Jökull - 01.01.2013, Page 107

Seismicity observed under Snæfellsjökull volcano
canoes such as Hrómundartindur, Hofsjökull or Ör-
æfajökull (Jakobsdóttir, 2008). The observed seismic
activity typically occurs in swarms and epicenters are
densely clustered around magma reservoirs or dyk-
ing volumes. Hypocentral depths of volcano related
seismicity reach down to 34 km but events are mostly
clustered within 2–6 km depth (Jakobsdóttir, 2008).
This paper reports the results of our three-months
field campaign and identifies for the first time the
occurrence of seismic activity beneath the Snæfells-
jökull volcano.
10 km
Figure 2. Spatial distribution of the recorded seismic
events around the Snæfellsjökull volcano. The loca-
tions of the seismometers are denoted by blue squares
and the dashed line marks the cross section that is
used for the plot of Figure 3. – Dreifing jarðskjálfta
undir Snæfellsjökli. Staðsetning jarðskjálftamælanna
er sýnd með bláum ferningum, en strikalínurnar
marka staðsetningu þversniðsins á 3. mynd.
DATA AND METHODS
We deployed five Trillium 240 broadband seismome-
ters on the Snæfellsnes peninsula from July to
September 2011, with the locations shown by squares
in Figures 1 and 2. Since our primary objective was
to determine the existence of seismic activity beneath
the Snæfellsjökull volcano, we deployed four stations
in the westernmost part of the peninsula, distributed
5 to 12 km distant from the volcano. We installed
our fifth seismometer approximately 21 km north-
west of the Ljósufjöll volcanic complex because our
second objective was to determine if seismicity oc-
curs between the Snæfellsjökull and the eastern parts
of the peninsula. The stations around Snæfellsjökull
were deployed either on lava flows or on hyaloclastic
rocks. Data were recorded at a sampling rate of 100
Hz. The recordings were manually inspected for seis-
mic events, which were located from hand-picked P-
and S-wave arrivals using the HYPOCENTER code
(Lienert and Havskov, 1995) and the 1D SIL veloc-
ity model introduced by Stefánsson et al. (1993) (see
Table 1). We considered only events showing a S-
wave arrival on the OLA, GLA and STA stations (see
Figure 2; GUF recordings were dominated by large
background noise and are excluded from this study)
and that at least allow for one P-wave pick at any
of the three stations. We used the ObsPy toolbox
(Beyreuther et al., 2010) for data processing.
Table 1. 1D velocity model used for this study, based
on Stefánsson et al. (1993) with 1.74 as vp/vs ratio.
– Einvítt hraðalíkan notað til skjálftastaðsetninga við
Snæfellsjökul.
Depth km vp km/s
0 3.53
1 4.47
2 5.16
3 5.60
4 5.96
6 6.50
9 6.73
20 7.20
RESULTS
During three months of continuous recording, we
identified and located 29 seismic events that were not
listed in the Icelandic seismic catalogue. Numerous
other microseismic events were also detected but are
not included here because the low signal to noise ra-
tio prevented reliable epicenter determinations. Nom-
inal epicenter uncertainties for the reported 29 seis-
mic events range from ±0.5 km to ±3.5 km in both
Longitude and Latitude but are likely underestimated.
However, uncertainties in the phase picking addition-
JÖKULL No. 63, 2013 107