ArticlePDF Available

Abstract and Figures

Thinning rates for the debris-covered Gangotri Glacier and its tributary glaciers during the period 1968-2014, length variation and area vacated at the snout from 1965-2015, and seasonal variation of ice surface velocity for the last two decades have been investigated in this study. It was found that the mass loss of Gangotri and its tributary glaciers was slightly less than those reported for other debris-covered glaciers in the Himalayan regions. The average velocity during 2006-2014 decreased by 6.7% as compared to that during 1993-2006. The debris-covered area of the main trunk of Gangotri Glacier increased significantly from 1965 until 2015 with the maximum rate of increase (0.8 ± 0.2 km2a−1) during 2006-2015. The retreat (9.0 ± 3.5 ma−1) was less in recent years (2006-2015) but the down-wasting (0.34 ± 0.2 m a−1) in the same period (2006-2014) was higher than that (0.20 ± 0.1 m a−1) during 1968-2006. The study reinforced the established fact that glacier retreat is a delayed response to climate change, whereas glacier mass balance is a more direct and immediate response. Therefore, it is recommended to study the glacier mass balance and not only the glacier extent, to conclude about any glacier’s response to climate change.
Content may be subject to copyright.
Overall recession and mass budget of Gangotri Glacier, Garhwal
Himalayas, from 1965 to 2015 using remote sensing data
ATANU BHATTACHARYA,
1
*TOBIAS BOLCH,
1,2
*KRITI MUKHERJEE,
1
TINO PIECZONKA,
1
JAN KROPÁČEK,
3
MANFRED F. BUCHROITHNER
1
1
Institut für Kartography,Technische Universität Dresden,Dresden,Germany
2
Department of Geography,University of Zürich Irchel,Zürich,Switzerland
3
Department of Geosciences,University of Tübingen,Tübingen,Germany
*Correspondence: Atanu Bhattacharya; Tobias Bolch <atanudeq@gmail.com;tobias.bolch@geo.uzh.ch>
ABSTRACT. Thinning rates for the debris-covered Gangotri Glacier and its tributary glaciers during the
period 19682014, length variation and area vacated at the snout from 1965 to 2015, and seasonal vari-
ation of ice-surface velocity for the last two decades have been investigated in this study. It was found
that the mass loss of Gangotri and its tributary glaciers was slightly less than those reported for other
debris-covered glaciers in the Himalayan regions. The average velocity during 200614 decreased by
6.7% as compared with that during 19932006. The debris-covered area of the main trunk of
Gangotri Glacier increased significantly from 1965 until 2015 with the maximum rate of increase
(0.8 ± 0.2 km
2
a
1
) during 200615. The retreat (9.0 ± 3.5 m a
1
) was less in recent years (2006
2015) but the down-wasting (0.34 ± 0.2 m a
1
) in the same period (20062014) was higher than that
(0.20 ± 0.1 m a
1
) during 19682006. The study reinforced the established fact that the glacier length
change is a delayed response to climate change and, in addition, is affected by debris cover, whereas
glacier mass balance is a more direct and immediate response. Therefore, it is recommended to study
the glacier mass balance and not only the glacier extent, to conclude about a glaciers response to
climate change.
KEYWORDS: corona and hexagon data, glacier mass balance, glacier retreat, glacier surface velocity
1. INTRODUCTION
The Himalayan region has one of the largest concentrations
of glaciers outside of the polar regions with a glacier cover-
age (including the Karakoram) of 40 800 km
2
(Bolch and
others, 2012). Himalayan glaciers are of interest for several
reasons. Water discharge from Himalayan glaciers contributes
to the overall Himalayan river runoff (Immerzeel and others,
2010) and the precipitation, along with snow and ice melt,
also affect the runoff considerably (Bhambri and others,
2011a). Singh and others (2008) and Bhambri and others
(2011a) reported that on an average, yearly snow and
glacier melt contributed 97% of water, measured at
Bhojbasa (4 km downstream from Gangotri Glacier snout
and 3780 m a.s.l.) to the Ganga Basin near the terminus
of Gangotri Glacier. Although far from the terminus of
Gangotri Glacier (Kaser and others, 2010), this percentage
decreases. In addition water discharge from Himalayan gla-
ciers is also important for irrigation and hydropower generation
(Singh and others, 2009). Gangotri Glacier is the largest
glacier in terms of length (30 km) and area (144 km
2
)in
the Garhwal Himalayas (Srivastava, 2012). The Bhagirathi
River originates from the snout (Gaumukh 3950 m a.s.l.)
of Gangotri Glacier, which is the main source of the River
Ganges. The glacier originates from the Chaukhamba
group of peaks (68537138 m a.s.l.) and flows northwest
towards Gaumukh (Bhambri and others, 2012). Gangotri
Glacier is one of the most sacred shrines in India, with
immense religious significance. Being the main source of
the River Ganges, the most sacred river to the Hindus, it
attracts thousands of pilgrims every year.
High resolution multitemporal and multispectral satellite
data have abundant potential to study the glaciers in terms
of extent, surface properties, surface velocity and temporal
mass balance (Bolch and others, 2010). Declassified
imagery such as Corona and Hexagon has proven to be espe-
cially useful data source for mapping historic extents of gla-
ciers and generation of digital terrain model (DTM) for
mass-balance studies (Surazakov and Aizen, 2010;
Pieczonka and others, 2011; Holzer and others, 2015;
Pellicciotti and others, 2015). These declassified imageries
can also be used for comparisons with glacier outlines
derived from topographic maps (Bhambri and Bolch, 2009).
Various methods have been applied for the on-going
mapping and monitoring of the Gangotri Glacier, which
have resulted in considerable differences in glacier retreat
rates (e.g. Srivastava, 2004; Kumar and others, 2008;
Bhambri and Chaujar, 2009; Bhambri and others, 2012).
The snout position of Gangotri Glacier was mapped first by
Auden, (1937) in 1935 using a plane table survey at a scale
of 1:4800. It was postulated from various geomorphological
features that the glacier retreated at a rate of 7.35 m a
1
from
1842 to 1935. Subsequent surveys were conducted by the
Geological Survey of India (GSI) to measure the retreat rate
of the Gangotri Glacier snout. Inherent inconsistency and un-
certainty in different methods are however still major issues.
Therefore, regular consistent monitoring is important for
Journal of Glaciology (2016), 0(0) 119 doi: 10.1017/jog.2016.96
©The Author(s) 2016. This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives
licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium,
provided the original work is unaltered and is properly cited. The written permission of Cambridge University Press must be obtained for commercial
re-use or in order to create a derivative work.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
improving our knowledge of glacier response to climate
change. In-situ measurements of glacier mass balance or
glacier velocity for a large debris-covered glacier, such as
Gangotri Glacier, are logically difficult and hence hardly
feasible due to its size and characteristics. In contrast to
glacier mass balance, glacier length change shows only the
indirect and delayed response of the glacier to climate
change. Given that the response time of large debris-
covered Gangotri Glacier is likely much longer than that of
smaller glaciers in the Garhwal region (Thayyen, 2008), the
determination of glacier mass balance is needed for precise
knowledge of the glacier health. Moreover, to understand
the glacier response to climate change, investigations of the
seasonal behaviour of glacier surface dynamics are also in-
dispensable. To our knowledge there are no published
studies addressing both the temporal mass balance and sea-
sonal variation of glacier surface velocity for this large debris-
covered glacier. Thus, the main goals of the present study are
(1) determine length and area variation at the snout of the
Gangotri and its tributary glaciers from 1965 to 2015 using
declassified imageries (KH-4A Corona, KH-9 Hexagon), im-
ageries from Landsat Mission and Advanced Spaceborne
Thermal Emission and Reflection Radiometer (ASTER) im-
ageries (2) to assess the geodetic glacier mass budget for
the last five decades using DTMs from declassified imageries
(KH-4A Corona) and ASTER imageries; and (3) to study the
seasonal ice surface velocity for the last two decades.
2. DESCRIPTION OF THE STUDY AREA
Gangotri Glacier is located in Garhwal Himal in Western
Himalaya (Fig. 1). It belongs to the Uttarkashi district of the
federal state of Uttarakhand in India. The glacier comprises
four main tributaries: two tributaries, Kirti Bamak and
Ghanohim Bamak, flow from the left and other two tributar-
ies, Swachhand Bamak and Maiandi Bamak, flow from the
right with respect to the main glacier flow (Srivastava,
2012). There are three other tributary glaciers, viz. Meru
Glacier (length 7.55 km) on left side and Chaturangi
Glacier (length 22.45 km) and Raktavarn Glacier (length
15.90 km) on right side (Srivastava, 2012), which have
been connected with the Gangotri Glacier in the past.
Bhambri and others (2011a) estimated that the Gangotri
and its tributary glaciers cover a total area of 210.60 km
2
,
29% of which is covered by debris. The average width of
the glacier is 1.5 km (Srivastava, 2012) and the estimated
glacier volume vary between 20 and 30 km
3
(Frey and
others, 2014). It covers an elevation range of 4000 to
7138 m a.s.l. (Srivastava, 2012).
Thayyen and Gergan, (2010) reported that the Garhwal
Himalayan glaciers are usually fed by summer monsoon
and winter snow fall. According to the recent study by
Maussion and others (2014), the glaciers in Garhwal
Himalaya are fed mainly by winter accumulation. The
maximum snowfall due to western disturbances usually
occurs from December to March as mentioned by Dobhal
and others (2008). Mean annual air temperature and
annual precipitation during 19712000 at Mukhim station,
(1900 m a.s.l.; 70 km from the snout of Gangotri
Glacier) shown in Figure 1, were found to be 15.4°C and
1648 mm respectively by Bhambri and others (2011a)using
the data recorded by Indian Meteorological Department
(IMD) and Snow and Avalanche Study Establishment
(SASE). The meteorological observatory at Bhojbasa
(3780 m a.s.l.), which is 4 km from the snout of Gangotri
Glacier (Fig. 1), recorded 11°C, 2.3°C and 546 mm
average annual maximum, minimum temperatures and
average winter snowfall respectively as reported by
Fig. 1 - Colour online, Colour in print
Fig. 1. Location of the study area in Himalaya (a) Bhagirathi Basin with Alkananda, Bhagirathi and Ganga River System (b) Gangotri and its
tributary glaciers. International boundaries are tentative only.
2Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
Bhambri and others (2011a). Precipitation data from this also
indicated on an average Gangotri and the surrounding areas
receives >15 mm of daily rainfall during the summer season
(Singh and others, 2005).
3. REVIEW OF CURRENT KNOWLEDGE
Gangotri Glacier is one of the best documented glaciers in
the Indian Himalaya with regards to the monitoring of its
snout position. It has been long observed in different
studies that the glacier has been retreating continuously
since 1935 (Auden, 1937). Based on the estimates reported
in articles (Tangari and others, 2004; Bhambri and Chaujar,
2009; Bahuguna and others, 2007), it can be stated that
Gangotri glacier retreated at a higher rate from 1970
2000. The rate of retreat was less both before 1970 and
after 2000 (Jangpangi, 1958; Vohra, 1981; Bhambri and
others, 2012; Srivastava, 2012). For more details of the pub-
lished results about the terminus retreat of Gangotri Glacier
see Figure 2 and Table 1 in the Supplement.
The areal extent of the Gangotri Glacier has been studied
from 1962 by Negi and others (2012), and a 6% glacier area
loss between 1962 and 2006 was found using SOI map and
Cartosat-1 data. A considerable reduction in glacier area was
also reported during the period from 1965 to 2006 (Kumar
and others, 2009, Bhambri and others, 2012). A 1962 SOI
map has been used as a baseline for a number of studies.
However, it should be pointed out that the map contains
some serious cartographic errors that resulted in an overesti-
mated delineation of glacier outline (Vohra, 1980; Raina and
Srivastava, 2008; Bhambri and Bolch, 2009; Raina, 2009;
Bhambri and others, 2011a;2012) In addition, interpretation
of debris-cover, shadow areas and seasonal snow on satellite
images are known to be some of the major challenges in
glacier inventories and glacier change studies (Bolch and
others, 2010; Paul and others, 2013).
It is evident from the above discussions that despite exten-
sive studies concerning glacier retreat, to the best of our
knowledge there are no published studies addressing tem-
poral mass balance and seasonal variation of glacier
surface velocity for the Gangotri Glacier in order to under-
stand the glaciers behaviour and its health. We, therefore,
present a multi-decadal assessment of the behaviour and re-
sponse of Gangotri and its tributary glaciers.
4. DATASETS
Remote sensing data were selected that had both complete
spatial coverage and suitable snow conditions. We selected
KH-4A Corona, KH-9 Hexagon, Terra ASTER, Landsat TM,
ETM+and OLI data for generating a glacier inventory, per-
forming snout monitoring and ice-surface velocity calculations
(Fig. 3 and Supplement Table 2). Among these, KH-4A Corona
stereo data of 1968 and Terra ASTER data for the year 2006 and
2014 were used for mass-balance studies. In addition, we used
SRTM3 dataset as a vertical reference for DTM generation.
4.1. KH-4A corona imagery
Corona images, declassified in 1995 and available in digital
format from 1996 (McDonald, 1995; Galiatsatos and others,
2008), were taken with two oblique viewing panoramic
cameras: a forward and a backward looking, each with a
15° tilt. This implies a stereo angle of 30° with a b/H ratio
of 0.54 (Pieczonka and others, 2011). The processing of
Corona stereo images for DTM generation has been well
described by Altmaier and Kany, (2002) and Lamsal and
others (2011). Corona KH-4A stereo-pairs of the study site
(acquired on 24 September 1965 and 27 September 1968)
were obtained from the USGS in a digital format scanned
at 3,600 dpi (7 microns).
4.2. KH-9 hexagon imagery
The KH-9 Mapping Camera (MC) System operated between
April 1973 (Mission 1205) and June 1980 (Mission 1216).
Fig. 2 - Colour online, Colour in print
Fig. 2. The annual retreat estimates of the Gangotri Glacier terminus as calculated by various authors. The horizontal bars correspond to the
observation period. The mid-point of each bar is marked by a blue point. An artificial shift of 0.1 m was introduced for several bars to improve
the legibility. The spread of the retreat values illustrates a large uncertainty of the estimates. For more details see the Table 1 in the Supplements.
3Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
Nearly 2 09 000 km
2
were recorded in trilap mode, 60
000 km
2
in bilap mode and 63 000 km
2
in mono mode
(Burnett, 2012; Pieczonka and Bolch, 2015). An area
of 250 × 125 km
2
is covered by one KH-9 scene with a
film resolution of about 85 lp mm
1
(line pairs per mm)
(Surazakov and Aizen, 2010). The data with a scan resolution
of 14 µm (1800 dpi) were used. The data were scanned by
USGS Earth Resources Observation and Science (EROS)
Centre. Two KH-9 MC stereo pairs from the mission 1216
(8 September 1980) and 1207 (24 November 1973) were
used for glacier mapping.
4.3. Landsat and terra ASTER
Data from the Landsat Mission provide a unique archive of
satellite imagery since the 1970s. For this study, the best
available cloud-free Landsat TM, ETM+and Landsat 8
scenes from the period 1993 to 2014 were downloaded
from the USGS GLOVIS website (glovis.usgs.gov).
ASTER imagery has been used for global observation of
land and ice since 2000 (e.g. Kääb and others, 2002).
ASTER Scenes from 2001 to 2014 with minimum cloud
cover were obtained under the umbrella of the Global
Land Ice Measurement from Space program (Kargel and
others, 2005) and were used for mapping, DTM generation
and surface velocity calculation.
4.4. SRTM DTM
The SRTM3 dataset (Farr and Kobrick, 2000), with a reported
absolute horizontal accuracy of 20 m and a vertical accuracy
of up to 10 m (Rodriguez and others, 2006), was chosen as
the vertical reference for collection of GCPs (Ground
Control Points). In order to overcome the radar-related data
gaps in the original DTM, a gap-filled SRTM3 DTM from
the Consultative Group on International Agricultural
Research (CGIAR) Version 4.1 (Jarvis and others, 2008) was
used.
5. METHODOLOGY
5.1. Glacier mapping
Glacier outlines for the year 2014 were generated based on
orthorectified ASTER data. The Band ratio NIR/SWIR was
used for mapping clean glacier ice. Thermal band informa-
tion of the ASTER data was also used for mapping
thin debris-covered region. Most of the debris-covered
regions were delineated manually using the slope gradient
and curvature information derived from ASTER DEM.
Additionally, shaded relief ASTER DEM was also used to
visually inspect and manually adjusted the glacier boundar-
ies (Bolch and others, 2007; Racoviteanu and others, 2008;
Bhambri and others, 2011b). The 2014 ASTER outlines
served as a basis for the manual adjustment for the other
periods.
5.2. Glacier length change calculation
Lines with 50 m separation, parallel to the main flow direc-
tion of the glacier, were drawn to calculate the length
change of Gangotri and its tributary glaciers (Koblet and
others, 2010; Bhambri and others, 2012). Length change
was calculated as the average change in distance between
two consecutive glacier outlines measured from the intersec-
tions of the lines with the glacier outlines. We also calculated
length changes in terms of the retreat along the central flow
line in order to compare with results derived from average
length change from the intersection of the lines with the
glacier outlines. Based on the outlines of the different
years, the area vacated near the snout for Gangotri and its
tributary glaciers were calculated (Fig. 4).
Fig. 3 - Colour online, Colour in print
Fig. 3. Data coverage of the utilized datasets for glacier delineation, DTM Processing and surface velocity estimation. For more details see the
Table 2 in the Supplements.
4Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
5.3. KH-4A Corona DTM processing
A KH-4A Corona DTM for the year 1968 was generated using
the Remote Sensing Software Package Graz (RSG) with a
fixed focal length of 609.60 mm. Four combinations of
forward and aft looking subsets were processed separately
in order to accommodate the entire Gangotri and its tributary
glaciers. GCPs were collected from terrain corrected Landsat
7 ETM+imagery (15 m panchromatic band, dated 15 and 22
October 1999) with SRTM3 as vertical reference. In order to
improve the sensor model, on average 225 automatically
selected tie points (TPs) for each pair of strips were also
used (Supplement Table 3). The stereo pairs of all the strips
were processed with a RMS of triangulation of <4 Pixels
(Supplement Table 3).
In order to assess glacier changes the DTMs should be
carefully co-registered so that all the pixels in the DTMs re-
present the same location and the elevation deviations of
the stable terrain are minimized (Pieczonka and others,
2013). We chose 2006 ASTER as the master/reference
DTM. KH-4A DTM (slave) was co-registered following the
approach described by Nuth and Kääb (2011) and corrected
using global trend surface analysis over gently inclined
non-glaciarized terrain using the method described by
Bolch and others (2008a) and, Pieczonka and others
(2013). The KH-4A DTM was resampled bilinearly to the
pixel size of the ASTER DTM (30 m) in order to reduce
the effect of different resolutions (Paul, 2008;Gardelle
and others, 2013). After co-registration all the DTM strips
were then mosaicked for mass-budget estimation. Before
mosaicing, the histograms of the overlapped regions were
examined visually and statistically by performing a statistic-
al significance test. For all the overlapping regions the
shape of the histograms, mean and SD values were
similar and differences between the height values estimated
from different corona strips were also statistically insignifi-
cant (p=0.18, 0.21 and 0.02 respectively for three over-
lapping regions from North). The mosaic operation has
been performed in ArcGIS 10.1.
5.4. Terra ASTER DTM processing
ASTER DTMs for the year 2006 and 2014 were generated
using PCI Geomatica Orthoengine 2014 selecting the
Toutins model (Toutin, 2002). A sufficient number of well
distributed GCPs were selected in a similar way as men-
tioned in Corona DTM processing (47 and 55 for 2006 and
2014 respectively). All stereo pairs were processed with an
RMS of triangulation of <1 Pixel (Supplementary
Table 3). A total of 70 and 65 TPs were used to improve
the sensor models for the 2006 and 2014 DTMs respectively.
The 2014 ASTER DTM was co-registered with the 2006
ASTER DTM using the same approach described above.
The overall quality of the generated raw DTMs appeared
promising as most of the glacier parts were almost fully repre-
sented. Data gaps mainly occur due to snow cover, cloud
cover and cast shadows. The difference images of the
DTMs of Gangotri and its tributary glaciers are shown in
Figure 5. Elevation differences of the off glacier terrain with
±50 scale of the study area are provided in Supplementary
Figure 1.
5.5. Data gaps and outlier handling
Data gaps in DTMs mainly occur for optical images in the
areas with less image contrast. In high-mountain areas this
is mainly related to the snow-covered accumulation
regions, areas with cast shadows and areas with cloud
cover. Therefore, outlier filtering for non-glacierized and gla-
cierized terrain is essential. For the non-glacierized terrain,
outliers are defined by the 1.5-fold interquartile range. The
68.3% quantile of the absolute elevation differences over
stable terrain was also calculated (Pieczonka and others,
2013) in order to take non-normality into account. The statis-
tics for non-glacierized terrain are shown in Table 1.
The thickness changes over the entire glaciers were not
homogeneous. It is well-known fact that there is lowest
surface elevation change at higher altitude accumulation
part of the glacier and maximum lowering in the glacier
Fig. 4 - Colour online, Colour in print
Fig. 4. Gangotri and its tributary glaciers outlines derived from different high-resolution satellite data overlay on Landsat 8 (2013) imagery.
Image shows retreat of glacier termini up to 890, 678, 394 and 378 m for Gangotri, Chaturangi, Raktavarn and Meru glaciers
respectively during 19652015.
5Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
front for retreating glaciers (Schwitter and Raymond, 1993;
Huss and others, 2010). Therefore, it is not suitable to
apply same threshold value for accumulation and ablation
region in order to identify the outliers (Pieczonka and
Bolch, 2015). Keeping this in mind, the outliers for glacier-
ized terrain were removed by using an elevation dependent
sigmoid function considering the nonlinear behaviour of
glacier thickness change. The maximum allowable thickness
change (Δh
MAX
) corresponding to a certain glacier elevation
(E
glacier
) was calculated by using the following equation as
mentioned by Pieczonka and Bolch (2015).
ΔhMAX ¼55 tanh 2π5EMAX EMIN
EGLACIER

×STDGLACIER:ð1Þ
Where, E
MAX
and E
MIN
are the maximum and minimum ele-
vation respectively, E
GLACIER
is the glacier elevation and
STD
GLACIER
is the overall standard deviation of the glacier
elevation differences. All values outside this range have
been filtered out as erroneous elevations. Finally, all data
gaps in the ablation and accumulation regions were filled
by means of ordinary kriging. We have used ordinary
kriging because it is more logical to assume a constant
mean in the local neighbourhood of each estimation point
rather than a constant mean for entire region. Moreover,
we also assumed an isotropic nature of the data with
stationary variance in order to simplify the model fitting (Li
and Heap, 2008).
5.6. Mapping uncertainty
Glacier outlines derived from various satellite datasets with dif-
ferent spatial resolutions, acquired at different times with
varying snow cover, cloud and shadow conditions have differ-
ent levels of accuracy. Uncertainty was therefore estimated for
all pairs of data used for length estimation. In this study, orthor-
ectified KH-4A Corona (1965 and 1968) and KH-9 Hexagon
data (1973 and 1980) were generated using the GCPs collected
from terrain corrected Landsat 7 ETM+imageries (15101999
and 22101999, 15 m panchromatic band) with SRTM3 as
vertical reference in RSG and ERDAS LPS Photogrammetry soft-
ware respectively. Sufficient numbers of GCPs were collected
(Supplementary Table 3) and same GCPs were used if they
were identifiable in both images. Similarly, ASTER data (2006
and 2014) were orthorectified in PCI Geomatica Orthoengine
2014 using the same GCPs and SRTM3 DTM. The length un-
certainty (U
L
) of each pair of data was calculated from the fol-
lowing formula (Hall and others, 2003).
UL¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R1
ðÞ
2þR2
ðÞ
2
qþRE ð2Þ
where, R
1
and R
2
are the spatial resolution of the image 1 and
image 2 respectively and RE is the rectification uncertainty. All
Table 1. Stable terrain statistics before and after co-registration (co-registration by Nuth and Kääb, 2011; Pieczonka and others, 2013)
RMSE
Z
MEAN MIN MAX MEDIAN SD NMAD Q 68.3
mm mm m mm m
ASTER (2006) KH-4 (1968)
Before adjustment 23.33 1.69 58.67 57.64 1.09 23.27 16.65* 6.27*
After adjustment (co-registration) 16.28 0.25 35.52 32.68 0.86 16.23
ASTER (2014) ASTER (2006)
Before adjustment 13.91 0.23 33.91 33.40 0.32 13.91 11.23* 2.13*
After adjustment (co-registration) 10.33 0.21 21.67 21.50 0.26 10.32
ASTER (2014) KH-4 (1968)
Before adjustment 26.60 0.67 68.86 67.04 0.32 26.59 14.03* 7.21*
After adjustment (co-registration) 18.36 0.13 39.75 38.01 0.11 18.36
NMAD, normalized median absolute deviation; SD, standard deviation; Q68.3, 68.3% quantile.
Fig. 5 - Colour online, Colour in print
Fig. 5. Total glacier thinning shown as difference image of respective DTMs during the period (a) 20061968 and (b) 20142006.
6Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
the imageries were rectified with KH-4 Corona (1965) image.
Theuncertaintiesestimatedfor all data pairs are given in Table 2.
A mapping inaccuracy of 2 pixels was assumed for the
outlines derived from KH-4A Corona data (4 m spatial reso-
lution), mapping inaccuracy of 1 pixel was assumed for KH-9
data (8 m spatial resolution), ASTER (15 m spatial reso-
lution) and Landsat OLI (15 m spatial resolution) and
mapping inaccuracy of half a pixel was assumed for
Landsat TM data (30 m spatial resolution). This led to an un-
certainty of 2.5% for the 1965 KH-4A Corona imagery and
3.1% for the 2015 Landsat 8 data. The overall uncertainty
for the area change was 4.0% considering the law of error
propagation (Pieczonka and Bolch, 2015).
5.7. Mass-budget uncertainty
The overall mass-budget uncertainty was estimated by asses-
sing the quality of the elevation products over glacierized as
well as non-glacierized terrain. Due to the presence of out-
liers, normalized median absolute deviation (NMAD) was
considered instead of standard deviation (SD) for the
quality criteria. However, the NMAD values for all cases
(Table 1) over stable terrain differ significantly with the
68.3% quantile. Therefore, considering the non-normality
of the elevation differences, the 68.3% quantile was used
as the quality criterion for glacier thickness change measure-
ments. Finally, the overall mass budget uncertainty (U
M
) was
calculated using Eq. 3(Pieczonka and Bolch, 2015).
UM¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Δh
t×
Δρ
ρw

2
þUDTM
t×ρI
ρw

2
s:ð3Þ
where, Δhis the measured glacier thickness change, tis the
observation period, ρ
w
is the density of water (999.972 kg
m
3
). U
DTM
is the overall thickness change uncertainty
which was assumed to be the 68.3% quantile value. The
ice density (ρ
I
) and ice density uncertainty (Δρ) were consid-
ered as 850 and 60 kg m
3
respectively (cf. Huss, 2013).
5.8. Multitemporal velocity estimation
Surface velocities of the Gangotri Glacier were calculated
from multi-temporal Landsat (TM, ETM+, OLI) and ASTER
3N data from 1993 to 2014 using Cosi-Corr (Leprince and
others, 2007). Cosi-Corr is widely used to measure glacier
surface velocity for pushbroom sensors like SPOT and
ASTER. Landsat data have some advantages over ASTER.
For example, it covers an area 9 times bigger than ASTER
and is available from early 1970s, whereas ASTER data is
only available from 2000. Landsat data are however, affected
by sub-pixel noise created by unknown attitude variation of
the satellites and imaging systems (Scherler and others,
2008). Nevertheless, image to image registration accuracy
of 5 m for ETM+sensor and 6 m for TM sensor have
been obtained by Lee and others (2004) and Storey and
Choate, (2004) respectively, which can be considered
within an acceptable limit because in most cases the displa-
cements of the glaciological features exceed this noise level
(Heid and Kääb, 2012). Moreover, to compare the results
obtained in Cosi-Corr, glacier surface velocity was also esti-
mated using another normalized cross-correlation (NCC) al-
gorithm implemented on Correlation Image Analysis
Software (CIAS) (Kääb and Vollmer, 2000) for the period
1993/94 (Fig. 6). Surface velocity along the central flow
line estimated from both the software was quite similar
(p=0.44). Therefore, glacier surface velocities were esti-
mated using Cosi-Corr for the remaining dataset.
The correlation analysis was performed using different
window sizes and steps for different datasets. Initially a rough
pixel-wise displacement was estimated using bigger window
size followed by the sub-pixel displacement determination
using smaller window size (Leprince and others, 2007).
Initial and final window sizes of 64 and 32 pixels with a step
of 2 pixels were used for Landsat TM and ETM+data
whereas a window size of 128 and 32 pixels with a step of 4
pixels were used for ASTER 3N and Landsat 8 dataset to
achieve an ice flow velocity map sampled at every 60 m.
Thecorrelation imageswere filteredto excludemiscorrelation
using three filtering steps. Initially low SNR pixels (SNR 0.90)
were excluded from the displacement map. Then the pixels
whose velocitydirection deviated ± 20
0
fromthe glacierflowdir-
ection were removed by using a directional filter (Kääb, 2005;
Kääb and others, 2005). Finally, a magnitude filter was also
used, considering the fact that the velocities do not change
abruptly, but rather, gradually (Scherler and others, 2008).
6. RESULTS
6.1. Length and area vacated near snout
This study revealed that Gangotri Glacier retreated 889.4 ±
23.2 m with an average rate of 17.9 ± 0.5 m a
1
from 1965 to
Table 2. Overall mapping uncertainty determined in this study (Hall and others, 2003)
Time period Spatial resolution
of the image
Registration
error
Overall uncertainty associated
for measurement glacier termini
mm m
CORONA KH-4 (1968) 4 3.92 9.57
HEXAGON KH-9 (1973) 7.6 4.86 13.45
HEXAGON KH-9 (1980) 7.6 5.77 14.36
LANDSAT TM (1993) 30 11.92 42.18
LANDSAT TM (1998) 30 12.56 42.82
ASTER 3N (2001) 15 10.58 26.10
ASTER 3N (2006) 15 6.41 21.93
ASTER 3N (2013) 15 6.65 22.17
ASTER 3N (2014) 15 6.81 22.33
LANDSAT 8 (2015) 15 7.01 22.53
7Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
2015 (Table 3). Variable retreat rates can be observed
during the entire observational period for the Gangotri
Glacier. The retreat rate was considerably lower between
1965 and 1968 but then increased during 19681973.
However, in recent years (200615) the retreat rate was
lower (9.0 ± 3.5 m a
1
) as compared with the period
19652006 (19.7 ± 0.6 m a
1
). Mean annual retreat of
0.9% could be observed during recent years (200615)
whereas the rate was significantly higher (2.2%)
between 1965 and 2006. Fluctuation of snout position
was also estimated along the central flow line and found
to be higher than that measured by averaging along the
front. Slightly lower average retreats were estimated for
the tributary glaciers (Table 3).
The area of Gangotri Glacier near its snout, considering
areas upto 2 km from the snout position, delineated from
the year 1965, shrank by 0.47 ± 0.07 km
2
with an average
of 0.01 ± 0.001 km
2
a
1
(0.33 ± 0.02%) between 1965 and
2015 (Table 4). Similar to the measured retreat rate, the
rate of area loss near the snout of Gangotri Glacier decreased
during the period of 200615 and was found to be 0.006 ±
0.002 km
2
a
1
(0.04 ± 0.01%), whereas from 1965 to 2006
the area shrinkage rate was 0.01 ± 0.001 km
2
a
1
(0.3 ±
0.02%). The total debris-covered area was 25.5 ± 1.0 km
2
(17.9 ± 0.7% of the whole ice cover) in 1965 which
increased to 38.7 ± 2.9 km
2
(27.3 ± 2.0% of the whole ice
cover) in 2015. Debris cover significantly increased during
recent time. During 19652006 and 200615 overall
debris cover increased by 4.4 ± 0.9% (0.15 ± 0.03 km
2
a
1
) and 4.9 ± 1.2% (0.8 ± 0.2 km
2
a
1
) respectively.
6.2. Glacier mass budget
Gangotri and its tributary glaciers experienced predominant
downwasting between 1968 and 2014 with an average thick-
ness decrease of 10.5 ± 7.2 m resulting in an average mass
loss of 0.19 ± 0.12 m w.e. a
1
(Table 5). The highest mass
loss could be observed during the period 200614 with an
average mass budget of 0.29 ± 0.19 m w.e. a
1
.
However, the mass loss rate during the period 19682006
was less as compared with the recent period and found to
be 0.17 ± 0.12 m w.e. a
1
. Average surface lowering along
longitudinal profiles with normalized length for Gangotri
and its tributary glaciers are shown in Figure 7. The profiles
are generated by applying a moving average with a band-
width of 150 m. Our results indicate an increased surface
lowering rate of Gangotri and its tributary glaciers could be
found during 200614 period (0.34 ± 0.2 m a
1
) as com-
pared with 19682006 period (0.20 ± 0.1 m a
1
). However,
the differences are not significant. Significant surface lower-
ing in the debris-covered glacier part only could be observed
for the Gangotri Glacier main trunk (Fig. 8). The average
surface lowering rate in the debris-covered part of the
Gangotri main trunk were 0.54 ± 0.3 m a
1
(19682006) and
0.8 ± 0.6 m a
1
(200614), clearly indicating that significant
thickness loss occurred despite thick debris cover. Our results
indicate that the debris-free region also thinned during the
investigated time. A slight thickening, especially in the accumu-
lation region of Gangotri and Raktavarn glaciers (Fig. 7)in
recent years (200614) is observed, possibly caused by an in-
crease in the amount of precipitation. However considering
the high uncertainty, detailed analyses of meteorological data
Fig. 6 - Colour online, Colour in print
Fig. 6. Velocity image of Gangotri Glacier System derived from images acquired on 29 October 1993 and 17 November 1994 based on
correlation of ortho images in Cosi-Corr and in CIAS software. Graph shows the velocity profile along the central flow line (Red line
marked in the image). The arrow lengths for both the images are not in the same scale.
8Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
Table 3. Total and average recession of Gangotri and its selected tributary glaciers length
Time period 196568 196873 197380 198093 199398 19982001 200106 200613 2013/14 2014/15
Total retreat (m) 20.2 ± 11.1 195.2 ± 16.5 182.8 ± 19.6 250.7 ± 44..5 73.2 ± 60.1 37.8 ± 50.1 48.8 ± 34.1 66.3 ± 31.1 9.1 ± 31.4 5.3 ± 31.7
Rate of retreat (m a
1
)6.7 ± 3.7 39.0 ± 3.3 26.1 ± 2.8 19.3 ± 3.4 14.6 ± 12.0 12.6 ± 16.7 9.8 ± 6.8 9.5 ± 4.4 9.1 ± 31.4 7.9 ± 31.7
Length change (m) along flow line +82.3 ± 11.1 510 ± 16.5 15.1 ± 19.6 165.7 ± 44..5 94.2 ± 60.1 39.6 ± 50.1 51.3 ± 34.1 80.2 ± 31.1 10.9 ± 31.4 7.3 ± 31.7
GANGOTRI GLACIER: total retreat =889.4 ± 23.2 m; rate of retreat =17.9 ±0.5 m and total central flow line retreat=1057.1 ± 23.2 m
Total retreat (m) 39.6 ± 11.1 136.9 ± 16.5 105.9 ± 19.6 154.8 ± 44..5 33.3 ± 60.1 67.1 ± 50.1 45.4 ± 34.1 78.4 ± 31.1 9.8 ± 31.4 6.4 ± 31.7
Rate of retreat (m a
1
)13.2 ± 3.7 27.4 ± 3.3 15.1 ± 2.8 11.9 ± 3.4 6.7 ± 12.0 22.4 ± 16.7 9.08 ± 6.8 11.2 ± 4.4 9.8 ± 31.4 9.6 ± 31.7
Length change (m) along flow line 38.7 ± 11.1 86.5 ± 16.5 98.1 ± 19.6 141.2 ± 44..5 28.4 ± 60.1 87.6 ± 50.1 81.1 ± 34.1 68.5 ± 31.1 11.0 ± 31.4 9.8 ± 31.7
CHATURANGI GLACIER: total retreat =677.6 ± 23.2 m; rate of retreat =13.6 ± 0.5 m and total central flow line retreat =650.9 ± 23.2 m
Total retreat (m) 78.6 ± 11.1 15.5 ± 16.5 57.8 ± 19.6 27.7 ± 44..5 16.4 ± 60.1 70.9 ± 50.1 20.1 ± 34.1 71.7 ± 31.1 10.7 ± 31.4 8.2 ± 31.7
Rate of retreat (m a
1
)26.2 ± 3.7 3.1 ± 3.3 8.3 ± 2.8 2.1 ± 3.4 3.3 ± 12.0 23.6 ± 16.7 4.0 ± 6.8 10.2 ± 4.4 10.7 ± 31.4 12.3 ± 31.7
Length change (m) along flow line 42.6 ± 11.1 5.3 ± 16.5 92.5 ± 19.6 13.2 ± 44..5 6.4 ± 60.1 114.7 ± 50.1 29.2 ± 34.1 12.2 ± 31.1 8.1 ± 31.4 4.1 ± 31.7
RAKTVARN GLACIER: total retreat =393.6 ± 23.2 m; rate of retreat =7.9 ± 0.5 m and total central flow line retreat =345.3 ± 23.2 m
Total Retreat (m) 70.4 ± 11.1 33.9 ± 16.5 21.4 ± 19.6 51.2 ± 44..5 29.4 ± 60.1 68.2 ± 50.1 101.5 ± 34.1 24.1 ± 31.1 5.6 ± 31.4 6.8 ± 31.7
Rate of Retreat (m a
1
)23.4 ± 3.7 6.8 ± 3.3 3.1 ± 2.8 3.9 ± 3.4 5.9 ± 12.0 22.7 ± 16.7 20.3 ± 6.8 3.4 ± 4.4 5.6 ± 31.4 10.2 ± 31.7
Length change (m) along flow line 72.9 ± 11.1 20.5 ± 16.5 21.8 ± 19.6 84.3 ± 44..5 38.8 ± 60.1 68.2 ± 50.1 47.6 ± 34.1 24.7 ± 31.1 6.2 ± 31.4 5.9 ± 31.7
MERU GLACIER: total retreat =377.6 ± 23.2 m; rate of retreat =7.6 ±0.5 m and total central flow line retreat =328.3 ± 23.2 m
Table 4. Total and average area vacated near snout of Gangotri and its selected tributary glaciers
Time period 196568 196873 197380 198093 199398 19982001 200106 200613 2013/14 2014/15
Total area (10
3
m
2
)17.7 ± 27.1 96.2 ± 28.0 88.2 ± 23.9 115.3 ± 32.1 39.1 ± 33.9 17.6 ± 25.4 39.8 ± 14.9 46.6 ± 13.1 4.5 ± 9.3 4.9 ± 10.0
Avg. area (10
3
m
2
a
1
)5.9 ± 9.0 19.2 ± 5.6 12.6 ± 3.4 8.9 ± 2.5 7.8 ± 6.8 5.9 ± 8.5 7.9 ± 3.0 6.6 ± 1.9 4.5 ± 9.3 7.4 ± 10.0
% Area (%) 0.01 ± 0.02 0.07 ± 0.02 0.06 ± 0.02 0.08 ± 0.02 0.03 ± 0.02 0.01 ± 0.02 0.03 ± 0.01 0.03 ± 0.01 0.003 ± 0.01 0.003 ± 0.01
GANGOTRI GLACIER: total area vacated =(469.8 ± 30.0) × 10
3
m
2
; average area vacated =(9.6 ± 0.6) × 10
3
m
2
a
1
; % area vacated =(0.33 ± 0.02)
Total area (10
3
m
2
)10.1 ± 13.2 37.8 ± 13.0 25.2 ± 11.3 33.7 ± 15.5 7.8 ± 15.5 16.5 ± 15.2 10.1 ± 13.7 23.7 ± 11.5 5.8 ± 6.4 2.1 ± 3.2
Avg. area (10
3
m
2
a
1
)3.4 ± 4.4 7.6 ± 2.6 3.6 ± 1.6 2.6 ± 1.2 1.6 ± 3.1 5.5 ± 5.1 2.0 ± 2.7 3.4 ± 1.6 5.8 ± 6.4 3.2 ± 3.2
% Area (%) 0.01 ± 0.02 0.05 ± 0.02 0.03 ± 0.02 0.05 ± 0.02 0.01 ± 0.02 0.02 ± 0.02 0.01 ± 0.02 0.03 ± 0.01 0.01 ± 0.01 0.003 ± 0.004
CHATURANGI GLACIER: total area vacated =(172.8 ± 9.8) × 10
3
m
2
; average area vacated =(3.5 ± 0.2) × 10
3
m
2
a
1
; % area vacated =(0.24 ± 0.01)
Total area (10
3
m
2
)19.6 ± 9.3 1.1 ± 8.1 12.3 ± 8.1 9.5 ± 11.4 5.4 ± 13.4 14.7 ± 14.8 1.6 ± 13.0 20.9 ± 10.1 4.6 ± 6.2 3.3 ± 4.6
Avg. area (10
3
m
2
a
1
)6.5 ± 3.1 0.2 ± 1.6 1.8 ± 1.2 0.7 ± 0.9 1.1 ± 2.7 5.0 ± 4.9 0.3 ± 2.6 3.0 ± 1.4 4.6 ± 6.2 5.0 ± 4.6
% Area (%) 0.05 ± 0.03 0.003 ± 0.02 0.03 ± 0.02 0.03 ± 0.03 0.02 ± 0.03 0.04 ± 0.04 0.004 ± 0.03 0.06 ± 0.03 0.01 ± 0.02 0.009 ± 0.01
RAKTAVARN GLACIER: total area vacated =(92.8 ± 8.3) × 10
3
m
2
; average area vacated =(1.9 ± 0.2) × 10
3
m
2
a
1
; % area vacated =(0.26 ± 0.02)
Total area (10
3
m
2
)2.0 ± 9.8 8.1 ± 9.1 5.1 ± 8.6 10.9 ± 12.9 6.5 ± 15.0 16.3 ± 14.6 29.1 ± 14.6 3.0 ± 8.7 1.8 ± 5.0 3.0 ± 4.2
Avg. area (10
3
m
2
a
1
)0.7 ± 3.3 1.6 ± 1.8 0.7 ± 1.2 0.8 ± 1.0 1.3 ± 3.0 5.4 ± 4.9 5.8 ± 2.9 0.4 ± 1.2 1.8 ± 5.0 4.6 ± 4.2
% Area (%) 0.03 ± 0.2 0.1 ± 0.1 0.08 ± 0.1 0.18 ± 0.2 0.11 ± 0.2 0.26 ± 0.23 0.47 ± 0.24 0.05 ± 0.1 0.03 ± 0.08 0.05 ± 0.07
MERU GLACIER: total area vacated =(85.8 ± 9.4) × 10
3
m
2
; average area vacated =(1.7 ± 0.2) × 10
3
m
2
a
1
; % area vacated =(1.4 ± 0.2)
9Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
are required to validate the statement, which are not available
with us.
6.3. Glacier surface velocity
The velocity measurements using Cosi-Corr show that
Gangotri Glacier is active throughout the tongue but the vel-
ocity varied slightly from 1993 to 2014 (Fig. 9). We picked a
profile along the central flow line of the Gangotri Glacier
main trunk and plotted the annual velocity (Fig. 10). We
Table 5. Mass loss and surface lowering during 19682014 of
Gangotri and its selected tributary glaciers
Observation
period
Average thickness
decreased
Uncertainty
(U
M
)
Average mass
loss
mma
1
m w.e. a
1
20061968 7.8 ± 6.3 0.14 0.17 ± 0.12
20142006 2.7 ± 2.1 0.23 0.29 ± 0.19
Total (1968
2014)
10.5 ± 7.2 0.14 0.19 ± 0.12
Fig. 7 - Colour online, Colour in print
Fig. 7. Thickness change along longitudinal profile with normalized length. During 20061968 (a) Gangotri and Raktavan Glaciers (b)
Chaturangi and Meru Glaciers, and during 20142006 (c) Gangotri and Raktavan Glaciers (d) Chaturangi and Meru Glaciers. The profiles
are generated applying a moving average with a bandwidth of 150 m. Light colour seam along the profile lines indicate 68.3% quantile of
the absolute elevation difference.
Fig. 8 - Colour online, Colour in print
Fig. 8. Thickness change of debris cover region along longitudinal profile of Gangotri Glacier main trunk during (a) 20061968 (b) 2014
2006. The profiles are generated applying a moving average with a bandwidth of 150 m. Light colour seam along the profile lines
indicate 68.3% quantile of the absolute elevation difference.
10 Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
extracted the displacement data along a profile that extends
from the accumulation zone, down to the toe of the
glacier. Due to the lack of visible surface features in the
snow covered region the correlation was not satisfactory.
However, annual displacement profiles in the lower
regions were consistent and the standard deviation among
Fig. 9 - Colour online, Colour in print
Fig. 9. Displacement image of Gangotri Glacier System derived from correlation of ortho images acquired on (a) 14 October 199622
September 1997 (b) 11 October 199815 October 1998 (c) 10 October 200315 October 2005 (d) 15 October 20059 October 2006 (e)
23 November 200826 November 2009 and (f) 29 October 201317 November 2014.
Fig. 10 - Colour online, Colour in print
Fig. 10. Comparison of annual velocity profile in different years along the central flow line of Gangotri Glacier main trunk.
11Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
data points from different displacement profiles were 7m
a
1
, except the 1996/97 displacement (11 m a
1
).
Annual surface velocities during 1998/99 (average vel-
ocity 50 ± 7.2 m a
1
) were slightly higher compared with
the 1993/94 period (average velocity 46 ± 7.5 m a
1
). It
can be assumed that higher surface velocities are probably
associated with higher temperatures during that period or
more precipitation in the previous years. No significant
annual velocity trends were visible in most parts of the
central flow line, except some portion of the ablation
region during 1996/97 (Fig. 9). The average surface velocity
after the October 1999 (46 ± 5.5 m a
1
) was found to be
slightly less than before.
Velocity differences in the debris-covered region between
200305 and 2005/06 were insignificant (p=0.1) and
almost indistinguishable. The differences during the years
2008/09 and 2013/14 were also insignificant (p=0.1).
However, significant differences (p=0.001) in the debris-
covered region can be observed by comparing 200305
and 2005/06 with 2008/09 and 2013/14.
The annual surface velocity between 2008 and 2009
(48 ± 4.8 m a
1
) was slightly higher compared with the
2013/14 period (43 ± 5.1 m a
1
) and that in the 200306
period was slightly higher (48 ± 6.1 m a
1
) than in the
200814 period (42 ± 4.9 m a
1
)(Fig. 10). The average vel-
ocity during 200614 (44.7 ± 4.9 m a
1
) decreased by
6.7% as compared with that during 19932006 (48.1 ±
7.2 m a
1
). Our results provide an indication that the
glacier velocity might have slightly decreased but the differ-
ences are not significant.
7. DISCUSSION
7.1. Glacier length and area change near snout
Gangotri Glacier has been studied extensively in terms of its
retreat rates. Estimated retreat rate and associated uncertainty
vary considerably in most of the studies due to inconsisten-
cies of the methods used. The majority of remote sensing-
based studies have used topographic maps provided by the
Survey of India (SOI) or coarse-resolution satellite data. The
retreat rate of the Gangotri Glacier is significantly lower as
compared with some previous estimation where 1962 topog-
raphy map was used. For instance, an average retreat rate of
38 m a
1
(total 1651 m) between 1962 and 2006 was
reported by Bhambri and Chaujar, (2009) using topography
map (1962) and ASTER data (2006). Similarly, Tangari and
others (2004) reported 1600 m (42 m a
1
) recession of
Gangotri Glacier using 1962 topography map and IRS-1D
data between 1962 and 2000. 1962 SOI topography map
and IRS-1C data (2000) were also used by Bahuguna and
others (2007) to estimate recession of Gangotri Glacier and
it was found to be 1510 m (40 m a
1
). Several studies
highlighted overestimated delineation of glacier outline in
1962 SOI topographic map (Vohra, 1980; Raina and
Srivastava, 2008; Raina, 2009 etc.). Therefore, it can be
assumed that the higher retreat of Gangotri Glacier is prob-
ably associated with the inconsistency of the SOI map.
On the contrary, our estimated retreat rate is in good agree-
ment with several other published results, where mainly satel-
lite data were used for retreat estimation. An average retreat
rate (20 m a
1
) similar to this study (17.9 ± 0.5 m a
1
)
was reported (Fig. 2 and Supplementary Table 1) in literatures
during different time periods (Kumar and others, 2009; Kargel
and others, 2011; Saraswat and others, 2013 etc.). Moreover,
total and average retreat (739.7 ± 27.5 m and 22.4 ± 0.8 m
a
1
) of Gangotri Glacier during 1968001 using low reso-
lution ASTER data was supported by the results obtained by
using high resolution IRS-1C PAN data (764 ± 19 m and
23.2 ± 0.6 m a
1
)byBhambriandothers(2012). An overes-
timated recession of Gangotri Glacier along the central flow
line (1057.1 ± 23.2 m) could be observed during the entire
observation period as compared with the retreat calculated
from the intersection of the glacier outlines with the lines
drawn parallel to the central flow line (889.4 ± 23.2 m).
Thus, measurements based along the centre flow line might
not provide clear evidence of overall retreat of the glacier
tongue. Therefore, averaging along the front is a more robust
method, which was already mentioned by Bhambri and
others (2012), and provides more reliable estimations. The
centre portion of the terminus of Gangotri Glacier advanced
slightly during 1968, but the averaging along the entire
glacier front during the period between 1965 and 1968
clearly indicated significant retreat (6.7 ± 3.7 m a
1
).
This study demonstrated that Gangotri Glacier lost an area
of 0.47 ± 0.03 km
2
(0.01 ± 0.001 km
2
a
1
) between 1965
and 2015 from its front. These results match well with the
remote sensing based study by Bhambri and others (2012)
and also with the study conducted by GSI using in-situ
field surveys by Srivastava (2004). Observation was taken
from 1965 to 2006 by Bhambri and others (2012) using
high-resolution satellite imageries (KH-4, KH-9, IRS-1C and
Cartosat-1) and it was found that 0.41 ± 0.03 km
2
(0.01
km
2
a
1
) area was vacated near the snout of Gangotri
Glacier. GSI study suggested 0.58 km
2
(0.01 km
2
a
1
)
area lost near the snout from 1935 to 1996. Moreover,
during 19682006 the area vacated near snout of Gangotri
Glacier (0.39 ± 0.03 km
2
and 0.01 ± 0.007 km
2
a
1
) esti-
mated from our study was also supported by Bhambri and
others (2011a) during the same observation period (0.38 ±
0.03 km
2
and 0.01 km
2
a
1
) by using KH-4 and Cartosat-1
data. However, the area loss during 200106 found in our
study is slightly higher (0.007 ± 0.003 km
2
a
1
) as com-
pared with the findings of Bhambri and others (2012)
(0.003 ± 0.002 km
2
a
1
). The difference is probably due to
the use of coarser resolution ASTER data compared with
IRS PAN and Cartosat-1 data. However, the results are
within the uncertainty range. Compared with other parts of
Garhwal Himalaya, the shrink rate of Gangotri Glacier
from this study was slightly higher (0.01 ± 0.001 km
2
a
1
)
during 19652015. For instance, Satopanth Glacier and
Bhagirathi Kharak Glacier shrank by 0.314 km
2
(0.007
km
2
a
1
) and 0.129 km
2
(0.002 km
2
a
1
) near their snouts
between 1962 and 2006 (Nainwal and others, 2008).
Similar values for Satopanth Glacier and Bhagirathi Kharak
Glacier were also estimated by Bhambri and others (2011a)
based on Corona and ASTER imagery during 19682006.
In addition, Pindari Glacier, Uttarakhand, lost 0.111 km
2
(0.0026 km
2
a
1
) at its front during 19582001 determined
by Oberoi and others (2000).
Several other studies have reported that debris cover
increased over time in the Himalaya (Iwata and others,
2000; Bolch and others, 2008a; Kamp and others, 2011).
Our study also estimated that the debris cover area of the
Gangotri Glacier main trunk increased significantly by
13.1 ± 2.1 km
2
or 9.2 ± 1.2% (0.2 ± 0.03% a
1
) between
1965 and 2014, which is similar to Bhambri and others
(2011a) who found that an area in upper Bhagirathi Basin
12 Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
increased by 11.8 ± 3.0% (0.31 ± 0.08% a
1
) during 1968
2006. Bhambri and others (2011a) also estimated that 83%
of the upper Bhagirathi Basin is covered by three large debris
covered glaciers: Gangotri, Raktavan and Chaturangi. This
study also supports the above mentioned findings for the
Gangotri Glacier during 19652014.
7.2. Glacier mass change
The mass-balance patterns in the Hindu Kush Himalayan
(HKH) region are highly variable due to the wide variation
of climatic conditions, different glacier features and different
geographic regions (Bolch and others, 2012; Gardelle and
others, 2013; Kääb and others, 2015). For instance, a slight
mass gain or balanced mass budget was observed in the
central Karakoram region, whereas moderate to high-mass
loss were observed in the Central/Eastern Himalaya and
Western Himalaya respectively during the recent decade
(Azam and others, 2012; Gardelle and others, 2013;
Vincent and others, 2013). However, to our knowledge no
mass-balance study has been published in peer-reviewed lit-
erature for Gangotri Glacier to date and mass-balance data in
the Himalayan region become sparser as we go back in time
(Bolch and others, 2012). Thus, the rate at which these gla-
ciers are changing remains poorly understood. This study
presents the longest mass-balance study using remote
sensing techniques for Gangotri Glacier and indicates that
the mass loss of Gangotri and its tributary glaciers (0.19 ±
0.12 m w.e. a
1
) from 1968 to 2014 is slightly less than
reported for other debris-covered glaciers in Himalayan
regions. For instance, in-situ measurements by Dobhal and
others (2008) for Dokriani Glacier (area 7km
2
, length 5
km and 20% debris cover) in the Garhwal Himalaya,
from 1992 to 2000 revealed a mass loss of 0.32 m w.e.
a
1
. Gautam and Mukherjee (1989) estimated mass lost of
0.24 m w.e. a
1
during 198188 for Tipra Glacier (area
7.5 km
2
, length 7 km, thick debris cover). Moreover,
various mass-balance studies were also conducted for
nearly debris-free Chhota Shigri Glacier (area 15.7 km
2
,
length 9 km) in Himachal Pradesh using different techni-
ques. For example, Azam and others (2012) estimated mass
loss of 0.67 ± 0.40 m w.e. a
1
during 200210 using glacio-
logical method and field investigation. Later on, Vincent and
others (2013) compared these results with geodetic measure-
ments and found a lower mass loss rate of 0.44 ± 0.16 m w.e.
a
1
during 1999/10.
Geodetic assessments showed a mass loss of 0.33± 0.18 m
w.e. a
1
for the heavily debris covered glaciers (Lirung
Glacier, area 6.4 km
2
; Shalbachum Glacier, area 11.5
km
2
; Langtang Glacier, area 53.5 km
2
and Langshisha
glacier, area 23.5 km
2
) in the upper Langtang valley,
central Himalaya, Nepal, during 197499 (Pellicciotti and
others, 2015). A similar value during 19702007 (0.32 ±
0.08 m w.e. a
1
) was also observed by Bolch and others
(2011) for ten glaciers (nine out of them were heavily
debris covered having debris area 36.3 km
2
), south and
west of Mount Everest using stereo Corona imageries, aerial
images and high-resolution Cartosat-1 data. However, the
debris covered Khumbu Glacier (length 12 km) in this
region lost 0.27 ± 0.08 m w.e. a
1
during the same period
(Bolch and others, 2011). It is worth mentioning that, in
spite of the lower overall mass budget during the observation
period of Gangotri and its tributary glaciers, significant
surface lowering could be observed in the debris-covered
part only.
Our study also estimated slope variations of Gangotri and
its tributary glaciers (Fig. 11) to interpret the thinning charac-
teristics, as mentioned by several researchers (Nuimura and
others, 2012; Pellicciotti and others, 2015). Five length pro-
files parallel to the central flow line were drawn and the
mean values were considered in order to avoid the ambigu-
ous selection of the central flow line (Pellicciotti and others,
2015). Nuimura and others (2012) observed higher surface
lowering of debris cover part in lower mean slope for large
glaciers and vice-versa for small glaciers in the Khumbu
region. They also mentioned that the debris covered glacier
can absorb large amounts of energy (Sakai and others,
2002) due to the presence of supraglacial ponds and ice
cliffs, which may increase melting rate (Pellicciotti and
others, 2015). Very rough topography (Fig. 11) and the pres-
ence of numerous supraglacial ponds on the debris covered
portion of the Gangotri Glacier were also confirmed by
Bhambri and others (2011a). Nuimura and others (2012)
also observed that the higher mass loss rate in debris
covered region occurred in moderate slope as compared
with steep slopes. A similar conclusion was also drawn by
Pellicciotti and others (2015) from their analysis of the gla-
ciers in the upper Langtang valley. It is also evident from
our study that for all glaciers, thinning is stronger for the
gentler slopes, especially in the lower sections of the tongues.
7.3. Glacier surface velocity
So far, few studies have investigated the velocity of the
Gangotri Glacier in different time periods. Scherler and
others (2008) observed an increasing glacier surface flow vel-
ocity with distance upstream from the terminus of the debris-
covered Gangotri Glacier using normalized cross-correlation
of optical imageries. Gantayat and others (2014) reported
that the maximum velocity varied from 61 to 85 m a
1
,
whereas minimum velocity varied from 5to15 m a
1
for the Gangotri Glacier main trunk during 2009/10 using
Landsat TM data. Similar results were also reported for
Gangotri Glacier using ASTER data by Saraswat and others
(2013). They reported that the velocity decreased from
70.2 ± 2.3 m a
1
to <30 ± 2.3 m a
1
from the accumula-
tion to the ablation region during 2009/10. Our analysis
during 2008/09 also produced similar values (maximum
71 m a
1
and minimum 13 m a
1
). Glacier surface vel-
ocities were also estimated in different parts of the
Himalayan region. For instance, Müller (1968) estimated
the surface velocity of the Khumbu Glacier during April
1956 to November 1956 at the Everest Base Camp (EBC)
and found 58 m a
1
whereas surface velocity was 28 m
a
1
at the transition between clean ice and debris-covered
ice. Bolch and others (2008b) also estimated glacier surface
velocity of the Khumbu Glacier based on the Ikonos (2000/
01) and ASTER (200103) data. The glacier surface velocity
varied from >50 m a
1
to <30 m a
1
in the upper debris-
free zone and decreased gradually towards the terminus.
Glacier surface velocities of the same glacier were also esti-
mated by Luckman and others (2007) using ERS data during
19922002 and it was found as 50 m a
1
at the EBC and
<20 m a
1
south of the transition between clean ice and
debris-covered ice.
The glacier surface velocity among the investigated gla-
ciers was also estimated during recent years (2008/09). The
13Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
average velocity of the Gangotri Glacier was significantly
higher (48 ± 4.8 m a
1
) than the Chaturangi Glacier
(20.1 ± 3.7 m a
1
) and Raktavarn Glacier (17.9 ± 4.0 m
a
1
) whereas, the average velocity of Meru Glacier was in
between (33.6 ± 5.3 m a
1
). Gangotri Glacier flow velocity
was also monitored during the late 70s using glaciological
methods (Srivastava, 2012). During 1977, average flow vel-
ocity near the snout of Gangotri Glacier was estimated as
44 m a
1
whereas 33 m a
1
flow velocity was reported
at the junction point between Gangotri and Chaturangi
Glacier. However, measurements based on one point of
the glacier are not representative for the entire glacier.
The processing errors were also examined for the loca-
tions of stable ground near the snout of the Gangotri
Glacier where the slope conditions were nearly the same
as the glacier (Saraswat and others, 2013). The location of
the stable ground and the errors associated with the process-
ing for all pairs of data are shown in the Supplementary
Figure 2. The mean bias varies from 1.43 m a
1
(2013/14)
to 5.77 m a
1
(200305) and the standard deviation varies
from 0.63 m a
1
(2013/14) to 3.01 m a
1
(1996/97), which
were quite similar to the reported values estimated by
Saraswat and others (2013) using ASTER dataset from 2005
to 2011 for Gangotri Glacier.
A relationship similar to Pellicciotti and others (2015) and
Holzer and others (2015) between recent surface velocity
(2008/09) and down-wasting during the entire observation
period (19682014) was also estimated through a profile
along the central glacial flow line for all investigated glaciers
(Fig. 12). The maximum down-wasting for most of the inves-
tigated glaciers in this study can be observed corresponding
to lower surface velocities, particularly within few kilometres
upward from their respective tongues. Sakai and others
(2000) investigated the importance of supraglacial ponds in
ablation process of the debris-covered Lirung Glacier in
Langtang valley, Nepal Himalayas. They have reported that
the heat absorption rate of a supraglacial pond is 7 times
higher than the average heat absorption of the entire
debris-covered region and more than half of the heat is
released through the water flow from the supraglacial
pond. Heat contained in the water expands the englacial
conduit and hence enhance internal ablation. Therefore, it
may be inferred from our study also that the considerable
down-wasting occurred due to the presence of supraglacial
ponds in the tongue, which may absorb large amount of in-
coming electromagnetic energy (Holzer and others, 2015).
Despite strong down-wasting together with lower surface vel-
ocity in the lower portion of the glaciers, our study estimated
significant retreat rate, though the rate is less in recent time
(Table 3). Hence, Gangotri Glacier behaves in this regard dif-
ferently than other debris-covered glaciers such as Khumbu,
Nuptse and Lhotse glaciers at Mount Everest, Nepal (Bolch
and others, 2011), Fedchenko Glacier in Pamir, Tajikistan
(Lambrecht and others, 2014), and Muztag Ata Glacier in
eastern Pamir (Holzer and others, 2015), which appeared
to have stagnant debris-covered tongues.
The velocity profile along the central flow line of Gangotri
Glacier was critically examined during 200810. Significant
annual velocity differences along the profile can be observed
from October 2008 to July 2009 and from July 2009 to
October 2010 (Fig. 13a). Earlier annual velocities along the
profile were faster than the most recent. However, it was
not clear from the results whether this velocity difference is
a true decrease over entire time period or only a seasonal
Fig. 11 - Colour online, Colour in print
Fig. 11. Normalized length profiles with average elevation difference during the period 20141968 (blue) and average slope estimated from
SRTM (orange), where the average results are from five parallel length profiles for each of the four glaciers: (a) Gangotri Glacier (b) Chaturangi
Glacier (c) Raktavarn Glacier and (d) Meru Glacier. Uncertainty range is the standard deviation (dotted); approximate debris limits (vertical
line). Curves of both elevation changes and slope were smoothed with a five-window moving average.
14 Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
effect. Therefore, seasonal effect was also studied similar to
the study mentioned by Scherler and others (2008).
Several studies demonstrated that glacier flow velocities
can vary significantly throughout the year (Anderson and
others, 2004; Bartholomaus and others, 2008). Harper and
others (2007) mentioned that highest surface velocity for
many mountain glaciers can be observed during spring to
early summer. In order to examine the seasonal effect of
Gangotri Glacier we investigated annual surface velocities
along the central flow lines over different periods of 2008
10 as described by Scherler and others (2008). A difference
in annual surface velocity in lower parts of the glacier
along the profile can be observed between the time period
starting in July 2008 and in October 2008. The velocity esti-
mated during October 2008/09 was relatively faster than the
velocity from July 2008 to October 2009 (Fig. 13b). This dif-
ference of surface velocities can be attributed due to the add-
ition of the slow surface velocity period during JulyOctober
(Scherler and others, 2008), when the flow velocity is rela-
tively slower as compared with the average annual velocity.
Therefore, it can be assumed that the velocity difference
during October 2008 to July 2009 and July 2009 to
October 2010 (Fig. 13a) as compared with October 2008
to October 2009 and July 2008 to October 2009 (Fig. 13b)
may be due to the effect of slower velocities. Thus it can
be concluded from this study that the higher melting and
consequent higher surface velocity may occur due to high
temperatures during the early summer. Such observations
are also reported by meteorological and hydrological
studies (Singh and others, 2006,2007) and remote sensing-
based studies (e.g. Scherler and others, 2008).
8. CONCLUSIONS
We have examined glacier length, area and elevation
changes for the Gangotri Glacier and its three major
Fig. 12 - Colour online, Colour in print
Fig. 12. The profile shows the surface velocities during 2008/09 (black) and corresponding down-wasting during 19682014 (red) for each of
the four glaciers: (a) Gangotri Glacier (b) Chaturangi Glacier (c) Raktavarn Glacier and (d) Meru Glacier. Flow velocity was measured in
upstream direction.
Fig. 13 - Colour online, Colour in print
Fig. 13. Annual surface velocity derived from the correlation of ortho-images (Landsat TM L1 T) during (a) October 2008October 10 and (b)
October 2008October 09. Light blue and light red seam along the profile lines indicate one sigma error. Figure shows the high contribution to
the displacement during the summer period. In the upper part of the glacier the red line is missing due to fresh snow cover.
15Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
tributaries Chaturangi Glacier, Raktavarn Glacier and Meru
Glacier, Garhwal Himalaya, India for the last five decades.
All four glaciers are heavily debris-covered on their tongues.
The seasonal variations of ice surface velocity for the last two
decades also have been investigated in this study. Our main
conclusions are as follows:
(1) The average retreat rate of the Gangotri Glacier during
the observed period (19652015) was found to be con-
sistent with the other reported values, which used satel-
lite data for comparisons. The retreat rate of Gangotri
Glacier declined from 19.7 ± 0.6 m a
1
during 1965
2006 to 9.0 ± 3.5 m a
1
during 200615. Rate of retreat
in this study, however, estimates less recession as com-
pared with the measurement obtained from 1962 SOI
map.
(2) This study also demonstrated that significant areas were
vacated near the front of Gangotri Glacier during the
investigated time. Our study also estimated that the
debris cover area of the Gangotri Glacier main trunk
increased significantly (0.2 ± 0.03% a
1
) in 2014 as
compared to 1965. Maximum average increase could
be found during the most recent period of study (2006
15).
(3) The mass loss for Gangotri and its tributary glaciers (0.19
± 0.12 m w.e. a
1
) during the observation period was
slightly less than those reported for other debris
covered glaciers in Himalayan regions. Our observation
also revealed that the recent down-wasting (200614)
was significantly higher than the previous period
(19682006). Despite the lower overall mass budget
during the observation period for Gangotri and its tribu-
tary glaciers, significant surface lowering could be
observed in debris-covered part. It was also observed
that even though the debris-free regions also possibly
thinned during the investigated time, there might be a
slight thickening in the accumulation region, especially
in Gangotri Glacier and Raktavarn Glacier in recent
years (200614). Our study on Gangotri and its tributary
glaciers also support earlier findings in the sense that for
debris-covered glaciers thinning is stronger for the gentler
slopes, especially in the lower sections of the tongues.
(4) Gangotri Glacier loses significantly mass despite thick
debris-cover. However, retreat rates in recent years
(200615) are less as compared with previous years
(19652006).
(5) Our study found a slightly lower surface velocity for
Gangotri Glacier after the year 2000 as compared with
mid 90s. Analysis of surface velocities also revealed
that there was a clear reduction in velocities from the ac-
cumulation to the ablation region during the entire obser-
vation period. The surface velocity of the Gangotri
Glacier during 2008/09 was found to be consistent
with the other reported values. The maximum down-
wasting occurred corresponding to the region of lower
surface velocities particularly within 12 km upward
from their respective tongues. It was also observed that
the tongue is active throughout, in contrast to other
Himalayan debris-covered glaciers.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit
http://dx.doi.org/10.1017/jog.2016.96.
ACKNOWLEDGEMENTS
A. Bhattacharya acknowledges research funding through
Alexander von Humboldt (AvH) foundation and T. Bolch
acknowledges funding through the German Research
Foundation (DFG, code BO 3199/21) and the European
Space Agency (Glaciers_cci project. code 4000101
77810IAM). ASTER data were provided at no cost by NASA/
US Geological Survey under the umbrella of the Global
Land Ice Measurements from Space (GLIMS) project. The
authors acknowledge the USGS Earth Resources
Observation and Science Center (EROS) for providing the
Landsat imageries. Special thanks are due to two anonymous
referees and the scientific editor of the paper for their careful
and constructive comments.
AUTHOR CONTRIBUTIONS
A.B. and T.B. designed the study and discussed the results. T.P.
and K.M. supported the DTM generation and J.K. the velocity
calculation. A.B. performed all analysis, generated the figures
and wrote the draft of the manuscript. All authors contributed
to the final form of the article.
REFERENCES
Ahmad S and Hasnain SI (2004) Analysis of satellite imageries for
characterization of glaciomorphological features of the
Gangotri Glacier, Ganga headwater, Garhwal Himalayas. In
Srivastava D, Gupta KR and Mukherjee S eds. Proceedings of
Workshop on Gangotri Glacier, 2628 March 2003, Lucknow,
India, No.80, 6167. Geological Survey of India (GSI), Calcuttta
Altmaier A and Kany C (2002) Digital surface model generation from
CORONA satellite images. ISPRS J. Photogramm. Remote Sens.,
56(4), 221235 (doi: 10.1016/S0924-2716(02)00046-1)
Anderson RS and 6 others (2004) Strong feedbacks between hydrol-
ogy and sliding of a small alpine glacier. J. Geophys. Res.,109,
F03005 (doi: 10.1029/2004JF000120)
Auden JB (1937) Snout of the Gangotri Glacier, Tehri Garhwal. Rec.
Geol. Surv. India,72, 135140
Azam MF and 10 others (2012) From balance to imbalance: a shift in
the dynamic behaviour of Chhota Shigri glacier, western
Himalaya, India. J. Glaciol.,58(208), 315324 (doi: 10.3189/
2012JoG11J123)
Bahuguna IM and 5 others (2007) Himalayan glacier retreat using
IRS 1C PAN stereo data. Int. J. Remote Sens.,28(2), 437442
(doi: 10.1080/01431160500486674)
Bartholomaus TC, Anderson RS and Anderson SP (2008) Response
of glacier basal motion to transient water storage. Nature
Geosci.,1,3337 (doi: 10.1038/ngeo.2007.52)
Bhambri R and Bolch T (2009) Glacier mapping: a review with
special reference to the Indian Himalayas. Prog. Phys. Geog.,
33(5), 672704 (doi: 10.1177/0309133309348112)
Bhambri R and Chaujar RK (2009) Recession of Gangotri glacier
(19622006) measured through remote sensing data. In
Proceeding of National Seminar on Management Strategies for
the Indian Himalaya: Development and Conservation. HNB
Garhwal University, Srinagar, India, vol. 1, 254264
Bhambri R, Bolch T, Chaujar RK and Kulshreshta SC (2011a) Glacier
changes in the Garhwal Himalayas, India 19682006 based on
remote sensing. J. Glaciol.,57(203), 543556 (doi: 10.3189/
002214311796905604)
Bhambri R, Bolch T and Chaujar RK (2011b) Mapping of debris-
covered glaciers in the Garhwal Himalayas using ASTER DEMs
and thermal data. Int. J. Remote. Sens.,32(23), 80958119
(doi: 10.1080/01431161.2010.532821)
Bhambri R, Bolch T and Chaujar RK (2012) Frontal recession of
Gangotri Glacier, Garhwal Himalayas, from 1965 to 2006,
16 Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
measured through high-resolution remote sensing data. Curr.
Sci.,102(3), 489494
Bolch T, Buchroithner MF, Kunert A and Kamp U (2007) Automated
delineation of debris-covered glaciers based on ASTER data. In
Gomarasca, M.A., ed. GeoInformation in Europe. Proceedings
of the 27th EARSeL Symposium, 47 June, 2007, Bozen, Italy.
Millpress, Rotterdam, 403410
Bolch T, Buchroithner MF, Pieczonka T and Kunert A (2008a)
Planimetric and volumetric glacier changes in the Khumbu
Himalaya 19622005 using Corona and ASTER data. J.
Glaciol.,54(187), 592600 (doi: http://dx.doi.org/10.3189/
002214308786570782)
Bolch T, Buchroithner MF, Peters J, Baessler M and Bajracharya S
(2008b) Identification of glacier motion and potentially danger-
ous glacier lakes at Mt. Everest area/Nepal using spaceborne
imagery. Nat. Hazard Earth. Syst. Sci.,8(6), 13291340 (doi:
10.5194/nhess-8-1329-2008)
Bolch T and 7 others (2010) A glacier inventory for the western
Nyainqentanglha Range and the Nam Co Basin, Tibet, and
glacier changes 19762009. Cryosphere,4, 419433 (doi:
10.5194/tc-4-419-2010)
Bolch T, Pieczonka T and Benn DI (2011) Multi-decadal mass loss of
glaciers in the Everest area (Nepal Himalaya) derived from stereo
imagery. Cryosphere,5, 349358 (doi: 10.5194/tc-5-349-2011)
Bolch T and 11 others (2012) The state and fate of Himalayan Glaciers.
Science,336(6079), 310314 (doi: 10.1126/science.1215828)
Burnett MG (2012) Hexagon (KH-9) Mapping Program and
Evolution. National Reconnaissance Office, Chantilly, Virginia
Dobhal DP, Gergan JT and Thayyen RJ (2008) Mass balance studies
of the Dokriani Glacier from 1992 to 2000, Garhwal Himalaya,
India. Bull. Glaciol. Res.,25,917
Farr TG and Kobrick M (2000) Shuttle radar topography mission pro-
duces a wealth of data. EOS Trans. AGU,81(48), 583585 (doi:
10.1029/EO081i048p00583)
Frey H and 9 others (2014) Estimating the volume of glaciers in
the Himalayan-karakoram region using different methods.
Cryosphere,8, 23132333 (doi: 10.5194/tc-8-2313-2014)
Galiatsatos N, Donoghue DNM and Philip G (2008) High resolution
elevation data derived from stereoscopic CORONA imagery with
minimal ground Control: an approach using Ikonos and SRTM
Data. Photogramm. Eng. Remote Sens.,74(9), 10931106 (doi:
10.14358/PERS.73.9.1093)
Gantayat P, Kulkarni AV and Srinivasan J (2014) Estimation of ice
thickness using surface velocities and slope: case study at
Gangotri glacier, India. J. Glaciol.,60(220), 277282 (doi:
10.3189/2014JoG13J078)
Gardelle J, Berthier E, Arnaud Y and Kääb A (2013) Region-wide glacier
mass balances over the PamirKarakoramHimalaya during 1999
2011. Cryosphere,7, 12631286 (doi: 10.5194/tc-71263-2013)
Gautam CK and Mukherjee BP (1989) Mass-balance vis-à-vis snout
position of Tipra bank glacier District chamoli, Uttar Pradesh. In
Proceedings of the national meet on Himalayan Glaciology,5
6
th
June. New Delhi, India, 141148
Hall DK, Bayr KJ, Schöner W, Bindschadler RA and Chien JYL (2003)
Consideration of the errors inherent in mapping historical glacier
positions in Austria from the ground and space. Remote Sens.
Environ.,86(4), 566577 (doi: 10.1016/S0034-4257(03)00134-2)
Harper JT, Humphrey NF, Pfeffer WT and Lazar B (2007) Two modes
of accelerated glacier sliding related to water. Geophys. Res.
Lett.,34, L12503 (doi: 10.1029/2007GL030233)
Heid T and Kääb A (2012) Repeat optical satellite images reveal
widespread and long term decrease in land-terminating glacier
speeds. Cryosphere,6, 467478 (doi: 10.5194/tc-6467-2012)
Holzer N and 5 others (2015) Four decades of glacier variations at
Muztagh Ata (eastern Pamir): a multi-sensor study including
Hexagon KH-9 and Pléiades data. Cryosphere,9, 20712088
(doi: 10.5194/tc-9-2071-2015)
Huss M (2013) Density assumptions for converting geodetic glacier
volume change to mass change. Cryosphere,7, 877887 (doi:
10.5194/tc-7-877-2013)
Huss M, Jouvet G, Farinotti D and Bauder A (2010) Future high-moun-
tain hydrology: a new parameterization of glacier retreat. Hydrol.
Earth Syst. Sci.,14,815829 (doi: 10.5194/hess-14-815-2010).
Immerzeel WW, van Beek LPH and Bierkens MFP (2010) Climate
change will affect the Asian water towers. Science,328(5984),
13821385 (doi: 10.1126/science.1183188)
Iwata S, Aoki T, Kadota T, Seko K and Yamaguchi S (2000)
Morphological evolution of the debris cover on Khumbu
Glacier, Nepal, between 1978 and 1995. In Nakawo M,
Raymond CF and Fountain A eds. Proceedings of Debris
Covered Glaciers, IAHS Publ. Seattle, vol. 264,311,
International Association of Hydrological Sciences (IAHS)
Jangpangi BS (1958) Report on the survey and glaciological study of
the Gangotri glacier, Tehri Garhwal District: Glacier No. 3, Arwa
Valley: Satopanth and Bhagirath Kharak glaciers, Garhwal
District, Uttar Pradesh. Mem. Geol. Surv. India, p.18
Jarvis A, Reuter HI, Nelson A and Guevara E (2008) Hole-filled
SRTM for globe Version 4. (available from the CGIAR_CSI
SRTM 90 m, database). http://srtm.csi.cgiar.org
Kamp U, Byrne M and Bolch T (2011) Mapping glacier fluctuations
between 1975 and 2008 in the Greater Himalaya Range of
Zanskar, South Ladakh. J. Mt. Sci.,8(3), 374389 (doi:
10.1007/s11629-011-2007-9)
Kargel JS and 16 others (2005) Multispectral imaging contributions
to global land ice measurements from space. Remote Sens.
Environ.,99(12), 187219 (doi: 10.1016/j.rse.2005.07.004)
Kargel JS, Cogley JG, Leonard GJ, Haritashya U and Byers A (2011)
Himalayan glaciers: the big picture is a montage. Proc. Natl.
Acad. Sci. USA,108(36), 1470914710 (doi: 10.1073/
pnas.1111663108)
Kaser G, Großhauser M and Marzeion B (2010) Contribution poten-
tial of glaciers to water availability in different climate regimes.
Proc. Natl. Acad. Sci. USA,107(47), 2022320227 (doi:
10.1073/pnas.1008162107)
Koblet T and 6 others (2010) Reanalysis of multi-temporal aerial
images of Storglaciären, Sweden (19591999)- part 1: determin-
ation of length, area, and volume changes. Cryosphere,4, 333
343 (doi: 10.5194/tc-4-333-2010)
Kumar K, Dumka RK, Miral MS, Satyal GS and Pant M (2008)
Estimation of retreat rate of Gangotri glacier using rapid static
and kinematic GPS survey. Curr. Sci.,94(2), 258262
Kumar R, Areendran G and Rao P (2009) Witnessing change:
Glaciers in the Indian Himalayas. WWF, India, pp. 48
Kääb A (2005) Combination of SRTM3 and repeat ASTER data for de-
riving alpine glacier flow velocities in the Bhutan Himalaya.
Remote Sens. Environ.,94(4), 463474 (doi: 10.1016/j.
rse.2004.11.003)
Kääb A and Vollmer M (2000) Surface geometry, thickness changes
and flow fields on creeping mountain permafrost: automatic ex-
traction by digital image analysis. Permafrost Periglac. Processes,
11(4), 315326 (doi: 10.1002/1099-1530(200012)11:4<315::
AID-PPP365>3.0.CO;2-J)
Kääb A and 6 others (2002) Glacier monitoring from ASTER imagery:
Accuracy and Applications. Proceedings of EARSeL-LISSIG-
Workshop Observing our Cryosphere from Space,1113
th
March, Bern, Number 2, 4353
Kääb A, Lefauconnier B and Melvold K (2005) Flow field of
Kronebreen, Svalbard, using repeated Landsat 7 and ASTER
data. Ann. Glaciol.,42(1), 713 (doi: http://dx.doi.org/10.3189/
172756405781812916)
Kääb A, Treichler D, Nuth C and Berthier E (2015) Brief communi-
cation: contending estimates of 20032008 glacier mass
balance over the Pamir-Karakoram-Himalaya. Crosphere,9,
557564 (doi: 10.5194/tc-9-557-2015)
Lambrecht A, Mayer C, Aizen V, Floricioiu D and Surazakov A
(2014) The evolution of Fedchenko glacier in the Pamir,
Tajikistan, during the past eight decades. J. Glaciol.,60(220),
233244 (doi: 10.3189/2014JoG13J110)
Lamsal D, Sawagaki T and Watanabe T (2011) Digital terrain mod-
elling using corona and ALOS PRISM data to investigate the distal
17Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
part of Imja Glacier, Khumbu Himal, Nepal. J. Mt. Sci.,8, 390
402 (doi: 10.1007/s11629-011-2064-0)
Lee DS, Storey JC, Choate MJ and Hayes RW (2004) Four years of
Landsat-7 on-orbit geometric calibration and performance. IEEE
Trans. Geosci. Remote Sens.,42(12), 27862795 (doi:
10.1109/TGRS.2004.836769)
Leprince S, Barbot S, Ayoub F and Avouac JP (2007) Automatic and
precise orthorectification, coregistration, and subpixel correl-
ation of satellite images, application to ground deformation mea-
surements. IEEE Trans. Geosci. Remote Sens.,45(6), 15291558
(doi: 10.1109/TGRS.2006.888937)
Li J and Heap AD (2008) A Review of Spatial Interpolation Methods
for Environmental Scientists. Geoscience Australia, Record 2008/
23, 137 pp. ISBN 978 1 921498 28 2
Luckman A, Quincey D and Bevan S (2007) The potential of satellite
radar interferometry and feature tracking for monitoring flow
rates of Himalayan glaciers. Remote Sens. Environ.,111(23),
172181 (doi: 10.1016/j.rse.2007.05.019)
Mattson LE, Gardner JS and Young GJ (1993) Ablation on Debris
Covered Glaciers: an example from the Rakhiot Glacier,
Punjab, Himalaya. Symposium at Kathmandu, Nepal, Nov.
1992- Snow and Glacier Hydrology, IAHS publ. 218, 289296
McDonald RA (1995) CORONA-success for space reconnaissance,
a look into the Cold War, and a revolution for intelligence.
Photogramm. Eng. Remote Sens.,61(6), 689720
Mukherjee BP and Sangewar CV (2001) Recession of Gangotri
glacier through 20th century. Geological Survey of India
Special Publication, Number 65, pp. 13
Maussion F and 5 others (2014) Precipitation seasonality and variabil-
ity over the Tibetan Plateau as resolvedby the High Asia Reanalysis.
J. Climate,27,19101927 (doi: 10.1175/JCLI-D-13-00282.1)
Müller F (1968) Mittelfristige Schwankungen der Ober-flaechen-
geschwindigkeiten des Khumbugletschers am Mount Everest.
Schweizerische Bauzeitung,86(31), 569573 (doi: http://dx.doi.
org/10.5169/seals-70102)
Nainwal HC, Negi BDS, Chaudhary M, Sajwan KS and Gaurav A
(2008) Temporal changes in rate of recession: evidence from
Satopanth and Bhagirath Kharak glaciers, Uttarakhand, using
Total Station Survey. Curr. Sci.,94(5), 653660
Naithani AK, Nainwal HC, Sati KK and Prasad C (2001)
Geomorphological evidences of retreat of the Gangotri glacier
and its characteristics. Curr. Sci.,80(1), 8794
Negi HS, Thakur NK, Ganju A and Snehmani (2012) Monitoring of
Gangotri glacier using remote sensing and ground observations.
J. Earth. Syst. Sci.,121(4), 855866 (doi: 10.1007/s12040-012-
0199-1)
Nuimura T, Fujita K, Yamaguchi S and Sharma RR (2012) Elevation
changes of glaciers revealed by multitemporal digital elevation
models calibrated by GPS survey in the Khumbu region, Nepal
Himalayas, 19922008. J. Glaciol.,58(210), 648656 (doi:
10.3189/2012JoG11J061)
Nuth C and Kääb A (2011) Co-registration and bias corrections of
satellite elevation data sets for quantifying glacier thickness
change. Cryosphere,5, 271290 (doi: 10.5194/tc-5271-2011)
Oberoi LK, Maruthi KV and Siddiqui MA (2000) Secular movement
studies of the selected glaciers in Pindar and Vishnuganga basins,
Almora and Chamoli districts, Uttaranchal. Geol. Surv. India
Rec.,135(8), 114115
Paul F (2008) Calculation of glacier elevation changes with SRTM: is
there an elevation dependent bias? J. Glaciol.,54(188), 945946
(doi: http://dx.doi.org/10.3189/002214308787779960)
Paul F and 19 others (2013) On the accuracy of glacier outlines
derived from remote sensing data. Ann. Glaciol.,54(63), 171
182 (doi: http://dx.doi.org/10.3189/2013AoG63A296)
Pellicciotti F and 5 others (2015) Mass-balance changes of the
debris-covered glaciers in the Langtang Himal, Nepal, from
1974 to 1999. J. Glaciol.,61(226), 373386 (doi: http://dx.doi.
org/10.3189/2015JoG13J237)
Pieczonka T, Bolch T and Buchroithner MF (2011) Generation and
evaluation of multitemporal digital terrain models of the Mt. Everest
area from different optical sensors. ISPRS J. Photogramm. Remote
Sens.,66(6), 927940 (doi: 10.1016/j.isprsjprs.2011.07.003)
Pieczonka T and Bolch T (2015) Region wide glacier mass budgets
and area changes for the Central Tien Shan between 1975 and
1999 using Hexagon KH-9 imagery. Global. Planet. Change,
128,113 (doi: 10.1016/j.gloplacha.2014.11.014)
Pieczonka T, Bolch T, Wie J and Liu S (2013) Heterogeneous mass
loss of glaciers in the Aksu-Tarim Catchment (Central Tien
Shan) revealed by 1976 KH-9 Hexagon and 2009 SPOT-5
stereo imagery. Remote Sens. Environ.,130, 233244 (doi:
10.1016/j.rse.2012.11.020)
Racoviteanu AE, Arnaud Y, Williams MW and Ordonez J (2008)
Decadal changes in glacier parameters in the Cordillera
Blanca, Peru, derived from remote sensing. J. Glaciol.,54(186),
499510 (doi: http://dx.doi.org/10.3189/002214308785836922)
Raina VK (2004) Is the Gangotri glacier receding at an alarming rate?
J. Geol. Soc. India.,64, 819821
Raina VK (2009) Himalayan glaciers: a state-of-art review of glacial
studies, glacial retreat and climate change. Kosi-Katarmal, Ministry
of Environment and Forests. G.B. Pant Institute of Himalayan
Environment and Development. (MoEFF Discussion Paper.),p.60
Raina VK and Srivastava D (2008) Glacier Atlas of India. Geological
Society of India, Bangalore, First Edition, ISBN: 81-85867-80-9,
p. 316
Rodriguez E, Morris CS and Belz JE (2006) A global assessment of
the SRTM performance. Photogramm. Eng. Remote. Sens.,
72(3), 249260 (doi: http://dx.doi.org/10.14358/PERS.72.3.249)
Sakai A, Takeuchi N, Fujita K and Nakawo M (2000) Role of supra-
glacial ponds in the ablation process of a debris-covered glacier
in the Nepal Himalayas. IAHS publ. 265 (Symposium at Seattle,
Washington, USA, Sep. 2000- Debris-Covered Glaciers,119130
Sakai A, Nakawo M and Fujita K (2002) Distribution characteristics
and energy balance of ice cliffs on debris-covered glaciers, Nepal
Himalaya. Arct. Antarct. Alp. Res.,34(1), 1219 (doi: 10.2307/
1552503)
Saraswat P and 5 others (2013) Recent changes in the snout position
and surface velocity of Gangotri glacier observed from space.
Int. J. Remote Sens.,34(24), 86538668 (doi: 10.1080/
01431161.2013.845923)
Scherler D, Leprince S and Strecker MR (2008) Glacier-surface vel-
ocities in alpine terrain from optical satellite imagery-Accuracy
improvement and quality assessment. Remote Sens. Environ.,
112(10), 38063819 (doi: 10.1016/j.rse.2008.05.018)
Schwitter MP and Raymond CF (1993) Changes in the longitudinal
profile of glaciers during advance and retreat. J. Glaciol.,39(133),
582590 (doi: http://dx.doi.org/10.3198/1993JoG39-133-582-590)
Singh P, Haritashya UK, Ramasastri KS and Kumar N (2005)
Prevailing weather conditions during summer seasons around
Gangotri Glacier. Curr. Sci.,88(5), 753760
SinghP, HaritashyaUK, KumarN and Singh Y (2006) Hydrological char-
acteristics of the Gangotri glacier, central Himalayas, India. J.
Hydrol.,327(12), 5567 (doi: 10.1016/j.jhydrol.2005.11.060)
Singh P, Haritashya UK and Kumar N (2007) Meteorological study
for Gangotri Glacier and its comparison with other high altitude
meteorological stations in central Himalayan region. Nord.
Hydrol.,38(1), 5977 (doi: 10.2166/nh.2007.028)
SinghP, HaritashyaUK and Kumar N (2008) Modelling and estimation of
different components of streamflow for Gangotri basin, Himalayas.
Hydrol. Sc. J.,53(2), 309322 (doi: 10.1623/ hysj.53.2.309)
Singh P, Polglase L and Wilson D (2009) Role of snow and glacier
melt runoff modeling in hydropower projects in the Himalayan
region. In Jain SK, Singh VP, Kumar V, Kumar R, Singh RD and
Sharma KD eds. Proceedings of the International Conference
on Water, Environment, Energy and Society (WEES2009), 12
16 January 2009, New Delhi, India. Vol. 1. National Institute of
Hydrology, Roorkee, 366371
Srivastava D (2004) Recession of Gangotri glacier. In Srivastava D,
Gupta KR and Mukerji S eds. Proceedings of Workshop on
Gangotri glacier, 2628 March, Lucknow, India, Geological
Survey of India, Special Publication, Number 80, pp. 2132
18 Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
Srivastava D (2012) Status Report on Gangotri Glacier. Science and
Engineering Research Board, Department of Science and
Technology, New Delhi, Himalayan Glaciology Technical
Report, No. 3, pp. 102
Storey JC and Choate MJ (2004) Landsat-5 bumper-mode geometric
correction. IEEE Trans. Geosci. Remote Sens.,42, 26952703
(doi: 10.1109/TGRS.2004.836390)
Surazakov A and Aizen V (2010) Positional accuracy evaluation of
declassified Hexagon KH-9 mapping camera imagery.
Photogramm. Eng. Remote Sens.,76(5), 603608 (doi: http://
dx.doi.org/10.14358/PERS.76.5.603)
Tangari AK, Chandra R and Yadav SKS (2004) Temporal monitoring
of the snout, equilibrium line and ablation zone of Gangotri
glacier through remote sensing and GIS techniques an
attempt at deciphering the climatic variability. In Srivastava D,
Gupta KR and Mukerji S eds. Proceedings of Workshop on
Gangotri glacier, 2628 March, Geological Survey of India,
Lucknow, India, Special Publication, Number 80, pp. 145153
Thayyen RJ (2008) Lower recession rate of Gangotri glacier during
19712004. Curr. Sci.,95(1), 910
Thayyen RJ and Gergan JT (2010) Role of glaciers in watershed hy-
drology: a preliminary study of a Himalayan catchment.
Cryosphere,4(1), 115128 (doi: 10.5194/tc-4-115-2010)
Toutin T (2002) 3D Topographic mapping with ASTER stereo data in
rugged topography. IEEE Trans. Geosci. Remote Sens.,40, 2241
2247 (doi: 10.1109/TGRS.2002.802878)
Vincent C and 10 others (2013) Balanced conditions or slight mass
gain of glaciers in the Lahaul and Spiti region (northern India,
Himalaya) during the nineties preceded recent mass loss.
Cryosphere,7, 569582 (doi: 10.5194/tc-7-569-2013)
Vohra CP (1980) Some problems of glacier inventory in the
Himalayas. Proceedings of the Workshop of Riederalp, 1722
September 1978, IAHS-AISH Publication, vol. 126, pp. 6774
Vohra CP (1981) Himalayan glaciers. In Lall JS and Moddie AD eds.
The Himalayan Aspect of Change. Oxford University Press, New
Delhi, India, 138151
MS received 9 February 2016 and accepted in revised form 8 July 2016
19Bhattacharya and others: Overall recession and mass budget of Gangotri Glacier, Garhwal Himalayas
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
... Several studies have estimated two-dimensional (i.e. glacier area, length or snout) glacier changes (Cogley 2011;Zemp et al. 2015;Romshoo et al. 2020a), which represents an assessment of indirect, delayed and filtered response of changing climate (Scherler et al. 2011;Gardelle et al. 2013;Bhattacharya et al. 2016). ...
... Low signal to noise ratio (SNR) values less than (0.9) were initially filtered out to remove the poorly correlated pixels. The SIV values > 200 m a −1 were also removed to exclude the incorrect SIV values caused by poor correlation (Bhattacharya et al. 2016;Bhushan et al. 2018). ...
... Landsat images offer good spatial and temporal coverage (Ding et al. 2016; Shukla and Garg 2020) but they also feature subpixel noise caused by attitude changes (Heid and Kääb 2012;Bhattacharya et al. 2016). The average image-to-image registration precision for ETM + and TM sensors is ~ 5 and ~ 6 m, respectively (Storey and Choate 2004;Bhattacharya et al. 2016), which is difficult to model due to their whisk-broom nature. ...
Article
Full-text available
The Himalayan glaciers provide water to a large population in south Asia for a variety of purposes and ecosystem services. As a result, regional monitoring of glacier melting and identification of the drivers are important for understanding and predicting future cryospheric melting trends. Using multi-date satellite images from 2000 to 2020, we investigated the shrinkage, snout retreat, thickness changes, mass loss and velocity changes of 77 glaciers in the Drass basin, western Himalaya, India. During this period, the total glacier cover has shrunk by 5.31 ± 0.33 km 2. The snout retreat ranged from 30 to 430 m (mean 155 ± 9.58 m). Debris cover had a significant impact on glacier melting, with clean glaciers losing ~ 5% more than debris-covered glaciers (~ 2%). The average thickness change and mass loss of glacier have been − 1.27 ± 0.37 and − 1.08 ± 0.31 m w.e.a −1 , respectively. Because of the continuous melting and the consequent mass loss, average glacier velocity has reduced from 21.35 ± 3.3 m a −1 in 2000 to 16.68 ± 1.9 m a −1 by 2020. During the observation period, the concentration of greenhouse gases (GHGs), black carbon (BC) and other pollutants from vehicular traffic near the glaciers increased significantly. Increasing temperatures, caused by a significant increase in GHGs, black carbon and other pollutants in the atmosphere, are driving glacier melting in the study area. If the current trend continues in the future, the Himalayan glaciers may disappear entirely, having a significant impact on regional water supplies, hydrological processes, ecosystem services and transbound-ary water sharing.
... However, with the present 7-decade long reconstruction, it is clear that both the glaciers gained some mass during Period I (1951Period I ( -1975. The accelerated mass wastage over Period III (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020) observed on DBG and CSG are also in agreement with several remote sensing as well as modelling studies that suggest a rapid post-2000 mass wastage (Bhattacharya et al., 2016;Maurer et al., 2019;Mandal et al., 2020). The present study, together with other studies, suggest a slightly positive MB period , a near-steady state and an accelerated wastage (post-2020) on DBG and CSG. ...
... The present study, together with other studies, suggest a slightly positive MB period , a near-steady state and an accelerated wastage (post-2020) on DBG and CSG. However, given the large uncertainties in our study as well as other studies compared here Bhattacharya et al., 2016;Maurer et al., 2019;Azam & Srivastava, 2020) these inferred MB states should be viewed with caution. ...
Article
Glacier mass balance (MB) and runoff relationship is poorly explored in the Himalaya due to limited in-situ data. A simplified glacio-hydrological model is used to reconstruct the long-term MB and runoff on Dokriani Bamak Glacier (DBG) in the Monsoon regime and Chhota Shigri Glacier (CSG) catchments in the Alpine regime of the Himalaya. Since 1950, DBG and CSG show limited mass wastage with mean annual glacier-wide MB of –0.09 ± 0.35 m w.e. a⁻¹ and –0.12 ± 0.28 m w.e. a⁻¹, respectively. The mean annual catchment-wide runoff are also almost similar 0.13 ± 0.01 10⁶ m³ a⁻¹ and 0.14 ± 0.01 10⁶ m³ a⁻¹ from DBG and CSG catchments, respectively over 1950-2020. Available in-situ short-term MB and runoff data from eight glacierized catchments do not show any significant MB-runoff relationship, except for CSG. The long-term modelled MB and runoff data on both DBG and CSG catchments suggest that the higher catchment-wide runoff correspond to more negative MB, and vice versa. This is in contrast to the previously suggested MB-runoff relationship on DBG catchment. Present detailed analysis also indicates that the hydrology of the DBG catchment is mainly dominated by the summer precipitation, while in CSG catchment, it is mainly controlled by the glacier-wide annual MB. In dry years, the corresponding negative annual MB on both the glaciers provide additional glacier-degraded runoff to the catchment streams that can be as high as 13% and 38% of total runoff on DBG and CSG catchments, respectively.
... The Gangotri glacier occupies an elevation range of 4000-7000 m above sea level. It flows 30 km northwest towards its snout (Bhattacharya et al., 2016), which feeds the River Bhagirathi, the source stream of the River Ganga. Due to its meltwater's contribution to the Rivers Bhagirathi and Ganga, the glacier requires continuous monitoring as it follows the worldwide trend of glacier retreat and mass loss. ...
... Due to its meltwater's contribution to the Rivers Bhagirathi and Ganga, the glacier requires continuous monitoring as it follows the worldwide trend of glacier retreat and mass loss. A vast literature exists on this glacier related to its geomorphological mapping, retreat, hydrology, mass balance, surface velocity (Bandyopadhyay et al., 2019;Rai et al., 2019;Agrawal et al., 2017;Bhattacharya et al., 2016). Albeit vital, the Gangotri glacier is rarely studied in terms of its surface facies. ...
Article
New techniques are being implemented to precisely monitor the changes in debris-covered glaciers for ascertaining their response to climate change. The super-resolution mapping (SRM) technique is implemented here to provide the low-cost and fine-resolution facies maps of debris-covered glaciers in the Indian Himalayan region using coarse-resolution satellite data. The implemented SRM algorithm is capable of preserving the spatial pattern of facies on the debris-covered glacier surface. Scale factor (sf) and surface heterogeneity are two possible determinants of SRM accuracy. Larger sfs decrease the SRM accuracy and its computing efficiency, and hence optimal sfs must be selected. Higher SRM accuracies are obtained for less heterogenic glacier surfaces and vice-versa. Uncertainty in the resultant maps arises due to the underlying natural factors (shadow/clouds) and seasonal differences in the image acquisition. Temporal implementation of SRM is feasible to assess the facies variations on a debris-covered glacier surface, facilitating time- and cost-effective monitoring of its response to climate change over large areas.
... Meanwhile, snout fluctuation in a glacier helps understand the frontal dynamics and disintegration patterns . Though quite distinct and easily observed, the length and area changes should be supplemented with data on other parameters such as glacier dynamics (mass balance and SIV) to comprehend the actual state of the glaciers (Bhattacharya et al., 2016;Garg et al., 2017a). A glacier is majorly divided into two zones: (1) Accumulation zone, where fresh snow is received and accumulates, which under the influence of gravity is transported to the lower elevation known as (2) Ablation zone, where melting takes place. ...
... We have, however, employed the feature tracking method on optical imageries (Table 1), performed in Co-Registration of Optically Sensed Images and Correlation (COSI-Corr) software. It is a suitable method for estimating the displacement of moving objects and has been successfully implemented in the Himalayan terrain (Bhattacharya et al., 2016;Garg et al., 2017a;Bhushan et al., 2018;Sam et al., 2018). The basic principle followed in this method is the sub-pixel correlation of two co-registered and orthorectified images acquired during different times, thereby estimating the displacement in the given time interval (Leprince et al., 2007;Scherler et al., 2011). ...
Article
Glacier specific studies, in a relatively unexplored terrain of Ladakh, hold immense importance tocomprehend not only the glacier response but also its synchronicity with the general regional trend. Accordingly, in this study, the Kangriz glacier in the Suru sub-basin, western Himalaya, has been taken up for multiparametric (area, terminal retreat, debriscover, snow line altitude, surface dynamics) assessment for the period 1971–2018. Results reveal an overall shrinkage of 3.3 ± 1.6%, with an expansion in the supraglacial debris cover by 45% (1971–2018). Concomitantly, the glacier surface velocity has reduced by 10.85 ± 5.68 ma⁻¹ (35%), from 31.2 ± 5.8 ma⁻¹ (1993/94) to 20.3 ± 1.7 ma⁻¹ (2017/18), with mass wastage of −0.52 ± 0.19 m w.e.a⁻¹ during 2000–17. The notable glacier degeneration is synchronous with regional warming (Tmax increase by 7%, Tmin increase by 43%, Tavg increase by 64%) and a decrease in precipitation by 3% (significant at α <0.05). Besides, frontal dynamics have changed recently, with an enhanced intensity of terminal retreat (2016–2018: 57 ± 13(Stdev) ma⁻¹). The overall glacier status suggests a degenerative pattern of the glacier, which is in sync with the other western Himalayan glaciers. In view of the recently amplified ice-calving events and rapid mass loss observed in the snout region, the frontal glacier morphology may change drastically in the coming years .
... (1961-2015) and precipitation (1951-2007) Note: Significance level: P < 0.05 *, P < 0.01 **, P < 0.001***. Patel et al. 2021) and in the wider Himalaya (Scherler et al. 2008;Quincey et al. 2009;Saraswat et al. 2013;Gantayat et al. 2014;Bhattacharya et al. 2016;Kraaijenbrink et al. 2016;Satyabala 2016;Shukla and Garg 2020). Based on in situ dGPS measurements, Patel et al. (2021) reported higher velocity for Samudra Tapu (64.3 ± 36.7 m/yr) and Sutri Dhaka (52.6 ± 17.3 m/yr) glaciers and comparatively slow movement in Gepang Gath (26.5 ± 12.9 m/yr) and Batal (6.2 ± 2.9 m/yr) glaciers in the Chandra basin. ...
Article
Full-text available
Spatiotemporal surface velocity measurements of the alpine valley type debris-covered Miyar Glacier of the Chandrabhaga (Chenab) basin, western Himalaya, were assessed based on the cross-correlation of Landsat images spanning nearly three decades (1992-2019). Long-term (1950-2015) temperature and precipitation trends were evaluated using Asian Precipitation Highly Resolved Observational Data Integration Towards Evaluation (APHRODITE) datasets. The mean velocity (1992-2019) of the Miyar Glacier is ∼29 m/yr, with spatial patterns revealing that the debris-covered tongue is nearly stagnant (∼5 m/yr) compared to the debris-free up-glacier zone (∼35 m/yr). The transition zone from clean to debris-covered ice in the mid-ablation area shows the highest long-term mean velocities of ∼60 m/yr during the observation period, likely resulting from a steep surface gradient and greater ice thickness than the other regions of this glacier. The slow-moving and nearly stagnant debris-covered area reveals the highest amount of surface lowering due to the expansion of supraglacial ponds. Miyar Glacier experiences summer speed-up of ∼67-80% in seasonal velocity compared to winter, interpreted as a result from enhanced basal sliding during summer months due to warmer temperatures inputting more meltwater into the subsurface drainage system. Inter-annual velocity variations are greatest in the upper glacier, with higher velocities observed more frequently in recent decades. Future work should aim to elucidate the causes of this pattern, considering the overall rising air temperature trend in the western Himalaya. ARTICLE HISTORY
... Literature shows that the average surface air temperature in the Himalayan region increased by 1 °C in the previous era (Telwala et al., 2013). Based on new literature, the ground truth and remote sensing studies of the Himalayan glaciers indicate a longterm climatic variability that may accelerate the melting and retreating of the glaciers of the Himalayan region (Bhattacharya et al., 2016). Other research has discovered that snowfall tendency has reduced, which has traditionally been an important source for the region's glaciers to reserve freshwater supplies. ...
Article
Full-text available
There are several causes for the increasing rate of deglaciation, such as global warming, increase in the concentration of black carbon, and extensive use of fossil fuels which causes the change in the overall climate system and shifting glacier ecosystem. This study was conducted on Pindari valley glaciers part of lesser Himalaya in Uttarakhand. This study investigates to (1) monitor and map change in the frontal length or the snout region of a glacier that can be studied with the help of remote sensing techniques and (2) evaluate the decadal and annual retreat rate of the glacier from 1972 to 2018. The study applies both the maximum likelihood classifier and NDSI spectral indices based classification for extracting the glacier region for different periods. This study reveals a significant amount of retreats taking place in the selected glaciers, Pindari, Sundardhunga, Kafni, and Baljuri base camp glaciers, from 1972 to 2018 as 1719.95 m, 1751.21 m, 1057.01 m, and 810.78 m, respectively. The highest amount of change is noticed in Pindari and Sundardhunga glaciers, higher than ~ 1700 m. The study analyses temporal variation of the annual and decadal retreat rate in the Pindari valley glaciers, which would be helpful for the further study of the other glaciers.
... The Normalized Median Absolute Deviation (NMAD) is not as sensitive to outliers as the Standard Deviation and has been recommended to evaluate DEM precision (Höhle and Höhle, 2009). Therefore, NMAD of non-glacial stable areas was utilized to assess uncertainty in surface elevation change (Bhattacharya et al., 2016) (Supplementary Table 1). We found 18 m (1.1 m − 1 ) average NMAD values that are within the range of previous estimates (e.g., Berthier and Brun, 2019). ...
Article
The Karakoram has a large concentration of surge-type glaciers, including 69 tributary glaciers, compared to 152 surge-type main or trunk glaciers. The paper addresses the interactions between tributary and trunk glaciers using digital elevation models (DEMs), surface displacement, field and archival reports. In particular, it explores the behavior and impacts of 13 tributary glacier surges on three trunk glaciers, namely the Hispar, Braldu and Panmah. Observations include five surge tributaries of Panmah, five of Braldu, and three of Hispar. We observed ASTER DEMs can help in some cases to detect surge signature where automated surface displacement does not detect the surge. We also observed substantial differences in surge dimensions, timing and histories of the main trunk glacier and their tributaries. East Braldu III tributary surged between 2000 and 2003, whereas East Braldu IV surged from 2003 to 2006, but in these periods, no other tributary shows surge signature. Between 2013 and 2016, Braldu trunk Glacier surged along with four tributaries out of five except West Braldu I. Volumes and geometry of ice transferred from tributary to trunk glaciers are unique to each case, but the surging ice melted rapidly in about 2 to 4 years for some cases such as Little Skamri and Drenmang. The tributary ice modified all studied trunk glacier dynamics, morphology, distribution of debris and hypsography. The ice transferred from tributaries such as Little Skamri and Drenmang blocked the flow of trunk Nobande Sobonde Glacier from 2004 to 2006. Such ice transfer by surge tributaries to the main trunk glacier is referred here as surge-modified ice. It introduces indirect and post-surge effects and complicates or delay in tracking glacier responses to climate change. Also, mass balance in surge-type and surge-modified glaciers differ from systematic direct responses to climate in non-surge-type glaciers. Therefore, more research and monitoring are required to address the distinct responses of such glaciers and individual tributaries to better understand the heterogeneity of surging glaciers in Karakoram.
... In order to reduce any errors, imagery with minimum cloud and seasonal snow cover were used. The image co-registration error in the Landsat imageries is acceptable for glaciological studies [47,48]. The error introduced due to orthorectification may result into some minor horizontal shift. ...
Article
Full-text available
Investigation of spatiotemporal variation in glacier velocity is imperative to comprehend glacier mass and volume loss as a function of their sensitivity to climate change. The long‐term glacier velocity record for Eastern Himalayan region is of utmost importance owing to its data scarcity and climate sensitivity. Here, we present a long‐term dataset spanning over more than two decades (1994‐2020) of glacier surface velocity for the entire Sikkim Himalaya by applying image correlation method on the multi‐temporal Landsat images. Our result demonstrates an average glacier surface velocity decline from 15.7±5.69 (1994/96) to 12.88±2.09 m yr‐1 (2018/2020) i.e. decline by ~15% during the period of investigation. Trend analysis shows decreasing trend in median velocity (32.2%) at a rate of 0.25 m yr‐1. Despite the general decline in average glacier velocity, rate of slowdown of individual glaciers is extremely heterogeneous (3.6‐20 m yr‐1). Our study shows that up to 32% of the observed heterogeneity in velocity variation can be explained by the variation in glacier size. The present study highlights that large glaciers with thick ice cover move faster as compared to small glaciers (even those situated on the steep slopes). The findings are significant and have direct implications for assessing future water availability scenarios and modeling glacio‐hydrology in the region.
... (1961-2015) and precipitation (1951-2007) Note: Significance level: P < 0.05 *, P < 0.01 **, P < 0.001***. Patel et al. 2021) and in the wider Himalaya (Scherler et al. 2008;Quincey et al. 2009;Saraswat et al. 2013;Gantayat et al. 2014;Bhattacharya et al. 2016;Kraaijenbrink et al. 2016;Satyabala 2016;Shukla and Garg 2020). Based on in situ dGPS measurements, Patel et al. (2021) reported higher velocity for Samudra Tapu (64.3 ± 36.7 m/yr) and Sutri Dhaka (52.6 ± 17.3 m/yr) glaciers and comparatively slow movement in Gepang Gath (26.5 ± 12.9 m/yr) and Batal (6.2 ± 2.9 m/yr) glaciers in the Chandra basin. ...
Article
Full-text available
Spatiotemporal surface velocity measurements of the alpine valley type debris-covered Miyar Glacier of the Chandrabhaga (Chenab) basin, western Himalaya, were assessed based on the cross-correlation of Landsat images spanning nearly three decades (1992-2019). Long-term (1950-2015) temperature and precipitation trends were evaluated using Asian Precipitation Highly Resolved Observational Data Integration Towards Evaluation (APHRODITE) datasets. The mean velocity (1992-2019) of the Miyar Glacier is ∼29 m/yr, with spatial patterns revealing that the debris-covered tongue is nearly stagnant (∼5 m/yr) compared to the debris-free up-glacier zone (∼35 m/yr). The transition zone from clean to debris-covered ice in the mid-ablation area shows the highest long-term mean velocities of ∼60 m/yr during the observation period, likely resulting from a steep surface gradient and greater ice thickness than the other regions of this glacier. The slow-moving and nearly stagnant debris-covered area reveals the highest amount of surface lowering due to the expansion of supraglacial ponds. Miyar Glacier experiences summer speed-up of ∼67-80% in seasonal velocity compared to winter, interpreted as a result from enhanced basal sliding during summer months due to warmer temperatures inputting more meltwater into the subsurface drainage system. Inter-annual velocity variations are greatest in the upper glacier, with higher velocities observed more frequently in recent decades. Future work should aim to elucidate the causes of this pattern, considering the overall rising air temperature trend in the western Himalaya.
Article
Full-text available
Meltwater from the cryosphere contributes a significant fraction of the freshwater resources in the countries receiving water from the Third Pole. Within the ESA-MOST Dragon 4 project, we ad-dressed in particular changes of glaciers and proglacial lakes and their interaction. In addition, we investigated rock glaciers in permafrost environments. Here, we focus on the detailed investigations which have been performed in the Poiqu River Basin, central Himalaya. We used in particular multi-temporal stereo satellite imagery, including high-resolution 1960/70s Corona and Hexagon spy images and contemporary Pleiades data. Sentinel-2 data was applied to assess the glacier flow. The results reveal that glacier mass loss continuously increased with a mass budget of −0.42 ± 0.11 m w.e.a−1 for the period 2004–2018. The mass loss has been primarily driven by an increase in summer temperature and is further accelerated by proglacial lakes, which have become abundant. The glacial lake area more than doubled between 1964 and 2017. The termini of glaciers that flow into lakes moved on average twice as fast as glaciers terminating on land, indicating that dynamical thinning plays an important role. Rock glaciers are abundant, covering approximately 21 km2, which was more than 10% of the glacier area (approximately 190 km2) in 2015. With ongoing glacier wastage, rock glaciers can become an increasingly important water resource.
Article
Full-text available
The geodetic method is widely used for assessing changes in the mass balance of mountain glaciers. However, comparison of repeated digital elevation models only provides a glacier volume change that must be converted to a change in mass using a density assumption. This study investigates this conversion factor based on a firn compaction model applied to simplified glacier geometries with idealized climate forcing, and two glaciers with long-term mass balance series. It is shown that the "density" of geodetic volume change is not a constant factor and is systematically smaller than ice density in most cases. This is explained by the accretion/removal of low-density firn layers, and changes in the firn density profile with positive/negative mass balance. Assuming a value of 850 ± 60 kg m<sup>−3</sup> to convert volume change to mass change is appropriate for a wide range of conditions. For short time intervals (&leq;3 yr), periods with limited volume change, and/or changing mass balance gradients, the conversion factor can however vary from 0–2000 kg m<sup>−3</sup> and beyond which requires caution when interpreting glacier mass changes based on geodetic surveys.
Article
Full-text available
Global warming is expected to significantly affect the runoff regime of mountainous catchments. Simple methods for calculating future glacier change in hydrological models are required in order to reliably assess economic impacts of changes in the water cycle over the next decades. Models for temporal and spatial glacier evolution need to describe the climate forcing acting on the glacier, and ice flow dynamics. Flow models, however, demand considerable computational resources and field data input and are moreover not applicable on the regional scale. Here, we propose a simple parameterization for calculating the change in glacier surface elevation and area, which is mass conserving and suited for hydrological modelling. The Δ h -parameterization is an empirical glacier-specific function derived from observations in the past that can easily be applied to large samples of glaciers. We compare the Δ h -parameterization to results of a 3-D finite-element ice flow model. As case studies, the evolution of two Alpine glaciers of different size over the period 2008–2100 is investigated using regional climate scenarios. The parameterization closely reproduces the distributed ice thickness change, as well as glacier area and length predicted by the ice flow model. This indicates that for the purpose of transient runoff forecasts, future glacier geometry change can be approximated using a simple parameterization instead of complex ice flow modelling. Furthermore, we analyse alpine glacier response to 21st century climate change and consequent shifts in the runoff regime of a highly glacierized catchment using the proposed methods.
Article
Full-text available
The first space mission to provide stereoscopic imagery of the Earth's surface was from the American CORONA spy satellite program from which it is possible to generate Digital Elevation Models (DEMS). CORONA imagery and derived DEMS are of most value in areas where conventional topographic maps are of poor quality, but the problem has been that until recently, it was difficult to assess their accuracy. This paper presents a methodology to create a high quality DEM from CORONA imagery using horizontal ground control derived from Ikonos space imagery and vertical ground control from map-based contour lines. Such DEMS can be produced without the need for field-based ground control measurements which is an advantage in many parts of world where ground surveying is difficult. Knowledge of CORONA image distortions, satellite geometry, ground resolution, and film scanning are important factors that can affect the DEM extraction process. A study area in Syria is used to demonstrate the method, and Shuttle Radar Topography Mission (SRTM) data is used to perform quantitative and qualitative accuracy assessment of the automatically extracted DEM. The SRTM data has enormous importance for validating the quality of CORONA DEMS, and so, unlocking the potential of a largely untapped part of the archive. We conclude that CORONA data can produce unbiased, high-resolution DEM data which may be valuable for researchers working in countries where topographic data is difficult to obtain.