Jökull - 01.01.2020, Page 70

The analog seismogram archives of Iceland
performed mostly by students of the University of Ice-
land. The scanning operation was later moved to the
Institute of Earth Sciences, University of Iceland.
THE DATA AND RESEARCH
POSSIBILITIES
The new data archive opens several lines of research
into the activity of the crust in Iceland, both volcanic
and seismic. The installation of seismographs in Ice-
land has often happened in response to large events.
The large earthquakes in the transform zones of Ice-
land in 1910 and 1912, for example, without doubt at-
tracted the attention of the international seismological
community to Iceland and the high activity there and
led to the installation of the first seismograph. The in-
crease in the number of instruments in the fifties was
influenced by the general belief that Katla, the feared
volcano of South Iceland, was likely to erupt around
1960 (Thorarinsson, 1960), an opinion based on its
previous eruption pattern. Otherwise, the central part
of the twentieth century was rather quiet if compared
to previous centuries. This changed in the sixties with
the eruption of Askja in 1961, the large earthquake
of Skagafjörður in 1963, and the prolonged eruptive
activity of Surtsey 1963–1967 (e.g., Sayadi et al.,
2020). Portable seismographs became available and
were used to study the eruptions of Hekla in 1970 and
Heimaey in 1973.
The new analog seismograph system, Landsnet,
came just in time to record the beginning of the Krafla
rifting episode 1975–1984 with its 20 diking events
and 9 eruptions. Several papers are based on data
from these events, but considerable part of the data set
still has not been analyzed to its full capacity. An ex-
ample of the data is shown in Figure 7, that displays
the beginning of the last and largest Krafla eruption
on September 4, 1984, as recorded at the Skinnas-
taður station, 30 km from the eruption site (Einarsson,
2018). The seismogram shows the precursory seismic
activity that began about three hours before the erup-
tion, and the characteristic low-frequency earthquake
about 9 minutes before the eruption. The earthquakes
then decrease as soon as the intruding dike reaches the
surface and feeds the eruption.
The Grímsvötn volcano erupted in 1983 after sev-
eral decades of quiescence, showing typical precur-
sory activity, earthquake swarm and eruption tremor
(Einarsson and Brandsdóttir, 1984). Soon after that
the analog network was expanded into the central
highland of Iceland in order to increase the moni-
toring capabilities for the volcanoes there. This net-
work showed the background activity of Bárðarbunga
and Grímsvötn volcanoes, in addition to the Loki
Ridge, the source area of jökulhlaups into the Skaftá
river (Björnsson and Einarsson, 1990). The enig-
matic bursts of tremor following the jökulhlaups were
also recorded by the network (Figure 8). The bursts
normally last a few tens of minutes and occur at the
end of the water release from the glacial cauldrons
that feed the floods. The similarity of the tremor to
eruption tremor from Grímsvötn led Björnsson and
Einarsson (1990) to suggest that they signified small,
subglacial volcanic eruptions, triggered by the sud-
den pressure drop when the water was released from
the overlying ice cauldron. Other authors suggest
that the tremor may be caused by flash boiling of the
geothermal system beneath the cauldron (Björnsson,
2003; Eibl et al., 2020). Seismicity in connection
with eruptive activity in Grímsvötn 1998 and 2004
was recorded by the network, also the seismicity of
Bárðarbunga volcano leading up to the Gjálp eruption
in 1996 (Einarsson et al., 1997).
In addition to the seismicity within Iceland, the
network also recorded seismic activity on adjacent
sections of the plate boundary. An example is shown
in Figure 9 of a large earthquake swarm in 1990 on the
Reykjanes Ridge, SW of Iceland, reported by Einars-
son (1993). The swarm originated at latitude 63◦N,
about 180 km away from the seismograph station
at Bjarnastaðir. Hundreds of events were recorded
within the next two days, at least 14 of magnitude 4
and larger.
The examples above are only a few selected sam-
ples from a list of remarkable events during the cen-
tury of analog seismographic coverage of Iceland. A
comprehensive list is available on the website seis-
mis.hi.is.
JÖKULL No. 70, 2020 67