Volcanic G r o u p detritus, w h i c h is characterized by very low A1203/Ti0, ratios (between 3.6 and 11). The paucity of ~ c ~ u r d o Volcanic Group detritus in C R P - 3 is a l s o indicated by low concentrations of the trace element Nb (Fig. 3). Small amounts of McMurdo Volcanic Group detritus may be present above -50 mbsf, but Nb contents below that level in CRP-3 are consistent with sediment supply s o l e l y from t h e F e r r a r Dolerite, the BeaconSmoothed CRP3 A12031Ti02
MVG 15 20 BS,LB,LF 25
l
\ \Fig. 2 - A l , O / T i O ratio profile for CRP-3. smoothed with a 5 - point moving average and plotted vs. subbottom depth. Labeled arrows indicate Al,O,/TiO, ratios calculated for potential source rock types from data of do er & Pyne (1989); MVG = McMurdo Volcanic Group. FD = Ferrar Dolerite. L F = Lasliley Formation.
LB = Lower Beacon Supergroup sediments. and BS = crystalline basement. Note irregular Al,O,/TiO, profile below -400 n~bsf-.
short-term fluctuations ("quasi-cycles") above -400 mbsf, a n d absence of overall trend in values above -400 mbsf.
Supergroup, and the Granite Harbour Intrusives. T h e absence of McMurdo Volcanic Group detritus also was noted during initial studies of CRP-3 clasts and sandstone framework grains (Cape Roberts Science Team, 2000).
I n the a b s e n c e of M c M u r d o Volcanic G r o u p detritus, the stratigraphic profile of A1203/Ti02 ratios (Fig. 2) c a n b e interpreted to record r e p e a t e d variations in relative inputs from the Ferrar Dolerite (with an A120,/Ti0, ratio of -19) and the Beacon Supergroup
+
"basement" (Granite Harbour Intrusives;both with A1203/Ti02 ratios of 22-24). S r contents (Fig. 4) help to distinguish inputs from the Beacon
Tab. 2 - Selected geochemical data for CIROS-l sediment source rocks (from Roser & Pyne. 1989). Major element abundances are weight % of the oxide; trace element abundances are in ppnl.
Source Terrane Si02 A I 2 0 3 Ti02 CaO N a 2 0 K 2 0 P205 CIA A12031Ti02 Nb Sr
Basement 65.25 16.14 0.72 3.88 3.63 3.69 0.17 48.7 22.4 13 516
Loner Beacon 89.73 5.7 0.24 0.68 0 2 1.44 0.02 64.6 23.8 5 47
Fenar Dolerite 57.17 15.83 0.83 8.67 2.65 1.28 0.11 42.4 19.1 8 144
McMurdo Volcanic 42.69 13.83 3.82 10.78 3.53 1 49 0.86 33.9 3.6 114 970
McMiirdo Volcanic 52.76 19.1 1.73 5.06 7.23 3.86 0.58 43 l l 204 951
Lashly Formation 77.86 12.7 0 53 l .06 1.72 2.26 0.03 63.7 24 12 147
Fig. 3 - Nb concentrations with depth in CRP-3. Labeled arrows indicate Nb contents for potential source rock types from data of Roser & Pyne (1989): MVG = McMurdo Volcanic Group. FD =
Ferrar Dolerite. L F = Lashley Formation. LB = Lower Beacon Supergroup sediments, and BS = crystalline basement. Low Nb contents indicate little or no supply of McMurdo Volcanic Group detritus to CRP-3.
Supergroup vs. those from "basement", and indicate that detritus from the Granite Harbour Intrusives is most important above -200 mbsf. Below that level, the low S r c o n t e n t s ( g e n e r a l l y - 1 0 0 p p m ) a r e consistent with sediment supply dominated by the Ferrar Dolerite and the Beacon Supergroup. Taken together, t h e A 1 7 0 3 / T i 0 , r a t i o s a n d S s c o n t e n t s suggest that sediments above -200 mbsf in CRP-3 were supplied from all three possible source rock types: the Ferrar Dolerite, the Beacon Supergroup, and the Granite Harbour Intrusives. Sediments below -200 mbsf in CRP-3, however, appear to have been supplied primarily from only two of these sources, the Ferrar Dolerite and the Beacon Supergroup.
For comparison to t h e s e geochemically based interpretations of sediment provenance, preliminary studies of clasts in C R P - 3 (Cape Roberts Science Team, 2000) identified a "lower clast assemblage", present below -150 mbsf and dominated by Beacon Supergroup and Ferrar Dolerite clasts, and an "upper clast assemblage", present above -150 mbsf a n d dominated only by Ferrar Dolerite clasts. "Basement"
clasts were observed discontinuously through the C e n o z o i c s e c t i o n , g e n e r a l l y i n low a n d v a r i a b l e
Fig. 4 - Sr concentrations with depth in CRP-3. Labeled arrows indicate Sr contents for potential source rock types from clatii of Roser & Pyne (1989): MVG = McMui-do Volcanic Group. I-'D = Ferrar Dolerite, L F = Lashley Formation. LB = Lower Beacon Supergroup sediments. and BS = crystalline basement. Increased Sr contents above -200 mbsf indicate increased i m p o r t a n c e ol' basement-derived detritus above that level.
proportions. The composition of sandstone framework grains also records the importance of Ferrar-derived material, especially above -180 mbsf, and input from the Beacon Supergroup (Cape Roberts Science Team, 2000; Smellie, this volume). In summary, then, all t h r e e d a t a s e t s ( m u d r o c k g e o c h e m i s t r y , clast c o m p o s i t i o n , a n d s a n d s t o n e f r a m e w o r k grain composition) indicate both the importance of the Ferrar Supergroup as a sediment source throughout the d e p o s i t i o n of C R P - 3 , and the i m p o r t a n c e of Beacon Supergroup sources below 150-200 mbsf.
These three data sets provide a less consistent picture of inputs from the Beacon Supergroup above 150-200 mbsf a n d f r o m t h e basement. T h e s e a r e inconsistencies in describing less important sources, however, a n d may arise f r o m either: 1 ) t h e p r e l i m i n a r y n a t u r e of t h e s e d a t a s e t s a n d their interpretations, or 2) differences in sample locations superimposed on variable inputs from these sources.
A s m o r e d e t a i l e d s t u d i e s are c o m p l e t e d , a comprehensive interpretation of sediment provenance will b e an important step toward understanding the tectonic a n d u n r o o f i n g histories of the a d j a c e n t Transantarctic Mountains.
T h e s t r a t i g r a p h i c variations i n scdiment provenance at CRP-3, identified from (lie A1,0,/Ti07 ratios (Fig. 2) and the Sr contents (Fig. 4), might be expected to also affect the CIA profile (l"ig. I ) . since unweathered Beacon Supergroup material has a higher apparent CIA (-64; Tab. 2) than unweatliered F e i ~ ~ r Dolerite (-42) or unweatlierecl basement (-49). Below -400 mbsf, however, the CIA and Al,O-/TiO, profiles show little covariation, suggesting that the observed CIA fluctuations are a conlplex signal, influenced both by s h i f t s in p r o v e n a n c e and by c h a n g e s in palaeoweathering. Deciphering these two influences, in order to develop a p a l a e ~ w e a t l i e ~ i n g curve for the lower 400 m of the CRP-3 Cenozoic sequence, will require m o r e detailed compositional modeling to remove the overprint of provenance; this will be the objective of future work. Above -400 mbsf, however, the C I A and A120,/Ti0., profiles show strikingly similar p a t t e r n s of s h o r t - t e r m variation ("quasi- cycles"), c l e a r l y r e c o r d i n g r e p e a t e d c h a n g e s in sediment provenance that strongly influence tlie CIA record. The cause(s) of the provenance changes is not known at present, but possibilities include episodic uplift in the Transantarctic Mountains and glacial advancelretreat. These possible explanations will be evaluated after the profiles of contributions from the three possible source rock types are calculated using linear mixing models. In addition, changes in source rock contributions will be compared to independent indicators of tectonism or palaeoenvironment, such as sequence stratigraphic boundaries.
A b o v e - 4 0 0 m b s f i n C R P - 3 , t h e C I A a n d A120JTi02 profiles show strikingly similar patterns of short-term variation; the overall trends of these two profiles above -400 mbsf, however, show a subtle, but perhaps important, difference. In the CIA profile, the "quasi-cycles" a r e superimposed on a general decrease in C I A values upsection; the "baseline"
value at -400 mbsf is -55, whereas the "baseline"
value above the shallowest "quasi-cycle" is 50 or less.
In contrast, the "baseline" for the "quasi-cycles" in the Al,OJTiO, profile remains relatively constant t h r o u g h o u t t h e i n t e r v a l a b o v e - 4 0 0 m b s f . T h i s d i f f e r e n c e in l o n g - t e r m b e h a v i o r s u g g e s t s t h a t palaeoweathering intensities were decreasing during deposition of the uppermost 400 m of CRP-3, even as shorter-term, but relatively consistent, provenance c h a n g e s a l s o i n f l u e n c e d t h e C I A r e c o r d . If t h i s h y p o t h e s i s i s c o r r e c t , t h e n t h e d e c r e a s e in palaeoweathering may record climatic deterioration (cooling andlor drying) between -33 M a and -31 Ma.
Testing this hypothesis will require more detailed compositional modeling to remove the overprint of provenance changes from the CIA record.
SUMMARY AND CONCLUSIONS
Sediments deposited at CRP-3 during the early Oligocenellate E o c e n e ( ? ) w e r e supplied f r o m t h e
k r r t i r Dolerite, tlie l3e;icon S u p c r g r o ~ ~ p , and local basement (tlie (;r;ini(c Iliirboiir I i i t r ~ ~ s i v e s ) ; t h e McMurclo Volcanic G r o u p supplied little o r n o nititerial. Elevated Sr contents suggest that basement- derived material is most important above -200 n ~ b s f , whereas Beacon Supergroup andlor Fcrrar Dolerite components dominate below that level. Below -400 nibsf, the CIA and AI.,03/Ti09 profiles show little or n o covariation, s u g ~ e s t i n g t t h a t the C I A record is influenced by both provenance and palaeoweathering.
Above -400 inbsf, however, the CIA and Al,O/riO, p r o f i l e s show strikingly s i m i l a r s h o r t - t e r m fluctuations, indicating a strong provenance influence on t h e C I A record. T h e c a u s e of s u c h repeated provenance changes, howcver, is presently unknown.
CIA values exhibit a general decrease when viewed over the entire interval above -400 mbsf. whereas A1203/Ti0, values remain uniform. This difference suggests that palaeoweathering intensity decreased during deposition of the upper 400 nl of CRP-3, as w o u l d be e x p e c t e d d u r i n g a t i m e of c l i m a t i c deterioration (cooling andlor drying). A more detailed i n t e r p r e t a t i o n of t h e p a l a e o w e a t h e r i n g a n d p a l a e o c l i m a t i c c o m p o n e n t of t h e C I A r e c o r d , however, will only be possible after the overprint of provenance changes has been removed usingmixing models and comprehensive analyses of major element and trace element data.
ACKNOWLEDGEMENTS - Lawrence Krissek was supported by Office of Polar Programs, NSF grant OPP- 9527008. Philip Kyle was supported by Office of Polar Programs. NSF grant OPP-9527329. The XRF facility at New Mexico Tech was partially funded by NSF grant EAR- 93 16467. We gratefully acknowledge the efforts of the Cape Roberts Project drillers and core processors. the Antarctica New Zealand support staff at the Cape Roberts camp, and the Antarctic Support Associates personnel at McMurdo Station. B. Roser. T.C. Homer, and W. Ehrmann provided helpful reviews of this manuscript.
REFERENCES
Barrett P.J.. 1997. Cape Roberts Project Science Plan. Antarctic Data Series. no. 20, Victoria University of Wellington. 59 p.
Barrett P.J. & Davey F.J.. 1992. Cape Roberts Project Workshop Report. Royal Society of New Zealand, Miscellaneous Series, 23. 38 p.
Cape Roberts Science Team, 2000. Studies form the Cape Roberts Project, Ross Sea Antarctica. Initial Report on CRP-3. T e r m Antartica, 7. 1-209.
Dietrich H.-G.. Klosa D. & Wittich C., 2001. Carbonate content in CRP-3 drillcore. Victoria Land Basin. Antarctica. This volume.
Fedo C.M.. Nesbitt H.W. & Young G.M.. 1995. Unraveling the effects of potassium metasomatism in sedimentary rocks and palaeosols. with in~plications for palaeoweathering conditions and provenance. Geology. 23. 921-934.
Hallett, R.B., & Kyle, P.R. 1993. XRF and INAA determinations of major and trace e l e m e n t s in Geological S u r v e y of J a p a n i g n e o u s and sedimentary rock s t a n d a r d s . G e o s t a n d a r d s Newsletter. 17. 127-1 33.
A'oliirr 299. 715-717.
N o r r i s l ~ K . & Cl,appcll L W . , 1977. X-rziy I'limrcscencc s ~ ~ ~ c I ~ ~ ~ ~ I ~ ~ I ~ ~ , 11,: 7 < 1 s s n w n J , (e<l.). sic^^/ Mel/?o</.s i u D e n - n i t i i n i t i r r M i m f a l n ~ y . 2ml al. Acinlemic Press. New York, 2 10.272.
Roser B.P. & I'yne A.R.. 1989. \Vliolcrock geochemislry. In: H$irrclt P.J. ( c d . ) , A n ~ : i r c t i c C m o z o i c I l i s l o r y l'rom Hie C I R O S - l D r i l l l i o l r M c M u ~ l o Soiuul, Aillarclicst. IjSIIi B t i I I e ~ i n , 245.
175-184.
S m c l l i e J.1-., 3 0 0 1 . H i s l o r y o f Oligocene erosion. uplil'l and u n r o o f i ~ ~ g <)l' t l' ~ r ~ ~ ~ ~ s ~ ~ ~ ~ l i ~ r c t i c ~ h4ottt11ai11s < l c < l ~ ' c d f r o ~ n S ~ I ~ ~ S I ~ ~ I I C d c ~ r i t : % I modes i n ClZ1'-3 d r i l l c ~ ~ r ~ . V i c l o r i s ~ I . n ~ l 13zvi11, A ~ ~ l ~ t ~ c l i c : ~ . This ~ ~ ~ ~ I ~ ~ ~ I ~ c . Young (;.M, & N e s b i t l H.W., 1998. Processes c o n l r o l l i n g l l i c
~ l i s ~ ~ i l ~ ~ t l i ~ ~ ~ , or 'K :m1 A I i n wc>~lI,crir,g ]profiles, siIicic1:istic sedimmls and sedimmlary rocks. J. . Y d . Resran-h. 6 8 . 448- 455.
