Conference PaperPDF Available

Interferometric control for mapping and quantifying the 2012 breakup of Matusevich Ice Shelf, Severnaya Zemlya

Authors:

Abstract and Figures

ERS, TDX and S1A SAR interferometric models calibrated with ICESat and CryoSat-2 altimetry data were successfully applied to mapping and quantifying recent breakup of the Matusevich Ice Shelf in Severnaya Zemlya. The assessment of the mapped changes showed that the ice shelf lost 123±2.5 km² or two thirds of its area and 9±0.3 km³ of its volume in the course of the past 5 years. The potential causes and aftereffects of the ice shelf disintegration were determined and the sensitivity of MIS remnants to climatic forcing was assessed. The mean annual temperature of-9°C proved to be a valid climatic threshold for the ice-shelf viability in the Russian Arctic. We suggest that the integral estimation of calving regime in Severnaya Zemlya previously published by other investigators is out-of-date and must be revised.
Content may be subject to copyright.
INTERFEROMETRIC CONTROL FOR MAPPING AND QUANTIFYING THE 2012
BREAKUP OF MATUSEVICH ICE SHELF, SEVERNAYA ZEMLYA
Aleksey Sharov (1), Dmitry Nikolskiy (2), Ksenia Troshko (3) and Zinaida Zaprudnova (3)
(1) DIGITAL - Joanneum Research, Steyrergasse 17, 8010 Graz, Austria, Email: aleksey.sharov@joanneum.at
(2) Sovzond, Shipilovskaya 28 A, Moscow, 115563, Russia, Email: nikolskiy@inbox.ru
(3) Faculty of Geography, Lomonosov Moscow State University, Lenin Mountains 2, 119992, Russia
ABSTRACT
ERS, TDX and S1A SAR interferometric models calibrated
with ICESat and CryoSat-2 altimetry data were successfully
applied to mapping and quantifying recent breakup of the
Matusevich Ice Shelf in Severnaya Zemlya. The assessment of
the mapped changes showed that the ice shelf lost 123±2.5
km² or two thirds of its area and 9±0.3 km³ of its volume in
the course of the past 5 years. The potential causes and
aftereffects of the ice shelf disintegration were determined and
the sensitivity of MIS remnants to climatic forcing was
assessed. The mean annual temperature of -9°C proved to be a
valid climatic threshold for the ice-shelf viability in the
Russian Arctic. We suggest that the integral estimation of
calving regime in Severnaya Zemlya previously published by
other investigators is out-of-date and must be revised.
1. INTRODUCTION
Permanent floating extensions of grounded glaciers
referred to as ice shelves belong to relatively
uncommon, rapidly vanishing and poorly documented
forms of present-day glaciation in the Arctic [1].
According to Wikipedia ice shelves are only found in
Antarctica, Greenland and Canada (2015). This
message is rather misleading, however, as it disregards
the existence of ice shelves in the Russian High Arctic.
The largest of Russian ice shelves, the Matusevich Ice
Shelf (MIS) in Severnaya Zemlya, 241 km² in size
(1953), lies south of 80°N thus being the southernmost
floating glacier in the Old World. All Canadian ice
shelves at Ellesmere Island are situated north of 82°N.
The recent series of ice shelf collapses in West
Antarctica, Greenland and Canadian Arctic raised, apart
from the discussion on their potential causes, a question
about the validity of empirical climatic thresholds
established somewhat earlier for the ice-shelf viability
in different geographic settings and epochs [2, 3, 4 and
5]. In this regard, MIS represents a unique object for
studying ice-shelf behaviour and stability of parent ice
masses in glaciomarine environments, which are
dynamically different from those in the West. MIS is
several orders smaller than ice shelves in the Antarctica
and Greenland and is, sometimes, referred to as “natural
model” of larger floating glaciers [6]. MIS is the main
producer of large tabular icebergs up to several
kilometres in size threatening ships and oil rigs in the
Laptev and Kara seas. Its breakup in 2012 exited
debates about the emergence of new icebergs and gave a
new impetus to glaciological research and remote
sensing activities in and around Severnaya Zemlya.
Joanneum Research in Graz has been conducting
detailed remote sensing studies of MIS and parent ice
caps using a synergetic combination of satellite
altimetry and interferometry as a part of the 5-year
programme of glacier change mapping in Severnaya
Zemlya carried out since 2009 [7]. The present paper
describes a new series of glacier interferometric models
and satellite image maps at 1:100,000 scale representing
the current state of the ice shelf and demonstrating
dynamic changes in its topography and rheology in the
past 80 years. The quantitative characteristics of main
calving events and associated ice loss processes derived
from space-borne ERS, TanDEM-X, CryoSat and
Sentinel-1 radar data, WorldView, LANDSAT and
ICESat optical data are validated and interpreted by
comparing with 30-year-old cartographic reference
models, air-borne radio-echo sounding data and 30-
year-long hydrometeorological and oceanographic time
series. The potential causes and collateral effects of the
ice shelf disintegration have been determined and the
sensitivity of MIS remnants to climatic forcing has been
assessed.
2. GEOGRAPHICAL SITUATION
MIS (79.8°N, 98.0°E) is a confluent ice shelf formed by
8 outlet glaciers which flow down from Rusanov Ice
Cap (956 km²), at the north, and Karpinsky Ice Cap
(2561 km²), at the south, into the 50-km-long
Matusevich Fjord situated on the northern coast of
October Revolution Island in the Severnaya Zemlya
archipelago [8]. These glacier tongues merge and build
a platform of floating ice with seemingly flat and
smooth albeit fissured surface. Gently sloping surface of
the ice platform with typical slopes of 0.2°, large depths
in the fjord, continuous propagation of internal tides
between the outer and the inner parts of the fjord,
periodic release of large tabular icebergs and the results
of multitemporal remote sensing studies and geodetic
works provide the evidence that the ice shelf is largely
afloat [6]. The floating state of MIS was accepted by all
investigators from the very beginning of its explorations
[9].
The Matusevich Fjord provides very favourable
conditions for the ice shelf viability (Fig.1, a). Owing to
its north-eastward orientation, the ice shelf is secluded
from strong insolation and is well protected from the
impact of winds and waves of the Laptev Sea. Several
small islands and banks support MIS in its central and
widest part. Water depths range from 323 m in the fjord
mouth to 258 m twelve kilometres inwards, decreasing
to 50 m between Arduous and Barrier islands and
increasing again to 128 m in front of Cape Fort, 17 km
upstream. Radio-echo soundings (RES) of 1968/69 and
1974/75 showed the potential existence of large depths
exceeding 200 m beneath the south-eastern part of MIS
[8]. The bottom profile in other parts remains uncertain.
a) b)
Figure 1. Location diagram (Sentinel mosaic, a); mean
annual temperature at Golomyanny Island (b, + in a)
The shallow region of the Laptev Sea eastwards of the
fjord mouth is characterized with perennial pack ice and
often calms, weak waves typically within 1 meter and
relatively low convection in surface waters to the depth
of only 5-10 meters. For most of the year the fjord and
adjacent straits are covered with immobile fast ice, but
even in winter, polynyas can be found at some locations
seawards of the fjord, and the zone of landfast ice is
relatively narrow. In general, the north-eastern coast of
October Revolution Island appears to be less bounded
with landfast ice than the western shore.
The regional climate influenced by continental air
masses is one of the most severe in the Arctic. The long-
term average air temperature is -14.8 °C, while the
mean annual temperature of the past decade is -11.8° C
(Fig. 1, b). Mean annual precipitation, typically in the
form of snow, is about 200 mm at sea level and 450 to
500 mm at higher altitudes. The climatic snowline
altitude is at 50 – 100 m, which is the lowest in the
Eurasian Arctic. Six large ice caps cover 58% of the
land area of October Revolution Island [8]. The mass
balance of both parent ice caps is negative. In the past
25 years Karpinsky and Rusanov ice caps lost 21 km³
and 8.5 km³ of their volumes respectively [7]. There
were no regular mass-balance measurements carried out
in situ on the MIS surface.
The average sea surface temperature (SST) along the
north-eastern coast of October Revolution Island is
about -0.8°C. The medium layer with warm water at
+1.5°С originating from the inflow of Atlantic waters
was revealed at 300 m depth. In warm summers
numerous areas of open seawater surrounding MIS are
found along ice-free coasts. The amplitude of local
semi-diurnal irregular tides measures 0.3 – 0.4 m,
although it can surpass 1 m under strong NE winds [10].
Seasonal and inter-annual variations of the sea level
recorded at the polar station on Golomyanny Island, 140
km westwards of MIS don’t exceed 0.7 m. Please refer
to the maps in Figs. 2 and 6 hereafter, for all geographic
locations, names and features.
3. COLLECTED MAPS AND DATA
Apart from remote sensing time series and stationary
records, the most consistent, thorough and condensed
factual knowledge on the MIS state and fluctuations is
compiled in the form of maps and journal publications
by Russian, British and American explorers. Yet, even
our extensive collection contains very few detailed
maps representing the entire MIS surface at large scale
with sufficient accuracy (Fig. 2).
Figure 2. Photogrammetric maps of MIS: observational
map 1:400, 000 by O.v.Gruber (1933, a); topographic
map 1:200,000 (1984, b) and RES profile (inset)
MIS was first explored and sketched by the expedition
of G.Ushakov and N.Urvantsev during two sledge
20
10
0
1999
2001
2003
2005
2007
2009
2011
2013
Mean-annual T, °C
4kmb)
a)
journeys across Matusevich Fjord in April-May and
June-July, 1931. The first photogrammetric map
showing the central part and marginal parts of the parent
ice caps at 1:25,000 scale with a contour interval of 20
m and the observational map representing the entire
MIS at 1:400,000 scale were produced by O.v.Gruber
based on materials of the strip-wise stereoscopic
photographic surveys performed from the airship Graf
Zeppelin in July 1931 (Fig. 2, a).
The entire outlines of MIS were placed on a large-scale
map after extensive air-borne photographic surveys of
the entire archipelago had been carried out by the
“Soyuzmorniiproject” Trust from 1951 to 1952. Air-
borne RES surveys of the ice shelf carried out in
1968/69 and 1973/74 allowed the ice thickness and
glacier bed topography along MIS margins to be
determined and mapped at small scales [8]. The next
aerial photographic survey of MIS was performed in
1984 and provided basic materials for the present
topographic map series of this glacier complex at scales
1:100,000 and 1:200,000 with contour intervals of 20
and 40 m (Fig. 2, b). The hydrographic chart No.11135
issued by the Russian Hydrographic Service in 1994 at
1:500,000 scale showing water depths in the Matusevich
Fjord was also at our disposal. Up-to-date topographic
maps and digital elevation models (DEMs) of the MIS
area are either non-existent or of limited quality and
coverage.
The space-borne interferometric radar data set covering
MIS area included:
two tandem ERS-1/2 SLCI pairs of 23/24.09.1995
and 22/23.03.1996. Both pairs were obtained from
ascending orbits with normal baselines of +5.5 m and
-46.2 m respectively under steady cold weather
conditions with high atmospheric pressure, zero
precipitation and spring tides;
one TanDEM-X HH-SRA SM pair of 05.05.2011
obtained from ascending orbits with perpendicular
cross-track baseline of 178.8 m (low atmospheric
pressure of 996.8 mb, solid precipitation of 4
mm/day, spring tide);
two repeat-pass Sentinel-1A Extra Wide Swath pairs
(GRD and SLC) of 09/21.10.2014 (1017/1029 mb,
zero precipitation and spring tides on both dates) and
06/18.12.2014 (1010/1024 mb, zero precipitation and
neap tides on both dates) taken from ascending orbits
with B = -70.9 m and single horizontal polarization;
CryoSat-SIRAL SARIn interferometric altimetry
data of 2010 -2012.
The space-borne optical data set included:
14 QuickBird and WorldView-1/2 quicklook images
of 2009-2014;
8 LANDSAT images of 1973-2014;
ICESat GLA06 lidar altimetry data (release 33) of
2003-2009.
30-year long records of hydrometeorological and
oceanographic data obtained from 3 coastal stations
(Golomyanny, Fedorova, and Vize) around MIS as well
as ancillary glaciological and oceanographic
publications were involved in the analysis of main
causes and consequences of the MIS breakup.
3. ICE SHELF MORPHOLOGY
The entire ice catchment area of Matusevich Fjord was
estimated at approx. 1100 km² with 58% of the area
belonging to Karpinsky Ice Cap. In [8] the total areas of
outlet basins feeding MIS are given as 93.2 km² and
179.5 km² for Rusanov and Karpinsky ice caps
respectively. RES surveys revealed that the north-
western periphery of Karpinsky Ice Cap rests on a bed
located 150 m to 350 m below sea level and the
tidewater glacier front is about 300 m thick [8]. The
submerged part of the glacier bed is about 12 km wide
(Fig. 2, b, inset). The outer south-eastern part of glacier
bed of Rusanov Ice Cap is relatively flat and lies
approximately at the sea level.
The very central part of MIS is surrounded with steep
coasts descending east- and westwards and outlet
glaciers flowing from Rusanov Ice Cap gradually merge
without sharp distinction into the shelf ice, so that it is
difficult to detect the ice shelf margin from the outside.
Also at the right coast, the marked break in slope is not
always detectable and precise delineation of the ice
shelf area represents a challenging task. Most explorers
draw inner shelf boundaries along the grounding line
(point A in Fig. 3, a), where the shelf ice leaves land
and begins to float. The projection of the grounding line
on the ice shelf surface is above the geoid.
Figure 3. Inferred longitudinal profile (not to scale, a)
and morphological scheme of MIS, inner part (b)
Both, the grounding line (A) and the coastline (C)
hidden by ice masses flowing down into the fjord can be
a)
b)
detected in radar interferometry data. Depending on the
ice thickness T0 and the glacier bed slope , well-
expressed hollows (D) can be seen rather often, but not
necessarily, close to the boundary of the transitional
zone (B). We observed such depressions in front of
Khodov (No.21), AARI (No.48) and Researcher’s
(No.49) outlet glaciers. In general, the longitudinal
profile of the ice shelf derived from hydrostatic
equilibrium conditions and proved by our explorations
in Franz Josef Land looks as shown in Fig. 3, a. Glacier
numbers are given according to [8].
The MIS highest elevations range from 14 meters at the
MIS front to 17 - 19 meters in the fjord centre and attain
21 meters in the innermost part, Red Bay (1984). Hence,
the MIS maximum ice thickness can be estimated at 160
- 200 meters assuming hydrostatic equilibrium and
using the buoyancy ratio of 0.105. For the year 1984 the
total MIS area was given as 217.2 km² and the ice shelf
volume was assessed at approx. 19 km³.
On closer examination the ice shelf morphology proves
to be complex (Fig. 3, b, adopted from [6]). MIS
elevations vary irregularly from several meters to
several tens of meters at spatial scales of lesser than 1
km. The ice thickness distribution and the ice flow
pattern are also very heterogeneous. Floating glacier
tongues coming in contact from opposite directions or
striking against islands and coastal rocks create a
succession of elongated ice rises and ridges rising up to
15-20 meters, which are well observed even in winter
images. In summer, numerous shallow meltwater lakes
and channels occur on the MIS surface, and depressions
between ice bulges are typically filled with meltwater or
rock debris. The same holds for large tabular icebergs.
The total width of all outlets terminating in MIS
measures 20 kilometres which is twice longer than the
outer-fjord width of 8 to 10 kilometres. Numerous
longitudinal fissures and crevasses on the surface of
MIS and most of its tributaries indicate the compressive
character of glacier ice flow in the fjord and especially
at its flanks [9].
Considering the complexity of MIS morphology and
variable dynamics we divided the ice shelf into three
sections (Fig. 6), c.f. [4]. The inner part of MIS fed by
outlet glaciers Nos. 21, 22 and 23 from the left and by
outlet glaciers Nos. 46, 47 and 48 coming from the right
is the most stable section of the ice shelf. This area
covering approx. 80 km² is referred to as MIS-A zone.
The outer part named MIS-B is built by the ice from
Researcher’s Outlet Glacier (No.49) and is the most
variable. The MIS-B section consists of three lobes,
shaped like a shuttlecock, and transitional zone or
“crown” between the ice shelf and the outlet glacier.
The transitional zone of MIS-B with the area of about
25 km² was first put on a map as the south-easternmost
floating part of MIS by E.Zinger and V.Koryakin in
1965. Yet, it was not treated as a part of MIS in other
cartographic publications. The third zone referred to as
MIS-C covers the transitional zone between the ice shelf
and Avsyuk Outlet Glacier (No.20) at the left coast of
Matusevich Shelf. This narrow section with the area of
approx. 4 km² was represented as a part of MIS on
Russian topographic maps and in [8], but was not
considered as such in [4, 11].
5. ICE SHELF ELEVATION MODELS
The first cartographic MIS elevation model (DEM0)
with 50-m posting was derived from the Russian
topographic map 1:200,000 (CI = 40 m) showing the
glacier state as surveyed in 1984. DEM0 represents
parent ice masses and ice-free areas with sufficient
accuracy and detailedness. The rms vertical accuracy of
DEM0 is typically given as one fifth to one third of the
contour interval on glaciers, i.e. 8 to 13 m rms. The
information contents of available topographic maps
showing, at best, only one contour line on the entire
MIS surface and only one spot height per 10 km² of the
ice shelf area are obviously insufficient for representing
the undulating character of the ice shelf surface. The
direct use of the overall cartographic elevation model
for glacier change detection and ice loss estimates is
still limited, since it doesn’t represent inner shelf
borders and other structural elements, e.g. outlet glacier
basins needed for glaciological measurements.
The mid-term interferometric elevation model (DEM1)
representing the MIS surface state of the 2000s was
derived from the pair of differential ERS-1/2 SAR
interferograms overlaid with ICESat GLA06 altimetry
data (Release 33). Precise delineation of MIS area was
manually performed using high-resolution optical
imagery and the best interferometric coherence image of
23/24.09.1995 with a mean coherence value of 0,7 (Fig.
4, a). In complex cases, e.g. in the south-eastern
confluent area, we additionally involved the second
coherence image [c.f.12]. The representation of ice shelf
elevations was essentially improved along and in the
close vicinity of 17 altimetric transects crossing the MIS
area with a track spacing of about 4 km. The impact of
long-term snow accumulation was neglected due to high
rates of melting on the MIS surface. Yet, MIS
elevations in between altimetric transects are error-
prone due to data age difference and pronounced tidal
effects on the DINSAR phase, especially in the central
and eastern part of MIS. The rms vertical error of DEM1
was given as ± 4 m [7].
The present-day elevation model (DEM2) with 25-m
posting applicable to the computer analysis of ice shelf
characteristics and volume changes was generated from
the TanDEM-X SAR interferometry data of 05.05.2011
and controlled using the concurrent CryoSat-2
interferometric radar and ICESat lidar altimetry data
(Fig. 4, b). DEM2 was further oriented by means of
ground control points and was levelled by referencing to
the landfast sea ice assuming the freeboard value of 0.25
m. MIS is largely in the state of hydrostatic equilibrium
and tides are relatively low. Hence the sea level
recorded at the time of satellite surveys can be used as a
datum plane for practical modelling of the glacier
complex. The precision of elevation data checked on the
northern fjord coast and at several ice-free islands with
known heights was characterized by a rms vertical error
of ± 2.7 m. Topographic contour lines on the MIS
surface were drawn at 10 meter intervals with the
additional contour depicted at the elevation of 5 m (Fig.
4, b). Ice shelf areas with relatively thin ice were
delineated in semi-automatic mode.
Figure 4. ERS-1/2 coherence image of 23/24.09.1995
(a); interferometric TDX DEM2 of 05.05.2011(b)
The comparison with the cartographic DEM0 showed
that the TDX elevation model is superior in representing
top heights and lows, depicting ice surface undulations
and compact glaciological features, such as ice ridges,
depressions, crevasses etc., yet without compromising
on positional accuracy. A dozen new islands
underpinning the ice shelf were first discovered in
DEM2 and, afterwards, verified in high-resolution
optical images. In DEM2 we detected six elevated and
crevassed ice shelf areas alternating with lower areas of
thin even ice, apparently of marine origin. MIS top
heights of up to 32 m asl are typically observed in front
of tributary glaciers that feed into the ice shelf, while
extensive areas of thinner ice with medial heights of
several meters are to be found along ice-free coasts. The
average elevation of the MIS sections was measured in
DEM2 as 10.5 m (MIS-A) and 13.8 m (MIS-B), and the
average ice thicknesses were given as 100 m and 131 m
respectively. The expected maximum ice thickness of
MIS was given as 220 m and the thickness of the
thinnest parts was estimated at 25 m. This observation
sheds some light on spatiotemporal variations of the
MIS iceberg production capacity.
7. MAPPING ICE SHELF DIMENSIONS AND
FLUCTUATIONS
DEM2 was successfully applied to geocoding
intermediate interferometric models, ortho-rectifying
multitemporal optical and radar images from
LANDSAT, WorldView, QuickBird and Sentinel
satellites and building image time series. The image
time series with the equalized pixel size of 50 m dating
back to 1973 were attached with the vector layers
derived from available maps representing the ice shelf
boundaries in 1931, 1952 and 1984. Separate sections of
the image time series were used for measuring ice shelf
dimensions, quantifying ice loss processes, locating and
dating main surging and calving events, and iceberg
monitoring in the MIS area. Afterwards, they served as
basic layers for the output map series and animations of
the MIS evolution in 1931-2014 at 1:100,000 scale
(UTM 46N, WGS84), which were included into the
“Online atlas of glacier fluctuations” accessible at
http://dib.joanneum.at/MAIRES/index.php?page=atlas.
A small-size copy of one of maps demonstrating MIS
disintegration is represented in Fig. 6.
The cartometric analysis of the resultant maps showed
that the largest dimensions (256 km²) of the ice shelf
were recorded at the very beginning of its explorations
in 1931. Since then, MIS experienced several retreats in
the 1950s and in the 1980s and advances in 1973 and in
1995 mentioned in [4]. Yet, the ice shelf area was never
larger than that in 1931. After the last advance in 1995,
the ice shelf retreated gradually. The essential
acceleration of MIS disintegration was observed in the
2000s. The present MIS disintegration began no later
than 2009 when the first large portion of the outer
margin was lost. Finally, the ice shelf retreated from the
shoal between Arduous and Barrier islands, split up into
three unequal parts and lost two thirds of its original
area of 1931 by late summer 2012. The ice shelf
collapse continued in 2013-2014.
MIS event icebergs were mostly observed in 2010 and
2012 with surprisingly few large icebergs found in
summers 2013 and 2014. In April 2010 we counted 14
large tabular icebergs of rhomboidal form wintering in
the fjord. The largest iceberg was 2.77 km long and 1.42
km wide. Seven large tabular icebergs from the 2012
breakup were observed in the fjord in the fall of 2012
and in spring 2013. The largest of them were 3.3 km x
1.0 km and 3.5 km x 1.1 km in size, the latter was
detected the inner fjord. The size of the largest iceberg
sighted at the eastern coast of Severnaya Zemlya in
a)
b)
August 2014 was 668 m x 336 m. Our observations
showed that, in the past decade, the time between
calving events varied from 1 to 2-3 years and the typical
size of tabular icebergs might indicate the possible rate
of ice flow at the glacier front. Hence, we supposed that
the velocity of Researcher’s Glacier exceeded 200 m/a.
The analysis of DINSAR and altimetry data showed
that, in the period of 1984-2012, the MIS thickness
decreased gradually in central and inner parts, and the
ice surface roughness increased drastically over the past
years. The maximum thinning up to 37 m was
discovered in front of Polyarnikov Glacier with frontal
velocity of less than 20 m/a and between Khodov and
Zhuravlev glacier tongues. According to DEM2, the
present elevation of the ice shelf surface in both areas of
extreme thinning does not exceed 2 and 3 meters
respectively. The present minimal ice thickness of MIS
was thus estimated at 25 meters.
The ice shelf dimensions in 1931-2014 are specified in
Table 1. Large tabular icebergs attached, frozen or
grounded close to the ice shelf were considered as a part
of MIS. The total amount of ice loss from MIS due to
calving, surface ablation and basal melting in the period
of 1984-2012 is estimated at 13±0.4 km³. The advance-
retreat cycles in the MIS observation history seem to be
less regular than it was mentioned in [4, 13]. The
present total area of MIS including MIS-A, MIS-B and
MIS-C sections with “newly” detected floating area
(“crown”) in the south-eastern part of MIS-B was
measured as 86 km² (IX, 2014) and the remaining ice
shelf volume was given as 6.4 km³, which is at least
three times smaller than it was after the previous
breakup in 1984-85. For the sake of data conformity the
values given in Tab. 1 don’t account for the presumably
floating part (25 km²) between the ice front of MIS-B
and the ice fault on Researcher’s Outlet Glacier, 5 km
upstream. Currently, the concave front-line of the MIS-
B section runs between the glacier contour lines 40 m
and 80 m represented in the Russian topographic map
with the glacier state of 1984. The present height of the
MIS-B front ranges from 11 m in the east to 24 m in the
west. The MIS-C section has nearly ceased to exist. Yet,
several small icebergs were observed quite recently
close to its southernmost front.
Our maps and quantitative estimates were verified and
approved during the “Kara-Summer-2014” expedition
to the Severnaya Zemlya archipelago on board the
research vessel “Academician Treshnikov” under the
leadership of A.V.Nesterov. Aerovisual observations
and ice-sounding radar surveys of MIS remnants from a
helicopter were carried out in August, 2014 and
provided up-to-date ice thickness values ranging from
30 to 200 meters. The maximum ice thickness of 208 m
was measured at the south-eastern MIS margin. In Red
Bay, the maximal ice thickness of 155 m was recorded
at the front of Zhuravlev Glacier. The ice fronts were
found in the same position as they were represented in
our maps. The average ice thickness of MIS remnants
determined along reliable profiles was given as 75 m.
The value seems to have been underestimated and we
suggest that it is not representative of the entire MIS.
Due to the sparse coverage of the area by flight tracks,
the absence of bottom returns over heavily crevassed
areas, unstable weather and inaccurate referencing, the
vertical accuracy of surveys was not better than 10 m.
8. ICE MOTION
There are very few instrumental records documenting
the rate of ice flow in the MIS region. Frontal velocities
of several northern tributaries were measured in the
field during the ablation season of 1965. The resultant
velocity diagrams with maximum velocity values
ranging from 50 m/a (Zhuravlev Glacier) to 110 m/a
(Khodov Glacier) can be found in Fig. 3, b). Although
some relevant surveys using EO data are currently
underway [13], the factual knowledge on ice flow
velocities at the right coast of Matusevich Fjord remains
extremely scarce.
Frontal velocities of 7 outlet glaciers flowing into the
Matusevich Fjord were recently measured with the aid
of satellite differential interferometry using ERS-1/2
INSAR data and the reference DEM2 (Fig. 5, a). It was
reported that at least four outlets conveying ice to MIS
from Rusanov Ice Cap, namely Khodov (No.21),
Zhuravlev (No. 22), Esenin (No.23) and Researcher’s
(No.49) glaciers are fast flowing with frontal velocities
exceeding 0.7 m/day or 170 m/a. This measurement
indicates the essential acceleration of ice flow in the
MIS area in the period of 1965-1995, but cannot be
related directly to the breakup event of 2012, however.
Further it was recognized that other small tributary
glaciers are slow moving with frontal daily velocities
ranging from 9 to 14 cm/day. Positive elevation changes
up to +29 m were detected in the accumulation zones of
Fairytale (No.46) and Polyarnikov (No.47) outlet
glaciers at the south-eastern coast of Fairytale Bay. Both
glaciers are in a quiescent phase.
Table 1. Ice shelf dimensions in 1931-2014
Year 1931 1952 1973 1984 1995 2009 2012 2014
Area, km² 256 241.1 245 217.2 242 184 71.7 61.1
Length, km 35.5 36 37 33 35 32.9 24.5 23.7
Volume, km³ - 22 - 19 - 15 - 6
Source [6, 9] [8, 11] [4, 7] [7] [7] [7] Present Present
Figure 5. MIS: ERS-1/2 DINSAR model of 23/24.09.1995
(a) and S1A EW3 fringe image of 06/18.12.2014 (b)
The replacement of DEM0 with DEM2 treated in this
paper improved the accuracy of DINSAR measurements
and showed that the daily velocity of ice flow in the
middle part of Researcher’s Glacier at a distance of 5
km from the glacier front was 0.5 m/day or 120 m/a.
The velocity of glacier flow increases notably
downstream and the maximum velocity value close to
the glacier front is supposed to be about 200 m/a. This
glacier demonstrates steadily extending character of ice
flow with predominantly transverse crevasse pattern.
Besides, we recognized that the character of ice flow on
Zhuravlev, Esenin and AARI glaciers changed from
compressive to extending and the crevasse pattern
changed correspondingly from longitudinal to transverse.
The interferometric measurement of ice flow velocities
with the aid of Sentinel-1 SAR data was limited by the
12-day repeat interval of satellite surveys, relatively low
coherence and significant phase noise on radar
interferograms of fast flowing tributaries (Fig. 5, b). We
detected several patches with good visibility of S1
interferometric fringes in the inner part of MIS and
recognized that these spots coincided with the areas of
thinner shelf ice between glacier tongues. We suggest
that the ice shelf undergoes tide-induced vertical
oscillations in some parts with thinner ice as a plausible
explanation for such finding. The S1 INSAR data of
06/18.12.2014 shows an extensive area of fast sea ice
attached to the fronts of MIS-A, MIS-B and MIS-C
sections. The frontal velocity of the eastern lobe of
Researcher’s Glacier was estimated at approx. 100 m/a
by analysing the fast-sea-ice motion induced by glacier
flow. The maximum glacier velocity at the centre of the
glacier front was determined as being close to 200 m/a
under the assumption that it is twice of that at the flank.
The analysis of concentric tide-related fringes proved
the existence of a shoal (50 m bsl) located half way
between Arduous and Barrier islands.
9. CAUSAL ANALYSIS AND CONCLUSIONS
In 2009, MIS front retreated from the shallow water
area between Arduous and Barrier islands and the ice
shelf disintegrated rapidly during the next 5 years.
Rapid climatic warming in the High Arctic is the most
obvious reason for the MIS disintegration in 2009 –
2014. A general negative balance of parent ice masses
was related to the climatic changes too [7]. The 2012
breakup of MIS is the largest glacier calving ever
reported from the Russian High Arctic. The annual rate
of ice loss from MIS given as 0.43 km³/a has yet to be
accounted for in the long-term mass balance research in
Severnaya Zemlya. We suggest that the rate of total ice
loss due to calving and basal melt in Severnaya Zemlya
previously specified in [7, 8] is out-of-date and must be
revised.
The main causes of the MIS breakup are summarized as
follows. The leap-year 2012 was the warmest and the
wettest in the history of meteorological observations at
the Golomyanny station. The mean annual temperature
of -9.3°C recorded in 2012 reached practically the
threshold of -9°C specified as climatic limit for the ice-
shelf viability in [5]. In 2012 the annual precipitation
amount in Severnaya Zemlya reached the record value
of 353 mm. The number of days with rain (61) was
twice larger than the long-term average value. Rains
with the intensity of up to 25-27 mm/day destroyed the
snow cover on the MIS surface in June-September and,
together with extensive meltwater on the ice-shelf
surface, contributed essentially to fracturing processes.
There were a series of severe storms with wind gusts up
to 104 km/hr and surges recorded at Golomyanny and
Fedorov meteorological stations in April-July 2012.
Very similar climatic records were registered at the Vize
station 400 km westwards of MIS.
In past years, SSTs in the northern Laptev Sea in
August were several degrees warmer than the 1982-
2010 August mean [14]. The inflow of warm surficial
and medium-layer waters into the Matusevich Fjord and
enhanced basal melting of the ice shelf is believed to be
one of the main causes for the ice shelf disintegration
and iceberg removal from the fjord. The high rate of
MIS thinning given as 1.3 m/a can be directly associated
with basal melting of MIS. The increase in pre-event
velocities of MIS tributaries, the transition to extending
a)
b)
ice flow, and structural weakening of the ice shelf might
impact its reaction to climatic forcing in the 2010s. We
suggest that the buttressing resistance of the ice shelf
decreases with the ice thickness well before the removal
of shelf ice from the fjord.
It was decided that the unfavourable combination of
long-term atmospheric and oceanic warming, heavy
precipitation in liquid form and strong winds in summer
as well as essentially negative mass balance of both
parent ice caps, structural weakening of the ice shelf
and, probably, low concentrations of sea ice along the
eastern coast of Severnaya Zemlya was the main driving
factor for the MIS event in 2012. Based on the seismic
history of the Laptev Sea, we excluded nonrecurring
one-time impacts, such as earthquakes and tsunami from
the causal analysis of the breakup. We suppose that,
under current environmental conditions, MIS-A and
MIS-B sections will vanish within the next decade.
Several areas with thin ice in the MIS-A section will
disappear first, followed by floating tongues of Esenin,
Zhuravlev and Khodov glaciers. The potential advance
of several smaller tributary glaciers having a positive
mass balance can decelerate MIS recession, but cannot
stop it. The Researcher’s Glacier will continue calving
and will likely undergo further retreat which will stop
when the glacier front reaches a new stable position, 7
to 8 km upstream from the present location.
Thanks to the wide terrestrial coverage of Sentinel-1
EWS data, we were able to compare the rate of MIS
disintegration with the concurrent ice loss rates from
two smaller ice shelves in Severnaya Zemlya, No.19 at
the eastern margin of Academy of Sciences Ice Cap and
No.70 at the southern margin of Karpinsky Ice Cap
situated 70 km to the north and 60 km to the south of
MIS respectively. The areas of these ice shelves were
measured in the 1950s as 5.7 km² and 11.1 km²
respectively [8]. To our surprise, we discovered that the
ice shelf No. 19 advanced and its area increased by 2.2
km² (2014). We explained this finding by the local
dynamics of the Academy of Sciences Ice Cap, colder
climate and the temporary excess of ice masses in the
upper part of this glacier mentioned in [7]. The ice shelf
No.70 built by two confluent outlet glaciers Nos. 66 and
67 on Karpinsky Ice Cap calves in fresh-water Fjord
Lake with the bathymetric mark of 98 m close to the
outer shelf margin with an elevation of 6 m (1984).
There is some doubt on the floating state of this glacier.
The glacier area measured in the WV02 optical image
of 2013 does not exceed 9 km², which means a relative
decrease of 20 % in approx. 30 years.
Finally, we conclude that the rapid collapse of MIS is a
new sign of climate change in the Eurasian High Arctic
and is consistent with the observations made recently in
the Canadian Arctic and Greenland, where large ice
shelves demonstrate similar behaviour. The mean
annual temperature of -9°C proved to be a valid climatic
threshold for the ice-shelf viability in the High Russian
Arctic. We expect an essential improvement of glacier
interferometric models with the advance of Sentinel-1B
satellite and anticipate further use of our products by
Russian colleagues for operational iceberg monitoring
in the Kara-Laptev region.
ACKNOWLEDGEMENTS
The present study was funded from the EuRuCAS FP7
INCO project. ERS, CryoSat-2 and TanDEM-X satellite
data were provided by ESA and DLR through AOP.6327
GEMINI and AOP.GLAC249 GEODESIA. S-1A SAR
EW data were obtained from the Sentinel-1 Scientific Data
Hub. ICESat altimetry data were made available by
NSIDC. Hydrometeorological data were downloaded from
the “TuTiempo” online archive of historical weather.
REFERENCES
1. Dowdeswell, J. et al. (1994). Evidence for floating ice
shelves in Franz Josef Land, Russian High Arctic. AAR.
26 (1), 86-92.
2. Münchov, A. et al. (2014). Interannual changes of the
floating ice shelf of Petermann Gletscher, North
Greenland, from 2000 to 2012. J. Glac. 60(221), 13J135.
3. Copland, L. et al. (2007). Rapid loss of the Ayles Ice
Shelf, Ellesmere Island, Canada. GRL, 34 (L21501),
doi: 10.1029/2007GL031809.
4. Williams, M. & Dowdeswell, J. (2001). Historical
fluctuations of the Matusevich Ice Shelf, Severnaya
Zemlya, Russian High Arctic. AAAR, 33 (2), 211-222.
5. Morris, E. & Vaughan, D. (2003). Spatial and temporal
variation of surface temperature on the Antarctic
Peninsula and the limit of viability of ice shelves.
Antarctic research series, Washington, AGU, 61-68.
6. Govorukha, L.S. (1989). Modern glaciation of the Soviet
Arctic. Leningrad, Gidrometeoizdat, 256 p.
7. Sharov, A. & Tyukavina, A. (2009). Mapping and
interpreting glacier changes in Severnaya Zemlya with
the aid of differential interferometry and altimetry. ESA
SP-677, 8 p.
8. Vinogradov, O.N. (1980). Catalogue of glaciers on
Severnaya Zemlya. Hydrometeoizdat, L., 79 p. (in Russ.)
9. Gruber, O. (1933). Über die Ausrüstung des “Graf
Zeppelin“, die Auswertungsmethoden und die bisherigen
Ergebnisse aus dem gewonnenen Aufnahmematerial.
Petermanns Mitteilungen. Gotha, 216, 68-77.
10. Gorshkov S.G. (1980). Atlas of the Arctic Ocean.
Leningrad, GUNO, 188 p.
11. Koryakin, V.S. (2014). What is happening to
Severnaya Zemlya’s glaciers? Nature, 11, 42-49 (Russ).
12. Rignot, E. et al. (2011). Antarctic grounding line
mapping from differential radar interferometry. GRL, 38
(L10504), doi: 10.1029/2011GL047109.
13. Willis, M.J. et al. (2013). Outlet glacier thickness and
velocity changes in response to the destruction of the
Matusevich Ice Shelf. AGU Fall Meeting, C13D-07.
14. Timmermans, M. & Proshutinsky, A. (2014). Arctic
Ocean surface temperature. Arctic Report Card: Update
for 2014. DOC, NOAA, 3 p.
Figure 6. Satellite image map 1:100,000 of MIS evhandleolution in 1931 - 2014
Full-text available
Article
Te methods of satellite monitoring of dangerous ice formations, namely icebergs in the Arctic seas, representing a threat to the safety of navigation and economic activity on the Arctic shelf are considered. Te main objective of the research is to develop methods for detecting icebergs using satellite radar data and high space resolution images in the visible spectral range. Te developed method of iceberg detection is based on statistical criteria for fnding gradient zones in the analysis of two-dimensional felds of satellite images. Te algorithms of the iceberg detection, the procedure of the false target identifcation, and determination the horizontal dimensions of the icebergs and their location are described. Examples of iceberg detection using satellite information with high space resolution obtained from Sentinel-1 and Landsat-8 satellites are given. To assess the iceberg threat, we propose to use a model of their drif, one of the input parameters of which is the size of the detected objects. Tree possible situations of observation of icebergs are identifed, namely, the «status» state of objects: icebergs on open water; icebergs in drifing ice; and icebergs in the fast ice. At the same time, in each of these situations, the iceberg can be grounded, that prevents its moving. Specifc features of the iceberg monitoring at various «status» states of them are considered. Te «status» state of the iceberg is also taken into account when assessing the degree of danger of the detected object. Te use of iceberg detection techniques based on satellite radar data and visible range images is illustrated by results of monitoring the coastal areas of the Severnaya Zemlya archipelago. Te approaches proposed to detect icebergs from satellite data allow improving the quality and efciency of service for a wide number of users with ensuring the efciency and safety of Arctic navigation and activities on the Arctic shelf.
Full-text available
Article
Glaciological investigations on the Severnaya Zemlya archipelago were resumed in 2013 when a new research station «Ice base Cape Baranova» had been organized by Arctic and Antarctic Research Institute in the North-West of the Island Bolshevik. In 2014–2015, the glaciological polygon named after Leonid Govorukha was established on glaciers Mushketov and Semenov-Tyan-Shanskiy. Two years of observations on the glaciers allowed us to estimate the mass balance of the Mushketov Glacier, which was positive in the 2013–2015. By the end of the melting periods, a superimposed ice was formed on the glacier with thickness of 4 cm in 2014 and 17 cm in 2015, on the average. A snow-firn mass with its vertical thickness exceeding 3 m had been found on the upper part of the Semenov-Tyan-Shansky Glacier. Based on analyses of summer air temperatures and precipitation at the meteorological station «The Golomyanny Island», we assumed that in 2013–2015 the mass balance was also positive on the other glaciers of the archipelago, located to the North of the studied glaciers on the Island of Bolshevik. Data of remote sensing of the catastrophic advancing of the outlet glacier from the Vavilov Ice Cap, obtained in 2013–2016, testify that for much longer period, i.e. during 25 years, conditions for the ice mass accumulation were favorable on the southern and eastern slopes of the Vavilov Ice Cap.
Full-text available
Article
The combination of satellite differential radar interferometry (DINSAR) and altimetry was successfully applied to the overall geometric modelling of glacier elevation changes in Severnaya Zemlya in the period from the 1980s to the 2000s. The 2-pass DINSAR orthomosaic composed of 12 ascending ERS-1/2 tandem SAR interferograms covering the entire archipelago was calibrated and de-ramped using the change signal obtained by differencing ICESat altimetric profiles taken in the 2000s and the reference elevation model representing the glacier state as of the 1980s. The glacier-wide change signal was derived from the calibrated differential phase and mapped at 1:500,000 scale. An integral assessment of glacier changes was carried out. High spatial correlation (> +0.91) between glacier accumulation, sea ice concentration and the magnitude of the geopotential was determined and the spatial asymmetry in the distribution of glacier changes and ice flow pattern was explained.
Full-text available
Article
Petermann Gletscher, NW Greenland, drains 4% of the Greenland Ice Sheet into Nares Strait. Its floating ice shelf retreated from 81 km to 46 km length during two large calving events in 2010 and 2012. We document changes in the three-dimensional ice shelf structure from 2000 to 2012 using repeated tracks of airborne laser altimetry and ice radio-echo sounding, ICESat laser altimetry, and MODIS visible imagery. The recent ice-shelf velocity, measured by tracking surface features between flights in 2010 and 2011, is ∼1.25 km a −1 , about 15-30% faster than estimates made before 2010. The steady-state along-flow ice divergence represents 6.3 Gt a −1 mass loss through basal melting (∼5 Gt a −1) and surface melting and sublimation (∼1.0 Gt a −1). Airborne laser altimeter data reveal thinning both along a thin central channel and on the thicker ambient ice shelf. From 2007 to 2010 the ice shelf thinned by ∼5 m a −1 , which represents a non-steady mass loss of about 4.1 Gt a −1 . We suggest that thinning in the basal channels structurally weakened the ice shelf and may have played a role in the recent calving events.
Full-text available
Article
On August 13, 2005, almost the entire Ayles Ice Shelf (87.1 km2) calved off within an hour and created a new 66.4 km2 ice island in the Arctic Ocean. This loss of one of the six remaining Ellesmere Island ice shelves reduced their overall area by ~7.5%. The ice shelf was likely weakened prior to calving by a long-term negative mass balance related to an increase in mean annual temperatures over the past 50+ years. The weakened ice shelf then calved during the warmest summer on record in a period of high winds, record low sea ice conditions and the loss of a semi-permanent landfast sea ice fringe. Climate reanalysis suggests that a threshold of >200 positive degree days year-1 is important in determining when ice shelf calving events occur on N. Ellesmere Island.
Full-text available
Chapter
Mapping surface air temperature in the Antarctic Peninsula region is made unusually difficult by: the scarcity of meteorological stations, strong climatic gra­ dients and recent rapid regional warming. We have compiled a database of 534 mean annual temperatures derived from measurements of snow temperature at around 10-m depth and air temperature measured at meteorological stations and automatic weather stations. These annual temperatures were corrected for inter­ annual variability using a composite record from six stations across the region. The corrected temperatures were then analysed using multiple linear regression to yield altitudinal and temporal lapse rates. A subset of 508 values were then used to produce a map of temperature reduced to sea level and for a specific epoch (2000 A.D.). The map shows the dramatic climate contrast (3-S°C) between the east and west coast of the Antarctic Peninsula in greater detail than earlier studies and also indicates that the present limit of ice shelves closely follows the-9°C (2000 A.D.) isotherm. Furthermore, the limit of ice shelves known to have retreat­ ed during the last 100 years is bounded by the-9°C and-S oC (2000 A.D.) isotherms, suggesting that the retreat of ice shelves in the Antarctic Peninsula region is consistent with a walllling of around-4°C.
Full-text available
Article
Examination ofdigital Landsat TM and MSS imagery of Franz Josef Land, Russian High Arctic. reveals a number of ice caps with apparently very low surface gra- dients at their seaward margins. The largest of these low gradient areas is 45 km'. The areas are dynamically a part of the parent ice mass, and have a marked break of slope at their inner margins. They generally occur in protected embayments and often have relatively deep water offshore. The presence of deep inter-island channels (up to 600 m) in the archipelago also suggests that deglaciation after the last glaciation may have proceeded rapidly due to enhanced iceberg calving. Tab- ular icebergs (maximum observed length 2.3 km) are produced from several of the low gradient ice cap margins today. Ice surface profiles, derived from analysis of vertical aerial photographs, show slopes of 0.5" on these features, as compared with 3.5 to 5" on other ice caps. At least some are likely to be floating ice shelves. They have similar ice surface gradients to a known ice shelf on Severnaya Zemlya. There is no requirement for deep water to occur beneath these features, but simply that they become buoyant over a significant part of their base. Glacier thinning, due to reduced mass balance since the termination of the Little Ice Age, may have contributed to the presence of these features. An origin for some of these low gradient margins by deformation of an unlithified substrate cannot be ruled out. Field radio-echo experiments could be used to test the interpretation of these features as ice shelves.
Full-text available
Article
The delineation of an ice sheet grounding line, i.e., the transition boundary where ice detaches from the bed and becomes afloat in the ocean, is critical to ice sheet mass budget calculations, numerical modeling of ice sheet dynamics, ice-ocean interactions, oceanic tides, and subglacial environments. Here, we present 15 years of comprehensive, high-resolution mapping of grounding lines in Antarctica using differential satellite synthetic-aperture radar interferometry (DInSAR) data from the Earth Remote Sensing Satellites 1–2 (ERS-1/2), RADARSAT-1 and 2, and the Advanced Land Observing System (ALOS) PALSAR for years 1994 to 2009. DInSAR directly measures the vertical motion of floating ice shelves in response to tidal oceanic forcing with millimeter precision, at a sample spacing better than 50 m, simultaneously over areas several 100 km wide; in contrast with earlier methods that detect abrupt changes in surface slope in satellite visible imagery or altimetry data. On stagnant and slow-moving areas, we find that breaks in surface slope are reliable indicators of grounding lines; but on most fast-moving glaciers and ice streams, our DInSAR results reveal that prior mappings have positioning errors ranging from a few km to over 100 km. A better agreement is found with ICESat's data, also based on measurements of vertical motion, but with a detection noise one order of magnitude larger than with DInSAR. Overall, the DInSAR mapping of Antarctic grounding lines completely redefines the coastline of Antarctica.
Article
Dramatic retreat of ice shelves in Antarctica in recent years, linked to climatic warming, is well documented. In contrast, the ice shelves of the Russian Arctic remain largely unstudied. A time-series analysis of the largest ice shelf in the Russian High Arctic, the Matusevich Ice Shelf, Severnaya Zemlya, was undertaken for the period 1931 to 1994 using georeferenced Landsat satellite imagery and published maps. The positions of three major ice margins in 1931, 1955, 1962, 1973, 1985, 1988, and 1994 are compared. The floating margin of the ice shelf underwent at least two cycles of retreat followed by periods of advance between 1931 and 1994. These periodic calving events produce tabular icebergs up to several kilometers in length. This process is typical of floating ice shelves in Antarctica and Greenland, whereas grounded ice margins in, for example, Svalbard, produce smaller icebergs much more frequently. There is little evidence that these calving events are related to climate change. Landsat imagery is also used to track the movement of 50 icebergs identified in 1985 imagery of Matusevich Fjord. Iceberg release from the fjord between 1985 and 1994 was extremely slow, with 48 of the icebergs observed in 1985 still trapped in the fjord in 1994. The icebergs from Matusevich Ice Shelf remain in the fjord for many years, probably due to either grounding on submarine moraines or trapping by shore-fast sea ice. Much of the sediment load of the trapped icebergs may be melted out and deposited beneath the sea-ice cover of Matusevich Fjord, and little iceberg-rafted debris of heterogeneous grain size will be transported to the Laptev Sea.
Article
Dramatic retreat of ice shelves in Antarctica in recent years, linked to climatic warming, is well documented. In contrast, the ice shelves of the Russian Arctic remain largely unstudied. A time-series analysis of the largest ice shelf in the Russian High Arctic, the Matusevich Ice Shelf, Severnaya Zemlya, was undertaken for the period 1931 to 1994 using georeferenced Landsat satellite imagery and published maps. The positions of three major ice margins in 1931, 1955, 1962, 1973, 1985, 1988, and 1994 are compared. The floating margin of the ice shelf underwent at least two cycles of retreat followed by periods of advance between 1931 and 1994. These periodic calving events produce tabular icebergs up to several kilometers in length. This process is typical of floating ice shelves in Antarctica and Greenland, whereas grounded ice margins in, for example, Svalbard, produce smaller icebergs much more frequently. There is little evidence that these calving events are related to climate change. Landsat imagery is also used to track the movement of 50 icebergs identified in 1985 imagery of Matusevich Fjord. Iceberg release from the fjord between 1985 and 1994 was extremely slow, with 48 of the icebergs observed in 1985 still trapped in the fjord in 1994. The icebergs from Matusevich Ice Shelf remain in the fjord for many years, probably due to either grounding on submarine moraines or trapping by shore-fast sea ice. Much of the sediment load of the trapped icebergs may be melted out and deposited beneath the sea-ice cover of Matusevich Fjord, and little iceberg-rafted debris of heterogeneous grain size will be transported to the Laptev Sea.
Modern glaciation of the Soviet Arctic
  • L S Govorukha
Govorukha, L.S. (1989). Modern glaciation of the Soviet Arctic. Leningrad, Gidrometeoizdat, 256 p.