absence of Pliocene and Pleistocene diatoms beneath most change. Part 2 (Antarctic Research Series, Vol. 60). Washington, of the Ross embayment. The model is consistent with avail- D.C.: American Geophysical Union. able stratigraphic, sedimentologic, and geomorphic data Drewry, D.J. 1983. Antarctica: Glaciological and geophysical folio. Cambridge, U.K.: Scott Polar Research Institute, University of from the Ross embayment. Despite the paucity of data, the Cambridge. cross-section presents a hypothesis regarding the stratigra- Engelhardt, H., N. Humphrey, B. Kamb, and M. Fahnstock, 1990. phy that may be preserved in the Ross embayment, including Physical conditions at the base of a fast moving ice stream. Sci- the Bentley Trough. The model can (and, I hope will) be ence, 248(4951), 57-59. tested in the coming decade(s?). Testing this model will Harwood, D.M., R.P. Scherer, and P-N. Webb. 1989. Multiple Miocene marine productivity events in West Antarctica as recorded in require additional sub-ice sampling in the deep interior Upper Miocene sediments beneath the Ross Ice Shelf (Site J-9). basins and the development of stratigraphic drilling beneath Marine Micropaleontology, 15, 91-115. thick glacial ice. Hayes, D.E., and L.A. Frakes. 1975. General synthesis: Deep Sea This research was supported by National Science Foun- Drilling Project 28. In D.E. Hayes and L.A. Frakes (Eds.), Initial dation grants OPP 92-20413 and OPP 94-96169. results of the Deep Sea Drilling Project. Washington, D.C.: U.S. Government Printing Office. Kellogg, D.E., and T.B. Kellogg. 1986. Diatom biostratigraphy of sedi- References ment cores beneath the Ross Ice Shelf. Micropaleontology, 32(1), 74-79. Alley, R.B., D.D. Blankenship, S.T. Rooney, and C.R. Bentley. 1989. Rooney, S., D.D. Blankenship, R.B. Alley, and C.R. Bentley. 1991. Seis- Sedimentation beneath ice shelves: The view from ice stream B. mic profiling of a sediment filled graben beneath ice stream B, Marine Geology, 85, 101-120. Antarctica, In M. Thompson, J. Crame, and J. Thompson (Eds.), Anderson, J.B., and L.R. Bartek. 1992. Cenozoic glacial history of the Geological evolution of Antarctica. Cambridge, U.K.: Cambridge Ross Sea revealed by intermediate resolution seismic reflection University Press. data combined with drill site information. In: J.P. Kennett and Savage, M.L., and P.F. Ciesielski. 1983. A revised history of glacial sed- D.A. Warnke (Eds.) The antarctic paleoenvironment: A perspective imentation in the Ross Sea region, In R.L. Oliver, P.R. James, and on global change. Part 1 (Antarctic Research Series, Vol. 56). Wash- J.B. Jago (Eds.), Antarctic earth science. Cambridge, U.K.: Cam- ington, D.C.: American Geophysical Union. bridge University Press. Barrett, P.J., (Ed.). 1989. Antarctic Cenozoic history from the CIROS-1 Scherer, R. 1991. Quaternary and Tertiary microfossils from beneath drillhole, McMurdo Sound. Department of Scientific and Indus- ice stream B: Evidence for a dynamic west antarctic ice sheet his- trial Research Bulletin, New Zealand, 245, 1-254. tory. Palaeogeography, Palaeoclimatology, and Pa laeoecology Barrett, P.J., and B.C. McKelvey (Eds.). 1986. Stratigraphy and Antarc- (Global Planetary Change Section), 90, 395-412. tic Cenozoic history from the MSSTS-1 drillhole, McMurdo Sound. Scherer, R. 1992. Diatom paleoproductivity and sediment transport in Department of Scientific and Industrial Research Bulletin, New west antarctic basins and the Neogene history of the west antarc- Zealand, 237,9-51. tic ice sheet. (Ph.D. Dissertation, Ohio State University, Colum- Bentley, C., and J. Clough. 1972. Antarctic subglacial structure from bus, Ohio.) seismic refraction measurements. In R.J. Adie (Ed.), Antarctic geol- Scherer, R.P. 1993. There is direct evidnce for Pleistocene collapse of ogy and geophysics. Oslo: Universitetsforlaget. the west antarctic ice sheet. Journal of Glaciology, 39(133), Bindschadler, R.A., B. Koci, and A. Iken. 1988. Drilling on Crary Ice 716-722. Rise, Antarctica. Antarctic Journal of the U.S., 23(5), 60-62. Scherer, R., Harwood, D., Ishman, S., and Webb, P. 1988. Micropale- Cooper, A.K., P. Barrett, K. Hinz, V. Traube, G. Leitchenkov, and H. ontological analyses of sediments from Crary Ice Rise. Antarctic Stagg. 1991. Cenozoic prograding sequences of the Antarctic Con- Journal of the U.S., 23(5), 34-36. tinental Margin: A record of glacio-eustatic and tectonic events. Webb, P.-N., T.E. Ronan, Jr., J.H. Lipps, and T.E. DeLaca. 1979. Marine Geology, 102, 175-213. Miocene glaciomarine sediments from beneath the southern Ross Cooper, A.K., S. Eittreim, U. Ten Brink, and I. Zayatz. 1993. Cenozoic Ice Shelf, Antarctica. Science, 203(4416), 435-437. glacial sequences of the antarctic continental margin as recorders Whillans, I.M., J. Bolzan, J., and S. Shabtaie. 1987. Velocity of ice of antarctic ice sheet fluctuations. In J.P. Kennett and D.A. Warnke streams B and C, Antarctica. Journal of Geophysical Research, (Eds.), The antarctic paleoenvironment: A perspective on global 92(B9), 8895-8902.

Volcanic activity and seismicity of , 1986-1994 RAYMOND R. DIBBLE, Victoria University of Wellington, Wellington, New Zealand PHILIP R. KYLE and MICHAEL J. SKOV, Department of Geoscience, New Mexico Institute of Mining and Technology, Socorro, New Mexico 87801

ount Erebus, a 3,794-meter (m) high, active volcano on 1985, when the lava lake was exhumed, small strombolian MRoss Island, Antarctica, has a permanent lake of molten eruptions have occurred at rates of between zero and at least anorthoclase phonolite magma in its . A 3-month eight per day. Between December 1986 and December 1990, period of sustained large strombolian eruptions starting in Victoria University of Wellington maintained surveillance of September 1984 buried the lake and ejected over 100,000 the lava lake (when cloud and solar battery power allowed) cubic meters of phonolite bombs and other ejecta. Since from a videocamera situated on the north rim of the main

ANTARCTIC JOURNAL - REVIEW 1994 11 crater (Dibble et al. 1988). The video signal was radioed to (figure 1, right) and were about 10 m in diameter. Again, erup- using a teleyision transmitter where it was tions were from the north lake, which convected less than the recorded on videotape. Following are brief descriptions of the south lake. The TV station was dismantled on 11 December volcanic activity and changes in the lava lake recorded using 1990. No observations were made during the 1991-1992 field the video monitoring system and ground-based surveillance season. In the 1992-1993 field season, rare small strombolian made during austral summer field seasons. eruptions occurred from two small lakes generally similar to In 1986-1987, there was one incandescent lake 2-10 m in those observed in November 1990. diameter near the center of the inner crater. It exploded like a On 19 October 1993, two large phreatic eruptions cannon about three times per day. By the austral summer of occurred. These are the first ever recorded at Mount Erebus, 1987-1988, the lake was on the north side of the inner crater. and they showered the north crater rim with pulverized It was 10 m in diameter and erupted large bursting globules of hydrothermally altered debris. The eruptions formed a new magma. A much smaller lava pool, which rarely erupted, crater about 50x20 m across near the edge of the formed near the south side of the inner crater, and near the floor above the lava lakes (figure 2). The earthquakes that northwest side, steam revealed a third vent, which was hidden accompanied the eruptions gave seismograms with cigar- from the videocamera. A camera failure limited surveillance shaped envelops and had durations of around 1 minute. The during the austral summer of 1988-1989 until January 1989. At earthquakes were well recorded by a seismometer at that time, the north and south lakes were almost frozen over, McMurdo Station. A helicopter piloted by Lt. Bob Brodin was but in February, a lava flow from the south pool formed a lake flying along the flank of Mount Erebus at an altitude of about at least 15 m long. In April, both lakes were about 5 m across. 1,850 m and recorded the first eruption at about 16:46 (local By November 1989, a fauklike crack had appeared at the base time). The crew observed a black mushroom-shaped cloud of the west wall of the inner crater, and the floor appeared to rising vertically about a thousand meters above the crater. be subsiding on the north side. The south lake was 13 m Wind dispersed the cloud to the north. across and rarely erupted, but east-to-west magma convec- An array of six vertical 1 -hertz seismometers monitors the tion within the lake was common. The north lake was again seismicity associated with the volcanic activity. The data from frozen over, but between 18 and 25 December 1989, lava the array are recorded at the Mount Erebus Volcano Observa- flowed north-northwest from it. After this, frequent small tory (MEVO) at McMurdo Station using helicorders and a PC explosions occurred from the hidden northwest vent until computer running XDETECT, an International Association of mid-January. A line of small incandescent vents formed Seismology and Physics of the Earths Interior (IASPEI) event- between the south and north lakes in March 1990, ending on recording program. The digital recordings are automatically 26 March at 10:00 universal time (UT) in a violent explosion, transferred to a Sun workstation and (after compression) sent which joined all the lava lakes into one of 25 m diameter (fig- to New Mexico Institute of Mining and Technology and Victo- ure 1, left) with flow toward the north. This persisted until TV ria University via the Internet, utilizing the STARS communi- transmission ceased on 19 April 1990. When video transmis- cation link between Antarctica and the United States. This sion resumed on 10 November 1990, a new solid flat lava floor allows year-round study of the volcano. had formed above that observed in March 1990. The north From December 1993 to June 1994, high-quality record- and south lakes had been reestablished in shallow pit craters ings of two or more seismic events per day have been made.

Figure 1. Scenes from the TV surveillance tape showing (left) the enlarged lava lake at 0446 UT on 13 April 1990 and (right) the two pools of lava in the new solid lava floor at 2300 UT on 30 November 1990. The view is from the north rim of the main crater and is 38 m wide at the distance of the lava lakes.

ANTARCTIC JOURNAL - REVIEW 1994 12 These result from small erup- - - 2- tions and earthquakes scattered . throughout the volcano to depths of 10 kilometers (km) below the summit. A provisional three-dimensional seismic veloc- it structure clown to LZ Km I)elow the summit of Erebus vol- cano has been obtained (Luo and l)ibble 1994) by a traveltime inversion technique similar to i\kj and Lee (1976). A three- ______dimensional grid model of the velocity structure was defined at 9 points on each of three 15-km square planes at depths of -1.5, 1 .5, and 4.5 km below sea level (BSL). A three-dimensional ray tracing routine (Cerveny, Mimes, and Psencik 1988, pp. 89-169) was used to calculate traveltimes Figure 2. View southward across the crater floor of Mount Erebus, showing a new crater formed by the through the model Velocity d two phreatic eruptions that occurred on 19 October 1993. In the foreground is the wall to the inner . .7 crater, the site of the permanent convecting lava lake of anorthoclase phonolite magma. Behind the new location corrections were solved crater is the main crater floor. The crater is over 50 m in diameter. simultaneously. Figure 3 shows contours of final P velocity on the three planes. The average velocities over the A. Depth -1 ,5k Vp(knils) A. Depth: -1 b r:. Vp(km/s) grids are 3.9±0.1 kilometers per second (km/ s) at 4.8 1. 48 -7728 .77. -1.5 km BSL, 4.1±0.1 km/s at 1.5 km BSL, and 4.5±0.1 km/s at 4.5 km BSL. In addition, there 4.4 44 4.2 42 are horizontal velocity variations of up to 20 per- W 4.0 W,.?: 40 cent. The plane at 4.5 km BSL has an area of -7T 32 .77 32 3.8 38 lower velocity about 5 km across centered below 3.6 3.6 the crater. It could be a magma chamber of 3.4 LO2(-) 3.4 rather lower velocity and diameter than shown 13.2 3.2 1670e S 1672y 16700 16720 in the figure. The plane at 1.5 km BSL has a less

B. Vp(km/s) B. Depthi 1 5krri Vkm/s) symmetrical area of low velocities extending - 5.1 - 5.1 -iT 28 .77 28 from beneath the lava lake toward the Hut Point 4.8 - 4.8 Peninsula. The top plane 1.5 km above sea level 46 - 4.6 shows no sign of a low-velocity region below the 4.4 44 lava lake. The diameter of the conduit at this 77 32 .77 32 - 4.2 4.2 depth must be much less than the 5-km resolu- it 4.0 4.0 tion of the study. 3.8 3.8 The rapid availability of digital seismic data 3.6 3.6 167OiY 1672O 16700 167 20 from Mount Erebus not only increases our abil- C. Depth: 4.5k Ti Vp(km/s) C. Depth: 4.5km Vkm/s) ity to provide effective volcanic monitoring but • . 5.7 57 .77 28 IJ"" - .77 28 also enhances our ability to build a large data- base through which an improved velocity model 5.3 might be created. 51 51 4.9 This research was supported by National 7T 32 .77 32 4.7 4-7 Science Foundation grant OPP 91-18056. Victo- 4,5 I 4.5 ria University of Wellington supported the 4.3 tomography study and the TV surveillance. We 4.1 4.1 167O 1672LY 167 00• 167 20 greatly appreciate the tremendous support from the VXE-6 helicopter squadron and the Berg Figure 3. Contours of seismic P velocity in the model obtained by interpolation and smoothing of the values determined on each depth plane. A. At 1.5 km above sea Field Center staff of Antarctic Support Associ- level. B. At 1.5 km below sea level. C. At 4.5 km below sea level. Left: First iteration. ates. Lt. Bob Brodin kindly provided photos of Right: Second iteration using corrected hypocenters. one of the 19 October 1993 phreatic eruptions

ANTARCTIC JOURNAL - REVIEW 1994 13 and Bob Mien, U.S. Geological Survey, provided the vertical Dibble, R.R, I.D. Barrett, K. Kaminuma, S. Miura, J. Kienle, C.A. air photo of the Erebus Crater. Rowe, P.R. Kyle, and W.C. McIntosh. 1988. Time comparisons between video and seismic signals from explosions in the lava References lake of Erebus volcano, Antarctica. Bulletin of the Disaster Pre- vention Research Institute, Kyoto University, 38(3), No. 337, 147-161. Aid, K., and W.H.K. Lee. 1976. Determination of three-dimensional Luo, X., and R.R. Dibble. 1994. A study of three-dimensional velocity velocity anomalies under a seismic array using P arrival times structure of Erebus Volcano, Antarctica. The 27th General Assem- from local earthquakes 1. A homogeneous initial model. Journal of bly of the International Association of Seismology and Physics of the Geophysical Research, 81(23), 4381-4399. Earths Interior, Wellington New Zealand, 10 January 1994. Cerveny, V., L. Mines, and L. Psencik. 1988. Complete ray tracing in Wellington, New Zealand: Organising Committee of the 27th Gen- three-dimensional structures. In D.J. Doornbos, (Ed.), Seismologi- eral Assembly of the International Association of Seismology and cal algorithms. London: Academic Press. Physics of the Earths Interior. [Abstract]

Argon-40largon-39 dating of Mount Erebus, , Antarctica RICHARD ESSER, M. HEIZLER, P. KYLE, and W.C. MCINTOSH, Department of Geoscience, New Mexico Institute of Mining and Technology, Socorro, New Mexico 87801

ount Erebus (77 032S 167°10E) is a 3,794-meter high Anorthoclase and plagioclase feldspars were separated Mactive volcano overlooking McMurdo Station on Ross from phonolite, trachyte, and hawaiite collected from the Island, Antarctica. Two major types of rocks dominate Mount summit and flanks of Mount Erebus over several field sea- Erebus: a core of less evolved basalt and a carapace of sons. Whole-rock separates were also obtained for those anorthociase phonolite. The former rock type crops out prin- lavas containing no abundant feldspar phenocrysts. All cipally on Fang Ridge, an eroded remnant of a proto-Erebus 40Ar/ 39Ar analyses were performed at the New Mexico volcano, and is composed predominantly of plagioclase- Geochronology Research Laboratory at New Mexico Institute phyric ne-hawalite. Smaller volumes of this rock type outcrop of Mining and Technology in Socorro, New Mexico. sporadically on the southwestern flanks of Mount Erebus, The anorthoclase feldspars from historically erupted often near the coast. The majority of exposed rock on Mount (1984) volcanic bombs, hereafter called summit phenocrysts, Erebus is anorthoclase phonolite which represents the were selected for analysis so that any deviation from an age youngest and current stage of growth. Anorthoclase phonolite of 0 would help identify the source of excess argon. By ana- lava flows are exposed around the flanks and inifil the major lyzing pure vs. glassy summit phenocrysts, a positive rela- summit caldera. The summit cone is mainly composed of tionship between the chlorine content and the apparent age anorthoclase phonolite ejecta erupted by explosive eruptions. of the anorthoclase was discovered (that is, the higher the Mechanical disintegration of the bombs has left a lag of chlorine content, the older the age). Because no chlorine is anorthoclase crystals up to 10 centimeters in length. Within thought to exist within the feldspar itself, we concluded that the bombs and lavas, anorthoclase feldspar is abundant excess argon is associated with the glass melt inclusions (approximately 30-40 percent) and is often riddled with melt (1,700 parts per million chlorine) present in the anorthoclase (glass) inclusions trapped during rapid growth. The high feldspars. potassium oxide (1(20) content (3-4 percent), abundance, and With the excess argon component now identified, it was large size potentially make the anorthoclase ideal for argon- desirable to remove as many of the contaminating melt 40/argon-39 (40Ar/39Ar) dating. inclusions as possible from the anorthoclase. In most cases, Previous attempts at dating lavas from Mount Erebus older samples containing greater than 5 percent glass often have produced suspicious results. Historically erupted produced discordant age spectra that were considerably anorthoclase feldspar phenocrysts yielded conventional higher in age than those produced with the same sample of potassium/argon (K/Ar) dates of over 200,000 years (Arm- lower glass content (less than 5 percent). Fortunately, for the strong 1978). Because there is no evidence for xenocrystic majority of samples displayed in this study, apparent ages contamination, the anomalously old ages are attributed to were derived from a "plateau" in the pure samples. "excess argon" as defined by Dalrymple and Lanphere The table shows several dates for Mount Erebus lavas as (1969). In this study, historically erupted phenocrysts as well determined by the 40Ar/39Ar method. A number of these as other samples from Mount Erebus previously dated by samples were not analyzed in duplicate; consequently an the K/Ar method were re-dated using the more accurate and accurate measure of their susceptibility to excess argon is not precise 40Ar/ 39Ar dating method (Merrihue and Turner available. However, pure Mount Erebus samples (approxi- 1966). mately 99 percent anorthoclase) older than approximately

ANTARCTIC JOURNAL - REVIEW 1994 14