CHAPTER
23
Remote sensing of glaciers in Afghanistan
and Pakistan
Michael P. Bishop, John F. Shroder Jr., Ghazanfar Ali, Andrew B.G. Bush,
Umesh K. Haritashya, Rakhshan Roohi, Mehmet Akif Sar|kaya, and Brandon J. Weihs
ABSTRACT
23.1
INTRODUCTION
Glaciers in Afghanistan and Pakistan are parts of
an Asian ‘‘critical region’’ having significant roles in
rising sea level, local and regional water resources,
natural hazards, and geopolitical stability. The two
countries lack fundamental and reliable quantitative information regarding glacier fluctuations. As
part of the Global Land Ice Measurements from
Space (GLIMS) project, we used satellite imagery
and field observations to assess a relatively large
number of glaciers in both countries. In Afghanistan, many glaciers have systematically been
observed to be retreating and downwasting. Many
glaciers have lost significant ice mass and have
evolved into numerous smaller individual ice
masses. Furthermore, the glaciers around the Kohi
Bandakha massif in southern Badakshan Province
are significantly more debris covered than other
regions in Afghanistan. In Pakistan, the situation
is more complex, as many glacier termini are variably stationary, advancing, or retreating. There
appears to be a spatial trend with more retreating
glaciers in the western Hindu Kush. To the east we
observe more advancing glaciers and surging
glaciers associated with an increase in precipitation.
These observations suggest that glacier response to
climate forcing is very different in Pakistan compared with conditions in the central and eastern
Himalaya.
The glaciers in Afghanistan and Pakistan represent
parts of an Asian ‘‘critical region’’ having significant impacts on rising sea level, local and regional
water resources, natural hazards, and geopolitical
stability (Haeberli et al. 1998, Bishop et al. 2007).
Little is known about spatiotemporal climate–
glacier responses due to a variety of factors including: (1) complex regional climate dynamics; (2) lack
of systematic data collection networks including
climate monitoring and glacier mass balance–
monitoring programs; (3) logistical difficulties
associated with rugged terrain; (4) lack of quality
baseline information including maps, aerial photos,
and geodetic control points; and (5) numerous technical issues involving information extraction from
satellite imagery. Consequently, the region lacks
fundamental and reliable quantitative estimates
regarding glacier distribution and ice volume, equilibrium line altitudes (ELAs), advance and retreat
rates, ablation rates, supraglacial lake development,
catastrophic outburst flood potential, as well as
glacier and regional mass balance estimates (Dyurgerov and Meier 2004, Haeberli 2004, Shroder and
Bishop 2010a, b).
Some glaciological studies, however, have been
conducted in both countries. In Afghanistan, Gilbert et al. (1969) worked on small glaciers near Mir
Samir in the central Hindu Kush, Braslau (1972) on
510
Remote sensing of glaciers in Afghanistan and Pakistan
Keshnikhan Glacier in the Wakhan Hindu Kush,
and Breckle and Frey (1976a, b) near the Pakistan
border. In Pakistan much of the early work was
concentrated on the ice associated with famous
peaks (e.g., Mason 1930, Auden 1935, Finsterwalder 1937, 1960, Mott 1950, Pillewizer 1956,
Loewe 1959). More recent glaciological studies
included the British International Karakoram
Project (IKP) (Goudie et al. 1984, Miller 1984),
the Chinese (BGIG 1979, 1980, Shi and Wang
1980, Wang et al. 1984), and research conducted
on numerous glaciers (e.g., Gardner 1986, Gardner
and Jones 1993, Mayer et al. 2006, Mihalcea et al.
2006).
The early Landsat series of satellites had insufficient resolution to do a fully adequate job of
inventorying the many small glaciers in Afghanistan, and the political situation and remoteness
was not conductive to field-based mapping of
glaciers. Photo-based mapping from the air was
the best option in the 1980s; Shroder (1980, 1989)
started the beginnings of a glacier inventory of
Afghanistan for the World Glacier Monitoring Service (WGMS) using photo-derived (from the air),
small-scale (1:100,000) maps to study glacier distributions. With the advent of better satellite imagery
and computer technology, several remote-sensing
and geographic information system (GIS) studies
were conducted. For example, Landsat multispectral scanner data were also used to provide baseline
information for the country (Shroder and Bishop
2010a). In addition, the Russians conducted
several studies (e.g., Maksimov and Perugina
1975, Kravtsova 1990, Kotlyakov 1996).
Remote-sensing studies of Pakistan glaciers
includes work by our group (e.g., Bishop et al.
1995, 1998, 1999, 2000, 2004, Shroder and Bishop
2010b), Russian scientists (e.g., Kotlyakov 1996,
Osipova and Tsvetkov 2002, 2003), the Italians
(e.g., Mayer et al. 2006, Smiraglia et al. 2007),
and others. Recently, Pakistani government organizations have become interested in inventorying glaciers and meltwater resources; this has resulted in
the production of a basic inventory based on Landsat Thematic Mapper data (Roohi et al. 2005,
Roohi 2007). Although the aforementioned field
and remote-sensing research has contributed to
our knowledge of selected glaciers in both
countries, there is an urgent need to better characterize glacier response to climate forcing at a
regional scale.
Previous research and regional extrapolation
based on modern studies and climate modeling
suggests that the glaciers in both countries should
be exhibiting terminus retreat and/or downwasting
(Mayewski and Jeschke 1979, Aizen et al. 2006,
Khromova et al. 2006, Mayer et al. 2006, Berthier
et al. 2007, Hasnain 2007). In Afghanistan, the
general lack of field studies and spatiotemporal
assessments makes it difficult to predict how
glaciers are responding to climate forcing, although
prior work (de Grancy and Kostka 1978, Haritashya et al. 2007, 2009) indicates that glaciers in
the Wakhan Corridor are retreating and downwasting. The situation in Pakistan appears to be more
complex due to extreme topography and increases
in precipitation in the eastern region (Fowler and
Archer 2006, Treydte et al. 2006, Roohi 2007).
Furthermore, there are many surging glaciers in
the Karakoram (Hewitt 1969, Wang et al. 1984,
Diolaiuti et al. 2003, Copland et al. 2011). Consequently, our objectives are to report on our new
spaced-based and field observations of glaciers in
Afghanistan and Pakistan, as part of the Global
Land Ice Measurements from Space (GLIMS)
project.
23.2
REGIONAL CONTEXT
23.2.1 Geology
The main glacierized orogens of Afghanistan and
Pakistan are the highly eroded product of collage
tectonics due to the continuing collision of the
Indo-Australian Plate with the Eurasian Plate and
isostatic rise of overthickened crust. Collectively,
this includes the multiple ranges of the Hindu Kush,
Pamir, Hindu Raj, Karakoram Himalaya, Nanga
Parbat Himalaya, and other diverse mountains
(Shroder 2011a, b), resulting in a high-altitude
setting with some peaks >8 km in altitude (Tables
23.1, 23.2). Both countries have a complex juxtaposition of geologic units that originally migrated
laterally across the Tethys Sea (now the Indian
Ocean) since Mesozoic time to collide and thrust
up multiple rock units into mountains. The structural framework of these mountains results from
the collision and suturing of parts of three crustal
plates; the Arabian Plate in the southwest, the
Eurasian Plate in the north, and the Indo-Pakistan
Plate (also known as the Indo-Australian Plate) in
the south and southeast. Several smaller slivers of
continental masses had broken off earlier as small
islands that moved across the Tethys, a volcanic
island arc formed in the seaway and then was
Regional context 511
Table 23.1. List of glaciated and glacierized mountain
ranges in Afghanistan (Hindu Kush, Pamir, Spin Ghar).
Map labels for some of these mountains are displayed
in Fig. 23.4.
Map label
Mountain range
1.0
Pamir
1.1
Northern Pamir (Tajikistan and
Khyrgistan)
1.2
Afghanistan Pamir
1.2.1
Badakshan Pamir
1.2.2
Wakhan Pamir
1.2.2.1
Greater Pamir
1.2.2.2
Lesser Pamir
2.0
Hindu Kush
2.1
Western Hindu Kush
2.2
Central Hindu Kush
2.3
Eastern Hindu Kush
2.4
Khwaja Mohammed (Northern) Hindu
Kush
2.5
Hesar range
3.0
Spin Ghar (Safed Koh or Koh-i-Safed
range)
crushed between the plates, while continental shelf
sediments on the plates were caught up in the
collisions, tilted on end, and metamorphosed.
Large-scale strike-slip faults, which along strike
have transpressive components, as well as thrust
fault elements, where the faults change direction
to oblique slip, constitute many of the suture zones
where rocks were accreted in plate collisions.
In Afghanistan, the main portions of the glacierized Hindu Kush and Pamir of the central and
eastern part of the country earlier were considered
only as part of an interplate marginal collision zone
(Shareq et al. 1977), but this was later refined by
Ruleman et al. (2007) to include regions of shear
zones and accreted terranes, transpressional plate
boundary mountains, and metamorphosed and
uplifted platform rocks. In addition, many of the
peaks with glaciers are formed of diverse plutonic
rocks, particularly granitoids, that were injected
into the many slivers of older rocks during platetectonic collisions.
In Pakistan, the glacierized mountains in the
north are considered geologically to be part of the
Karakoram and Himalaya Crystalline Thrust Zone
(Kazmi and Rana 1982), which is subdivided from
south to north into the Himalayan crystalline
schuppen (stacked thrust sheets) zone, the Nanga
Parbat-Haramosh massif, the Kohistan volcanic
island arc and calc-alkaline magmatic belt, and
the Karakoram (Tethyan) fold belt of shelf sediments that are now largely metasedimentary rocks
into which granitoid batholiths have intruded.
Whereas the Himalaya continue to be formed
mainly by overthrusting of the Tibetan Plateau
(Eurasian Plate side) over the Indian subcontinent,
the lateral ranges from the Pamir to the Hindu
Kush represent the opposite sense of overthrusting
with strike-slip tectonics of a lateral extrusion of
crust emanating from near the northwestern node
of the Himalayan arc (Abers and Bryan 1988). In
the Hindu Kush, tectonics are transpressive rather
than directly convergent, and crustal thicknesses
and mountain heights decrease with distance
(southwestward) away from the plate node represented by the Karakoram. Also being west of the
main monsoonal flows from the Indian Ocean, the
Hindu Kush in Afghanistan are far less heavily
glacierized than the Himalaya and derive their snow
accumulation exclusively during nonsummer
months, unlike parts of the Himalaya, where some
glaciers accumulate all year long. In Afghanistan
many small glaciers and glacierets occupy and continue to erode cirques that were left by the retreat of
Pleistocene and early Holocene ice. Compared with
the Hindu Kush of Afghanistan, glacierization
also is much greater in Pakistan at the plate node
represented by the Karakoram, where tectonics are
strongly convergent, crust is thick, mountains are
high and rising isostatically, and monsoonal
moisture supply is much greater than in semiarid
Afghanistan. Most deep valleys in the high mountains of Pakistan and the Wakhan of Afghanistan
were, and some remain today, filled deeply with ice
or ice-dammed lakes (Owen et al. 2002, 2005,
Bishop et al. 2003, Seong et al. 2007). The Pamir
in some regards are transitional between the
Karakoram and Hindu Kush styles of tectonic
deformation; the Pamir have almost no monsoonal
moisture supply but, occurring at a higher latitude,
have greater benefit from westerly winter storms;
consequently, the Pamir have an intermediate level
512
Remote sensing of glaciers in Afghanistan and Pakistan
Table 23.2. List of glaciated and glacierized mountain ranges in Pakistan: Lesser Pamir,
eastern Hindu Kush, Spin Ghar, and western (Punjab) Himalaya. Map labels are
displayed in Fig. 23.4.
Map label
1.2.2.2
Mountain range
Lesser Pamir
2.3
Eastern Hindu Kush
3.0
Spin Ghar (Safed Koh or Koh-i-Safed Range)
4.0
Hindu Raj
5.0
Kohistan ranges (Lesser Himalaya)
5.1
Dir Kohistan
5.2
Swat Kohistan
5.3
Indus Kohistan
6.0
Great Himalaya
6.1
Nanga Parbat Massif
6.2
Pangi range
7.0
Inner Himalaya
7.1
Deosai Mountains
7.2
Zanskar range
7.3
Ladakh range
8.0
Lesser (South) Karakoram Himalaya of the Trans-Himalaya
8.1
Naltar Mountains
8.2
Rakaposhi-Haramosh range (including Spantik-Sosbun Mountains)
8.3
Masherbrum range (including Shimshak and Mangor Gusar Mountains)
8.4
Saltoro-Kailas ranges (South Saltoro Mountains)
9.0
Greater Karakoram Himalaya of the Trans-Himalaya
9.1
Lupghar Mountains
9.2
Batura Muztagh
9.3
Hispar Muztagh
9.4
Panmah Muztagh
9.5
Baltoro Muztagh
9.6
Siachen Muztagh
9.7
Saser Muztagh
10.0
Lesser (North) Karakoram Himalaya of the Trans-Himalaya
10.1
North Ghujerab Mountains
10.2
South Ghujerab Mountains
10.3
Wesm Mountains
Regional context 513
of glacierization between that of the Hindu Kush
and the Karakoram.
During the Pleistocene and Holocene, ice caps
and large valley glaciers were more common and
widespread in both Pakistan and Afghanistan.
Catastrophic breakout floods from lakes near
glacier termini produced huge floods and transported large erratics out into the lowlands on many
occasions (Shroder et al. 1989, Cornwell 1998).
23.2.2 Topography
The topography in Afghanistan and Pakistan,
which has enhanced past glaciations and present
glacierization, is a function of the general rise in
altitude from west to east and south to north for
both countries. Altitude and relief culminate northeast across the Pamir and Hindu Kush in Afghanistan at the entrance to the Wakhan Corridor in the
area of Noshaq peak, which is >7 km in altitude.
Similarly, the north of Pakistan has the high Hindu
Kush, Hindu Raj, Great Himalaya (Nanga Parbat),
and especially the Karakoram Himalaya in the far
northeast, which has more >8 km high peaks than
anywhere else in the world. Swath profile analysis
using SRTM 90 m data clearly depicts the trends in
altitude and relief (Figs. 23.1, 23.2).
23.2.3 Climate
The climate of Afghanistan and Pakistan can be
characterized using NCEP/NCAR reanalysis data,
which are a synthesis of worldwide observations
passed through a numerical climate model. The
global climatological dataset is represented as a
regular grid with a 2:5 2:5 horizontal resolution
(NCEP daily global analysis data provided by the
NOAA/OAR/ESRL PSD, Boulder, Colorado,
U.S.A., from their website at http://www.cdc.noaa.
gov/). Data shown here have been averaged over
the 55-year period spanning 1950–2005. Across
Afghanistan there is a large difference in annual
mean temperature between the southwest, where
temperatures are on average 20–25 C, and the
northeast, where temperatures are 0 C to 5 C,
and into the Wakhan Corridor, where temperatures
drop to 5 to 10 C as elevation increases (Fig.
23.3a). The 25–30 C difference in annual mean temperature across the country is combined with an
equally dramatic difference in precipitation (Fig.
23.3b), with values ranging from virtually zero in
southwest–central Afghanistan to more than 300
mm yr1 in the northeast along the windward side
Figure 23.1. Afghanistan swath profile analysis
results. Minimum and maximum altitudes and relief
are depicted across the country using north–south
and west–east profiles.
of the Hindu Kush, and less than 200 mm yr1 in
the Wakhan Corridor on the lee side of the Hindu
Kush.
Pakistan also has very large spatial gradients in
temperature and precipitation since its southern
margin is a coastal region, whereas its northern
border is at high elevation in the Central Karakoram. Mean annual temperatures along a transect
from the coastal to mountainous zones range from
more than 25 C to less than 5 C in the mountains.
Climatologically, Pakistan’s southern margin is
outside the direct influence of the westerly summer
monsoon, which makes landfall over India, and
therefore the southern half of the country receives
on average less than 100 mm yr1 of precipitation.
The influence of the monsoon in Afghanistan and
Pakistan is rather complex. Although the summer
monsoon makes landfall over northwestern coastal
514
Remote sensing of glaciers in Afghanistan and Pakistan
central Pakistan into India, bringing relatively
high wintertime precipitation to these regions
(Fig. 23.3d).
23.2.4 Glaciers
23.2.4.1
Figure 23.2. Pakistan swath profile analysis results.
Minimum and maximum altitudes and relief are
depicted across the country using north–south and
west–east profiles.
India, leaving southern Pakistan primarily free
from its influence, monsoon winds encounter the
western Himalaya in northwest India at which point
a branch of the monsoon winds tracks towards the
northwest along the mountain margin into northern
Pakistan and across the border into northwestern
Afghanistan. The signature of this branch of the
monsoon winds is seen in the summer precipitation
rate (Fig. 23.3c). Over central and southern
Afghanistan during the summer monsoon, winds
are strong, northerly, and relatively dry. In the
winter, however, precipitation in the northern zones
of both countries is primarily influenced by the
westerly winds crossing Iran. Westerlies encountering the Hindu Kush split into a branch that flows
across northern Afghanistan into Uzbekistan and
Tajikistan, and another that flows across north–
Afghanistan
The glaciated mountain ranges in Afghanistan (Fig.
23.4) contain an apparent genetic continuum of
glacier types (Fig. 23.5), ranging from more or less
debris-free white-ice glaciers, through progressively
more debris accumulations on the ice. This debris
load accumulation can increase slowly or rapidly
through ablation variation controlled by climate
fluctuation, or abruptly through landslides, which
in the Badakshan part of Afghanistan is controlled
by high seismicity. As the supraglacial debris load
increases on any glacier and ultimately covers
almost all of the ice, debris-covered ice topography
becomes complex. The surface may have curvilinear
ice cliffs protruding through the debris cover, as
well as subglacial ice tunnel portals of meltwater
rivers near the glacier terminus. In the presence of
plentiful meltwater, the ice slides easily over the
substrate, basal slip occurs, and therefore the
glacier terminus itself is commonly not steep. Icecored moraines occur where heaps and hummocks
of low activity or stagnant debris-covered glaciers
have downwasted to the point of minimal to no
forward movement. On the other hand, where the
frontal movement of internal ice beneath debris is
slowed or stopped through increased friction or
being frozen to its bed, a steep front at the angle
of repose (SFAR) develops, and an ice-cored or icecemented rock glacier can result.
The issue of rock glacier terminology as generic
or genetic (Vitek and Giardino 1987) has been an
undercurrent in technical meetings and the literature for over a quarter century, mainly as a variance
between some North American (Johnson 1987),
British (Whalley and Martin 1992, Whalley and
Azizi 2003) and Continental (Barsch 1987a, b,
1996, Kääb and Reichmuth 2005) points of view
as to whether the origin of the features was glacial,
permafrost, or landslide, and whether their mechanics of motion are primarily due to internal deformation of ice, movement along subice slip surfaces,
or variations of hydrostatic pressure beneath icerich confining layers in debris (Shroder and Sewell
1985, Shroder 1987). In Afghanistan, without
detailed fieldwork, morphologic criteria were determined using only satellite imagery, and we cannot
Regional context 515
Figure 23.3. NCEP/NCAR annual mean reanalysis data for the period 1950–2005 of (a) surface air temperature
( C) and (b) precipitation (mm d1 ). Seasonal precipitation over the same time period is shown for (c) July and (d)
January. Figure can also be viewed as Online Supplement 23.1.
Figure 23.4. Mountain regions in Afghanistan and Pakistan. Region names are listed in Tables 23.1 and 23.2.
Figure can also be viewed in higher resolution as Online Supplement 23.2.
516
Remote sensing of glaciers in Afghanistan and Pakistan
readily address the origins of many of these features.
The main morphologic criteria of the SFAR of
the rock glaciers we studied is typified by steep,
light-colored frontal zones composed of coarse
rubble that occurs underlying the darker varnished
boulders capping the lower gradient surfaces of
landforms (Fig. 23.6). In a few cases these can be
confused with debris-covered rampart glaciers,
which occur where advancing glaciers with plentiful
supraglacial, englacial, and subglacial debris
sources move from a low-incline side valley out over
a steep main valley wall and frontal debris piles up
at the angle of repose. Such rampart glaciers in
Afghanistan generally have exposed ice cliffs on
their upper surfaces which allow differentiation
from rock glaciers (Fig. 23.7); some rampart
glaciers can morphologically resemble ice-cored
rock glaciers.
Numerous glacier types (as defined in the
methodology) exist in northeastern Afghanistan
and tend to have orientations predominantly to
the northeast, which is most likely a reflection of
minimum ablation from topographic shielding,
although a partial influence from structural control
on topography and some geographic sampling bias
cannot be discounted. Furthermore, we discovered
that only four of the total number of glaciers
(regardless of how you classify them) show any
evidence of terminus advancement, whereas the
remainder are either without much change in either
the position of the terminus or white-ice contact
with debris cover, or they have retreated.
Figure 23.5. Diagrams of glaciers and rock glaciers in
the Hindu Kush and western Himalaya. WI ¼ white ice
of a glacier; SFAR ¼ steep front at the angle of repose;
small triangles are rock fragments and rock debris;
EI ¼ exposed ice cliffs; H ¼ hummocks; T ¼ talus slope;
1 ¼ clean white-ice glacier with little debris and terminus slope with no steep front at the angle of repose;
2 ¼ debris-covered glacier with exposed white ice at
the head and a debris-covered terminus slope with
no steep front at the angle of repose; 3 ¼ ice-cored
moraine with profuse debris overlying more or less
stagnant downwasting ice with no exposed ice and
plentiful hummocks because of extensive differential
downwasting; 4 ¼ ice-cored rock glacier with exposed
white ice in the head region, possibly partially debriscovered but exposed ice cliffs, and a steep debriscovered front at the angle of repose; 5 ¼ ice-cemented
rock glacier with talus at the head, ice-cemented rock
debris throughout the interior, and a steep debris-covered front at the angle of repose.
23.2.4.2
Pakistan
Pakistan exhibits some of the world’s longest midlatitude glaciers in the higher altitude ranges (Fig.
23.4) due to snow accumulation from both winter
westerlies and the summer monsoon. There is also
an abundance of rock glaciers throughout the
country, although they are difficult to assess via
satellite imagery because of their small sizes and
heavy debris cover (Fig. 23.8). Pakistan is estimated
to have > 5,218 glaciers (Roohi 2007), although we
consider this to be an underestimate given that rock
glaciers have not been adequately inventoried, and
numerous glacierets cannot be adequately mapped
using satellite imagery. Nonetheless, an estimate of
glacier ice reserves indicates > 2,700 km3 of storage
(Roohi 2007); this figure is unaffected by the smaller
glaciers not included in the total number count. In
Regional context 517
Figure 23.6. Synthetic oblique view (satellite imagery draped over digital elevation model) looking southwest,
from a viewpoint above the Sanglech Valley in Badakshan of the Kohi Bandakha massif showing three different
combinations of rock fragments and ice. (A) The dark-colored rock-varnished Sanglech rock glacier (left and lower
center image) with its pronounced lobate toe with transverse ridges and furrows, as well as the steep front at the
angle of repose (SFAR; arrow) that highlights with its light color the unvarnished finer clastics that have settled
down between the interstices of coarser clasts. (B) The main Sanglech Glacier (upper right and center image)
whose exposed ice cliffs and lighter colored rock fragments reveal it to be an active debris-covered glacier. (C)
Between the two glaciers mentioned above (picture center) can be seen the Sanglech ice-cored moraine. The strong
color contrasts between the Sanglech rock glacier (A) and the Sanglech Glacier (B) in part reflect lithological
differences as well as sediment transport. Sanglech Glacier (B) exhibits clast interaction, as opposed to the
piggyback transport and lack of clast interaction on rock glacier A, coupled with varnish accumulation. Rocks
on active debris-covered glaciers are constantly being moved up and down through the ice by way of ice thrusting,
crevassing, melting and refreezing of the ice, and pervasive jostling and clast interaction that collectively keep rock
varnish to a minimum. Large clasts on the surface of rock glaciers tend to be carried piggyback without much clast
interaction so that rock varnish can accumulate effectively. Ice-cored moraines tend to be inactive, with debris
accumulating progressively and thickening over ice that is melting away in the lower reaches. Only rock glaciers
develop light-colored SFARs where the uppermost rock-varnished surface of coarse rock fragments contrasts
markedly with light-colored SFARs that have mixtures of coarse and fine clasts (picture from Google Earth).
addition, 2,420 glacier lakes have been identified,
among which 1,328 are characterized as major lakes
> 0:02 km2 , 52 of which are classified as potentially
dangerous with a high risk of outburst flooding
(Roohi 2007). Lake formation, size, magnitude,
and spatial frequency is thought to have increased
based on multitemporal image interpretation,
although detailed analysis to estimate seasonal
518
Remote sensing of glaciers in Afghanistan and Pakistan
Figure 23.7. Synthetic oblique view (satellite imagery draped over digital elevation model) of debris-covered
rampart glacier 8:5 km northeast of Kohi Bandakha that formed when a debris-covered glacier moved out of its
cirque and into the adjacent valley over a ramp of its own debris piled up at the angle of repose (arrow). Such
rampart glaciers superficially resemble rock glaciers with the rampart front at the angle of repose but the exposed ice
cliffs on the top, and further evidence of a thin debris cover over the ice highlights the difference (picture from
Google Earth).
and annual baseline fluctuations in supraglacial
lake size and frequency has not been established.
23.3
METHODOLOGY
The data utilized for assessing glaciers in Afghanistan and Pakistan included a variety of satellite
sensor imagery depending on issues of data quality
and temporal availability. In Afghanistan, the relatively small size of glaciers and rock glaciers prohibited the use of satellite imagery such as Landsat
TM and ASTER, as the spatial resolution of the
sensors prohibits accurate identification and differentiation. Consequently, high-resolution satellite
imagery coupled with digital terrain information
(available from Google Earth) was used to assess
the topographic character of alpine glaciers and
differentiate them from rock glaciers, as the SFAR
is a defining characteristic. Larger alpine glaciers in
Afghanistan were assessed using readily available
satellite imagery. Furthermore, we used highresolution digital Corona data to assess glacier conditions. All Corona imagery was co-registered to
orthorectified ASTER imagery based on the selection of image control points.
Access to declassified Corona satellite photos
from 1961 enabled direct comparison with various
topographic maps made from photos taken from
the air shortly before (Glicken 1960) which were
Methodology
519
Figure 23.8. Ground photo (1993) of a rock glacier in the high-altitude Khunjerab Pass area of the North Gujerab
Mountains of northern Pakistan. Note the extensive debris lobes and steep front at the angle of repose characteristic
of rock glaciers. Figure can also be viewed in higher resolution as Online Supplement 23.3.
produced by American (U.S. Department of
Defense) and Soviet sources, as well as the Afghan
Cartographic Institute in Kabul. In addition, we
utilized a variety of other satellite imagery (ETM,
MSS) in the ensuing decades up to the present to
map changes in the glaciers, but made our measurements with ASTER imagery from 2006. Because
ice-cored moraines and rock glaciers have long
timescales for significant changes due to their protective debris, ‘‘deflation’’ occurs very slowly as ice
melts out. Because satellite imagery does not effectively permit detecting subtle changes in downwasting and retreat of such glaciers, we concentrated
mainly on detecting change in the uppermost
white-ice streams where they were present, rather
than glacier or rock glacier termini.
We selected 341 glacier/rock glacier combinations in six different mountain ranges in the Hindu
Kush, Badakshan Pamir, and Wakhan Pamir of
northeastern Afghanistan. We characterized the
glaciers of these ranges into those that are: (1) dominantly white-ice glaciers that are not covered with
much debris; (2) debris-covered glaciers with plentiful meltwater or dominantly ice-cored moraines,
commonly without much meltwater; (3) rampart
glaciers; (4) ice-cored rock glaciers; and (5) icecemented rock glaciers. Both ice-cored and icecemented rock glaciers tend to have steep fronts
(termini) at the angle of repose (SFAR) because
basal slip is restricted at least in their lower
reaches, whereas ice-cored moraines, debriscovered glaciers, and white-ice glaciers are generally
characterized by basal slip in their lower sections at
these latitudes and altitudes so that no SFAR
occurs, except where the slope changes abruptly
as with rampart glaciers.
We also estimated the advance/retreat rates for
larger glaciers in Afghanistan and Pakistan. All
imagery was orthorectified and multitemporal
imagery co-registered to obtain the most reliable
changes in glacier terminus position. Visual
examination of registration results revealed no
detectable registration errors. Table 23.3
provides a listing of the data used for the Wakhan
520
Remote sensing of glaciers in Afghanistan and Pakistan
Table 23.3. Satellite imagery and acquisition dates
used to estimate glacier terminus change for regions
in Afghanistan and Pakistan.
Region
Wakhan Pamir
Hindu Raj
Batura/
Hispar Mustagh
Nanga Parbat
Imagery
Acquisition
date
Landsat MSS
ASTER
9-Aug-1976
23-Jul-2003
Landsat MSS
ASTER
26-Sep-1972
9-Jun-2007
Landsat MSS
ASTER
4-Aug-1973
13-Sep-2004
Topographic map
ASTER
1934
13-Sep-2004
Pamir region of Afghanistan and regions within
Pakistan.
Advance/retreat rate calculations followed a
rigorous procedure to produce the highest quality
estimates. Visual interpretation of multitemporal
satellite imagery and spectrally enhanced imagery
was used to identify glacier terminus positions.
Specifically, we used the farthest downvalley location of the terminus. In all cases, terminus positions
were easily identifiable in the imagery due to
variations in surface biophysical conditions and/
or topography (i.e., glacier surface versus forefield
characteristics that influence spectral reflectance).
For some glaciers, subtle spectral reflectance variations were enhanced using false-color composites,
image ratioing, or principal component analysis.
We then utilized a transect approach to compute
the linear distance from terminus locations. This
was accomplished by establishing a glacier flow
transect that characterizes the azimuthal direction
of glacier flow downvalley. We made sure that each
glacier sampled had a relatively consistent linear
flow direction, and that advance or retreat did
not occur in a curvilinear fashion. Therefore, terminus positions were projected perpendicular to the
flow–azimuth orientation in order to compute the
change distance. This approach is required to
account for changes in the shape of the terminus
of a glacier through time, as the maximum downvalley location varies across the width of the glacier.
Glacier terminus change rates were computed based
on calculated distances and image acquisition dates
and times. Leap years were taken into account when
calculating advance/retreat rates. The inherent
error associated with estimating the change distance
is a function of the spatial resolution of the sensors,
the azimuth orientation of glacier flow, and registration error.
23.4
CASE STUDIES
23.4.1 Afghanistan
23.4.1.1
Mir Samir
We assessed 25 glaciers and associated features
(ice-cored moraines, rock glaciers) around the
Mir Samir massif in the Central Hindu Kush just
south of the Panjshir River valley (Fig. 23.4; region
2.2). Mir Samir is a prominent glacial horn composed of faulted gneisses and granites that rises to
an altitude of 5,809 m above a gipfelflur (summit
area) of accordant summit topography that
averages around 5,000 m. Glaciers on the north side
of Mir Samir are among the few in Afghanistan that
have ever had any real assessment of their mass
balance regimes (Gilbert et al. 1969). Gilbert et
al. thought that the real climatic snowline was a
little above the highest summits but that the steep
north-facing slopes and shadowed cirques provided
topographic conditions for small glaciers to persist.
Annual snow accumulation for 1958–1965 was
thought by them to average some 1,300 kg m2
for the eight prior years of snow accumulation they
observed in the bergschrund at 4,900 m. Maximum gross ablation was calculated to be 40 kg
m2 d1 on one particular day but was generally
between 10 and 20 kg m2 d1 . Net ablation in
July–August 1965, measured as stream discharge,
averaged 3,600 m3 d1 , or 9 kg m2 d1 . Slow
recession was observed to be the average condition
for the glacier. Halt stages of glacial retreat were
interpreted from morainal topography and lichenometry to be at altitudes of 4,000, 4,600, and
4,800 m, the latter dated at a minimum of 400
years ago.
Mir Samir East Glacier (Yakhchaal-i Sherq or
AF5Q1212135, Table 23.4) and West Glacier
(Yakhchaal-i Charb or AF5Q1212143, Table
23.4) below the north face of Mir Samir were
mapped and estimated by Gilbert et al. (1969) to
be 1 km2 and 0:5 km2 , respectively. Considerable downwasting and backwasting of both these
glaciers has occurred in the intervening 4þ decades
since Gilbert and his colleagues were at the site.
This has left light-colored glacier forefields, moraines, and bare bedrock exposed between the older
moraines and the ice fronts of today. In fact, West
Case studies
Glacier in 1965 can be seen on ASTER imagery
from 2006 and Google Earth imagery to have
downwasted sufficiently across a rock rib so that
it has now split into two glaciers (AF5Q1212143.1
and AF5Q1212143.2, Fig. 23.9).
Most of the 25 glaciers surveyed around Mir
Samir were white-ice glaciers (17). There were a
few ice-cored moraines (3) and rock glaciers (5).
Glacier features in this region are all small, with a
range of lengths of 0:42–3.47 km, an average
length of 1:25 km, and an average width of
0:44 km. From 1961 to 2006 the white-ice glaciers
here had an average retreat of 0.2 km and average
retreat rate of 5 m yr1 .
23.4.1.2
Daste Wer
This area directly south of Mir Samir has lithologies
and altitudes similar to those around Mir Samir. It
has a total of 30 glaciers and related cryospheric
features, 26 of which are ice-cored moraines,
oriented predominately to the northwest, three
are northeast-facing rock glaciers, and one is a
white-ice glacier. The ice-cored moraines are generally headed by white-ice deposits, thus suggesting
a glacier-like origin. The average length of all
glaciers and glacier-like features sampled is 0.99
km and the widths average 0.38 km. From 1961
to 2006 white-ice contacts exposed at the heads of
ice-cored moraines showed an average retreat of
0.19 km (average retreat rate of 4 m yr1 ).
23.4.1.3
Balacomar
To the east of Mir Samir, the Balacomar range has
25 glaciers and similar cryospheric landforms that
we analyzed in imagery; 15 of these are ice-cored
moraines that face north-northwest, 8 are rock
glaciers with the same aspect, and 2 are white-ice
glaciers facing southwest. Unlike neighboring
regions, the glaciers in this region tend to be shorter
and wider, with an average length of 1.1 km and a
width averaging 4.1 km. From 1961 to 2006 the
white-ice streams of the ice-cored moraines show
an average retreat of 0.06 km and a retreat rate
of 1.3 m yr1 .
23.4.1.4
Bandaka
The Kohi Bandaka massif is characterized by six
major peaks > 6,000 m in altitude, and a number of
others almost as high (Shroder and Bishop 2010a).
The main peak of Bandaka is 6,843 m in altitude.
The mountain is composed of resistant gneisses of
521
Meso-Archean age that have been forced up at a
pronounced (probably transpressional) bend in the
major Panjshir–Central Badakshan strike-slip fault
system that exists there. The result is an upstanding
massif of sufficient altitude to form numerous
masses of glacier ice. We assessed 42 glaciers on
the Koh-i-Bandaka (Bandakhor, in some usages)
massif located on the south side of the northern
Hindu Kush for which we have 45 years of repeat
satellite imagery (1961–2006). Of these glaciers 21
have multiple tributaries, some as many as 6 (Table
23.5).
Grötzbach (1990) first mapped > 30 Blockstrom,
which in his German usage (oral commun. from
Grötzbach to Shroder in 1977) most closely translates to ‘‘rock glacier’’ (Giardino et al. 1987). Similarly, Shroder and Bishop (2010a) also mapped the
same region and showed a similar number of
glacier-like forms, which they considered to be rock
glaciers or ice-cored moraines. Such debris-laden
cryospheric features on this massif and adjacent
mountain ridges are numerous at high elevations.
The abundant debris may relate to the fact that
Koh-i-Bandaka is situated almost directly over
the main seismic source beneath the Hindu Kush
(Wheeler and Rukstales 2007). This high seismicity
generates copious rock fragments that cascade off
cliffs in rock falls and ice avalanches to generate
thick debris covers on the glacier ice. It is not clear,
however, why some of these glaciers have such
plentiful rock debris on them, in some cases to
the point where almost no ice shows on the surface,
and they have become classic ice-cored rock glaciers
with SFAR, whereas a few others that are directly
adjacent have comparatively little debris. In addition, some of the features more closely resemble icecored moraines where the glacier ice is covered
almost completely with thick rock debris and does
not exhibit a rock glacier–defining SFAR (Fig.
23.6).
Glaciers and rock glaciers on the Kohi Bandaka
massif constitute a number of types of an apparent
genetic continuum. Examples include: (1) white-ice
glaciers with little debris (Fig. 23.10); (2) slightly to
heavily debris-covered glaciers (Fig. 23.10); (3)
downwasted (‘‘deflated’’) ice glaciers with lateral
and terminal moraines; (4) ice-cored moraines with
exposed ice patches in the cirques, possible ice cliffs
in lower reaches, and no SFAR (Fig. 23.6); (5) icecored rock glaciers with exposed ice patches in the
cirques and a SFAR (Fig. 23.6); (6) ice-cemented
rock glaciers, generally with only talus exposed at
the top, and with SFAR; and a very few rampart
Name
E
NNE
NE
NNE
E
NE
E
ESE
NNW
NNW
NNE
AF5Q1212138
AF5Q1212139
AF5Q1212133.4
AF5Q1211050
AF5Q1211051
AF5Q1211052
AF5Q1211053
AF5Q1212131
AF5Q1212132
AF5Q1212133.1
Azimuth
AF5Q1212134
ID
Glacier
Longitude
(70 E)
11 0 06.100 00
11 0 16.270 00
10 0 51.268 00
11 0 49.367 00
11 0 13.600 00
12 0 03.190 00
11 0 02.960 00
10 0 56.720 00
13 0 18.673 00
12 0 44.964 00
10 0 46.355 00
Latitude
(35 N)
35 0 30.527 00
37 0 43.229 00
37 0 50.050 00
34 0 32.719 00
32 0 15.300 00
32 0 05.360 00
32 0 38.720 00
33 0 35.590 00
34 0 36.835 00
34 0 25.596 00
34 0 36.267 00
Location
6.74
5.58
4.88
0.00
0.00
0.00
0.00
3.72
1.63
2.79
6.28
Retreat a
(m yr1 )
Lake size changes. White-ice retreat 0.29 km 1961–2004
New lakes, lake removals and size changes. White-ice retreat
0.24 km 1961–2004
New lakes, lake removals and size changes. White-ice retreat
0.21 km 1961–2004
Little change between 1961 and 2004
Little change between 1961 and 2004
Lake size changes, new lakes, little white-ice change 1961–2004
New supraglacial lake in 1999
New lakes. White-ice retreat 0.16 km 1961–2004
Lakes removed. White-ice retreat 0.07 km 1961–2004
Lake removal and size changes. White-ice retreat 0.12 km 1961–2004
New lakes and lake size changes. White-ice retreat 0.27 km
1961–2004
Changes
ASTER-derived parameters (1961–2004)
Table 23.4. Glaciers and related features (ice-cored moraines, rock glaciers) near the peak of Mir Samir (5,809 m) in the central Hindu Kush.
522
Remote sensing of glaciers in Afghanistan and Pakistan
a
SW
WNW
NW
ENE
E
NE
AF5Q1212146.1
AF5Q1212146.2
AF5Q1212147
AF5Q1211054.1
AF5Q1211054.2
AF5Q121155
NDC ¼ no detectable change.
W
AF5Q1212143.2
WNW
AF5Q1212142
West Glacier B
SE
AF5Q1212137
W
E
AF5Q1212136
AF5Q1212143.1
ENE
AF5Q1212135
SSE
AF5Q1212133.3
Mir Samir West
Mir Samir East
NE
AF5Q1212133.2
11 0 10.540 00
10 0 30.719 00
10 0 31.774 00
10 0 15.324 00
10 0 20.373 00
09 0 26.770 00
09 0 17.345 00
08 0 51.314 00
09 0 12.854 00
09 0 16.440 00
10 0 10.928 00
10 0 41.941 00
10 0 42.329 00
14 0 11.706 00
34 0 30.578 00
35 0 14.265 00
35 0 56.288 00
37 0 01.990 00
37 0 16.895 00
36 0 03.115 00
35 0 30.771 00
35 0 30.771 00
35 0 13.475 00
34 0 40.820 00
33 0 59.765 00
33 0 44.801 00
33 0 58.483 00
34 0 17.659 00
1.86
2.79
0.00
2.79
3.49
1.86
5.58
1.86
6.05
8.14
6.05
8.84
2.09
11.40
White-ice retreat 0.08 km 1961–2004. Ablated to patches of ice
White-ice retreat 0.12 km 1961–2004
Little change between 1961 and 2004
White-ice retreat 0.12 km 1961–2004
Lake removed and lake size changes. White-ice retreat 0.15 km
1961–2004
New lake. White-ice retreat 0.08 km 1961–2004
White-ice retreat 0.24 km 1961–2004
Lake size changes. White-ice retreat 0.08 km 1961–2004
White-ice retreat 0.26 km 1961–2004
White-ice retreat 0.35 km 1961–2004
White-ice retreat 0.26 km 1961–2004
New lakes and lake size changes. White-ice retreat 0.38 km
1961–2004
Lake removal. White-ice retreat 0.09 km 1961–2004
New lakes, lake removals, and size changes. White-ice retreat
0.49 km 1961–2004
Case studies
523
524
Remote sensing of glaciers in Afghanistan and Pakistan
Figure 23.9. Synthetic oblique view (satellite imagery draped over digital elevation model) to the northeast of Mir
Samir (5,809 m) at the Yakhchaali Gharb or Mir Samir West Glacier. Since 1965 when this glacier was first mapped
in detail, it has downwasted sufficiently across a rock rib (vertical up arrow) such that now it has split into two
glaciers (Table 23.4, AF5Q1212143.1 and AF5Q1212143.2). The westernmost ice patches in the cirque on the
right (vertical down arrow) are now little more than thin dead ice remnants, whereas the main glacier in the cirque in
the left center still retains its bergschrund in spite of severe downwasting into its pronounced bowl-shaped form of
the present day (picture from Google Earth).
glaciers that have an apparent SFAR that on close
inspection is collapsed terminal moraine (Fig. 23.7).
Results indicate that the 42 glacier-like forms
measured on the Bandaka massif range from 1–
10 km in length and average 3:2 km long and
0:7 km wide. All the features with exposed ice
have lost ice mass in the near half century of
imagery assessed; most rock glaciers probably have
thinned somewhat but quantitative estimates are
difficult to assess. Most supraglacial lakes have
changed position and size over the observation
time.
We assessed 17 rock glaciers, 15 ice-cored moraines, and 10 white-ice glaciers. The orientation
azimuth of these different types is unusual in that
they are somewhat preferentially oriented, such that
the rock glaciers are located on west and southwest
slopes, the ice-cored moraines tend to face southeast, and the white-ice glaciers are oriented northeast, southeast, and southwest. The reasons for
these differences are not clear. The retreat of white
ice exposed at the tops of some ice-cored rock
glaciers averaged 0.44 km, with an average rate of
retreat of 11 m yr1 , whereas the retreat of ice-cored
moraines was 0.28 km (7 m yr1 ) and of white-ice
glacier termini was 0.44 km (11 m yr1 ).
23.4.1.5
Badakshan
Glaciers in Badakshan occur throughout the higher
parts of this province’s mountain ranges, especially
on the main Kohi Safed Khirs just west of the
main Central Badakshan Fault that trends approximately north–south through the north–central part
of the province. To the east of this fault system,
glaciers also occur in the regional Kohi-Hazar
Chashma directly north and south of Lake Shewa,
especially where the peaks rise to nearly 5,000 m all
along the bank of the Panj River (Abi Panj). This
river emerges from the Wakhan Corridor to the
AF5XA03111A
AF5XA03111C
AF5XA03111G
AF5XA03110F
AF5XA03110A
AF5XA03110B
AF5XA03110C
AF5XA03110D
AF5XA03110E
Sanglech B
Sanglech C
Sanglech D
Sanglech E
Sanglech F
Dshurm A
Dshurm C
Dshurm G
AF5XA03111
Sanglech A
N
AF5XA03115
A032110
SE
AF5XA03112
Dshurm E
Tergaram
SSE
AF5XA03111E
Dshurm
NNE
SE
ESE
NE
SSE
NNE
ESE
SE
SE
W
E
SE
AF5XA0313
Sanglech
Azimuth
ID
Name
Glacier
Longitude
(70 E)
01 0 10.617 00
59 0 44.816 00
59 0 29.821 00
00 0 04.140 00b
04 0 04.592 00b
58 0 55.036 00
00 0 27.85 00b
59 0 45.312 00
59 0 33.162 00
01 0 40.473 00b
59 0 49.437 00
02 0 55.67 00b
58 0 43.466 00
58 0 52.989 00
00 0 55.095 00b
Latitude
(36 N)
10 0 53.084 00
15 0 00.625 00
16 0 04.820 00
16 0 12.540 00
13 0 12.220 00
09 0 52.631 00
12 0 50.819 00
12 0 05.290 00
11 0 45.163 00
10 0 12.130 00
12 0 29.189 00
09 0 39.518 00
15 0 40.465 00
15 0 40.397 00
13 0 57.323 00
Location
0.00
2.75
2.50
1.50
20.00
11.75
3.50
13.75
3.00
3.00
0.00
25.50
0.00
0.00
0.00
Retreat a
(m yr1 )
Little change between 1961 and 2001
(continued)
New lakes and lake size changes 1961- 2001. White-ice retreat
0.11 km
Increase in lake number. White ice retreat 0.1 km 1961–2001
Lake size changes. White-ice front retreat 0.06 km 1961–2001
White-ice front retreat 0.8 km 1961–2001
White-ice front retreat 0.47 km 1961–2001
White-ice front retreat 0.14 km 1961–2001
White-ice front retreat 0.55 km 1961–2001
White-ice front retreat 0.12 km 1961–2001
Upper white-ice retreat 0.12 km 1961–2001
Possible surge 1973–1999
New lakes. Lake position and size changes 1961–2001. White ice
retreat 1.02 km
Lake size increase 0.005 km2 1961–2001
Many lakes changed position and size 1961–2001
Little change between 1961 and 2001
Changes
ASTER-derived parameters (1961–2001)
Table 23.5. Various glaciers and related types (ice-cored moraines, rampart glaciers, rock glaciers) on the Kohi Bandaka massif in southern Badakshan,
Afghanistan. Letter designations indicate tributaries to a single glacier, although in a few cases downwasting and backwasting may have separated formerly
contiguous glaciers from each other.
Case studies
525
A032122
A032123
A03219
Achdan North
Achdan South
Sanglech RG
WNW
A032119
A032121
W
A032118
Dasmana East
W
A032117
A032120
WSW
A032116
Bandana
WSW
A032111I
Sakhi East 4
ENE
ESE
SE
SSW
WSW
WNW
A03218
NW
A03216
Sakhi East 3
W
AF5XA0314
WSW
NNW
AF5XA0312.4
A03218A
N
AF5XA0312.3
Sakhi East 2
N
AF5XA0312.2
Dshurm I
Azimuth
ID
Glacier
Name
Table 23.5 (cont.)
Longitude
(70 E)
01 0 50.298 00b
05 0 16.57 00b
06 0 11.668 00b
57 0 52.111 00
57 0 1.864 00
57 0 53.423 00
57 0 53.423 00
57 0 53.423 00
57 0 34.668 00
57 0 26.682 00
57 0 20.929 00
56 0 46.682 00
58 0 32.991 00
58 0 13.701. 00
02 0 23.419 00
02 0 23.419 00
03 0 23.474 00b
Latitude
(36 N)
14 0 13.396 00
13 0 29.939 00
13 0 20.538 00
16 0 33.644 00
16 0 07.523 00
11 0 51.248 00
11 0 51.248 00
11 0 51.248 00
15 0 43.214 00
15 0 13.706 00
14 0 55.186 00
14 0 36.485 00
08 0 53.825 00
06 0 41.842 00
08 0 48.639 00
08 0 48.639 00
09 0 20.115 00
Location
0.00
0.00
0.00
0.00
7.25
10.50
12.25
10.75
19.25
0.00
0.00
0.00
6.00
10.50
0.00
0.00
0.00
Retreat a
(m yr1 )
Little change between 1961 and 2001
Many lakes changed position and size 1961–2001
Many lakes changed position and size 1961–2001
Many lakes changed position and size 1961–2001
Upper white-ice stream retreat 0.29 km 1961–2001.
Lake 2.04 km2 1961 to 4.8 km2 2001
White-ice retreat 0.42 km 1961–2001
White-ice retreat 0.49 km 1961–2001
White-ice retreat 0.43 km 1961–2001
White-ice retreat 0.77 km 1961–2001
Increased debris cover, separated tributary white-ice streams
Increased debris cover, separated tributary white-ice streams
Increased debris cover, separated tributary white-ice streams
White-ice retreat 0.24 km 1961–2001
White-ice retreat 0.42 km 1961–2001. New lake in 1999
Glacier ablated entirely between 1961 and 2001. No lake changes
Rock glacier. Drained lake 1961–2001
Lake drained between 1999 and 2001. No white-ice retreat
Changes
ASTER-derived parameters (1961–2001)
526
Remote sensing of glaciers in Afghanistan and Pakistan
b
a
AF5XA03112A
AF5XA03111H
NDC ¼ no detectable change.
This coordinate is 71 E.
A032114
NNE
AF5XA03111F
Tigarah
ENE
AF5XA03111D
Dshurm H
A032113
WSW
AF5XA03111B
Dshurm F
Sakhi East 1
NNE
AF5XA0317
Dshurm D
SSE
SW
NNE
SSE
NE
AF5XA0312.1
Dshurm B
SSW
A03211
Sakhi West
57 0 37.188 00
59 0 8.6 00
59 0 11.004 00
59 0 58.657 00
01 0 26.999 00b
01 0 05.374 00b
02 0 14.594 00b
02 0 58.923 00b
57 0 53.423 00
59 0 51.072 00
13 0 24.991 00
14 0 59.893 00
16 0 11.844 00
13 0 46.822 00
13 0 56.322 00
16 0 04.639 00
13 0 23.325 00
12 0 45.823 00
11 0 51.248 00
09 0 21.164 00
32.50
00.00
8.25
3.25
0.00
5.50
2.75
8.50
0.00
16.84
Ice front retreat 1.3 km 1961–2001
Increased debris cover, separated tributary white-ice streams
White-ice retreat 0.33 km 1961–2001
Little change between 1961 and 2001. White-ice retreat 0.13 km
Many lakes changed position and size 1961–2001
White-ice retreat 0.22 km 1961–2001
White-ice retreat 0.11 km 1961–2001
White-ice retreat 0.34 km 1961–2001
Many lakes changed position and size 1961–2001
White-ice front retreat 0.64 km 1961–1999
Case studies
527
528
Remote sensing of glaciers in Afghanistan and Pakistan
Figure 23.10. Synthetic oblique view (satellite imagery draped over digital elevation model) looking northnortheast at Sakhi East Glacier 2 (Table 23.5) on left (vertical up arrow). Sakhi East Glaciers 3 and 4 occur to
the right center on the north slopes of the Kohi Bandakha massif (6,843 m), the main peak of which is out of the
picture just to the right in this view. In recent years Sakhi East Glacier 2 has retreated from the light-colored glacier
forefield in the lower center of the picture and left a new lake (horizontal arrow) (picture from Google Earth).
southeast of Badakshan and turns north to constitute the international border with Tajikistan.
Badakshan has three important areas of glaciers
that we describe as follows: (1) YagardaChukshash; (2) Lali-Mishash; (3) Tshali Dara.
The Yagarda-Chukshash area of the southern
Badakshan Pamir close to the entrance to the
Wakhan Corridor is dominated by resistant
Archean-aged gneisses that have a number of
north/south-trending shear zones. This orientation
of rock ridges has helped produce more than 33
glaciers, with a dominant orientation to the northeast, an average length of 2.25 km and an average
width of 0.45 km. Of the 33 glaciers 14 are rock
glaciers, 11 are ice-cored moraines, and 8 are whiteice glaciers. Only one of these white-ice glaciers
advanced 0.25 km and the termini of the rest
remained stationary. The exposed white-ice streams
above the ice-cored rock glaciers retreated an average of 0.27 km and their retreat rate was 6 m yr1 .
Ice-cored moraines in this region neither advanced
nor retreated. Directly north of Lake Shewa in
Badakshan, the local Lali-Mishash range rises in
a series of more than 10 peaks greater than 4,600
m, with the highest at 4,837 m; and all are
surrounded by glaciers, of which we have assessed
48. The region is dominated by 44 white-ice glaciers
with only a few (2) debris-covered glaciers/ice-cored
moraines and 2 rock glaciers. Comparison of nearly
half a century of satellite imagery of these glaciers
shows that almost all have been in retreat, and
many are surrounded by light-toned glacier forefields that characterize long-term downwasting
and backwasting. Average white-ice glacier retreat
was 100 m and the retreat rate was 2 m yr1 . The
meltwaters of these ice masses flow into the Abktal
and the Abzijad Rivers that drain into the north
arm of landslide-dammed Lake Shewa (Shroder
and Weihs 2010).
We measured 110 glaciers and similar features in
the local mountains of Tashali Dara in northern
Badakshan, and found a mixture of glacier types,
although 58 are predominantly clean white-ice
glaciers. There are 31 rock glaciers and 21 ice-cored
Case studies
moraines. There were no advances for any of the
white-ice glaciers, and their retreat averages 9 m
yr1 , with a retreat rate of 0.2 m yr1 . The white
ice exposed at the heads of the 31 ice-cored rock
glaciers retreated an average of 12 m yr1 at an
average rate of 2 m yr1 . The white ice of the icecored moraines had an average retreat of 3 m yr1 .
23.4.1.6
Wakhan Pamir
We assessed 30 glaciers in the Wakhan Pamir
(Table 23.6). The glaciers are relatively small to
medium in size, and terminus positions occur within
an altitude range of 4,500–5,000 m. Although we
could not assess debris-cover conditions due to
snow cover present in available ASTER imagery,
we note from prior publications (de Grancy and
Kostka, 1978) and Google Earth imagery that
many of these glaciers are heavily debris covered
in their lower reaches, and currently exhibit surface
geomorphological evidence of rapid downwasting.
It should be noted that these glaciers do not occur
in the WGMS database.
Many glaciers in the region show the growth of
small lakes near their termini, and this would likely
represent a response to a climate shift of some type.
Some are supraglacial, some are moraine-dammed
lakes, and others are glacier dammed and formed
where glacier tributaries have detached. For
example, Google Earth views (Digital Globe and
SPOT images) show about six small lakes near
the termini of East and West Little Ali Su Glaciers
in the Wakhan Corridor (37 09 0 45 00 N, 73 10 0 52 00 E).
These lakes and most of those now appearing near
many glaciers in the region generally were not
present in 1975 or have grown since then. The comparison dataset consists of detailed maps made by
Austrian mountain climbers in 1975 under the
leadership of Robert Kostka, a consummate photogrammetrist. He used a stereo-phototheodolite
(Wild), which took pictures looking down from
all the peaks they climbed and made excellent topographic maps at 1:50,000 and 1:25,000 of all the
glaciers. Their base was the 1:100,000 topographic
maps made from aerial photos taken in the late
1950s and early 1960s.
Our retreat results (Table 23.6) indicate that 28 of
the 30 glaciers are retreating, with an average
retreat distance of 294 m over the 27 years. Retreat
distance and downwasting variability is high, and
the larger glaciers appear to have changed more
than the smaller ones. For example, five large
glaciers (Western Ali Su, Eastern Ali Su, Western
529
Bay Tibat, Eastern Bay Tibat, Tila Bay, and Ptukh)
exhibited the greatest retreat distances. In addition,
a large compound valley glacier, Northern Issik,
has retreated and become a valley glacier, as its five
tributaries have partly detached and now seem to be
at least two independent smaller glaciers with considerable apparently stagnated ice-cored moraine in
their lower reaches. The two separated tributaries to
the northeast exhibit geomorphological patterns of
downwasting and thinning, with the lateral moraine
of the main glacier acting as an end moraine or icecored moraine for the detached tributary glaciers
(Fig. 23.11).
Image analysis also reveals that the larger glaciers
with high average retreat rates exhibit relatively
high debris cover at the terminus, which is presumably the result of increased ablation. Smaller
glaciers at higher altitudes do not have as much
debris cover. Glaciers with lower average recession
rates commonly exhibit signs of surface downwasting. For example, Zemestan and Ptukh Glaciers
have significantly downwasted. The terminus
regions of these glaciers are topographically
entrenched within relatively higher lateral and end
moraines. Glacier geomorphological patterns provide strong evidence that downwasting is a major
part of glacier mass loss in this region, rather than
strict terminus recession. Paul et al. (2007) suggested that frontal glacier recession is commonly
coupled to lateral glacier thinning of a similar magnitude, although some glaciers in the region do not
appear to be downwasting. Large glacier forefields
exist beyond the 1976 terminus positions for many
glaciers, indicating that rapid retreat also occurred
prior to 1976 in the Wakhan Pamir.
23.4.2 Pakistan
23.4.2.1
Hindu Raj
Hundreds of small to moderate-sized glaciers occur
in the Hindu Raj located in the extreme northwest
of Pakistan.There are no summits over 7,000 m, but
there are high peaks like Koyo Zom (6,889 m),
Ghamobar Zom (6,518 m), Buni Zom (6,551 m),
Dasbar Zom (6,072 m), and Chiantar Chish (6,273
m). In general, alpine glaciers are located predominately on the northern side of the Hindu Raj, close
to the Wakhan Corridor. The largest glaciers occur
on the northern faces of the peaks. Chiantar Glacier
is the largest at 32 km long. The southern side of
AF5X14220057
AF5X14220061
AF5X14220080
AF5X14220081
AF5X14220083
AF5X14220084
AF5X14220100
AF5X14220105
AF5X14220117
AF5X14221001
AF5X14221002
AF5X14221003
AF5X14221004
—
—
—
Tila Bey
—
E Bey Tibat
W Bey Tibat
E Ali Su
W Little Ali Su
E Little Ali Su
Western Ali Su
West Issik
ID a
—
Name
Glacier
E
N
NE
NE
N
N
N
E
NE
N
NW
N
NW
Azimuth
Longitude
(73 E)
27 0 59.04 00
23 0 38.38 00
22 0 32.24 00
21 0 20.30 00
18 0 24.04 00
18 0 16.59 00
16 0 20.91 00
13 0 00.89 00
10 0 52.31 00
07 0 20.58 00
06 0 33.03 00
08 0 52.70 00
10 0 34.82 00
Latitude
(37 N)
13 0 37.06 00
12 0 35.14 00
09 0 06.97 00
08 0 22.59 00
08 0 14.49 00
10 0 39.05 00
11 0 00.64 00
10 0 00.00 00
09 0 44.75 00
05 0 35.03 00
06 0 07.04 00
05 0 35.02 00
07 0 23.35 00
Location
07 0 51.47 00
07 0 29.70 00
06 0 52.85 00
06 0 52.63 00
11 0 43.58 00
12 0 06.80 00
13 0 00.10 00
11 0 04.56 00
10 0 08.57 00
09 0 12.74 00
09 0 36.71 00
14 0 15.17 00
14 0 19.54 00
Latitude
(37 N)
09 0 20.99 00
08 0 27.11 00
07 0 49.12 00
07 0 19.96 00
10 0 44.62 00
13 0 06.90 00
15 0 51.78 00
19 0 34.02 00
20 0 25.13 00
21 0 11.03 00
21 0 36.37 00
24 0 03.14 00
27 0 05.67 00
Longitude
(73 E)
Terminus position
ASTER-derived parameters (7/23/2003)
4776
4620
4810
4819
4690
4541
4600
4718
4387
4804
4747
4645
4676
Z
(m)
02.2177
16.1968
03.8227
03.5664
06.1749
11.1563
09.4084
03.7308
20.4291
02.9600
02.4417
07.5597
03.1331
Area
(km2 )
11.0
14.2
NDC
04.8
30.2
36.7
11.4
07.2
03.3
NDC
06.0
03.9
06.9
Retreat b
(m yr1 )
Table 23.6. Sampled alpine glaciers over the Wakhan Pamir, Afghanistan. Glacier information extracted from ASTER multispectral data using WGS 84 Spheroid
and Datum. Average annual retreat rate estimates are from August 9, 1976 to July 23, 2003. Glacial debris coverage could not be estimated due to snow cover in
the upper ablation zones of the glaciers.
530
Remote sensing of glaciers in Afghanistan and Pakistan
b
a
AF5X14221006
AF5X14221007
AF5X14230029
AF5X14230038
AF5X14230072
AF5X14230073
AF5X14230076
AF5X14230086
AF5X14230091
AF5X14231001
AF5X14231002
AF5X14231003
AF5X14231004
AF5X14231005
AF5X14231006
AF5X14231007
—
—
Zemestan
Ptukh
—
—
—
—
—
—
—
—
—
—
—
—
NE
NE
S
E
SE
S
SW
S
SE
E
NE
NE
SE
E
SW
W
W
11 0 32.69 00
23 0 12.47 00
23 0 29.83 00
14 0 09.24 00
18 0 27.36 00
24 0 48.14 00
24 0 21.54 00
21 0 28.16 00
24 0 09.01 00
29 0 54.27 00
17 0 05.28 00
14 0 48.58 00
12 0 34.39 00
22 0 58.02 00
24 0 40.09 00
29 0 45.80 00
31 0 02.92 00
08 0 13.56 00
10 0 02.39 00
10 0 53.30 00
03 0 09.60 00
06 0 26.57 00
04 0 14.40 00
04 0 29.67 00
06 0 40.30 00
11 0 18.24 00
12 0 04.63 00
06 0 55.93 00
08 0 04.19 00
06 0 46.16 00
07 0 36.03 00
08 0 51.36 00
13 0 29.40 00
13 0 04.60 00
Glacier identification number follows World Glacier Inventory guidelines.
NDC ¼ no detectable change.
AF5X14221005
—
13 0 36.68 00
14 0 03.70 00
08 0 12.08 00
07 0 41.30 00
06 0 02.56 00
06 0 09.86 00
06 0 03.73 00
11 0 18.14 00
10 0 43.91 00
06.27.90 00
05 0 10.03 00
04 0 42.59 00
05 0 20.10 00
03 0 10.84 00
10 0 26.91 00
09 0 56.84 00
08 0 09.69 00
31 0 56.50 00
31 0 12.53 00
24 0 39.36 00
24 0 30.89 00
14 0 15.90 00
14 0 48.84 00
15 0 43.83 00
30 0 43.49 00
25 0 15.10 00
22 0 49.72 00
24 0 37.70 00
25 0 01.35 00
19 0 09.41 00
16 0 01.58 00
22 0 51.30 00
22 0 25.20 00
09 0 42.28 00
4540
4600
4849
4846
4875
4812
4745
4788
4872
4804
4835
4863
4911
4804
5005
4922
4857
03.8426
04.3370
02.1312
04.5936
06.8352
12.0400
03.1414
03.8039
03.2758
05.6955
01.4306
00.8426
07.1872
05.2316
01.4439
01.6293
04.5687
10.3
08.9
10.1
03.9
34.6
23.7
06.8
10.6
08.3
18.4
02.8
01.3
18.6
17.0
06.7
03.9
04.3
Case studies
531
532
Remote sensing of glaciers in Afghanistan and Pakistan
23.4.2.2
Figure 23.11. Modified version of fig. 21 of Shroder
and Weihs (2010) showing some of the recent variation
of glaciers in the Greater Pamir of the Wakhan Corridor.
The base map, transient snowlines, and the distribution
of debris-covered ice were taken from Exploration
Pamir 75 (Patzelt 1978) and none of these mapped
data reflect present-day conditions.
the range exhibits lower altitudes and consequently
smaller glaciers.
Very little is known about glacier fluctuations in
the region. Consequently, we assessed 85 glaciers
using multitemporal Landsat and ASTER imagery
acquired between 1972 and 2007. These glaciers
exhibited a variety of sizes, orientations, and altitudes. A sample of glacier fluctuation results are
presented in Table 23.7.
The vast majority of sampled glaciers ( 73%)
were found to be retreating, whereas 24% had
advanced and 3% exhibited no detectable change
in terminus position. Larger glaciers with lower
terminus altitudes exhibited greater retreat distances than smaller high-altitude glaciers. Climatic
patterns for the region are similar to those of the
Karakoram, although the magnitude of annual
precipitation is much less.
Central Karakoram
The glaciers in this region range from relatively
small to some of the largest in Pakistan (Table
23.8). They receive most of their accumulation in
the winter and spring from westerlies, although they
do receive summer accumulation due to penetration
of the monsoon (Wake and Mayewski 1993). They
exhibit significant debris cover. The larger glaciers
are found in major valleys and smaller glaciers are
contained in valleys oriented perpendicular to the
major valleys. Furthermore, the region exhibits
numerous surging glaciers (Hewitt 1969, 1988,
1998, Kargel et al. 2005, Copland et al. 2011).
Our change detection studies, coupled with fieldwork in the 1980s, 1990s, and 2005 indicate that
these glaciers are fluctuating differently, depending
on numerous factors such as location, debris cover,
and topographic variation, all of which strongly
control ablation and accumulation. Terminus fluctuations are highly varied and some small glaciers
are advancing. We present our longer term terminus
fluctuations based on a comparison of data from
Keyhole imagery acquired on August 4, 1973 and
ASTER imagery acquired on September 13, 2004.
Numerous glaciers exhibit downwasting patterns
and the development of supraglacial lakes. These
include Batura, Hispar, and Barpu. Batura Glacier
has retreated 430 m over the 31-year period and is
rapidly downwasting. Downwasting has resulted in
retreat of the white-ice stream (retreat rate of 91
m yr1 ), as the debris cover is redistributed due to
downwasting ice and exposure of englacial load.
The frequency and size of supraglacial lakes has
also increased, and outburst floods occur when
impounded water breaks through lateral moraine
deposits.
Hispar Glacier also shows geomorphological
evidence of retreat and rapid downwasting (Fig.
23.12). Near the terminus, thick supraglacial debris
cover retards supraglacial lake development,
although the terminus has retreated 238 m. At
higher altitudes in the ablation zone, shallow debris
cover and the presence of ice cliffs enhances
ablation and the development of supraglacial lakes.
This glacier also exhibits a surging tributary glacier
(Pumari Chhish). A few other large glaciers that we
did not sample, such as Chogo Lungma, also show
a similar surface-downwasting pattern.
The Hunza region, however, does not appear to
exhibit significant supraglacial lake development on
the numerous moderate to smaller glaciers there,
and proglacial lake development is minimal at the
Case studies
present time. We identified only one proglacial lake
at the terminus of Passu Glacier which has been
present for at least two decades (Fig. 23.13). This
lake has been involved in many minor but locally
destructive glacier lake outburst floods.
Hunza’s glaciers are retreating at variable rates
(Table 23.8). The highest retreat rates are associated
with moderate to large-size glaciers, and include
Batura, Gulmit, and Passu. Conversely, other large
glaciers such as Hispar, Barpu, and Bualtar exhibit
lower retreat rates. Over this time period 35% of the
glaciers exhibited terminus advance. There does not
appear to be a relationship between glacier orientation and advance. The highest advance rate was
associated with Gurpi Glacier which advanced
573 m (Fig. 23.14). This glacier receives its mass
from snow avalanches that produce an upper
glacier width of 800 m, whereas the average lower
terminus width is 300 m. This glacier has not been
reported to have surged in the past, and there is no
surficial evidence of surging. This suggests that the
advance may be caused by positive mass balance,
although flow dynamics such as increased basal slip
could also cause this to happen. Several other
glaciers, however, have advanced with no reported
history of surging. These include Ghulkin, Gorambar, Gujerab N, Karun SW, and Momhil Glaciers.
Yengutz Har Glacier also advanced 236 m,
although it is known to have surged in the past
(Fig. 23.15). Conversely, Bualtar Glacier has surged
in the past (Hewitt 1988), although it has retreated
180 m. The terminus is also downwasting,
whereas surge lobes, flow loops, and complex ice
fracture patterns clearly document its chaotic
behavior (Fig. 23.16). Other glaciers in this region
have also surged including Balt Bare, Yazghil, and
Khurdopin.
In the upper Hunza Valley region, the Batura and
Hispar Muztagh Mountains host Batura Glacier
(58 km) and Hispar Glacier (61 km), respectively,
which are among the longest of the world’s midlatitude glaciers. Numerous glaciers occur in this
region, including the well-known Passu, Ghulmet,
and Gulkin directly south of Batura, and some five
to ten surging glaciers in the Shimshal Valley north,
south, and west of the Hispar Mustagh Mountains.
Surging glaciers in Shimshal Valley are 10 km
long, but Balt Bare in the east and Pumari Chhish
in the south are only a few kilometers in length. In
the Lesser (South) Karakoram where the Rakaposhi-Haramosh range also generates glaciers in
the lower Hunza Valley region, most glaciers are
< 10 km long, with a few being 15 km in length.
533
Chogo Lungma Glacier, however, which is east of
Hunza and flows east-southeast, is 33 km in
length.
Glacier fluctuations in the K2 region of the
Himalaya–Karakoram–Hindu Kush buck the
worldwide trend by having a relatively high frequency of advancing glaciers and surging glaciers
(Copland et al. 2011, Bolch et al. 2012, see also
Chapter 24 of this book by Racoviteanu et al. on
the Himalaya). Our analysis has documented extensive surging of many glaciers in addition to those
identified and described by Hewitt (2007). Liligo
Glacier (Fig. 23.17) serves as a classic example,
where its terminus has reached Baltoro Glacier
and produced a large ephemeral proglacial lake that
periodically drains. Furthermore, satellite imagery
reveals that many glaciers exhibit stationary termini
or have advanced in this region. Our analysis indicates a temporal Karakoram anomaly of precipitation and advancement as first suggested by Hewitt
(2005) and later confirmed by Scherler et al. (2011)
and Gardelle et al. (2012).
23.4.2.3
Nanga Parbat Himalaya
We sampled 20 glaciers in the Nanga Parbat massif
(Table 23.9). These glaciers receive summer and
winter snow accumulation, are relatively small to
medium in size, and exhibit significant variability in
debris cover and depth, although debris cover percentages tend to be fairly high by world and Himalaya–Karakoram–Hindu Kush (HKH) standards.
Ice flow velocities also vary significantly given the
variations in glacier characteristics (Shroder et al.
1999). Furthermore, many of these glaciers are avalanche fed and do not exhibit a connected accumulation zone (e.g., Sachen, Patro, Lotang). They vary
in type from compound glacier systems in larger
valleys to simple alpine glaciers in confined subalpine basins, to small glacierets occuring along highaltitude ridges.
Our space-based assessments coupled with fieldwork from 1993 to 1997 indicate that, in general,
most of the larger glaciers have been retreating over
the past decade, although multitemporal change
detection studies reveal that several glaciers have
advanced over selected time periods. Collectively,
our results indicate the presence of terminus oscillations that could be related to ice flow dynamics
caused by earthquakes or avalanche periodicity,
or potentially to variations in mass balance. Here
we only report on long-term terminus fluctuations
based on comparison of data from Kick’s 1934
WGMS ID
PK5Q1314B103
PK5Q1314B100
PK5Q1314B099
PK5Q1314B098
PK5Q1314B097
PK5Q1314B096
PK5Q1314B095
PK5Q1314B094
PK5Q1314A093
PK5Q1314A088
Name
Madil
Shetor
Porarilio
Kotalkash
Koyo
Pechus
Chhateboi
Darkot
Barbin
Chiantar
Glacier
WNW
NW
NW
N
N
N
N
NNW
NNW
NW
Azimuth
Longitude
(73 E)
00 0 27.4079 00
05 0 31.1024 00
08 0 29.1202 00
12 0 00.0246 00
14 0 02.6684 00
15 0 48.7015 00
18 0 25.9713 00
22 0 18.4110 00
24 0 51.5295 00
46 0 27.5883 00
Latitude
(36 N)
41 0 08.5826 00
42 0 51.2363 00
44 0 09.9876 00
44 0 30.1084 00
45 0 28.4101 00
44 0 19.8832 00
44 0 22.1526 00
45 0 30.6242 00
47 0 26.5931 00
46 0 25.3418 00
Location
48 0 44.2089 00
49 0 12.2533 00
47 0 27.1747 00
48 0 54.4216 00
48 0 02.9843 00
47 0 25.9167 00
48 0 14.2598 00
47 0 46.8737 00
47 0 34.1944 00
42 0 56.0998 00
Latitude
(36 N)
73 33 0 18.3854 00
73 22 0 39.1557 00
73 20 0 00.6821 00
73 17 0 07.9865 00
73 15 0 25.0605 00
73 14 0 01.6763 00
73 11 0 29.6266 00
73 06 0 54.1111 00
73 03 0 08.3799 00
72 59 0 00.6941 00
Longitude
(73 E)
Terminus position
ASTER-derived parameters (2007)
3,686
3,912
3,668
3,344
3,398
3,599
3,471
3,902
3,244
3,851
Z
(m)
27.3
15.3
25.5
5.6
10.7
10.6
14.8
23.5
1.5
17.4
Rate a
(m yr1 )
Table 23.7. Sampled alpine glaciers over the Hindu Raj Region, Pakistan. Glacier information extracted from ASTER and Landsat multispectral data using WGS
84 spheroid and Datum. Average annual advance (þ) or retreat () rate is estimated between 1972 and 2007.
534
Remote sensing of glaciers in Afghanistan and Pakistan
a
NE
NW
W
PK5Q1314C106
PK5Q1314A090
East Ghamu Bar
West Ghamu Bar
Gazin
Chikzar
E
S
E
E
E
SW
SE
NE
Unnamed
Borum Bar
Aghost Bar
Ghalsapar
Mushk Bar
Unnamed
North Haiz
Chhateboi
NDC ¼ no detectable change.
EES
Unnamed
N
NE
East NE Ghamu Bar
23 0 46.3898 00
22 0 04.0389 00
19 0 02.7737 00
03 0 40.1396 00
32 0 42.5789 00
22 0 53.9256 00
13 0 39.7862 00
09 0 14.7278 00
06 0 32.5432 00
06 0 52.5174 00
05 0 42.6587 00
44 0 40.8652 00
53 0 49.8214 00
53 0 32.1344 00
36 0 20.4950 00
36 0 37.6001 00
37 0 28.2934 00
36 0 27.8464 00
44 0 22.4627 00
43 0 28.4497 00
36 0 28.4392 00
41 0 09.3106 00
37 0 36.1689 00
34 0 37.3675 00
32 0 41.7095 00
41 0 10.1006 00
39 0 53.4368 00
46 0 43.4624 00
50 0 25.5677 00
38 0 22.5366 00
39 0 44.2903 00
31 0 54.1204 00
33 0 46.3434 00
37 0 52.5617 00
39 0 25.1766 00
38 0 04.1642 00
42 0 36.7689 00
47 0 47.5244 00
36 0 21.4761 00
38 0 23.5353 00
38 0 32.7938 00
38 0 07.1356 00
73 55 0 12.2349 00
73 54 0 45.6266 00
73 43 0 42.3609 00
73 09 0 08.6547 00
73 08 0 29.8703 00
73 08 0 44.0410 00
73 08 0 52.7239 00
73 12 0 17.7680 00
73 24 0 29.4255 00
73 32 0 18.5027 00
72 59 0 15.5719 00
73 17 0 28.3110 00
73 24 0 25.5566 00
73 24 0 43.2979 00
3,535
4,113
3,750
3,367
3,445
3,297
3,911
3,516
3,501
3,811
3,371
3,664
2,720
2,939
NDC
4.9
3.6
12.6
13.0
3.4
6.2
24.9
15.9
4.3
6.3
9.7
17.2
13.6
Case studies
535
b
a
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
PK5Q
Barpu
Batura
Bualtar
Ghulkin
Ghutulji
Gorambar
Ghujerab N
Gurmit
Gurpi
Gutumi
Hispar
Karun E
Karun SW
Karun N
Lupghar Yaz
Momhil
Passu
Silking
Silking NE
Yengutz Har
This coordinate is 75 E.
This coordinate is 35 N.
WGMS ID
Name
Glacier
N?
NW
N
E
N
N
NW
SW
NE
W
NW
S
E
NW
N
N
E
N
E
NW
Azimuth
00
00
00
06 44.2188
0
00
12 0 04.2300 00
10 52.9932
0
28 07.5972
0
20 40.7868
0
00
57 53.9676
0
00
41 0 31.5240 00
40 29.5536
0
00
00a
42 47.5020
0
05 39.3252
0
00 0 46.5048 00a
00a
00a
23 0 03.9084 00
02 10.2948
0
05 59.5536
0
02 0 10.4856 00a
00
00
37 0 42.2688 00
35 47.3712
0
37 16.4424
0
00 0 00.0000 00
00
00
00 0 00.0000 00
44 28.8600
0
47 03.8940
0
37 0 48.3204 00
00
00
59 0 11.8500
22 08.4396
0
24 34.8264
0
05 0 31.1208 00a
00
00
40 0 06.3732 00
58 09.4548
0
44 09.5424
0
44 34.2060
00 0 15.5628 00
00
00
00
05 0 23.3556 00
23 13.1388
0
25 34.4820
0
08 59.8020
0
00 0 00.0000 00
00 0 00.0000 00
00
50 0 47.6304 00
09 0 41.0400 00
0
Longitude
(74 E)
Latitude
(36 N)
Location
00
00
00
00
00
00
00
00
00
00
09 21.0780
0
00
12 0 28.4580 00
13 20.5788
0
27 17.9208
0
30 22.9428
0
28 0 51.1248 00
40 0 19.4772 00
35 14.5032
0
37 42.4740
0
09 0 47.3076 00
01 0 09.8832 00
19 36.4584
0
23 57.8076
0
41 0 46.5360 00
07 0 26.6700 00
25 55.1064
0
24 56.9988
0
15 19.1664
0
30 0 41.2092 00
13 0 17.9076 00
Latitude
(36 N)
00
00
00
00
00
00a
00a
00
58 53.3856
0
00
40 0 44.4648 00
39 17.3088
0
00
00a
52 37.1676
0
02 58.1496
0
01 0 13.0440 00a
01 0 20.1720 00a
00 17.2944
0
08 43.3716
0
01 0 12.4464 00a
36 0 19.1700 00
44 19.2804
0
50 29.9652
0
04 0 35.1516 00a
00 0 49.0824 00a
58 31.0404
0
52 51.5424
0
45 33.2100
0
51 0 25.5312 00
47 0 22.3332 00
Longitude
(74 E)
Terminus position
3,561
4,420
3,275
2,583
2,916
3,424
4,272
4,657
4,151
3,097
3,229
3,032
2,938
4,456
3,862
3,984
2,465
2,295
2,630
2,830
Z
(m)
ASTER-derived parameters (9/13/2004)
09.5493
01.9872
16.4877
57.2541
74.1880
17.8732
15.3354
06.3383
08.5593
00.0000
09.3651
11.8647
13.8304
07.3985
21.2272
09.0981
29.6480
71.0609
0.0000
102.5525
Area
(km2 )
35
11
18
05
19
35
20
04
32
00
31
23
32
13
44
33
30
24
00
22
Debris
(%)
7.6
4.9
4.4
9.7
10.2
2.6
5.9
7.1
5.0
7.7
6.8
18.4
10.2
6.8
12.8
5.1
5.7
5.8
13.8
2.4
Rate
(m yr1 )
Table 23.8. Sampled alpine glaciers over the Batura and Hispar Muztagh regions in Pakistan. Glacier information extracted from ASTER multispectral data
using WGS 84 Spheroid and Datum. Average annual advance (þ) or retreat () rate is estimated between August 4, 1973 to September 13, 2004.
536
Remote sensing of glaciers in Afghanistan and Pakistan
Case studies
537
Figure 23.12. ASTER (bands 3, 2, 1, RGB) false-color composite image (September 13, 2004) of Hispar Glacier
in the Hispar Muztagh Mountains. Downwasting patterns vary along the main trunk of the glacier, with smaller
variations in debris thickness near the terminus, and highly variable debris thickness farther upglacier as ice cliffs and
supraglacial lakes enhance surface ablation. The tributary glacier on the far right is Pumari Chhish (arrow) which
surged in 1996.
Figure 23.13. Ground photograph of the terminus of Passu Glacier in Pakistan in 1984. The large proglacial lake
has resulted from terminus retreat and still remains to this day. Figure can also be viewed in higher resolution as
Online Supplement 23.4.
538
Remote sensing of glaciers in Afghanistan and Pakistan
Figure 23.14. ASTER (bands 3, 2, 1, RGB) falsecolor composite of extensively debris-covered Gurpi
Glacier in the Batura Muztagh Mountains. Glacier
accumulation is from frequent snow and ice avalanches
that produce a significant ice mass compared with the
lower ablation zone. Numerous avalanche lobes of
deposition can be seen above supraglacial debris cover.
Steep topography results in a highly reflective ice fall
connecting the upper and lower debris-covered portions of ice flow.
1:50,000-scale topographic map (from original
work of Richard Finsterwalder in 1934) and
ASTER imagery acquired on September 13, 2004.
Noteworthy visual changes in the character of
glacier surfaces include debris cover variation
and the development/evolution of proglacial lakes.
In general, the maximum altitude of debris
cover increased over the 1934–2004 period, which
suggests a possible increase in downwasting. Visual
examination of Landsat MSS imagery acquired
during the 1977 ablation season reveals that this
change occurred sometime before 1977. Consequently, supraglacial debris cover patterns have
not significantly changed within the past few decades. Furthermore, based on our fieldwork in the
1990s, terminus debris depths can range from 1 to
5 m.
Supraglacial lake development also appears to be
minimal. The existing supraglacial lakes are relatively small in size and are not widely distributed,
which further suggests minimal downwasting. Proglacial lake development/evolution has occurred in
only two locations. A proglacial lake ( 1:8 km2 )
has developed in front of Lichar Glacier on the
north side of the massif. In addition, the proglacial
lake associated with Tap Glacier on the south side
of Nanga Parbat has increased in size to 0:8 km2
(Fig. 23.18). Another lake ( 2:9 km2 ) has recently
appeared and is the result of a landslide that has
covered the terminus of Buldar Glacier and blocked
the Buldar Basin valley (Fig. 23.19).
It is difficult to compare ASTER-derived glacier
areas with the work of Kick (1980) and data in the
WGMS database because of methodological differences. Numerous tributary glaciers were partitioned
from the main trunk glaciers and given different ID
numbers. For the eight glaciers that can be effectively compared, several have not significantly
changed in area, and some differences are the result
of cartographic generalization used in producing
the 1934 map. Several glaciers, however, exhibit
significant areal differences, and these warrant discussion.
Buldar Glacier on the north side of Nanga Parbat
is a compound valley glacier that is avalanche fed
(Fig. 23.19). Its area in 1934 was 19:3 km2 versus
17:7 km2 in 2004. The change is due to significant
terminus retreat, and the retreat/disappearance of a
large tributary glacier that originated from Buldar
Ridge, and flowed into the westernmost tributary
glacier of Buldar Basin in 1934. Buldar RG2
Glacier on the eastern side of Buldar Basin has also
decreased in area from 1:16 to 0:83 km2 , although
there is no detectable change in terminus position.
This represents a change in terminus configuration,
as the glacier terminus is retreating more on the
south-facing portion of the basin compared with
the north. The most significant change in area anywhere on Nanga Parbat is associated with Rupal
Glacier (54:8 to 46:7 km2 ). The decrease in area is
primarily caused by terminus retreat which is estimated to be 13:9 m yr1 .
Other glaciers are also retreating at variable rates
(Table 23.9). The highest retreat rates are associated
Case studies
539
Figure 23.15. Ground photograph of Yengutz Har Glacier in Pakistan in 1984. It has a reported history of surging
and the steep terminus configuration depicts an advancing glacier.
Figure 23.16. Three-dimensional perspective (looking west) of Bualter Glacier using high-resolution satellite
imagery draped over a digital elevation model (Google Earth). Note the ice flow loops which are indicative of glacier
surging. Surge lobes occur (arrow) near the terminus. Barpu Glacier occurs in the lower right-hand portion of the
perspective.
540
Remote sensing of glaciers in Afghanistan and Pakistan
Figure 23.17. Liligo Glacier with proglacial lake in July 2005 on the south side of Baltoro Glacier (to the right).
Liligo is a surging glacier and light-colored trimlines document the redistribution of ice mass.
Figure 23.18. Tap Glacier with moraine-dammed lake in July 1996 (left foreground) on the south side of Nanga
Parbat. The debris-covered termini of Shigiri and Rupal Glaciers are located farther up the Rupal Valley.
Case studies
541
Figure 23.19. ASTER (bands 3, 2, 1, RGB) false-color composite of Buldar Glacier on the north side of Nanga
Parbat. A large landslide (arrows) has blocked the Buldar Valley floor, thereby altering glacier meltwater flow
c
b
a
PK5Q13021506
PK5Q13003207
PK5Q13003209
PK5Q13021504
PK5Q13002701
PK5Q13003101
PK5Q13003102
PK5Q13003302
PK5Q13020602
PK5Q13003211
PK5Q13003211
PK5Q13021514
PK5Q13021511
PK5Q13002803
PK5Q13003103
PK5Q13021515
PK5Q13020801
PK5Q13021509
PK5Q13021510
PK5Q13021507
Bazhin
Buldar
Buldar RG2
Chongra
Diamir
Ganalo NW
Ganalo NE
Lichar W
Lotang
Momacha 1a
Momacha 1b
Mazeno
Mazeno E
Patro
Raikot
Rupal
Sachen
Shaigiri
Shaigiri W
Tap
SE
S
SE
E
NE
N
NW
S
S
NW
NW
NE
NW
N
NE
NW
SE
NW
N
SE
Azimuth
NDC ¼ no detectable change.
Terminus position altered by landslide.
Estimate based on data from 1961 to 2004.
WGMS ID
Name
Glacier
00
13 01.5384
0
12 26.0568
0
12 35.5428
0
19 35.5764
0
08 43.4688
0
18 38.7828
00
00
00
00
00
00
35 27.5460
0
32 19.5396
0
33 14.8896
0
45 16.5672
0
29 50.0208
0
36 58.5180
0
00
00
00
00
00
00
31 0 32.9844 00
17 0 46.1112 00
0
31 0 25.9932 00
00
00
00
00
00
12 0 24.3900 00
43 14.5776
0
42 57.3624
0
44 14.2800
0
42 03.3084
0
34 47.2440
0
29 0 43.4544 00
00
00
00
00
00
00
34 0 10.6968 00
31 25.4532
0
12 0 31.1220 00
24 22.8966
0
24 12.2868
0
21 39.1608
0
26 20.1948
0
18 01.5948
0
19 0 09.4188 00
15 06.0336
0
41 0 17.8188 00
16 0 48.5148 00
00
43 0 26.7276 00
41 29.3208
0
23 0 01.8960 00
21 58.8780
00
37 0 47.8596 00
14 0 01.9140 00
0
Longitude
(74 E)
Latitude
(35 N)
Location
00
00
00
00
00
00
00
11 36.0420
0
11 10.2048
0
10 56.4168
0
21 03.3156
0
10 05.1564
0
22 35.5296
0
00
00
00
00
00
00
19 0 00.0624 00
10 0 42.8484 00
10 0 44.9112 00
24 31.5648
0
24 31.0536
0
22 24.0744
0
26 34.1232
0
19 46.6752
0
19 0 49.2852 00
17 13.2108
0
13 0 56.3340 00
23 0 11.7852 00
24 01.5768
0
11 0 44.7072 00
Latitude
(35 N)
00
00
00
00
00
00
00
36 53.5680
0
32 37.8312
0
35 26.1924
0
47 23.4312
0
34 00.8220
0
35 04.5348
0
00
00
00
00
00
00
30 0 16.0524 00
31 0 40.9152 00
30 0 07.6716 00
42 59.0544
0
42 49.0968
0
45 47.4192
0
41 58.3836
0
35 11.2308
0
34 0 52.2084 00
26 50.9568
0
43 0 10.5528 00
43 0 03.1944 00
40 56.2944
0
40 0 27.0192 00
Longitude
(74 E)
Terminus position
3,610
4,458
3,614
3,368
3,691
3,209
3,780
4,322
4,529
4,630
4,551
3,690
4,695
3,784
3,903
3,551
2,925
4,589
3,451
3,328
Z
(m)
ASTER-derived parameters (9/13/2004)
03.8458
02.5776
05.8878
10.7037
46.6955
36.9287
07.6239
03.6007
07.0400
00.2956
00.5993
03.8042
00.2728
06.2572
05.5423
37.5483
23.8561
00.8338
17.7770
18.1847
Area
(km2 )
18
32
37
86
39
14
30
25
42
10
12
68
15
39
37
23
34
09
36
36
Debris
(%)
1.0
2.7
3.0
1.0
13.9
3.1
7.3
6.0
1.1
10.5
7.1
2.5
2.0
0.5
NDC
9.7 c
8.6
NDC
17.6 b
12.1
Rate a
(m yr1 )
Table 23.9. Sampled alpine glaciers over the Nanga Parbat massif, Pakistan. Glacier information extracted from ASTER multispectral data using WGS 84
Spheroid and Datum. Average annual advance (þ) or retreat () rate is estimated between 1934 and 2004.
542
Remote sensing of glaciers in Afghanistan and Pakistan
Regional synthesis 543
with Bazhin, Buldar, Momacha 1a, and Rupal
Glaciers. Several glaciers, however, did not retreat
over this time period. Thirty-five percent of our
sample were found to exhibit no detectable change
or terminus advance. Advances were found with
northwest-facing glaciers (Diamir and Patro),
northeast orientation (Ganalo NE), and southern
orientations (Mazeno and Mazeno E). These
glaciers have not previously been reported as
surging, and we do not see any surge-type characteristics. It is difficult to explain these advances and
the highly varied advance/retreat rates over the
massif. Given a lack of climate and glaciological
data, we suspect that multiscale topographic effects
related to ablation and/or accumulation are primarily responsible.
23.5
REGIONAL SYNTHESIS
23.5.1 Afghanistan
Our combined historical assessment of the glaciers
of Afghanistan reveals several important observations. They have systematically been observed to be
downwasting and backwasting, and the glaciers
around the Kohi Bandakha massif in southern
Badakshan Province are more debris covered than
other regions in Afghanistan.
Glacier downwasting and retreat have been
observed over 25 years ago but such information
has not been readily available (Shroder and Bishop
2010a). The westernmost glaciers in Afghanistan in
the Kohi Baba range of the Hindu Kush located in
the central part of the country in Bamiyan Province
have been long observed to be severely declining in
mass (Kargel et al. 2005). Only one glacier in
Afghanistan, the one near Mir Samir discussed
above, has ever had any mass balance work done
on it, and it has now downwasted so much that it
has divided into two distinct ice masses, with the
westernmost part now appearing as a thin remnant
of what appears to be a largely stagnant ice mass.
In similar vein, our detailed measurements of the
glaciers in the Wakhan Pamir also show nearly
universal downwasting and backwasting, with the
establishment of many new supraglacial and proglacial lakes where ice has melted away. Some of
these may become hazardous in the future if their
moraine dams become unstable.
As noted in a study done three decades ago, but
only recently published (e.g., Shroder and Bishop
2010a), many glaciers around the Kohi Bandaka
massif were mapped as predominantly debris covered, even though the high altitude would allow for
much greater development of debris-free ice as is
common elsewhere in the Hindu Kush, Pamir, and
western Himalaya at similar altitudes. The exceptionally high seismicity beneath and directly around
this massif, now known to be among the highest in
the world (Wheeler and Rukstales, 2007), with > 20
earthquakes of magnitude > 7 in the past century
and a half, means that unusually large debris loads
are concentrated in the cirques from the copious
landslides produced by strong ground motion. In
combination with suspected long-term negative
mass balances and consequent debris cover
increases, especially in recent years as shown in
satellite imagery, the result has been the development of an unusually profuse continuum of debriscovered glacier forms. Thus we observe: (1) active
debris-covered glaciers with exposed semicircular
ice cliffs; (2) ice-cored moraines that have no
exposed ice cliffs, and that are either stagnant, or
which do not appear to have had much movement;
and (3) ice-cored rock glaciers with plentiful transverse ridges and furrows, as well as SFARs that
indicate renewed movement after periods in which
debris cover has built up such that no ice is exposed
in the lower parts. In contrast, elsewhere in northern Badakshan and in the Wakhan Pamir, such
features of extensively debris-covered ice are far less
common, and alpine glaciers are less debris covered.
23.5.2 Pakistan
We have made many new discoveries regarding the
nature of glacier fluctuations in Pakistan. Our analysis of satellite imagery has sampled a large number
of glaciers that vary in size and orientation, from
the eastern Hindu Kush to the Central Karakoram
in the northeastern part of the country. Our results
reveal that a large number of glaciers exhibit
retreating, stationary, or advancing termini. Our
data suggest the presence of a longitudinal gradient
in the frequency of advancing and retreating
glaciers, such that glaciers in the westernmost portion of Pakistan (Hindu Kush) appear to be dominantly retreating, whereas towards the east an
increasing number of glaciers have been observed
to be advancing, with the greatest number of advancing glaciers occuring in the Baltoro Mustagh
region, near K2 mountain. Detailed analysis of
these advancing glaciers reveals a large number of
non-surging glaciers and the spatiotemporal anomaly of glacier surging in the Karakoram (Hewitt
544
Remote sensing of glaciers in Afghanistan and Pakistan
2005, 2007, Copland et al. 2009, Quincey et al.
2009).
This longitudinal gradient in the frequency of
advancing glaciers is most likely related to climate
forcing. Previous climatological, hydrological, and
glaciological research suggests that the northeastern
portion of the country has received more precipitation, and that glaciers may be responding differently
to climate forcing compared with glaciers elsewhere
in the Himalaya. Our analysis of TRMM data and
ERA-40 climate reanalysis data clearly reveals
recent increases in the magnitude of precipitation
over time, from the Hindu Raj region to the K2
region. Furthermore, the magnitude of accumulated precipitation is greatest near K2. This indicates that a strong orographic precipitation effect is
at work and supports the results of Gardelle et al.
(2012) who report a slight mass gain of Karakoram
glaciers. Furthermore, temperature patterns from
climate station data and ERA-40 reanalysis data
reveal decreases in temperature during the summer
months. Analysis of ice flow velocity fields in the
Karakoram region support an interpretation of
climate forcing and relatively rapid glacier
response, as ice flow velocity anomalies have been
detected at altitude, and increased snow accumulation, abundant meltwater, and basal sliding can
produce terminus advances (Quincey et al. 2009).
We would not expect nonsurging glaciers to
respond this quickly to increased precipitation
given the deformation flow and longer response
times; however, topographic variations with altitude that govern surface irradiance can significantly
alter ablation rates and meltwater availability to
facilitate basal sliding.
The climate and glacier response patterns in this
region of the western Himalaya are unique. Given
the dominance of glacier retreat patterns in other
regions of the Himalaya, it is necessary to understand the climate dynamics in Pakistan. Clearly,
decreasing summer temperatures and increasing
snow accumulation trends raise doubt about the
intensification of the southwest monsoon as a forcing factor, as this would generate warmer summer
temperatures and more rainfall precipitation rather
than snow accumulation. Examination of precipitation data clearly reveals that the most northern
regions are dominated by precipitation from
westerlies. We postulate that increased precipitation
and decreased summer temperatures are a result of
ENSO-related variations, as they affect the monsoon and westerlies. Climate-modeling simulations
will further investigate this possibility.
23.6
ACKNOWLEDGMENTS
This work was funded by the National Aeronautics
and Space Administration under the NASA OES02 program (award NNG04GL84G), as well as a
grant from the National Academy of Sciences
(award PGA-P280421). ASTER data courtesy of
NASA/GSFC/METI/Japan Space Systems, the
U.S./Japan ASTER Science Team, and the GLIMS
project.
23.7
REFERENCES
Abers, G., and Bryan, C. (1988) Thrusting of the Hindu
Kush over the southeastern Tadjik Basin, Afghanistan:
Evidence from two large earthquakes. Tectonics, 7(1),
41–56.
Aizen, V.B., Kuzmichenok, V.A., Surazakov, A.B., and
Aizen, E.M. (2006) Glacier changes in central and
northern Tien Shan during the last 140 years based
on surface and remote sensing data. Annals of Glaciology, 43, 1–13.
Auden, J.B. (1935) The snout of the Biafo Glacier in
Baltistan. Geological Survey of India Records, 68,
400–413.
Bamber, J.L., and Payne, A.J. (Eds.) (2004) Mass Balance
of the Cryosphere: Observations and Modelling of Contemporary and Future Changes, Cambridge University
Press, Cambridge, U.K.
Barsch, D. (1987a) The problem of the ice-cored rock
glacier. In: J.R. Giardino, J.F. Shroder Jr., and J. Vitek
(Eds.), Rock Glaciers, Allen & Unwin, Boston, MA,
pp. 45–54.
Barsch, D. (1987b) Rock glaciers: An approach to their
systematics. In: J.R. Giardino, J.F. Shroder Jr., and
J. Vitek, Rock Glaciers, Allen & Unwin, Boston, MA,
pp. 41–44.
Barsch, D. (1996) Rockglaciers: Indicators for the Present
and Former Geoecology in High Mountain Environments, Springer-Verlag, Berlin.
BCIG (1979) The Batura Glacier in the Karakoram
Mountains and its variations. Scientia Sinica, 22,
958–974 [Batura Glacier Investigation Group].
BCIG (1980) Karakoram Batura Glacier: Exploration and
Research, Science Press, Beijing [Batura Glacier Investigation Group].
Baudo, R., Tartari, G., Vuillermoz, E., and Shroder, J.F.,
Jr. (Eds.) (2007) Mountains: Witnesses of Global
Changes, Vol. 10: Research in the Himalayas and Karakoram, Elsevier Science, Amsterdam, The Netherlands.
Berthier, E., Arnaud, Y., Rajesh, K., Ahmad, S.,
Wagnon, P., and Chevallier, P. (2007) Remote sensing
estimates of glacier mass balances in the Himachal
References 545
Pradesh (Western Himalaya, India). Remote Sensing of
Environment, 108(3), 327–338.
Bishop, M.P., Shroder, J.F., Jr., and Ward, J.L. (1995)
SPOT multispectral analysis for producing supraglacial debris-load estimates for Batura Glacier, Pakistan. Geocarto International, 10(4), 81–90.
Bishop, M.P., Shroder, J.F., Jr., Hickman, B.L., and
Copland, L. (1998) Scale dependent analysis of satellite
imagery for characterization of glacier surfaces in the
Karakoram Himalaya. Geomorphology, 21, 217–232.
Bishop, M.P., Shroder, J.F., Jr., and Hickman, B.L.
(1999) High resolution satellite imagery and neural
networks for information extraction in a complex
mountain environment. Geocarto International, 14(2),
17–26.
Bishop, M.P., Kargel, J.S., Kieffer, H.H., MacKinnon,
D.J., Raup, B.H., and Shroder, J.F., Jr. (2000) Remote
sensing science and technology for studying glacier
processes in high Asia. Annals of Glaciology, 31,
164–170.
Bishop, M.P., Shroder, J.F., Jr., and Colby, J.D. (2003)
Remote sensing and geomorphometry for studying
relief production in high mountains. Geomorphology,
55, 345–361.
Bishop, M.P., Barry, R.G., Bush, A.B.G., Copland, L.,
Dwyer, J.L., Fountain, A.G., Haeberli, W., Hall,
D.K., Kääb, A., Kargel, J.S. et al. (2004) Global
land-ice measurements from space (GLIMS): Remote
sensing and GIS investigations of the Earth’s cryosphere. Geocarto International, 19(2), 57–84.
Bishop, M.P., Shroder, J.F., Jr., Haritashya, U.K., and
Bulley, H.N.N. (2007) Remote sensing and GIS for
alpine glacier change detection in the Himalaya. In:
R. Baudo, G. Tartari, E. Vuillermoz, and J.F. Shroder
Jr. (Eds.), Mountains: Witnesses of Global Changes,
Vol. 10: Research in the Himalayas and Karakoram,
Elsevier Science, Amsterdam, The Netherlands.
Bolch, T., Kulkarni, A., Kääb, A., Huggel, C., Paul, F.,
Cogley, J.G., Frey, H., Kargel, J.S., Fujita, K., Scheel,
M. et al. (2012) The state and fate of Himalayan
glaciers. Science, 336, 310–314.
Braslau, D. (1972) The glaciers of Keshnikhan. In: K.
Gratzl (Ed.), Hindukusch-Osterreichische Forschungs
Expedition in den Wakhan 1970, Akademische Drucku. Verlagsanstalt, Graz, Austria, pp. 112–116 [in
German].
Breckle, S.W., and Frey, W. (1976a) Beobachtungen zur
heutigen vergletscherung der hauptkette des zentralen
hindukusch. Afghanistan Journal, 3(3), 95–100 [in
German].
Breckle, S.W., and Frey, W. (1976b) Die höchsten Berge
im zentralen Hindukusch. Afghanistan Journal, 3(3),
91–94 [in German].
Copland, L., Pope, S., Bishop, M.P., Shroder, J.F.,
Clendon, P., Bush, A.B.G., Kamp, U., Seong, Y.B.,
and Owen, L.A. (2009) Glacier velocities across the
central Karakoram. Annals of Glaciology, 50(52), 41–
49.
Copland, L., Sylvestre, T., Bishop, M.P., Shroder, J.F.,
Seong, Y.B., Owen, L.A., Bush, A.B.G., and Kamp, U.
(2011) Expanded and recently increased glacier surging
in the Karakoram. Arctic, Antarctic, and Alpine
Research, 43(4), 503–516.
Cornwell, K. (1998) Quaternary break-out flood sediments in the Peshawar basin of northern Pakistan.
Geomorphology, 25, 225–248.
de Grancy, R.S., and Kostka, R. (Eds.) (1978) Grosser
Pamir, Akademische Druck Verlagsanstalt, Graz,
Austria [in German].
Diolaiuti, G., Pecci, M., and Smiraglia, C. (2003) Liligo
Glacier, Karakoram, Pakistan: A reconstruction of the
recent history of a surge-type glacier. Annals of Glaciology, 36(1), 168–172.
Dyurgerov, M.B., and Meier, M.F. (2004) Glaciers and
the study of climate and sea-level change. In: J.L.
Bamber and A.J. Payne (Eds.), Mass Balance of the
Cryosphere: Observations and Modelling of Contemporary and Future Changes, Cambridge University Press,
Cambridge, U.K., pp. 579–621.
Finsterwalder, R. (1937) Die gletscher des Nanga Parbat.
Zeitschrift für Gletscherkunde, 25, 57–108 [in German].
Finsterwalder, R. (1960) German glaciological and geological expeditions to the Batura Mustagh and
Rakaposhi Range. Journal of Glaciology, 3(28), 787–
788.
Fowler, H.J., and Archer, D.R. (2006) Conflicting signals
of climate change in the upper Indus Basin. Journal of
Climate, 19, 4276–4293.
Gardelle, J., Berthier, R., and Arnaud, Y. (2012) Slight
mass gain of Karakoram glaciers in the early twentyfirst century. Nature Geoscience, 1, 1–4.
Gardner, J.S. (1986) Recent fluctuations of Rakhiot
Glacier, Nanga Parbat, Punjab Himalaya, Pakistan.
Journal of Glaciology, 32(112), 527–529.
Gardner, J.S., and Jones, N.K. (1993) Sediment transport
and yield at the Raikot Glacier, Nanga Parbat, Punjab
Himalaya. In: J.F. Shroder Jr. (Ed.), Himalaya to the
Sea: Geology, Geomorphology and the Quaternary,
Routledge, London, pp. 184–197.
Giardino, J.R., Shroder, J.F., Jr., and Vitek, J. (1987)
Rock Glaciers, Allen & Unwin, Boston, MA.
Gilbert, O., Jamieson, D., Lister, H., and Pendlington, A.
(1969) Regime of an Afghan glacier. Journal of Glaciology, 8(52), 51–65.
Glicken, M. (1960) Making a map of Afghanistan. Photogrammetric Engineering, 26(5), 743–745.
Goudie, A.S., Brunsden, D.N., Collins, D.N., Derbyshire, E., Ferguson, R.I., Hashmet, Z., Jones,
D.K.C., Perrot, F.A., Said, M., Waters, R.S. et al.
(1984) The geomorphology of the Hunza Valley, Karakoram Mountains, Pakistan. In: K.J. Miller (Ed.),
The International Karakoram Project, Cambridge University Press, Cambridge, U.K., pp. 359–410.
546
Remote sensing of glaciers in Afghanistan and Pakistan
Grötzbach, E. (1990) Afghanistan: Eine geographische
Landeskunde (Wissenschaftliche Länderkunden), Wissenschaftliche Buchgesellschaft, Darmstadt, Germany
[in German].
Haeberli, W. (2004) Glacier and ice caps: Historical background and strategies of world-wide monitoring. In:
J.L. Bamber and A.J. Payne (Eds.), Mass Balance of
the Cryosphere: Observations and Modelling of Contemporary and Future Changes, Cambridge University
Press, Cambridge, U.K., pp. 559–578.
Haeberli, W., Hoelzle, M., and Suter, S. (Eds.) (1998) Into
the Second Century of Worldwide Glacier Monitoring:
Prospects and Strategies, UNESCO, Paris.
Haritashya, U.K., Bishop, M.P., Shroder, J.F., Jr., Bush,
A.B.G., and Bulley, H.N.N. (2009) Space-based assessment of glacier fluctuations in the Wakhan Pamir,
Afghanistan. Climatic Change, 94, 5–18.
Hasnain, S.I. (2007) Shrinking cryosphere in South Asia.
In: R. Baudo, G. Tartari, E. Vuillermoz, and J.F.
Shroder Jr. (Eds.), Mountains: Witnesses of Global
Changes, Vol. 10: Research in the Himalayas and
Karakoram, Elsevier Science, Amsterdam, The Netherlands..
Hewitt, K. (1969) Glacier surges in the Karakoram Himalaya (Central Asia). Canadian Journal of Earth
Sciences, 6, 1009–1018.
Hewitt, K. (1988) Catastrophic landslide deposits in the
Karakoram Himalaya. Science, 242, 64–67.
Hewitt, K. (1998) Glaciers receive a surge of attention in
the Karakoram Himalaya. EOS Transactions, 79, 104–
105.
Hewitt, K. (2005) The Karakoram anomaly? Glacier
expansion and the ‘‘elevation effect’’, Karakoram
Himalaya. Mountain Research and Development,
25(4), 332–340.
Hewitt, K. (2007) Tributary glacier surges: An exceptional concentration at Panmah Glacier, Karakoram
Himalaya. Journal of Glaciology, 53, 181–188.
Johnson, P.G. (1987) Rock glacier: Glacier debris systems
or high-magnitude low-frequency flows? In: J.R.
Giardino, J.F. Shroder Jr., and J. Vitek, Rock Glaciers,
Allen & Unwin, Boston, MA, pp. 175–192.
Kääb, A., and Reichmuth, T. (2005) Advance mechanisms of rock glaciers. Permafrost and Periglacial Processes, 16, 187–193.
Kargel, J.S., Abrams, M.J., Bishop, M.P., Bush, A.,
Hamilton, G., Jiskoot, H., Kääb, A., Kieffer, H.H.,
Lee, E.M., Paul, F. et al. (2005) Multispectral imaging
contributions to global land ice measurements from
space. Remote Sensing of Environment, 99, 187–219.
Kazmi, A.H., and Rana, R.A. (1982) Tectonic map of
Pakistan. Geological Survey of Pakistan, 1, 1.
Khromova, T.E., Osipova, G.B., Tsvetkov, D.G., Dyurgerov, M.B., and Barry, R.G. (2006) Changes in glacier
extent in the eastern Pamir, Central Asia, determined
from historical data and ASTER imagery. Remote
Sensing of Environment, 102, 24–32.
Kick, W. (1980) Material for a glacier inventory of the
Indus drainage basin: The Nanga Parbat massif. In:
World Glacier Inventory, Proceedings of the Ridderalp
Workshop (IAHS-AISH Publication No. 126), International Association of Hydrological Sciences,
Rennes, France, pp. 105–109.
Kotlyakov, V.M. (Ed.) (1996) Maps of the Hindu Kush
and Karakoram, The World Atlas of Snow and Ice
Resources, Institute of Geography, Russion Academy
of Sciences, Moscow, pp. 233–238.
Kravtsova, V.I. (1990) Snow cover mapping of Afghanistans mountains with space imagery. Mapping Sciences
and Remote Sensing, 27(4), 295–302.
Loewe, F. (1959) Some observations of the radiation
budget and of the ablation of glacier ice in the Nanga
Parbat region. Pakistan Journal of Science, 11(5), 227–
236.
Maksimov, E.V., and Perugina, N.N. (1975) Opyt glaciomorformetricheskogo analiza kart Hindu Kush (transcribed from Cyrillic). Vsesoiuznoe Geograficheskoe
Obshchestvo Izvestiia, 107(3), 238–243 [in Russian].
Mason, K. (1930) The glaciers of the Karakoram and
neighborhood. Geological Survey of India Records,
63, 214–278.
Mayer, C., Lambrecht, A., Belò, M., Smiraglia, C., and
Diolaiuti, G. (2006) Glaciological characteristics of the
ablation zone of Baltoro Glacier, Karakoram, Pakistan. Annals of Glaciology, 43(1), 123–131.
Mayewski, P.A., and Jeschke, P.A. (1979) Himalayan and
trans-Himalayan glacier fluctuations since ad 1812.
Arctic, Antarctic, and Alpine Research, 11, 267–287.
Mihalcea, C., Mayer, C., Diolaiuti, G., Lambrecht, A.,
Smiraglia, C., and Tartari, G. (2006) Ice ablation and
meteorological conditions on the debris-covered area
of Baltoro Glacier, Karakoram, Pakistan. Annals of
Glaciology, 43(1), 292–300.
Miller, K.J. (Ed.) (1984) The International Karakoram
Project, Cambridge University Press, Cambridge, U.K.
Mott, P.G. (1950) Karakoram survey, 1939: A new map.
Geographical Journal, 116(1/3), 89–95.
Osipova, G.B., and Tsvetkov, D.G. (2002) Investigations
of complex glacier dynamics by space images. News of
Academy of Sciences; Geographical Series, 3, 29–38.
Osipova, G.B., and Tsvetkov, D.G. (2003) Monitoring
of surging glaciers: What does it give to us? Priroda
(Nature), 4, 3–13.
Owen, L.A., Finkel, R.C., Caffee, M.W., and Gualtieri,
L. (2002) Timing of multiple late Quaternary glaciations in the Hunza Valley, Karakoram Mountains,
northern Pakistan: Defined by cosmogenic radionuclide dating of moraines. Geological Society of
America Bulletin, 114, 593–604.
Owen, L.A., Finkel, R.C., Barnard, P.L., Ma, H., Asahi,
K., Caffee, M.W., and Derbyshire, E. (2005) Climatic
and topographic controls on the style and timing of
Late Quaternary glaciation throughout Tibet and the
Himalaya defined by 10 Be cosmogenic radionuclide
References 547
surface exposure dating. Quaternary Science Reviews,
24, 1391–1411.
Patzelt, G. (1978) Gletscherkundliche Untersuchungen
im Grossen Pamir (Glacier research in the Greater
Pamir). In: R.S. de Grancy and R. Kostka (Eds.),
Grosser Pamir, Akademische Druck Verlagsanstalt,
Graz, Austria, pp. 131–149 [in German].
Paul, F., Kääb, A., and Haeberli, W. (2007) Recent
glacier changes in the Alps observed by satellite: Consequences for future monitoring strategies. Global and
Planetary Change, 56, 111–122.
Pillewizer, W. (1956) Der Rakhiotgletscher am Nanga
Parbat im Jahr 1954. Zeitschrift für Gletscherkunde
und Glazialgeologie, 3(2), 181–199 [in German].
Quincey, D.J., Copland, L., Mayer, C., Bishop, M.P.,
Luckman, A., and Belo, M. (2009) Ice velocity and
climate variations for Baltoro Glacier, Pakistan. Journal of Glaciology, 55(194), 1061–1071.
Roohi, R. (2007) Research on global changes in Pakistan.
In: R. Baudo, G. Tartari, E. Vuillermoz, and J.F.
Shroder Jr. (Eds.), Mountains: Witnesses of Global
Changes, Vol. 10: Research in the Himalayas and
Karakoram, Elsevier Science, Amsterdam, The Netherlands.
Roohi, R., Ashraf, A., Hussain, S., Naz, R., Mool, P.K.,
and Bajracharya, S.R. (2005) Farm Mechanization
Options under Climate Change Scenarios in Pakistan
(technical report), Water Resources Research Institute,
National Agricultural Research Center, Islamabad,
Pakistan.
Ruleman, C.A., Crone, A.J., Machette, M.N., Haller, K.
M., and Rukstales, K.S. (2007) Map and Database of
Probable and Possible Quaternary Faults in Afghanistan
(USGS Open-File Report No. 2007-1103), U.S. Geological Survey, Reston, VA.
Scherler, D., Bookhagen, B., and Strecker, M.R. (2011)
Spatially variable response of Himalayan glaciers to
climate change affected by debris cover. Nature
Geoscience, 4, 156–159.
Seong, Y.B., Owen, L.A., Bishop, M.P., Bush, A.B.G.,
Clendon, P., Copland, L., Finkel, R., Kamp, U., and
Shroder, J.F., Jr. (2007) Quaternary glacial history of
the Central Karakoram. Quaternary Science Reviews,
26, 3384–3405.
Shareq, A., Chmyriov, W.M., Alekseev, K.F., Dronov,
V.I., Gannon, P.J., Lubemov, B.K., Karfarskiy, A.K.,
Malyarov, E.P., and Rossovskiy, L.N. (1977) Mineral
Resources of Afghanistan, Second Edition (UNDP
Project No. AFG/74/012), United Nations, New York
[United Nations Development Program].
Shi, Y., and Wang, W. (1980) Research on snow cover in
China and the avalanche phenomena of Batura Glacier
in Pakistan. Journal of Glaciology, 26(94), 25–30.
Shroder, J.F., Jr. (1980) Special problems of glacial inventory in Afghanistan. In: World Glacial Inventory Proceedings Reideralp Workshop, September 1978 (IAHSAISH Publication No. 126), International Association
of Hydrological Sciences, Rennes, France, pp. 142–
147.
Shroder, J.F., Jr. (1987) Rock glaciers and slope failures:
High plateaus and La Sal Mountains, Colorado Plateau, Utah, USA. In: J.R. Giardino, J.F. Shroder Jr.,
and J. Vitek, Rock Glaciers, Allen & Unwin, Boston,
MA, pp. 193–238.
Shroder, J.F., Jr. (1989) Glacierized areas of Afghanistan.
In: W. Haeberli, H. Bosch, K. Scherler, G. Østrem, and
C.C. Wallen (Eds.), World Glacial Inventory, Status
1988, IAHS(ICSI)-UNEP-UNESCO, Teufen, Switzerland, pp. C39–C40 [International Commission on
Snow and Ice].
Shroder, J.F., Jr. (2011a) Himalaya. In: V.P. Singh,
P. Singh, and U.K. Haritashya (Eds.), Encyclopedia
of Snow, Ice and Glaciers, Springer-Verlag, Dordrecht,
The Netherlands, pp. 510–520.
Shroder, J.F., Jr. (2011b) Hindu Kush. In: V.P. Singh,
P. Singh, and U.K. Haritashya (Eds.), Encyclopedia of
Snow, Ice and Glaciers, Springer-Verlag, Dordrecht,
The Netherlands, pp. 523–525.
Shroder, J.F., Jr., and Bishop, M.P. (2010a) Glaciers of
Afghanistan. In: R.S. Williams Jr. and J.G. Ferrigno
(Eds.), Satellite Image Atlas of Glaciers of the World:
Asia (USGS Professional Paper No. 1386-F), U.S.
Geological Survey, Reston, VA, pp. F167–F199.
Shroder, J.F., Jr., and Bishop, M.P. (2010b) Glaciers of
Pakistan. In: R.S. Williams Jr. and J.G. Ferrigno
(Eds.), Satellite Image Atlas of Glaciers of the World:
Asia (USGS Professional Paper No. 1386-F), U.S.
Geological Survey, Reston, VA, pp. F201–F257.
Shroder, J.F., Jr., and Sewell, R.E., Jr. (1985) Mass
Movement in the La Sal Mountains, Utah (Special
Studies Volume), Utah Geological and Mineral Survey, Salt Lake City.
Shroder, J.F., and Weihs, B.J. (2010) Geomorphology of
the Lake Shewa landslide dam, Badakshan, Afghanistan, using remote sensing data. Geografiska Annaler,
92A(4), 471–486.
Shroder, J.F., Jr., Khan, M.A., Lawrence, R.D., Madin,
I.P., and Higgins, S.E. (1989) Quaternary glacier
chronology and neotectonics in the Himalaya of northern Pakistan. In: L.L. Malinconico and R.J. Lillie
(Eds.), Tectonics of the Western Himalayas (GSA
Special Paper No. 232), Geological Society of America,
Boulder, CO, pp. 275–294.
Shroder, J.F., Jr., Scheppy, R., and Bishop, M.P. (1999)
Denudation of small alpine basins, Nanga Parbat
Himalaya. Arctic, Antarctic, and Alpine Research,
31(2), 121–127.
Singh, V.P., Singh, P., and Haritashya, U.K. (Eds.)
(2011) Encyclopedia of Snow, Ice and Glaciers,
Springer-Verlag, Dordrecht, The Netherlands.
Smiraglia, C., Mayer, C., Mihalcea, C., Diolaiuti, G.,
Belò, M., and Vassena, G. (2007) Ongoing variations
of Himalayan and Karakoram glaciers as witnesses of
global changes: Recent studies of selected glaciers. In:
548
Remote sensing of glaciers in Afghanistan and Pakistan
R. Baudo, G. Tartari, E. Vuillermoz, and J.F. Shroder
Jr. (Eds.), Mountains: Witnesses of Global Changes,
Vol. 10: Research in the Himalayas and Karakoram,
Elsevier Science, Amsterdam, The Netherlands.
Treydte, K.S., Schleser, G.H., Helle, G., Frank, D.C.,
Winiger, M., Haug, G.H., and Esper, J. (2006) The
twentieth century was the wettest period in northern
Pakistan over the last millennium. Nature, 440, 1179–
1182.
Vitek, J.D., and Giardino, J.R. (1987) Rock glaciers: A
review of the knowledge base. In: J.R. Giardino, J.F.
Shroder Jr., and J. Vitek (Eds.), Rock Glaciers, Allen &
Unwin, Boston, MA, pp. 1–26.
Wake, C.P., and Mayewski, P.A. (1993) Regional distribution of monsoon and desert dust signals recorded in
Asian glaciers. Geophysical Research Letters, 20(14),
1411–1414.
Wang, W., Huang, M., and Chen, J. (1984) A surging
advance of Balt Bare Glacier. In: K.J. Miller (Ed.), The
International Karakoram Project, Cambridge University Press, Cambridge, U.K., pp. 76–83.
Whalley, W.B., and Azizi, F. (2003) Rock glaciers and
protalus landforms: Analogous forms and ice sources
on Earth and Mars. Journal of Geophysical Research,
108, doi: 10.1029/2002JE001864.
Whalley, W.B., and Martin, H.E. (1992) Rock glaciers,
II: Models and mechanisms. Progress in Physical Geography, 16, 127–186.
Wheeler, R.L., and Rukstales, R.S. (2007) Seismotectonic
Map of Afghanistan and Adjacent Areas (USGS OpenFile Report No. 2007-1104, Afghanistan Project Product No. 154), U.S. Geological Survey, Reston, VA.
Williams, R.S., Jr., and Ferrigno, J.G. (Eds.) (2010)
Satellite Image Atlas of Glaciers of the World: Asia
(USGS Professional Paper No. 1386-F), U.S.
Geological Survey, Reston, VA.