Jökull - 01.12.2007, Page 52

Þorbjarnardóttir et al.
During 2006 a geothermal power plant was under
construction in the Hengill area. In February, water
was injected into a borehole on three occasions. After
recording seemingly induced microearthquakes in the
vicinity of the borehole, two extra seismometers were
installed to monitor crustal processes. The induced
earthquakes were relocated using a double-difference
method, where the majority of hypocenters clustered
along a 1 km near-vertical plane northeast from the
borehole. The borehole was 2.2 km deep, which cor-
relates well with the location of the earthquakes at 1–
2 km depth (Vogfjörð and Hjaltadóttir, 2007).
Along the South Iceland Seismic Zone earthquake
activity was low, though fairly constant. No earth-
quake sequences were recorded, but aftershocks were
concentrated along the June 2000 faults, as in previ-
ous years.
Katla volcano
There are two seismically active areas beneath the
Mýrdalsjökull ice cap, within the Katla caldera and in
the currently more active Goðabunga area to the im-
mediate west of the caldera (Figure 1). Seismicity in
the Goðabunga area has followed a seasonal pattern
for the last decades, where the number of earthquakes
in the second half of the year is many times higher
than in the first half. Increased pore pressure in the
crust and the decreased mass of the overlying ice cap
are possible influences on the seismicity rate (Einars-
son and Brandsdóttir, 2000; Jónsdóttir et al., 2007).
In July 1999, earthquake and geothermal activity
increased beneath the Mýrdalsjökull ice cap. A short-
lived jökulhlaup took place (Sigurðsson et al., 2000;
Roberts et al., 2003), a new cauldron formed within
the Katla caldera, pre-existing ones deepened (Guð-
mundsson et al., 2000) and crustal uplift increased
(Sturkell et al., 2003). In 2001 the fall seismicity be-
neath Goðabunga was especially intense and contin-
ued into 2002. Though earthquake activity decreased
slightly over the summer months, it was continuous
throughout the year. This period of unrest continued
to the end of 2004, when the earthquake activity re-
sumed its seasonal nature. At roughly the same time,
uplift decreased within the Katla caldera (Sturkell et
al., 2008).
In 2006 the usual seasonal seismicity beneath
Goðabunga and seismicity within the Katla caldera
were observed. In addition, an unusual swarm of
earthquakes was detected beneath Entujökull, north
of Goðabunga (Figure 5). Unlike the seismic events
beneath Goðabunga, which are characterized by low-
frequency energy, the earthquakes beneath Entujökull
were high-frequency events. Twelve earthquakes,
0.3–2.4 in magnitude, were located at a depth of
7 km from 13–19 November. The hypocenter distri-
bution of these events delineates a vertical, northwest-
striking fault plane. Focal mechanisms suggest left-
lateral movement along the apparent fault.
Vatnajökull Region
From late December 2004 to September 2006, six
seismic stations were installed in an area north of the
Vatnajökull ice cap. The stations were added to the
SIL network to monitor possible induced movement
in the crust associated with the construction of the hy-
droelectrical power plant at Kárahnjúkar (Jakobsdóttir
et al., 2005). The filling of the Kárahnjúkar reservoir,
Hálslón, began at the end of September 2006. On 22
October a magnitude 1.4 earthquake was located at
about 17 km depth east of the reservoir. This was the
only earthquake recorded in 2006 that is possibly as-
sociated with crustal subsidence caused by the forma-
tion of the reservoir.
The addition of the Kárahnjúkar seismic stations
increased the number of earthquakes detected by the
SIL system in the region from northwest Vatnajökull
to the Askja caldera and Herðubreið. The magnitude
of completeness (Mc) has decreased by about 0.3 for
northwest Vatnajökull (Figure 6A), and about 0.5 for
the Askja-Herðubreið region (Figure 6B).
The largest earthquake sequence in the Vatnajök-
ull region in 2006 was a mainshock-aftershock se-
quence, just southwest of Kistufell in the northwest-
ern sector of the ice cap (Figure 7). The mainshock
occurred on 24 September with a magnitude of 3.7 at
a depth of 11.5 km. The seismicity was most intense
during the first three days, but the sequence lasted for
nine days. This is the largest earthquake sequence that
has been recorded in this area, but recorded seismic-
ity has increased since mid 2005. After relocation,
the majority of hypocenters in the sequence clustered
50 JÖKULL No. 57