Skip navigation

Please note that this site is an archive, and is not being updated.

You are in:  Home » News items »

The volcanic sequence of the subglacial eruption in Vatnajökull

THE SEQUENCE OF EVENTS - 29 Sept. to 1 Nov. 1996

The Nordic Volcanological Institute, gave the first reports of this eruption:

"On September 29, 1996 at 10:48 an earthquake of magnitude 5 on the Richter scale was detected within the Vatnajökull icecap in SE Iceland. This event was followed by an intense earthquake swarm with a large number of small events with intermittent larger quakes of magnitude 3-4 on the Richter scale. In the morning of October 1st an over flight discovered a subsidence bowl and cracks in the glacier surface at a location where an eruption had occurred in 1938. On Oct 2nd, in the early morning, an over flight observed that an eruption had broken through the ice. Rhythmic explosions resulted in black ash clouds rising to a height of 500 m while the buoyant eruption column rose to 3000 m.

The Vatnajökull ice sheet [USGS Landsat 1 MSS image] is a temperate glacier covering about 8300 km2 in the SE part of Iceland. Volcanic fissure systems of the Mid-Atlantic Ridge plate boundary are partly covered by the western part of the ice sheet. Two major volcanic centers lie beneath the ice, the Bardarbunga volcanic centre and the Grimsvotn volcanic centre both with large subglacial caldera depressions. The last eruption within the Bardarbunga centre occurred in 1910, but eruptions on the fissure system have occurred in 871 AD, 1477 AD and 1862 AD, all producing substantial amounts of lava. The Grimsvotn centre is the more active of the two with an eruption frequency during past centuries close to one eruption per decade. The last eruption occurred in 1983. As Bardarbunga the Grimsvotn centre is a part of a a fissure system which includes the Laki fissure, which in 1783 produced about 12-14 km3 of basaltic lava. Within the ice filled Grimsvotn caldera intense geothermal activity continuously melts the ice to form a subglacial lake, which at intervals of 5 to 10 years is emptied along subglacial channels to create large floods (jökulhlaup) on the alluvial plain, Skeidararsandur (literally "Spoon River Sands"), on the Icelandic south coast. The lake was last emptied in 1995 and the water level is presently low.

The present eruption fissure is located between these two volcanic centres with a direction parallel to the regional tectonic lineament. The subglacial topography directs meltwater from the erupting fissure toward the Grimsvoetn caldera which is rapidly filling. By the evening of Oct. 1st the ice cover above the subglacial lake had risen 10-15 m."

ICELAND REVIEW reported on 10 October 1996

A newly formed 3.5 km long rift has been torn into the surface of the Vatnajökull glacier. Scientists only discovered the gorge yesterday after the weather cleared to allow the first flights over the icecap in almost a week. The cleft lies just to the south of the main crater which was formed at the beginning of the month by a volcanic eruption forcing its way through 450 metres of glacial ice. Scientists estimate that the southern end of the newly formed ravine is 200 metres wide, but that the cleft is much wider at its northern end nearest the crater. The eruption continued unabated yesterday, belching enormous towers of black volcanic ash, smoke, steam and detritus high above the cracked surface of the glacier. At one stage yesterday there were massive explosions of ash from the crater at regular 10-15 minute intervals. Meanwhile the Grímsvötn subglacial basin, a huge depression under the icecap, continues to fill with meltwater. There has, as yet, been no sign of the large scale flooding of the surrounding countryside warned of by scientists.

Magnus Halldorsson at the Science Institute, University of Iceland, reports [updated 28 Oct]:

October 11-14
The amplitude of the eruption tremor, thought to reflect the vigour of the eruption, decreased steadily. The tremor was last seen in the late afternoon of October 13, indicating that the eruption had ceased. Observations from an aircraft on October 11 showed that the height of the eruption column had decreased markedly. On October 12 a little island was seen within a gigantic glacier canyon at the eruption site. No explosive activity has been seen after October 14.

October 16
The jökulhlaup, i.e. draining of the eruption meltwater which has been accumulating under the ice shelf on the Grímsvötn caldera lake, is imminent. The water level in Grímsvötn is now at 1504 m according to a GPS instrument which scientists from the Nordic Volcanological Institute deployed on the ice shelf on October 11. Although the eruption is over, 500-700 m3/s of meltwater is currently flowing into Grímsvötn. Normally jökulhlaups start when water is able to penetrate beneath the ice dam but before the lake level has risen high enough to lift the ice dam, which is at about 1505-1510 m elevation. The lake level was at 1455 m a.s.l. when the last jökulhlaup started in April 1996. The total volume of the lake was 1.5 km3 of which 1.1 km3 drained out during the jökulhlaup which increased gradually to a peak in 3 weeks.

Today about 3 km3 of water are stored in the caldera lake Grímsvötn and the discharge may increase more rapidly than in the normal jökulhlaups, even reach a peak in 3 days (as in the jökulhlaup of 1938) because warm water draining from the eruption site melts the ice walls of the watercourse beneath the glacier and they expand rapidly.

The highest crater along the 7-8 km long hyaloclastite ridge which has now formed beneath the glacier is 250-300 m higher than the former glacier bedrock, i.e. at about 1500 m a.s.l. The top of the crater forms an island within a meltwater lake in a 500 m wide ice canyon. The gorge is 3-5 km long and sits within a 30 km2 depression within the glacier. The ice walls in the canyon are about 150 m high where they sit on the new ridge. The glacier was about 450 m thick at this site prior to the eruption.

This eruption is the forth largest in Iceland during this century and is estimated to have produced 0.6-0.7 km3 of tephra or 0.4 km3 of solid rock. Only the eruptions in Katla 1918, Hekla 1947, and Surtsey are larger.

October 17.
The water level of Grímsvötn has now risen to 1505 m.
The inflow of water from eruption site during the last 4 days is estimated to be 400-500 m3/s. A few microearthquakes within the Grímsvötn region whereas the The Bárđarbunga region is quiet.

October 18
Eruption site cooling as first snow falls on mountain.
Snow covered for the first time the new mountain within the crater formed by the Vatnajökull eruption.
The canyon [image from Magnus Halldorsson] is now estimated to measure around 400m wide and between 1-2km in length. At one point during its period of high activity the canyon was measured at around 3km in length.From Iceland Review.

October 21
The water level of Grímsvötn has now risen to 1507 m.
Microearthquake activity within the Grímsvötn region. The Bárđarbunga region is quiet.

October 25
The water level of Grímsvötn is now at 1509.5 m.
Inflow of water from eruption site is estimated at 200 m3/s. Microearthquake activity in the Grímsvötn region has decreased. The Bárđarbunga region is quiet.

October 30.
The eruption ended on October 13th, no eruptive activity has occurred since. Meltwater continious to flow from the eruptive site towards the subglacial Grimsvotn caldera lake, but at a much reduced rate than before. For the last week the flow of water has been on the order of 100 cubic meters per second, sufficient to raise the water level of the lake by about 0.25 meters per day. Rate of flow is decreasing with time, as the eruptive material becomes cooler.

November 1.
The water level of Grímsvötn is still at similar level as October 25.
Microearthquake activity in the Grímsvötn region is still ongoing, whereas the Bárđarbunga region is quiet.