"The eruption continued through May, strengthening considerably at the end of the month. At 1900 on the 30th, following reports of heavy ash eruption during the day, Civil Defence and government authorities were informed that activity had intensified by the declaration of an alert. No other action is to be taken at this stage.

"A trend of increasing activity began on 6 May, and from the 16th, significantly stronger activity persisted. White vapours crowned the summit until 22 May, forming a column about 1 km high. Grey emissions began to color the base of this column on the 7th. Much stronger activity occurred on the 16th and 17th when heavy ashfalls took place in the summit region and light ash deposits (less than 1 mm) were registered at the NW coast. At this time, convoluted grey ash clouds were rising from the crater.

"During the evening of 19 May a period of about 30 minutes of strong volcanic tremor was recorded. It was discovered during an aerial inspection of the crater on the 20th that 5 new vents had opened. Mixed activity was observed, ranging from steady, convoluted, pale grey ash and vapour emission, to continuous black ash jets, to frequent Surtseyan ash and vapour projections from vents in a shallow lake. Activity continued in a similar fashion over the next few days with variable numbers of vents active.

"On the 24th, a new higher level of activity commenced. Ash emissions were markedly more voluminous, resulting at times in the formation of an ash haze in the caldera. A plume of ash driven by the wind stretched out about 10-15 km from the summit. Ashfalls were heavy in the summit region but only light at the W and NW coasts.

"Similar activity persisted through the remainder of the month, but on 30 May much denser ash emission took place, resulting in moderate to heavy ashfalls on the lower W flanks. An ash plume 50-60 km long formed W of Karkar. At one village 2 km from the W coast an ash deposit 1 cm thick was measured on the morning of the 31st. Severe damage to vegetation was reported from W coast areas.

"During the last week of May the stronger ash eruptions were accompanied by a higher level of seismic activity, and explosive sounds were heard at villages on the lower W flanks. However, levels of volcanic tremor remained relatively low compared with pre-eruptive seismic activity in 1978. Discrete volcanic earthquakes were registered on most days in May.

"The crater lake, which had been observed intermittently since 23 April, has not been seen since 22 May. Rapid infilling of the crater with ejecta began on 24 May. On 31 May part of Bagiai Cone collapsed into the crater, reducing its depth by about 60 m to an estimated 50-80 m.

"A program of gravity measurements which commenced on 13 April has shown a trend of steadily increasing gravity values at the three stations on the caldera floor. The gravity change averaged 0.03 mgal. None of the other stations, situated at coastal, mid-flank, and summit locations, showed any trends. This suggests that the causative agent is spatially restricted to the environs of the inner caldera. Levelling measurements at four dry tilt stations spread from the coast to the inner caldera rim showed no evidence of flank deformation. It is uncertain at present whether gravity effects are caused by elevation changes of the caldera floor relative to the rest of the volcano, or to a change in the density structure of the volcano.

"Copious volumes of muddy water have been observed flowing off the slopes of Bagiai on several days since 13 May, and mudflows on parts of the caldera floor have blanketed underlying lava flows. When smaller, similar streams of water were running off Bagiai in April, it was considered that this was condensation from the emission cloud. However, the large volume of liquid observed on several days in late May suggests that the source in this case may be springs."

Geological Summary. Karkar is a 19 x 25 km forest-covered island that is truncated by two nested summit calderas. The 5.5 km outer caldera was formed during one or more eruptions, the last of which occurred 9,000 years ago. The steep-walled 300-m-deep, 3.2 km diameter, inner caldera was formed sometime between 1,500 and 800 years ago. Cones are present on the N and S flanks of this basaltic-to-andesitic volcano; a linear array of small cones extends from the northern rim of the outer caldera nearly to the coast. Recorded eruptions date back to 1643 from Bagiai, a pyroclastic cone constructed within the inner caldera, the floor of which is covered by young, mostly unvegetated andesitic lava flows.

Information Contacts: C. McKee, RVO.