1953-57 ERUPTION OF TULUMAN VOLCANO: RHYOLITIC VOLCANIC ACTIVITY IN THE NORTHERN BISMARCK SEA

M. A. REYNOLDS

AMAX PETROLEUM (AUSTRALIA) INC., 1'.0. BOX J663, PERTH, AUSTRALIA

J. G. BEST

:KAJUARA MINING CORPORATION PTY LTD,, P.O. BOX 728, NORTH SYDNEY, AUSTRALIA

R. W. JOHNSON

BUREAU OF MINERAL RESOURCES GEOLOGY AND GEOPHYSICS, P.O. BOX 378, CANBERRA CITY, AUSTRALIA

GEOLOGICAL SURVEY OF PAPUA NEW GUINEA MEMOIR 7, 1980

______

Page 5

S U M M A R Y

Rhyolitic volcanic activity off the southern tip of Lou lsland in the St Andrew Strait area of northern Papua New Guinea began in June 1953. By the end of the eruption in January 1957, two new islandsÑthe Tuluman IslandsÑhad been formed. Seven phases of volcanic activity and eight principal centres of eruption are recognised. The earlier phases were characterised y dominantly submarine activity and produced fields f floating vesicular lava masses. The climax of the eruption was reached in the fifth phase, between February and June 1955, when five different vents were active. Activity then declined, and there followed a seven-month period of quiescence. In November 1956 the final phase took place, producing subaerial lava flows. Many of the volcaniclastic deposits of the Tuluman Islands were eroded away after the end of the eruption, and the lava flows of the final phase of activity now make up the greater part of the larger of the Tuluman Islands. The Tuluman rhyolites are chemically similar to those found on other islands in St Andrew Strait and are thought to have originated by partial melting of the basaltic crust that overlies a conjectured mantle hot spot beneath the strait.

I N T R O D U C T I O N

Fourteen Quaternary volcanoes in Papua New Guinea are known to have erupted during the last 100 years (Cooke and Johnson, 1978). With one exception, these active volcanoes are associated with seismically active late boundaries, and occupy volcanic provinces characterised by basaltic and andesitic rock compositions. The exception is Tuluman volcano in the northern Bismarck (Fig. l). Tuluman is not associated with a plate boundary. Moreover, its rhyolitic rocks are in marked compositional contrast to the basaltic, andesitic, and dacitic rocks produced at the other active volcanoes in Papua New Guinea. Tuluman is a dominantly submarine volcano about 1 km south of the southwestern tip of Lou Island, one of the Admira]ty Islands (Fig. 2). Manus Island is the largest and best known of the Admira]ty group; its main centre is Lorengau,45 km north-northwest of Tuluman (Fig. 1). From the north coast of Manus Island, Miklouho Maclay (1885) (1) observed volcanic activity in March 1883, and concluded that it may have been from Lou Island. However, another possible site was the 130 m-deep shoal just south of Lou Island, which is shown on the Admiralty Chart prepared in 1944. This shoal may have represented a submarine volcano that subsequently became the foundation of the Tuluman Islands. In any event, no new islands that may have formed during the 1883 activity have survived, and there appear to be no other reports pf activity during the next 70 yeatrs - that is ubtil June 1953 when submarine volcanic activoity marked the the beginning of a 31/2 year eruptivwe period.

Footnote 1 Miklouho-Maclay (1885, p. 965) reported his observations as follows: " . . . when at anchor on the north coast of the Great Admiralty Island [Manus Island], I witnessed the eruption of a volcano on the south coast of the island or on one of the small islands south from the big island. It might very likely, have been the volcano on the small island calle by the natives Loo [Lou], and from which they obtain the obsidian for their weapons and implements. It was during the night of March 28th [1883], and I could see a large halo as from an immense fire, and two or three times heavy thunderlike rolling noises were heard, followed by distinct flashes like columns of fire on the horizon."

********************************************************************

p.21:

TABLE 2

PERIODS OF ACTIVITY OF THE TULUMAN CONE

Case   Place  Period         		Weeks	Type of Activity

1	A	27 June-6 July 1953			11/2		Submarine*
	D	20 Oct.-6 Nov. 1954  		2/1/2		Submarine
	E	16 May-6 June 1955  		3			Submarine
	E	6-20 June 1955	 			2			Minor explosive and vapour
	F	20 Sept. 1955							Vapour emission
		TD = 2 yr 3 mth

2	B	14 Nov. 1953-18 Feb. 1954	14			Submarine (Intermittent)
	C	9-11 and 21-27 July 1954	11/2		Submarine reaching surface	
	E	10-12 Feb. 1955				1/2			Minor explosive
		TD = 1 yr 3 mth

3	C	11-14 July 1954				1/2			Surface explosive
		TD = 3 days
1/2
4	E	11-15 Feb. 1955				1/2			Surface explosive
	F	21 July-2 Aug. 1955			2			Minor axplosive & vapour
	F	15 Oct.-25 Nov. 1955		4			Minor axplosive & vapour
	G	24 Nov.-4 Dec. 1956			1			Minor axplosive & vapour
		TD = 1 yr 10 mth

5	E	11-13 and 16-23 Feb. 1955	11/2		Submarine, reaching surface
	E	24 Feb.-11 Mar. 1955         2			Surface explosive & effusive
		TD = 1 mth

6	E	12 Mar.-9 May 1955			 9			Submarine (Intermittent)
		TD = 2 mth

-	F	26-28 Sept., 3-7 Oct. 1955    1			Submarine

7A	G	7-26 Nov. 1956				  3			Mainly vapour emission
	G	27 Nov.-11 Dec. 1956		  2			Surface explosive

7G	G	30 Nov.-2 Dec. 1956			  1/2 		Submarine
	G	11 Dec. 1956-28 Jan. 1957     7  		Surface effusive & minor explosive
		TD = 21/2mth

  TD is total duration of cone life, between first and last recorded activity.

********************************************************************

page 41

SUMMMARY AND CONCLUSIONS

1. Tuluman is one of several Quaternary volcanoes the St Andrew Strait area of northern Papua New Guinea. These volcanoes are not associated with any of the currently active plate boundaries in Papua New Guinea and are thought to have developed over a mantle hot spot beneath the Bismarck Sea.

2. Tuluman volcano lies off the southern tip of the arcuate Lou Island and was not definitely known to exist before June 1953, when a 31/2-year period of intermittent volcanic activity began. Seven phases of activity (A to G) and eight main centres of eruption (Cones 1-6, 7A, 7B) are identified. Except for Cone 3, all the centres occupy a 2 km-long zone, trending northeast.

3. Phases A to D were characterised by submarine extrusion of lava flows from Cones 1 and 2, producing floating masses of lava, but at Cone 3 during Phase C, an island was formed, and a remnant of this still remains today. The climax of the eruption was in Phase E during which Cones 1, 2, 4, 5 and 6 were all active. Explosive activity built up a large island consisting of the products of Cones 2, 4 and 5, but during and after the following 1 hase F much of it was eroded away. After a seven l onth interval of apparent inactivity, Phase G began, 1 iving rise to subaerial lava flows, some of which (from one 7B) were joined to the remnant of Cones 2 and 4. uch of the volcaniclastic part of the island was tes which had accumulated in and partly choked thevents during quiescent periods, or to condensation of vapour within the clouds. At least some of this material may have been sulphur or its compounds. Material of similar origin is thought to have been responsible for the yellow froth on the sea during some periods of submarine activity.

Gases such as sulphur diox;de and hydrogen sulphide were present in minor quantities; amounts were never sufficient to cause discomfort near the active centres, although stainless steel objects (e.g. wrist watches) sometimes became heavily tarnished. The origin of the faint smell resembling burning tar noticed on some occasions was not determined.

____

page 42

removed by erosion during the following twenty years, but the lava flows remain intact at the present day.

4. Some periods of activity appeared to coincide with solstice, equinox, and syzygyÑin particular, the climactic explosive events of Phase E in early March 1955 when the sun and moon were both close to the zenith of Tuluman.

5. The Tuluman rocks are alkali-rich rhyolites similar in composition to those found on Lou and Pam Islands. All these rocks are thought to represent partial melts of the basaltic crust believed to underlie St Andrew Strait.

6. The Tuluman eruption never reached catastrophic proportions; there was no loss of human life and very little damage to property. No substantiated accounts of activity have been reported from the area since 1957, but because the Tuluman eruption appears to represent the latest stage in a progression of rhyolitic volcanic activity along an arc centred on St Andrew Strait (and defined principally by twelve volcanic centres on Lou lsland), further activity on the southern extension of the Lou-Tuluman arc would not be unexpected. The arc may be the trace of a nascent ring fault that outlines a cauldron block; this is speculation, but the possibility should be borne in mind when the long-term eruptive potential of the area is considered.

Return to The Electronic Volcano

Last Revision June 17,1995