Jökull - 01.01.2021, Síða 42

Magnússon et al.
glacial eruption can melt large volumes of ice and
trigger powerful jökulhlaups. The Katla central vol-
cano beneath the Mýrdalsjökull ice cap (Figure 1) is
renowned for violent eruptions that are often accom-
panied by devastating jökulhlaups and tephra fall (e.g.
Eyþórsson, 1945; Þórarinsson, 1975; Larsen, 2000,
2010; Larsen et al., 2013; Óladóttir et al., 2008,
2014; Smith and Haraldsson, 2005; Guðmundsson et
al., 2005a, 2005b, 2007, 2008, 2013; Elíasson et al.,
2005, 2006; Björnsson et al., 2000; Tómasson, 1996;
Russel et al., 2010; Björnsson, 2010, 2017). Katla
has also been identified as one of the largest volcanic
sources of CO2 on the planet (Ilyinskaya et al., 2018).
In view of the above, Mýrdalsjökull was an ob-
vious target for early efforts to survey and map the
bedrock of Icelandic glaciers. The first attempt ap-
plied seismic reflection at nine survey sites in 1955
(Rist, 1967a), suggesting an ice thickness in the cen-
tral plateau of 200–370 m. In 1977 a prototype of
an analogue low frequency radar designed for use on
temperate glaciers (Sverrisson et al., 1980) was used
to survey a few ice thickness profiles on the cen-
tral plateau of Mýrdalsjökull (Björnsson, 1978). The
radar system was designed to operate at frequencies
low enough (3–10 MHz) to penetrate up to 1000 m
thick temperate ice and reflect back enough energy
from the bedrock for detection of ice thickness. In
temperate ice water pockets and tunnels on various
length scales up to tens of metres scatter energy from
frequencies higher than tens of MHz, leaving little or
no energy reflected back from the bed. An overview
of the use of the low frequency radar systems to sys-
tematically map the bedrock of all major ice caps in
Iceland, starting in 1978, was given by Björnsson and
Pálsson (2020). The survey of Mýrdalsjökull in 1977
showed an undulating bed with 500 to 600 m thick
ice in the central part and confirmed the existence of
a 100 km2 caldera below Mýrdalsjökull (Björnsson,
1978), already suggested from interpretation of Land-
sat images (Sigbjarnason, 1973). A distinct internal
reflector at ∼300 m was observed and interpreted as
the 1918 tephra layer. In 1991 a second RES-survey
was done (Figure 2a,b), now with a fully developed
version of the analogue RES-system, and Loran-C
and GPS for navigation, covering most of Mýrdals-
jökull, including the central part, with sounding lines
at 1–2 km intervals. From this survey a topographic
map (referred hereafter as a Digital Elevation Model;
DEM) of both the surface and bedrock was created
(Björnsson et al. (2000). This allowed determination
of ice and water divides, estimation of the ice volume,
location of the caldera rim and twelve subglacial geo-
thermal areas, and discussion of how eruption sites re-
late to the bedrock topography. In the 1991 survey the
internal layer seen in the 1977 profiles was visible in
a large portion of the caldera, again interpreted as the
1918 tephra layer. A study of the depth of the layer
suggested an average mass balance in the caldera
since 1918 of 3.5–4.5 m a−1 water equivalent (Brandt
et al., 2005). The ice thickness was also surveyed at
about 70 sites on Sólheimajökull (see Figure 1a for
place name locations), an outlet glacier in southwest
Mýrdalsjökull (MacIntosh et al., 2000). The glaciol-
ogy group of the Institute of Earth Science (IES) ad-
ditionally conducted RES point measurements at the
northeast corner of the caldera plateau (at Entujökull
outlet), profiles across the K7 depression (ice caul-
dron, Figure 1a) that formed in 1999, and between
Kötlukollar and Háabunga in 2000. An RES-profile
was collected at Kötlujökull in 2003, plus 60 RES
point measurements on the Kötlujökull snout in 2004
(Pálsson et al., 2005).
Since 2001 the mass balance of Mýrdalsjökull has
been measured in most years at a few sites in a col-
laborative effort of the Iceland Glaciological Society
(JÖRFÍ), IES and the Icelandic Met Office (Ágústs-
son et al. 2013). The thickness of winter snow de-
posited in the elevation range 1300–1500 m asl on the
caldera plateau is typically 10–12 m by Mid-May, cor-
responding to a winter balance of 5–6 m (water equiv-
alent thickness). This is comparable to the winter
accumulation measured on the 1800 m high plateau
of Öræfajökull (Guðmundsson, 2000), the site where
highest precipitation levels in Iceland have been mea-
sured. Ablation rates on Mýrdalsjökull (as a func-
tion of elevation) are similar to those recorded on
the Breiðamerkurjökull outlet of southern Vatnajökull
(Ágústsson et al., 2013), resulting in an annual accu-
mulation that typically varies between 2–5 m (water
equivalent thickness) on the caldera plateau.
40 JÖKULL No. 71, 2021