Jökull - 01.01.2013, Síða 102

H. Ágústsson et al.
eral be different with regard to the time of year based
on the different vertical structure of wind and temper-
ature in the impinging air masses and whether solid
or liquid precipitation is predominant in the simula-
tion of each event. Due to the difference in terminal
fall speeds, winds will carry falling snow slightly fur-
ther onto the plateau than they will carry rain, thus on
average resulting in a different spatial distribution of
precipitation depending on the season (Figure 5).
CONCLUDING REMARKS
Mass balance data from the plateau of Mýrdalsjökull
has been described and explored. The winter snow
layer reaches a thickness of nearly 13 metres at the
survey site at the southern edge of the ice cap. The
smallest observed thickness is approx. 6 m at the top
of the ice cap, slightly above the plateau. The water
equivalent for the winters 2001 and 2007–2012 lies
in the range 3.4–7.8 mwe, depending on the location.
The observed summer mass balance varies greatly and
ranges from -0.9 – -3.1 mwe and there is consequently
considerable spatial and temporal variability in the an-
nual mass balance in the accumulation area at the ice
cap plateau (2.1–5.9 mwe).
The observed winter balance of the ice cap and
observations of precipitation from two lowland sta-
tions south of the glacier, have been used to find a
first estimate of the precipitation falling at the sites
during summer, giving between 1–1.8 m of water.
The mean annual precipitation (sum of measured win-
ter balance and estimated summer precipitation) dur-
ing 2007–2011 is as great as 8 m, which is simi-
lar to that estimated for Öræfajökull (Guðmundsson,
2000), but somewhat higher than predicted by previ-
ous linear and dynamical studies (Crochet et al., 2007;
Rögnvaldsson et al., 2004). The measured winter bal-
ance at the sites is successfully reproduced by high
resolution simulations of precipitation from a state of
the art numerical atmospheric model. The simulated
precipitation field verifies the precipitation gradient
expected a priori, and furthermore indicates that the
maximum values may be close to 10 m of water, at the
southeast limit of the plateau. The observational data
as well as the atmospheric simulations indicate that
a large part of the precipitation on the ice cap falls
in liquid form, possibly between 30–50% on an an-
nual basis, but a more detailed analysis is beyond the
scope of this paper. Refreezing of winter rain within
the snow pack contributes to its relatively high den-
sity.
It is important that the systematic mass balance
measurements on Mýrdalsjökull are continued. Ad-
ditional measurements of the mass balance could be
done on the upper parts of the outlet glaciers as well
as on the southeast edge of the plateau where a precip-
itation maximum is expected. Measurements in the
ablation zone are also important and have in fact been
initiated at an outlet glacier reaching the lowlands on
the south side of the ice cap. The observational data
is necessary for verifying mass balance models, simu-
lations of precipitation, as well as meltwater contri-
bution to the glacial rivers. The data may further-
more improve the understanding of the response of
the glaciers to a warming climate and changing pre-
cipitation trends, and for other basic glacial and atmo-
spheric research.
Acknowledgments
The mass balance measurements on Mýrdalsjökull re-
ported here have been conducted by members of The
Iceland Glaciological Society (Jöklarannsóknafélag
Íslands). We thank many society volunteers for
their efforts. Eiríkur Lárusson and Bergur Einars-
son helped organize the trips and contributed sig-
nificantly to their success. The National Power
Company (Landsvirkjun), the Institute of Earth Sci-
ences of the University of Iceland (Jarðvísindastof-
nun Háskólans), the Reykjavík volunteer search and
rescue team (Hjálparsveit skáta í Reykjavík), as well
as the Icelandic Meteorological Office (Veðurstofa
Íslands) provided equipment and vehicles. We thank
Benedikt Bragason at Arcanum glacier tours for his
help. The RÁV-project was supervised by Haraldur
Ólafsson and Ólafur Rögnvaldsson at the Institute for
Meteorological Research (Reiknistofa í veðurfræði /
Belgingur) with a grant from the Icelandic research
fund (RANNÍS). We acknowledge the constructive
comments from three anonymous reviewers, which
led to improvements in the manuscript.
102 JÖKULL No. 63, 2013