Jökull - 01.01.2019, Page 48

Historical accounts of pre-eruption seismicity in Iceland
areas. Earthquakes were also felt on several occa-
sions after this first eruption, most likely associated
with lateral dike propagation along the northern fis-
sure swarm of the Askja volcanic system and erup-
tions at Sveinagjá (Figure 1) in the fissure swarm dur-
ing 1875–1876. The activity culminated on March 29,
1875 with a large explosive eruption in the Askja cen-
tral volcano, accompanied by a large caldera collapse,
where the 200 m deep caldera Öskjuvatn was formed
(Sigurdsson and Sparks, 1977, 1978). Little infor-
mation is available about the temporal relationship of
these eruptions with earthquakes.
1921–1933: Several small effusive eruptions occurred
within the main caldera and in the immediate vicin-
ity of the Askja volcano during this period. The only
seismograph in operation was in Reykjavík at a dis-
tance of 260 km. It had a low amplification and
did not record any events associated with this activ-
ity (Brandsdóttir, 1992).
1961: A lava eruption began on a short fissure near the
eastern rim of the main Askja caldera on October 26.
It was preceded by increased geothermal activity and
a few small earthquakes on October 6. The beginning
of the eruption was accompanied by recorded earth-
quakes but the temporal relationship is unclear (Thor-
arinsson and Sigvaldason, 1962; Brandsdóttir, 1992).
The eruption lasted until the beginning of December.
Other areas
1151: Eruption near Trölladyngja on the Reykjanes
Peninsula (Figure 1) with earthquakes and casualties.
[Eldur uppi í Trölladyngjum, húsrið og mann-
dauði (Ísl. ann.,1847, p. 62)]
1211: Eruptions off the Reykjanes Peninsula, at El-
dey, and on the peninsula. Earthquakes. Spatial or
temporal relationships uncertain.
[Eldur kom upp úr sjó fyrir sunnan Reykjanes.
Sörli Kolsson fann Eldeyjar hinar nýju, en hinar horf-
nar er alla æfi höfðu staðið. . . . (Ísl. ann. 1847, p.
88).]
1240: Large earthquakes in south Iceland. Red sun.
Fire off Reykjanes.
[Landskjálftar miklir fyrir sunnan land. Sól rauð.
Eldur fyrir Reykjanesi (Ísl. ann., 1847, p. 114)]
1311: An earthquake during the night between Jan-
uary 10 and 11 was followed on January 25 by an
eruption in the "eastern glaciers", probably meaning
Vatnajökull glacier. May be mixed up with stories
about eruption of Katla, see above.
[Landskjálfti nóttina milli 10. og 11. janúar; 25.
janúar eldur uppi í Austurjöklum . . . (Thoroddsen,
1899)]
1619: An eruption occurred in late July with earth-
quakes and ashfall, but the eruption site is unknown
(Thorarinsson, 1967). Possibly in western Vatnajök-
ull.
DISCUSSION
This study is complimentary to a recent study of short-
term precursory seismic activity associated with re-
cent volcanic eruptions in Iceland (Einarsson, 2018),
based on instrumental recording. It was demonstrated
that all 23 confirmed eruptions during the period
1970–2014 had a measureable precursory seismic ac-
tivity, usually in the form of an intense swarm of small
earthquakes. The precursor time, i.e. the time from
the first detected event until the eruption began, var-
ied between 15 minutes and 13 days, and in half of
the cases it was shorter than two hours. This time was
even quite variable for different eruptions of the same
volcano. For the nine Krafla eruptions during 1975–
1984 the range was 15 minutes to seven hours, for
the eruptions of 1983, 1998, 2004 and 2011 of Gríms-
vötn volcano it was 90 minutes to nine hours. Hekla
stands out for its short precursor times of 25, 23, 30,
and 79 minutes for the eruptions of 1970, 1980, 1991,
and 2000, respectively. The present study was meant
to expand the scope of the instrumental study, and to
re-interpret the historical data in light of the present
understanding of the dynamics of the volcanoes. The
results are in general agreement with the instrumental
data of recent eruptions, that eruptions are preceded
by seismic activity, and the precursor times are of the
order of hours and sometimes days. It is also con-
firmed that the seismicity associated with eruptions of
Hekla is rather weak and the precursor times are short.
The historical data have limitations, of course.
Only felt earthquakes are documented so the study is
limited to large events and to volcanoes that are suf-
ficiently close to inhabited areas. It is very probable
JÖKULL No. 69, 2019 47