Jökull - 01.01.2019, Qupperneq 38

Historical accounts of pre-eruption seismicity in Iceland
more related to that of the volcanoes of the volcanic
flank zone to the south. In addition to the variable tec-
tonic position of the volcanoes there is also a variable
environment in which the eruptions take place. An
eruption that would be effusive under subaerial condi-
tions, may turn out explosive if the volcano is covered
by a glacier or is in an oceanic environment off the
coast. The high frequency of eruptions and variability
in the expression of the volcanism make Iceland a pre-
ferred area for the study of volcanic phenomena. For
the local population it is of high interest to monitor
the volcanoes with the ultimate goal to forecast their
activity and issue warnings of impending eruptions.
Einarsson (2018) investigated the relationship be-
tween eruptions and precursory seismic activity for
Icelandic eruptions during recent decades for which
instrumental data are available. All eruptions since the
early seventies were shown to have detectable short-
term seismic precursors. Furthermore, it was docu-
mented that in 2/3 of all cases the precursory activity
was detected soon enough to issue a warning to the
public before the eruption started. In the present paper
an attempt is made to extend the study to a time period
where monitoring was entirely based on visual obser-
vations rather than instrumental monitoring we know
today, and compile all cases of documented precur-
sory seismic activity. The paper includes all known
cases of premonitory seismicity of eruptions of Katla
and Öræfajökull. These volcanoes have not erupted
within the time of instrumental coverage, so no in-
strumental data are available so far.
DATA
Thoroddsen (1899, 1925) continued the compila-
tion of Hannes Finnsson and other authors of earth-
quakes and volcanic activity during Historical time
in Iceland (since the ninth century) as documented
in written form. A good part of his work was based
on historic annals, some of which are now accessi-
ble on the internet (https://ia601406.us.archive.org/-
14/items/slenzkirannlars00sigugoog/slenzkirannlars-
00sigugoog.pdf). Thorarinsson (1967, 1975) revised
some of his compilations and emphasized that the
historical reliability of the different annals is very var-
ied. Contemporary, or nearly contemporary annals
are most valuable, whereas annals written long after
the events are notoriously unreliable. The descrip-
tions of eruptions in old annals are often quite cryptic
and subject to interpretation. The written documents
must be re-interpreted in light of new understanding
of the activity of the volcanoes gained by continuing
research. Therefore, in this paper, where appropriate
the original Icelandic text is repeated for the benefit
of later researchers who might want to test alternative
interpretations. A few of these cases were pointed out
by Björnsson and Einarsson (1981).
For obvious reasons most of the documented cases
are related to activity of the most active volcanoes,
Katla and Hekla. The two other members of this
class of very active volcanoes, Grímsvötn and Bárð-
arbunga, are hardly mentioned in the documents be-
cause of their very remote locations in the interior of
Iceland. Major rifting episodes are documented, like
those of the Krafla activity 1724–1746 in the North-
ern Volcanic Zone (Sæmundsson, 1726) and the Laki
eruption of 1783–1784 in the Eastern Volcanic Zone
(e.g., Steingrímsson, 1788). The Reykjanes Penin-
sula Oblique Rift went through an active phase that
lasted about three centuries and ended in 1240 CE
(Sæmundsson and Sigurgeirsson, 2013). Very few
documents mention this activity and little is known
about the course of events. No volcanic activity is
known there since then. Similarly, the only major vol-
canic event known in the Western Volcanic Zone, the
Hallmundarhraun eruption, took place in early His-
torical times (Jóhannesson, 1989) and is hardly men-
tioned in written documents.
DOCUMENTED CASES OF
EARTHQUAKES ASSOCIATED WITH
ERUPTIONS
Katla
Katla is one of the most frequently erupting volca-
noes in Iceland. The central volcano is partly covered
by the Mýrdalsjökull glacier (Björnsson et al., 2000)
and the 700 m deep caldera is filled to the rim with
ice (Figure 2). Eruptions in Historical times have
been basaltic and have occurred within the caldera.
They have therefore been accompanied by massive
JÖKULL No. 69, 2019 37