POLISH
POLAR
RESEARCH
11
3—4
259—266
1990
Włodzimierz K O W A L E W S K I , Stanisław R U D O W S K I and S. Maciej Z A L E W S K I
Department of Polar and Marine
Research
Institute of G e o p h y s i c s
Polish A c a d e m y of Sciences
Księcia Janusza 64, 01-452 Warszawa,
POLAND
Seismoacoustic studies
within flooded part of the caldera
of the Deception Island,
West Antarctica
A B S T R A C T : T h e results of the detailed seismoacoustic profilling (CSP, b o o m a r ) are
presented. The investigation has been carried out in February 1985 and 1988 during two
G e o d y n a m i c a l Expeditions organized by the Institute of G e o p h y s i c s of the Polish A c a d e m y of
Sciences. The b o o m a r
penetration
of the caldera floor went d o w n
to
150 msec.
Four
seismoacoustic units of volcanic formations have been determined. T h e unit A corresponds to
pre-caldera series and occurred only in the border part of the flooded caldera. T h e unit contains
mainly pyroclastic rocks (consolidated agglomerates and tuffs) and probably s o m e inter­
calations of lavas. The units B, C and D fill up the caldera b o t t o m and correspond
to
post-caldera series. T h e units are c o m p o s e d of pyroclastic rocks, containing also materials
redeposited by iahars, glaciers, landwaters and by wind. The units C and D (the youngest one)
were certainly deposited under water. All the units are cut by n u m e r o u s faults, vents and other
types of intrusions. T h e larger faults, en echelon type, are situated around the b o t t o m and form
a ring-fracture. Caldera w a s formed by succesive stages of collapsing. This process is not
finished
yet and volcanic activity is still alive (especially in the western part of the flooded
caldera).
Key
w o r d s : Antarctica, D e c e p t i o n volcano, geophysics and geological structures.
Introduction
This report describes the results of marine seismoacoustic investigations
conducted in the flooded caldera (Port Foster) of the Deception Island during
the Polish Geodynamical Expeditions of 1984/85 and 1987/88, organized by
the Institute of Geophysics of the Polish Academy of Sciences and led by
Professor A. Guterch.
A preliminary seismoacoustic profilling was carried out in February 1985
(Kowalewski et al. 1987) aboard the vessel Jantar. It was the first seismoacous-
260
Włodzimierz K o w a l e w s k i et
al.
tic study in this area. In February 1988 the studies were continued. The
detailed seismoacoustic investigations were taken also aboard the vessel
Jantar. Numerous seismoacoustic profiles were made with profile lines situated
ca. every 250 m. The investigations were taken by CSP (continuous seis­
moacoustic profiling) used EG and G (Boomar and Sparker) equipment.The
boomar penetration of the caldera bottom went down to 150 msec. As a result,
it was possible to distinguish and determine the bottom structure of the flooded
caldera, to a depth of ca. 150—200 m beneath the level of the bottom. The
profiles were located by radar bearings on the characteristic objects on the
coast.
The research was sponsored by C P B P 03.03. Programme of the Polish
Academy of Sciences.
Geographical location and general topography
Deception Island which is one of the two active volcanoes in the Antarctic,
is situated in the Bransfield Strait near the southwestern end of the South
Shetland Islands archipelago (Fig. 1), ca. 15 km south of the Livingstone Island.
The geographical coordinates of the island are 62°57' S and 60°38' W.
Fig. 1. Locality m a p of the D e c e p t i o n Island
261
Seismoacoustic studies of D e c e p t i o n Island
The ring-shaped island (Fig. 2), 14 km in diameter, is an intermittently
active, composite stratovolcano built by successive volcanic eruptions from
a variety of vents, from the Tertiary (?) to Recent times with intervening periods
60°3OW
Witt
1967-1970
62°55'-
1969
-63°00'S
63°00'S —
Q
2
[H3 S
JiJ8
EHHfo
3
LA> H S
I ^|10
[C
Q i 2
13
60°3ffW
6oyo'
Fig. 2. G e o m o r p h o l o g i c a l sketch of caldera of the D e c e p t i o n Island. O n s h o r e elements, marked
mainly after G o n z a l e s - F e r r a n et al. (I971) and Newhall and Dzurisin (1988): 1 — c a l d e r a rim,
2 — craters, 3
1969 fissures, 4 — fumaroles, 5 — dated eruption centers, 6
new small cone,
active (mainly steam emission) in 1985 and 1988; 7 — limits of main ice fields. Offshore elements,
based o n seismoacoustic profile m a d e during Polish G e o d y n a m i c a l Expeditions 1984/85 and
1987/88: 8 — uneven surface of an accumulative area of the pyroclastic rocks c o n n e c t e d with c o n e s
and lahars, 9
relatively flat surface of the accumulative area of the pyroclastic
waterlain settled and horizontal bedded, 10
materials
c o n e s o n the floor, 11 — identified limits of the
y o u n g (lakkolit-like ?) intrusions (pyroclastic agglomerates ?) only slightly reflected in the surface
relief; 12
main faults, 13
lines of the selected seismoacoustic profiles (PI.
1-3).
262
Włodzimierz K o w a l e w s k i et
al.
of subsidence (Ashcroft 1972). The average height of the island is ca. 300 m, the
maximum height being 542 m above sea level, and ca. 1000 m above the
surrounding sea bottom.
The present relief (PI. 3, Fig. 2; PI. 4) of the island is a result of the past and
present volcanic activity (cones, craters, maars, lava flows etc.) combined with
activity of the slope processes (mainly lahars) and glaciers, activity (moraines,
sandurs etc.). Glaciers cover a considerable part of the island area.
The caldera, ca. 8—10 km in diameter, is largely occupied by sea. The
irregular-shaped landlocked harbour Port Foster (ca. 3—5 km in diameter) has
an average depth of ca. 130 m, reaching a maximum of 180 m. The narrow
entrance, Neptun Bellows, ca. 500 m wide and ca. 300 m deep joins Port Foster
with the open sea.
Previous works and historical volcanic activity
The onshore part of the island is relatively well investigated. The results of
the geological studies have been summarized in a number of works, especially
by Adie (1964), Gonzales-Ferran et al. (1971), Baker et al. (1975, after Newhall
and Dzurisin 1988). Two main geological series were distinguished: the preand post-caldera. The pre-caldera series contain lavas and pyroclastics
(agglomerates). Lavas are predominantly sodic andesites or basaltic andestites
(Newhall and Dzurisin 1988). Within the post-caldera series pyroclastic rocks
are dominant. Pyroclastic rocks containing consolidated agglomerates, tuffs,
non-consolidated young volcanic ashes and materials of the lahars — predomi­
nate on the exposed part of the island. The past and present volcanic activity
(marked with fissures, craters, cones and fumaroles) is associated with the
complex of ring-fracture around the caldera.
The offshore part of caldera, that is the sea floor of Port Foster, has been
much less investigated. One profile of the seismic refraction was made during
expeditions of the Department of Geology of the University of Birmingham
(Ashcroft 1972). During the two Polish Geodynamical Expeditions of 1979/80
and 1984/85 some reflection and refraction studies were also carried out. The
results will be published shortly. Also, Spanish-Argentine-Chilean Expedition
made a geophysical research of this area in February 1988.
Discovered in 1820, Deception Island has never been properly settled. The
only visitors to this area had been sailors, seal-hunters and whalers who used
Port Foster as a good harbour and as their temporary base during the summer
It was only in the mid 1940 s that the first permanent research stations, British
next Chilean and Argentine were established. They were in operation until
1969. The scantiness of reports on the island volcanic activity (as listed below
263
Seismoacoustic studies of D e c e p t i o n Island
mainly after Newhall and Dzurisin 1988) reflects the periodic presence of
people in this area.
1829: T h e occurrence of n u m e r o u s fumaroles and subterranean s o u n d s were reported.
1842: Reports of strong volcanic activity concerning 13 vents in the southern part of the island.
H a w k e s (1961, after Adie 1964) associates this with y o u n g volcanic lavas e x p o s e d in the
region of Mt. K i r k w o o d . R o o b o l (1973, after N e w h a l l and Dzurisin 1988) suggests that the
volcanic activity of this period w a s probably c o n n e c t e d with big fummaroles.
1912
and 1917: O n the basis of an analysis of the pyroclastic materials interbedded with ice t w o
eruptions have been postulated (Orheim 1972).
1923: W h a l e m e n observed shore subsidence to the southwest of Port Foster. „Boiling" water
washed paint off the ship hull.
1930: A n earthquake w a s observed. Part of Port Foster floor near the Whalers' Bay sub­
sided 4.5 m.
1967: After a series of earth tremors a powerfull eruption occured from four vents situated a l o n g
the ring-fructure and fractures were formed. As a result this part of the shore of Port Foster
changed significantly (Gonzales-Ferran et al. 1971).
1969: Strong eruptions were registered. N e w fissure (5 k m long) w a s formed a l o n g the ring fracture
o n the east side of the caldera. 2 0 — 3 0 new vents were formed within new fissure. T h e
Chilean research station w a s completely destroyed. T h e British station w a s partly buried
under a layer of pyroelastic materials carried d o w n from caldera slope by lahars. T h e crews
of the stations, together with the crew of the Argentine station which had remained intact,
had to be evacuated.
1970: Stations located around D e c e p t i o n Island (at a m a x i m u m distance of 170 km) reported
a profuse fall of volcanic ashes (Gonzales-Ferran et al. 1971). In 1972 Shultz (after
Newhall
and Dzurisin 1988) reported landscape changes in the northwestern part of the caldera.
1984: During the Polish G e o d y n a m i c a l Expedition in February, Dr. E. Perchuć discovered a small
(ca. 10 m height) new volcanic c o n e , situated in the south end of the fissure formed during
the eruption o 1969. T h e activity of the c o n e (also in February 1988) w a s limited mainly to
emission of steam.
1987: O n 23 July local seismic activity was recorded at a research station o n K i n g G e o r g e Island.
O n satellite images of this area a plume w a s seen, t h o u g h t to be a volcanic cloud (Newhall
and Dzurisin 1988).
Seismoacoustic studies within flooded caldera
The bottom structure of the flooded caldera (Port Foster) was investigated
in detail by the continuous profiling. Four major seismoacoustic units were
distinguished and described (PI. 1—3). A general lithological interpretation of
these units was introduced on the basis of our knowledge of the island onshore
geology. The specifield correlation between the seismoacoustic units and
onshore geology structure is very difficult and reguires research.
Unit A. The occurence of the unit A was identified only in border part of
flooded caldera (PI. 1, Fig. 1). Most probably unit A corresponds with
pre-caldera series and is limited by large en echelon faults, dipping downward
to the center of caldera (Fig. 2, Pis. 1—3). These faults correspond with caldera
264
Włodzimierz K o w a l e w s k i et
al.
rims in the successive stages of its development. The unit A certainly consists
mainly of pyroclastic agglomerates (strongly consolidated) and probably also
interbedded lavas (locally).
Unit B. The unit B probably corresponds with a some older part of the
post-caldera series. This unit fills up the center of flooded caldera and takes up
the largest part of it. The maximum thickness of unit B is ca. msec, but its lower
limit in the central part (like as lower limit of the unit A) has not been
identified. Unit B is mainly composed of consolidated pyroclastic agglomerates
and tuffs, associated with slopes of cones around the caldera rim. In our
opinion this unit was formed before the opening the entrance and filling the
caldera depression by sea water.
Unit C. The unit C was formed in its lower part (sub-unit Cj) mainly as a result
of sudden deposition on the flooded caldera floor of large quantities of different
size particles of the pyroclastic materials carried down by large lahars from
onshore part (PI. 1, Fig 2). The upper part of the unit C (sub-unit C ) contains
pyroclastic materials of associated cones or lahars and laterally interbedded
layers of the waterlain deposition (common in central part of the floor). All the
unit C has thickness ca. 50 msec in the central part of flooded caldera and from
5 to 15 msec in its border part. Dotted area (Fig. 2) on the floor corresponds
with this unit.
2
Unit D. The unit D, the youngest of all fourth units, fills the depressions and
covers the elevations of the floor as a more or less thick carpet i.e. the unit
D lies discordantly on the older units (PI. 1, Fig. 1). The cover it forms is
approximately 1 to 25 msec thick. It is composed of interbedded layers of tuffits
and agglomerates. The lower part of this unit (sub-unit D ) contains layers of
lahars with intercalations of the layers built of materials gravitionally deposited
through the water. These waterlain deposited and distinctly horizontal
stratificated layers predominant in the upper part of the unit D (sub-unit D ) .
This sub-unit fills the center of the flooded caldera floor and occurred on its
bottom surface. The dashed area (Fig. 2) correponds with the surface of the unit
D, but only where the unit is thicker than 1—5 msec.
t
2
The surface of the caldera floor is so hard (cf. Ashcroft 1972) that it is
impossible to take up core samples from the bottom by piston corer. The
pyroclastic sediments deposited under water are quickly cemented by ashes in
fine fraction that fill interstices between bigger particles. In addition, con­
solidation may be connected also with high temperature. On the shores where
fumaroles were active, some kind of the hard plates were observed in 1988.
These „beach rocks" occurred in shallow part of the bottom and were formed
very recently from the loose materials creeped and slided from the coast
(formed after the 1970 eruptions).
Seismoacoustic studies of D e c e p t i o n Island
265
All the units are cut by numerous faults, vents and other kind of the
intrusions. The bigger faults are situated around the caldera's floor as a ringfracture in its border part (e.g. see faults in the east side of the profile 7W on the
PI. 1, Fig. 1). Some of the faults, especially in the western part of the caldera's
floor are still active {see Pis. 1 and 2).
Volcanic vents have different sizes and occurred in the western part of the
floor. Most probably they are composed of pyroclastic agglomerates. Some
vents continue to the cones rise above the present floor level. All these cones
and vents are young. The oldest ones are younger than the unit A, the youngest
ones were probably formed during 1969-1970 eruptions.
In the southwestern part of the caldera floor large young intrusions were
identified (Fig. 2), may be of lakkolit type. The border part of the one of them is
presented in the east side (PI. 2, Fig. 2) and west (PI. 3, Fig. 1) sides.
Conclusions
On the ground of the detailed seismoacoustic profiling (CSP — boomar)
the structure of the caldera in its flooded part was determined. Good
penetration up to ca. 150—200 m beneath bottom level and good resolution
were obtained. Four main seismoacoustic units were distinguished. The unit
A probably corresponds to pre-caldera series and contains consolidated
pyroclastics, agglomerates and tuffs and probably intercalations of the lavas.
The units B, C and D belong to the post-caldera series and contain different
pyroclastic materials also strongly consolidated. The unit B was formed mainly
as a slope of cones surrounding the caldera rim. Units C and D are also
abundant in the volcanic debris redeposited from the onshore part by lahars,
glaciers, landwater, shore and slope processes and by wind.
All the units are cut by numerous faults and vents. The evidences of young
and even today vital volcanic activity are common on the bottom, especially in
its western part. The cones and vents which occurred here are undoubtedly of
the post-caldera age. The youngest ones are associated with 1969—1970
eruptions.
Our results suggest that caldera was formed undoubtedly by collapse of
central part of the single composite strato volcano {cf. Adie 1964, Baker et al.
1969, Gonzales-Ferran et al. 1971, Ashcroft 1972, Newhall and Dzurisin 1988).
Collapsing occurred in successive stages, more or less en block, along ring faults
— en echelon in types. The subsidence is still active.
The opening of the entrance in caldera rim and flooding the caldera bottom
by sea was determined as formed after deposition of the unit B. The unit C and
D were undoubtedly deposited under water on the flooded caldera floor.
266
W ł o d z i m i e r z K o w a l e w s k i et
al.
References
Adie R.J. 1964. G e o l o g i c a l history. — In: R. Priestley, R.J. Adie and G. Robin (eds), Antarctic
Research. Butterworth, L o n d o n ; 118—126.
Ashcroft W. A. 1972. Crustal structure of the S o u t h Shetland Islands and Bransfield Strait. —
Br. Antarct. Surv. Sci. Repts., 66:
1-43.
Baker P. E. 1969. A v o l c a n o erupts beneath the Antarctic ice. — G e o g r a p h . M a g a z i n e , L o n d o n ;
11: 1 1 5 - 1 2 6 .
Baker P. E., D a v i e s T . G . and R o o b o l M . J . 1969. Volcanic activity at D e c e p t i o n Island in 1967 and
1969. — Nature, 224: 5 5 3 — 5 6 0 .
Gonzales-Ferran O., M u n i z a g a F. V. and M o r e n o H. R. 1971. Sintesis de la e v o l u t i o n volcanica de
isla D e c e p t i o n . — Publ. I N A C H Ser. Cie., 1: 1—14.
K o w a l e w s k i W., R u d o w s k i S., Zalewski S. M. and Ż a k o w i c z K. 1987. Investigations of the seafloor
of the caldera of D e c e p t i o n Island by C S P m e t h o d s . — Proc. 5-th Intern. S y m p . Antarctic
Earth Sci., C a m b r i d g e , vol. abstr., 81 p.
K o w a l e w s k i W., R u d o w s k i S. and Zalewski S. M. 1989. B u d o w a dna z a t o p i o n e j kaldery wyspy
Deception
(Antarktyka
Zachodnia),
na
podstawie
profilowania
sejsmoakustycznego.
— Mat. XVI S y m p . Polar., Toruń.
N e w h a l l Ch. G. and Dzurisin D . 1988. Historical unrest at large calderas of the world, 2. — Bull.
G e o l o g i c a l Surv., U S A , sp. issue,
1013-1019.
Schultz C h . H . 1972. Eruption at D e c e p t i o n Island, Antarctica, 1970. — Bull. G e o l . Soc. Amer., 83:
2837-2842.
Received July 5, 1990
Revised and accepted July 20, 1990
Streszczenie
W y s p a D e c e p t i o n stanowi wulkan z kalderą p o ł ą c z o n ą z o t w a r t y m m o r z e m w ą s k ą cieśniną
(fig. 1 i 2). W pracy p r z e d s t a w i o n o rezultaty profilowania s e j s m o a k u s t y c z n e g o ( C S P , b o o m a r )
w y k o n a n e g o p o d c z a s Ekspedycji G e o d y n a m i c z n y c h d o Antarktyki Z a c h o d n i e j (pl. 1—4), zor­
ganizowanych
przez Instytut Geofizyki
PAN
w latach
1984/85 i 1987/88. U z y s k a n o
dobrą
penetrację d o g ł ę b o k o ś c i ok. 2 0 0 m poniżej p o z i o m u dna zatopionej części kaldery. W y r ó ż n i o n o
cztery g ł ó w n e j e d n o s t k i sejsmoakustyczne. Jednostka A o d p o w i a d a z a p e w n e tzw. seriom prekald e r o w y m i z b u d o w a n a jest z piroklastycznych, silnie s k o n s o l i d o w a n y c h a g l o m e r a t ó w i tufów,
p r a w d o p o d o b n i e z wtrąceniami law. Jednostki B, C i D należą d o serii p o s t k a l d e r o w y c h i są
z b u d o w a n e z r ó ż n o r o d n e g o , s k o n s o l i d o w a n e g o materiału piroklastycznego. J e d n o s t k a B zawiera
głównie o s a d y s t o ż k ó w otaczających kalderę. Jednostki C i D , u t w o r z o n e p o d w o d ą (już p o zalaniu
kaldery w o d ą m o r s k ą — p o p o w s t a n i u bramy w pierścieniu kaldery), zawierają także znaczne
ilości kruszywa w u l k a n i c z n e g o r e d e p o n o w a n e g o z lądu przez lahary, l o d o w c e , w o d y
lądowe,
procesy brzegowe i s t o k o w e oraz przez wiatr. Jednostki są pocięte przez liczne uskoki i k o m i n y
wulkaniczne. D z i a ł a l n o ś ć wulkaniczna jest nadal żywa, zwłaszcza we wschodniej partii
dna
kaldery.
Rezultaty w y k o n a n y c h b a d a ń wskazują na z a p a d l i s k o w e p o c h o d z e n i e kaldery w w y n i k u
osiadania, mniej lub bardziej en block,
centralnej partii p o j e d y n c z e g o , z ł o ż o n e g o stratowulkanu.
Osiadanie n a s t ę p o w a ł o etapami wzdłuż u s k o k ó w typu s c h o d o w e g o , p o ł o ż o n y c h d o o k o ł a scianv
kaldery. N a j m ł o d s z e z nich są związane z w y b u c h a m i
1969—1970.
P O L . P O L A R R E S . , V O L . 11
W . K O W A L E W S K I et at., P L . 1
1. Selected seismoacoustic profile No. 7 W (boomar record). Location on the Fig. 1. Seismoacoustic units:
A — pyroclastic agglomerate and tuffs, pre-caldera series; B — pyroclastic agglomerates and tuffs, mainly slopes
of cones, older part of the post-caldera series; C — pyroclastic materials deposited on the flooded caldera floor,
post-caldera series; sub-unit C, mainly lahars and locally slopes of cones; sub-unit C lahars interbedded layers of
tuflits, D — the youngest part of post-caldera series, sub-unit D , waterlain settled pyroclastic materials and
lahars, sub-unit D mainly materials waterlain settled horizontal bedding, only some intercalations of lahars,
W intrusions, pyroclastic agglomerates riddle or not dykes. Faults are marked. Horizontal lines mark two-time
travel every 10 msec
2
2
2. Selected seismoacoustic profile No. 4S (boomar record). Explanations on PI. 1, Fig. 1
POL. POLAR
RES.. V O L .
11
W. K O W A L E W S K I
et al., P L . 2
1. Selected seismoacoustic profile N o . 2 0 W (boomar record). Explanations o n PI. 1, Fig. 1
2. Selected seismoacoustic profile N o . 2 1 W ( b o o m a r record). Explanations o n PI. 1, Fig. 1
POI
P O I AR
RES., VOI
II
W. K O W A L E W S K I
et al.. P L . 3
1. Selected seismoacoustic profile N o . 2 3 W (boomar record). Explanations o n PI. 1, Fig. 1
2. Southern part of the caldera. Mt. K i r k w o o d — region of the eruption in 1842 — in the back. A n
excellent e x a m p l e of the maar in the first plane. P h o t o by S. R u d o w s k i , February
1988
POL. P O L A R
RES.. VOL.
11
W. K O W A L E W S K I
el al., P L . 4
1. N o r t h w e s t e r n part of the caldera, the region of eruptions 1 9 6 7 — 1 9 7 0 , new c o n e s and craters in the
back. A fragment of caldera rim (pre-caldera lavas?) in the first plane. P h o t o by S. R u d o w s k i ,
February
1988
2. T h e „Black Glacier" in the east side of the caldera. Thick pyroclastic layers interbedded ice.
A m o r a i n e below, at the glacier. P h o t o by S. R u d o w s k i , February
1988