Geochronologic evidence for Early Cretaceous volcanic activity on Barton Peninsula, King George Island,

Hyeoncheol Kim, Jong Ik Lee, Moon Young Choe, Moonsup Cho, Xiangshen Zheng, Haiqing Sang & Ji Qiu

Ages of six volcanic and plutonic rocks on Barton Peninsula, King George Island, were determined using “Ar/39Ar and K-Ar isotopic systems. The “Ar/”Ar and K-Ar ages of basaltic andesite and diorite range from 48 My to 74 My and systematically decrease toward the upper stratigraphic section. Two specimens of basaltic andesite which occur in the lowermost sequence of the peninsula, however, apparently define two distinct plateau ages of 52-53 My and 119-120 My. The latter is interpreted to represent the primary cooling age of basaltic andesite, whereas the former is interpreted as the thermally-reset age caused by the intrusion of Tertiary granitic pluton. The isochron ages calculated from the isotope correlation diagram corroborate our interpretation based on the apparent plateau ages. It is therefore likely that volcanism was active during the Early Cretaceous on Barton Peninsula. When the K-Ar ages of previous studies are taken into account with our result, the ages of basaltic andesite in the northern part of the Barton Peninsula are significantly older than those in the southern part. Across the north-west- south-east trending Barton fault bounding the two parts, there are significant differences in geochronologic and geologic aspects.

H. Kim. Isotope Research Team, Koreo Ba.tic Science Institute. Yeo-Em Dorig 52, Rtsung-Ku. Tuejon 305-333, South Korea; J. I. Lee & M. Y. Ciioe. Polar Science Lcihornfor?,, Korea Oceari Research & De~~elopitieniInstitute, Str-dorrg 12 70, Anxiti 425- 170, South Korea; M. Clio, Dept. of Geological Scrmces, Seoul Nutionnl Uiiiwrsitv. Seoiil 151-742, South Koreu; X. ZlienR, H. Sang & J. Qiu. Institute of Geology. Chinese Acndernq of Sciences, Beijirig 100029. Cliiiiu.

The . situated along the sandstones and conglomerates, then by basalt to Pacific margin of the Peninsula (Fig. I ), dacite bas interbedded with andesitic and represent a magmatic arc of Jurassic to Quatern- rhyolitic ignimbrites (Smellie. Pankhurst et al. ary age. Volcanic sequences and plutonic intru- 1984; Arche et al. 1992; Willan et al. 1994; sions in these islands built upon a pre-volcanic. Smellie, Liesa et al. 1995; Hathway & Lomas sialic basement of pre-Jurassic schists and 1998). deformed sedimentary rocks. The latter is pri- King George Island is the largest island in manly composed of late Paiaeozoic-early Meso- the South Shetland Islands and mainly consists zoic turbiditic submarine fan deposits (Smellie. of volcanic and plutonic rocks of calc-alkaline Liesa et al. 1995). Marine sedimentary rocks affinity. Petrography and geochemistry of King containing Late Jurassic-Early Cretaceous fossil George Island have been well documented by faunas are succeeded by terrestrial volcanic previous workers (Barton 1965; Birkenmajer

Kim et al. 2000: Polur Research 19f21, 251-260 25 1 Fig. I. Structural elements of King George and Nelson islands (after Birkenmajer 1983)

1983: Smellie, Pankhurst et al. 1984: Tokarski biotite ages (60-63 My) for mafic plutonic rocks. 1988; Kang & Jin 1989; Lee et al. 1996). Recent geochronologic data from Barton Peninsu- Ferguson (192 I) has classified the stratiform la, however, suggest Eocene volcanic and plutonic volcanic complexes of King George Island into an activity (Park 1989; Lee et al. 1996). K-Ar whole- older and a younger series, based on the observa- rock ages of volcanic rocks range from 35.5 to tion that the Andean intrusive suite intrudes the 48.5 My, whereas those of plutonic rocks mostly older series but never the younger series. The older from 42.1 to 45.2 My (Park 1989). K-Ar biotite series was tentatively correlated with Jurassic ages of plutonic rocks are in the range of 41.2- rocks of . whereas the younger series 41.9My (Lee et al. 1996). Because of intrinsic with Cenozoic rocks (Tyrrell 1921). The temporal problems, such as the Ar loss in K-Ar isotopic gap in these volcanic complexes was confirmed by ages, further discussion on the apparent discre- Birkenmajer (1980a-d): Cardozo Cove and Martel pancy is not warranted. Inlet Groups at Admiralty Bay (Fig. Ic), pre- In this paper, we present “Ar/39Ar as well as K- dating the Andean intrusions, are correlated with Ar ages of volcanic and intrusive rocks on Barton the older series of Ferguson (19211, whereas the Peninsula, King George Island. “Ar/”Ar ages King George Island Supergroup correlates with the provided us with new geochronologic constraints younger series. Furthermore, based on lithology not only for Cretaceous volcanic activity on King and stratigraphy, Birkenmajer (1983) divided King George Island but also the temporal relationship of George Island into three tectonic blocks bounded volcanism in the South Shetland Islands. by strike-slip faults (Fig. Ic). The Andean intru- sives are restricted to the central block, or the so-called “backbone”, corresponding to the up- thrown Barton Horst to which Barton Peninsula Geological setting and sample belongs (Birkenmajer, Francalanci et al. 1991). description Birkenmajer (1983) and Kang & Jin (1989) reported Palaeocene K-Ar whole rock and Rb-Sr Barton Peninsula is composed of volcanic and

252 Geochronologic evidence for Early Cretaceous volcanic activity on Barton Peninsula Fig. 2. Geologic map of Barton and Weaver peninsulas, King George Island, Antarctica.

Marian Cove

a Dhnite

Agglomrate I Tuffaceous U 4 sandstone and s111Stom R LOWW basalt

-Fault -

plutonic rocks together with minor thin-bedded The agglomerate occurs as a ca. 10 m thick layer sandstone and siltstone (Fig. 2). Volcanic rocks in along the southern coast of the peninsula. Both the peninsula mainly consist of basalt, basaltic agglomerate and lapilli tuff contain a variety of andesite, agglomerate and lapilli tuff, whereas volcanic glasses and breccias, but lithic fragments plutonic rocks include granodiorite and diorite. of crystalline rocks are absent. Acidic dykes are common in diorite. Fine-grained Basaltic andesites overlain by the lapilli tuff are diorite, corresponding to the lower part of granitic widespread in the peninsula and generally por- pluton, consists of plagioclase, clinopyroxene, phyritic in texture. Phenocrysts consist primarily hornblende, biotite and Fe-Ti oxides with minor of plagioclase together with minor clinopyroxene quartz. Medium-grained granodiorite, representing and orthopyroxene. Holocrystalline groundmass of a main body of the pluton, is composed of basaltic andesite commonly shows flow textures plagioclase. quartz, alkali feldspar, hornblende, defined by the preferred orientation of plagioclase biotite and Fe-Ti oxides. The ubiquitous occur- microlites. rence of miarolitic cavities indicates that this Volcanic rocks on Barton Peninsula are affected pluton was emplaced at a shallow level in the by hydrothermal alteration (So et al. 1995) and upper crust. Petrological and geochemical studies low-temperature thermal metamorphism (Kim et by Lee et al. (1996) further suggest that the al. 1995). Kim et al. (1995) defined two meta- plutonic rocks are vertically zoned. The pluton is morphic zones, changing from a calcite-chlorite bounded to the south-west by the north-west- zone to a amphibole-chlorite zone toward the south-east trending Barton fault, which apparently pluton on the peninsula. Mineral assemblages of penetrates the whole peninsula. metavolcanics are represented by epidote + chlor- Stratification of volcanic rocks is well-pre- ite f hornblende f actinolite f calcite (+plagio- served to the north of the pluton along the sea clase, quartz, opaque minerals), suggesting the cliff. Bedding planes of the volcanic sequences greenschist to lower-amphibolite facies meta- generally strike north-east or north-west and morphism. Further details on thermal metamorph- consistently dip south at 1040’ NE or NW (Kang ism will be reported elsewhere (Kim et al. 2000). & Jin 1989). The lapilli tuff mainly occurs to the We have analysed three basaltic andesites and north of the Barton fault and locally to the south. one diorite for the “Ar/”Ar age determination and

Kim et al. 2000: Polar Research 19(1),251-260 253 two basaltic andesites for the K-Ar age determina- CuO and sponge titanium and then introduced tion. All samples were collected from the northern directly to RAG-10 mass spectrometer for argon part of the Barton fault. Fresh samples were isotope analysis. Apparent ages were corrected for chosen to minimize the effect of argon loss or mass discrimination, interference of K and Ca to contamination, but most of these samples con- Ar isotope, and "Ar radioactive decay. The tained small amounts of secondary minerals such measured correction factors for the interference as chlorite and white mica. In addition, all of of K and Ca to Ar isotope are ("Ar/39Ar)K amphibole crystals of diorite are altered comple- = 3.05 x lo-', ("Arl'7Ar)Ca = 2.64 x lop4 and tely to actinolite. ("AI-/~'A~)~~= 6.87 x lo-' (Hu, Wang et al. 1985). A half life of 35.1 days was adopted to correct the radioactive decay of "Ar. Uncertainty Analytical procedure is quoted as one sigma and does not include that of the calculated J-factor. Isochrons were regressed "hrt"'Ar method following the method of York (1969). The analysis for "'Ar/"Ar age determination was K-Ar method carried out at the Institute of Geology, Chinese Academy of Sciences, following the procedures Quantitative analysis of argon was performed by described by Hu, Wang et al. (1985). The samples isotopic dilution method using a "Ar spike at the were ground into sizes of 60-80 mesh and Institute of Geology, Chinese Academy of packaged in aluminum foil together with the Sciences. Samples were fused at 1450 C-15OO'C flux monitors. The flux monitors include ZBJ for 40 minutes. CuO and sponge titanium were hornblende (132.8 f 3.1 Mya), ZBH-25 biotite used as purifiers and 38Ar spike was added during (132.7 f 2.8 Mya) and international standard sam- the purification process. The purified argon was ple, BSP-I hornblende (2060 f 18 Mya). Both directly introduced into RAG- 10 mass spectro- samples and flux monitors were evacuated and meter for isotopic analysis. Potassium concentra- sealed hermetically in quartz ampoules. K2S04 tion was measured by a flame photometer. and CaF2 were added into each package to monitor The decay constants and isotopic abundance the interfering reactions "K(n,p)"Ar, ratio of K used in the age calculation are as 11 K(n.a)40Ar. "OCa(t~,a)~~Ar,4"Ca(n,na)36Ar and follows: A = 5.543 x 10-'oy-', I.,{= 4.962 x "Ca(n,a)"Ar. Then, two basaltic andesite (BPO5 10-'oy-l, A, = 0.581 x IO-'"y-'. 'OK/K = and BP13) samples and one diorite (BP03) sample 0.01 167 (atomic %) (Steiger & Jager 1977). were put in the B4 channel of 49-2 Reactor for 4320 minutes and irradiated with fast neutron flux (0.56 x n/cm2sec). The total amount of the Results integrated fast neutron flux is 1.45 x 10l8 dcm'. On the other hand, one basaltic andesite sample The K-Ar whole-rock ages of two basaltic (BP04) was placed in the B8 channel of 49-2 andesites are listed in Table 1. The K-Ar age of Reactor for 3174 minutes and irradiated with BP09 is older than that of BP12, which was instantaneous fast neutron flux (4.22 x 10" n/ collected from a stratigraphically upper part near cm'sec) and the total amount of the integrated the Barton fault (Fig. 2). The '"Ar/"Ar incre- fast neutron flux is 8.04 x 10" n/cm2. A 0.5 mm mental heating data and ages of basaltic andesite thick cadmium foil was used as a shield to prevent and diorite are listed in Table 2. The results of the interference of slow neutrons. During the 40Ar/39Ar whole-rock ages are shown in Fig. 3. irradiation, the sample box was cooled by water, Isotope correlation ages of basaltic andesite 42°C at the outlet. and was rotated 2 to 8 times per (BP04, BP05 and BP13) range apparently from minute to eliminate a transverse gradient of 53 & 2 My to 1 18 f 4 My, whereas that of diorite neutron flux. (BP03) is 48.8 -+ 1.5 My. After cooling to a safe dosage, the irradiated Specimen BP04 records an internally discordant samples were put in an Ar-extraction system to 40Ar/39Ar age spectrum that apparently defines carry out step heating analysis. A high frequency two "plateau" ages: 52 ? I My from low-tempera- oven was used to heat the samples at each step for ture fractions (steps 2 to 4) and 119 +. 1 My from 20 minutes. The extracted argon was purified by high-temperature fractions (steps 6 to 9) (Fig. 3a).

254 Geochronologic evidence for Early Cretaceous volcanic activity on Barton Peninsula Tnhlr I. K-Ar whole-rock age\ of basaltic andesite

BPO9 0. I761 0.75 o.9xiox 68.1 0.004383 73.9 It 4.4 BPI1 0.2780 1.04 0.83859 79.8 0.002702 -15.9 * 3.8

These plateau data define, respectively. two rock ages of basaltic andesite decrease progres- isochron ages in 36Ar/40Ar versus 39ArT"'Ar sively from the stratigraphically lower part (BP04 isotope correlation diagram (Fig. 3b). Four high- and BP05) to the upper part (BP13) (Fig. 4). The temperature steps, recording similar apparent ages, apparent ages from low-temperature fractions of yield a plateau isotope correlation age of BP04 and BP05 indicate that they were partly reset 118 f 4 My with a 40Ar/"6Ar intercept value of by the emplacement of plutonic rocks about 290 f 3. These values are indistinguishable from 49 Mya (BP03). The "Ar/"Ar ages from high- the plateau age and the atmospheric ratio of 295.5, temperature fractions. however, indicate the pre- respectively. On the other hand, three or four steps sence of Cretaceous volcanic activity on Barton at temperatures of 460Y-9OO"C yield an isochron Peninsula. On the other hand, K-Ar and 4"Ar/"Ar age of 53 f 2 My which is identical to the apparent ages of BPI2 and BP03, respectively, are con- "plateau" age of 52 f I My within error range. sistent with the previously reported ages (Fig. 4), The 40Ar/36Arintercept value of 288 rf: 4 is larger suggesting an early Tertiary magmatism. than the 4oAr/3yArratio in the present-day atmos- phere, indicating the presence of minor excess 40 Ar in low-temperature fractions. Discussion and geologic implications Specimen BP05 yield a '"Ar/"'Ar apparent age spectrum with characteristics similar to those of Birkenmajer ( 1989) suggested considerable differ- BP04, defining two apparent "plateau" ages of ences in stratigraphic succession, ages and charac- 53.1 f 1.5 Myand 120.4f 1.6My(Fig.3c).These teristics of three tectonic blocks: i.e. Fildes Block, ages are consistent with the plateau ages of BP0-t. Barton Horst and Warszawa Block. Barton Horst, The old plateau age is based on three high- upthrown relative to Fildes and Warszawa blocks temperature steps comprising >70% of the 3"ArlC and more deeply eroded than the other two blocks, released. Furthermore, the CdK ratio of these steps exposes the plutonic rocks intruding into the old (Table 2) are consistently high, suggesting the volcanic strata. outgassing from actinolite-hornblende phenocrysts. Although Mesozoic volcanic activities are In the isotope correlation diagram (Fig. 3d), data suggested in the Fildes Peninsula and Admiralty points of low-temperature fractions are too concen- Bay of King George Island (Ferguson 1921: trated in a small area to define an isochron age, high- Davies 1982: Birkenmajer 1983; Birkenmajer. temperature steps yield a plateau isotope correlation Narebski et al. 1983a), no Mesozoic isotope age age of 105 k 5.5 My. Because the isotope correle- of volcanic rocks are reported from Barton tion ages do not assume any "Ar/"Arratio, they are Peninsula. K-Ar and Rb-Sr ages of volcanic rocks considered to be more reliable than the apparent in the Barton Horst apparently range from plateau ages. Hence, the 105 rf: 5.5 My age is Paleocene to Eocene (Birkenmajer, Narebski et geologically significant and is interpreted to be al. 1983a, b; Kawashita & Soliani 1988; Park close to the formation age of basaltic andesite BPOS. 1989). The K-Ar ages of volcanic specimens Specimens BP13 and BP03 record concordant collected from the southern part of the Barton fault whole-rock '"Ar/'"Ar apparent age spectra (Figs. (Park 1989) (Fig. 4) range from 35.5 to 48.5 My. 3e, g) and define the isochron ages of 60.6 f These ages are consistent with the K-Ar age of 1.4 My and 48.8 f 1.5 My, respectively. which basaltic andesite, BPI2 (45.9 f 3.8 My). are identical to plateau age within error range As mentioned above. all of the analysed speci- (Figs. 3f. g). mens were slightly to moderately altered and/or When we consider the stratigraphic relationship experiencsd low-grade metamorphism. Nonethe- of each sample, the K-Ar and 40Ar/"'Ar whole- less, the 4"Ar/3yAr ages obtained in this study are

Kim et al. 2000: Polnr- Reseorch 19(2),251-260 255 VIN Tubk 2. Incremental heating data for '"Ar/"Ar age determination of basaltic andesite and diorite. m Temp. '"Arc: '0Ar*/'9ArK 3'1 Arc: Apparent age Step ('C) (J"Ar/"Ar)M (ZhAr/39Ar)k, (77Ar/3"Ar)M ('xAr/74Ar)M (10- l2 mol) (+I u) (%) (*I c, My)

B PO4 1 460 3 I S380 0.1026 I .974 1 0.1154 0.18 1.37 f 0.10 I .3 35.0 f 2.0 (basaltic andesite) 2 650 8.5246 0.0205 0.5591 0.0765 0.85 2.11 fO.03 6.2 50.0 f 3.0 J = 0.01339 3 800 4.4828 0.0079 0.9136 0.0614 1.36 2.19 f 0.02 10.0 52.0 f 1.0 w = 0.35 g 3 900 4.8691 0.0089 0.7387 0.0382 3.33 2.28 f 0.02 25.2 54.0 t 2.0 5 I000 16.4520 0.0435 1.8761 0.1581 0.72 3.76 t 0.05 5.3 89.0 f 5.0 6 1100 9.241 3 0.0145 2.2729 0.0497 1.64 5.13 f 0.03 12.1 120.0 f 3.0 7 I200 7.4205 0.008 1 1.3234 0.026 I 3.68 5.1 I f 0.03 27.0 119.0 i 1.0 8 I300 11.2760 0.0223 4.8123 0.0659 1.19 5.05 t 0.04 8.7 118.0 f 4.0 9 1450 16.8750 0.0396 1.0322 0. I167 0.57 5.28 f 0.05 4.2 123.0 f 6.0

BP05 1 520 643.6700 2.1632 2.7477 0.4694 0.1 I 7.44 f 2.10 1.6 137.9 f 36.1 (basaltic andesite) 2 650 2 13.2300 0.7 165 1.7686 0.1654 0.27 2.53 f 0.60 3.8 48.2 c 12.6 J = 0.01 068 3 780 92.7100 0.3058 2.1633 0.0825 0.63 2.88 f 0.30 8.9 54.6 t 5.4 W = 0.25 g 4 900 103.2500 0.3428 0.0575 0.0894 0.79 2.77 f 0.30 11.1 52.5 f 5.3 5 I000 76.2400 0.2309 13.2300 0.1023 0.65 6.41 t 0.24 9.2 119.5 f 4.5 6 1100 2 1.2 100 0.0534 13.3340 0.1043 2.49 6.56 f 0.07 35.0 122.1 f 1.8 7 1250 33. I800 0.0988 13.8130 0.1302 1.85 6.29f0.11 26.1 117.3 t 2.5 rn 8 1450 148.7500 0.4722 5.7 13-3 0.1875 0.3 1 10.30 f 0.50 4.4 187.8 f 12.2 4 G BPI3 I 450 38.40 10 0.1160 2.355 I 0.0760 0.53 3.43 t 0.12 1.6 83.4 f 2.4 (basaltic andesite) 2 640 11.9180 0.0333 1.0763 0.0247 1.57 1.89 f 0.04 4.7 36.2 t 1.3 J = 0.01068 3 780 I 1.4430 0.0278 1.5310 0.0233 1.93 3.36 f 0.04 5.9 63.7 f 1.3 w = 0.25 g 3 900 9.0152 0.0197 1.0447 0.0 I89 2.84 3.27 f 0.03 8.6 61.9f 1.1 5 I050 11.1320 0.0274 1.6743 0.0255 2.28 3.19 i: 0.03 6.9 60.4 f 0.9 < 6 1200 6.3286 0.0105 0.5639 0.0171 9.20 3.36 f 0.02 28.1 63.6 t 0.8 0 7 I350 5.7537 0.0089 0.5280 0.0215 13.70 3.13 f 0.02 31.8 59.3 f 0.7 e, 2. 8 I500 45.7140 0.1343 3.9559 0.0771 0.75 6.50 f 0.13 2.3 121.1 f 3.9

BP03 460 140.9100 0.4545 13.2 120 0.1212 0.14 8.25 t 0.45 0.2 152.4 f 8.6 (diorite) 640 86.6670 0.2805 3.1329 0.0821 0.42 3.91 i: 0.27 0.5 73.8 f 5.2 J = 0.01068 750 38.4350 0.1241 1.7825 0.0442 0.63 2.02fO.12 0.8 38.5 f 2.5 0 w = 0.20 g 850 5.0259 0.009 1 3.0652 0.0355 16.60 2.56 f 0.02 21.6 48.7 f 1.5 W 950 5.5235 0.0105 0.6537 0.0168 9.22 2.46 f 0.02 12.1 46.8 f I .4 6 1050 3.5172 0.0066 0.302.5 0.0 159 18.70 2.57 f 0.02 24.4 48.8 f I .5 1200 4.2857 0.0059 0.3985 0.0 149 25.60 2.56 f 0.02 33.5 18.6 t I .5 1350 9.4872 0.0235 0.9765 0.020 I 5.03 2.62 f 0.03 6.5 49.8 f 1.7 1500 114.1030 0.3389 15.7870 0.2051 0.17 9.79 f 0.36 03 179.5 t 9.5

.Ar*: radiogenic Ar: J: calculated J-factor; W: sample weight: M: measured value: "Arti: calculated '"Ar originated from potassium 150 - : (a) BPO~ +-1,2=119*1 My + - I 2 Yr” - 100 - 100 0) ul - m - c c E m 50 50 a a a , 1,,=52+1My

0 111111111, 0 10 20 30 40 50 60 70 80 90 100 Fraction of 39ArK released 0.004 I 0.004

0.003 0.003

L 1 = 60.6 f 1.4 l,, = 53 i2 a , My Mv I 4QArl’’Ar=294.7 i2.3 0.002 ;0.002 4: 0 0.001 0.001

0.000 0.000 0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.1 0.2 0.3 0.4 3PAr/‘OAr 3gArl‘OAr

200 200 (S) BP03 zX 150 W W CI) m m 100 m I 100 t - c c E E m f--- 1 = 48.4 i0.5 My -b 50 CL a a 50 a 4 0 0 l--===- 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Fraction of 39ArK released Fraction of 39Ar, released

0.003 0.003 1,=48.8*1.5 My

L ‘OArI 1sAr=295.i f 2.1 L 4 0.002 0.002 .L a 0 0.001 0.001

0.000 0.000 0.0 0.1 0.2 00 0.1 0.2 0.3 0.4 39Arl‘OAr 39Ar/‘OAr

Fig. 3. ‘”Ar/”Ar apparent ages and isotope correlation diagrams of hlrsdtic andesite and diorite from Barton Peninsula, King George Island. Antarctica. Abbreviations: P = apparent plateau age; 1 = isochron age: 1 = low-temperature fraction,; 2 = high-temperature fractions. (a-f) are for basaltic andesite. and (g-h) for dioritr.

Kim et al. 2000: Polar Reseurch 19(2), 251-260 257 580 47' w 58" 44' w Fig. 4. Map showing the U' I 4"Ar/3"Ar and K-Ar ages compiled from this and previous studies of Barton Peninsula. King George Island. Filled circles represent the results of this study. Filled and open squares deiiote the results of Park (1989) and Lee et al. (1996), respectively. Filled stars refer to the locations of plant fossils reported by Chun et al. (1994). Abbreviations: tg, = biotite age; tw = whole- rock age: tl = isochron age; t,, = low-temperature fraction isochron age: t12 = high- temperature fraction isochron age.

0 c-

considered to be reliable because the initial volcanic rocks on Barton Peninsula. Furthermore, 4"Ar/'6Ar ratios of analysed specimens are con- plant fossils documented from the Jurassic strata sistent with that of present atmospheric argon of , , are also (295.5). Furthermore, the apparent plateau ages reported from the Cardozo Cove Group of the are compatible with isochron ages, which support Barton Horst (Zastawniak 1981; Birkenmajer & this interpretation. The old isochron ages deter- Zastawniak 1986). The Cardozo Cove Group of mined in this study thus suggest an Early the Admiralty Bay (Fig. Ic) is correlated with the Cretaceous volcanic activity on Barton Peninsula, volcanic sequence of Barton Peninsula (Birken- whereas the young ages are compatible with the majer 1982). Thus, we cannot preclude the Eocene ages previously determined from the occurrence of Mesozoic volcanic and volcaniclas- plutonic rocks. Thus, the latter are interpreted to tic rocks on Barton Peninsula. be the ages reset by the thermal effect during the Based on the similarity of K-Ar ages, Jwa et al. emplacement of plutons. (1 992) suggested that the displacement ofthe Barton Plant fossils of late Palaeocene to early Horst with respect to the Fildes Block is negligible. Oligocene ages occur in reddish to black shales K-Ar and 40Ar/39Arages from the eastern part of intercalated with agglomerates in the southern part pluton, however, indicate the presence of Early to of Barton Peninsula (Fig. 4) (Del Valle et al. 1984; Late Cretaceous volcanism on Barton Peninsula and Chun et al. 1994). These results are incompatible do not coincide with an age of 40-60 My for the with the K-Ar and "Ar/39Ar ages determined by Fildes Block. The agglomerate, the lowermost this study, as well as with the previous suggestion volcanic unit of southern Barton Peninsula, contains of a Mesozoic age of volcanic rocks on the Barton Eocene flora fossils but is absent in the north- Peninsula (Davies 1982; Birkenmajer, Narebski et western peninsula across the Barton fault. On the al. 1983b). On the other hand. Tokarski et al. other hand, lapilli tuff abruptly diminishes in its ( 1987) suggested a possible presence of Mesozoic occurrence to the south of the Barton fault.

25 8 Geochronologic evidence for Early Cretaceous volcanic activity on Barton Peninsula The apparent discrepancies in geologic and Ackiio~vlrdgenieiirs.- This work was financially supported by geochronologic data across the Barton fault can be the Ministry of Maritime Afhirs and Fisheries, South Korea, through KORDI (PP98001 04), and by the Nature Science accounted for by the presence of significant Foundation of China (19672107t, the Chinese Academy of displacement across the fault. This fault is thought Sciences (KZ951-AI-205-02) and the Ministry of Science and to be a transform fault which is related to the Technology, China (98-927-0 1-06-05), opening of the Bransfield Strait. Previously determined K-Ar ages from the volcanic rocks of the southern part of Barton Peninsula (Park 1989) References are well correlated with Eocene volcanism having widely occurred in the Fildes and the Warszawa Arche. A,, L6pez-Martinez. J. & Martinez de Pison. E. 1992: blocks of King George Island. Sedimentology of the Miers Bluff Formation, , South Shetland Islands. In Y. Yoshida et al. (eds.): Hu, Zheng et al. (1996) reported Late Cretac- Recunf progress in Antarctic earth scieiice. Pp. 357-362. eous to late Eocene (92-40 Mya) ‘“Ad3‘Ar and K- Tokyo: TeWd Scientific Publishing. Ar ages in the northern Fildes Block of King Barton. C. M. (ed.) 1965: The geology of the Sourh Shetland George Island. These ages successively decrease Islands: 111. Thc striitigruphy of King Georye Iskind. Br. toward the north-east, and Hu. Zheng et al. (1996) Aiirarcr. Suiv. Sci. Rep. 44. Birkenmajer. K. 1980a: Tertiary volcanic-sedimentary succes- have attributed this systematic change to the sion at Admiralty Bay. King George Island (South Shetland gradual migration of the volcanic centre (Pank- Island, Antarctica). Stud. Geol. Pol. 64, 7-65. hurst & Smellie 1983: Smellie, Pankhurst et al. Birkenmajer, K. 1980b: Report on geological investigations of 1983). Our result, however, indicates a large King George Island. South Shetland Islands (West Antarc- temporal gap along a vertical set of volcanic tica), in 1978-1979. Stud. Geol. Polon. 64, 89-105. Birkenmajer, K. 1980~:A revised lithostratigraphic standard for sequences in a narrow area. Furthermore. recent the Tertiary of King George Island. South Shetland Islands geochronologic studies on Livingston Island (). Bull. Pol. Acod. Sci.. Edi Sci. 27( In). indicate an Early to Late Cretaceous (137- 49-51, 73 Mya) volcanic activity (Smellie, Pallhs et al. Birkenmajrr, K. 1980d: Geology of Admiralty Bay, King 1996: Zheng. Hu et al. 1997; Zheng, Sang et al. George Island (South Shetland Islands) - an outline. Pol. Polar Res. /(/). 29-54. 1998) corroborating our result. It is thus likely that Birkenmajer. K. 1982: Mesozoic stratiform volcmic-sedimen- at least parts of King George and Livingston tary succession and Andean intrusions at Admiralty Bay, islands are composed of Early Cretaceous volcanic King George Island (South Shetland Islands. Antarctica). sequences. SlUd. Gcd. Pol. 74(3). 105-154. Birkenmajer, K. 1983: Late Cenozoic phases of block-faulting on King George Island, South Shetland Islands, West Antarctica. Bull. Pol. Acad. Sci., Eurfh Sci. 30(IRJ,21-32. Birkenmajer, K. 1989: A guide to Tertiary geochronology of King George Island. West Antarctica. Pol. Polar Rrs. /O(4), 55.5-579. Conclusions Birkenmajer. K.. Francalanci, L. & Peccerillo. A. 1991: Petrological and geochemical constraints on the genesis of 1 ) “Ar/3’Ar step-heating whole-rock ages deter- Mesozoic-Cenozoic magmatism of King George Island, mination of basaltic andesite from Barton Penin- South Shetland Islands. Antarctica. Anrarcr. Sci. 3(3).793- sula in King George Island suggest two separate 30s. magmatic activities: Early Cretaceous and middle Birkenmajer. K., Narebski, W., Nicoletti. M. & Petrucciani. C. 19X3a: Late Cretaceous through late Oligocene K-Ar ages of Eocene. the King George Island Supergroup volcanics, South Shet- 2) “oAr/3yAr and K-Ar ages of volcanic rocks in land Islands (West Antarctica). Bull. Pol. Acud. Sci.. Ecirrli the northern pluton part are compatible with the Si. .iO(.i’/4/j), 133-143. stratigraphic relationship among sedimentary and Birkenmdjer K.. Narehski, W.. Nicoletti, M. & Petrucciani, C. volcaniclastic sequences. In spite of thermal 19X3h: K-Ar ages of the “Jurassic volcanics” and “Andean” intrusions of King George Island. South Shetland Islands resetting caused by the intrusion of pluton. (West Antarctica). Bull. Pol. Acad. Sc.i.. E(irtl7 Sri. 3O(MJ. 4oAr/39Ar step-heating ages of basaltic andehite 121-131. preserve the evidence for Early Cretaceous Birkenmajer. K. & Zastawniak. E. 1986: Plant remains from the volcanic activity in the northern Barton Peninsula. Dufayel Island Group (early Tertiary’?),King George Island. 3) Geologic and geochronologic differences South Shetland Islands (West Antarctica). Acrci Pdaeobor. across the north-west-south-east trending fault Chun, H. Y., Chang. S. G. & Lee, J. I. 1994: Biostratigraphic indicate the presence of tectonic discontinuity on study on the plant fossils from the Barton Peninsula and Barton Peninsula. adjacent areas. J. Paleonrol. Soc. Koreci /O( / ), 69-81.

Kim et al. 2000: Polar Rrsenrch 19(2),251-260 259 Davies, R. E. S. 1982: The geology of the Marian Cove area, Pankhurst. R. J. & Smellie, J. L. 1983: K-Ar geochronology of King George Island and Tertiary age for its supposed Jurassic the South Shetland Islands, Lesser Antarctica: apparent volcanic rocks. Br. Anturcr. Surv. Bull. 51, 151-166. lateral migration of Jurassic to Quaternary island arc Del Valle, R. A., Diaz, M. T. & Romero. E. J. 1984: Preliminary volcanism. Earth Planer. Sci. Lett. 66, 214-222. report on the sedimentites of Barton Peninsula, 25 de Mayo Park, B.-K. 1989: K-Ar radiometric ages of volcanic and Island (King George Island), South Shetland Islands. plutonic rocks from the Barton Peninsula. King George Argentine Antarctica. Contrib. I/ut. Aiitn'rt. Argent. 308. 1- Island, Antarctica. J. Geol. Sue. Koreu 25(3),495497. 19. Smellie, J. L.. Liesa. M.. Muiioz, J. A,, Sibat, F., Pallhs, R. & Ferguson, D. 1921: Geological observations in the South Willan, R. C. R. 1995: Lithostratigraphy of volcanic and Shetland Islands, the Palmer Archipelago and Graham Land. sedimentary sequences in central Livingston Island. South Antarctica. Truns. R. Soc. Edinburgh 53(l),29-55. Shetland Islands. Antcrrct. Sci. 7(1), 99-1 13. Hathway, B. & Lomas, S. A. 1998: The Upper Jurassic-Lower Smellie. J. L.. Pallas, R.. Sibat. F. & Zheng. X. S. 1996: Age and Cretaceous Byers Group, South Shetland Islands, Antarctica: correlation of volcanism in central Livingston Island, South revised stratigraphy and regional correlations. Cretac. Res. Shetland Islands: K-Ar geochemical constraints. J. South 19,4347. Ainer. Earth Sci. 9(3/4).265-272. Hu. S. L., Wang, S. S.. Sang, H. Q. & Qiu, J. 1985: An Smellie. J. L., Pankhurst, R. J.. Thomson, M. R. A. & Davies, R. application of the fast-neutron activation dating technique to E. S. (eds.1 1984: The geology of the South Sherlarid Islunds: approach the age of early emplacement of Jiuling gmno- VI. Srratigraphy, geochentistry arid evolution. Br. Antarct. diorite intrusion of Jiangxi province. ACIUPerrolog. Sin. I. Sun,. Sci. Rep. 87. 29-34. So. C.-S., Yun. S.-T. & Park. M.-E. 1995: Geochemistry of a Hu. S. L., Zheng, X., Molan, E. & Birkenmajer. K. 1996: fossil hydrothermal system at Barton Peninsula, King George ""Ar13'Ar and K-Ar age datings on the volcanic rocks in Island. Anrarct. Sci. 7, 63-72. northern coast of King George Island, West Antarctica. Steiger, R. H. & Jager. E. 1977: Subcommission on geochron- Korean J. Polar Res. 7(1). 11-21. ology convention on the use of decay constants in geo- and Jwa. Y.-J., Park, B.-K. & Kim. Y. 1992: Geochronology and cosmochronology. Earth Planet. Sci. Lett. 36, 359-362. geochemistry of the igneous rocks from Barton and Fildes Tokarski, A. K. 1988: Structural analysis of Barton Peninsula peninsulas. King George Island: a review. In Y. Yoshida et al. (King George Island, West Antarctica): an example of (eds.): Rccenf progress in Antcirctic earth scirncr. Pp. 439- volcanic arc tectonics. Srud. Geol. Pol. 95. 53-63. 442. Tokyo: Terra Scientific Publishing. Tokarski. A. K., Danowski, W. & Zastawniak. E. 1987: On the Kang. P.-C. & Jin, M.3. 1989: Petrology and geologic age of fossil flora from Barton Peninsula. King George Island, structures of the Barton Peninsula. King George Island, West Antarctica. Pol. Polar Res. 8(3).293-302. Antarctica. In H. T. Huh et al. (eds.): Antcrrctic science: Tyrrell. G. W. 1921: A contribution to the petrography of the geology and biology. Pp. 121-135. Seoul: Korea Ocean South Shetland Islands, the Palmer Archipelago and the Research and Development Institute. Kawashita, K. & Soliani, E., Jr. 1988: A Rb-Sr isochron diagram Danco Land Coast, Graham Land. Antarctica. Trms. R. Soc. for the Znosko Formation (Cardozo Cove Group), Edinburgh 53, 57-79. Admiralty Bay, King George Island, Antarctica. Ser. Cienr.. Willan, R. C. R., Pankhurst, R. J. & HervC, F. 1994: A probable Insr. Antcirt. Chi/. 38, 59-66, Early Triassic age for the Miers Bluff Formation, Livingston Kim, H.. Cho, M. & Lee, J. 1. 1995: Low-pressure thermal Island. South Shetland Islands. Armrcr. Sci. 6. 401308. metamorphism of volcanic rocks in the Barton Peninsula, York, D. 1969: Least-squares fitting of a straight line with King George Island. Antarctica. In Y. Kim et al. (eds.): The correlated errors. Earth Plfinet. Sci. Lett. 5. 320-324. Zastawniak, E. 1981: Tertiary leaf flora from the Point ,fi~urt/iSeoul Irtternutional Symposiuni on Anrarcric Science - Geology qf the South Shetla~id ldands. Pp. 25-27. Seoul: Hennequin Group of King George Island (South Shetland Korea Ocean Research and Development Institute. Islands, Antarctica): preliminary report. Stud Geoliig. Pol. Kim, H., Cho. M. 2s Lee, J. 1. 2000 (unpubl. ms.): Thermal 72, 97-108. metamorphism of volcanic rocks in the Barton Peninsula, Zheng. X. S., Hu, S. L., Liu, J. Q. & Sabat, F. 1997: New King George Island, Antarctica. 4"Ar/37"Arage evidence for the Cretaceous volcanic rocks of Lee, J.-I., Hwang, J., Kim, H., Kang. C. Y., Lee, M. J. & Nagao, the Mount Bowles Formation in Livingston Island, South K. 1996: Subvolcanic zoned granitic pluton in the Barton and Shetland Islands. Chin. J. Polar Sci. N(2j. 89-95. Weaver peninsulas, King George Island. Antarctica. In K. Zheng, X. S., Sang, H. Q..Qiu. J., Liu. J. Q.. Lee, J.-I. & Hwang, Moriwalj et al. (eds.): Proceedings uf' rhr 15rh NlPR J. 1998: Isotopic age of the volcanic rock in Byers Peninsula, Symposium 011 Antcirctic Grosciences. Pp. 7&90. Tokyo: Livingston Island, West Antarctica. Chin. J. Polar Re.$. 10(1), National Institute of Polar Research. 1-10.

260 Geochronologic evidence for Early Cretaceous volcanic activity on Barton Peninsula