Jökull - 01.01.2011, Blaðsíða 76

S. Magnúsdóttir and B. Brandsdóttir
anfari, 1884). In contrast to the events in 1867 and
1872, formation of fissures in or near Húsavík was
not documented but more than 50 events were felt
there November 2–6th (Fjallkonan, 1884). Some
of the November events may have originated further
east on the HFFS. This earthquake sequence was fol-
lowed by a M 6.3 earthquake in Kelduhverfi in Jan-
uary 1885.
The January 25th, 1885 earthquake sequence
caused considerable structural damage in western
Kelduhverfi (Figure 6) where large surface fissures
and grabens were created (Norðanfari, 1885a,b;
Thoroddsen, 1905). A 100 m wide N-S aligned rift,
with up to 2 m of subsidence (Dokkin) and nu-
merous smaller fissures formed in the sand north-
west of Víkingavatn (Norðanfari, 1885a,b; Elías-
son, 1976a,b). An increase in geothermal activity
was observed around Lón (Haraldsson and Johnsen,
1986). Þórarinn Stefánsson (2002) from Grásíða
(born 1878) stated in his memoirs that during the
summer of 1885 many of the new fissures were
impassable. When measured in 1935, the base of
Dokkin lay 80 cm below the water table of lake Vík-
ingavatn (Elíasson, 1976a). The Dokkin depression
is similar to depressions forming in eastern Keldu-
hverfi during the Krafla rifting episode.
Although most fissures from 1885 within the
sandy soil around Víkingavatn have since disap-
peared, the Dokkin subsidence is still visible and
has a direction close to N15◦E (Magnúsdóttir,
2010). According to Þorgeir Þorgeirsson, farmer at
Grásíða, buried fissures beneath the farm are aligned
NE-SW (Þ.Þ., pers. comm. 27.08.2009) similar to
fissures further south (Figure 5e). Both Lónin and
Víkingavatn are likely to have been created by rift-
ing and subsidence similar to the Skjálftavatn lake
in eastern Kelduhverfi, which formed following the
Krafla rifting event in 1975–1976.
The magnitude of the 1885 earthquake has been
estimated at 6–6.5 (Tryggvason, 1973) based on a
location near the center of Öxarfjörður, ∼25 km
northeast of Víkingavatn (Figure 4). Halldórsson
(2005) revised the location to the shore north of
Víkingavatn with a magnitude of 6.3. Stefánsson et
al. (2008) assigned a preferred left-lateral strike-slip
movement along a N14◦E striking fault to the near
shore epicenter, based on relative locations of micro-
seismic swarms by Rögnvaldsson et al. (1998).
The 1885 event threw people off their feet and
farmhouses at Víkingavatn, Grásíða and Sultir were
severely damaged (Norðanfari, 1885b). Based on in-
tensity reports the 1885 earthquake had a Mercalli
magnitude close to VIII in western Kelduhverfi and
V-VI in Þistilfjörður (Austri, 1885), 55 km to the
east which translates to a Richter magnitude of 5.8–
6.6 (Halldórsson, 2005). The largest 1975 rifting
event in eastern Kelduhverfi MS 5.2 was clearly felt
in Þistilfjörður albeit seemingly with less intensity
than the 1885 event.
A M 6.3 event within 5–10 km of Víkingavatn
would thus be expected to have intensities close to
VIII at the Víkingavatn and Grásíða farms (Figure
6) whereas a M 6.3 event 25 km offshore would
have lower intensities, VI-VII. Similarly, based
on the magnitude relationship derived for Iceland
(Halldórsson et al., 1984), a M 7.5 event 25 km off-
shore would be required in order to generate inten-
sity VIII in western Kelduhverfi.
Based on intensity reports, the 1885 event orig-
inated close to Víkingavatn. The earthquake most
likely has a strike-slip focal mechanism as rift zone
normal faulting earthquakes associated with rifting
are not known to reach magnitude 6. Normal fault-
ing earthquakes during the Krafla rifting episode did
not exceed MS 5.2 (PDE monthly listings, USGS).
Therefore, we conclude that the 1885 event was trig-
gered by rifting within the ÞFS under similar cir-
cumstances as the MS 6.4 1976 Kópasker event.
By comparison, the 1934 MS 6.2 Dalvík earth-
quake had an intensity of VIII-IX within a 10 km ra-
dius (Þórarinsson, 1937). The Kópasker earthquake
caused considerable structural damage within a 10–
20 km radius indicating that its intensity was similar
or slightly higher than the 1885 event. The Kópasker
event was widely felt in N-Iceland (Dagur, 1976).
Based on damage reports we estimate its intensity to
be VIII at Kópasker and NE Kelduhverfi.
CONCLUSIONS
Normal faults within the 70–80 km long (N-S) and
7–8 km (E-W) wide Þeistareykir fissure swarm have
an average azimuth close to N25◦E whereas the av-
76 JÖKULL No. 61, 2011