Jökull - 01.01.2020, Qupperneq 10

Glacier extent in Iceland, 1890–2019
are published in numerous scientific papers, theses
and reports, and many are not in digital format. The
most complete glacier extent data sets are from the
LIA maximum (here denoted with “∼1890” although
the LIA maximum may have been reached earlier
for some glaciers), 1945–1946, 1970–1980, ∼2000,
2007–2013, 2014, 2017 and 2019. Some glaciers have
a higher number of outlines than others (see the cap-
tion of Figure 12 for references). We have chosen
the years with most complete glacier coverage to be
included in the data set and omitted years with data
covering only a single or a few glaciers.
Several different definitions of a glacier exist. The
GLIMS definition is as follows (Raup and Khalsa,
2010, p. 4): “A glacier or perennial snow mass, identi-
fied by a single GLIMS glacier ID, consists of a body
of ice and snow that is observed at the end of the melt
season. . . . This includes, at a minimum, all tributaries
and connected feeders that contribute ice to the main
glacier, plus all debris-covered parts of it. Excluded is
all exposed ground, including nunataks.” The Interna-
tional Association of Hydrological Sciences (IAHS)
glossary definition (Cogley et al., 2011, p. 45) is: “A
perennial mass of ice, and possibly firn and snow,
originating on the land surface by the recrystalliza-
tion of snow or other forms of solid precipitation and
showing evidence of past or present flow.”
Although the Icelandic glacier outline data set is
intended for GLIMS it was not possible to fully ad-
here to the GLIMS definition since the outlines are
based on existing delineations of glacier margins us-
ing data of many different origins. In particular, out-
lines based on glacial geomorphological evidence (to
delineate the maximum LIA extent) may be assumed
to be closer to the IAHS rather than GLIMS defini-
tion, because terminal and lateral moraines are formed
in areas affected by “past or present flow” of ice. Out-
lines from the map of Icelandic glaciers (Sigurðsson
et al., 2017) are partly based on oblique aerial pho-
tographs from several different times where an effort
was made to exclude perennial and seasonal snow,
thus these outlines are also closer to the IAHS than
the GLIMS definition.
It is a challenge to determine the glacier bound-
ary for debris-covered glaciers (e.g. Paul et al., 2013),
which should according to the GLIMS definition be
included within the glacier polygon. The main ar-
eas where debris-covered glacier snouts are no longer
connected with the active glacier front are Rjúpna-
brekkujökull and the outlets north of Köldukvíslar-
jökull of northwestern Vatnajökull, covering 25 km2
in 2019, the north and east flowing outlets of Kverk-
fjöll (northern Vatnajökull), covering 5 km2 in 2019
and Klofajökull (northern outlet of Eiríksjökull), cov-
ering 2 km2 in 2019. These areas show little surface
lowering in recent decades.
When these debris-covered glacier parts become
detached from the active glacier, flat proglacial areas
characterized by sandur plains, sometimes crossed by
glacier rivers, may emerge and the adjacent ice mar-
gin on the inside of these can be considered the “ac-
tive” margin of the glacier in question. These debris-
covered parts have been delineated separately and
submitted to GLIMS as polygons of debris-covered
glacier. They are also delineated within the main
outline of the respective glacier, as required by the
GLIMS definition and are, therefore, included in
the calculated glacier area reported here. We have
only defined such debris-covered-glacier polygons
for moraine or dead-ice fields that do not seem to
be part of the active glacier and not for dirty or
debris-covered glacier snouts that participate in the
terminus variations and thickness changes of the re-
spective glacier. Ice-cored terminus moraines and
debris-covered dead-ice buried in sandur plains or
completely detached and located far from the glacier
are, however, not included within our glacier outline
database. We have not systematically mapped such
areas in the neighbourhood of glaciers in Iceland,
which is beyond the scope of this work. The area
of polygons showing debris-covered glacier within
the glacier outlines may be subtracted from the total
area of the glacier in question to obtain the area of
the active glacier. This smaller glacier area may be
more appropriate to use in some analyses of glacier
area changes than the total area including moraine or
dead-ice fields, see for example the analysis of Aðal-
geirsdóttir et al. (2020) of glacier changes in Iceland
since ∼1890 based on volume–area scaling.
JÖKULL No. 70, 2020 7