Jökull - 01.12.1984, Page 21

Amax.
mm.
Fig. 6. Time history of the tremor bursts, as recorded on the KV seismograph. Maximum amplitude of
each burst (in mm on the seismogram) and the duration of the burst are indicated on a time axis.
6. mynd. Óróahviður samkvœmt skjálftamœlinum á Kvískerjum. Stœrsta útslag hverrar hviðu er sýnt á
tímaás, ásamt tímalengd hviðunnar.
was highest during the period from 15h on May
28 to 17 h on May 29. After that the activity
subsided rapidly, but small bursts could be seen
until 01:38h on June 2.
Spectral characteristics of the tremor are diffi-
cult to discern in the analog records, but it seems
as if the background tremor has a more
monochromatic character than the tremor bursts.
The predominant frequency is slightly lower than
3 Hz, except during two short periods on May 28
and 29, when the frequency appeared to be
slightly higher than 3 Hz. The predominant fre-
quencies of the tremor bursts were between 2 and
5 Hz.
INTERPRETATION AND DISCUSSION
Since the seismic observations are practically
the only physical measurements available on the
1983 Grímsvötn eruption, their interpretation re-
lies to a large extent on experience from other
recent eruptions in Iceland, in particular the
1980-81 Hekla eruptions and the 1975-81
Krafla eruptions.
The long term increase in seismicity prior to
the eruption bears close resemblance to increased
activity observed in Krafla before some of the
eruptions and rifting events, both with respect to
time scale, seismicity rate, and maximum magnit-
ude. In Krafla this seismicity occurred at small
depth within the caldera and above the inflating
magma chamber (Einarsson 1978). The earthqu-
akes were associated with brittle fracturing of the
crust in the stress field produced by increased
magma pressure in the chamber. By analogy we
suggest that the earthquakes during the last few
months before the Grímsvötn eruption were ass-
JÖKULL34. ÁR 19