13)

    Glacier changes in the Ravi basin, north-western Himalaya (India) during the last four decades (1971–2010/13) Pritam Chand, Milap Chand Sharma PII: DOI: Reference:

S0921-8181(15)30095-3 doi: 10.1016/j.gloplacha.2015.10.013 GLOBAL 2349

To appear in:

Global and Planetary Change

Received date: Revised date: Accepted date:

17 March 2015 12 October 2015 22 October 2015

Please cite this article as: Chand, Pritam, Sharma, Milap Chand, Glacier changes in the Ravi basin, north-western Himalaya (India) during the last four decades (1971–2010/13), Global and Planetary Change (2015), doi: 10.1016/j.gloplacha.2015.10.013

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Glacier Changes in the Ravi basin, North-Western Himalaya (India) during

PT

the last four Decades (1971-2010/13)

RI

Pritam Chand* and Milap Chand Sharma

Centre for the Study of Regional Development (CSRD), Jawaharlal Nehru University (JNU),

SC

New Delhi 110067, India

NU

*Contact No.: +91-9650966260 & *e-mail for the corresponding author: [email protected]

MA

Glacier Changes in the Ravi basin, North-Western Himalaya (India) during

D

the last four Decades (1971-2010/13)

TE

Abstract: A glacier inventory of the Ravi basin, north-western Himalaya has been generated for the year 2002 using Landsat ETM+ and ASTER Global DEM (GDEM V2) as the baseline data

AC CE P

for the change analysis. The Ravi basin consists of 285 glaciers (> 0.02 km2) covering an area of 164.5 ± 7.5 km2, including 71 debris-covered glaciers with an area of 36.1 ± 2.1 km2 (22 % of total glacierised area) in 2002. Change analysis based on Corona KH-4B (1971), Worldview (2010) and Landsat 8 OLI/ TRIS (2013) images was restricted to a subset of 157 glaciers (covering an area of 121.4 ±5.4 km2 in 2002) due to cloud cover. Glacier area decreased from 125.8 ± 1.9 km2 (1971) to 119.9 ± 4.8 km2 (2010/13), a loss of 4.7 ± 4.1 % or 0.1 ± 0.1 % a-1. The glacier recession rate has decreased, to a minimum for the recent decades (2002-2010/13). The debris-covered glacier area increased by 19.24 ±2.2 % (0.5 ±0.05% a–1) in the Ravi basin. However, there were significant variation in its sub-basins i.e. in Budhil and Upper Ravi subbasin, where the debris-covered area increased by 28.6 ±3.1% (0.7 ±0.1% a–1) and 14 ±1.6% (0.3 ±0.04% a–1), respectively, between 1971 and 2010/13. Field investigation of selected 1

ACCEPTED MANUSCRIPT glaciers (2010-2014) support glacier recession trend from remote sensing data. Glacier retreat rates in the Ravi basin were lower than previously reported for selected glaciers in the similar

PT

basin and other basins (e.g. Chenab, Beas, Parbati, Baspa and Tirungkhad) of the Himachal

RI

Himalaya.

SC

Keywords: Glacier Changes; ASTER GDEM; Debris-covered glaciers; Remote sensing;

NU

Himachal

MA

1. Introduction

The glaciers and snow cover of the Himalaya-Karakoram (H-K) region play significant roles in

D

the regional climatic system. The H-K region has the largest concentration of snow cover and

TE

glaciers outside the polar regions, with a total glacier cover of ~40,800 km2 (Bolch et al., 2012)

AC CE P

involving 9575 glaciers in the Indian Himalaya (Raina and Srivastava, 2008). The glaciers and snow cover of the H-K region influence the overall runoff in lowland rivers, recharge river-fed aquifers, provide water for hydro-power, agriculture, ecosystems and eventually contribute to regional/global sea-level (Dyurgerov and Meier, 2005; Thayyen and Gergan, 2010). Glaciers being physically complex and dynamic systems are sensitive to climate changes (Benn and Evan, 2010). These ice bodies are key indicators for assessing climate change in the Himalayan region, especially areas without having climatic observations instrumentation facility (Barry, 2006; Bolch et al., 2012). Since the end of the Little Ice Age (LIA) (~1850s), Himalayan glaciers have been in a general state of recession (Bhambri and Bolch, 2009; Mayewski and Jeschke, 1979). Himalayan glaciers studies indicate that many glaciers show an increased, receding-trend over the past few decades (Bolch et al., 2008; Kulkarni et al., 2007), and many glaciers have stable

2

ACCEPTED MANUSCRIPT fronts since 2000 (Bahuguna et al., 2014; Bhambri et al., 2013). Glaciers in the Karakoram region, show long-term irregular behavior with frequent advances and possible slight mass gain

PT

since 2000 (Bhambri et al., 2013; Bolch et al., 2012; Gardelle et al., 2013; Hewitt, 2011; Kääb et

RI

al., 2012). This irregular behavior of Himalayan glaciers in general could be attributed to local/regional topography (Oerlemans 1989; Haeberli, 1990), local/regional climatic system

SC

(Kargel et al., 2005), glacier hypsometry (Furbish and Andrews, 1984), the characteristics and

NU

thickness of supraglacial debris cover on the glacier surface (Bolch et al., 2008; Scherler et al., 2011) ,the glacier size and ratio of accumulation area to total area (Kulkarni et al., 2007),

MA

contributions from tributary glaciers (Nainwal et al., 2008) and their geometrical/morphological properties (Mehta et al., 2014).

TE

D

Glacier monitoring typically involves ground-based measurements, aerial reconnaissance, and satellite imaging (Bhambri and Bolch, 2009; Hubbard and Glasser, 2005). Although field

AC CE P

investigation is an essential component of glaciology to cover the complexity of a glacier system and is highly recommended, given the large magnitude of investigation and the difficult terrains of glaciers, only a limited number of glaciers can be investigated in field by glacier terminus change measurements (e.g. Chand and Sharma, 2015; Mehta et al., 2011) and mass balance studies (Pratap et al., 2015a). Multi-temporal and multi-spectral remotely sensed data enable mapping and monitoring of glaciers with large spatial scales at regular temporal intervals (Bolch et al., 2010; Paul et al., 2013; Racoviteanu et al., 2009). Glaciers changes for most of the basins located (e.g. Beas, Chenab and Sutlej) in the Himachal Himalaya have been measured using historic glaciers outlines either using topography maps from the Survey of India (SoI) or coarser spatial resolution satellite datasets (e.g. Landsat MSS) (Kulkarni et al., 2011;Kulkarni et al., 2007; Kulkarni and Rathore, 2005; Pandey and

3

ACCEPTED MANUSCRIPT Venkataraman 2013; Mir et al., 2013). However, many studies have reported inaccuracies in representation of glacial extent on the SoI topographical maps (Bhambri and Bolch, 2009;

PT

Chand and Sharma, 2015; Raina, 2009). It is further observed that the coarser resolution

RI

satellite datasets (e.g. Landsat MSS) make it difficult to map glacier terminus accurately, especially in the case of debris-cover region. Thus, the declassified high spatial resolution

SC

imagery of Corona and Hexagon acquired during the same period of 1960's and 1970's provides

NU

great potential to derive the historic glacier outlines for comparison with contemporary glacier outlines derived from high resolution satellite images (Bhambri and Bolch, 2009; Bolch et al.,

MA

2010). In addition, best of our knowledge there is no published study on the Ravi basin addressing glacier change in association with other variables (e.g. change in debris cover) using

TE

D

high resolution Corona images except for the two glaciers i.e. Manimahesh and Tal glaciers of Budhil by the Geological Survey of India (GSI, 2005) based on the SoI topographic maps and

AC CE P

field observation. Thus, the main goals of this study are to: i) generate a complete and up-to-date glacier inventory for the Ravi basin using Landsat ETM + (2002) images (aided by ASTER Data of 2002) and documentation of glacier characteristics to the GLIMS data base and the Randolph Glacier Inventory (RGI; Arendt et al., 2014); ii) analyse glacier area changes in the Ravi Basin over the past 40 years; and iii) elucidate the possible impacts of climate variables on glacial changes in the study area.

2. Study Area The Ravi basin is located in the south-eastern part of Chamba district and north-eastern part of Kangra district in Himachal Pradesh, India. It covers, three sub-basins of the Ravi River: Siul, Budhil and the upper Ravi (Fig.1). On the basis of geographical division of Himalaya, it is located in the Himachal Himalaya which is regionally a part of north-western Himalaya. Ravi is 4

ACCEPTED MANUSCRIPT a major tributary of the river Indus. It originates from the Raigarh Glacier (snout ~4050 m a.s.l.). It is the second largest valley glacier (8.8 km2) in the Bara-Bhangal region of Pir-Panjal range

PT

(Fig. 1b). The Ravi flows in a north-west direction for most of its course. It follows the strike between two parallel mountain ranges of the Dhaula-Dhar to the south and Pir-

RI

Panjal to the north (Marh, 1986). The total study area is ~4900 km2 with altitude varying from

SC

765 to 6125 m a.s.l. (Fig.1b). The geomorphic history of the Ravi basin can be associated with

NU

the Pleistocene to Holocene climatic changes (Marh, 1986). The Balu is a largest glacier of the Ravi basin with an area of 10.3 km2, located in the south-east Pir-Panjal range of the Ravi basin

MA

(Fig.1b).

The climate of the study area is transitional between the winter-dry climate of the Indo-Gangetic

TE

D

plains (CWg by Koppen’s classification) and the highland climates (H) of the western Himalaya (Spate and Learmonth, 1967). The basin lies between the transition zone of the maximum

AC CE P

(Dharamshala, Kangra region) and minimum (Lahaul) precipitation areas of Himachal Pradesh (Bookhagen et al., 2006; Marh, 1986; Fig 1). The northern aspect of the Dhaula-Dhar range is in a rain shadow, and therefore, experiences minimum precipitation as compared to southern aspect of the Pir-Panjal range. For example, a small town Holi (~1830 m a.s.l.) located in the rain-shadow has experienced an average of 816 mm annual precipitation, whereas Bharmour (~ 2150 m a.s.l.) in the Pir-Panjal ranges has experienced an average of 1180 mm annual precipitation as an average of 20-25 years (Bhagat et al., 2004). The mean annual air temperature (MAAT) is 19.54 C0 at the Chamba station (~ 920 m a.s.l., situated on the bank of Ravi river) and the annual precipitation is 861 mm (Pareta and Pareta, 2014) (Fig. 1c). The highest amount of rainfall occurs in July with >20% of the annual average. The Indian summer monsoon (June to September) accounts for 56.9 % (486.4 mm a-1) of the average annual rainfall, while mid-latitude

5

ACCEPTED MANUSCRIPT westerlies (November to March) contribute 32.4 % (278.5 mm a–1) as recorded in the period 1990– 2013 (Pareta and Pareta, 2014). The hydrology of the Ravi basin is mainly controlled by

PT

spring snowmelt and the summer monsoon rainfall (HPSEB, 2004). Field investigations and

RI

satellite images indicate that most of the valley glaciers in the study area have extensive supra-

SC

glacial debris in the ablation zone (Fig. 1b).

3. Methods

MA

3.1 Data Source and Image Rectification

NU

Insert Figure 1

Glacier mapping, inventory and change analysis was derived from several temporal, multi-

TE

D

spectral and medium to high resolution satellite image sources: i) Corona KH-4A, ii) Landsat 4/5 Thematic Mapper (TM) iii) Landsat 7 Enhanced Thematic Mapper (ETM+), iv) Landsat8

AC CE P

Operational Land Imager (OLI), v) Advanced Space borne Thermal Emission and Reflection Radiometer (ASTER), and vi) WorldView-2 (Table 1). The Corona and Landsat datasets were acquired from the United States Geological Survey (USGS; http://earthexplorer.usgs.gov/). Most of the satellite images were selected for the end of the ablation period with minimum seasonal snow cover and a high solar position to avoid deep shadows (Paul and Svoboda, 2009, Table 1). In addition, the SoI topographic maps (52 D 11 and 15, 1963 and 1968 edition) at 1 : 50,000 scale with 40 m contour interval (planimetric accuracy ±12.5 m and elevation accuracy ±6.5 m) (Raju and Ghosh 2003) were used for comparison of glacier outlines from the SoI topographic maps and high resolution Corona image for pervious studied glaciers (e.g. Manimahesh and Tal glaciers). The ASTER GDEM V2 (30 m spatial resolution) from Japan Space Systems (http://gdem.ersdac.jspacesystems.or.jp/) was used as reference DEM for

6

ACCEPTED MANUSCRIPT semiautomatic delineation of drainage basin and extraction of glacier topographic parameters. Seven Corona KH-4B images (image frame ~14 km х 188 km) of 27 September 1971 were used

PT

to extract the historic extent of glaciers. The Landsat TM images of 1989 were used to map the

RI

extent of glaciers in ~1990s. The Landsat ETM+ L1T images (panchromatic and multi-spectral) for the year 2002 were selected as base line datasets for glacier inventory. This is close to the

SC

2000CE standard which is recommended for a global glacier inventory as baseline information to

NU

facilitate global glaciological applications (Paul et al., 2009). ASTER image (2002) was additionally used to assist and identify the glacier terminus from Landsat ETM+ (2002) images.

MA

The 2013 ortho-rectified Landsat 8 OLI/TRIS L1T data were used to map latest glacier outlines. A high resolution WorldView-2 (2010) image with limited swath (~16.4 km at nadir) facilitated

D

to accurately identify glaciers terminus for 28 glaciers in the upper Ravi sub-basin (Fig. 1b). All

TE

the selected Landsat TM/ETM+ imageries are available as ortho-rectified in the processing level

AC CE P

L1T, but the Landsat TM image of 1989 was only processed to L1G (±250 σ). This latter image shows a slight horizontal shift of a pixel (∼30 m) as compared to 2002 Landsat ETM+ reference image; hence, it was co-registered to the reference image using the projective transformation algorithm available in Erdas Imagine 10 (Bhambri et al., 2011). This algorithm includes DEM (i.e. ASTER GDEM V2) which refine the co-registration process as compared to simple polynomial method. Image to image rectification was also carried out for WorldView-2 by Landsat ETM+ PAN imagery using 32 GCPs. Twelve common geographical points were identified on both the images to assess positional accuracy of WorldView-2. The horizontal shift between base Landsat ETM+ PAN image and corresponding WorldView-2 (2010) image was 4.9 m (1.9 pixel). Insert Table 1

7

ACCEPTED MANUSCRIPT

Corona images are acquired using a non-metric panoramic camera on a satellite which scans the

PT

selected area from one side to the other (Goossens et al., 2006). The image scale is depended upon the nadir angle (Beck et al., 2007). Most of Corona image strips used in the Ravi basin

RI

were situated near this nadir point minimising distortion in terms of the area on the extreme side

SC

of the strip. A spline adjustment method inbuilt in ESRI ArcGIS 10.1 was used as a geo-

NU

referencing method for Corona images. This provides results within an acceptable error (< 15 m) (Bhambri et al., 2011). Owing to a complex geometry and large size (~1GB) of the Corona

MA

image, 12 subsets of 7 corona images for the entire Ravi basin were generated. Stable roads/roads intersection, stream junctions, edges of rocky outcrops and prominent linear

D

geological structure (e.g. lineaments) were used for GCP’s. For each generated Corona subset,

TE

35 to 212 GCPs were acquired from Landsat PAN Imagery (2002) for co-registration (Fig. 2). In

AC CE P

addition, to assess positional accuracy, 24 common location points such as stable river junctions were identified carefully in both Landsat ETM+ PAN image and Corona subsets. The horizontal shift between the both images was 5.3 m (1.8 pixels). Insert Figure 2

3.2 Glacier Mapping, Inventory and Changes The classification of the Global Land Ice Measurements from Space (GLIMS) initiatives (http://www.glims.org/MapsAndDocs/guides.html) were adopted to map the glacier boundaries from satellite images. The area above the bergschrund and ice bodies not directly connected to the glacier were excluded during the mapping of glacier outlines as they contribute infrequent and indirect snow and ice to the glacier mass through avalanches and creep flow (Bhambri et al., 2011; Racoviteanu et al., 2008). The clean glacier ice areas were mapped from all the Landsat 8

ACCEPTED MANUSCRIPT images by the well-established semi-automated TM3/TM5 (RED /SWIR) band ratio approach followed by a 3×3 median filter to eliminate isolated pixels (Bhambri et al., 2013; Bolch et al.,

PT

2010; Frey et al., 2012; Racoviteanu et al., 2008). Binary images were generated using threshold

RI

values between 1.8 and 2.1, and then converted into vector polygon. The resultant glacier vector polygons were visually checked and manually improved to eliminate misclassified pro-glacial

SC

lakes, seasonal snow cover and shadow areas. The minimum size of mapped glaciers included in

NU

our inventory was 0.02 km2 as per Frey et al. (2012) and Bajracharya and Shrestha (2011). Manual digitization was preferred to map the debris-covered glacier areas as spectral signature

MA

on satellite images does not differentiate debris-covered ice with surrounding bare moraines (Bhambri et al., 2011; Paul et al., 2009). Identification of debris-covered glaciers termini was

TE

D

rather difficult using Landsat TM/ETM+/OLI imagery alone. Therefore, additional high resolution PAN-sharpened multispectral images of Landsat ETM (2002) /Landsat 8 OLI (2013)

AC CE P

and DEM (ASTER GDEM V2, using geo-morphometric techniques) were used for identification of the glacier terminus. A Brovey transform image fusion technique was used to merge three multispectral bands with the higher spatial resolution panchromatic band of Landsat 7 ETM+ and Landsat 8 OLI image (Bahuguna et al., 2004; Csatho et al., 2005). The aim was to identify the glacier terminus and its morphology by increased the spatial resolution from 30 m to 15 m. High resolution datasets of ASTER (2002) and WorldView-2 (2010), Bhuvan 2D/3D (India web-based GIS tool, http://bhuvan.nrsc.gov.in/bhuvan_links.php#) and Google Earth (GE) were used as additional sources to improve the glacier outlines of debris-covered glaciers (e.g. Paul et al., 2013). We used several signs of movement (identified based on overlays of multi-temporal images); emerging melt water streams at the end of the terminus; breaks in surface slope; spectral color differences; and the presence of small melt water ponds for determination of the

9

ACCEPTED MANUSCRIPT most likely position of the glacier termini (e.g. Bhambri et al., 2013; Bolch et al., 2010). The historic glacier extents from 1971 Corona images were manually delineated. Field investigations

PT

of selected glaciers (e.g. Shah, Manimahesh, Tal, Laihas, Sarni, Thamsar and Tapni) using

RI

handheld non-differential Global Positing System (GPS, Garmin GPSMAP 76Cx ± 5-10 m) , field photographs and field mapping of recessional moraines were also facilitated the

SC

determination of glacier termini between 2010-14.The contiguous ice masses were separated into

NU

their drainage basins using semi-automated approach for hydrological divides using ASTER GDEM V2, as proposed by Bolch et al. (2010) and further checked and manually improved

MA

where required. We assumed that the divides remained fixed over the time under investigation. This approach avoids errors that may occur due to the different delineation of the upper glacier

TE

D

boundary, e.g. under different snow conditions (Bhambri et al., 2013). Topographic information (e.g. mean, median, slope, aspect) was computed from the ASTER GDEM V2 using Geographic

AC CE P

Information System (GIS) tools as suggested by Paul et al. (2009). The characteristics of glacier distribution were examined by statistically analyzing the relations between topographic parameters and glacier area. Change detection of glacier between 1971 and 2010/2013 was carried out on 157 glaciers. Rest of the glaciers remained obscured by cloud cover on the Corona images (1971). The Landsat TM scenes for the years 1989 was partially suitable for glacier mapping due to seasonal snow cover. Thus, a subset of 54 glaciers, out of the 157 glaciers, were compared for change detection between 1971- 1989 and 1989-2002 (Fig. 1b). This glacier subset is considered representative as its members are in different area size classes between 0.05 and 8.3 km2 (2002), and elevation ranges from 3938m to 5182m a.s.l.. For glacier change analysis, we overlaid the glacier outlines of 2002 on the corona images (1971) as proposed by previous studies (Bhambri et al., 2011; Bolch et al., 2008; T. Bolch et al., 2010).

10

ACCEPTED MANUSCRIPT The overlay adjustments were restricted to the lower part of the glaciers as no visible changes could be identified in the upper accumulation region, and snow cover often hamper correct

PT

identification at high-elevation areas (Bhambri et al., 2013; Bolch et al., 2010). Several rock

RI

glaciers has been identified above 3500 m a.s.l. during field in the study area (Fig. 3a). Similar observation of rock glaciers reported in other regions of H-K by Owen and

SC

England (1998). Frey et al. (2012) excluded rock glaciers from the glacier inventory; likewise

NU

rock glaciers were not included in our glacier inventory. Rock glaciers were differentiated from debris-covered glaciers, mainly by the presence of tongue-like or lobate shapes, or a missing

MA

well-defined accumulation area, or ridge-and-furrow surface structures and steep termini at or near the angle of repose that can be identified in satellite images (Elizabeth Martin and Whalley,

Insert Figure 3

AC CE P

TE

D

1987; Rangecroft et al., 2014; Scotti et al., 2013).

3.3 Mapping uncertainty

The present study used different spatial and spectral resolution datasets for different time period with varying conditions at the time of the acquisition (e.g. seasonal snow, shadow and clouds). Therefore, the issue of uncertainty and its propagation in glacier mapping based on remote sensing deserves proper consideration and it is crucial to ascertain the accuracy and significance of the results. Previous studies reported a mapping uncertainty of ± 2–4% for clean-ice glaciers (Paul et al., 2002) and while in perfect conditions, some studies achieved less than half-pixel accuracy (Bolch et al., 2010). We used the 2002 ETM+ scene with a 15 m panchromatic band as the basis of the inventory and estimated the mapping uncertainty for each glacier based on a

11

ACCEPTED MANUSCRIPT buffer around the glacier margins as suggested by Granshaw and Fountain (2006) and Bolch et al. (2010). The buffer size was chosen to be half of the estimated shift caused by misregistration

PT

as only one side can be affected by the shift (Bolch et al., 2010). Therefore, 7.5 m buffer size

RI

used for the outlines of the Landsat 7 ETM+, Landsat 8 OLI and further considered same for

SC

Landsat 5 TM as assumed the uncertainty is similar for the mapped outlines of the TM scenes (Table 1). Besides, a smaller buffer size of 2.5 m was used for the Corona and WorldView-2

NU

images by taking the positional accuracy (e.g. 5.2 m and 4.9 for Corona and WorldView-2 respectively) into account. This method includes the relative higher error of small polygons as a

MA

small glacier has relatively more edge pixels (Bolch et al., 2010b). The average mapping uncertainty was 4.5 % for the base 2002 ETM+ image, 2.6 % and 4.6 % for the 1989 TM and

TE

D

2013 Landsat 8 OLI images, respectively and 1.5 % and 1.3 % for the 1971 Corona images and 2010 WorldView-2 images respectively. The lower mapping uncertainties for 1989 and 2010/13

AC CE P

(4 %, average of glaciers boundary mapped from WorldView-2 and Landsat 8) are mainly due to the lower (54) and, on average, larger glaciers for the area change analysis. Further, uncertainty was evaluated by comparing the glacier outlines derived from the multispectral bands with 30 m resolution of a 2010 Landsat TM image, and those derived from a high resolution WorldView-2 image of the same year for a sample of six selected glaciers as suggested by Paul et al. (2013). The resultant uncertainty was ±2.4 % which is within the range of the uncertainties reported by previous studies (Bhambri et al., 2013;Paul et al., 2013). The area change uncertainty is estimated according to standard error propagation, as the root sum square of the uncertainty for the outlines mapped from different sources (Bhambri et al., 2013). Using this approach, both the uncertainty of the upper glacier area, where the area changes are usually minor, but could not be adjusted in some instance due to adverse snow conditions, and the mapping uncertainty of

12

ACCEPTED MANUSCRIPT outlines of the individual years could be addressed (e.g. Bhambri et al., 2013). The location uncertainty for the common outlines of upper accumulation parts are insignificant and well

PT

within the uncertainty calculated with the buffer method (e.g. Bolch et al., 2010; Bhambri et al.,

RI

2013).

SC

4. Results

NU

4.1 Glacier inventory and characteristics

There are 285 glaciers identified in the Ravi basin from Landsat ETM+ images (2002), covering

MA

an area of 164.5 ± 7.5 km2 (Fig. 1b and Table 2). Out of 285 glaciers, 71 have debris-covered tongues and comprising 36.1 ± 2.1 km2 (22 % of the total glacierised area). A range of small to

D

large valley glaciers are identified in the basin, ranging in area from 0.02 to 10.3 km2 with a

TE

mean size of 0.6 km2. The mean size of glaciers in the study area is comparatively lower than

AC CE P

other glacierised basins of the H-K region e.g. Shyok (1.4 Km2), Chenab (1.1 km2), Bhagirathi (1.3 km2), Saraswati/Alaknanda basin (3.7 km2), Ganga (1.1 km2) and Brahmaputra (1.2 km2) (Bajracharya, and Shresta, 2011; Bhambri et al., 2013; Frey et al., 2012).Out of the total 285 glaciers, <1.0 km2 comprise 87.4 % (249) of the total number, but contribute only 39.2 % of the total area (Fig. 4). The > 5 km2 size class has 5 glaciers, which is only 1.8 % of the total number of glaciers, but this class covers 35.6 km2 ± 1 km2 (21.7 % ) (Fig. 4). This conforms to the pattern in other regions of the H-K (Bajracharya, and Shresta, 2011; Bhambri et al., 2013; Frey et al., 2012). Most of the larger glaciers (>5 km2) are located in the Pir-Panjal range of the Upper-Ravi sub-basin and their presence can be attributed to the topographical factors e.g. high altitudinal range, extensive catchment morphology and the precipitation gradient as the rainfall convert into snowfall in the upper higher reaches and nourishment the accumulation area of the glaciers.

13

ACCEPTED MANUSCRIPT

PT

Insert Table 2 & Figure 4

RI

The mean elevation of the glaciers ranges from 3941 to 5822 m a.s.l., an average of 4828 m a.s.l.. Moreover, the median elevation of the glaciers, which is widely used for the estimation of

SC

long-term ELA based on topographic data (Braithwaite and Raper, 2009) , is similar to the mean

NU

elevation, with ranges from 3896 to 5859 m a.s.l., and an average of 4828 m a.s.l. (Fig. 4 & 6d). The mean elevation is lower than that found in central and western Himalayan basins like

MA

Alaknanda (5380 m), Bhagirathi (5544 m), Yamuna (5083 m), Sutlej (5436 m), Chenab (5064 m), Indus (5404 m), and Shyok (5868 m) (Frey et al., 2012). Additionally, the mean elevation

TE

D

for different glacier size class varies as it is higher for the smaller size glacier as compare to large glacier (Fig. 4).In addition, the elevation range also vary according to glaciers size, at a

AC CE P

maximum for larger glacier and minimum for small glaciers (Fig. 6b). It infers that the larger valley and compound valley glaciers tend to extend down to lower elevations, while smaller glaciers have higher elevated termini. Glaciers with size class of 2-5 km2 have lower mean elevations (~4629 m) as compared to other size class glaciers which further suggest their sensitivity to warming climate as there terminus at lower elevation. The average elevation versus east /west position (longitudinal) shows that elevation tended to slightly increase to the east (Fig. 5b). It suggested that the spatial distribution of the glaciers follow the basin topography as it increase towards the eastern side of the study area. A similar plot for elevation versus north /south position (latitudinal) showed a slightly more complex pattern, but the overall pattern shows the increase trend of elevation towards the highest peaks of the Pir-Panjal range (Fig. 5b). Moreover, the climate dependence becomes noticeable when looking at the spatial distribution of

14

ACCEPTED MANUSCRIPT the glacier‐specific mean elevation values, as the precipitation gradient decreases towards the inner eastern part in the Ravi basin (Bookhagen et al., 2006). The distribution of glacier area by

PT

elevation (i.e. the hypsometry) of total glacier, clean ice, debris-covered ice surface and sorted by

RI

glacier size class is depicted in Fig. 5a. Most of the glaciers in the different size class are distributed between 4600 and 4800 m a.s.l. with smaller size-class glacier generally at higher

SC

elevation compared to large size glacier. In addition, the glaciers in size class >5 km2 having

NU

another peaks in hypsometry curve at different elevations (below ~4500 m) because of debriscovered parts, which account for 37.1 % of glacier area in this size class (Fig. 5a). Moreover, it

MA

has been reported that thick debris-cover is one of the controlling factors in the glacier dynamics and lowers the steady-state Area Accumulation Ratio (AAR), and thus a larger ablation area is

AC CE P

Evan et al. 2010).

TE

D

required to balance accumulation which further influenced the glacier hypsometry (Benn ant

Insert Figure 5

Glaciers are located on moderate slopes averaging 27.3°. Similar average slopes were reported by Frey et al. (2012) for glaciers in other basins of north-western Himalaya with range from 23.9° to 30.9°. The smaller glaciers are on a higher slope as compare to the large valley glaciers in the basin (Fig. 4 and 6c). It suggested that the larger valley glaciers (> 5 km2) with gentle slope having most of the ice volume as compare to the glaciers with having steeper slope (Frey et al., 2014, 2012). The maximum number of glaciers are oriented towards west (~24.7 %) followed by south (~24.4 %) and southwest (~16.6 %). On an average more than half of the glaciers (~51.6 %) are located on southward aspect (Fig. 6a). The debris-covered glacier area

15

ACCEPTED MANUSCRIPT percentage in the Ravi basin (22 %) and its sub-basin, e.g. Budhil (23.7 %) and Upper Ravi (22.3 %) is comparatively higher than the others basins in north-western Himalaya (average of all

PT

basins 15 %) e.g. Shyok (2%), Indus (9 %), Jhelum (12 %), Chenab (16 %), Sutlej (10 %),

RI

Yamuna (16 %) and Bhagirathi (18 %) (Frey et al., 2012). The higher percentage of debris cover can be attributed to the steepness of the ice-free zone above the snow line, surface conditions of

SC

the headwall where the debris mantle originates, the higher glacier catchment relief, and the

NU

spatial distribution of potential debris supply (PDS) slopes (Nagai et al., 2013; Regmi and Watanabe, 2009; Scherler et al., 2011). Kääb, (2005) and Nagai et al., (2013) observed in Bhutan

MA

Himalaya that glaciers south of the main range having considerably higher amount of debriscovered area than northbound glaciers and significant correlation between surface area of

TE

D

southwest-facing PDS slopes and debris-covered area, respectively. To investigate whether this relationship of debris cover exists in our study region, we averaged spatial aspect distribution of

AC CE P

PDS slope for the glaciers which having more than 15 % debris cover area to the total area. It found that more than 60.8 % of PDS slope for these glaciers (> 15% debris cover area) are in south-west or south facing. Thus, the suggested explanation of more intensive debris supply from the PDS slope and steep rock faces surrounding the glaciers in the south-west or south might apply here too. Besides, the surface temperatures on these slopes (south-west or south) fluctuate around the melting point, providing large amounts of debris under the influence of the diurnal freeze–thaw cycle as same reported in Bhutan Himalaya (Nagai et al., 2013).

Insert Figure 6

4.2 Glacier area change

16

ACCEPTED MANUSCRIPT The present study observed significant loss in glacier area since 1971 in the Ravi basin. During the period 1971-2010/13, glacier area changed from 125.9 ± 1.9 km2 (1971) to 120 ± 4.8 km2

PT

(2010/13), a decrease of 5.9 ± 5.2 km2 or 4.7 ± 4.1 % (0.14 ± 0.12 km2 a-1 or 0.1 ± 0.1 % a-1)

RI

(Table 3 & Fig. 7, 8). The number of glaciers slightly increased from 157 (1971) to 159 (2010/13) due to the disintegration of glaciers (Fig. 7). Loss in glacier area ranged from 0 %

SC

to 43.7 % during 1971-2010/13. At the sub-basin level, the glacierised area decreased by a-1

) and 3.5 ± 2.8 km2

NU

0.2 ± 0.2 km2 (0.004 ±0.004 km2 a-1), 2.2 ± 2.2 km2 (0.05 ± 0.05 km2

(0.08 ± 0.07 km2 a-1) for Siul, Budhil and Upper-Ravi during the same period. The debris-

MA

covered glacier area increased by 19.2 ± 2.2 % (0.5 ± 0.1 % a–1) between 1971 and 2010/13. In total, 49 glaciers were covered with debris in 1971, increasing to 56 in 2010/13. In the sub-basins

TE

D

of Ravi, the change rate of debris-covered glacier area is varying (Table 3). The increase in debris-cover could be attributed to higher ratio of PDS slope area to the total glacier area and

AC CE P

higher glacier catchment relief especially for the large valley glacier (e.g. Tal, Shah, Laihas, Tapni and Raigarh, Fig. 1b). Several other studies also reported increased of debris-covered area across the Himlayan region e.g. Bhambri et al. (2011), Bolch et al. (2008) and Pratap et al. (2015).

Insert Table 3 &Figure 7

Decadal glacier changes examined in details for the 54 mapped glaciers (1989) in the Ravi basin. These glaciers (~54) lost 1.9 ± 2.8 km2 (2.7 ± 3.9 % or 0.2 ± 0.2% a–1) of their area from 1971 to 1989, 1.4 ± 3.8 km2 (1.9 ± 5.4 % or 0.2 ± 0.4% a–1) from 1989 to 2002, and 0.9 ± 3.4 km2 (1.3 ± 5.0 % or 0.2 ± 0.4% a–1) from 2002 to 2010/13. It shows that the rate of glacier recession has

17

ACCEPTED MANUSCRIPT decreased during the last decade i.e. 2002-2010/13 as compared to 1989-2002 as reported for

PT

other glaciers across the Himalayan region (Bahuguna et al., 2014).

RI

Insert Figure 8

SC

In the Ravi basin, during 1971-2010/13, glaciers with area between 2-5 km2 lost maximum

NU

percentage of their area (6.2 ± 3.0 % or 0.2 ±0.1% a-1), followed by glaciers with areas <1 km2 (5.5 ± 5.7 % or 0.1 ± 0.1% a-1), and glaciers with areas >5 km2 (1.5 ± 2.6 % or 0.04 ±0.06% a-1)

MA

(Fig. 8). The glaciers sized between 2-5 km2 witnessed comparatively higher area loss due to their lowest mean/minimum mean elevation (i.e. compartively lower terminus elevation) and

D

higher percentage of clean-ice area (45.5 %). Moreover, significant variation occurs among the

TE

glaciers sized <1 km2, with maximum retreat rate for the smaller glacier with clean-ice cover

AC CE P

(Fig. 8). During 1971-2010/13, the total glaciers, clean ice glaciers and debris-covered glaciers lost 4.7 ± 4.1 % (0.1 ±0.1 a-1), 6.9 ± 4.5 (0.2 ±0.1 a-1) and 3 ± 3.8 % (0.1 ±0.1 a-1) respectively (Table 3). This show that the clean ice-covered glaciers lost with amost twice annual rate than debris-covered glaciers in the Ravi basin. Overall, results show variation in retreat rate among the different clean ice-covered glacier size clases, being higher for the small clean ice-covered glacier compared to large ice-covered glaciers. However on an average, the glaciers with extensive debris-covered (>40-50 % of their total area) (e.g. Tal, Raigarh, Shah glaciers) areas in their ablation zones were in the minimum loss percentage of glacial area as compared to clean ice-covered or less debris-covered glaciers (<10-15 % of their total area) (e.g. Manimahesh, Nlkora, Kalihen glaciers). Scherler et al. (2011), also reported lower reccession rates and stable fronts of heavily debris-covered glaciers in the Himalayan region.

18

ACCEPTED MANUSCRIPT The 157 glaciers (2002) mapped in the Ravi basin were at elevations ranging from 3482 to 5860 m a.s.l. with >50% of the glacier area distributed at elevations of 4600–5000 m a.s.l.

PT

(Fig. 9 c,d). The glaciers with median elevation range from 4800 to 5000 m a.s.l. show

RI

significant loss in their area, followed by those in the 5000-5200 m a.s.l.. It indicates that the glaciers with median elevations closer to their maximum elevations or average glaciers median

SC

elevation (~4828) of the basin are losing more area due to having comparatively smaller

NU

accumulation area (Bajracharya et al., 2014). During 1971-2010/13, glaciers with southward aspects (including southeast, south and southwest) had change rates of -4.5 ± 3.7 % (-6.7 ± 4.4

MA

%, -4 ± 4.2 % and -4.1 ± 3.2% respectively) while glaciers with northward aspects (including northeast and northwest) had change rates of -5.4 ± 7.9% (-12.5 ± 8.4 % and -3.3 ± 7.7 %

TE

D

respectively) (Fig. 9b). In addition, the glaciers with west and east aspect had change rates of -4.4 ± 4.5% and -10.04 ± 5.2 % respectively. Although, in terms of absolute area change,

AC CE P

glaciers with southwest aspects lost maximum area of 1.8 ±1.4 km2, followed by glaciers with west (-1.5 ±1.5 km2), south (1.2 ±1.3 km2) and south east (0.8 ±0.5 km2) aspect. The change rate of northwest and east facing glaciers had the largest acceleration because all 5 northwest-facing glaciers and 7 out of 11 east facing glaciers were smaller than 0.2 km2 and covered with clean ice, while their retreat rate changed significantly during the last four decades. The glaciers with slope between 150 and 250 show comparatively higher retreat rate (Fig. 9a). These above factors indicates that, the response of the glacier to changes in the climate is largely governed by individual glacier morphology/shape (e.g. length, slope and orientation), their size, mean elevation and characteristics of clean ice and debris cover on their surface in the study area.

Insert Figure 9

19

ACCEPTED MANUSCRIPT

5. Discussion

PT

5.1 Glacier inventory and changes

RI

Our study presents an improved glacier inventory for the Ravi basin, based on excluding

SC

seasonal snow patches through visual interpretation of temporal Landsat ETM+ images, high spatial resolution images of Bhuvan 2D/3D and Google Earth, along with limited

NU

field checks. The presented inventory will provide important input for other glaciological and hydrological studies (e.g. modeling). We identified 285 glaciers with a total area of 164.5 ± 7.5

MA

km2 in the Ravi basin for 2002. Geological Survey of India (2008, Raina and Srivastava, 2008), ICIMOD (2004, Bhagat et al., 2004), ICIMOD (2011, Bajracharya and Shresta, 2011) and Frey

TE

D

et al. (2012) under the GlobGlacier project (glacier outlines of this project contributed into RGI v4) have carried out significant work on a glacier inventory for the Himachal Himalaya,

AC CE P

including the Ravi basin (Table 4). There is noticeable differences in the total number of glaciers and their area in these inventories (Table 4). To investigate the reason for the differences in glacier area and numbers in different inventories , vector shape-file derived from: i) GSI (2008) report; ii) glacier outlines of ICIMOD (2004 and 2011); and iii) GlobGlacier/RGI v4 were overlaid with the outlines generated in present study using 2002 Landsat ETM+ PAN image (Fig. 10c). A significant difference (~39.7 % or 188 glaciers) in the total number of glacier is observed on comparison of present inventory that with the GlobGlacier/RGI v4 database. This difference can be attributed to inadvertent inclusion of seasonal snow/ice patches and/or snow/ice aprons at the base of steep slopes or avalanche slope and seasonal snow on paleo-cirques in the GlobGlacier/RGI v4 database, arising out of automated mapping (Fig. 10 a, b, Table 4).The GSI glacier inventory included a lower number of glaciers as compare to

20

ACCEPTED MANUSCRIPT our glacier inventory probably due to scale limitations of topographic maps (1:50,000), and misinterpretation of glacier outlines from aerial photographs due to seasonal snow and debris

PT

cover (Raina and Srivastava, 2008; Bhambri and Bolch, 2009). In comparison with ICIMOD

RI

glacier inventory (2004 and 2011), the variation can be attributed to: i) differences in interpretation of debris-covered glaciers; ii) temporal differences in terms of acquired images and

SC

mapping period; iii) differences in classification/definitions of glacier areas/boundary; and iv)

NU

contiguous ice masses may be counted as a single entity or could be subdivided into multiple

MA

glaciers depending on the purpose of the inventory (Bhambri et al., 2013).

TE

D

Insert Table 4 & Figure 10

The areal changes of the glaciers in the Ravi basin confirm an expected and published trend of

AC CE P

glacier retreat as reported by Shukla and Dutta, (2005). However, the rate of retreat is less than previously estimated from SoI Toposheet analysis. For example, our study suggests that Manimahesh and Tal glacier shrank by 0.2 ±0.01 km2 (0.005 ±0.0002 km2 a-1) and 0.01 ± 0.003 km2 (0.0002 ±0.0001 km2 a-1), respectively, between 1971 and 2013, whereas Shukla and Dutta (2005) reported that the Manimahesh and Tal glaciers reduced by 0.7 km2 (~ 0.02 km2 a-1) and 0.7 km2 (~ 0.02 km2 a-1) respectively between 1963/1968 and 2005 (Chand and Sharma, 2015). As per the Shukla and Dutta (2005), Tal Glacier had not only thinned but the lower part got detached from the main glacier after 1963, based on the SoI maps. However, Corona (1971) image shows that there had been minor changes for the main trunk of Tal Glacier and the glacier retained two identifiable bodies; one active and the other remained as detached dead ice (Fig 11). This fact is further reinforced through field survey (2013) as well as multi-temporal satellite

21

ACCEPTED MANUSCRIPT images of Corona (1971), Landsat ETM (2002), Landsat 8 OLI (2013) and GE (2013) images (Fig 11) (Chand and Sharma, 2015).The average recession rate for the detached lower dead ice

PT

part of the former glacier was estimated at ~ 304.3 ±33.6 m (7.2 ±0.8 m a-1) with a total area loss

RI

of 0.09 ± 0.007 km2 (0.002 ± 0.0002 km2 a-1) between 1971 and 2013 (Fig. 11b). The main reason for the discrepancy is probably the interpretation of the glacier terminus on topographic

SC

maps and in coarser resolution datasets which is known to be a significant challenge in glacier

NU

terminus mapping and inventory (Bhambri and Bolch, 2009; Bolch et al., 2010). The use of highresolution satellite image conceivably provides more consistent results than topographic maps

MA

and coarse resolution satellite datasets, and thus provides a valuable resource for monitoring the historic extent of the glacier. However, discrepancies in the rectification of Corona images are

TE

D

principal source of uncertainty in our estimations of glacier change. The main challenges with Corona data are the complex image geometry, unknown parameters of the satellite camera and

AC CE P

its interior orientation specifications. The adopted spline method in the present study for Corona image rectification is laborious and requires interpreter expertise for the collection of GCPs. Moreover, the data are valuable for providing insight into glacier changes since the 1960s not only for the Himachal Himalaya but other remote areas of the Himalaya as well.

Insert Figure 11 Glacier shrinkage in the upper Ravi sub-basin (3.5 ± 2.8 km2) was found marginally higher than adjoining Budhil sub-basin (2.2 ± 2.2 km2) of Ravi. The difference in recession patterns of glacier in the sub-basins may be explained by the quantity and distribution of clean ice and debris cover, sub-basins topography (e.g. glacier altitudinal range, slope, aspect, catchment relief), glaciers geometry and nature of accumulation on glaciers (e.g. avalanches, direct snowfall

22

ACCEPTED MANUSCRIPT and hanging/tributary glaciers) which vary considerably from glacier to glacier in the sub-basins of Ravi basin. The slower rate of area loss in the Budhil sub-basin (0.05 ± 0.05 km2 a-1)

PT

compared with the upper Ravi sub-basin (0.08 ± 0.07 km2 a-1) from 1971 to 2010/13 can be

RI

explained by: i) the higher percentage of debris covered glaciers and change in it during the past four decades in Budhil sub-basin as compare to upper Ravi sub-basin which observed

SC

comparatively lower retreat rate and further suggested the impact of increase debris cover on the

NU

glacier dynamics.; ii) the higher altitudinal range (542 m) of the glaciers in the Budhil sub-basins than in the upper Ravi sub-basin (492 m); and iii) the upper Ravi basin comprises clean ice-

MA

covered glaciers of southern side (i.e. Dhaula-Dhar range) which are comparatively small in size and respond faster than larger debris covered glaciers to perturbations in climate (Bahr et al.,

TE

D

1998).

Several studies have reported that debris cover has increased over time and debris-covered

AC CE P

glaciers shown lower recession rate as compared to clean ice-covered glaciers in the Himalaya (Bhambri et al., 2011; Bolch et al., 2008; Iwata et al., 1980; Kamp et al., 2011). The study find that debris-covered glacier area increased significantly by 19.2 ± 2.2 % (0.5 ± 0.1 % a–1) between 1971 and 2010/13 in the basin. This supports the assumption that, concomitantly with the reccession of glacier with varying topographical charcteristc e.g ratio of PDS area to the total glacier area, the higher catachment relief with steep slope and its ratio to glacier area, the debriscovered area increased in the basin. Several studies highlighted the importance of the role of supraglacial debris cover in the glacier dynmaics in response to climate change by altering surface ablation rates and spatial patterns of mass loss (Benn and Lehmkuhl, 2000; Kirkbride, 2000; Nakawo et al., 1999). In addition, experimental and short-period (ablation season) studies suggest that a thick debris cover reduces ablation,where as a thin debris layer increases ice melt

23

ACCEPTED MANUSCRIPT underneath (Mattson et al., 1993; Pratap et al., 2015b; Sakai et al., 2000; Singh et al., 2000). Moreover, the higher percentage of debris-covered ice in the Ravi basin suggests that the

PT

supraglacial debris cover may significantly define the nature of glacier dynamics in terms of

RI

terminus fluctuations, while further modifying the mass balance of many glaciers in the basin (Benn et al., 2012; Dobhal et al., 2013; Scherler et al., 2011). Although there are not specifc

SC

reasons for the stable terminus and comparatively lower recession rate of debris-covered glaciers

NU

in the basin and thus, it needs further investigations. This would involve debris supply and its transport on glacier surface, as well as the measuement of debris thickness and cover and their

MA

effects on glacial dynmaics.

The present study preferred ASTER GDEM V2 over other available open source DEMs for

TE

D

Himalaya region e.g. SRTM 3v4 and recently available national level CartoDEM (for whole India) and CartoDEM V1.1 R1 (for selected region of India including the western Himalaya) due

AC CE P

to following reasons: i) 27 % of glacier area in the basin comes under the interpolated regions of SRTM3v4 data voids where the accuracy are systematically low and more uncertainty is reported for interpolated regions (Frey et al., 2012), ii) the gross errors would reduce the quality of the topographic parameters and further wrongly demarcate the location of catchment divides which affect the overall glacier extent (Frey et al., 2012) and, iii) the initial analysis shows better accuracy for ASTER GDEM V2 as compared to CartoDEM V1/V1.1 R1 for the present study area and its adjoining basin area i.e. Beas as reported by Singh et al., (2015). Moreover, there can be future applications of CartoDEM V1, especially of CartoDEM V1.1 R1 for the glacier studies in the Himalayan region but being in initial phase of development the study preferred ASTER GDEM V2 over CartoDEM (V1 and V1.1 R1).

24

ACCEPTED MANUSCRIPT 5.2 Comparison with other basins of Himalaya The present study indicates that the recession rate of Ravi basin glaciers 4.6 ± 4.1 % (0.1 ± 0.1%

PT

a-1) from 1971 to 2010/2013 is less than published reported rates for Himachal Himalaya and

RI

other Himalayan regions. In the Himachal Himalaya, the recession rate of glaciers is reported comparatively higher than in the Ravi basin, e.g. Kulkarni et al. (2007) reported a glacier retreat

SC

rate of 21 % (~0.5 % a–1) in Chenab basin, 22 % (~0.6 % a–1) in Parbati, a sub-basin of Beas and

NU

19 % (~0.5 % a–1) in Baspa, a sub-basin of Sutlej. In addition, there were noticeable variations within sub-basins of Chenab as glaciers in the Bhaga, sub-basin of Chenab, retreated at 30 %

MA

(~0.8 a-1) whereas in the adjoining sub-basin of Chandra and Miyar (for 111 glaciers), glaciers retreats were 20 % (0.5 a-1) and 8 % (0.2 a-1),respectively, from 1962 to 2001/2004 (Kulkarni,

D

2012) based on SOI topography map and LISS III images (spatial resolution 23.5 m). Mir et al.

TE

(2013) reported a noticeable deglaciation of 26.1% (~0.6 a-1) for the mapped 34 glaciers in

AC CE P

Tirungkhad sub-basin of Sutlej during 1966-2011 using the SoI toposheets and Landsat satellite images. However, Pandey and Venkataraman (2013) reported a comparatively lower retreat rate of 2.5% (~0.1 a-1) for the 15 glaciers in Chandra and Bhaga sub-basins of Chenab (1980-2010) using the remote sensing datasets of Landsat MSS/TM and Indian remote sensing satellite (IRS) LISS-III/AWFIS. Similarly, we found comparatively lower recession rate (i.e. 4.6 ± 4.1 %) using high resolution datasets. Our study highlight that apparent higher rate of glacier retreat in the Himachal Himalaya could be the result of overestimation of glacier cover in the old datasets of the SoI toposheets which is in line of other studies in Himalaya based on comparison of topography map and Corona images (Bhambri et al., 2011; Bolch et al., 2012, 2008; Chand and Sharma, 2015). In comparison to other parts of Himalaya, the Ravi basin also reported comparatively lower retreat rate. For instance, on the basis of SoI topographic maps and IRS

25

ACCEPTED MANUSCRIPT images, Kulkarni (2012) estimated glaciers of Bhut basin retreat from 1962 to 2001 in the Jammu and Kashmir (J&K) Himalaya at 10 % (~0.3 % a–1), retreat in Warwan basin at 21 % –1

) and retreat in Zanskar at 9 % (~0.3 % a

–1

). However, Bhambri et al. (2011)

PT

(~0.6 % a

RI

estimated comparatively less glacier loss at 4.6 ± 2.8% (~0.1 ±0.1 % a-1) based on high resolution Corona and Cartosat-1 imageries from 1968 to 2006 in the Bhagirathi and

SC

Saraswati/Alaknanda basin of Garhwal Himalaya which is similar to our study (0.1 ± 0.1% a-1)

NU

based on high resolution Corona, Landsat ETM+ PAN and WorldView-2 images. In Nepal Himalaya, the study carried out by Bajracharya and Mool (2006) in the Tamor river basin,

MA

eastern Nepal reported that 5.88% (~0.2% a-1) of the glacier area was deglaciated during 19702000. However, Bolch et al. (2008) had reported a loss of 5.2% (~0.12% a-1) in glacier area

TE

D

from 1962 to 2005 based on Corona and ASTER images in the Khumbu Himalaya which is close-similar to our present study. For the Sikkim Himalaya, Basnett et al. (2013) reported the

AC CE P

glaciers loss as an area of 3.27 ± 0.77% (~0.2 ± 0.1% a-1) between 1989/90 and 2010. However, in the nearby Bhutan Himalaya, Bajracharya et al. (2014) reported a 23.3 ± 0.9% (~0.8 ±0.03 % a-1) glacial area loss between 1980 and 2010. In contrast, there was heterogeneity found in glacier response in Karakoram region. For example, based on space imagery, Bhambri et al. (2013) reported an overall slight area gain of 0.1 ± 3.5 % for the glaciers in the upper Shyok valley from 1973 to 2011. This suggests that the Karakoram area is showing a different response to climate change than other parts of the Himalaya. In addition, for Hindu Raj Himalaya, Sarıkaya et al. (2013) reported that most of the glaciers (70.6%) retreated during the last four decades 1972-2007, although some glaciers advanced (17.6%) or exhibited no detectable change in terminus position (11.8%). High reductions in glacier area of the Himachal Himalaya, Nepal, Bhutan Himalaya and advancing and surging effects of the Karakoram glaciers do not appear in

26

ACCEPTED MANUSCRIPT the Ravi basin of Himachal Himalaya. Such complexity and variations invites a rigorous and systematic analysis of the Himalayan glaciers instead of direct comparisons and extrapolations of

PT

results from well-studied glaciers in some of the basin to others poorly-observed glacierised basin. Glacier recession rate (0.1 a-1) of Ravi basin, Himachal Himalaya is similar to the Khumbu

RI

Himalayan glaciers (0.1 % a–1; Bolch et al., 2008) and Garhwal Himalaya (0.1 % a–1,Bhambri et

NU

SC

al., 2011) and these studies used high resolution historical datasets of Corona and Hexagon.

5.3 Climate Considerations

MA

The Ravi basin a part of Himachal Himalaya, regionally comes under north-western Himalaya, is alternatively influenced by the southwest summer monsoon and winter mid-latitude westerlies: a

TE

D

climatic regime that distinguishes this region from the eastern and Karakoram Himalaya. Pareta and Pareta (2014) reported that the average temperature (Tavg) of the Chamba town (~968 m

AC CE P

a.s.l), which lies within the Ravi basin, has increased by 0.5 °C from 19.3 °C for Period-I (19902001) to 19.8 °C for Period-II (2002-2013). The retreating behavior of glaciers in the basin is mainly attributed to climate change through their dependence on temperature and precipitation regimes as same analyses using NCEP/NCAR (1950-2014) reanalysis data in the absence of nonavailability of long term climate data within the basin. The trend of temperature and precipitation of NCEP/NCAR (1950-2014) reanalysis data was analysed for the grid (32.5ᵒ N, 77.5ᵒ E) located within the basin using Mann-Kendall method (Chand et al., 2015). The study shows that the winter (DJF) average temperature for the selected grid within the basin has slightly increased during the past half-century (1950-2014). The rate of increased trend is significance and thus it is important that slight change in temperature has an impact on the glacier fluctuation. Besides, the continuous increase in average temperature for the winter month is also more significant in terms

27

ACCEPTED MANUSCRIPT of glaciers fluctuation as it may be one of the causative factor to simultaneously occurring of ablation during the winter month in the Ravi basin (Kulkarni et al., 2010). Additionally,

PT

Bhutiyani et al. (2007) also found a significant warming of 1.6°C over the last century with

RI

significant increase of ~3.2°C in winter average maximum temperature (Tmax) during the past two decades for north-western Himalaya. In addition, Shekhar et al. (2010) reported that seasonal

SC

Tavg over the western Himalaya has increased by 2°C with an increase of 2.8°C in annual Tmax

NU

between 1984/85 and 2007/08 in the western Himalaya, while annual minimum temperature (Tmin) increased by about 1°C during a similar period in this region. Dash and others (2007)

MA

observed an increase of 1°C in annual Tmax over the western Himalaya with an annual Tmin decrease of 1.9°C during 1955–72, followed by an increasing trend over more recent decades.

TE

D

Their results show that the seasonal Tmax (0.9°C) and Tmin (0.7°C) have increased during the period 1988–2008 in the Pir-Panjal ranges of the western Himalaya. In contrast, summer cooling

AC CE P

has been reported in some parts of the western Himalaya and upper Indus basin during the latter part of the 20th century (Yadav et al., 2004; Fowler and Archer, 2006). Pareta and Pareta (2014) reported that the precipitation of the Chamba tehsil from previous decade of 1990-2001 to last decade (2002-2013) has reduced from 1037.9 mm to 684 mm (decreased 354 mm) with decrease in number of days of rain throughout the year. Additionally, the winter and pre-monsoon precipitation from NCEP-NCAR data shows decreasing trend at 5% and 1% significance level at the rate of -0.01 and -0.02 mm/year, respectively (Chand et al., 2015). Moreover, in the northwestern Himalaya, a significant decreasing trend has been reported in the monsoon precipitation during the period 1866–2006 (Bhutiyani et al., 2009). Moreover, Dimri et al. (2008) reported reduced snowfall over the western Himalaya in a warming climate. Reduction of snowfall can be explained by the combined effect of climate change and meso-scale influences of the mountains.

28

ACCEPTED MANUSCRIPT Shekhar et al. (2010) also reported decreasing trend in snowfall by 280 cm over the Pir-Panjal range in past few decades. Rana et al. (2008) also reported decreasing trend in snow fall for the

PT

21 sites in Sutlej basin of Himachal Himalaya during 1984-2005.

RI

However, the heterogeneity in trend of climatic parameters on the regional scales along with the influence of topographical parameters as well as glacier morphology, thickness and distribution

SC

of debris-covered area makes it difficult to determine the significant effects of climate

NU

parameters on glacier variability for the Ravi basin because : i) limited availability of climate data especially near the glaciers (e.g. >4000 m a.s.l.), or within the sub-basins; ii) availability of

MA

reanalysis climate data for the study area has to be validated before conclude the complex glacier–climate interactions, and iii) glacier changes in the Ravi basin of Himachal Himalaya

TE

D

show irregularities in amount, rate and time of occurrence during the study period which mainly depend on glacier response (time) to local climate, topography, glacier specific geometry,

AC CE P

characteristics of debris cover on the glacier surface and the glacier catchment morphology. Therefore, the availability of long term meteorological data and field based direct mass-balance measurements within/nearby, or adjoining basin and including geodetic estimates of glacier mass changes will provide a valuable database and further improve knowledge on the response of glaciers to climatic parameters in the Ravi basin of Himachal Himalaya.

6. Conclusion This study provides a comprehensive glacier inventory and multi temporal glacier change for the Ravi basins (Himachal Himalaya) of the north-western Himalaya from 1971 to 2010/13. The major findings and conclusion drawn from the present study are as follows:

29

ACCEPTED MANUSCRIPT  The Ravi basin consists of 285 glaciers (larger than 0.02 km2in size) with a total area of 164.5 ± 7.5 km2 including 71 debris covered glaciers with an area of 36.1 ± 2.1 km2 (22

PT

% of the total glacierised area) in 2002.

RI

 Glacier area decreased from 125.9 ± 1.9 km2 (1971) to 120 ± 4.8 km2 (2010/13), a loss of 4.7 ± 4.1 % or 0.1 ± 0.1 % a-1. In contrast to the basin-wide ice loss, the absolute area loss

SC

in the upper Ravi sub-basin 3.5 ± 2.8 km2 (4.7 ± 3.7 %) was marginally higher than in the

NU

Budhil sub-basin 2.2 ± 2.2 km2 (4.6 ± 4.7 %).This difference can be attributed mainly to the differences in mean glacier size. The glacier retreat rate over the period 1971–

MA

2010/13 is comparatively lesser than other basins in Himachal Himalaya. The glacier recession rates have decreased subsequently in the observation periods, reaching a

D

minimum in the recent decades (2002-2010/13).

TE

 Over the study period (1971-2010/13), glaciers with an area between 2-5 km2 lost 6.2 ±

AC CE P

3.0 % (0.2 ±0.1% a-1) of their ice, followed by glaciers with areas <1 km2 i.e. 5.5 ± 5.7 % (0.1 ± 0.1% a-1). These rates are significantly more than for larger glaciers (>5 km2) which lost 1.5 ± 2.6 % (0.04 ±0.06% a-1) of their extent. This indicates that the average reduction rate is influenced by glacier size.  The debris-covered glacier area increased by 19.2 ± 2.2 % (0.5 ± 0.1 % a–1) in the Ravi basin with significant variation in its sub-basins. For example in Budhil and Upper Ravi sub-basin, the area was increased by 28.6 ± 3.1% (0.7 ± 0.1 % a–1) and 14 ±1.6 % (0.3 ±0.04% a–1), respectively. Possible reasons behind the increased debris cover on glacier surface and its effect on the glacier dynamics need to be investigated further.  This study suggests that the earlier estimation exclusively based on the SoI maps are highly erroneous. Therefore, a reassessment and recalculation of all such areas where

30

ACCEPTED MANUSCRIPT these maps were used is needed to correct records for the glacier recession/area changes. High spatial resolution images of declassified Corona and recently WorldView-2 are

PT

efficient for mapping the change in historical and contemporary glacier terminus and its

RI

morphology, respectively. Additionally, the recent images of Landsat 8 will not only fill gaps in the glacier inventory database but also support further detailed studies on glacier

SC

terminus position and its associated morphological changes, volumetric changes, surging

NU

mechanism and lake outburst modelling of ice/moraine dammed lakes. Acknowledgements: We are thankful to the University Grant Commission, New Delhi for

MA

financial support for this work. The authors are also grateful to Jawaharlal Nehru University, New Delhi for providing the research facilities. We also thank USGS for providing Landsat

TE

D

TM/ETM+/OLI and Corona data. The first author is grateful to Mr. Bruce Raup, GLIMS (http://www.glims.org/)) and Mr. Ian Gilbert, DigitalGlobe Inc./DigitalGlobe 8 Band Challenge

AC CE P

Committee, for providing ASTER and WorldView-2 satellite data, respectively, for this research at no cost. The authors acknowledge Prof. Julian Orford and Prof. Ashok Kumar Shingvi for their valuable suggestions and proof-read of this manuscript.

31

ACCEPTED MANUSCRIPT References

PT

Arendt, A., Others, and 77, 2014. Randolph Glacier Inventory [v4]: a dataset of global glacier outlines. doi:http://www.glims.org/ RGI/randolph.html

SC

RI

Azam, M.F., Wagnon, P., Ramanathan, A., Vincent, C., Sharma, P., Arnaud, Y., Linda, A., Pottakkal, J.G., Chevallier, P., Singh, V.B., Berthier, E., 2012. From balance to imbalance: a shift in the dynamic behaviour of Chhota Shigri glacier, western Himalaya, India. J. Glaciol. 58, 315–324. doi:10.3189/2012JoG11J123

NU

Bahr, D.B., Pfeffer, W.T., Sassolas, C., Meier, M.F., 1998. Response time of glaciers as a function of size and mass balance: 1. Theory. J. Geophys. Res. 103, 9777. doi:10.1029/98JB00507

MA

Bahuguna, I.M., Kulkarni, A. V., Nayak, S., 2004. DEM from IRS-1C PAN stereo coverages over Himalayan glaciated region—accuracy and its utility. Int. J. Remote Sens. 25, 4029– 4041. doi:10.1080/01431160310001652376

TE

D

Bahuguna, I.M., Rathore, B.P., Brahmbhatt, R., Sharma, M., Dhar, S., Randhawa, S.S., Kumar, K., Romshoo, S., Shah, R.D., Ganjoo, R.K., Ajai, 2014. Are the Himalayan glaciers retreating? Curr. Sci. 106, 1008–1013.

AC CE P

Bajracharya, S.., Mool, P.., 2006. Impact of global climate change from 1970s to 2000s on the glaciers and glacial lakes in Tamor Basin, eastern Nepal. Bajracharya, S.R., Maharjan, S.B., Shrestha, F., 2014. The status and decadal change of glaciers in Bhutan from the 1980s to 2010 based on satellite data. Ann. Glaciol. 55, 159–166. doi:10.3189/2014AoG66A125 Bajracharya, S.R., Shrestha, B., 2011. The Status of Glaciers in the Hindu Kush-Himalayan Region. Bajracharya, S. R., & Shresta, B., 2011. The Status of Glaciers in the Hindu Kush-Himalayan Region, ICIMOD. Barry, R.G., 2006. The status of research on glaciers and global glacier recession: a review. Prog. Phys. Geogr. 30, 285–306. doi:10.1191/0309133306pp478ra Basnett, S., Kulkarni, A. V., Bolch, T., 2013. The influence of debris cover and glacial lakes on the recession of glaciers in Sikkim Himalaya, India. J. Glaciol. 59, 1035–1046. doi:10.3189/2013JoG12J184 Beck, A., Philip, G., Abdulkarim, M., Donoghue, D., 2007. Evaluation of Corona and Ikonos high resolution satellite imagery for archaeological prospection in western Syria. Antiquity. 81, 161–175.

32

ACCEPTED MANUSCRIPT Benn, D.I., Lehmkuhl, F., 2000. Mass balance and equilibrium-line altitudes of glaciers in highmountain environments. Quat. Int. 65-66, 15–29. doi:10.1016/S1040-6182(99)00034-8

PT

Bhagat, R., V, K., C, S., PK, M., SR, B., 2004. Inventory of Glaciers, Glacial Lakes and Identification of Potential Glacial Lake Outburst Floods (GLOFs) Affected by Global Warming in the Mountains of Himalayan Region.

RI

Bhambri, R., Bolch, T., 2009. Glacier mapping: a review with special reference to the Indian Himalayas. Prog. Phys. Geogr. 33, 672–704. doi:10.1177/0309133309348112

NU

SC

Bhambri, R., Bolch, T., Chaujar, R.K., Kulshreshtha, S.C., 2011. Glacier changes in the Garhwal Himalaya, India, from 1968 to 2006 based on remote sensing. J. Glaciol. 57, 543–556. doi:10.3189/002214311796905604

MA

Bhambri, R., Bolch, T., Kawishwar, P., Dobhal, D.P., Srivastava, D., Pratap, B., 2013. Heterogeneity in glacier response in the upper Shyok valley, northeast Karakoram. Cryosph. 7, 1385–1398. doi:10.5194/tc-7-1385-2013

TE

D

Bhutiyani, M.R., Kale, V.S., Pawar, N.J., 2007. Long-term trends in maximum, minimum and mean annual air temperatures across the Northwestern Himalaya during the twentieth century. Clim. Change 85, 159–177. doi:10.1007/s10584-006-9196-1

AC CE P

Bhutiyani, M.R., Kale, V.S., Pawar, N.J., 2009. Climate change and the precipitation variations in the northwestern Himalaya: 1866-2006. Int. J. Climatol. 30, 535–548. doi:10.1002/joc.1920 Bolch, T., Buchroithner, M., Pieczonka, T., Kunert, A., 2008. Planimetric and volumetric glacier changes in the Khumbu Himal, Nepal, since 1962 using Corona, Landsat TM and ASTER data. J. Glaciol. 54, 592–600. doi:10.3189/002214308786570782 Bolch, T., Kulkarni, A., Kääb, A., Huggel, C., Paul, F., Cogley, J.G., Frey, H., Kargel, J.S., Fujita, K., Scheel, M., Bajracharya, S., Stoffel, M., 2012. The state and fate of Himalayan glaciers. Science 336, 310–4. doi:10.1126/science.1215828 Bolch, T., Menounos, B., Wheate, R., 2010a. Landsat-based inventory of glaciers in western Canada, 1985–2005. Remote Sens. Environ. 114, 127–137. doi:10.1016/j.rse.2009.08.015 Bolch, T., Menounos, B., Wheate, R., 2010b. Remote Sensing of Environment Landsat-based inventory of glaciers in western Canada , 1985 – 2005. Remote Sens. Environ. 114, 127– 137. doi:10.1016/j.rse.2009.08.015 Bolch, T., Yao, T., Kang, S., Buchroithner, M.F., Scherer, D., Maussion, F., Huintjes, E., Schneider, C., 2010. A glacier inventory for the western Nyainqentanglha Range and the Nam Co Basin, Tibet, and glacier changes 1976–2009. Cryosph. 4, 419–433. doi:10.5194/tc-4-419-2010

33

ACCEPTED MANUSCRIPT Bookhagen, B., Fleitmann, D., Nishiizumi, K., Strecker, M.R., Thiede, R.C., 2006. Holocene monsoonal dynamics and fluvial terrace formation in the northwest Himalaya, India. Geology 34, 601. doi:10.1130/G22698.1

PT

Braithwaite, R.J., Raper, S.C.B., 2009. Estimating equilibrium-line altitude ( ELA ) from glacier inventory data. Ann. Glaciol. 50, 127–132.

SC

RI

Chand, P., Sharma, M.C., 2015. Frontal changes in the Manimahesh and Tal Glaciers in the Ravi basin, Himachal Pradesh, northwestern Himalaya (India), between 1971 and 2013. Int. J. Remote Sens. 36, 4095–4113.

MA

NU

Chand, P., Sharma, M.C., Prasad, R.N., 2015. Heterogeneity in Fluctuations of Glacier with Icecover, Debris-cover and Proglacial-lake in the Upper Ravi Basin, Himachal Himalaya (India) During the Past Four Decades (1971-2013), in: Singh, R., Schickhoff, U., Mal, S. (Eds.), Climate Change and Dynamics of Glaciers and Vegetation in the Himalaya. Springer, Netherlands, p. In Printing. Csatho, B.M., Van Der Veen, C.J., Tremper, C.M., 2005. Trimline mapping from multispectral Landsat ETM+ imagery. Géographie Phys. Quat. 59, 49. doi:10.7202/013736ar

TE

D

Dash, S.K., Jenamani, R.K., Kalsi, S.R., Panda, S.K., 2007. Some evidence of climate change in twentieth-century India. Clim. Change 85, 299–321. doi:10.1007/s10584-007-9305-9

AC CE P

Dimri, A.P., Kumar, A., Satyawali, P.K., Ganju, A., 2008. Climatic variability of weather parameters over the western Himalayas: a case study. Proc. Natl. Snow Sci. Work. 11–12. Dobhal, D.P., Mehta, M., Srivastava, D., 2013. Influence of debris cover on terminus retreat and mass changes of Chorabari Glacier, Garhwal region, central Himalaya, India. J. Glaciol. 59, 961–971. Dyurgerov, M.B., Meier, M.F., 2005. Glaciers and the Changing Earth System: A 2004 Snapshot. Occas. Pap. (University Color. Boulder. Inst. Arct. Alp. Res. 117. Elizabeth Martin, H., Whalley, W.B., 1987. Rock glaciers: part 1: rock glacier morphology: classification and distribution. Prog. Phys. Geogr. Fowler, H.J., Archer, D.R., 2006. Conflicting signals of climatic change in the Upper Indus basin. J. Clim. 19, 4276–4293. Frey, H., Machguth, H., Huss, M., Huggel, C., Bajracharya, S., Bolch, T., Kulkarni, A., Linsbauer, A., Salzmann, N., Stoffel, M., 2014. Estimating the volume of glaciers in the Himalayan–Karakoram region using different methods. Cryosph. 8, 2313–2333. doi:10.5194/tc-8-2313-2014

34

ACCEPTED MANUSCRIPT Frey, H., Paul, F., Strozzi, T., 2012. Compilation of a glacier inventory for the western Himalayas from satellite data: methods, challenges, and results. Remote Sens. Environ. 124, 832–843. doi:10.1016/j.rse.2012.06.020

PT

Furbish, J D , Andrews, J.T., 1984. The use of hypsometry to indicate long-term stability and response of valley glaciers to changes in mass transfer. J. Glaciol. 30, 199–211.

SC

RI

Gardelle, J., Berthier, E., Arnaud, Y., Kääb, a., 2013. Region-wide glacier mass balances over the Pamir-Karakoram-Himalaya during 1999–2011. Cryosph. 7, 1263–1286. doi:10.5194/tc-7-1263-2013

NU

Goossens, R., De Wulf, A., Bourgeois, J., Gheyle, W., Willems, T., 2006. Satellite imagery and archaeology: the example of CORONA in the Altai Mountains. J. Archaeol. Sci. 33, 745– 755. doi:10.1016/j.jas.2005.10.010

MA

Haeberli, W., 1990. Glacier and permafrost signals of 20th-century warming. Ann. Glaciol. 4, 99–101.

TE

D

Hewitt, K., 2011. Glacier Change, Concentration, and Elevation Effects in the Karakoram Himalaya, Upper Indus Basin. Mt. Res. Dev. 31, 188–200. doi:10.1659/MRD-JOURNALD-11-00020.1 HPSEB, 2004. Chamba Hydro-electric project Himachal Pradesh.

AC CE P

Hubbard, B., Glasser, N.F., 2005. Field Techniques in Glaciology and Glacial Geomorphology. Wiley-Blackwell. Iwata, S., Watanabe, O., Fushimi, H., 1980. Surface Morphology in the Ablation Area of the Khumbu Glacier. J. Japanese Soc. Snow Ice 41, 9–17. doi:10.5331/seppyo.41.Special_9 Kääb, a., 2005. Combination of SRTM3 and repeat ASTER data for deriving alpine glacier flow velocities in the Bhutan Himalaya. Remote Sens. Environ. 94, 463–474. doi:10.1016/j.rse.2004.11.003 Kääb, A., Berthier, E., Nuth, C., Gardelle, J., Arnaud, Y., 2012. Contrasting patterns of early twenty-first-century glacier mass change in the Himalayas. Nature 488, 495–8. doi:10.1038/nature11324 Kamp, U., Byrne, M., Bolch, T., 2011. Glacier fluctuations between 1975 and 2008 in the Greater Himalaya Range of Zanskar, southern Ladakh. J. Mt. Sci. 8, 374–389. doi:10.1007/s11629-011-2007-9 Kargel, J.S., Abrams, M.J., Bishop, M.P., Bush, A., Hamilton, G., Jiskoot, H., Ka, A., Kieffer, H.H., Lee, E.M., Paul, F., Rau, F., Raup, B., Shroder, J.F., Soltesz, D., Stainforth, D., Stearns, L., Wessels, R., 2005. Multispectral imaging contributions to global land ice

35

ACCEPTED MANUSCRIPT measurements from space. Remote Sens. Environ. 99, 187–219. doi:10.1016/j.rse.2005.07.004

PT

Kirkbride, M.P., 2000. Ice marginal geomorphology and Holocene expansion of debris-covered Tasman Glacier, New Zealand., in: Nakawo, M., Raymond, C., Fountain, A. (Eds.), DebrisCovered Glaciers. IAHS Publication 264, pp. 211–217.

RI

Kulkarni, A. V, 2012. Monitoring Himalayan cryosphere using remote sensing techniques. J. Indian Inst. Sci.

NU

SC

Kulkarni, A. V, Bahuguna, I.M., Rathore, B.P., Singh, S.K., Randhawa, S.S., Sood, R.K., Dhar, S., others, 2007. Glacial retreat in Himalaya using Indian Remote Sensing satellite data. Curr. Sci. 92, 69–74.

MA

Kulkarni, A. V, Rathore, B.P., 2005. Alarming retreat of Parbati glacier , Beas basin , Himachal Pradesh. Curr. Sci. 88, 1844–1850. Kulkarni, A. V., Rathore, B.P., Singh, S.K., 2010. Distribution of seasonal snow cover in central and western Himalaya. Ann. Glaciol. 51, 123–128. doi:10.3189/172756410791386445

TE

D

Kulkarni, A. V., Rathore, B.P., Singh, S.K., Bahuguna, I.M., 2011. Understanding changes in the Himalayan cryosphere using remote sensing techniques. Int. J. Remote Sens. 32, 601–615. doi:10.1080/01431161.2010.517802

AC CE P

Marh, B.S., 1986. Geomorphology of the Ravi River. Inter-India Publications. Mattson, L.E., Gardner, J.S., Young, G.J., 1993. Ablation on debris covered glaciers: an example from the Rakhiot Glacier,Punjab Himalaya, in: IAHS Publ.218 (Symposium at Kathmandu 1992– Snow and Glacier Hydrology). pp. 289–296. Mayewski, P.A., Jeschke, P.A., 1979. Himalayan and Trans-Himalayan Glacier Fluctuations Since AD 1812. Arct. Alp. Res. 11, 267–287. Mehta, M., Dobhal, D.P., Bisht, M.P.S., 2011. Change of Tipra Glacier in the Garhwal Himalaya, India, between 1962 and 2008. Prog. Phys. Geogr. 35, 721–738. doi:10.1177/0309133311411760 Mehta, M., DPDobhal, Kesarwani, K., Pratap, B., Kumar, A., Verma, A., 2014. Monitoring of glacier changes and response time in Chorabari Glacier, Central Himalaya, Garhwal, India. Curr. Sci. 107, 281–289. Mir, R.A., Jain, S.K., Saraf, A.K., Goswami, A., 2013. Glacier changes using satellite data and effect of climate in Tirungkhad basin located in western Himalaya. Geocarto Int. 29, 293– 313. doi:10.1080/10106049.2012.760655

36

ACCEPTED MANUSCRIPT Nagai, H., Fujita, K., Nuimura, T., Sakai, A., 2013. Southwest-facing slopes control the formation of debris-covered glaciers in the Bhutan Himalaya. Cryosph. 7, 1303–1314. doi:10.5194/tc-7-1303-2013

RI

PT

Nainwal, H.C., Negi, B.D.S., Chaudhary, M., Sajwan, K.S., Gaurav, A., 2008. Temporal changes in rate of recession: Evidences from Satopanth and Bhagirath Kharak glaciers, Uttarakhand, using Total Station Survey. Curr. Sci. 94, 653–660.

SC

Nakawo, M., Yabuki, H., Sakai, A., 1999. Characteristics of Khumbu Glacier, Nepal Himalaya: recent change in the debris-covered area. Ann. Glaciol. 28, 118–122. doi:10.3189/172756499781821788

NU

Owen, L. a, England, J., 1998. Observations on rock glaciers in the Himalayas and Karakoram Mountains of northern Pakistan and India. Geomorphology 26, 199–213. doi:10.1016/S0169-555X(98)00059-2

MA

Pandey, P., Venkataraman, G., 2013. Changes in the glaciers of Chandra–Bhaga basin, Himachal Himalaya, India, between 1980 and 2010 measured using remote sensing. Int. J. Remote Sens. 34, 5584–5597. doi:10.1080/01431161.2013.793464

TE

D

Pareta, K., Pareta, U., 2014. Climate Change Impact on Land and Natural Resource in Chamba Tehsil of Himachal Pradesh State , India. Int. J. Sci. Technoledge 2, 38–48.

AC CE P

Paul, F., Barrand, N.E., Baumann, S., Berthier, E., Bolch, T., Casey, K., Frey, H., Joshi, S.P., Konovalov, V., Bris, R. Le, Mölg, N., Nosenko, G., Nuth, C., Pope, A., Racoviteanu, A., Rastner, P., Raup, B., Scharrer, K., Steffen, S., Winsvold, S., 2013. On the accuracy of glacier outlines derived from remote-sensing data. Ann. Glaciol. 54, 171–182. doi:10.3189/2013AoG63A296 Paul, F., Barry, R.G., Cogley, J.G., Frey, H., Haeberli, W., Ohmura, A., Ommanney, C.S.L., Raup, B., Rivera, A., Zemp, M., 2009. Recommendations for the compilation of glacier inventory data from digital sources. Ann. Glaciol. 50, 119–126. Paul, F., Kääb, A., Maisch, M., Kellenberger, T., Haeberli, W., 2002. The new remote-sensingderived Swiss glacier inventory: I. Methods. Ann. Glaciol. 34, 355–361. doi:10.3189/172756402781817941 Paul, F., Svoboda, F., 2009. A new glacier inventory on southern Baffin Island , Canada , from ASTER dataᵒ: II . Data analysis , glacier change and applications. Ann. Glaciol. 50, 22–31. Pratap, B., Dobhal, D.P., Bhambri, R., Mehta, M., Tewari, V.C., 2015a. Four decades of glacier mass balance observations in the Indian Himalaya. Reg. Environ. Chang. doi:10.1007/s10113-015-0791-4

37

ACCEPTED MANUSCRIPT Pratap, B., Dobhal, D.P., Mehta, M., Bhambri, R., 2015b. Influence of debris cover and altitude on glacier surface melting: a case study on Dokriani Glacier, central Himalaya, India. Ann. Glaciol. 56, 9–16.

RI

PT

Racoviteanu, A.E., Arnaud, Y., Williams, M.W., Ordoñez, J., 2008. Decadal changes in glacier parameters in the Cordillera Blanca, Peru, derived from remote sensing. J. Glaciol. 54, 499– 510. doi:10.3189/002214308785836922

SC

Racoviteanu, A.E., Paul, F., Raup, B., Jodha, S., Khalsa, S., Armstrong, R., 2009. Challenges and recommendations in mapping of glacier parameters from spaceᵒ: results of the 2008 Global Land Ice Measurements from Space ( GLIMS ) workshop , Boulder , Colorado , USA. Ann. Glaciol. 50, 53–69.

MA

NU

Racoviteanu, A.E., Williams, M.W., Barry, R.G., 2008. Optical Remote Sensing of Glacier Characteristics: A Review with Focus on the Himalaya. Sensors 8, 3355–3383. doi:10.3390/s8053355 Raina, V., Srivastava, D., 2008. Glacier Atlas of India. Bangalore.

TE

D

Raina, V.K., 2009. Himalayan glaciers: a state-of-art review of glacial studies, glacial retreat and climate change. Delhi.

AC CE P

Rana, R.S., Bhagat, R.M., Kalia, V., Lal, H., 2008. Impact of climate change on shift of apple belt in himachal Pradesh, in: ISPRS Archives XXXVIII-8/W3 Workshop Proceedings: Impact of Climate Change on Agriculture. pp. 131–137. Rangecroft, S., Harrison, S., Anderson, K., Magrath, J., Castel, A.P., Pacheco, P., 2014. A First Rock Glacier Inventory for the Bolivian Andes. Permafr. Periglac. Process. 25, 333–343. doi:10.1002/ppp.1816 Regmi, D., Watanabe, T., 2009. Rockfall activity in the Kangchenjunga area, Nepal Himalaya. Permafr. Periglac. Process. 20, 390–398. doi:10.1002/ppp.664 Sakai, A., Takeuchi, N., Fujita, K., Nakawo, M., 2000. Role of supraglacial ponds in the ablation process of a debris-covered glacier in the Nepal Himalayas. Sarıkaya, M.A., Bishop, M.P., Shroder, J.F., Ali, G., 2013. Remote-sensing assessment of glacier fluctuations in the Hindu Raj, Pakistan. Int. J. Remote Sens. 34, 3968–3985. doi:10.1080/01431161.2013.770580 Scherler, D., Bookhagen, B., Strecker, M.R., 2011. Spatially variable response of Himalayan glaciers to climate change affected by debris cover. Nat. Geosci. 4, 156–159. doi:10.1038/ngeo1068

38

ACCEPTED MANUSCRIPT Scotti, R., Brardinoni, F., Alberti, S., Frattini, P., Crosta, G.B., 2013. A regional inventory of rock glaciers and protalus ramparts in the central Italian Alps. Geomorphology 186, 136– 149. doi:10.1016/j.geomorph.2012.12.028

PT

Shekhar, M.S., Chand, H., Kumar, S., Srinivasan, K., Ganju, a., 2010. Climate-change studies in the western Himalaya. Ann. Glaciol. 51, 105–112. doi:10.3189/172756410791386508

SC

RI

Shukla, S., Dutta, S., 2005. Generation Of Baseline Data On Secular Movement Of Selected Glaciers In Ravi Basin , Chamba District , Himachal Pradesh ( Tal and Manimahesh Glaciers ).

NU

Singh, M.K., Gupta, R.D., Bhardwaj, A., Ganju, A., 2015. Scenario-Based Validation of Moderate Resolution DEMs Freely Available for Complex Himalayan Terrain. Pure Appl. Geophys. doi:10.1007/s00024-015-1119-5

MA

Singh, P., Kumar, N., Ramasastri, K.S., Singh, Y., 2000. Influence 0f a fine debris layer on the melting f snow and ice on a Himalayan glacier, in: IAHS Publ.264 (Symposium at Seattle 2000–Debris Covered Glaciers). pp. 63–70.

TE

D

Thayyen, R.J., Gergan, J.T., 2010. Role of glaciers in watershed hydrology: a preliminary study of a “Himalayan catchment.” Cryosph. 4, 115–128. doi:10.5194/tc-4-115-2010

AC CE P

Yadav, R.R., Park, W., Singh, J., Dubey, B., 2004. Do the western Himalayas defy global warmingᵒ? 31, 1–5. doi:10.1029/2004GL020201

39

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

Fig. 1

40

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

Fig. 2

41

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

Fig. 3

42

TE AC CE P

Fig. 4

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

43

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

Fig. 5

44

AC CE P

Fig. 6

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

45

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

Fig. 7

46

Fig. 8

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

47

AC CE P

Fig. 9

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

48

Fig. 10

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

49

AC CE P

Fig. 11

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

50

ACCEPTED MANUSCRIPT Fig. 1. (a) Location of study area in the Himalaya with major river systems of the Indian subcontinent. (b) Glacier coverage in the Ravi basin, based on Landsat ETM+ (2 August 2002)

PT

imagery; stars represent glaciers mapped from Landsat TM image (08 October 1989), (c)

RI

Climate data for Chamba (~924 m a.s.l.)

SC

Fig. 2. Satellite images of Kugti Glacier, Budhil basin (see Fig. 1b for location). (a) Rectified

NU

subset of Corona image (28 September 1971) based on the spline method with similar-year glacier outline. (b) Landsat ETM+ PAN image (02 August 2002) with Landsat ETM + and

MA

Corona glacier outlines.

TE

D

Fig. 3. Rock glacier in the Manimahesh valley of the Ravi basin (see Fig. 1b for location), (a)

AC CE P

Field Photograph (2013) (b) Corona (1971) satellite image.

Fig. 4. Distribution of the number of glaciers, glacier area, mean glacier elevation and mean slope as per glacier size class.

Fig. 5. Characteristics of glaciers in 2002, (a) Hypsometry of clean-ice, debris-covered, total glaciers and glaciers with different size classes (b) latitude/longitude distribution of glaciers v/s mean elevation. Glacier inventory and its characteristics derived from Landsat ETM+ PAN (2 August 2002) image and ASTER GDEM V2. Triangle, circle and cross represent glaciers of Siul, Budhil and Upper Ravi sub-basin of Ravi.

51

ACCEPTED MANUSCRIPT Fig. 6. Characteristics of glaciers in 2002 (a) glacier area v/s aspect, (b) glacier area v/s elevation range, (c) glacier area v/s slope, (d) glacier v/s median elevation. Glacier inventory and its

PT

characteristics derived from Landsat ETM+ PAN (2 August 2002) image and ASTER GDEM

RI

V2. Triangle, circle and cross represent glaciers of Siul, Budhil and Upper Ravi sub-basin of

SC

Ravi.

NU

Fig. 7. Glacier change for 157 glaciers in the Ravi basin, Himachal Himalaya from 1971 to 2010/13 (red and blue glacier outlines derived from Corona (1971) and Landsat 7/8

MA

(2013)/WorldView-2 (2010) images respectively).

TE

D

Fig. 8. Glacier area change as per size class, (a) glacier area change for 157 glaciers during 19712010/13, (b) Glacier area change for 54 glaciers during 1971-1989-2002-2010/13, and (c) glacier

AC CE P

area change for 209 glaciers during 2002-2010/13. We added <1 km2 size class and total glaciers in graph for comparisons.

Fig. 9. Scatter Plots of (a) glacier area change v/s slope, (b) glacier area change v/s aspect, (c) glacier area change v/s median elevation, (d) glacier area change v/s elevation zone for 157 glaciers during 1971-2010/13. Triangle, circle and cross represent glaciers of Siul, Budhil and Upper Ravi sub-basin of Ravi.

Fig. 10: Glacier inventory (a) misinterpretation of seasonally snow-covered areas, (b) misinterpretation of glacier boundary, background image Landsat ETM+ 15 October 2000 (c) overlay of glacier outlines of different inventory e.g. GSI (2008), ICIMOD (2004, 2011),

52

ACCEPTED MANUSCRIPT GlobGlacier (2002) (Frey and others 2012) on present study glacier outlines (2002), background image Landsat ETM+ 2 August 2002, (d) misclassified Ice divides (dark black line represent the

PT

erroneous glacier boundary) (background image, Landsat ETM+ 15 October 2000). White, white

RI

dotted and black arrows represent misinterpretation of seasonal snow covered patches, erroneous

SC

glacier boundary and ice/basin divides.

NU

Fig. 11. Recession of Tal glacier, Budhil sub-basin of Ravi. (a) Corona image (28 September 1971) with same-year glacier outline, (b) Google Earth (GE) image (26 September 2013) with

MA

glacier outline of 1971 and 2013 (Note: GE image used to show the location of snout and comparison while it was also mapped in field 2013) (c) field photographs towards upward

TE

D

direction of glacier (2013), (d) field photographs towards downward direction of glacier (2013)

AC CE P

shows the main body of Tal glacier.

53

ACCEPTED MANUSCRIPT Table 1. Database for the present study Spatial resolution (m) 3 3

70 MM X DS1115-2282DA066//DA068 70 MM X DS1115-2282DF063/DF062/DF057/DF058/DF061

Landsat 5 TM

9 October 1989

30, 60

ETP147R37_5T19891009/ETP147R38_5T19891009

< 20 m

3 October 2010

30, 60

LT51470382010276KHC00

< 20 m

2 August 2002

15, 30, 60

LE71470372002214SGS00

< 15m

28 October 2002

15, 30, 60

LE71480372002301SGS00

<15 m

28 June 2013

15, 30 , 100

LC81480372013179LGN01

< 15m

25 September 2013

15, 30 , 100

LC81470382013268LGN00

< 15m

27 October 2013

15, 30 , 100

LC81470382013300LGN00

< 15m

ASTER

28 October 2002

15,30,90

AST_L1A_003_10282002054845_11112002172853

< 15 m

WorldView-2

4 November 2010

0.4, 2.5

10NOV04055517-M3DS-052430842030_01_P001

< 5m

RI

SC

NU

MA D

Landsat 8 OLI/TRIS

Accuracy RMS x,y <6m

TE

Landsat 7 ETM+

Scene/Product/Path and Row ID

PT

Corona KH-4B

Date of Acquisition 28 September 1971

AC CE P

Satellite/Sensor

54

ACCEPTED MANUSCRIPT Table 2. Glacier parameters for the Ravi basin (2002) and its sub-basins (2002)

2002

Sub-basins (2002) Budhil

Upper-Ravi

Average elevation minimum (m)

4623

4369

4642

4648

Average elevation maximum (m)

5038

4765

5072

5054

Average elevation mean (m)

4828

4576

4854

4846

Average elevation median (m)

4828

4579

4854

4847

Average elevation range (m)

415

395

430

406

Minimum Elevation (m)

3473

3754

3473

3636

Max Elevation (m)

6024

5465

5732

6024

Mean slope (0)

27.33

27.04

28.50

26.39

Mean Aspect

SW (207.8)

W (252.9)

SW (206.8)

S (201.4)

NU

MA

0.69 (7.6%)

13.65 (23.6%) 21.78 (22.3%) 44.15 (76.4%) 75.77 (77.7%)

D

Debris-covered glacier area (km2) 36.123 (22%)

PT

Suil

RI

Ravi

SC

Parameters

8.41 (92.4%)

64.41

4.21

26.93

33.27

1-2

24.23

4.89

8.58

10.76

2-5

40.15

22.30

17.85

5-11

35.66

Total glacierized area (km2)

164.45

Area by glacier size (km2)

AC CE P

<1

128.33 (78%)

TE

Clean-ice glacier area (km2)

35.66 9.1 (5.5%)

57.8 (35.2%)

97.5 (59.3%)

55

ACCEPTED MANUSCRIPT Table 3. Changes in total ice area, clean-ice area and debris-covered ice area in the Ravi basin, Himachal Himalaya between 1971, 1989, 2002 and 2010/13 based on spaceborne imagery. Ice Extent

125.84 ± 1.87 121.39 ± 5.43

159

90.4 ±4.92

103

201013

119.99 ± 4.82

159

88.21 ±4.27

103

1971 (Coron a) 201013 1971 (Coron a) 201013 1971 (Coron a) 201013

3.29 ±0.06

7

3.12 ±0.17 47.25 ±0.73

7

3.29 ± 0.06 3.12 ±0.17 37.68 ±0.66

1971 (Coron a) 1989

71.6 ±0.92

201013

71.8 ±2.57

91

69.67 ±2.66 68.3 ±2.64 67.4 ±2.21

91

56

7

-

-

7

-

-

36

9.58 ±0.06

32.75 ±1.79 58.24 ±0.99

32

12.32 ±0.29 17.06 ±0.09

52.34 ±2.31

64

66

TE

61

26.65 ±0.15

No of Glacier s 49

30.99 ± 0.51 31.78 ± 0.55

AC CE P

2002

45.07 ±2.08 75.3 ±1.08

Km

56

23

25

19.46 ±0.26

27

54.96 ±0.82

36

16.63 ±0.09

16

54

52.36 ±2.34 50.1 ±2.35 49.38 ±1.93

36

17.32 ±0.32 18.2 ±0.29 18.01 ±0.29

18

54

34 34

Clean Ice

Debris Cover

19712010/1 3 19712002

-5.85 ±5.17

10.98 ±4.6 -8.8 ±5.21

5.13 ±0.57

-4.65 ±4.11

19.24 ±2.15

-3.54 ±4.56

11.07 ±4.64 -8.87 ±5.26

4.34 ±0.53

-2.18 ±6.52

0.79 ±0.75

-1.15 ±5.98

-2.42 ±7.21

2.53 ±2.42

-0.16 ±0.18

-0.16 ±0.18

-

-4.97 ±5.61

-4.97 ±5.61

-

Siul

19712010/1 3

-2.19 ±2.2

-4.93 ±1.91

2.74 ±0.3

-4.63 ±4.67

13.07 ±5.06

28.61 ±3.13

Budh il

19712010/1 3

-3.51 ±2.79

-5.9 ±2.51

2.4 ±0.27

-4.66 ±3.70

10.13 ±4.32

14.04 ±1.61

Uppe r Ravi

19712010/1 3 19711989 19892002 20022010/1 3

-4.2 ±2.4

-5.58 ±2.09

1.38 ±0.3

-5.86 ±3.35

8.31 ±1.8

Ravi

-1.92 ±2.82 -1.35 ±3.75 -0.90 ±3.44

-2.61 ±2.48 -2.26 ±3.32 -0.71 ±3.04

0.68 ±0.33 0.89 ±0.43 -0.19 ±0.41

-2.69 ±3.93 -1.97 ±5.38 -1.32 ±5.04

10.15 ±3.81 -4.74 ±4.52 -4.32 ±6.34 -1.42 ±6.06

20022010/1 3

-4.46 ±5.74 -1.4 ±7.26

19712010/1 3

Debris Cover

Total Area

Basi n/Su bbasin

Ravi

16.29 ±1.98

29

52

54

Clea n ice

PT

1971 (Coron a) 2002

59

99.19 ±1.72

No of Glacie rs 108

Total Area

RI

Km

% Change Rate

Period

SC

No of Glacie rs 157

Debris Cover

NU

Km

Clean Ice

MA

Total Area

D

Year

Change Rate

20 20

4.11 ±2 5.12 ±2.49 -1.03 ±2.23

Table 4: Comparison of glacier inventory in the Ravi basin S. No.

Inventory

Number of Glaciers

Area (Km2)

Minimum Glacier Size (Km2)

Data Used (Year)

References

56

ACCEPTED MANUSCRIPT 193

-

2.

ICIMOD (2004)

198

235.2

-

3.

ICIMOD (2011)

217

113.6

0.02

4.

GlobGlacier, (2012) (Part of RGI 4)

473

213

0.02

5.

Present Study

285

164.5 ± 7.5

SOI Toposheet and Aerial Photography IRS-1D LISS-III (2000-2002) Landsat ETM+ (2005±3)

PT

172

RI

GSI (2008)

NU

SC

1.

Present Study

AC CE P

TE

D

MA

0.02

Landsat ETM + (2002) and ALOS Palsar (2007) Landsat ETM + (2002)

Raina and Srivasatva, 2008 Bhagat et al., 2004 Bajracharya and Shrestha, 2011 Frey et al., 2012

57

ACCEPTED MANUSCRIPT Highlights Analyse glacier change for Ravi basin, Himachal Himalaya (HH) from 1971 to 2010/13.



Glacier loss 4.6 % of their area, comparatively lesser than other basins in HH.



Discuss the glacier inventories issues and provide updated inventory for Ravi basin.



Discussed the potential reason for increase of debris-cover area.



Discuss the issues of mapping historic extent of glacier from the SoI topomap.

AC CE P

TE

D

MA

NU

SC

RI

PT



58