History and Surveillance of Volcanic Activity on Jan Mayen Island

A. G. SYLVESTER Department of Geological Sciences, University of California, Santa Barbara, California 93106 U.S.A.

Abstract Jan Mayen is a small isolated Arctic island located on or very near the .junction of the southern end of Mohns ridge and the central part of the Jan Mayen fracture zone. The alkaline petrology and tectonic setting of Jan Mayen are similar to some other volcanic islands akmg the mid-Atlantic ridge, particularly the Westmann Islands of lceland. Both the Westmann Islands and Jan Mayen are underlain by a relatively thick oceanic(?) clnast, about 15 km thick, and recent eruptions were initiated and accompanied by earthquakes with loci from 25 to 30 km deep. The northern halt of the island is dominated by Mt. Beerenberg, a major central volcano composed mainly of alkali olivine basalt. The oldest exposed lavas are of upper Pleistocene age. Flank eruptions on Beerenberg in 1732, 1818 and 1970, together with historic and cartographic evidence for similar eruptions in the 16'" century' and the middle oF the 19'" century, suggest an eruption I:requency of 150 years _+ 75 years. A three-station seismograph network, six surface tilt-measuring sites, and a gravity and levelling profi!e comprise the main features o1 the present stlrveillance system.

Introduction A m a j o r v o l c a n i c e r u p t i o n o c c u r r e d on J a n M a y e n in 1970 s i m u l t a n e o u s l y with, o r w i t h i n a few h o u r s a f t e r a series of m o d e r a t e l y s t r o n g e a r t h q u a k e s (SIGGEaUD, 1972). G e o m o r p h i c e v i d e n c e of r e c e n t v o l c a n i c a c t i v i t y is a b u n d a n t o n t h e i s l a n d , e r u p t i o n s h a v e b e e n o b s e r v e d in 1732 (ANI~EaSON, 1746) a n d 1818 (ScoRESBY, 1820), a n d e a r t h q u a k e a c t i v i t y is c o m m o n . N e v e r t h e l e s s , t h e 1970 e r u p t i o n w a s so u n e x p e c t e d a n d i m p r e s s i v e , a n d a p p e a r e d t o c o n s t i t u t e s u c h a n imm e d i a t e t h r e a t to the i m p o r t a n t w e a t h e r a n d n a v i g a t i o n s t a t i o n at

--

314

--

Olonkinbyen, that personnel of the station were abruptly evacuated to the Norwegian mainland even though the station was 30 km from the eruption site. As it turned out, the evacuation was a costly and unnecessary undertaking, but one that drammatically pointed out a serious and previously unrecognized hazard to the continued operation of the station. It became clear that alternatives to total evacuation were needed in the event of future eruptions, because the island is often ice-bound, preventing easy evacuation by ship, and frequent storms and fog make evacuation by aircraft an impossibility at any given time. Thus, in addition to development of new emergency contingency plans, a seismograph network was established upon the island in 1972 to provide surveillance of future activity and to give, if possible, a warning about when and where an eruption is likely to occur so that the station personnel itself might decide what actions to take.

Tectonic

Setting

Jan Mayen is located close to the intersection of the southern end of Mohns ridge and the central part of Jan Mayen fracture zone (Fig. 1). As such, it is the northernmost volcanic island on the midAtlantic ridge system and the northernmost active volcano in the world. Nord Jan, the northern half of the 380 km 2 island, is dominated by the large shield volcano, Beerenberg, which is about 25 km in diameter at sea level and is capped by a nearly symmetrical summit lava cone reaching an elevation of 2277 m above sea level (Fig. 2). Eruptive activity is presumed to have occurred on Jan Mayen since late middle Tertiary time on the basis of wave-terrace and glaciation morphology (FITCH, 1964) and tephra chronology studies in deep-sea cores (SvLVESTER, 1976); however K-Ar age determinations show that the oldest emergent lavas on Nord Jan are of upper Pleistocene age (FITCH et al., 1965). The location of Jan Mayen may also be described as being at the north end of the Jan Mayen ridge, a narrow submarine ridge which stretches southward almost as far as the Icelandic Plateau. Rather than being a typical spreading ridge, seismic reflection and refraction studies (JOHNSONand HEEZEN, 1967; SORNES and NAVRESTAD,1975) and recent deep-sea drilling data (TALWANIe t al., 1975) show that the eastern half of the ridge is a well-developed continental rise consisting of

315

an east-dipping layered sequence u n c o n f o r m a b l y overlain b y nearly flat-lying beds from 100 to 300 m thick. JOHNSON and HEEZEN (1967) suggested that the ridge is a detached relict of the Greenland continental rise. They concluded, in turn, tha the b a s e m e n t beneath the layered sequence must be continental, and that it and the layered sequence split off from Greenland from l0 to 30 m.y. ago. The GLOMAR CHALLENOER failed to penetrate the b a s e m e n t when it drilled on Jan Mayen

5,

1.

s r ~a,.,~ ~ p . ~ r . ¸'

S"

S!

pr~l'S

fr~lgl~fe

Z0~e

it,cl,~e

/

I |

%< ..... "

%

:',x:z:

..........:4!

Fro. I - Idealized tectonic s e t t i n g s of m i d - A t l a n t i c volcanic i s l a n d s , J a n M a y c n . Westm a n n I s l a n d s ( m o d i l i c d a l t e r WXRD, 1971), Sl. Paul'~ R o c k s {mc*dified al:im TIIOMPS0N a n d MELSON, 1972), a n d A s c e n s i o n [ s l a n d (moclificd a l t e r V:',~ AXI~EI. e t al., 1973}.

ridge in summer, 1974, but p r e d o m i n a n t l y tcrrigenous sandstone and sandy m u d s t o n e of late Eocene and early Oligocene age were encountered beneath the u n c o n f o r m i t y and were tentativelv regarded as supporting evidence for these hypotheses (TALWANI el al., 1975). Seismic refraction studies ( S O R N E S and NAVRESTAD, 1975) also show that the crust thickens to a b o u t 18 km beneath Jan Mayen. This contrasts with crustal thickness of about 6 km elsewhere in this part of the Norwegian-Greenland Sea, but the extent of the thickened area ut crust is not well-known, because of limited data. The voluminous and persistent activity from Beerenberg is apparently due to the tectonic relationship b e t w e e n Mohns ridge and

- -

316

- -

Jan Mayen fracture zone, but the exact nature of that relationship is not clear. Elsewhere, such as in Iceland (WARD, 1971) and on other mid-Atlantic volcanic islands (Fig. 1), major active central volcanos are located at ridge crest-fracture zone intersections or at kinks in the ridge crest. Presumably these structural anomalies constitute zones of crustal weakness which are exploited and maintained as eruption loci for some time. The mid-Atlantic volcanoes do not originate upon a ridge crest - - Iceland being on exception - - but form at ~, favored locations ~ off to one side of a ridge (MENARD, 1969) and the eruptive products are generally characterized by alkaline olivine hasalts (THOMPSON and MELSON, 1972). Such favored locations apparently persist for several million years, but according to MENARD (1969), their origin and reason for persistence are not yet completely understood. Among the various hypotheses is that of MENARD and ATWATER (1968; 1969) which postulates that intrusion and extrusion may occur in fracture zones if the spreading direction is re-oriented so that a component of extension across the fracture zone accompanies regular passive transform separation. In such a case the spreading rate across a fracture zone would be slower than that across the adjacent ridge and is a function of the sine of the angle made by the ridge and fracture zone. VAN ANDEL et al. (1969, 1971) have used this idea to explain the geometry and heat flow anomalies of the Vema fracture zone, as have THOMPSON and I~IELSON (1972) for the petrology of undersaturated alkaline mafic and ultramafic plutonic intrusive rocks in the St. Paul's fracture zone (Fig. 1). The volcanic activity in the Westmann Islands of Iceland, including Surtsey and the recent eruption on Heimaey (Fig. 1), occurs at the intersection of the east end of the Reykjanes fracture zone and the southern extension of the east rift zone (WARD, 1971). Ascension Island (Fig. 1) also lies near a m a j o r ridge-fracture zone intersection, but its origin cannot be ascribed to the simplified re-orientation model outlined above (VAN ANDEL et at., 1973). The location and persistence of Beerenberg as a central volcano may be explained either by ,, leakage ~, from the Jan Mayen fracture zone, as postulated above for other mid-Atlantic volcanic islands, or by eruption from a short segment of a spreading ridge caught between e n d c h e l o n segments of Jan Mayen fracture zone. The dominantly alkaline petrologic character of the lavas is not typical of midocean ridge basalts, however, and available recent hathymetric data (JOHNSON, 1975) do not conclusively show that this part of Jan Mayen

317 --

fracture zone is b r o k e n into e n d c h e l o n segments. VAN ANDEL e t al. (1971) postulate that spreading takes place along the edges of a fracture zone, and that the intrusions are p r o b a b l y thin dikes parallel to the trend of the fracture zone. Although B e e r e n b e r g is on the edge of the fracture zone, eruptive activity typically occurs along fissure zones trending northeast, nearly perpendicular to the Jan Mayen fracture zone. Short faults have been found perpendicular to fracture zones (DE'I:RmK el al., 1973), but it does not seem likely that the long northeast-trending rift zones on Jan Mayen are of this nature. Thus, the ultimate origin of Jan Mayen and its relationship to the neighboring geotectonic structures would seem to require a m o r e complicated explanation than currently available data permit. However, the petrology and tectonic setting of Jan Mayen are s h o w n to be remarkably similar to that of H e i m a e y in the W e s t m a n n Islands of Iceland, suggesting that the origin and n a t u r e of the volcanic activity on Jan Mayen may be typical of volcanic islands with similar tectonic settings.

Structure

and

Petrology

Jan Mayen island presents a rugged landscape, particularly a r o u n d Beerenberg where nearly vertical wave-cut cliffs rise as high as 500 m t-rom the sea. Beerenberg itself is blanketed by snow fields and abouf 20 gliaciers, some of which flow directly into the sea. S u m m i t eruptions have melted some of the gliaciers, resulting in flooding and mudflows down the flanks of Beerenberg (Fvrcn, 1964). Fractures, faults. dikes and long fault-controlled(?) sea-cliffs strike northeast-southwest parallel to the strike of Mohns ridge and WNW-ESE parallel to the Jan Mayen fracture zone. Most of the island is c o m p r i s e d of relatively thin a n k a r a m i t i c and trachybasaltic lava llows and dikes rich in olivine and chromiandiopside xenocrysts, together with tephra, agglomerate, interlayered scree and till (FITCH, 1964). Trachyte plugs occur in Sor Jan (CaRSTENS. 1961, 1963), and occasional trachyte and quartz trachyte xenoliths are found in the volcanic rocks in Nord Jan (F1TCH, 1964). Tile summit lava cone on Beerenberg is formed of a very distinctive glomeroporphyritic basalt rich in large plagioclase phenocrysts (Frrcl-l, 1964). Nord Jan was constructed in three stages (Fvrcll, 1964): 1) Formation of the main shield volcano by eruption of thin flows and py-

--

318

--

roclastics from a central vent and from radial fractures followed by a m a j o r explosive phase, 2) Eruption of more viscous lavas and formation of the s u m m i t lava cone, and 3) Eruptions from northeast-trending fissures on the flanks of the dome. The Havhestberg Formation of probable latest Pleistocene age records at least one major explosive episode in the shield-building first stage. It is a basalt pumicetuff several meters thick composed mostly of agglomeritic or blockbearing basalt sillar or non-welded ignimbrite deposited as an ashflow, and lesser amounts of air-fall and water-lain tuff (FITCr~, 1964). The high proportion of TiO2 and K20, the strongly xenocrystic nature, and the high degree of differentiation of the alkali olivine basalts of Nord Jan are consistent with derivation from depths of 35-70 k m by partial melting of a pyrolitic magma, and with the interpretation that the resultant magmas rose rapidly and directly to the surface (HAWKINS and ROBERTS, 1972; WEIGAND, 1972). The trachytes at Sor Jan, however, may have been derived by differentiation in shallow reservoirs f r o m an ankaramitic-trachy-basaltic parent from which clinopyroxene was removed (HAWKINSand ROBERTS, 1972).

Historic

Activity

Jan Mayen is so remote, is so often shrouded in low clouds and fog, and has been so rarely inhabited in historic time that reports of volcanic activity are few, and even those reports were generally discounted until the 1970 eruption. Since that eruption, however, field investigations have found many areas of lava and tephra that appear to be quite recent. The earliest suggestion of volcanic activitity on the island was in a book published in 1558 in Venice by Nicolo and Antonio Zeno. The book contains a description of the St. Thomas Zenobium Cloister which was said to have been built of stones cast glowing hot from the mountain, and which was heated by canals of hot water. The cloister was located upon an island which, with respect to the position of Iceland on a map accompanying the book, would correspond to Jan Mayen (RICHTER, 1968). Descriptions of volcanic activity in 1732 (ANDERSON, 1746) and 1818 (Scom~sBv, 1820) have been discounted by several authors writing prior to the 1970 eruption. Now, however, these reports seem more credible in light of the 1970 eruption and because of the recent

- -

319

- -

discovery of fumarolic activity a few kilometres n o r t h e a s t of Eggoya (Fig. 2), w h e r e the explosive eruptions in 1732 and 1818 are t h o u g h t to have taken place (FITCH, 1964). Kossletta, a lava plain at the n o r t h end of Jan Mayen (Fig. 2), was probably f o r m e d between 1820 and 1882 (SIGGERUD, personal commun., 1973). The plain is clearly shown on the 1:100,000 m a p of the 1882-1883 Austrian expedition, but not on Scoresby's 1820 map. Otherwise the two maps generally correspond to one a n o t h e r and

....7 " I "

' " "-

,£}/,.,~']"~1970

Area

i Beerenbe}gi. (~{~7~,gruption

JeW

.... '

,

'-"~:"~"

NORD

JAN MAYI:N , JME 1732, 1818 - ,' '\,~ .... lE,upt, on Areo ( Y ) "

SBR

-,?JAN MAYE'N j-J

( [cl(;. 2 - L o c a t i o n m a p (black clots).

s h o w i n g the a r e a o1 the

1970 e r u p t i o n

al~d s e i s m o g r a p h

sitc~

arc fairly accurate. Indeed, the Austrian map was the best available map of Jan Mayen until 1954 (RIcH'rt-'.R, 1968). Kokssletta is such a large and prominent geographic feature (4 km along, I km wide) that it seems highly unlikely Scoresby could have overlooked it, particularly in view of the general accuracy of the rest of his map. After having walked across Kokssletta and the newly f o r m e d 1970 lava plain in August 1973, the writer concluded that the Kokssletta plain f o r m e d m u c h mor~ recently than 2500-3000 years ago as FITCH (1964) postulated on the basis of wave-terrace and glacial geomorpholugy. t h e writer based his conclusion upon the presence of delicate flow structures on the surface of lava flows, the relatively minor degree

- - 320 - of wave erosion, the paucity of vegetation relative to that on older flows, and the slight amount of drifted and windblown sand on the plain. Owerview of the 1970 E r u p t i o n

The 1970 eruption (SIGGERUD, 1971, 1972) probably began on 18 September simultaneously with or within hours after a strong earthquake was felt at Olonkinbyen (Fig. 2). Earthquakes were also felt on 19 September (Table 1), but station personnel were not aware an eruption was in progress largely because of an intense storm and its heavy cover of low clouds over the island from 18 to 21 September. On 20 September a Japanese commercial aircraft reported a high cloud over the island similar to those observed in Japanese volcanic eruptions, and German and Italian commercial aircraft reported seeTABLE 1 - T h e l a r g e s t e a r t h q u a k e s Data from the catalogue

Date

Time (GMT)

at commencement o f 1970 J a n f o r t h e I.S.C., Edinbtu-gh.

°N

oW

Depth (kin)

Mayen

Mag.

eruption.

N. of Stations

1)

18 Sept.

02 06 30.1 ± 0.24

71.27 ± 0.039

7.3 - 0.12

28 ~ 3.6

5.1

160

2)

19 Sept.

20 57 10.3 ± 0.62

71.23 ± 0.087

8.0 7 0.25

33 (assumed)

4.4

19

3)

19 Sept.

21 32 49

+_ 1.3

71.3 ~: 0.17

8.1 :~ 0.42

33 (assumed)

4.2

17

4)

19 Sepl.

21 58 06

± 2.8

71.6 = 0.12

6.6 ~ 0.49

69 - 30

(3.9)

17

ing fire and smoke shortly thereafter. The weather cleared on 21 September, permitting visual observation of the smoke and steam cloud from Olonkinbyen, and the eruption site was observed directly from Norwegian aircraft and ships sent from the mainland. Subsequent aerial, sea, and ground investigations (SW,6ERUD, 1972) found that lava erupted initially from a northeast- trending fissure on the northeast flank of Beerenberg. The fissure was about 6 km long and 600 m above sea level at its southwest end; it descended nearly to sea level at the northeast end (Fig. 2). Activity soon concentrated in five major vents approximately from 1 to 1.5 km apart. Lava erup-

- -

321

- -

ted explosively f r o m the topographically highest vents in fountains up to several h u n d r e d meters high. The lava then s t r e a m e d d o w n steep n a r r o w canyons to the sea, w h e r e the flows coalesced to f o r m a delta with an ultimate v o l u m e of 0.5 km 3. The sub-aerial part of the delta was a b o u t 4 km long and 1 km wide late in O c t o b e r before it was partially eroded by strong waves and drift-ice. E a r t h q u a k e activity was high t h r o u g h o u t the eruption. According to SIGGERUD (1972), from 200 to I450 e a r t h q u a k e s per day were recorded with a seismic event c o u n t e r from 29 S e p t e m b e r until 12 October. Average counting levels were from 600 to 800 events per day. The c o u n t e r was located in the middle of the island 25 km from the nearest crater. The b a c k g r o u n d was set for surf waves of a full storm, and only separate events 70 % above the b a c k g r o u n d were counted. When the instrument ceased to function in the beginning of November, six weeks after the beginning of the eruption, the count rate was 476 earthquakes p e r day. This is a very high level ot activity when c o m p a r e d to the n u m b e r s of local, non-volcanic events; the greatest n u m b e r of e a r t h q u a k e s ever recorded in one day on Jan Mayen was 90 during an e a r t h q u a k e s w a r m in J a n u a r y 1973 comprising 305 separate events (NAvRESTAD and SORNES, 1974). The eruption a p p a r e n t l y ceased in mid-October, but s m o k e clouds were observed intermittently from March 1971 to June 1971 (SmGERUO, 1972). Small quantities of ash and steam were e r u p t e d from the s u m m i t crater of Becrenberg and the satellite cone of Eggoya after June, 1971; however, steam has frequently been observed from Eggoya since establishement of the w e a t h e r station around 1920. Fumarolic activity and recent faulting were observed in the eruption area in August 1973, and Beerenberg and Egg~ya were still emitting intermittent puffs of steam (SvLvESTER, 1974). It seems rather astonishing that a m a j o r volcanic eruption could go undiscovered for three days by nearly 40 men only 30 km a w a y In fact the eruption w o u l d have remained u n o b s e r v e d even longer had not the commercial aircraft been diverted n o r t h w a r d from regular more southerly routes, and if the w e a t h e r over Jan Mayen had not cleared. This only serves to illustrate how Jan Mayen's remoteness, p o o r weather, and history of long periods of uninhabitation may have combined to hide other historic eruptions. Thus, e r u p t i o n s m a y occur m o r e frequently than every 150 years L 75 years indicated from the observed and suspected eruptions in the 16th century, 1732, 1818, i820-1882, and 1970. 9 ll

- - 322 - -

Seismological Studies Regional Studies The seismicity of the Norwegian-Greenland Sea was clarified by SYKES (1965) who showed that epicenters of moderate earthquakes recorded from 1965 to 1964 defined what has since been recognized as a system of spreading ridges and transform fracture zones representing the Arctic extension of the mid-Atlantic ridge system. This was confirmed by HUSEBYE et al. (1974) using NOAA data through 1972. The few available first motion studies are also compiled by HUSEBYE et al. (1974) and yeld interpretations generally compatible with strike-slip movement on the fracture zones and normal faulting on ridge crests.

Local Studies Seismographs have been operated on Jan Mayen from 1962 to May 1970 and from 1971 to the present (NAVRESTADand SORNES, 1974). Unfortunately, the interval of inoperation in 1970 included the fivemonth period before the 1970 eruption and several months afterwards. A vertical component seismograph (Trolldalen) was located near Olonkinbyen from 1962 to 1963. It was converted from a smoked-drum to a pen-recorder and moved a few kilometres up Trolldalen in 1963 where it operated continuously and without incident until May 1970. At that time it was considered too costly for the amount of information it yielded on the seismicity of the Norwegian-Greenland Se~a, and it was dismantled. In retrospect, this proved to be an unfortunste decision both from scientific and eruption-forecasting points of view. Prior to the eruption neither local nor remote recordings of seismic activity gave any reason to suspect that Jan Mayen, more than any part of the Kolbeinsey ridge-Jan Mayen fracture zoneMohns ridge system was due to have a volcanic eruption in September 1970. Had the seismograph in Trolldalen been in operation during the summer and early fall of 1970, it is possible that premonitory activity, if it occurred, might have been recorded. It is questionable, however, if it would have been recognized as premonitory in view of the rather frequent occurrence of local earthquakes and earthquake swarms in the Jan Mayen region. Moreover, many months often elaps10

- - 323 ed between the recording and interpretation of Jan Mayen earthquake data, so that it is unlikely that the eruption might have been forcasted had the Trolldalen seismograph been in operation. There is no question, however, that the records would have provided much useful information for understanding future seismic and volcanic activity on Jan Mayen. From 1962 to 1970 local seismic activity around Jan Mayen was characterized by frequent earthquake swarms and moderate single shock-aftershock sequences within 20-80 km of Trolldalen (NAVRESTAD and SORNES, 1974). Since the 1970 eruption, the level of activity has been rather high 5-20 km northwest of Beerenberg relative to that before the eruption.

10~

9°

8°

7~

6°

5c

o

3

• I

JT~

e2

,.~,~

ERUPTION

J q71 0°

i --

~

--~

FIG. 3 - E p i c e n t e r s o1' t h e f o u r l a r g e s t e a r t h q u a k e s T a b l e I [ o r [ocal d a t a .

--~70 5° preceding

the

1970 ~.,,-uption. S e e

Data are given in Table l for the main earthquakes associated with the beginning of the 1970 eruption, and the epicenters are plotted in Figure 3. ZoB~N (1972) determined a focal rnechanism for the first main shock. Using 20 stations, he concluded that the earthquake occurred on a nearly vertical strike-slip fault, the northeast-southwest strike of which is parallel to the strike of the eruption fissure zone. According to Zobin's interpretation, the relative movement on the fault was right-lateral, which agrees with B~RKENMA.IER'S (1972) analysis of en #chelon fissures and tension gashes. The 28 km focal depth of the first main earthquake is unique and of particular interest, because 1) it is deeper than most mid-ocean

- - 324 - ridge earthquakes, and 2) e a r t h q u a k e s w i t h comparable focal depths occurred one day before the 1973 H e i m a e y e r u p t i o n in the W e s t m a n n Islands, Iceland. This is significant because the tectonic setting, crustal structure and volcanic petrology of the W e s t m a n n Islands (PALMASON and SAEMUNDSSON, 1974) are r e m a r k a b l y similar to Jan Mayen (Table 2). According to BJ/3RNSSON a n d EINARSSON (1975), the Heimaey e a r t h q u a k e s are the deepest to be located w i t h confidence in Iceland, lying f r o m 8 to 18 k m below the u p p e r m a n t l e b o u n d a r y (Vp -- 7.2 k m / s e c ) where partially m o l t e n rocks are expected. Tectonic earthTABLE2- Comparison of the Beerenberg 1970 and Heimaey 1973 (Iceland) eruptions. 1an Mayen 1970

Tectonic setting

On JMFZ near intersection with Mohns ridge

Heimaey 1973

On Reykjanes FZ near intersection with eastern Neovolcanic zone (WARD, 1971) 14-15 km (P~.LMASON,1971)

Crustal thickness (Layer 3)

18 km (SoaNES and NAVaESTAD,1975)

Focal depths of main earthquakes

28-30 km (ISC, Edinburgh this paper)

20-30 km (BJiJaNSSON and EINARSsoN, 1975)

Eruption mechanism

Initially from NE-trending rift; ultimately from five central craters (SI6ur~RvD, 1972)

Initially from NNE-trending rift ; ultimately from one central crater

Petrology

Alkali olivine basalt (WEt6A.~D,1972)

Mugearite-hawaiite (JAKOBSSONet al., 1973)

quakes and e a r t h q u a k e s w a r m s are c o m m o n in Iceland, but they originate only at depths less t h a n 5 to 8 km. If the deeper e a r t h q u a k e s are f o r e r u n n e r s of Icelandic eruptions, they constitute an i m p o r t a n t criterion to distinguish volcanic e a r t h q u a k e s w a r m s f r o m harmless tectonic s w a r m s (BJ6RNSSON and EINARSSON, 1975), at least at the south end of the Neovolcanic zone in s o u t h e r n Iceland where the a b r u p t increase in crustal thickness (PALMASON, 1971) and the dominance of alkalic lavas point t o w a r d a deeper m a g m a source t h a n is t h o u g h t to be typical elsewhere in Iceland (J~OBSSON, 1972). The 12

--

325

close tectonic and petrologic similarities b e t w e e n the recent H e i m a e y and B e e r e n b e r g eruptions suggest that the occurrence of ,, deeper ~ focus e a r t h q u a k e s at Jan Mayen m a y precede e r u p t i o n s there, too.

Current

Surveillance

Introduction Several factors place tight constraints upon any system one might p r o p o s e to monitor all geophysical p h e n o m e n a that might presage a volcanic eruption c;n Jan Mayen. These factors include severe w e a t h e r conditions that virtually limit field w o r k to a few weeks during the summer, poor accessibilty that limits placement of field installations requiring more than annual servicing to the central part of Jan Mayen - - far from Beerenberg - - and the relative hazard to personnel and p r o p e r t y that necessarily limits the scope of investment and effort. Clearly the most practical surveillance system w o u l d be a ,, remote-sensing ,, systern - - one which records signals from field installations at a central location and requires little or no field maintenance, but one which provides real-time information to the personnel on the island. The following sections describe a seismograph n e t w o r k and a simple tiltmeter that comprise the present surveillance systcrn on Jan Mayen. It is quite limited by most standards. Ideally, for example, a seismograph should be placed north of Beerenberg to telemeter data to Olonkinbyen, to the mainland, or to a satellite. Unfortunately, this is b e y o n d the scope of the project at this time. The objective of the tiltmeter study was to d e t e r m i n e w h e t h e r or not crustal tilt is occurring, and if so, how its m a g n i t u d e and direction relate to the tectonic and volcanic activity. The present surface tiltmeter system appears to be affected by surface phenomena, although analysis of available data does not define precisely what these p h e n o m e n a are or how to successfully remove them. Installation of bore-hole tiltmeters would certainly help to overcome some of these problems.

Seismic Sludies The present seismograph n e t w o r k is c o m p r i s e d of three stations as summarized in Table 3 and located in Figure 2. All s e i s m o m e t e r s are Geotech, model S-13 with free periods of 1 sec. Signals are trans13

326 - m i t t e d b y surface cables to a central receiver at Olonkinbyen w h e r e they are recorded on magnetic tape. The vertical geophone at JMI is visually r e c o r d e d as well on a helicorder. In the event of an earthquake, a threshold trigger on the vertical JMI s e i s m o m e t e r starts a six-channel visual r e c o r d e r writing at a chart speed of 1 c m / s e c for 10 seconds and then for 55 seconds at a chart speed of 0.1 cm/sec. All five s e i s m o m e t e r s and coded timing signals are transcribed f r o m TABLE3 Station data for the Jan Mayen seismograph network. -

Commenced Operation

Station

Location

Components

JMI

70°55'41.9" N, 08°4Y50.9" W

1 vertical 2 horizontal

Jan. 1972

JMW

71°01'43.0" N, 08°25'41.I" W

1 vertical

Sept. 1972

JME

70~59"23.Y' N, 08q7'48.7" W

1 vertical

Sept. 1972

the magnetic tape with a delay time of 18 seconds. This permits the o b s e r v e r to determine differences in P-wave arrival times with a precision b e t t e r than ~ 0.1 second. The o b s e r v e r can then graphically locate the epicenters on a set of isoline maps using S-P arrivals for all three stations and using differences in P-wave arrivals a m o n g all three stations. The isoline m a p s are based u p o n crustal velocity calibration studies carried out by the Seismological Observatory in August, 1973 (SORNES and NAVRESTAD, 1975). A local magnitude scale has been developed (WESTRE, 1975) using the m e t h o d of coda duration (BISZTRICSANY, 1958; LEE et al., 1972). Magnitudes of events greater than Mb = 5 are o b t a i n e d from U. S. Geological Survey (Boulder, Colorado) and ISC (Edinburgh, Scotland) reports. Preliminary real-time location of e a r t h q u a k e epicenters is done daily at Olonkinbyen so that the island personnel and seismologists in Bergen m a y be alerted immediately if p r e m o n i t o r y activity is detected. Otherwise, detailed processing of data m u s t be delayed 4-6 w e e k s until the magnetic tapes are received in Bergen. At any time, however, the island personnel m a y c o m m u n i c a t e with the Seismological O b s e r v a t o r y b y radio regarding their observations, interpretations and possible emergency m e a s u r e s if an eruption seems imminent. 14

327

Crustal Movements Studies Two phenomena, which h a d no k n o w n precedent on Jan Mayen, suggested that m a j o r crustal uplift and tilt m a y have been associated with the 1970 eruption: 1) The tilting of a 100 m high radio a n t e n n a at Lagunevollen to as m u c h as 30 cm out of plumb, and 2) A 75 cm uplift of the sea-water intake well at Olonkinbyen with respect to sea level. Both of these p h e n o m e n a were observed in spring 1971 after the cessation of the eruption, and it is not known if the movements were gradational, or abrupt, nov even when began. Indeed, the tilting of the antenna may have been due to differential stress within the supporting cables, and it m a y be argued that it was a non-tectonic effect which increased gradually to the point of being noticed only after the eruption. The intake well at Olonkinbyen is built within a massive lava flow. The intake pipe was originally set to be p e r m a n e n t l y below sea level and was so for several years before the eruption. During and after the eruption, however, it was above sea level at low tide. Ultimately 75 cm had to be cut off the top of the pipe to reduce it to its original level below sea level. Such indications of large movements, if they were indeed volcano-tectonic, would not be considered e x t r a o r d i n a r y near the edifice of an active volcano. Geodetic and tiltmeter studies have shown that the edifices of Hawaiian and Japanese volcanos d e f o r m greatly and may increase in elevation several metres as m a g m a rises into their edifices and inflates them (MOGI, 1958). These d e f o r m a t i o n s are generally documented by repeated precision levelling, triangulation, and laser-ranging, tide gauges, and tiltmeters (DECKER and KINOSItVI'A, 1972). A tide gauge was installed in the sea-water intake well (near JML in Fig. 4) in the s u m m e r , 1974, by the geodetic division of the Norwegian Geographic Survey. A precision level line, about 4 km long, was run from the well to the aircraft control tower (JMF in Fig. 4). The tide gauge records continuously, and the data are analysed periodically by the Geographic Survey. The level line will be resurveyed each summer, weather permitting. Six tilt-measuring sites were established on Jan Mayen in 1973 at sites as far apart from one a n o t h e r - - but still accessible - - as possible. Sites were selected according to locations of suitable existing concrete slab foundations. Their locations and orientations arc' shown in Figure 4 and given more exactly in Table 4. Measurements 15

328 - are taken with a simple tiltmeter w h i c h is essentially a 1.2 m long hollow b a r of stainless steel in the center of w h i c h is m o u n t e d a machinist's level b u b b l e having a 10-second rating. At one end of the b a r is a m i c r o m e t e r which is u s e d to level the b a r and to provide m e a s u r e m e n t s . The b a r is placed u p o n small steel plates fixed permanently to concrete floor slabs, and readings are t a k e n daily, weekly, or as f r e q u e n t l y as weather, accessibility, and w o r k schedules

I°

t~

71 °

FIG. 4- Location map showing tilt-measuring sites.

permit. The data include t e m p e r a t u r e s of the b a r and the m a x i m u m and m i n i m u m daily t e m p e r a t u r e as r e c o r d e d by the Jan Mayen weather station. Reproducibility of a single m e a s u r e m e n t is ± 0.005 mm, whereas sensitivity of the i n s t r u m e n t is calculated to be a b o u t five parts in l0 s, which is from one to two o r d e r s of magnitude less than m o s t s t a n d a r d tiltmeters, b u t which ought to be sufficient to docum e n t direction and m a g n i t u d e of tilts large enough to. cause an antenna to tilt 30 c m and to cause a p o r t i o n of the island to be uplifted 75 cm. R a w data are plotted at Olonkinbyen and then are sent every 4-6 weeks to Bergen for c o m p u t e r processing. The direction and plunge of the tilt vector are calculated and plotted as s h o w n in Figure 5. Correlation analyses b e t w e e n tilt at various sites a n d temp e r a t u r e s are also done by the c o m p u t e r . 16

--

329

--

N o w a f t e r m o r e t h a n t w o y e a r s o f o p e r a t i o n , sufficient d a t a h a v e a c c u m u l a t e d f o r all sites, e x c e p t J M G , to d r a w t h e f o l l o w i n g conclusions: 1) R e l a t i v e l y l a r g e tilts a r e r e c o r d e d at s i t e s l o c a t e d ion ~, b e d r o c k >> (JML, J M B , J M F ) a n d l e s s e r tilts on s i t e s l o c a t e d u p o n s a n d f o u n d a t i o n s (JMC, J M H ) . Th{s is c o n s i s t e n t w i t h t h e e x p e r i e n c e w i t h t i l t m e t e r sites in C a l i f o r n i a ( R e x ALLEN, U. S. G e o l o g i c a l S u r v e y , p e r s o n a l c o m m u n . , 1972) a n d H a w a i i (R. I. TILLING, p e r s o n a l c o m m u n . , 1975). T:tULE 4 - Station data for the Jan Ma3en till-measuring sites. Site

Location

Azi,mtth.~ o[ l . e g . ~

Freqttetwy o[ Observation

L~tabliM#cd

.IML

70"55.0" N, 08"44.0" W

300,030

Evcl3' other day

2(} August 1973

JM B

70"56.6" N, 08"40.4' W

040,:130

Daily

20 April 1973

JMF

70"56.6' N, 08"40.3' W

130 920 -

Every other day

27 June 1973

JMC

70"58,1'W N 08"33.2"

340,070

WeckN"

27 June 1973

JMH

70:'58.0" N, 08"41.l' W

187,277

Whenever possible

19 August 1973

J MG

71"00.8" N. 08"27.9" W

(}45,135

Whenever poss ib le

21 August 1973

2) Tilt a z i m u t h s a r e fairly c o n s t a n t at m o s t s i t e s d u r i n g the o b s e r v a t i o n p e r i o d s h o w n in F i g u r e 5 e x c e p t at J M B w h e r e the azim u t h of the tilt v e c t o r h a s r o t a t e d t h r o u g h 360" in the l a t t e r p a r t of 1973 a n d the e a r l y p a r t of 1974. 3) T h e c h a r a c t e r a n d / o r r a t e of tilt c h a n g e d at all sites at t h e t i m e s of t w o m o d e r a t e local e a r t h q u a k e s felt at O l o n k i n b y e n d u r i n g the e n t i r e p e r i o d of s t u d y . One of t h e s e e a r t h q u a k e s is i n d i c a t e d in F i g u r e 5. T h e o t h e r o c c u r r e d on April 16, 1975, a b o u t 10 k m n o r t h w e s t of JMC w i t h Mb = 6.5. 4) T h e r e is n o e v i d e n t o r c o n s i s t e n t c o r r e l a t i o n b e t w e e n tilt a n d t e m p e r a t u r e v a r i a t i o n s of t e m p e r a t u r e - i n d u c e d p h e n o m e n a , s u c h as f r o s t heave. 12

--

330 - -

5) Significant tilt has not occurred at JMC, which is located near the radio antenna that'tilted 30 cm out of plumb about the time of the 1970 flank eruption of Beerenberg.

1t -

!

~]

ar~

~J~

~

J97.

g I J~a

i ~

dM

~- ,o.

F~6. 5 - Representative tilt record for 1973-74. Arrows show tilt azimuth with respect to north average over 10-day intervals. Dashed line in October 1973 indicates date of occurrence of the only earthquake felt at Olonkinbyen during this time period.

I t is p o s s i b I e t h a t t h e l a r g e t i l t s r e c o r d e d b y t h e t h r e e s i t e s n e a r Olonkinbyen reflect deformation along nearby faults which are presumed to be present from the geomorphology, but which have never 18

- -

331

- -

been mapped. Thus, the available data clearly indicate the need for a better understanding of the local geological structure. Moreover, even though there is no correlation between temperature, frost heave, and tilt, it would be an advantage to place the tiltmeter beneath the surface where surface effects could be ruled out with greater confidence. Thus, future plans include acquisition and placement of continuous-recording bore-hole tiltmeters in some existing bore-holes around Olonkinbyen and in holes that would have to be drilled nearer Beerenberg.

Miscellaneous S t u d i e s

Several volcanoes in other parts of the world sometimes give ambiguous seismic or crustal strain information of impending eruptions. However, periodic m e a s u r e m e n t s of variations in gas temperature and composition (ELSKEN, 1972; IWASAKL 1972; TONANI, 1972) geomagnetism and gravity (YoKoYAMA, 1972), and heat flow (MoxI1AM, 1972) have often been useful for forecasting eruptions. On Jan Mayen the temperatures of gases emitted at Eggaya are measured continuously and carried by cable to Olonkinbyen where they are recorded. No significant variations have been observed since measurements began in 1971. Nineteen gravity stations were established between Beerenberg and the south end of the island in the summer, 1974, by the Norwegian Geographic Survey. The objective is to measure regional vertical crustal movement, but the precision is several orders of magnitude less than can be obtained by precision levelling (DECKER, 1973). The stations will be reoccupied about once a year or at least as frequently as weather permits. Plans have also been made for additional crustal strain meas~ urements on Jan Mayen, such as triangulation and laser-ranging, together with periodic infra-red imagery and geomagnetic measurements. While such studies may provide inttrsting long-term information, they may be of little help if premonitory changes are of periods ot less than a year, because detection would be precluded in the winter, for example, when field studies are not possible. Moreover, surface phenomena, such as frost heave, may make geodetic measurements inconclt,sive or useless. 19

- - 332 - -

Evaluation of the Hazard and Surveillance System Historic data suggest a m i n i m u m eruption frequency of about 150 years +_ 75 years. This, together with the fact that historic activity has been characterized by fissure eruptions only on the flanks of Beerenberg, suggest that future activity will be sufficiently far from Olonkinbyen in space and time that the volcanic hazards are relatively low to personnel stationed on the island. Indeed, after having witnessed both the 1970 Jan Mayen eruption and the 1973 Heimaey eruption in the Westmann Islands of Iceland, an eruption which was similar in m a n y ways to the 1970 Jan Mayen eruption (SYLVESTERet al., 1973), SmGERUD (personal commun., 1973) asserted that Olonkinbyen would be in real danger only if an eruption broke out on the cliffs above the station, in Trotldalen, or beneath the flagpole of the station itself. He maintains that the hazards to personnel can be reduced considerably by careful emergency planning, including establishment of a few widely scattered temporary evacuation camps with enough supplies to last a week or so until total evacuation could be effected if necessary. The principal hazards would be from lava flows and falling tephra. Mud-flows and floods from melting glaciers would be a factor on the flanks of Beerenberg, and to a lesser degree in Trolldalen. The present seismograph network, with certain improvements and the addition of bore-hole tiltmeters, would constitute a system that ought to provide sufficient information to give adeguate warning for the personnel to take appropirate measures for their own safety. In the meantime, an adequate test of the surveillance system and the emergency planning m u s t wait until the next eruption. If historic activity is a reliable guideline, however, that test may not be forthcoming for many decades.

Acknowledgments I became involved in this project while on a two-year assignment in Bergen, Norway, for the University of California. Facilities of the Seismological Observatory of the University of Bergen were kindly placed at m y disposal by the observatory director, M. Sellevoll, and the project director, A. Sornes. Two trips were made to Jan Mayen through the courtesy of the Norwegian Defense Communication Administration. Many useful discussions were held with T. Siggerud of 20

- - 333 - the Norwegian Polar Institute, O. Hagen and C. A. Gloersen of the Norwegian Defense Communication Administration, T. Navrestad and J. Mathisen of the Seismological Observatory, and personnel on Jan Mayen, particularly K. Sandvik and J. Kverndal. G. L. Johnson, E. S. Husebye, and S. Bj6rnsson made preprints of their papers available. The computer programs for reduction, plotting and analysis of tiltmeter data were written by O. Vaagen of the Seismological Observatory. Figures were drafted by Masaoki Adachi and Dave Crouch. Earlier versions of the manuscript benefited from the constructive comments of B. M. Crowe, R. W. Decker, T. Siggerud, A. Sornes, and P. L. Ward. It is indeed a pleasure to express my thanks to all these people and institutions for their help and warm hospitality which has made the experience in Bergen and on Jan Mayen most instrvctive and exciting. References AM~ErsoN, J.. 1746, Nachrichten yogi Island, GriJnland zmd der Slrasse Davis. Hamburg. ,~IRKENMA.IER, K., 1972, Geoteclonic Aspecls o[ tile BeeJenberg I/olcemo E r u p t i o n 1970. Ja~ Mayen Island. Act~ Gcologica Polonica, 22, p. 1-15. B!~:z'lrlCS~.XY, E., 19.58, A N e w .I/leltmd for lilt" Delermi.aliot~ of Mag.ilttde of Earthqttakes. Geofiz. Kozlcmen., 7, p. 69-90. BJiirxss0N, S., and El~,~rss0N, P., 1975, Seismicity o/ Iceland, In: KRISI'J,'~NSSOX,L. (Ed.), Geodynamics of Iceland aml the North A t l a . i i c Atea, p. 225-239, Proc. NATO Advanced Study Institute, Reykiavik, Iceland, 1-7 July, 1974. D. Reidel Publishing Cu., Dordrecht-Holl,and. C',l~s'll-xs, H,, 1961, Cri.stobalite-Trtwhytes o[ .la~7 M a y e . Norsk Pularinsl, Sky. hr. 121, p. 3-10. - - , 1963, Lavas of the, Sottthern Part o[ Jal~ M a r e . . Norsk Polarinst..~,vbuk t'or 1961, p. 69-82. DEcKEr, R. W., 1973, State-o[-the-arz in Volcatto Forecasling. Bull. Voleanul., 37, p. 372-393. and Kl'qoSlll'r.x, W. T., 1972, Geodetic Measuremettts. In: The Sttrveillance arm Prediction of Volcanic activity: A Review o[ Methods and Techniques, p. 47-74. UNESCO Earth Science Series Pub. No. 8. Paris. l-)t~-rr]cl<, R, S., Mt~l~ll:, J. D., Lt'vt~.dnK, B. P. and M~c[m~u.J~, K. C., 1973, Near-bottom Observations o[ all Active T r a . s [ o r m Fazdt (,14id-Athm;ic Ridge at 37"N1. Nature Phys. Science, 246, p. 59-61. ELSKE~S, I., 1972, The Chemistry o/ Volcanic Gases aPTd Its Com,~exion with Prediction o[ Volcanic Activity, In: The Surveillance and Prediction of Volcamc ,4ctivity: A Review o/ Methods amt Techniques, p. 139-144, UNESCO Earth Science Series Pub. No. 8, Paris. FHCH, F. J., 1964, The D e v e l o p m e . t of the BeerepTberg Volcano, .lan Mayen. Prc~_', Geol. Assoc., 75, p. 133-165. - - , Gr~STV, D. L. and Mu.~£r J. A., I965, Polassitim-Argon Ages of R o c k s [rural Jan Mave~ and an Outline o[ Its Volcanic Hi~tc~ry. Nature, 207, p. 1349-13~1. 21

-

-

334

--

HAWKINS, T. R. W. and ROBERTS, B., 1972, The Petrology of the Volcanic and Intrusive Rocks of Nord.Jan, Jan Mayert. Norsk. Polarinst. Arbok for 1970, p. 19-41. HUSEBYE, E. S., GJOYSTDAL, H., BU/gGUM, H. and ELDHOLM, O., (ms.), The seismicity of the Norwegian and Greenland Seas and Adjacent Continental Shelf Areas. IWASaKt, I., 1972, Chemical Surveillance and Prediction of Volcanic Activity. In: The Surveillance and Prediction of Volcanic Activity: A Review of Methods and Techniques, p. 131-137. UNESCO E a r t h Science Series Pub. No. 8, Paris. JArOSSSON, S. P., 1972, Chemistry and Distribution Pattern of Recent Basaltic Rocks in Iceland. Lithos, 5, p. 365-386. - - , PEDERSEN, A. K., RiSNSBO, J. G. and MELCHOIR-LARsF.~"~,L., 1973, Petrology of Mugearite-Hawaiite: Early Extrusives o,f tho 1973 Heimaey Eruption, Iceland. Lithos, 6, p. 203-214. JOHNSON, G. L., 1975, Morphology of the Mid-ocean Ridge between Iceland and the Arctic Basin. In: KRISTJANSSON, L. (Ed.), Geodynamics o[ Iceland and the North Atlantic Area, p. 49-62. Proc. NATO Advanced Study Institute, Reykjavik, Iceland, 1-7 July, 1974. D. Reidel Publishing Co., Dordrecht-Holland. - and HEEZEN, B. C., 1967, The Morphology and Evolution of the NorwegianGreenland Sea. Deep-Sea Research, 14, p. 755-771. LEE, W. H. K., BENNETT, R. E. and MEAGI-I~, K. L., 1972, A Method for Estimating Magnitude of Local Earthquakes from Signal Duration. U. S. Geol. Surv. Open File Report. MENAI~O, H. W., 1969, Growth of Drifting Volcanoes. J. Geophys. Res., 74, p. 4827-4837. and AXWATER,T., 1968, Changes in Direction o/ Seafloor Spreading. Nature, 219, p. 463-467. and - - , 1969, Origin of Fracture Zone Topography. Nature, 222, p. 1037-1040. Moor, K., 1958, Relations between the Eruptions of Various Volcanoes and Deforma. tion of the-Ground~'Fmrfaces around Them.-Bull. "E~rrth~luake.Res. Inst. (Japan), 36, p. 99-134. MoxnAu, R. M., i9720 Thermal Surveillance of Volcanoes. In: The Surveillance and Prediction of Volcanic Activity: A Review of Methods and Techniques, p. 103-124. UNESCO Earth Science Series Pub. No. 8, Paris. N~.VI~ST.~, T., and S~RNZS, A., 1974, The Seismicity around Jan Mayen. Norsk Polarinst. Arbok for 1972, p. 29-40. P~LMASON, G., 1971, Crustal Structure of Iceland #ore Explosion Seismology. Rit. 40, Reykjavik, Soc. Sci. Island, 187 pp. and S.~EMUNOSSO~, K., 1974, Iceland in Relation to the Mid-Atlantic Ridge. Ann. Review, Earth and Planet. Sci. Letters, 2, p. 25-50. RICHTER, S., 1968, Historie. In: MuNcH, J. S., ed., Jan Mayen, p. 30-38, Ottar (Troms~ Museum), n. 56. RGaERTS, B. and HAWKJNS, T. R. W., 1965, The Geology of the Area al"ound Nordkapp, Jan Mayen. Norsk Polarinst..~rbok for 1963, p. 25-48. SCORESaV, W., 1820, Account o[ the Arctic Regions. 1, p. 154M69, Edinburgh. SIGGL~t~, T., 1971, Vulkanutbruddet pd Jan Mayen, h~sten 1970. Naturen, p. 451-473. ., 1972, The Volcanic Eruption on Jan Mayen 1970. Norsk Polarinst. ~,rbok for 1970, p. 5-18. SYKES, L. R., 1965, The Seismicity of the Arctic. Seism. Soc. America Bull., 55, p. 501-518. -

-

22

335

--

SVLVES~R, A. G., 1974, Sur[ace Faulting a n d Fumarolic Activity since the 1970 Beerenberg Eruption, Jan Mayen. Norsk Geol. Tidsskr., 54, p. 385-393. - - , 1976, Petrography of Volcanic Ashes in Deep-sea Cores near Jan Mayen Island; Sites 338, 345-350 DSDP Leg 38. In: TALWANL M. a n d UalNTSEV, G., el al., Initial Reports of the Deep Sea Drilling Project. 38, Washington, U. S. Government Printing Office. - - , SICGEB~t-t), T., NAvrt~s'r,~a. T., MArrrIIS[~N, J. and H.~G~, O., 1973, Surveillance o f Volcanic Activity i~ Iceland as~ Illustrated by the Heinzaey, 1973, Eruptio~z. Seismological Obs., Univ. of Bergen, Spec. Rpt., 21 pp. S~R~ES, A., and N~VREs'rAD, T., 1975, Seismic S,trvey ol Jalt Mayen. Norsk Polarinst. Arbok For ]973. T,~LWaXL M., UI)l~'rsEv, G.. BJORKLt~I), K., C,~STO~, V. N. D., FaAS, R. W., KnantM, G. M., MORRIS, D. A., Mt~LLER, C., NII.S}-N, T. H., vAN ]~lly'rE, J. E., WARNKE, D. A. and WH]lr, S. M., 1975, DSDP Leg ]8 in the Norwegian-Greenltmd Seas. Geotimes, 20, no. 2. T,~zt~-:vI-', H., 1972, .4 Dynamic Approach to the Problem o[ Forecasting Volcanic Paroxysms. In: The Surveillance and Predictio~ o/ Volcanic Activity: A R e v i e w o/ Methods and Techniques, p. 127-130, UNESCO Earth Science Series Pub. no. 8, Paris. TIIOMFSO~, G. and MI~:l.so~, W. G., 1972, The Petrology o[ Oceanic Crust across Fracture Zones in the Atlantic Ocean: Evidence o[ ,t New Kind o[ Sea Floor Spreading. Jour. Geol., $0, p. 526-538. To~,.,xN], F., 1972, Concepts and Tech~tiques [or the Geochemical Forecastbtg o[ Volcanic Eruptions. In: The Surveillance and. Prediction o[ Volcanic Activity: A Review o[ Methods anti Techniques, p. 145-166, UNESCO Earth Science Series Pub. no. 8, Paris. VA~ A~I)t~L, T. H., PIHI.l.l~s, J. D. and \;o~ HF.I~zI~t4, R. P., 1969, Ri[ting Origin [or the V e m a Fracture Zone in the North Atlantic. Earth and Planet. Sci. Letters, 5, p. 296-300. - - , VON HERZEN, R. P. and PIIlLLtI'S, J. D., 1971, The Vema Fracture Zoue and the Tectonics o[ Transverse Shear Zones in Oceanic Crustal Plates. Mavinc Geophys. Res., I, p. 261-283. - - , RI-~A, D. K., vo.~ HEIIZEN, R. P., and Hosl,~b:s, H., 1973, Ascension Fractttre Zone, Ascensio~ Island, and the Mid-Athmtic Ridge. Geol. Soc. America Bull, 84, p. 1527-1546. WARu, P. L., 1971, N e w Interpretation o[ the Geology o[ Iceland. Geol. Soc. America Bull., 82, p. 2991-3012 Wt~lt;~xl), P. W., 1972, Btdk-rock and Mineral Chemistry o/ Recettt Jan Mayelt BasalL~. Norsk Polarinst. ~rbok for 1970, p. 42-52. WEs'r~:, S., 1975, Richter's lokale magnitude og total sig~ah,arighet /~r lokale jord~kjeh, pd .lan Mayen. Hovedfag thesis, Scismol. Obs., Univ. Bergen. Norway, 95 pp. YO~OVAMA, I., |962, Gravimetric, Magnetic and Electrical Methods. In: The Surveillance and Prediction of Volcanic Activity: A Review o[ 14e;hods a~td Techniqt~es. p. 75101, UNESCO Earth Science Series Pub. no. b, Paris. Zoln.~, V. M., 1972, Focal Mechanism o[ V o l c a , ic E a r t h q u a k e s Bull. Volcanol., 36, p..561-571.

Mamc~cript received: Feb. 1975 Revised ms received: Jan. 1976 23