These Adnan Versionfinale

UNIVERSITE MONTPELLIER 2
SCIENCES ET TECHNIQUES DU LANGUEDOC
THESE
pour obtenir le grade de
DOCTEUR DE L'UNIVERSITE MONTPELLIER 2
Discipline : Eaux Continentales et Société
Ecole Doctorale : Systèmes intégrés en Biologie, Agronomie,

Géosciences, Hydrosciences et Environnement (SIBAGHE)
présentée et soutenue publiquement
par
Adnan Ahmad TAHIR
Le 21 septembre 2011
Title:
Impact of climate change on the snow covers and glaciers in the Upper Indus
River Basin and its consequences on the water reservoirs (Tarbela Dam) –
Pakistan
Titre :
Impact du changement climatique sur les couvertures neigeuses et les glaciers
dans le Haut Bassin de l'Indus et ses conséquences sur les ouvrages
hydrauliques (Réservoir de Tarbela) – Pakistan
JURY
M. Eric SERVAT Président

M. Pierre CHEVALLIER Directeur de thèse
M. Yves ARNAUD Co-directeur de thèse
M. Vazken ANDREASSIAN Rapporteur
M. Antonello PROVENZALE Rapporteur
Mme. Isabella ZIN Examinateur
1
ACKNOWLEDGEMENTS
It is with great emotion that I address my thanks to all institutions and people who supported
me during the completion of this research: without their help and encouragement, this
doctoral thesis would never have happened.
I express my thanks to Vazken ANDREASSIAN and Antonello PROVENZALE, who

accepted to assess my Ph.D. dissertation in spite of their professional engagements.
I want to thank especially:
 Higher Education Commission (HEC) of Pakistan to award me a scholarship during

the three years of this thesis;
 La Maison des Sciences de l’Eau (MSE) and Laboratoire Hydrosciences who

welcomed me into their premises, particularly in the library, where Catherine PICQ and
Joelle CANER were always present to help me;
 Water and Power Development Authority (WAPDA) and Pakistan Meteorological

Department (PMD), for their exchange of the valuable hydrological and climate data to
complete my research;
 Pierre CHEVALLIER, my advisor, whose encouragement, guidance and support from

the initial to the final level enabled me to develop an understanding of the subject. I am
highly impressed by his broad vision and kindness. Despite his quite busy schedule, he
never refused to have discussions;
 Thesis committee members, Yves ARNAUD (co-director of my thesis), Etienne

BERTHIER, Benoit HINGRAY, Luc NEPPEL and Patrick WAGNON, who supported me
with their valuable suggestions and comments during all my thesis work and they
showed their availability, whenever we needed them to evaluate the different stages of
my thesis;
 My colleagues, with whom I shared my scientific ideas and they became my friends on
this occasion: Mathieu LE COZ, Maïmouna Ibrahim, Karima Benhattab, Mahamadou
Koita, Mathieu Coustau, Marie SAVEAN, Paulina P. LOPEZ, Alexandre NOU and Julie
Carreau;
 All the members of HSM and especially Kristine GUJDA, Nadine PERES, Sandra
LAGOUTTE, Roger SARAIVA and Catherine MARCHAND, who were always very kind
to help me in the administrative matters;
 All my friends, especially Muhammad Ashraf, Saqib Ali and Ilyas CRESPIN, who
encouraged and prayed for me during all the difficult moments of my thesis;
 My family, who never forgot me in their prayers;
 Eric Servat, director of UMR HydroSciences Montpellier, for welcoming me during these
three years and provided me with the logistics support required. I was happy to work in
pleasant environment, surrounded by a friendly team;
Lastly, I offer my regards and blessings to all of those who supported me in any respect
during the completion of my thesis.
Adnan Ahmad TAHIR
2
TABLE OF CONTENTS
Résumé étendu de « l’Introduction générale » (en français) 5

General Introduction 8
1. Background and importance of this study 8
2. Main objectives of our study 12
3. Organisation of thesis 12
Résumé étendu du « Chapitre 1 » (en français) 14
Chapter 1. General area of study 16
1.1 Upper Indus River Basin (UIB) 16
1.2 Hydrological regime of the UIB 19
1.3 Snow cover change in UIB 21
1.4 Water resources management in UIB 24
1.5 Sub-catchments of UIB focused in this study 26
Chapter 2. Snow cover dynamics and the hydrological regime 29
2.1 Brief introduction and background 29
2.2 Description of the data sets used in this study 32
2.2.1 ASTER GDEM 32
2.2.2 Hydrological and climate data 34
2.2.3 MODIS snow cover 35
2.2.4 ASTER satellite images 37
2.3 Analysis of snow cover dynamics and hydrological regime of the Hunza 39
River basin (published research paper in HESS)
2.3.1 Some additional results of climate data analysis for Hunza River basin 73
2.4 Snow cover dynamics and hydrological regime of the Astore River basin 80
(western Himalaya)
2.4.1 Characteristics of study area 80
2.4.2 Data sets and treatment 82
2.4.3 Results and discussion 82
2.5 Summary of the results found in this study for the snow cover dynamics 96
and hydrological regime analysis in the Hunza and Astore River basins
2.6 Conclusion 98
Chapter 3. Snowmelt runoff modeling 103
3.1 Brief introduction and background 103
3.2 Description of Snowmelt Runoff Model (SRM) 105
3
3.3 SRM application in the Hunza River basin (publishced research paper in 106
JOH)
3.4 Application of snowmelt runoff model in the other snow and glacier fed 146
sub-catchments of the Upper Indus River Basin
3.4.1 Gilgit River basin 148
3.4.2 Astore River basin 156
3.4.3 Shyok River basin 164
3.5 Comparison of Hunza, Gilgit, Astore and Shyok sub-catchments 172
3.5.1 Comparison of the main characteristics 174
3.5.2 Comparison of the SRM efficiency for four sub-catchments of UIB 177
3.6 Conclusion 179
Chapter 4. Sustainability of water resources in the Indus basin 182
4.1 Indus basin water resources and their confronting challenges 182
4.2 Current water supply and demand: Tarbela Dam 188
4.3 Future water supply and demand: Post Diamer Basha Dam perspective 191
4.3.1 Water allocation for irrigation and domestic use from Diamer Basha 193
Dam
4.3.2 Effect of Diamer Basha dam on the sedimentation control 195
4.3.3 Hydropower generation capacity of Diamer Basha Dam 196
4.4 Change in water availability in the Upper Indus River Basin (Present and 199
future scenarios)
4.5 Impact of the environmental changes on the Tarbela and Diamer Basha 202
system
4.6 Conclusion 205
Résumé étendu de la « Conclusion générale » (en français) 207
General conclusion 210
References 215
List of Figures 223
List of Tables 230
Annex 232
Abstract / Résumé 233
4
Résumé étendu de L’INTRODUCTION GENERALE
Le Pakistan (Fig. 1), avec la sixième population du monde (175 millions d’habitants,
United Nations, 2010) possède une économie très largement basée sur l’agriculture.
Plus de 60% de ses habitants vit en zones rurales et leur moyens de subsistance
dépendent directement ou indirectement des activités agricoles. Ce secteur apporte
aussi la matière des industries agro-alimentaires nationales, comme celles du sucre,
du cuir ou des textiles. Toutefois, l’importance de l’agriculture pour la croissance du
Pakistan et le bien-être de ses habitants ne doit pas être surévaluée. Le Pakistan est
un pays aride à subaride et la contribution des précipitations est très limitée dans les
régions d’agriculture irriguée. Au cours de son histoire, les habitants se sont adaptés
aux précipitations faibles et mal distribuées en s’installant au bord des rivières et en
gérant avec soin les ressources locales en eau. L’une des plus grandes civilisations
humaines – la Civilisation de la Vallée de l’Indus (3300–1300 av. J.C, qui a culminé
entre 2600–1900 av. J.C) – a fleuri au bord des berges de l’Indus
(http://en.wikipedia.org/wiki/Indus_Valley_Civilization). Le faible rendement de
l’utilisation de l’eau par l’homme a considérablement changé avec l’apparition de
l’irrigation à grande échelle au XIXème Siècle. Le système d’irrigation du bassin de
l’Indus (Indus Basin Irrigation System, IBIS) est devenu le plus grand réseau continu
d’irrigation de la planète. La majeure partie de ce réseau d’irrigation est alimenté par
l’Indus et ses deux affluents orientaux, la Jhelum et la Chenab (Fig. 2). L’historien
grec Hérodote (Vème Siècle av. J.C.) décrivait l’Egypte comme « un don du Nil » et
encore comme « un pays gagné par les égyptiens et donné par le Nil ». Ces citations
peuvent également s’appliquer à la relation entre le Pakistan, l’Indus et ses affluents
principaux, les rivières Jhelum, Chenab, Ravi, Beas et Sutlej (Fig. 2).
Le bassin versant de l’Indus s’étend sur quatre pays, le Pakistan, l’Inde, la Chine et
l’Afghanistan avec la plus grande partie au Pakistan. L’économie de l’eau au
Pakistan a dû faire face à plusieurs grands défis au cours du dernier demi-siècle.
L’un de ces défis a été le traité de l’Indus signé par l’Inde et le Pakistan en 1960 sous
l’égide de la Banque Mondiale. Ce traité accorde au Pakistan l’autorité sur les eaux
de l’Indus, de la Jhelum et de la Chenab. Les terres irriguées du Punjab, la plus
grande province agricole du pays, étaient donc privées par ce traité des eaux vitales
des rivières Ravi, Beas et Sutlej. Un hiatus existait entre les apports principaux des
rivières à l’ouest et la grande zone irriguée à l’est. Ce constat a déterminé en 1974 la
5
construction du plus grand barrage en terre de la planète sur l’Indus à Tarbela (Fig. 1
et 2). Tarbela est le premier réservoir contrôlant l’Indus et la plupart des apports
annuels emmagasinés viennent du haut bassin et de la fonte de neige et des glaciers
des chaînes de l’Hindukush-Karakoram-Himalaya (HKH). Cette eau est ensuite
distribuée dans les terres irriguées à l’aval par un large réseau de barrages, de
canaux et de petits cours d’eau.
En d’autres mots, la durabilité de l’économie et de la sécurité alimentaire
pakistanaise dépend des ressources neigeuses et glaciaires des chaines
montagneuses HKH et de leur gestion. Tout changement dans les ressources en eau
disponibles lié à la variabilité climatique, à des facteurs socio-économiques ou aux
actions politiques internationales aura un impact sérieux sur la sécurité alimentaire et
l’environnement au Pakistan. Il est par conséquent important pour la gestion des
ressources en eau disponible d’étudier la dynamique de la couverture neigeuse et
les régimes hydrologiques du haut bassin de l’Indus dans un contexte de variabilité
climatique. Les objectifs principaux et les méthodes pour les atteindre sont présentés
ci-dessous en suivant l’organisation de ce mémoire.
Principaux objectifs de cette étude.

Cette étude s’intéresse à trois principales fonctions (déjà exposées) du haut bassin
de l’Indus (Upper Indus River Basin, UIB) :
a) La dynamique de la couverture neigeuse, l’impact du changement climatique sur la
cryosphère (neige et glace) et le régime hydrologique de l’UIB ;
b) La modélisation de la fonte de la neige dans l’UIB avec des hypothèses sur la
variabilité du climat futur ;
c) La gestion de la ressource en eau dans l’UIB avec des hypothèses sur les variabilités
futures du climat et de la demande en eau.
La réalisation de ces objectifs commence avec l’analyse du produit satellitaire
« neige » MODIS (MOD10A2) pour estimer les changements de la couverture
neigeuse dans notre région d’étude. L’analyse des données hydrologiques et
climatiques est ensuite abordée pour comprendre le régime hydrologique de la
région.
Les données satellitaires MODIS de couverture neigeuse et les données
hydrologiques et climatiques (précipitation et température) des stations d’observation
au sol ont d’abord été appliquées au modèle SRM (Snowmelt Runoff Model) pour
6
simuler les écoulements journaliers des bassins étudiés. L’application satisfaisante
de SRM a permis ensuite de simuler les écoulements sur la base de scénarios
climatiques pour le futur.
Une présentation des réservoirs existants et futurs est enfin faite en lien avec des
scénarios actuels et futurs de demande en eau pour assurer la gestion de l’eau de la
région d’étude.
Organisation du mémoire
Ce mémoire est structuré en quatre chapitres après la présente introduction
générale.
Le premier chapitre expose le contexte général de la région choisie pour cette
thèse. Le second chapitre comprend l’étude réalisée sur la dynamique de la
couverture neigeuse et le régime hydrologique de l’UIB, incluant un article de
recherche publié dans la revue internationale « Hydrology and Earth System
Sciences ». Le modèle SRM (Snowmelt Runoff Model) a été appliqué aux quatre
principaux sous-bassins de l’UIB où la neige et la glace jouent un rôle hydrologique
significatif (Hunza, Gilgit, Astore and Shyok) (Fig. 2.1) et les résultats sont exposés
dans le chapitre 3. Ce chapitre répond au second objectif de cette étude. Les
résultats de l’application de SRM sont inclus sous la forme d’un article de recherche
accepté pour la publication dans la revue internationale « Journal of Hydrology ». Le
chapitre 4 est orienté vers le troisième objectif de cette étude. Il dresse un tableau
de la gestion actuelle et future des ressources en eau avec des changements de
débits, de population et de demande en eau.
A la fin de ce mémoire, la conclusion générale du mémoire expose les implications
possibles de la recherche conduite pour des applications futures concernés par le
changement climatique et l’hydrologie de montagne dans le haut bassin de l’Indus.
7
GENERAL INTRODUCTION
1. Background and importance of this study

Pakistan (Fig. 1), a country with sixth largest population (175 million) in the world
(United Nations, 2010), is dependent highly on its agricultural based economy.
Approximately, more than 60 percent of country's population living in rural areas is
directly or indirectly dependent on agriculture for their livelihood. Moreover, this
sector provides raw material to domestic agro-based industries, such as sugar,
leather, and textiles. Hence, the importance of agriculture to the economic growth of
Pakistan and the well being of its people cannot be overstated. Pakistan is an arid to
semi arid country and the rainfall contribution is very little in the irrigation of
agricultural lands. Throughout history, people have adapted to the low and poorly
distributed rainfall by either living along river banks or by careful management of local
water resources. One of the greatest of human civilizations—the Indus Valley
Civilization (3300–1300 BCE; mature period 2600–1900 BCE) —flourished along the
banks of the Indus River (http://en.wikipedia.org/wiki/Indus_Valley_Civilization). The
low level balance of man and water was dramatically shifted with the advent of large-
scale irrigation technology, in the nineteenth century. The Indus Basin Irrigation
System, IBIS became the largest contiguous irrigation system in the world. A major
proportion of the water supply to this irrigation system is contributed by the Indus
River and its two eastern tributaries—Jhelum and Chenab (Fig. 2). The Greek
historian Herodotus (5th century BC) described Egypt as ―the gift of the Nile‖ and as
―a land won by the Egyptians and given them by the Nile‖. These remarks could
equally be applied to the relationship between Pakistan and the River Indus and its
major tributaries, the Jhelum, Chenab, Ravi, Beas and Sutlej (Fig. 2).
8
Fig. 1 A map of Pakistan, presented with its provincial/administration boundary lines. Five provinces (Punjab, Sindh, Balochistan,
Khyber Pakhtunkhwah and Gilgit Baltistan) are presented with their major cities.
9
The catchment area of the River Indus is located in 4 countries (Pakistan, India,
China and Afghanistan) with the largest part in Pakistan. The water economy of
Pakistan has faced some big challenges during the last half century. One of these
challenges was the Indus Water Treaty signed by India and Pakistan (brokered by
the World Bank), in 1960. According to this treaty, Pakistan has the rights to the
waters of Indus, Jhelum and Chenab rivers. Irrigated lands of Punjab (Fig. 1) (main
producer of Agriculture) were deprived from the life giving waters of Ravi, Beas and
Sutlej rivers according to this treaty. A mismatch was created between the location of
Pakistan’s main water carriers (western rivers) and the major irrigated areas in the
east. This led in 1974 to the construction of the world’s largest earth-fill dam, the
Tarbela dam (Figs. 1 and 2) on the River Indus. Tarbela is the first controlling storage
on the River Indus and most of the annual Upper Indus River influx stored at Tarbela
is derived from the snow and glacier melt in Hindukush-Karakoram-Himalaya (HKH)
ranges. This stored water is then supplied to the downstream irrigated lands through
a network of barrages, canals and small watercourses.
In other words, the sustainability of Pakistan’s economy and food security is
dependent on these water resources (snow and glaciers) of HKH ranges and their
management. Any change in these available water resources through climate
variability, socio-economic factors or international policies will have a serious impact
on food security and the environment in Pakistan. Therefore it is very important to
study the snow cover dynamics and the hydrologic regime of this Upper Indus region
under climate variability to manage the available water resources. The main
objectives and the method of achievement of these objectives are presented in the
next section following by the organisation of this thesis.
10
Fig. 2 Schematic diagram of the Indus River System in Pakistan with its major
tributaries (Source: WAPDA). (See fig. 4.1 for a detailed scheme)
11
2. Main objectives of our study
This study is oriented towards three important aspects (discussed previously) of the
Upper Indus River Basin (UIB):
a) Snow cover dynamics, impact of climate change on the cryosphere (snow and ice)
and the hydrological regime in the Upper Indus River Basin (UIB);
b) Modeling snowmelt runoff in the UIB under future climate variability;
c) Water resources management in the UIB under future climate change and water
demand.
The accomplishment of these objectives starts with the analysis of MODIS snow
cover product (MOD10A2) to estimate the snow cover area change of the region
under study. The analysis of the hydrological and climate data are then treated and
discussed to understand the hydrological regime of the area.
MODIS snow cover satellite data and the hydrological and climate data (precipitation
and temperature) of the ground stations was used in the snowmelt-runoff model
(SRM) to simulate the daily stream flows from the studied catchments. The
successful application of the snowmelt-runoff model led us to simulate the stream
flows under future climate change scenarios using the model SRM.
An overview of the existing and future water reservoirs is presented under the
present and future water demand scenarios to sustain the water resources
management in the area of study.
3. Organisation of thesis
The thesis is structured in four chapters following this section of general introduction.
Chapter 1: First chapter outlines the general area of study chosen for this thesis.
We present the different characteristics of the zone of study in this chapter. The
geography, snow cover change and hydrograph of the study area are presented to
understand the hydrological regime of the area. The satellite data used to present the
characteristics of the study area are ASTER Global Digital Elevation Model (GDEM)
and MODIS (MOD10A2) snow cover data.
Chapter 2: Second chapter is comprised of the study completed on the snow cover
dynamics and hydrological regime of the UIB. This chapter is oriented towards the
first objective of this study. We selected a high altitude sub-catchment of UIB, named
Hunza River basin (central Karakoram) (Fig. 2.1), for the detailed analysis of snow
cover dynamics and hydrological regime. This chapter contains the study results in
12
form of a research article accepted and published in an international journal
―Hydrology and Earth System Sciences‖ and some extra results including the snow
cover dynamics in Astore River basin (western Himalaya) (Fig. 2.1) to compare the
central Karakoram and western Himalaya snow cover change.
Chapter 3: The Snowmelt Runoff Model (SRM) was applied on the four main snow
and glacier fed sub-catchments of UIB (Hunza, Gilgit, Astore and Shyok) (Fig. 2.1)
and the results are presented in chapter 3. This chapter addresses the second
objective of this study. Hunza River basin was studied in detail for the snowmelt
runoff modeling over four hydrological years and under different future climate
scenarios (notably temperature and snow cover change). SRM application results in
Hunza basin are included in this chapter in the form of a research article accepted for
publication in an international journal ―Journal of Hydrology‖.
Chapter 4: Chapter 4 is oriented towards the third main objective of this study. This
chapter concerns about an overview of the present and future water resources
management under changed stream flow, population and water demand. The stream
flow in the all tributaries of Upper Indus River was modeled using SRM under the
climate change scenario of year 2075 and the results are presented in this chapter
with some technical data of Tarbela and proposed Diamer Basha dams.
At the end of thesis a general conclusion of our study is presented with the possible
implications of this study in future research regarding the climate change and
mountain hydrology of Upper Indus River Basin.
Overall, this study will help us to understand the hydrological processes of the Upper
Indus River Basin and to improve water resources management in the perspective of
these hydrological processes.
13
Résumé étendu du CHAPITRE 1
CONTEXTE GENERAL DE L’ETUDE
Ce chapitre comprend les fondamentaux et les caractéristiques du site général de

l’étude, le haut bassin de l’Indus (Upper Indus River Basin - UIB à l’amont du barrage
de Tarbela (Fig. 1.1). L’UIB a une superficie totale de 206 000 km² à Tarbela, selon
nos calculs à partir du modèle numérique de terrain GDEM. Il inclut les sous-bassins
de Hunza, Gilgit, Astore, Shigar and Shyok (Fig. 1.1). Près de 11,5% (22 000 km²) de
sa surface totale est couverte par des glaciers permanents (dont plusieurs des plus
grands glaciers de vallée de la planète) faisant d’elle la plus large zone englacée
hors des régions polaires et du Groenland (Hewitt, 2001; Hewitt, 2007). L’UIB a une
altitude moyenne de 4750 m avec presque 60% de sa superficie au-dessus de
4500 m et 12% au-dessus de 5500 m (presque équivalent à la surface englacée)
(Fig. 1.2).
Les apports en eau à Tarbela sont mesurés à Besham Qila, situé 80 km environ à
l’amont de Tarbela (Fig. 1.1), avec un débit moyen annuel de 2 410 m3/s (équivalent
à 370 mm de lame d’eau) (Fig. 1.3). Une précipitation moyenne annuelle d’environ
400 mm est estimée dans notre étude à partir des données disponibles aux
différentes stations de l’UIB. La première source d’écoulement vient des
précipitations de faible intensité d’hiver et de printemps apportées par les circulations
d’ouest. La dénivelée du bassin s’inscrit entre 455 m et 8611 m (cette dernière
correspond au sommet du K2 – second plus haut sommet de la planète) (Fig. 1.1) et
la première conséquence est que le climat varie beaucoup à l’intérieur du bassin. Le
débit écoulé est une combinaison de ruissellement rapide dans la partie la plus aval
de l’UIB et du débit plus lent issu de la neige et des glaciers dans les parties hautes
du bassin. Plus de 60% de l’écoulement annuel de l’UIB vient de la fonte de la neige
dans le HKH (Bookhagen and Burbank, 2010).
Le couvert neigeux sur la région de l’UIB varie de 10% à 70% (Fig. 1.4), estimés à
partir de 450 images MODIS MOD10A2. Le couvert neigeux de la région atteint un
maximum de 70-80% pendant la période d’accumulation d’hiver (décembre à janvier)
et un minimum de 10-15% en été (juin à septembre) lors de la fonte de la neige (Fig.
1.4 & Fig. 1.5). Les hautes eaux issues de la fonte neigeuse et glacière pendant les
mois d’été constituent la clé du remplissage des réservoirs (Tarbela) à l’exutoire du
14
bassin, lequel alimente ensuite le système d’irrigation du bassin de l’Indus (IBIS)
pendant les périodes de basses eaux, afin d’irriguer les importantes cultures de blé
et de coton.
Il n’y a pour l’instant qu’un seul réservoir dans l’UIB, celui du barrage de Tarbela.
Presqu’un tiers de la demande en irrigation du pays à l’aval de l’UIB est satisfaite
uniquement par ce barrage. Celui-ci était aussi une protection contre de nombreux
événements extrêmes en stockant un maximum d’eau grâce à un réservoir de
grande capacité. Aujourd’hui il ne suffit plus à satisfaire les demandes croissantes
d’alimentation, d’énergie et d’eau domestique, essentiellement liées à une
croissance accélérée de la population. En sus, sa capacité de stockage diminue à
cause d’une sédimentation rapide. La capacité originale de Tarbela a déjà été réduite
de presque 30% entre 1974 et 2009 (Khan and Tingsanchali, 2009). En limitant le
stockage, les crues sont moins bien contrôlées dans la zone aval. Un exemple récent
est la crue dévastatrice de 2010 qui a affecté un tiers de la surface du Pakistan (Fig.
1.6) et, concerné directement ou indirectement 2,5 millions de personnes (BBC,
2010).
Il est par conséquent important d’étudier les impacts du changement climatique sur la
couverture neigeuse, les glaciers et le régime hydrologique de l’UIB, ainsi que de
simuler les débits avec des scénarios de climat futur pour améliorer la gestion des
ressources en eau.
Pour mieux comprendre les changements du couvert neigeux et le régime

hydrologique de l’UIB, l’étude est conduite à l’échelle de sous-bassins. Quatre sous-
bassins ont été sélectionnés, ceux des rivières Hunza, Gilgit, Astore and Shyok
parce qu’ils disposent de séries importantes de données hydrologiques et
climatologiques. La superficie enneigée est estimée pour les sous-bassins de Hunza
et d’Astore durant une période de 10 ans (2000-2009), ainsi que pour tout le bassin
de l’UIB (la couverture neigeuse de l’UIB est présentée en Fig. 1.4). Ces deux sous-
bassins ont été choisis à cause de leurs différences de régime climatique et de
situation. Le bassin de Hunza est localisé dans la chaine centrale du Karakoram et
celui de l’Astore dans les piémonts sud de l’Himalaya occidental (Fig. 1.1).
15
CHAPTER 1
GENERAL AREA OF STUDY
This chapter encloses the background and characteristics of the general area of
study i.e. Upper Indus River Basin. The general characteristics and a global digital
elevation model of UIB is presented in section 1.1 followed by the hydrological
regime of the catchment in section 1.2. Snow cover change in UIB over 10-years
period is presented in section 1.3 and this chapter ends with a brief description of the
water resources management in the UIB.
1.1 Upper Indus River Basin (UIB)

The catchment area of the Indus River upstream of Tarbela is called the Upper Indus
River Basin (UIB) (Fig. 1.1). An Advanced Spaceborne Thermal Emission and
Reflection Radiometer (ASTER) global digital elevation model (GDEM) (24.18 m x
24.18 m grid size) of the catchment area is presented in Fig. 1.1, with some essential
geographical layers and main sub-catchments. The UIB has a total catchment area of
206,000 km² at Tarbela, as calculated from GDEM, in this study. The Upper Indus
River Basin includes the Hunza, Gilgit, Astore, Shigar and Shyok sub-basins (Fig.
1.1). Nearly 11.5% (22,000 km²) of the total area of the UIB is covered by perennial
glacial ice (including most of the largest valley glaciers) making it the largest area
outside the polar and Greenland regions (Hewitt, 2001; Hewitt, 2007). The
hypsometric curve of UIB and the percentage area between 500-m altitudinal layers
are estimated from ASTER GDEM and presented in Fig. 1.2. UIB has a mean
elevation of 4750 m with almost 60% of its total area above an elevation of 4500 m
and 12% of its area (almost the same area is glacier covered) above 5500 m (Fig.
1.2). A description of the hydrological regime and the cryosphere (snow cover and
glacier) change in this region (Upper Indus River Basin) is given in the next sections.
16
Fig. 1.1 ASTER Global digital elevation model (GDEM) of Upper Indus River Basin (UIB) and the location of UIB.
17
Area per m elevation band ()
0 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
9000
8500
8000
7500
7000
6500
6000
Elevation (m)
5500
5000
4500
4000
3500
3000
2500
2000
1500 Area per 500-m elevation band
Hypsometric curve
1000
500
0
104
2.0x104
3.0x104
4.0x104
5.0x104
6.0x104
7.0x104
8.0x104
9.0x104
105
1.1x105
1.2x105
1.3x105
1.4x105
1.5x105
1.6x105
1.7x105
1.8x105
1.9x105
2.0x105
2.1x105
0
Cumulated area (km²)

Fig. 1.2 Hypsometric curve of the Upper Indus River Basin and the distribution of area under each 500-m elevation band/layer.
18
1.2 Hydrological regime of the UIB
Inflow to Tarbela is measured at Besham Qila, situated approximately 80 km
upstream of Tarbela (Fig. 1.1), with a mean annual flow of 2410 m 3/s (i.e. 370 mm of
water depth equivalent) (Fig. 1.3) as calculated in this study from the flow records of
1969‒2008 (data provided by the Surface Water Hydrology Project, SWHP). An
average annual precipitation of approximately 400 mm is estimated in this study from
available data records of different climate stations (Fig. 1.3) in UIB. These
precipitation records are available only below 4700 m elevation and almost 50% of
UIB area (Fig. 1.2) is above this elevation where the maximum snow accumulation
occurs.
The climatic records in the UIB are different from the eastern Himalayas (Fowler and
Archer, 2006; Young and Hewitt, 1990). A major part of the basin (90%) is in the rain
shadow of the Himalayas and not affected by the summer monsoon (Immerzeel et
al., 2009). The primary source of water for UIB is originated from low intensity winter
and spring precipitation brought by westerly circulations. The altitude within the basin
ranges from 455 m to 8611 m, (with 8611 m corresponding to the K2 peak — 2nd
highest summit on the earth) (Fig. 1.1) and as a result the climate varies greatly
within the basin. Archer (2003) suggested that UIB can be divided into three
hydrological regimes as a) high altitude catchments (e.g. Hunza and Shyok) with
summer runoff derived mainly from the snow and glacier melt with concurrent energy
input in the form of temperature b) middle altitude catchments situated at the extreme
west of Himalaya (e.g. Astore) where the summer runoff is defined by preceding
winter precipitation and c) low altitude (foothill) catchments that have a runoff regime
controlled by the current winter or summer rainfall.
Stream flow is a combination of fast rain runoff in the lower part of the UIB basin and
slow snow and glacier runoff of the higher parts of the basin (Ali and Boer, 2007;
Archer, 2003). The melt water component is extremely important and the primary
source for irrigation of the entire Indus basin. More than 60% of annual flow in the
Upper Indus River is contributed by the snowmelt in the HKH (Bookhagen and
Burbank, 2010). The change in snow cover area for UIB is given in the next section.
19
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
12000 0
11000 20
10000 40
Avergae discharge (1969-2008)

9000 Minimum avergae discharge 60
year (1993)
Monthly precipitation (mm)

Maximum average discharge
8000 year (1973) 80
Astore basin precipitation
Discharge (m /s)
Gilgit basin precipitation

7000 100
3
Shyok basin precipitation

Hunza basin precipitation
6000 UIB average precipitation 120
5000 140
4000 160
3000 180
2000 200
1000 220
0 240
Time period (Monthly)
Fig. 1.3 Mean monthly discharge (1969‒2008) of Upper Indus River at Besham Qila (estimated from the data available in this study)
and mean monthly precipitation in Astore, Gilgit, Shyok, and Hunza basin.
20
1.3 Snow cover change in UIB
The Moderate Resolution Imaging Spectroradiometer (MODIS) snow products were
selected to calculate the snow cover percentage in UIB. The MODIS/Terra Snow
Cover 8-Day L3 Global 500-m Grid (MOD10A2) used for this study contains data
fields for maximum snow cover extent over an 8-day repeated period (Hall et al.,
2006, updated weekly). A data set of 450 processed MOD10A2 (V005) images
available from March 2000 to December 2009 was downloaded from
http://nsidc.org/cgi-bin/snowi/search.pl. WGS 1984 UTM ZONE 43N projection
system was used for UIB. When the percentage of cloud cover exceeded 20% on a
specific date, the record was eliminated and the average snow cover on this date
was estimated by interpolating linearly between the previous and the next available
cloud-free images.
The average snow cover area in the Upper Indus River Basin changes from 10% to
70% (Fig. 1.4). Snow cover in the area is at a maximum 70‒80% in the winter
(December to February) snow accumulation period and at a minimum 10‒15% in the
summer (June to September) snow melt period (Fig. 1.4 & Fig. 1.5). The year 2009
was the maximum snow year with an average snow cover of 49% and the year 2007
was the minimum snow year with an average snow cover of 34%, during the last
decade. The discharge in the River Indus is directly correlated to the change in snow
cover area. The discharge is at its low during the snow accumulation period and it
starts to increase in April (Fig. 1.3) as the mean temperature in the catchment starts
to increase, especially on low elevation snow cover areas. The discharge in the River
Indus reaches at its maximum in July as the glacial melt starts after the melting of
fresh snow in the catchment. This high flow derived from snow and glacier melt is the
key to fill the storage reservoirs (Tarbela) at the outlet of catchment which is then
supplied to the Indus Basin Irrigation System during low flow periods to irrigate the
important crops of wheat and cotton. Water resources of UIB and their management
under the availability of stored volume of water are highlighted in the next section.
21
(Data source: MODIS (MOD10A2) snow product)
Snow cover area, SCA ()
0
10
20
30
40
50
60
70
80
90
100
05-Jan
13-Jan
21-Jan
29-Jan
06-Feb
14-Feb
22-Feb
02-Mar
10-Mar
18-Mar
26-Mar
03-Apr
11-Apr
19-Apr
27-Apr
05-May
13-May
21-May
29-May
06-Jun
14-Jun
22-Jun
30-Jun
08-Jul
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
16-Jul
24-Jul
01-Aug
09-Aug
17-Aug
25-Aug
02-Sep
Time period (days classification)
10-Sep
18-Sep
26-Sep
04-Oct
Minimum average SCA year (2007)
Maximum average SCA year (2009)
12-Oct
20-Oct
Average SCA over 10 years (2000-2009)
28-Oct
05-Nov
13-Nov
21-Nov
29-Nov
07-Dec
15-Dec
23-Dec
Fig. 1.4 Percentage snow cover in the Upper Indus River Basin estimated by analysing 450 MOD10A2 images (2000‒2009).
31-Dec
22
Fig. 1.5 MODIS (MOD10A2) satellite images presenting the average snow cover area for each month in the Upper Indus River
Basin over the year 2009.
23
1.4 Water resources management in UIB
There is only one storage capacity, Tarbela Dam, in the Upper Indus Basin. Tarbela
is a multipurpose dam constructed in 1974 which revolutionized the country’s
agriculture by increasing the irrigation command area and cultivable land in Pakistan.
Tarbela along with Mangla dam (on River Jhelum) increases total cultivable area by
20% and these two satisfy almost one third of the country’s electricity demand
(Archer et al., 2010). Almost one third of country’s irrigation water demand
downstream of UIB is satisfied alone by Tarbela reservoir. It also prevents many
extreme event discharges by storing maximum of water in large capacity reservoir.
Today, the present water reservoirs are not sufficient to satisfy the country’s
increasing food, energy and domestic water demands, notably due to the fast
increasing population. Additionally, the present storage capacities are decreasing by
fast sedimentation. Initial capacity of Tarbela is already reduced by almost 30% in
2009 (Khan and Tingsanchali, 2009). This is giving rise to the reduced water store
and increasing floods in the downstream areas.
A recent example is the devastating flood of 2010 in Pakistan, which affected one
third of country’s surface area (Fig. 1.6) and 2.5 million people directly or indirectly
(BBC, 2010). This flood was a combination of climate factors and the lack of national
decisions on water use. July and August, 2010 were already the second hottest
months on record (NOAA, 2010) that makes the high runoff from snow and glacier
melt, whereas, the monsoonal rainfall (July-August) above than average in Pakistan
made it worst for the rivers already filled to capacity.
It is, therefore, important to study the climate change impacts on the snow covers,
glaciers and hydrological regime of UIB and to simulate the UIB discharge under the
future climate change scenarios to sustain the water resources management.
24
Fig. 1.6 Map showing the flood affected areas of Pakistan in 2010 and the track of
flood wave along the Indus River. (Sources: USAid & Pakistan Meteorological
Department)
25
1.5 Sub-catchments of UIB focused in this study
For the better understanding of snow cover change and hydrological regime of the
Upper Indus River Basin, this study is carried out on the sub-catchment level. Four
sub-catchments i.e. Hunza, Gilgit, Astore and Shyok river catchments, are selected
for the detailed study. These sub-catchments were selected, because, the maximum
hydrological and climate data for UIB is available from these four sub-basins.
The snow cover area is estimated for Hunza and Astore sub-catchments for a 10-
years period (2000‒2009) along with the snow cover area estimation in overall UIB
(snow cover area for UIB is presented in Fig. 1.4). These two sub-catchments are
selected because of their different climate regime and location. Hunza River basin is
situated in the central Karakoram and the Astore River basin is situated at the
southern foothills of western Himalaya (Fig. 1.1). Regarding the long and time
consuming process of the treatment of MOD10A2 satellite images, the snow cover
area for other sub-catchments (Gilgit and Shyok) was estimated only over two
hydrological years to calibrate the snowmelt runoff model (discussed in chapter 3).
The snow cover dynamics and hydrological regime of the Hunza and Astore River
basins are presented in the next chapter.
26
Résumé étendu du CHAPITRE 2
DYNAMIQUE DE LA CRYOSPHERE ET REGIME HYDROLOGIQUE
La vie économique du Pakistan dépend pour une large part de son agriculture, qui, à
son tour, est dépendante de l’irrigation à travers un vaste réseau de barrages, de
prises d’eau et de canaux à partir de l’Indus et de ses tributaires. La plus grande part
de l’eau prélevée sur l’Indus à Tarbela provient des sous-bassins situés dans les
montagnes du Karakoram, de l’Himalaya et de l’Hindukush, alimentés par la neige et
les glaciers. La région qui inclut ces sous-bassins est appelée « haut bassin de
l’Indus » (UIB) et ses ressources en eau dépendent essentiellement de la fonte de la
neige et des glaciers. Il est pour cette raison essentiel d’y comprendre la dynamique
de la cryosphère dans une perspective de gestion des ressources en eau. Une base
de données du produit satellitaire de couverture neigeuse issu du capteur MODIS
(MOD10A2) et des données hydrométéorologiques (débits, température et
précipitation) sont utilisées pour analyser la dynamique de la cryosphère et le régime
hydrologique de l’UIB.
Dynamique du couvert neigeux

La base de données satellitaires du capteur MODIS avec le produit de couverture
neigeuse MOD10A2 a été choisie pour analyser les changements entre mars 2000 et
décembre 2009 du couvert neigeux des bassins versants de Hunza et d’Astore. Ce
produit est libre d’accès et facile à traiter ; c’est donc une bonne option pour évaluer
sur le long terme l’extension de la cryosphère sur ces bassins. Une comparaison du
produit MODIS avec des images ASTER à résolution plus fine confirme la
validité/fiabilité de ces données satellitaires de couvert neigeux pour des hautes
altitudes. Au cours des 10 années de données cryosphériques analysées, une
tendance à l’augmentation des précipitations hivernales, la stabilité des températures
moyennes et une tendance à la diminution des écoulements dans le bassin de
Hunza suggèrent que la région de ce bassin est sujette à une légère augmentation
de la couverture cryosphérique, particulièrement en haute altitude. La même
tendance a été trouvée pour le couvert neigeux du bassin d’Astore, et plus largement
pour l’UIB, liée à l’augmentation des précipitations d’hiver et à la tendance à la
baisse des températures d’été dans la région. L’augmentation des écoulements à
Astore contraste avec ceux de Hunza et peut être la conséquence d’une tendance à
27
l’augmentation des pluies de mousson d’été dans le bassin. Ces changements de
couvert neigeux peuvent être utilisés comme hypothèses de climat futur pour simuler
les débits de la rivière Hunza à l’aide du modèle SRM (voir chapitre 3). L’impact du
réchauffement global n’est pas sensible, car une large partie du bassin s’étend à des
hautes altitudes où la température reste négative presque toute l’année.
Régime hydrologique
L’écoulement de la rivière Hunza est principalement influencé par les températures
moyennes d’été et d’hiver, alors que le débit de l’Astore présente une corrélation
positive significative avec les précipitations de l’hiver précédent. De plus, la zone
hydrologique la plus active du bassin de Hunza se situe au-dessus de 4500 m (zone
C). L’étude annuelle et saisonnière de la dynamique du couvert neigeux montre que
sa variabilité est fortement inversement corrélée dans les deux bassins (Hunza et
Astore) avec les températures moyennes d’été qui produisent de forts débits. C’est
pour cela qu’il est très important d’estimer les accumulations hivernales de neige à
haute altitude afin de prévoir les écoulements d’été et finalement d’améliorer la
gestion des réservoirs aval comme celui de Tarbela. Enfin, une chronique
d’observation des chutes de neige dans les zones d’accumulation pourrait être utile
pour comprendre plus efficacement le régime hydrologique de la région.
Les résultats de tendance pour les variables climatiques incitent à modéliser
l’écoulement de fonte de neige sous contrainte de variabilité climatique
(augmentation de l’accumulation de neige en hiver et température moyenne d’été
constante ou légèrement décroissante à Hunza et Astore) afin d’estimer les futurs
écoulements disponibles dans l’UIB pour une meilleure gestion des réservoirs. La
modélisation avec le modèle SRM dans les sous-bassins de l’UIB fait l’objet du
chapitre 3 de ce mémoire.
28
CHAPTER 2
SNOW COVER DYNAMICS AND THE HYDROLOGICAL REGIME
Foreword: This chapter is mainly oriented by a research paper published in HESS

(Hydrology and earth system sciences) on snow cover dynamics and hydrological
regime in the central Karakoram sub-catchment (Hunza) of Upper Indus River Basin
(UIB), included in the second part (section 2.3) of this chapter – In the first part
(sections 2.1 and 2.2), some additional information is given in order to complete the
reading of the paper and to improve the statements and research presented in the
paper. In addition, an analysis of snow cover dynamics and hydrological regime in a
western Himalayan sub-catchment (Astore) of UIB is detailed in section 2.4, in order
to compare its behaviors with the Karakoram region.
2.1 Brief introduction and background

Agriculture based economy of Pakistan is dependent on the irrigation waters supplied
by the River Indus and its tributaries. Most of the flow abstracted from the River Indus
at Tarbela is contributed by its sub-catchments (Fig. 2.1) situated in the Karakoram,
Himalaya and Hindu Kush Mountains and the area comprised of these sub-
catchments is called the Upper Indus River Basin (UIB). The tributaries of the UIB
such as the Shyok, Shigar, Astore, Gilgit and Hunza (Fig. 2.1), are mainly snow and
glacier-fed and have a little effect of monsoon rainfall system (except Astore basin)
because of the high altitude mountain barrier. These catchments contribute nearly
60% of the total Upper Indus flow as calculated by the analysis of stream flow
records in this study; the management of Tarbela reservoir depends to a large extent
on this inflow. No detailed investigation of snow and ice processes or their relevance
to hydrological regime has taken place in most areas of these high mountain ranges.
The active hydrological regions, which generate a bulk of water for the UIB, have no
or scarce climate observation; particularly above 4000 m. Hydrometeorological
investigations are very important in this hydrologically active region at elevations
between 4000 to 7000 m where most of the precipitation falls in the form of snow.
Remote sensing techniques are the only way to analyze snow and glacier cover in
these remote mountains. An understanding of the snow cover dynamics and its
influence on the hydrological regimes of the sub-catchments of UIB is critical for the
management of the water resources of Pakistan.
29
In this phase of the present study, we focused the high altitude Hunza River basin
(central Karakoram region) i.e. one of the snow and glacier fed sub-catchments of the
UIB, and the middle altitude Astore River basin (western Himalaya region) (Fig. 2.1).
Snow cover dynamics in these catchments is analysed by using a remote sensing
snow cover dataset and the climate and hydrological data of gauging stations
situated within or near the catchment boundary (at different altitudes) are used to
study the hydrological regime in this area. A detailed description of the available data
for this study is presented in the section 2.2 and 2.3.
30
Fig. 2.1 Location of Upper Indus River Basin and its sub-catchments.
31
2.2 Description of the data sets used in this study
2.2.1 ASTER GDEM
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER)
Global Digital Elevation Model (GDEM) released by NASA and Japan's Ministry of
Economy, Trade and Industry (METI) to the worldwide public on June 29, 2009 was
used to delineate the basin division studied. It is based on ASTER images (see
section 2.2.4, for the description of ASTER images). The ASTER GDEM is available
for high-latitude and steep mountainous areas not covered by SRTM3 (METI and
NASA, 2009). SRTM (Shuttle Radar Topography Mission) consists of a specially
modified radar system that flew onboard the Space Shuttle Endeavour during an 11-
day mission in February of 2000. SRTM obtained elevation data on a near-global
scale to generate the most complete high-resolution digital topographic database of
earth. The horizontal spacing of the SRTM3 DEM is 90x90 m and the absolute
vertical accuracy is better than 9 m (Farr et al., 2007).
As stated on the web page (http://www.ersdac.or.jp/GDEM/E/4.html), the ASTER
GDEM was created by processing and stereo-correlating the 1.3 million-scene
ASTER archive of optical images, covering land surfaces between 83°N and 83°S
and is composed of 22,600 1°-by-1° tiles. Tiles that contain at least 0.01% land area
are included. The ASTER GDEM is in the GeoTIFF format with geographic lat/long
coordinates and a 1-arc-second (approximately 30 m depending on latitude) grid of
elevation postings. It is referenced to the WGS84/EGM96 geoid. Pre-production
estimated accuracies for this global product were 20 m at 95% confidence for vertical
data and 30 m at 95% confidence for horizontal data.
The ASTER GDEM tiles are downloaded electronically from the
https://wist.echo.nasa.gov/~wist/api/imswelcome/ web page for the Hunza and Astore
River basins. The global digital elevation model (GDEM) for the Hunza River basin is
presented in Fig. 2.2 and for Astore in section 2.4 of this chapter.
32
Fig. 2.2 ASTER Global digital elevation model (GDEM) of the Hunza River basin.
33
2.2.2 Hydrological and climate data
Stream flow measurement in Pakistan is mostly carried out by the Surface Water
Hydrology Project of the Water and Power Development Authority (SWHP-WAPDA)
with the earliest records commencing in 1960. The database of daily flows for the
Hunza River gauged at Dainyor Bridge (Fig. 2.1) from 1966 to 2008 (~40 years) and
for Astore Rive gauged at Doyian (Fig. 2.1) from 1974 to 2007 (~33 years) was made
available for this analysis. The Hunza River at Dainyor Bridge represents the
moderately high runoff catchments in the centre of the Karakoram where the flow is
dependent on the mean spring and summer temperatures whereas the Astore River
is one of the highest runoff catchments in UIB where the annual runoff depends
largely on the winter precipitation.
Climatological measurement is under the responsibility of the Pakistan

Meteorological Department (PMD), which maintains stations with standard
measurements including temperature, precipitation, humidity and wind speed. These
stations have quite long data records with earliest data in 1950s. Stations established
by the PMD are mainly located in lower-elevation valleys, so there is no PMD station
established in the high altitude Hunza River basin. The temperature and precipitation
records at Gilgit installed by PMD just outside the Hunza basin have been used by
the researchers in the past years for testing runoff controls. The WAPDA has
maintained a network of automatic weather stations at high altitudes (located
between 1600 m and 4700 m) of the UIB from 1993 to improve runoff forecasting to
the Indus River and to establish different hydropower projects in the future. Apart
from the usual instruments, some were equipped with ultrasonic snow-depth sensors.
These sensors enable the investigation of seasonal snow cover characteristics in
combination with the measurement of radiation, temperature and humidity (Winiger et
al., 2005).
This network comprises three stations in the Hunza River Basin as well (Figs. 2.1 &
2.2). The records from these stations are available from 1999 to 2008. There are
three climate stations in the Astore River basin (Fig. 2.1). One is installed by PMD
and two stations are installed and managed by WAPDA at high altitudes (detailed
description is given later in section 2.4).
34
2.2.3 MODIS snow cover
selected to calculate the snow cover on our study area. MODIS/Terra Snow Cover 8-
Day L3 Global 500 m Grid (MOD10A2) used for this study contains data fields for
maximum snow cover extent over an 8-day compositing period and a chronology of
snow occurrence observations in HDF-EOS (Earth Observation System) format,
along with corresponding metadata. MOD10A2 consists of 1200x1200-km tiles of
500-m resolution data gridded in a sinusoidal map projection (Hall et al., 2000,
updated weekly).
The MODIS/Terra V005 data set available from March 2000 to December 2009 is
downloaded from http://nsidc.org/cgi-bin/snowi/search.pl. MODIS cryosphere data is
based on a snow mapping algorithm that employs a Normalized Difference Snow
Index (NDSI) and other criterion tests (Hall et al., 2000, updated weekly; Hall et al.,
2002). The snow mapping algorithm differentiates pixels as snow, ice lakes, cloud,
water, land or other. Snow extent is the primary variable of interest in this data set.
Version 5 (V005) of the MOD10A2 snow products is the latest version and provides
the best quality of data used in this study. Our present database used in this study
consists of 453 processed MOD10A2 images for each of the Hunza and Astore River
basin. As illustration, one MODIS image to show the average snow cover area for
each month over the year 2009 in Hunza River basin is shown in Fig. 2.3. MODIS
snow cover images for Astore River basin are presented in section 2.4.
35
Fig. 2.3 MODIS (MOD10A2) satellite images presenting the average snow cover area for each month in the Hunza River basin over
the year 2009.
36
2.2.4 ASTER satellite images
ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) is an
imaging instrument flying on a satellite called Terra launched in December 1999 as
part of NASA’s Earth Observing System (EOS). The ASTER instrument consists of
three separate instrument subsystems (Visible and Near Infrared [VNIR], the
Shortwave Infrared [SWIR] and the Thermal Infrared [TIR]) operating in different
spectral regions that have 15-m, 30-m and 90-m spatial resolutions, respectively
(Lopez, 2007; Sarwar and Bill, 2003).
The Normalized Difference Snow Index (NDSI) calculated with equation (1), has
been widely used for effective snow cover mapping using various sensors (Dankers
and de Jong, 2004; Hall et al., 2002; Nagler et al., 2008; Sirguey et al., 2008). The
NDSI was applied to ASTER images to enhance the snow cover extent.
(1)
The aim of this spectral index is to enhance the snow and ice targets and distinguish
them from vegetation, bare ground and clouds. The NDSI index does not differentiate
between snow and snow-covered ice as stated by Lopez (2007), so both of them are
included in the results obtained with the NDSI.
Four ASTER images, nearly cloud-free, in different seasons, were downloaded from
http://glovis.usgs.gov/ImgViewer/Java2ImgViewer.html to validate the MODIS snow
cover product, in this study. The dates of these images are 16 July 2000, 30 April
2001, 26 June 2001 and 28 February 2002. A visual comparison of four ASTER and
MODIS images is presented in Fig. 2.4. A comparison of percentage snow and ice
cover area on these four images is given later in table 2.4. The results of this
comparison suggest that the MODIS (MOD10A2) snow product is reliable to use in
the snow cover analysis for the present and future studies.
37
Fig. 2.4 A comparison of ASTER and MODIS images, on 4 different seasons.
38
2.3 Analysis of snow cover dynamics and hydrological regime of the Hunza River
basin
The analysis of snow cover dynamics and hydrological regime of the Hunza River
basin was carried out using the satellite and gauging stations data described above.
The results obtained are presented next in detail in the form of a research article
published (Tahir et al., 2011a) in the journal of ―Hydrology and Earth System
Sciences (HESS)‖.
39
Snow cover dynamics and hydrological regime of the Hunza River basin,
Karakoram Range, Northern Pakistan
Adnan Ahmad TAHIR1,*, Pierre CHEVALLIER1, Yves ARNAUD2 & Bashir

AHMAD3
1
Laboratoire Hydrosciences – UMR 5569 (CNRS, IRD, Montpellier University 1&2), CC57, Université
Montpellier 2, 34095 MONTPELLIER CEDEX 5, France. (E-mail: chevalli@msem.univ-montp2.fr,
Pierre CHEVALLIER)
2
Laboratoire des Transferts en Hydrologie et Environnement – UMR 5564 (CNRS, IRD, Université
Joseph Fourrier, Grenoble INP), LGGE, 54 rue Molière, Domaine Universitaire, BP 96, 38402 SAINT
MARTIN d’HERES CEDEX, France. (E-mail: yves.arnaud@ird.fr)
3
Director (Environment), Natural Resources Division, Pakistan Agricultural Research Council, G1/5,
ISLAMABAD, Pakistan. (E-mail: bashirad@hotmail.com)
* Corresponding author. [E-mail address: tahir@msem.univ-montp2.fr; uaf_adnan@hotmail.fr (Adnan

Ahmad TAHIR)], Tel.: +33 467 14 90 64, Fax. +33 467 14 47 74
Abstract
A major proportion of flow in the Indus River is contributed by its snow- and glacier-
fed river catchments situated in the Himalaya, Karakoram and Hindukush ranges. It
is therefore essential to understand the cryosphere dynamics in this area for water
resource management. The MODIS MOD10A2 remote-sensing database of snow
cover products from March 2000 to December 2009 was selected to analyse the
snow cover changes in the Hunza River basin (the snow- and glacier-fed sub-
catchment of the Indus River). A database of daily flows for the Hunza River at
Dainyor Bridge over a period of 40 years and climate data (precipitation and
temperature) for 10 years from three meteorological stations within the catchment
was made available to investigate the hydrological regime in the area. Analysis of
remotely sensed cryosphere (snow and ice cover) data during the last decade
(2000‒2009) suggest a rather slight expansion of cryosphere in the area in contrast
to most of the regions in the world where glaciers are melting rapidly. This increase in
snow cover may be the result of an increase in winter precipitation caused by
westerly circulation. The impact of global warming is not effective because a large
part of the basin area lies under high altitudes where the temperature remains
negative throughout most of the year.
40
1. Introduction and background
An agricultural country, Pakistan is highly dependent on the Indus irrigation system,
one of the largest irrigation networks in the world (SIHP, 1990). The Indus River
emerges from the Tibetan Plateau and flows toward northern areas of Pakistan
where it changes its direction toward the south and flows into the Arabian Sea (Fig.
2.5). It has a controlling storage at Tarbela dam as the river descends from the
mountains. The catchment area upstream of Tarbela reservoir is called the Upper
Indus River Basin (UIB) (Fig. 2.5), which contributes the main inflow to the main
course of the Indus River at high elevations as a result of snow and glacier melt; the
lower catchment is rain-fed. Tarbela is the first major structure on the Indus River and
supplies the flow to the Indus Irrigation System to irrigate the agricultural lands of
Punjab (a province of Pakistan), the dominant producer of agriculture products in the
country. Inflow to Tarbela is measured at Besham Qila, situated approximately 80 km
upstream of Tarbela (Fig. 2.6), with a mean annual flow of 2410 m 3/s (standard
deviation for mean annual flow = 309 m 3/s), according to the SWHP (Surface Water
Hydrology Project) flow records from 1969 to 2008.
The UIB has a total catchment area of 206,000 km² as calculated in this study.
Nearly 11.5% of the total area (22,000 km²) of the UIB is covered by perennial glacial
ice including most of the largest valley glaciers, the largest area outside the polar and
Greenland regions (Hewitt, 2001; Hewitt, 2007). The Greater Karakoram Range (Fig.
2.5), situated in UIB, has an extensive formation of glaciers due to high altitudes
(Young and Hewitt, 1990), covering an area of 16,300 km² with about 13,000 km² of
cover within Pakistan (Mercer, 1975). Most of the annual precipitation in the UIB falls
in the winter and spring and originates from the west (Young and Hewitt, 1990). The
high mountains of Hindukush-Karakoram-Himalaya (HKH) in the UIB decrease the
effect of the monsoon in the catchment area, whose influence weakens
northwestward; therefore, the climatic records in the UIB are different from the
eastern Himalayas (Fowler and Archer, 2005; Young and Hewitt, 1990).
41
Fig. 2.5 Location of the Upper Indus River Basin (UIB) with main controlling storage
at Tarbela Dam.
The main active hydrological zone for the Upper Indus River Basin is situated in the
high-altitude Karakoram ranges. Archer (2003) explained that the Indus River
catchment at Partab Bridge (Fig. 2.6) has nearly 36% of its area above 5000 m in
elevation and is mainly fed by glaciers and snow melt in this area. Several authors
(Hewitt et al., 1989; Wake, 1989; Young and Hewitt, 1990) reported that 80% of the
flow of the Upper Indus River is contributed by less than 20% of its area, essentially
from the zones of heavy snowfall and glacierised basins above 3500 m in elevation.
Liniger et al. (1998) stated that some 90% of the lowland flow of the Indus River
System originates from the Hindukush, Karakoram and western Himalaya mountain
areas. Maurer et al. (2003) stated that the presence of snow in a river catchment
area strongly affects the moisture that is stored at the surface and available for future
runoff. The summer runoff is highly correlated with the summer mean temperature in
42
these high-altitude sub-catchments of the UIB, mostly covered with permanent snow
pack and glaciers (Archer, 2003). Linear regression analysis by Archer (2003)
indicates that a 1°C rise in mean summer temperature would result in a 16%
increase in summer runoff into the Hunza River.
Satellite image observation is very helpful in these remote regions of the world to
examine cryosphere dynamics. Sirguey et al. (2009) reported that satellite remote
sensing is a powerful tool to estimate the snow-covered area in remote and
inaccessible areas. MODIS snow cover products have been widely used to estimate
the snow cover area by researchers in several regions of the world. Tekeli et al.
(2005) compared the MODIS snow cover images with ground observations and found
that MODIS determined the snow cover satisfactorily even in the rough terrain river
basin located in eastern Turkey. Maurer et al. (2003) found a significant improvement
in snow cover determination using MODIS snow cover products, especially in
topographically complex portions of the Columbia River basin, as compared to the
operational NOHRSC (National Operational Hydrologic Remote Sensing Center)
snow cover images. Immerzeel et al. (2009) investigated the effects of snow cover
dynamics on the discharge of the Upper Indus River and concluded that stream flows
can be predicted with a high degree of accuracy using MODIS snow cover data in a
hydrological runoff model. Lee et al. (2005) reported that MODIS snow cover
products have sufficient accuracy for stream flow prediction using the Snowmelt
Runoff model (SRM) in the snowmelt-dominated basin of the Upper Rio Grande
basin.
43
Fig. 2.6 ASTER Global digital elevation model (GDEM) of the Upper Indus River Basin and gauging station network within the
boundary of Pakistan (Hunza River catchment location in the UIB is shown with a thin boundary line). The projection of the map is
WGS1984 UTM 43N.
44
While the rapid melting of the ice cover is reported from High Asia, the Karakoram
Range shows a more nuanced picture as highlighted by the IPCC’s 4 th synthesis
report (IPCC, 2007). Hewitt (2005) reported that central Karakoram is the largest of
those very few areas where expansion of glaciers has been noted. Some of the
largest glaciers in the Karakoram Range have undergone rapid thickening since the
mid-1990s; 13 glaciers of intermediate size (10–20 km in length) and 16 high-altitude
tributaries were observed to be advancing (Hewitt, 2005; Hewitt, 2007). These
changes were observed only in the highest central parts of the Karakoram. This is
contrary to most of the glaciers in the world reported to be shrinking for the last
several decades, including the neighbouring Greater Himalaya analysed by Berthier
et al. (2007). This contrast in glacier evolution shows a climate change pattern in
Karakoram that differs from that in the Greater Himalaya (Fowler and Archer, 2006).
Hewitt (2005, 2007) stated that a unique climate regime in this area and exceptionally
high-altitude ranges of the ice masses may be the important factors in this expansion,
contrary to other regions.
The management of Tarbela reservoir depends to a large extent on the summer

inflow contributed by the snow- and glacier-fed tributaries situated in the Karakoram
Range. The Hunza River at Dainyor Bridge is representative of the moderately high
runoff catchments in the centre of the Karakoram where a significant proportion of the
flow is derived from cryosphere melt. It nearly doubles the runoff rate along with the
Gilgit River in the overall Indus catchment at Partab Bridge station (Fig. 2.6), i.e. at its
confluence point with the Indus River. It is, therefore, important to monitor the
seasonal snow cover in snow-fed catchments for several purposes such as
hydrometeorology, flood forecasting and water resource management. This study is
the first to detail and to investigate the cryosphere dynamics based on the remotely
sensed MODIS snow cover data (MOD10A2) over a period of 10 years in the Hunza
River basin (situated in the central Karakoram region). The main objectives of this
study were to investigate:
a) The climate variations within the Hunza River basin;
b) The relevance and validity of remotely sensed MODIS snow cover data in the
catchment;
c) The hydrological behaviour of the catchment in response to climate variables
(precipitation and temperature);
45
d) The cryosphere dynamics and impact of climate change on the cryosphere in the
high-elevation mountainous Hunza River basin;
e) The annual and seasonal correlation between snow cover, mean temperatures and
stream flow in the Hunza River basin.
The study area is described in section 2 of this paper followed by the description of
the data sets and analysis of the data in section 3. The results obtained from this
study are described and discussed in section 4. Section 5 presents the conclusions
drawn from this study.
2. Study Area (Hunza River basin)

The present analysis of cryosphere dynamics was undertaken in the Hunza River
basin (basin area, 13,733 km2) (Fig. 2.6), situated in the high mountainous region of
central Karakoram, with approximately 4463 km² of catchment area at an elevation
over 5000 m. The area distribution in different elevation bands in the catchment is
shown by the hypsometric curve in Fig. 2.7.
46
Area per 500-m elevation band (%)
0 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
8000
7500
7000
6500
6000
5500
5000
Elevation (m)
4500
4000
3500
3000
2500
2000
1500
1000
Area per 500-m elevation band
500 Hypsometric curve
0
0
500
10000
10500
11000
11500
12000
12500
13000
13500
14000
1000
1500
2000
2500
3000
3500
4000
4500
5000
5500
6000
6500
7000
7500
8000
8500
9000
9500
Fig. 2.7 Hypsometric curve of the Hunza River basin and distribution of the area by layer under each 500-m elevation band.
47
Some key features of the Hunza River basin are given in table 2.1.
River flow gauging station Dainyor Bridge

Latitude 35° 56' N
Longitude 74° 23' E
Elevation of gauging station 1450 m
Drainage area 13,733 km2
Glacier-covered area 4688 km2 (Akhtar et al., 2008)
Glacier cover percentage 34%
Mean elevation (computed from 4631 m
ASTER GDEM)
Area above 5000 m 32.5 %
No. of meteorological stations 3 (Installed by WAPDA)
Khunjerab Ziarat Naltar
4730 m 3669 m 2858 m
Table 2.1 Characteristics of the study area (Hunza River basin).
The snow cover area in winter is approximately 80% and decreases to 30% in
summer. The mean total annual precipitation is 170 mm at Khunjerab (4730 m), 225
mm at Ziarat (3669 m) and 680 mm at Naltar (2858 m) according to the 10-year
record (1999–2008) of the three climatic stations in the catchment. The Hunza River
has a mean annual flow of 323 m 3/s (i.e. 742 mm of water depth equivalent) gauged
at Dainyor Bridge (Fig. 2.6), according to the 40-year (1966–2008) flow record of the
SWHP-WAPDA. However, the present precipitation records are not representative of
the runoff at the outlet. There are two main explanations for this problem.
1) The most active hydrological region for the Hunza River lies above 5000 m in
elevation where maximum snowfall and accumulation occurs, as confirmed by Young
and Hewitt (1990). All the major meteorological stations of the region are located
below this altitude range and we have no data records of the water equivalent of the
snow accumulation in this elevation zone. Hewitt (2005, 2007) reported that there is a
five- to tenfold increase in precipitation over the elevation of 5000 m, and a large
drop in temperature. Hewitt (2007) stated that about 90% of the total glaciated area
in the Karakoram Range lies between 5000 and 6000 m, where most of the
accumulation zones lie. The Karakoram Range receives maximum precipitation at
48
elevations greater than 3500 m (Hewitt, 1986). Precipitation in the form of snow
contributes the large moisture surplus for the UIB (Wake, 1989). The Batura Glacier
Investigation Group, BGIG (1976) recorded a net winter accumulation of 1030 mm
water equivalent (w.e.) at 4840 m on Batura Glacier situated in the Hunza River
basin. An annual snow accumulation ranging from 900 mm to 1900 mm w.e. in the
4900- to 5400-m elevation range was reported by Hewitt et al. (1989) and Wake
(1989) in the Biafo Glacier basin of the Karakoram Range. Winiger et al. (2005)
derived the total annual precipitation for different altitudinal zones using a combined
snow cover and ablation model. The estimated precipitation for northwest Karakoram
was more than 1700 mm year−1 above 5500 m.
2) The other factor that marks this contrast is the fact that the precipitation gauges
present at high altitudes can catch only 20–30% of the precipitation, while the rest is
distributed outside the gauges by strong winds. This factor may be the reason for the
low precipitation level recorded at the Khunjerab climate station situated just below
5000 m in elevation. According to Sevruk (1985, 1989) and Førland et al. (1996), the
greatest of the systematic errors in precipitation gauging is wind-induced error (flow
distortion), especially in the case of snowfall where losses can be on average 10–
50%, despite the many gauges equipped with wind shields.
3. Data: Sources and analysis
3.1 Topography
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER),
Global Digital Elevation Model (GDEM) was used to delineate the catchment
boundary studied. The ASTER GDEM is available for high-latitude and steep
mountainous areas not covered by SRTM3 (METI and NASA, 2009). It is based on
ASTER images (see section 3.3).
Nine downloaded ASTER GDEM tiles were mosaicked and then treated further using
the ARC Map tool (version 9.3, © ESRI) to delimit the Hunza River basin. The study
area was then extracted and superimposed by the gauging stations and other
necessary feature layers. Three different altitudinal zones were extracted from the
GDEM study area for detailed analysis of snow cover distribution. Each of the three
zones contains one climate station so we can study each zone snow cover change in
relation to its own climate. The characteristics of these zones are given in table 2.2.
49
Elevation range Mean elevation Area Area Climate
Zone
(m) (m) (%) (km2) station
A 1432–3300 2365.5 11 1541 Naltar
B 3301–4300 3800.5 25 3413 Ziarat
C 4301–7849 6075.0 64 8779 Khunjerab
Total 100% 13,733
Elevation range and mean elevation for each zone was computed from ASTER global digital
elevation model (GDEM).
Table 2.2 Characteristics of the three elevation zones extracted from the GDEM of
the Hunza River basin and their corresponding climate station.
3.2 Hydrometeorology
The database of daily flows for the Hunza River gauged at Dainyor Bridge was made
available for this analysis from 1966 to 2008, with deficient data from 2005 to 2007.
Hydrology Project of the Water and Power Development Authority (SWHP-WAPDA),
with the earliest records beginning in 1960. Daily mean flows have been published in
annual reports and have been checked and digitised in a database prepared by the
German Technical Development Agency (GTZ) acting as advisors to WAPDA.
The temperature and precipitation records at Gilgit just outside the basin have been
used by the researchers in past years for testing runoff controls because there were
no climatic stations within the basin. We used the meteorological data available, at
the Gilgit climate station for almost 50-year period and at three high-altitude stations
(Fig. 2.6 and Table 2.1) from 1999 to 2008 within the Hunza River basin established
by WAPDA, although the records for these data were not particularly long (~10
years).
The available hydrological and meteorological data were entered in the Hydraccess
database, © IRD, developed by Vauchel (2005), for easy use in the future.
Regression analysis was performed on the climate data (temperature and
precipitation) available at high-altitude stations (Naltar, Ziarat and Khunjerab) within
the Hunza basin and the Gilgit climate station (Fig. 2.6) just outside the catchment
boundary, to investigate the climate correlations between these stations.
Regression analysis was also performed between annual and seasonal (summer and
winter) stream flow data at Dainyor and climate data from the Hunza basin climate
50
stations and the Gilgit station to identify the climate station that was the most
representative of discharge at Dainyor. Moreover, this analysis was performed to
investigate the main controlling factors of runoff for the Hunza River at Dainyor
Bridge. The Pearson correlation (Rodgers and Nicewander, 1988) and Kendall rank
correlation (Kendall, 1975; Kendall and Gibbons, 1990) tests were used to evaluate
the relationship between different hydrological and climate variables of all the
gauging stations mentioned above.
3.3 Snow cover
selected to calculate the snow cover percentage on our study area. The
MODIS/Terra Snow Cover 8-Day L3 Global 500 m Grid (MOD10A2) used for this
study contains data fields for maximum snow cover extent over an 8-day repeated
period (Hall et al., 2006, updated weekly) and has a resolution of approximately 500
m covering the Hunza River basin completely. A data set of 450 processed
MOD10A2 (V005) images available from March 2000 to December 2009 was
downloaded from http://nsidc.org/cgi-bin/snowi/search.pl.
The available MODIS images, from the 2000–2009 period on an 8-day classification
basis, were mosaicked and projected with the WGS 1984 UTM ZONE 43N projection
system. The Hunza River basin area was then extracted from this mosaicked scene
to assess the snow and ice cover (cryosphere) percentage in the study area over a
10-year period. When the percentage of cloud cover exceeded 20% on a specific
date, the record was eliminated and the average snow cover on this date was
estimated by interpolating linearly between the previous and the next available cloud-
free images. The snow cover area was also calculated for the different altitudinal
zones to investigate the snow cover dynamics in these zones over a period from
2000 to 2009. A version of non-parametric Mann-Kendall trend test for the
seasonal/periodic data (Hirsch et al., 1982; Hirsch and Slack, 1984) was applied to
identify the trends in time series data; the values of this trend test are represented by
Kendall’s tau (τ) coefficient.
Higher-spatial resolution ASTER (Advanced Spaceborne Thermal Emission and
Reflection Radiometer) images were used for snow cover extent extraction to
validate MODIS snow cover products. The ASTER instrument operates in different
spectral regions that have 15-m, 30-m and 90-m spatial resolutions (Lopez, 2007;
Sarwar and Bill, 2003). Four ASTER images, nearly cloud-free, in different seasons,
51
were downloaded from http://glovis.usgs.gov/ for use in this study. The dates of these
images are 16 July 2000, 30 April 2001, 26 June 2001 and 28 February 2002. The
validation was done by comparing the images from ASTER and MODIS snow cover
products on the same dates. As an additional analysis, these satellite images were
sliced into three elevation zones (description of these zones is given in table 2.2) to
investigate the sensibility of MODIS snow cover data at different altitudes and slopes.
The Normalized Difference Snow Index (NDSI) has been widely used for effective
snow cover mapping using various sensors (Dankers and de Jong, 2004; Nagler et
al., 2008; Sirguey et al., 2008; Hall et al., 2002). The NDSI was applied to the ASTER
images to enhance the snow cover extent. The aim of this spectral index is to
enhance the snow and ice targets and distinguish them from vegetation, bare ground
and clouds. The NDSI index does not differentiate between snow and snow-covered
ice as stated by Lopez (2007), so both of them are included in the results obtained
with the NDSI.
4. Results and discussion
4.1 Climate data variation analysis
Correlation coefficient values for the trend of different climate variables (monthly
temperatures and precipitation) between the Hunza basin climate stations and the
Gilgit climate station are given in table 2.3. Pearson and Kendall rank correlation
tests (significance level, p = 5%) were applied to evaluate the relationship between
these variables. A highly significant correlation was found for maximum, minimum
and mean temperature trends between all the stations of the Hunza basin and the
mean temperatures at the neighbouring Gilgit climate station. The correlation
coefficient value was at least 0.97 in each case. Regression analysis of seasonal and
annual temperatures at three Hunza basin climate stations ranging in elevation from
2858 m to 4730 m gives lapse rates ranging from 0.48 to 0.76°C.100m−1.
As discussed before, the absolute values of precipitation are significantly
underestimated and cannot be exploited in the framework of the hydrological
balance. Nevertheless we took as a hypothesis that for the same precipitation gauge
the observations are relatively consistent and they records give us the relative trend
of precipitation from one month to other. Therefore, the correlation is carried out for
the distribution of precipitation from one month to other and not for the volume.
Correlation coefficient values for monthly precipitation show that there is a significant
correlation between the Hunza basin climate stations (Table 2.3). A maximum
52
correlation was found between Ziarat and Naltar (r = 0.37). The minimum correlation
for monthly precipitation was found between Ziarat and Khunjerab (r = 0.25). Almost
the same behaviour was shown by the monthly precipitation correlation between the
Gilgit climate station and the Hunza valley stations and a maximum significant
correlation (r = 0.62) was found between the Gilgit and Naltar climate stations. These
correlations are made between short data series available from the Hunza climate
stations and results based on this should be verified with long data series (once
available) in the future.
53
Precipitation Ziarat Naltar Gilgit
Khunjerab 0.25 (0.32) 0.33 (0.26) 0.29 (0.33)

Ziarat 0.37 (0.31) 0.38 (0.29)
Naltar - 0.62 (0.47)
Mean precipitation in the Hunza River basin 0.62 (0.45)
Temperature Ziarat Naltar Gilgit
Tmax Tmin Mean T Tmax Tmin Mean T Tavg
Khunjerab 0.99 (0.92) 0.97 (0.88) 0.99 (0.92) 0.99 0.99 (0.92) 0.99 0.98 (0.88)
(0.92) (0.92)
Ziarat - - 0.99 0.97 (0.89) 0.99 0.98 (0.87)
(0.92) (0.92)
Naltar - - - - 0.99 (0.88)
Avg. of Hunza basin climate stations mean temperature 0.98 (0.88)
Tmax = Daily maximum temperature, Tmin = Daily minimum temperature, Mean T = Daily mean temperature, T avg = Monthly mean temperature. Values outside
and within the ( ) are Pearson and Kendall rank correlation coefficient values, respectively. All the values are different from 0 with p = 0.05.
Table 2.3 Monthly climate data (precipitation and temperature) correlation coefficients between three climate stations of the Hunza
River basin (1999–2007) and Gilgit climate station (1999–2007).
54
The altitude within the Hunza River basin varies greatly, which is why the climate
records at different stations within the basin are not strongly correlated. Moreover,
the Hunza basin climate stations are situated at much higher altitudes than the Gilgit
station and therefore better represent the climate records (temperature, precipitation,
etc.) of the Hunza River basin at different varying elevations. Nearly 98% of the
Hunza River basin has an elevation greater than 2000 m and the Gilgit station
(~1460 m) does not suitably represent the high-altitude climate conditions of the
catchment, which receives much more snow at high altitudes than the Gilgit River
basin.
4.2 MODIS snow cover product validation
The snow cover area on cloud-free ASTER images was compared with MODIS snow
cover images on the same dates (seasonally significant) to validate the MODIS snow
cover product as given in table 2.4. The validation was done for the total as well as
for zone-wise snow cover area of the images. The results obtained for the validation
of MODIS images with ASTER images (which have higher-spatial resolution than
MODIS) suggests that MODIS snow products are reliable in estimating the snow
cover area in the Hunza River basin. The zone-wise comparison suggests that the
low-elevation, high-relief areas have a larger mismatch than the higher-elevation,
lower relief areas.
The NDSI was applied to ASTER images and the index (threshold) values greater
than or equal to 0.82 were considered as snow and ice cover. The index value was
calibrated by visual and pixel inspection of two ASTER and MODIS images. In
comparison, the index value of 0.9 was used to calculate the snow cover extent from
ASTER images over the Pamir region (Afghanistan) by Haritashya et al. (2009).
Comparison of MODIS (MOD10A2) image and ASTER images taken on 28 February
2002 (Fig. 2.8) showed nearly the same percentage of snow cover area. MOD10A2
snow products are 8-day averages and the time span of this particular MODIS snow
product (compared with ASTER in Fig. 2.8) is from 25th February 2002 to 4th March
2002.
55
Zone-wise snow cover Total snow cover
Date of Elevation
area (%) area (%)
images zone
MODIS ASTER MODIS ASTER
A 9 1
16/07/2000 B 6 3 34 33
C 57 59
A 5 4
30/04/2001 B 32 24 65 63
C 88 88
A 0 2
26/06/2001 B 10 11 30 32
C 62 64
A 29 41
28/02/2002 B 75 74 84 85
C 97 96
Table 2.4 Comparison of zone-wise and total snow cover area on MODIS and
ASTER images for validation of MODIS images. (The main characteristics of the
elevational zones are given in table 2.2).
56
Fig. 2.8 (a) MODIS snow cover image and (b) ASTER snow cover image for the Hunza River basin. MODIS (MOD10A2) snow
product is 8-day average data and the time span of this particular image presented here is from 25 th February 2002 to 04th March
2002.
57
4.3 Hydrological behaviour of the Hunza River basin
The Pearson and Kendall rank correlation coefficient values (p = 5%) obtained for the
correlation between annual and seasonal stream flow at Dainyor and the climate data
from the Hunza basin climate stations and the Gilgit station are given in table 2.5. No
significant correlation was found between annual precipitation of all the climate
stations analysed and the stream flow at Dainyor Bridge. The maximum correlation
coefficient value was found between annual precipitation at Khunjerab and runoff at
Dainyor (r = 0.26). Hewitt et al. (1989) and Wake (1989) stated that the annual
precipitation for Hunza River basin above 5000 m in elevation is on the order of
1800–2000 mm. This elevation and precipitation are instrumental in creating large
glaciers and perennial snow, which slowly moves down to the ablation zone and
contributes to the river flow in summer. An analysis of seasonal runoff and
precipitation showed no significant correlation between summer precipitation and
runoff at Dainyor for almost all the climate stations, as given in table 2.5. A significant
inverse correlation was found between summer precipitation at Gilgit and runoff at
Dainyor (r = −0.22). The reason for this negative correlation is associated to more
precipitation and increased cloudiness which leads to less incoming short wave
radiation and higher albedos due to fresh snowfall and therefore less energy
available for snowmelt resulting in lower discharges (Archer, 2003). This may also be
attributable to the minimum effect of the monsoon regime in the Hunza basin
because of the Karakoram high mountainous barrier.
58
Climate data Stream flow in Hunza River at Dainyor Bridge
Annual correlation Summer correlation Winter and spring
(January to (April to September) correlation (October
December) to March)
a) Hunza climate
stations
Precipitation
Khunjerab 0.26 (0.17) 0.17 (0.11) 0.10 (0.00)
Ziarat 0.02 (0.05) −0.01 (0.01) 0.02 (0.02)
Naltar 0.03 (0.02) −0.04 (−0.01) −0.01 (−0.05)
Avg. of Hunza basin 0.08 (0.08) - (0.04) 0.01 (−0.04)
climate stations
Winter precipitation Summer stream flow
(Oct-Mar) (April – September)
Khunjerab - 0.12 (0.06) -
Ziarat - 0.08 (0.06) -
Naltar - 0.07 (0.00) -
Avg. of Hunza basin 0.10 (0.03)
climate stations
Daily mean
temperature
Khunjerab 0.79 (0.69) 0.80 (0.66) 0.64 (0.36)
Ziarat 0.79 (0.67) 0.81 (0.66) 0.61 (0.32)
Naltar 0.77 (0.66) 0.73 (0.56) 0.64 (0.36)
Avg. of Hunza basin 0.79 (0.69) 0.81 (0.66) 0.65 (0.36)
climate stations
b) Gilgit climate
station
Precipitation 0.04 (0.1) −0.22 (−0.1) −0.01(−0.18)
Winter precipitation - −0.04 (−0.06) -
(Oct-Mar)
Daily mean
0.79 (0.68) 0.89 (0.81) 0.70 (0.26)
temperature
Values outside and within the ( ) are Pearson and Kendall rank correlation coefficient values, respectively.
(Bold figures: values are different from 0 with p = 0.05).
Table 2.5 Annual and seasonal correlation coefficients between stream flow data at
Dainyor (1999–2007) and a) daily climate data (precipitation and temperature) of Hunza
basin climate stations (1999–2007), b) monthly climate data (precipitation and
temperature) of Gilgit station (1999–2007).
59
Analysis of correlation coefficient values suggests that there is a significant positive
correlation between the annual and seasonal temperature records and discharge at
Dainyor (Table 2.5), which indicates that river flows are driven by snowmelt (resulting
by the increase in mean temperature) in the catchment. This fact is explained by
Archer (2003) that the peak seasonal (summer) and daily flows in the Karakoram
region are caused by the availability of heat energy which melts the snow pack and
the water stored in the form of snow and ice. The maximum correlation coefficient
value (r = 0.79) was found for annual correlation between runoff at Dainyor and mean
temperatures at Khunjerab, Ziarat and Gilgit. For summer temperature and runoff, the
Gilgit climate station (situated outside the basin) showed a strongly significant
correlation (r = 0.89). Khunjerab (r = 0.81) and Ziarat (r = 0.81) (situated within the
basin) also had significant positive correlations with runoff at Dainyor. The correlation
coefficient between winter and spring temperatures and runoff was also found to be
significantly positive and the coefficient value was not less than 0.61 for any station.
This result can also be explained by the fact that snow and ice melt are driven by
short wave (SW) and long wave (LW) energy balance at the surface in early spring
and summer (Sicart et al., 2005; Sicart et al., 2006) and a higher air temperature is
only one of the outcomes of this balance. Moreover, the climate variables
(temperature and precipitation) observed at all the climate stations in the Hunza River
basin are almost as well correlated with Hunza river runoff as the Gilgit climate
station variables (Table 2.5).
The regression analysis results presented in table 2.5 suggest that zone C
(corresponding to the Khunjerab climate station) is the most active hydrological zone
for Hunza River flows. Both the correlation coefficient values for precipitation and
mean temperature at Khunjerab and runoff at Dainyor are higher than or equal to
other catchment stations (Naltar and Ziarat). This result can also be observed in Fig.
2.9, where zone B (corresponding to the Ziarat climate station) and zone C
(corresponding to the Khunjerab climate station) present a significant contribution to
river flow in the form of snow melt in summer (April to September). Zone C has the
largest proportion (64%) of the Hunza basin area as given in table 2.2 and hence a
small drop in the percentage of snow cover in zone C in summer will have a
significant impact on catchment runoff.
60
100
90
80
(Data source: MODIS snow product)
Snow cover area, SCA (%)
70
60
50
40
30
20
10
0
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Trend analysis: Data time period (Daily)
Zone A
Slope x(day),  
Zone A
Max. snow period trend,  , Min. snow period trend,  Zone B
Zone B Zone C
Zone C
Slope  x (day),  Slope  x(day), 
Max. snow period trend, , Min. snow period trend,   Max. snow period trend,  , Min. snow period trend,  
Fig. 2.9 Snow cover distribution in three different altitudinal zones of the Hunza River basin. Increasing trend (though not significant)
can be noted in zone C. Seasonal Mann Kendall’s trend analysis is also presented for maximum (November-February) and
minimum (June-September) snow cover periods. (Kendall’s tau (τ) coefficient values presented in bold figures indicate an
increasing trend with p<0.05).
61
4.4 Snow cover dynamics in the Hunza River basin
MODIS image analysis for snow cover dynamics over a 10-year period suggests that
the cryosphere area in the Hunza River basin has an increasing tendency (though
the trend is not statistically significant), as shown in Fig. 2.9. Analysis of Kendall’s tau
(τ) coefficient value indicates an upward tendency of the snow cover area in zone C
over an elevation of 4300 m. This tendency is stronger in the maximum snow periods
from November to February in all the altitudinal zones, notably in zone C (τ = 0.25)
(Fig. 2.9). In the minimum snow cover periods from June to September, a slight
increasing trend is noted in zone C (τ = 0.02) (Fig. 2.9); this expansion may result
from the constancy in the mean temperatures and an increasing trend in the
catchment’s annual precipitation. Fowler and Archer (2006) found that summer
temperatures (July to September) which are the key for glacial melt, have been falling
at many valley stations in the Karakoram in the period 1961 to 2000. Hussain et al.
(2005) indicated similar temperature falls in the monsoon and pre-monsoon period for
the high mountain regions. In addition, a significant increase in the precipitation is
reported in the Upper Indus Basin in both winter and summer during the period
1961‒1999 (Archer and Fowler, 2004).
The increasing trends of precipitation continue to feed the high altitudes, particularly
zone C, and results in the form of expanding snow cover in the area. A greater
proportion of the catchment area (i.e. almost 32.5% of the catchment area) is higher
than 5000 m in elevation, as shown in Fig. 2.7 and presented in table 2.1 and nearly
the same percentage of area (30%–40%) is snow covered all year, as illustrated by
the snow cover distribution in Fig. 2.10. This is the area where the mean temperature
remains below freezing throughout the year and the snow cover is expanding as a
result of the increasing precipitation trend.
62
Snow cover area (%)
images.
20
30
40
50
60
70
80
90
05-Jan
13-Jan
21-Jan
29-Jan
06-Feb
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
14-Feb
22-Feb
02-Mar
10-Mar
18-Mar
26-Mar
03-Apr
11-Apr
19-Apr
27-Apr
05-May
Average SCA over 10-years period

13-May
21-May
29-May
06-Jun
14-Jun
22-Jun
30-Jun
08-Jul
16-Jul
24-Jul
01-Aug
09-Aug
17-Aug
25-Aug
Time period (8 days classification)
02-Sep
10-Sep
18-Sep
26-Sep
04-Oct
12-Oct
20-Oct
28-Oct
05-Nov
13-Nov
21-Nov
29-Nov
07-Dec
15-Dec
23-Dec
Fig. 2.10 Percentage snow cover area (SCA) in the Hunza River basin calculated by analysing 450 MODIS (MOD10A2)
31-Dec
63
This snow cover area expansion may be associated to the glacier advancing
(especially surging) as reported by different researchers as follows. Hewitt (2005,
2007) found that in the late 1990s there was widespread evidence of glacier
advancing mainly in high level glaciers in the central Karakoram. Analysing the
response of Himalayan glaciers to the climate change, Scherler et al. (2011) has
found that more than 50% of the glaciers observed in the Karakoram region
(westerlies-influenced) are advancing or stable in contrast to the retreating glaciers of
eastern Himalaya (monsoon-influenced) and Tibetan Plateau. Research on glacial
mass balance in the central Karakoram region by Hewitt (1998, 2005) on the basis of
extensive field experience over several decades also contrast with reports elsewhere
of shrinking glaciers. The longer records of climate data at high-altitude weather
stations and satellite images may help researchers to understand the dynamics of
snow cover in the area. In the future, automatic weather stations or snow pillows
should be installed over 5000 m in elevation to assess the precipitation patterns in
the most active hydrological zone of the Karakoram Range (WAPDA, oral
communication, 2009).
The current hypothesis for this phenomenon of increasing cryosphere area is an
increase in winter precipitation mainly due to westerly circulation, as explained by
Hewitt (2005) and Archer and Fowler (2004). The impact of global warming is not
noticeable because of the high altitudes in the Hunza River basin. Even if global
warming is real, the mean temperature remains negative or very low over a large part
of the basin area. The increasing precipitation in the form of snowfall continues to
feed the snow cover and the accumulation zones of glaciers in zones B and C.
64
4.5 Correlation between snow cover change, climate variables and stream flow in
the Hunza River basin
Snow cover in the area is at a maximum 70–80% in the winter and spring seasons
(December to April) and at a minimum 30–40% in the summer (July to September),
as shown in Fig. 2.10. This change in the snow cover has a significant inverse
correlation with the mean temperatures and discharge in the Hunza River basin, as
shown in Fig. 2.11 and table 2.6. The relationship between standardised values of
four variables (precipitation, snow cover, mean temperature and discharge) is
presented in Fig. 2.11. The values were standardised using the normal deviate
formula, i.e. the distance of one data point from the mean, divided by the standard
deviation of the distribution. This indicates a continuous depletion in the snow cover
in summer as the average temperature in the area increases, resulting in increasing
discharge. In winter, as the temperature decreases, the snow cover starts to increase
and hence discharge decreases. The monthly difference in the Hunza River
discharge at Dainyor is inversely correlated (Pearson’s correlation coefficient =
−0.73; Kendall’s rank correlation = −0.56) with the monthly difference in the snow
cover, as shown in Fig. 2.12, which indicates that the change in Hunza River
discharge is dependent on the snow cover change in the area. It is likely that the
large snow cover difference has a large impact on discharges because a larger snow
cover area may also result in higher snow depths (at least in low-slope areas).
65
4
Pearson's correlation:
Tavg vs Q  0.85, Q vs SCA   0.89, Tavg vs SCA   0.80
Kendall's rank correlation:
Tavg vs Q  0.77, Q vs SCA   0.66, Tavg vs SCA   0.58
3
2
Standardised value
-1
-2 Precipitation (Khunjerab)
Discharge (Q)
Mean temperature (Tavg)
Snow cover area (SCA)
-3
02
03
04
05
06
07
08
09
10
11
12
02
03
04
05
06
07
08
09
10
11
12
02
03
04
05
06
07
08
09
10
11
12
02
03
04
05
06
07
08
09
10
11
12
02
03
04
05
06
07
08
09
10
11
12
2000
2001
2002
2003
2004
Fig. 2.11 Correlation between standardised values of precipitation (Khunjerab), snow cover, average temperature and discharge (at
Dainyor) in the Hunza River basin (on a monthly basis). All the correlation values are significant with a significance level, p<0.05.
66
800
Pearson correlation coefficient = r =  0.73 (P<0.05)
Spearman correlation coefficient = r =  0.75 (P<0.05)
Kendall correlation coefficient = r =  0.56 (P<0.05)
600
Linear regression line
95% confidence interval
400
Monthly difference in discharge (m /s)
3
y = (-1576) x - 1.660
200
-200
-400
-600
-800
-30 -20 -10 0 10 20 30
Monthly difference in snow cover (%)
Fig. 2.12 Correlation between monthly difference in snow cover and monthly difference in discharge at Dainyor in the Hunza River
basin. All the correlation values are significant with a significance level, p<0.05.
67
Pearson and Kendall’s rank correlation coefficient values (p = 5%) were used to
evaluate the correlation between the snow cover change and a) climate variables
(the mean temperature, precipitation) and b) stream flow in the Hunza River, as given
in table 2.6. There are more variables which influence the river runoff (e.g.
evaporation, sublimation, melting permafrost, different ablation procedures on debris
covered or debris free glaciers) but these have not been taken into consideration in
this study. These variables become important when applying a theoretical hydro-
climatic model of mountain systems. No significant correlation was found between
the snow cover change and precipitation in the Hunza basin. This may be due to the
fact that the precipitation gauges are mostly under sampled in the high-altitude
catchments due to a number of errors, as explained by Førland et al. (1996) and
therefore cannot efficiently represent the snow cover change at high altitudes. The
maximum inverse correlation was found between the snow cover change in the
Hunza basin and the mean temperatures observed at Khunjerab and Ziarat. A high
significant inverse correlation was found between snow cover change and summer
and annual discharges (Table 2.6). This correlation indicates that the stream flow in
the Hunza River strongly depends on the snow cover change and the mean
temperatures in the catchment. This is in accordance with the result found by the
Bookhagen and Burbank (2010) that the western Himalayan catchments (e.g. Indus
River catchment) are fed by the snowmelt up to 50% as a fraction of the total annual
discharge.
68
Snow cover change in the Hunza River basin
Annual Summer correlation Winter and spring
correlation (April to correlation
Climate data (January to September) (October to March)
December)
a) Hunza climate
stations
Precipitation
Khunjerab −0.26 (−0.18) −0.17 (−0.11) −0.09 (−0.09)
Ziarat −0.02 (−0.05) −0.01 (−0.03) 0.04 (0.01)
Naltar −0.01 (−0.02) 0.07 (0.02) 0.01 (0.00)
Avg. of Hunza basin −0.07 (−0.08) 0.01 (−0.04) 0.01 (−0.01)
climate stations
Daily mean
temperature
Khunjerab −0.80 (−0.56) −0.78 (−0.56) −0.42 (−0.21)
Ziarat −0.80 (−0.57) −0.80 (−0.57) −0.43 (−0.25)
Naltar −0.78 (−0.54) −0.72 (−0.48) −0.40 (−0.21)
Avg. of Hunza basin −0.80 (−0.56) −0.80 (−0.56) −0.43 (−0.23)
climate stations
b) Stream flow −0.88 (−0.65) −0.81 (−0.68) −0.64 (−0.26)

Values outside and within the ( ) are Pearson and Kendall rank correlation coefficient values,
respectively. (Bold figures: values are different from 0 with p = 0.05).
Table 2.6 Annual and seasonal correlation coefficients between snow cover change,
Hunza River basin (2000–2008), and a) daily climate data (precipitation and
temperature) (2000–2008), b) daily stream flow data at Dainyor bridge (2000–2008).
The analysis of 25 years of flow records (1980–2004) for the Hunza River shows that
the annual flow decreases with time, as shown in Fig. 2.13. In contrast to the
precipitation data of Hunza River basin, the discharge data at Dainyor Bridge is
available over a long period and the quality of data is quite reliable. With shorter time-
series, (Archer and Fowler, 2004) associated the decreasing river discharges of the
high-altitude central Karakoram catchments with the long-term storage of additional
ice.
69
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
1200 0
50
1100
100
1000
150
Precipitation at Gilgit (mm/year)

900
Runoff at Dainyor (mm/year)
200
800
250
700 300
350
600
400
500
Precipitation (mm/year) 450
Runoff (mm/year)
400 Linear regression line
95% confidence interval 500
300 y = (3.3616) x  761.71

550
200 600
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Time period (Annual)
Fig. 2.13 Annual flow trend in the Hunza River at Dainyor Bridge and total annual precipitation at Gilgit over 25 years (1980–2004).
70
5. Conclusion
The Hunza River basin is one of the main tributaries of the Indus river irrigation
system that contributes nearly one-fifth of the Upper Indus flow at Partab Bridge.
Situated in the high-elevation zone of central Karakoram, the Hunza river flow
depends mainly on the accumulation of snow at high altitudes and energy input
indirectly related to temperature. This suggests that meteorological data collection
methods should be improved at the high-altitude stations present within the basin
area to study the snow cover and runoff controlling factors more effectively. The
following conclusions can be drawn by analysing the results obtained in this study:
 The climate stations present within the Hunza River catchment area may be used to
study the behaviour of stream flow at Dainyor Bridge in the future.
 MOD10A2 remote sensing cryosphere product is a very valuable product and will
greatly enhance the hydrological predictions in the remote areas. In addition, MODIS
MOD10A2 product is free of charge and easy to treat and therefore is a good option
to assess the percentage cryosphere areas in the Hunza and other remote sub-
catchments of UIB over the long term. However, there are some errors and
uncertainties associated to this product; for example, it has a larger mismatch with
higher-spatial resolution ASTER images on low elevation regions with higher relief.
 The stream flow in the Hunza River is influenced mainly by winter precipitation, but
also by the mean summer and winter temperatures of its catchment. Moreover, the
most active hydrological region of the Hunza River basin seems located above an
elevation of 4500 m in zone C.
 The 10-year analysis of the remotely sensed cryosphere data and a declining
discharge trend in the Hunza River observed in this study, in combination with the
research of Archer and Fowler (2004), Hewitt (2005) and Fowler and Archer (2006)
who show an increasing winter precipitation trend and a decreasing trend in summer
mean temperatures, suggest that the Hunza River basin is a region undergoing a
slight expansion in the cryosphere area, especially at high elevations.
 The study of annual and seasonal snow cover suggests that the snow cover change
has a highly inverse correlation with the catchment’s summer mean temperatures,
which results in a large amount of stream flow. Therefore, it is very important to
estimate the winter snow accumulations (November to February) at high altitudes to
forecast the peak summer flow (May to September) and ultimately to improve the
management of the Tarbela reservoir.
71
A detailed study is needed on the other high-altitude catchments of the Upper Indus
region (e.g. Shigar and Shyok) that contribute a substantial amount of flow to the
Indus River and are mainly snow- and glacier-fed. The combination of satellite snow
cover data and ground data (e.g. snow-pits) or information based on snow cover-
runoff models is an appropriate approach to overcome the poor gauging of
precipitation data at these high altitudes. Snow depletion curves calculated in this
study for each altitudinal zone of the Hunza River basin can further be used (after a
critical evaluation), as input to the hydrological models with a snow component, e.g.
the Snowmelt Runoff Model developed by Martinec (1975), to simulate and forecast
the stream flow at Dainyor Bridge as well as to study the climate change impact on
snow cover and stream flow dynamics in the future. Ultimately, this study will help to
improve integrated water resource management in the Tarbela reservoir.
Acknowledgement
Adnan Ahmad Tahir was financially supported by the Higher Education Commission
of Pakistan within the framework of a France-Pakistan collaboration program for
overseas studies. This financial support is gratefully acknowledged and appreciated.
The authors extend their thanks to the Water and Power Development Authority
(WAPDA) and the Pakistan Meteorological Department (PMD) for contributing their
hydrological and meteorological data, respectively. The authors also wish to thank
NASA and Japan's Ministry of the Economy, Trade and Industry (METI) for providing
ASTER GDEM and to the Global Land Ice Measurements from Space (GLIMS)
project for providing four ASTER images for glacier and snow cover estimation. The
authors also acknowledge the anonymous reviewers for their valuable, rigorous and
scrupulous comments, which considerably improved the content of this manuscript.
72
2.3.1 Some additional results of climate data analysis for Hunza River basin
Monthly precipitation variation (magnitude) comparison:

An analysis of monthly precipitation (magnitude) variation between the Hunza basin
climate stations and the Gilgit climate station is presented in Fig. 2.14. These stations
are situated in an elevation range of 1460‒4700 m. The highest elevation climate
station is Khunjerab (4730 m) following by the middle altitude Ziarat (3669 m) and
Naltar (2858 m) climate stations with the Gilgit (1460 m) valley climate station
situated at the lowest elevation of these four stations. The quantity of total annual
precipitation received at all these climate stations varied significantly due to the effect
of elevation. The mean total annual precipitation is 170 mm at Khunjerab, 225 mm at
Ziarat and 680 mm at Naltar according to the 10-years data record (1999–2008) of
WAPDA on the three climate stations within the catchment. The mean total annual
precipitation is 132 mm at Gilgit according to the 50-years data record of PMD.
Precipitation magnitude at Gilgit is well below than the Hunza basin climate stations
(Fig. 2.14).
However, the present precipitation records do not represent the Hunza River basin
runoff 323 m3/s (i.e. 742 mm of water depth equivalent) gauged at Dainyor Bridge
(Fig. 2.2). Two main explanations for this problem are detailed earlier in section 2.3 of
this chapter. The precipitation magnitude gauged at higher altitudes seems well
under the average annual precipitation values (1500‒2000 mm) estimated by some
researchers (Hewitt, 1986; 2005; 2007, Winiger et al., 2005) during the field
experience or derived by modeling techniques. This difference in precipitation is due
to the well known errors of precipitation gauges at high altitudes as explained by
Sevruk (1985, 1989) and Førland et al. (1996). The precipitation gauges present at
high altitudes can catch only 20–30% of the precipitation, while the rest is distributed
outside the gauges by strong winds. This factor may be the reason for the low
precipitation values recorded at the Khunjerab and Ziarat climate stations situated
above an elevation of 3600 m.
73
Precipitation (mm) Precipitation (mm) Precipitation (mm) Precipitation (mm)
0
50
100
150
200
0
50
100
150
200
0
50
100
150
200
0
50
100
150
200
1999 1999 1999 1999

04 04 04 04
07 07 07 07
10 10 10 10
2000 2000 2000 2000
04 04 04 04
07 07 07 07
10 10 10 10
and Gilgit climate station.

2001 2001 2001 2001
04 04 04 04
07 07 07 07
10 10 10 10
2002 2002 2002 2002
04 04 04 04
07 07 07 07
10 10 10 10
2003 2003 2003 2003
04 04 04 04
07 07 07 07
10 10 10 10
2004 2004 2004 2004

04 04 04 04
07 07 07
07
10
10 10

10 Time period (Monthly)

2005 2005 2005 2005
04 04 04
04
07 07 07
07
10 10 10
10
2006 2006 2006 2006
04 04 04
04
07 07 07
07
10 10 10
10
2007 2007 2007
2007 04
04 04 04
07
07 07 07
Ziarat
Naltar
10
Gilgit
Khunjerab
10 10 10
2008 2008 2008
2008
Fig. 2.14 Monthly variations of precipitation at the Hunza River basin climate stations
74
Seasonal and annual correlation of precipitation and the stream flow
(additional comments on table 2.5 in the paper):
A significant positive correlation was found between winter and spring precipitation
variation, for almost all the climate stations of the Hunza basin and summer stream
flow at Dainyor (Table 2.5). This correlation was expected because most of the
precipitation in this period falls in the form of snow and accumulates at high altitudes
in the Hunza River basin. This snow contributes to the runoff on melting in summer
when the temperature increases in the valley. The highest correlation was found
between winter precipitation at Khunjerab and summer flow at Dainyor. This result is
supported by the fact that almost 60% area of the Hunza River basin has an altitude
greater than 4500 m and only the Khunjerab climate station is situated above this
altitude (Table 2.2) to represent the climate of this altitudinal zone. Young and Hewitt
(1990) also confirmed that the most active hydrological region for the Hunza River
lies above 5000 m in elevation where maximum snowfall and accumulation occurs. In
addition, the highest correlation value describing this zone can also be explained by
the fact that almost all the winter and spring precipitation received above an elevation
of 4500 m is in the form of snow and melts only in summer when the mean
temperatures rise above 0°C in this zone represented by the Khunjerab climate
station.
Trend analysis of short time period meteorological data (Hunza River basin)
and its comparison to a long time period data trend of Gilgit climate station:
An analysis of mean monthly temperatures (trend and magnitude) and average
annual precipitation (trend and magnitude) variation between the Hunza basin
climate stations and the Gilgit climate station is presented in Fig. 2.15 and Fig. 2.16,
respectively. A version of non-parametric Mann-Kendall trend test for the
seasonal/periodic data (Hirsch et al., 1982; Hirsch and Slack, 1984) was applied to
identify the trends in time series data; the values of this trend test are represented by
Kendall’s tau (τ) coefficient. This coefficient value varies between +1 and −1. A
positive coefficient value denotes the increasing trend in time series data and versa
negative coefficient value the contrary. A coefficient value of 0 denotes no trend in
time series data. These trend analyses are based on a short series of data available
from the catchment, therefore, the long data series of mean temperatures
(1966‒2007) and total annual precipitation (1951‒2007) of the neighbouring Gilgit
75
climate station (Fig. 2.15 and Fig. 2.16, respectively) is analysed to verify these
trends based on short data.
a) Trend analysis of mean temperature:
The magnitude of mean temperature observed at the Hunza basin climate stations is
well below than that at the Gilgit climate station, as shown in Fig. 2.15. An analysis of
the temperature variations (Fig. 2.15) from 1999 to 2008 using Kendall’s tau (τ)
coefficient value indicates no significant trend at all the climate stations of Hunza
basin. The mean temperatures seem almost constant over last decade (1999‒2008)
in the Hunza River basin. The trend analysis is based on short data series so it is
verified with the long data series (1966‒2007) available at neighboring Gilgit climate
station which has mean temperature magnitude more than Hunza basin but still the
mean temperature trend may give a representation of Hunza basin (Fig. 2.15). The
Gilgit station mean temperature data was analysed and the trends were found almost
similar to that of the Hunza basin climate stations.
Hunza River runoff is controlled by temperature seasonality and mainly fed by
glaciers and snow that melt in relation to increasing temperatures in summer as
explained in section 2.3. A seasonal trend analysis on the summer mean
temperatures of Karakoram region climate stations (notably Gilgit) was carried out by
Fowler and Archer (2006) who found that summer temperatures (July to September)
which are the key for glacial melt, have been falling at many valley stations in the
Karakoram in the period 1961 to 2000. Hussain et al. (2005) indicated similar
temperature falls in the monsoon and pre-monsoon period for the high mountain
regions.
76
(a) 20
Khunjerab
Mean temperature (°C) 15 Naltar
Ziarat
10
Trend analysis:
5 Khunjerab:
Kendall's tau coefficient = 0.02
0
Slope of linear trend line = 0.0004
-5 Naltar:
-10
-15
Ziarat:
-20 Kendall's tau coefficient = 0.02
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
(b)
30
Mean temperature (°C)
25
20
15
10
0
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Time period (Monthly) Gilgit climate station Trend analysis (Gilgit)
Linear regression line Kendall's tau coefficient = 0.008
95% confidence interval Slope of linear trend line = 0.0002
Fig. 2.15 Mean monthly temperature variation, (a) the Hunza River basin climate stations over 10-year period (1999‒2008) and (b)
Gilgit climate station over a period of 41-year (1966‒2007). Kendall’s tau (τ) coefficient values are presented for the trend analysis
of linear regression line with p<0.05.
77
b) Trend analysis of average annual precipitation:
Analysis of total annual precipitation (Fig. 2.16) by Kendall’s tau (τ) coefficient value
at all three climate stations of the Hunza basin indicates a slight increasing trend at
Ziarat (corresponding to zone B, table 2.2) and Khunjerab (corresponding to zone C,
table 2.2), over the period from 1999 to 2007. These trend analyses are based on a
short series of data available from the catchment and for a simple justification of
these previous trend results, the long data series of total annual precipitation
(1951‒2007) of the neighbouring Gilgit climate station (Fig. 2.16) was analysed and
the trends were found similar to that of the Hunza basin climate stations. Archer and
Fowler (2004) also reported a significant increase in the precipitation in the Upper
Indus Basin in both winter and summer during the period 1961‒1999.
The trend analysis comparison of Hunza basin and Gilgit climate variables and the
high-altitude variation of the Hunza River basin suggest that the data available at
Hunza basin climate stations may be used for current and future studies of the Hunza
river catchment. The results obtained by the trend analysis of mean temperature and
total annual precipitation of the Hunza basin also strengthen the conclusion of
section 2.3 that the Hunza basin is going under slight expansion of cryosphere area
possibly due to the stable or decreasing summer temperature and increasing winter
and spring precipitation. Whereas, the declining river flow at Dainyor (Fig. 2.13) may
be the result of a slight decreasing trend in summer mean temperature which
normally generates maximum of annual flow during summer months.
78
(a) 800
Khunjerab
Precipitation (mm) 700 Naltar
600 Ziarat
Trend analysis:
500
Khunjerab :
400 Kendall's tau coefficient = 0.06
Slope of linear trend line = 0.008
300
Naltar:
100
Ziarat:
1999 2000 2001 2002 2003 2004 2005 2006 2007 Slope of linear trend line = 0.02
(b)
250
225
Precipitation (mm)
200
175 Gilgit
125
100 Trend analysis:
75 Gilgit:
50 Slope of linear trend line = 0.34
25
1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Fig. 2.16 Total annual precipitation trend, (a) over 9-year period (1999‒2007) in the Hunza River basin climate stations and (b) over
57-year period (1951‒2007) on Gilgit climate station. Kendall’s tau (τ) coefficient values are presented for the trend analysis of
linear regression line with p<0.05.
79
2.4 Snow cover dynamics and hydrological regime of the Astore River basin
(western Himalaya)
2.4.1 Characteristics of study area
The Astore River basin (drainage area, 3,990 km 2) (Fig. 2.17), is situated in the high-
altitude extreme western Himalaya region, with a mean catchment elevation of 4285
m. The drainage area and the mean catchment elevation are estimated from the
ASTER GDEM (described previously in section 2.2). Three climate stations with
precipitation gauges are installed at different altitudes within the catchment and one
station just outside the catchment boundary (Fig. 2.17). The Astore climate station is
installed and managed by the PMD and have a quite long data series of more than
50 years (1954 to 2007). The Rama and Rattu climate stations are installed at the
high altitudes by WAPDA and the data record is available from 1995 to 2008 for
these stations. The mean total annual precipitation is 794 mm at Rama (3000 m), 723
mm at Rattu (2570 m) and 501 mm at Astore (2168 m) according to the available
data records of the four climate stations in the catchment. The Astore River has a
mean annual flow of 137 m3/s (i.e., 1084 mm of water depth equivalent) gauged at
Doyian hydrometric station (1583 m) (Fig. 2.17), according to the SWHP-WAPDA 33-
year (1974–2007) flow record.
The present precipitation records are not representative of the runoff at the outlet
because of the lack of climate data records at altitudes above than 4000 m in this
basin and the well-known gauging errors as explained previously (section 2.3). The
Astore river flow in summer (June-September) is influenced significantly by both
summer monsoon rainfall and the snow and glacier melt. Approximately 154 km² of
catchment area is at an elevation above 5000 m (as calculated from the GDEM
shown in Fig. 2.17) and almost the same area (7%) is glaciated as calculated from
the GLIMS shape file of the Upper Indus Basin. The hypsometric curve of the Astore
River basin is presented in the chapter 3 (Fig. 3.22).
80
Fig. 2.17 ASTER Global digital elevation model (GDEM) of the Astore River basin.
81
2.4.2 Data sets and treatment
Remote sensing satellite data (ASTER and MODIS snow cover data) and ground
observation data (stream flow and climate data) sources and their treatment are
explained in detail in sections 2.2 and 2.3.
2.4.3 Results and discussion

Snow cover change in Astore River basin:
The snow cover area (SCA) for the Astore River basin is estimated from the MODIS
snow cover images (MOD10A2) over a period of 10 years (2000‒2009). The average
SCA (magnitude) varies from approximately 95% in winter to 7% in summer (Fig.
2.18). The monthly snow cover (magnitude) variation in the Astore River basin for the
year 2009 is shown in Fig. 2.19. Snow accumulation period starts in October and the
maximum snow cover reaches in a range of 90‒95% in January (Figs. 2.18 & 2.19).
Snow melt period starts in early April and the minimum snow cover is observed
during August and September when the snow cover area drops in a range of 5‒10%
(Figs. 2.18 & 2.19). A large drop from 95% to 7% may be attributed to the small
catchment area and a mean elevation lower than Hunza River basin. Astore River
basin has only 4‒5% of its area above 5000 m elevation, whereas, this percentage
reaches to 33% in the Hunza River basin. An increase in mean temperature brings
the smaller area with lower altitude out of snow cover more rapidly than a bigger area
with higher altitude. Similarly, a snow fall can cover the smaller area more rapidly
than a bigger area depending on the topographic characteristics (slope, vegetation
and ground surface temperature etc.) of the catchment. The year 2009 represents
the maximum average snow cover and the year 2007 represents the minimum
average snow cover over the last decade.
82
(Data source: MODIS (MOD10A2) snow product)
Snow cover area, SCA ()
0
10
20
30
40
50
60
70
80
90
100
05-Jan
13-Jan
21-Jan
29-Jan
06-Feb
14-Feb
22-Feb
02-Mar
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
10-Mar
18-Mar
26-Mar
03-Apr
11-Apr
19-Apr
27-Apr
05-May
13-May
21-May

29-May
06-Jun Average SCA over 10 years (2000-2009)
14-Jun
22-Jun
30-Jun
08-Jul
16-Jul
24-Jul
01-Aug
09-Aug
17-Aug
25-Aug
02-Sep
10-Sep
18-Sep
26-Sep
04-Oct
12-Oct
20-Oct
28-Oct
05-Nov
13-Nov
21-Nov
29-Nov
07-Dec
15-Dec
Fig. 2.18 Percentage snow cover area (SCA) in the Astore River basin calculated by analysing 450 MODIS satellite images.
23-Dec
31-Dec
83
Fig. 2.19 MODIS (MOD10A2) satellite images presenting the average snow cover area for each month in the Astore River basin
over the year 2009.
84
Trend analysis of the snow cover variation in Astore River basin:
Snow cover dynamics/trend analysis over a 10-year period in the Astore River basin
is presented in Fig. 2.20. It suggests that the cryosphere area in the Astore River
basin has an increasing tendency (though the trend is not statistically significant)
(Fig. 2.20). Analysis of Kendall’s tau (τ) coefficient value indicates an upward
tendency (τ = 0.03) of the snow cover area in the catchment. This tendency is noted
both in the maximum snow periods from November to February (τ = 0.06) and in the
minimum snow cover periods from June to September, a slight increasing trend (τ =
0.07) (Fig. 2.20). The trend analysis of climate variables (mean temperature and
precipitation) may help to determine the factors for this increasing snow cover trend.
Trend analysis of the annual and seasonal (spring, winter and summer) mean
temperatures in the Astore river basin by Fowler and Archer (2006) indicates a
decreasing tendency in the catchment temperature over 40 years (1961‒1999) data
record. Hussain et al. (2005) indicated mean temperature falls in the monsoon and
pre-monsoon period for the high mountain regions of HKH. In addition, a significant
increase in the precipitation is reported in the Upper Indus Basin in both winter and
summer during the period 1961‒1999 (Archer and Fowler, 2004). The expansion in
snow cover area of the Astore River basin may be the result of decreasing trend in
the mean temperatures and an increasing trend in the catchment’s annual
precipitation as reported above. The increasing trends of precipitation continue to
feed the high altitudes and results in the form of expanding snow cover in the area.
These reasons of an increasing tendency in snow cover area are very much similar
to the Hunza River basin.
This snow cover area expansion may be associated to the glacier advancing
(especially surging) as reported in section 2.3 of this chapter. Analysing the response
of Himalayan glaciers to the climate change, Scherler et al. (2011) has found that
more than 50% of the glaciers observed in the western Himalaya and Karakoram
region (westerlies-influenced) are advancing or stable in contrast to the retreating
glaciers of eastern Himalaya (monsoon-influenced) and Tibetan Plateau. The current
hypothesis for this phenomenon of increasing cryosphere (snow and ice) area is an
increase in winter precipitation, as explained by Hewitt (2005) and Archer and Fowler
(2004).
85
100
90
80
(Data source: MODIS snow cover product)
70
60
50
40
30
20
10
0
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Time period (8-days classification data)
Seasonal/periodic Mann Kendall's trend analysis:
Slope = x, Kendall's tau coefficient for 10-years SCA,   Snow cover line
Max. snow (snow accumulation) periods trend,  
95% confidence interval line
Min. snow (snow melt) periods trend, 
Fig. 2.20 Snow cover trend in the Astore River basin during the year 2000‒2009. Increasing trend (though not significant) can be
noted over the 10-years data period. Seasonal Mann Kendall’s trend analysis is presented for overall data period, maximum
(October-March) and minimum (April-September) snow cover periods.
86
An increasing tendency of cryosphere area in the two sub-catchments (Hunza and
Astore, situated at different latitudes) of the Upper Indus River basin indicates that
the tendency may be the similar in the rest of Upper Indus River catchment. To verify
this indication, an overall snow cover area over the same 10-years period was
analysed for the Upper Indus River Basin. The trend analysis results in the same
increasing tendency as for the other sub-catchments (Fig. 2.21). This increasing
snow cover tendency may be due to the large concentration of high altitude mountain
ranges like central Karakoram range in this region. The other factor that makes UIB
cryosphere change tendency different from the eastern Himalaya is the climate
change trend. When the eastern Himalaya region is under the threat of retreating
glaciers due to global warming; a distinct situation is reported in the western
Himalaya and Karakoram region with increasing winter precipitation and decreasing
summer temperature trends at high altitudes of UIB (Hewitt, 2005; 2007, Archer and
Fowler, 2004). These climate change trends in UIB are playing important role to
increase the cryosphere area especially in the central Karakoram region of Pakistan.
87
90
80
70
60
50
40
30
20
10
0
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
04
07
10
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Snow cover area line
Seasonal/periodic Mann Kendall's trend analysis: Linear regression line
Slope  x, Kendall's tau coefficient for 10-years SCA,    95% confidence interval line
Fig. 2.21 Snow cover trend in UIB during the year 2000‒2009. Increasing trend (though not significant) can be noted over the 10
years data period.
88
Correlation between snow cover change, climate variables and stream flow in
the Astore River basin:
The relationship between standardised values of four variables (precipitation, snow
cover, mean temperature and discharge) is presented in Fig. 2.22. The values were
standardised using the normal deviate formula as explained in section 2.3, i.e. the
distance of one data point from the mean, divided by the standard deviation of the
distribution. Snow cover in the area is at a maximum 90–95% in the winter and spring
seasons (December to April) and at a minimum 5–10% in the summer (July to
September), as shown in Fig. 2.18. This change in the snow cover has a significant
inverse correlation with the mean temperature and discharge in the Astore River
basin, as shown in Fig. 2.22. This indicates a continuous depletion in the snow cover
in summer as the average temperature in the area increases, resulting in increasing
discharge. In winter, as the temperature decreases, the snow cover starts to
increase. The discharge in Astore River decrease in winter as the snowmelt
decreases because of freezing temperatures. The correlation between monthly
precipitation at high altitude (3000 m) climate station of Rama and the snow cover
area is very different than that given in table 2.6 for Hunza River basin. A correlation
value of 0.52 was found by applying the Pearson’s correlation between the monthly
precipitation at Rama and the snow cover area of the Astore River basin (Fig. 2.22).
An analysis of seasonal relationship between different variables on different seasons
(winter and summer) was also carried. Some important results are given on Fig. 2.22.
89
6
Pearson's correlation:
Tavg vs Q  0.75, Q vs SCA   0.56, Tavg vs SCA   0.89, Precipitation at Rama (P)
P vs SCA  0.52, Q vs P   0.28 Discharge (Q)
5 Mean temperature (Tavg)
Spearman's rank correlation:
Tavg vs Q  0.88, Q vs SCA   0.75, Tavg vs SCA   0.89, Snow cover area (SCA)
P vs SCA  0.54, Q vs P   0.31
4 Seasonal correlation analysis: (Pearson's correlation)
Winter precipitation vs summer discharge 0.76
Winter precipitation vs winter SCA 0.44
Standardised value
3 Summer discharge vs summer Tavg = 0.52
-1
-2
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
03
05
07
09
11
2000
2001
2002
2003
2004
2005
2006
2007
2008
Fig. 2.22 Correlation between standardised values of precipitation (Rama), snow cover area, mean temperature and discharge (at
Doyian) in the Astore River basin (on a monthly basis). All the correlation values are significant with a significance level, p<0.05.
90
350
300
Pearson correlation coefficient = r =  0.39 (P<0.05)
250 Spearman rank correlation coefficient = r =  0.48 (P<0.05)
200
Monthly difference in discharge (m3/s)
150
100
50
-50
-100
-150
-200
-250
-300
-350
-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60
Monthly difference in snow cover ()

Fig. 2.23 Correlation between monthly difference in snow cover and monthly difference in discharge at Doyian in the Astore River
basin. Correlation values are significant with a significance level, p<0.05.
91
A significant positive correlation is found between the winter precipitation and
summer discharge (r = 0.76) (Fig. 2.22) which indicates the accumulation of snow
during the winter precipitation which generates the discharge on melting in summer
season. The same correlation with a higher coefficient value was found by Archer
(2003) for the winter precipitation and summer flow in the Astore River basin. The
winter snow cover area is significantly correlated to the winter precipitation with a
correlation value of 0.44 but there was no correlation found between Hunza
catchment SCA and precipitation (as given in table 2.6). The correlation between
summer discharge and summer mean temperature (r = 0.52) is much lower than that
in the Hunza River basin (r = 0.80, on average as given in table 2.5). These
correlation results are much different from Hunza catchment correlation results where
we found that the river flows are more dependent on the temperature seasonality
than the winter precipitation. However, these differences may also be due to the lack
of quality precipitation data of the Hunza River basin. The monthly difference in the
Astore River discharge at Doyian is inversely correlated (Pearson’s correlation
coefficient = −0.39; Spearman’s rank correlation = −0.48) with the monthly difference
in the snow cover, as shown in Fig. 2.23, which indicates that the change in the
Astore River discharge is dependent on the snow cover change in the area.
However, this correlation value is smaller than that in the case of Hunza River basin
because Astore River discharge is also influenced by summer monsoon rainfall more
than Hunza River flow. The discharge gets the peak during July to September by
increasing snow and glacier melt. It is likely that the large snow cover difference has
a large impact on discharges because a larger snow cover area may also result in
higher snow depths (at least in low-slope valley areas).
Overall, the highest correlation values are found between the discharge, temperature
and SCA change which indicates that the Upper Indus River flow depends mainly on
the snow and glacier melt rate which increases with the increasing temperature in the
region. This is in accordance with the result found by Bookhagen and Burbank (2010)
that the western Himalayan catchments (e.g. Indus River catchment) are fed by the
snowmelt up to 50% as a fraction of the total annual discharge.
92
Trend analysis of stream flow, winter precipitation and summer temperature of
Astore River basin:
The analysis of 33 years of flow records (1974–2007) for the Astore River at Doyian
indicates that the annual flow increases with time, as shown in Fig. 2.24. This is in
contrast to the trend of flows in the Hunza River where the stream flow are
decreasing with time possibly due to the cooling temperatures at higher altitudes
which decelerate the snow and glacier melt especially in summer. This increasing
discharge trend in the Astore River may not be associated to snow and glacier melt
because there is an increasing tendency of cryosphere area as found in this study
and by other researchers (discussed earlier in this chapter). This may be attributed to
the increasing trend of winter and summer precipitation as reported by Archer and
Fowler (2004).
The climate variables which are instrumental for Astore River flows and snow cover
area expansion (winter precipitation and summer mean temperature) are analysed to
verify the trend reported by Archer and Fowler (2004) and Archer (2003). The trend
analysis was performed on the long data series available from the Astore climate
station. The analysis for the winter precipitation (December to February) over a
period (1954‒2007) of 53 years and summer mean temperature (June to August)
over a period (1952‒2000) of 48 years, at Astore climate station is presented in Fig.
2.25 with the linear regression lines and trend line equations. An increasing tendency
in winter precipitation and a decreasing tendency in summer mean temperature can
be noted by the linear trend line and equation. These results are in accordance with
the results of Archer and Fowler (2004) and Archer (2003) with a less significant
tendency in our analysis.
93
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2000 0
100
1900
200
1800 300
1700 400
500
Precipitation at Astore (mm)

1600 600
Runoff at Doyian (mm)
1500 700
Precipitation (mm) 800
1400 Runoff (mm) 900
1300 95% confidence interval 1000
y = (7.389) x - 13635.152 1100
1200
1200
1100 1300
1000 1400
1500
900 1600
800 1700
1800
700
1900
600 2000
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Fig. 2.24 Annual flow trend in the Astore River at Doyian and total annual precipitation at Astore climate station over 25 years
(1980–2004).
94
300
Total winter precipitation, DJF (mm)

Precipitation (a)

95% confidence interval line
y  x 
200
150
100
50
0
52
54
56
58
60
62
64
66
68
70
72
74
76
78
80
82
84
86
88
90
92
94
96
98
00
02
04
06
08
10
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
Summer mean temperature, JJA (°C)
22
(b)
21
20
19
18
Mean temperature
17 95% confidence interval
y  x 
16
52
54
56
58
60
62
64
66
68
70
72
74
76
78
80
82
84
86
88
90
92
94
96
98
00
02
04
06
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
20
20
20
20
Fig. 2.25 Trend analysis of climate data at Astore station (a) winter precipitation (DJF) (1954‒2007) and (b) summer mean
temperature (JJA) (1952‒2000).
95
2.5 Summary of the results found in this study for the snow cover dynamics and
hydrological regime analysis in the Hunza and Astore River basins:
A summary of the snow cover trend results, hydrological regime characteristics and
the correlation results between the stream flow, snow cover area, temperature and
precipitation, is presented in table 2.7. A detailed discussion on these results is
presented previously in this chapter. Hunza River basin is a high altitude snow and
glacier fed catchment situated in central Karakoram region (with a large
concentration of glaciers) which is dependent on the temperature seasonality to
generate the runoff in summer and spring seasons. Astore River basin, on the other
hand, is a middle altitude high runoff catchment situated in the western Himalaya on
the south of Karakoram which has a good correlation between the summer stream
flow and the annual and previous winter precipitation totality. The snow cover area is
noticed to be increased slightly in both catchments due to an increasing winter
precipitation in the area. An increasing tendency in the stream flow of Astore River at
Doyian may be the result of increasing summer monsoon rainfall which does not
influence Hunza River flow to a large extent due to less intrusion of summer
monsoon into central Karakoram region.
The results found and discussed in this chapter for the snow cover dynamics and
hydrological regime of the two snow and glacier fed sub-catchments (Hunza and
Astore) (situated at the north and south latitudes of the Upper Indus River Basin)
suggest that the UIB is a region going under the increasing tendency of cryosphere
area possibly due to the increasing winter precipitation and decreasing summer
temperature trends and that the UIB depends much more on snow and glacier melt
flows than on the rainfall runoff. These results are confirming previous studies: Hewitt
(1998; 2005; 2007), Scherler et al. (2011), Archer (2003), Archer and Fowler (2004)
and Bookhagen and Burbank (2010).
96
Hunza River basin Astore River basin
River flow gauging station Dainyor Bridge Doyian
Latitude 35° 56' N 35° 33' N
Longitude 74° 23' E 74° 42' E
1) Snow cover dynamics: SCA variation= 30‒80% SCA variation= 5‒95%
a) SCA tendency over 10-years Zone A:
(2000‒2009): τ = 0.01 (increasing tendency)
τ = 0.03 (increasing
(Kendall’s tau (τ) coefficient value) Zone C:
tendency)
τ = 0.03 (increasing tendency)
b) SCA tendency over snow Zone A:

accumulation periods (November to τ = 0.14 (increasing tendency) τ = 0.06 (increasing
February) Zone C: tendency)
c) SCA tendency over snow melt periods Zone A:
(July to September) τ = 0.00 (constant trend) τ = 0.07 (increasing
Zone C: tendency)
2) Hydrological regime:
a) Data record:
Stream flow data 1966‒2008 (~40 years) 1974‒2007 (~33 years)
Precipitation 1999‒2008 (~10 years) 1954‒2007 (~53 years)
Temperature 1999‒2008 (~10 years) 1954‒2007 (~53 years)
MODIS snow cover data 2000‒2009 (~10 years) 2000‒2009 (~10 years)
No. of meteorological stations 3 (Naltar, Ziarat and Khunjerab) 3 (Astore, Rattu and
Rama)
Mean annual runoff 742 mm 1084 mm
Total annual precipitation 680 mm at Naltar 501 mm at Astore
225 mm at Ziarat 723 mm at Rattu
170 mm at Khunjerab 794 mm at Rama
b) Correlations (Annual):
Q vs SCA −0.89 −0.56
Q vs Tavg 0.85 0.75
Q vs P 0.08 −0.28
P vs SCA −0.07 0.52
SCA vs Tavg −0.80 −0.89
c) Correlations (Seasonal):
Winter P vs winter SCA 0.01 0.44
Summer Q vs winter P 0.01 0.76
Summer Q vs summer Tavg 0.81 0.52
d) Trend analysis:
Stream flow Decreasing trend Increasing trend
Precipitation Increasing trend Increasing trend
Mean temperature Constancy Decreasing trend
P = precipitation, SCA = snow cover area, Q = discharge, Tavg = Mean temperature
Table 2.7 A summary of the results found in this study and previously for the snow
cover dynamics, hydrological regime and trend analysis of the Hunza and Astore
River basins.
97
2.6 Conclusion
The following conclusions can be drawn by analysing the results obtained in this
chapter:
The climate stations present within the Hunza River catchment area can replace the
Gilgit climate station to study the behaviour of stream flow at Dainyor Bridge in the
future. Moreover, a data series of snow fall measurement at snow accumulation
zones may be helpful to efficiently understand the hydrological regime of the area.
The MODIS MOD10A2 remote sensing cryosphere product is free of charge and
easy to treat and therefore is a good option to assess the cryosphere areas in the
Hunza and Astore River basins over the long term. A comparison of MODIS product
with finer resolution ASTER images proved the validation/reliability of this remote
sensing snow cover data for high elevations in this region.
The stream flow in the Hunza River is influenced mainly by the mean summer and
winter temperatures of the catchment whereas the discharge in the Astore River has
a significant positive correlation with the previous winter precipitation. Moreover, the
most active hydrological region of the Hunza River basin is present above an
elevation of 4500 m in zone C.
The 10-year analysis of the remotely sensed cryosphere data, an increasing winter
precipitation trend, constancy in mean temperatures and declining discharge trend in
the Hunza River basin suggest that the Hunza River basin is a region undergoing a
slight expansion in the cryosphere area, especially at high elevations. The same
tendency in the snow cover change is found in the Astore River basin and eventually
in the Upper Indus catchment due to increasing winter precipitation and cooling trend
of summer mean temperatures in the area. The increase of Astore River flows in
contrast to the Hunza basin may be the result of increasing tendency of summer
monsoon rainfall in the basin. This change in snow cover area of UIB may be used as
a future climate change hypothesis to simulate the Hunza River runoff by using the
Snowmelt-Runoff Model (see next chapter).
The study of annual and seasonal snow cover dynamics suggests that the snow
cover change has a highly inverse correlation with the summer mean temperatures in
both sub-catchments (Hunza and Astore) of UIB, which results in a large amount of
stream flow. Therefore, it is very important to estimate the winter snow accumulations
at high altitudes to forecast the summer flow and ultimately to improve the
management of the water reservoirs downstream like Tarbela reservoir.
98
The results obtained for climate variables trend suggest modeling the snowmelt
runoff under climate variability (increasing winter snow accumulation and a constant
or slightly decreasing mean summer temperatures in Hunza and Astore) to estimate
the future available inflows in the Upper Indus Basin for the better management of
water reservoirs. This analysis of snowmelt runoff modeling in the sub-catchments of
UIB is presented next in the chapter 3 of this study.
99
Résumé étendu du « CHAPITRE 3 »
MODELISATION DU DEBIT DE FONTE DE LA NEIGE
Une proportion majeure de l'écoulement de l'Indus vient des bassins situés dans la
chaîne du Karakoram alimentés par la neige et les glaciers. C'est pour cela qu'il est
essentiel d'estimer le débit de fonte de neige de ces bassins (avec des
enregistrements de précipitation absents ou limités) dans une optique de gestion
des ressources en eau. Le modèle SRM (Snowmelt Runoff Model) associé au
produit neige du capteur satellitaire MODIS a été choisi pour simuler les débits
journaliers et pour étudier l'impact du changement climatique sur ces débits dans
les bassins de l'Indus sous influence nivo-glaciaire.
SRM a d'abord été appliqué au bassin de Hunza, puis aux autres bassins de l'UIB.
Application de SRM au bassin de Hunza : Les résultats obtenus montrent que le
modèle SRM, sur la base d'un facteur degré-jour, simule correctement le débit
journalier. L'efficacité de SRM dans ces bassins de haute altitude peut être
attribuée à l'usage de la représentation de la cryosphère par le produit satellitaire
MOD10A2 comme un forçage équivalent à une production d'eau.
L'analyse de l'impact du changement climatique montre que l'hydrologie du bassin
de Hunza sera modifiée selon différents scénarios. L'application de SRM avec des
scénarios de changement de température moyenne, de précipitation et de couvert
neigeux conduit à un doublement des débits d'été vers le milieu du siècle actuel.
Cette évolution résulte, semble-t-il, d'une température croissante, produisant une
augmentation des écoulements d'été dans la région. Les écoulements de fonte de
neige semblent se décaler en lien avec la remontée de la température au
printemps. Cette analyse suggère que de nouveaux réservoirs soient envisagés
pour stocker les écoulements d'été et répondre aux besoins d'irrigation
supplémentaires à l'aval pendant les périodes de basses-eaux, ainsi qu'à la
demande croissante d'énergie, au contrôle des crues et à l'alimentation en eau
potable. Bien que les scénarios climatiques utilisés dans cette étude aient été
choisis a priori ou déduits d'études précédentes, ils fournissent une information utile
pour la gestion des ressources en réponse aux besoins futurs du pays.
Application de SRM aux autres bassins de l'UIB : Le succès de l'application de
SRM au bassin de Hunza a conduit à son application aux autres sous-bassins à
dominante nivale de l'UIB. SRM a donc été appliqué à trois autres sous-bassins
100
(Gilgit, Astore et Shyok). Les résultats sont présentés dans la section 3.4 de ce
chapitre. Ils sont assez satisfaisants et confirment que l'application de SRM pour
simuler et prévoir les débits journaliers dans ces sous-bassins dominés par les
processus nivo-glaciaires peut apporter une aide à une gestion intégrée des
ressources en eau dans le système d'irrigation de l'Indus (IBIS).
Comparaison des caractéristiques principales et de l'efficacité de SRM pour
les quatre sous-bassins de l'UIB : Le bassin de Shyok, avec la plus haute altitude
moyenne présente la plus grande proportion de son bassin (58%) au-dessus de
5000 m, suivi par le bassin de Hunza (34%). Les températures moyennes
mensuelles (Fig. 3.33) à l'altitude moyenne de chaque bassin sont estimées à partir
des données disponibles à la station la plus proche, corrigées par le facteur de
gradient d'altitude établi pour chaque bassin. Shyok présente la température la plus
basse, suivi respectivement par Hunza, Astore et Gilgit. La température apparaît
comme inversement proportionnelle à l'altitude moyenne. Gilgit possède la
température la plus élevée avec une altitude significativement inférieure à celle des
trois autres bassins. Le cumul de précipitation (moyenne des précipitations des
stations localisées dans le bassin) est le plus élevé à Astore, suivi par Shyok,
Hunza et Gilgit. Cela peut s'expliquer par le fait qu'Astore est situé sur le piémont
sud de la chaîne himalayenne et reçoit beaucoup plus de précipitations de mousson
d'été que les autres bassins.
Une comparaison de l'efficacité du modèle SRM pour les années hydrologiques
2000-01 et 2003-2004 fait l'objet du tableau 3.10 selon les critères de Nash-Sutcliffe
(NS) et de différence de volume (Dv). Une corrélation entre les débits observés et
calculés pour chaque bassin durant les deux années hydrologiques est présentée
Fig. 3.34. On peut considérer que le modèle SRM a correctement simulé les débits
de tous les sous-bassins étudiés dans ce chapitre. Une valeur maximale de NS de
0,93 pour l'année hydrologique 2000-01 a été obtenue pour le bassin de Hunza et de
0,94 à Shyok en 2003-04 (tableau 3.10). Les hauts débits sont surestimés dans les
bassins de Hunza et d'Astore, et sous-estimés dans le bassin de Gilgit (Fig. 3.34).
Mais surtout les performances du modèle ont été bonnes avec une valeur minimale
du coefficient de corrélation de 0,90 (Fig. 3.34) sur toutes les périodes et dans tous
les sous-bassins. Ces résultats sont encourageants pour l'utilisation de SRM dans
ces bassins nivo-glaciaire de haute altitude de l'UIB dans un but de gestion des
101
ressources en eau, tout en restant attentif aux sur- et sous-estimations des valeurs
les plus élevées des débits.
102
CHAPTER 3
SNOWMELT RUNOFF MODELING
Foreword: This chapter is mainly oriented by a research paper (on modeling

snowmelt runoff in the Hunza River basin) accepted for publication in the Journal of
Hydrology, included in the section 3.3 of this chapter. Some additional information is
given in the sections 3.1 and 3.2 of this chapter, in order to complete the reading of
the paper and to improve the statements and research presented in the paper. The
section 3.4 presents the application of Snowmelt Runoff Model (SRM) on the other
snow and glacier fed sub-catchments (Gilgit, Astore and Shyok) of the Upper Indus
River Basin. A comparison of these four sub-catchments (Hunza, Gilgit, Astore and
Shyok) of UIB is presented and discussed in section 3.5.
3.1 Brief introduction and background

The Upper Indus Basin contains the high mountain ranges of Hindukush, Karakoram
and Himalaya. The Karakorum and Himalaya mountains constitutes the heaviest
snow and glacial cover in main land. The Karakoram mountains are lying immediately
north of the western part of the greater Himalaya. It is the highest of the southwest
central Asian mountain systems. This region has many peaks higher than 6000 m. It
has the largest concentration of glaciers on mainland Asia and outside high latitudes.
There are many glaciers over 50 km in length and more than 20 km wide (Nabi,
2009). The perennial snow and ice cover, exceeding 16000 km 2, comprises a huge
fresh water store in a generally arid, drought-prone region as mentioned in the
previous chapter. Snow and glacial melt water contributes a major proportion of the
Indus River flow.
There are many models for snowmelt runoff forecasting. The snowmelt simulation
process varies widely in complexity, from single variable indices of melt to complete
energy balances. Each model is most reliable in the area where it has developed and
may require extensive calibration when applied elsewhere. Irrespective of the choice
of method, modeling of the spatial distribution of snow cover and melt usually is
accomplished by dividing a watershed into a number of smaller units based on
topography such as elevation bands and hill slopes or by geometrical subdivision into
grid squares (Nabi, 2009). Within individual land units the snowmelt, internal
changes, water transmission, storage and other processes are parameterized or
103
described by empirical formulas. The snowmelt models can be placed into two
categories; Energy based Models and Temperature Index Model (Maidment, 1993).
The details of these models are described in next sections.
Energy Balance Models

These models are also described as ―physically based‖ because these models use
equations that describe the physics of different processes in each component of the
energy balance. This approach applies the law of conservation of energy to a control
volume for calculating snowmelt. Some examples of energy balance/physically based
snow models are TOPKAPI model (Ciarapica and Todini, 2002), SHE model (Abbott
et al., 1986), SNAP model (Albert and Krajeski, 1998), ISNOBAL model (Link and
Marks, 1999; Winstral and Marks, 2002), and UEB model (Tarboton and Luce, 1996).
The primary advantage of using physically based snow process models, like all
physically based hydrological models, is that they can be applied in a wide range of
conditions and environments. Their big disadvantage is the large amount of input
data required for calibration and testing. These data may be available for experiment
plot simulations, but present major difficulties when these models are utilized in
spatially distributed frameworks.
Temperature Index Models

Many researchers recommended the use of degree day approach for snowmelt
modeling due to the data constraints. The temperature index (TI) also known as
degree-day models to predict daily rates of snowmelt is accepted to be a
considerable simplification of the full energy balance. These models physically lump
all the components of the surface energy balance into a degree-day melt factor,
which is a proportionality coefficient that calculates melt rates on the basis of air
temperature (normally in excess of some threshold value) alone. Several operational
models used to estimate snowmelt runoff from mountainous area use a temperature
index approach, including the Snowmelt Runoff Model (SRM) (Martinec, 1975) and
HBV (Bergström, 1975). They have also been used extensively as research tools to
investigate melt runoff and mass balance on glaciers.
Different hydrological models with a snow component are used to simulate the daily
stream flows in snow-and glacier-fed catchments. Few examples of their diversity are
given here:
104
 the HBV hydrological model integrated with degree day method was used by Sorman
et al. (2009) to simulate and forecast the discharges in the mountainous catchments
of eastern Turkey.
 The HEC-1 hydrological model coupled with degree day index was used by Verdhen
and Prasad (1993) to simulate the stream flow in a sub-basin of River Beas.
 The World Meteorological Organization (WMO, 1986) has listed almost 18 snowmelt
models and six of these i.e. the SSARR Model, HEC-1, NWSRFS Model, PRMS
Model, SRM and GAWSER Model , have been demonstrated as valuable operational
models and thought to have a high potential for future operational use.
3.2 Description of Snowmelt Runoff Model (SRM)
The Snowmelt-Runoff Model (SRM) is conceptual, deterministic degree day model
used for simulation and forecasting daily rainfall and snow-melt runoff in the
mountainous catchment. It can be applied to evaluate the effect of a changed climate
on seasonal snow cover and runoff. SRM was developed by Martinec (1975) for
small European basins. Due to satellite remote sensing of snow cover measurement
it is possible that SRM model can be applied to large catchments. The model has
been applied to Ganges River basin, which has an area of 917,444 km2 and
elevation up to 8,840 m a.s.l. (Martinec et al., 2007). SRM model has been applied in
more than hundred catchments worldwide.
In general SRM model can be applied to spatial and temporal scale. Runoff
computations by SRM appear to be relatively easily understood. SRM was also
successfully tested by the World Meteorological Organization with regard to runoff
simulations (WMO, 1986) and to partially simulated conditions of real time runoff
forecasts (WMO, 1992). The model require a known or estimated discharge value as
initial condition then it can be run for an unlimited number of days, as long as the
input variables such as temperature, precipitation and snow covered area are
provided.
The main equation of the model computes the water produced from snowmelt and
from rainfall, superimposed on the calculated recession flow and transforms this into
daily discharge from the basin.
Model Accuracy Criteria

The accuracy criteria used for the evaluation of SRM, the Nash-Sutcliffe coefficient
NS and the difference of volume DV are defined as in equations 3.1 and 3.2:
105
(3.1)
where
NS = Nash-Sutcliffe coefficient (a measure of model efficiency)
Qi = measured daily discharge
Q'i = simulated daily discharge
Qavg = average daily discharge for the simulation year or simulation season
n = number of daily discharge values
(3.2)
where
DV = difference between the total measured and simulated runoff
V = measured runoff volume
V' = simulated runoff volume
3.3 SRM application in the Hunza River basin

The Snowmelt Runoff Model (SRM) was applied, in the first step, on the snow and
glacier fed Hunza River basin (Fig. 3.1) over a period from 2000 to 2004. The results
obtained are explained in next section in the form of a research article accepted for
publication (Tahir et al., 2011b) in the Journal of Hydrology.
106
Fig. 3.1 Location of the Hunza River basin and its GDEM (global digital elevation model) to show the elevation variation in the
catchment.
107
Modeling snowmelt-runoff under climate scenarios in the Hunza River basin,
Karakoram Range, Northern Pakistan
Adnan Ahmad TAHIRa,*, Pierre CHEVALLIERa, Yves ARNAUDb, Luc NEPPELa &
Bashir AHMADc
a
Laboratoire Hydrosciences (UMR 5569 - CNRS, IRD, Montpellier Universities 1&2), CC57, Université
Montpellier 2, 34095 Montpellier Cedex 5, France. (E-mail: pierre.chevallier@ird.fr, Pierre
CHEVALLIER ; neppel@msem.univ-montp2.fr, Luc NEPPEL)
b
Laboratoire d’étude des Transferts en Hydrologie et Environnement, UMR 5564 - CNRS, IRD,
Grenoble University 1, Grenoble INP, LGGE, 54 rue Molière, Domaine Universitaire, BP 96, 38402
Saint Martin d’Hères Cedex, France. (E-mail: yves.arnaud@ird.fr)
c
Director (Environment), Natural Resources Division, Pakistan Agricultural Research Council, G1/5,
Islamabad, Pakistan. (E-mail: bashirad@hotmail.com)
*Corresponding author. [E-mail address: tahir@msem.univ-montp2.fr; uaf_adnan@hotmail.fr (Adnan

Ahmad TAHIR)], Tel.: +33 467 14 90 64, Fax. +33 467 14 47 74
Abstract
A major proportion of flow in the Indus River is contributed by its snow and glacier-fed
river catchments situated in the Karakoram Range. It is therefore essential to
estimate the snowmelt runoff from these catchments (with no or scarce precipitation
records) for water resources management. The snowmelt runoff model (SRM)
integrated with MODIS remote-sensing snow cover products was selected to
simulate the daily discharges and to study the climate change impact on these
discharges in the Hunza River basin (the snow- and glacier-fed sub-catchment of the
Indus River). The results obtained suggest that the SRM can be used efficiently in the
snow- and glacier-fed sub-catchments of the Upper Indus River Basin (UIB). The
application of the SRM under future climate (mean temperature, precipitation and
snow cover) change scenarios indicates a doubling of summer runoff until the middle
of this century. This analysis suggests that new reservoirs will be necessary for
summer flow storage to meet with the needs of irrigation supply, increasing power
generation demand, flood control and water supply.
Keywords: Upper Indus River Basin (UIB); Hunza River basin; MODIS; water
resources management; Snowmelt-Runoff Model; climate change.
108
1. Introduction
The economy of Pakistan, an agriculture-based country, is highly dependent on the
Indus irrigation system. The Indus River is one of the major water carriers of South
Asia emerging from the Tibetan Plateau and the Himalayas. It has a controlling
storage at Tarbela (Fig. 3.2) as the river leaves the mountains. Tarbela is the first
major structure on the Indus River and supplies the flow to the Indus Irrigation
System to irrigate the agricultural lands of Punjab and Sindh (provinces of Pakistan),
the dominant producer of agricultural products in the country. Inflow to Tarbela is
measured at Besham Qila (Indus basin area at Besham = 201,388 km²), situated
approximately 80 km upstream of Tarbela (Fig. 3.2), with a mean annual flow of 2410
m3/s according to the SWHP (Surface Water Hydrology Project) flow records from
1969 to 2008. The catchment area upstream of Tarbela reservoir is called the Upper
Indus Basin (UIB) (Fig. 3.2), which contributes the major part of inflow into the Indus
River from high elevations as a result of snow and glacier melt.
The active hydrological zone of the Upper Indus Basin lies in the high-altitude
Himalaya and Karakoram ranges. Several authors (Archer and Fowler, 2004; Hewitt
et al., 1989; Wake, 1989; Young and Hewitt, 1990; Bookhagen and Burbank, 2010)
reported that more than 65% of the annual flow of the Upper Indus River is
contributed by the seasonal and permanent snowfields and glacierised areas above
3500 m in elevation. Liniger et al. (1998) stated that some 90% of the lowland flow of
the Indus River System originates from the Hindukush, Karakoram and western
Himalaya mountain areas.
109
Fig. 3.2 Location map of study area (Hunza River basin).
110
The management of Tarbela reservoir depends to a large extent on the summer
inflow contributed by the snow- and glacier-fed tributaries situated in the Karakoram
Range. The presence of snow in a basin strongly affects the moisture that is stored at
the surface and is available for future runoff (Maurer et al., 2003). The summer runoff
is highly correlated with the summer mean temperature in these high-altitude sub-
catchments of the UIB, mostly covered with permanent snow pack and glaciers
(Archer, 2003). The Hunza River at Dainyor Bridge (Fig. 3.2) represents the
moderately high runoff catchments in the center of the Karakoram where a significant
proportion of the flow is derived from cryosphere melt. It nearly doubles the runoff
rate along with the Gilgit River in the overall Indus catchment at Partab Bridge station
(Fig. 3.2), i.e., at its confluence point with the Indus River. Climate change in these
sub-catchments will directly influence the inflow to the Indus irrigation system. Linear
regression analysis by Archer (2003) indicates that a 1°C rise in mean summer
temperature would result in a 16% increase in summer runoff into the Hunza and
Shyok River due to accelerated glacier melt. Analytical studies representing
temperature increases of 1–3°C in the western Himalayan region suggest an
increase in glacial melt runoff by 16–50% (Singh and Kumar, 1997). It is therefore
important to use reliable hydrological tools to simulate runoff, resulting from the snow
and glacier melt, under future climate change scenarios in the high-altitude
catchments to assess the water supply available for future needs. Regional climate
investigations by the IPCC (2007) indicate that the median temperature will rise up to
3.7°C in Central Asia by the end of 21st century. According to the IPCC (2007), the
temperature increase in the Himalayan region has been greater than the global
average of 0.74°C over the last 100 years. This temperature increase may be
associated specifically to the Greater Himalaya but not to the Hindukush and
Karakoram region. The Karakoram region shows a different climate trend from that of
the Greater Himalaya (Fowler and Archer, 2005). The Karakoram region is influenced
by the westerly circulations which bring the maximum of precipitation in winter in the
form of snow whereas the Greater Himalayan region is under the influence of
monsoon regime. This climate variability in the region is explained in detail by Fowler
and Archer (2005) and Tahir et al. (2011a). The future climate variables simulated by
Akhtar et al. (2008) using a regional climate model indicate that the annual mean
temperature will rise up to 4.8°C and annual mean precipitation up to 16% until the
end of the 21st century (2071–2100).
111
Snowmelt Runoff Model
Different hydrological models with a snow component (e.g., Sorman et al., 2009;
Verdhen and Prasad, 1993; WMO, 1986) are used to simulate the daily stream flows
in snow- and glacier-fed catchments.
Nevertheless, most of the hydrological models are not satisfactory for daily stream
flow simulation and forecasting in the high-altitude catchments where the snowmelt is
a major factor in the water cycle (Martinec et al., 2007). One of the widely applied
models to simulate and forecast the daily stream flows in these types of catchment is
the Snowmelt-Runoff Model (SRM) developed by Martinec (1975), which uses
remotely sensed cryosphere data as basic input (Georgievsky, 2009). Most recently,
it has also been applied to evaluate the effect of climate change on seasonal snow
cover and runoff (Martinec et al., 2007).
The SRM had successfully undergone tests by the World Meteorological
Organization with regard to runoff simulations (WMO, 1986) and to partially simulate
conditions of real-time runoff forecasts (WMO, 1992). Due to the progress of satellite
remote-sensing of the cryosphere, the SRM has been applied to larger and larger
basins. To date the model has been applied by various agencies, institutes and
universities to over 100 basins, situated in 29 countries (Martinec et al., 2007). It is
currently based on a simple degree-day method. The daily precipitation, air
temperature and snow cover area are the input data. In addition to the input
variables, a number of basin characteristics such as basin area, zone area (in zone-
wise application) and the hypsometric (area-elevation) curve are also needed. The
daily water produced from the snowmelt and rainfall is superimposed on the
calculated recession flow and transformed into daily discharge from the catchment
according to equation 3.3:
(3.3)
where: = average daily discharge (m3/s)

runoff coefficient, with Cs referring to snowmelt and Cr to rain
degree-day factor, DDF (cm. oC−1 d−1), T = number of degree-days (°C.d)
S = ratio of the snow covered area (SCA) to the total area,
n = sequence of days
P = precipitation contributing to runoff, A = area of the basin or zone
k = recession coefficient (Xc and Yc).
112
The detailed description of the variables and parameters used in this equation is
given by Martinec et al. (2007).
In the Himalayas and its surroundings, Immerzeel et al. (2009, 2010) investigated the
effects of snow cover dynamics on the discharge of the Upper Indus River and
concluded that stream flows can be predicted with a high degree of accuracy using
MODIS snow cover data in the SRM. Bookhagen and Burbank (2010) found that the
SRM combined with MODIS snow cover and TRMM rainfall data significantly
improves regional runoff modeling for the Himalayas. Prasad and Roy (2005) applied
the SRM to the Beas River basin and found that the observed and computed
discharges were well correlated. Hong and Guodong (2003) found a Nash-Sutcliffe
coefficient value of 0.87 when simulating the stream flow in the Gongnaisi River basin
in the western Tianshan Mountains by applying the SRM.
In other regions of the world where the impact of the snow cover is significant, Lee et
al. (2005) reported that the SRM has sufficient accuracy for stream flow prediction
using the MODIS snow cover products in the snowmelt-dominated basin of the Upper
Rio Grande basin. Georgievsky (2009) tested the SRM as a short-term forecasting
model and concluded that the model can be used for short-term runoff forecasts in
the mountain and foothill areas of the Krasnodar reservoir basin. Sirguey et al. (2009)
applied the SRM to the Waitaki catchment, New Zealand, and found that the
simulation of daily stream flows is reliable and that the SRM has the potential to
provide short- to medium-term forecasts of water availability in these catchments.
The previous studies of applying the SRM to mountainous river catchments of
Himalaya and other regions of the world suggest that the model can efficiently be
applied in the snow- and glacier-fed Hunza River basin to simulate and forecast the
daily stream flows as well as to study the effect of future climate change on river
runoff. The other factor that suggests the use of SRM is its weak sensibility towards
the precipitation forcing. Since, the precipitation data available from the high altitude
catchments is not of very good quality so we preferred to use the SRM which is more
sensitive to the daily snow cover and temperature input. In this study, the SRM
(WinSRM version 1.12) was applied to the Hunza River basin using observed and
APHRODITE (Yatagai et al., 2009) precipitation data (described in section 3.3),
which will show the reliability of APHRODITE precipitation products for hydrological
modeling in the remote regions of the UIB. This study will also help manage the water
reservoirs (existing Tarbela and proposed Diamer Bhasha) (Fig. 3.2) of the UIB for
113
irrigation, hydropower generation, flood control and water supply by providing a
reliable estimation of the future snowmelt runoff in one of the main snow- and glacier-
fed tributaries of the Indus River under different future climate change scenarios.
The main objectives of this study were to:
a) Evaluate the SRM’s ability to simulate the daily stream flows in the snow- and glacier-
fed Hunza River basin;
b) Investigate the reliability of APHRODITE precipitation data for modeling snowmelt
runoff in the high-altitude river catchment;
c) Simulate stream flow in the Hunza River basin under different future climate change
scenarios for water resources management in the UIB.
The study area is described in section 2 of this manuscript followed by the description
of the data sets and treatment of data in section 3. The principle and method of the
SRM is described in section 4. The results obtained from this study and discussion
on these results are described in section 5 following by the conclusions drawn in
section 6.
2. Study area: Hunza River basin
The present evaluation of the SRM is undertaken in the Hunza River basin (drainage
area, 13,733 km2) (Fig. 3.2), situated in the high-altitude central Karakoram region,
with a mean catchment elevation of 4631 m. Approximately 4463 km² of catchment
area is at an elevation above 5000 m and almost the same area (33%) is glaciated
(Akhtar et al., 2008), as shown in Fig. 3.3. Area distribution for different altitudinal
zones in the catchment is shown by the hypsometric curve in Fig. 3.4 and given in
table 3.1.
A previous paper described the hydrological behaviors and regime of the Hunza
River basin at Dainyor Bridge (Tahir et al., 2011a). The main elements are
summarized here.
The snow cover area in the Hunza River basin varies from approximately 80% in
winter to 30% in summer. Three climate stations with precipitation gauges are
installed by WAPDA at different altitudes within the catchment, as shown in Fig. 3.3.
The mean total annual precipitation is 170 mm at Khunjerab (4730 m), 225 mm at
Ziarat (3669 m) and 680 mm at Naltar (2858 m) according to the 10-year record
(1999–2008) of the three climate stations in the catchment. The Hunza River has a
mean annual flow of 323 m3/s (i.e., 742 mm of water depth equivalent) (standard
deviation for mean annual flow = 69 m 3/s), gauged at Dainyor Bridge (Fig. 3.2),
114
according to the SWHP-WAPDA 40-year (1966–2008) flow record. Obviously,
present precipitation records are not representative of the runoff at the outlet
because of the well-known gauging errors at high altitudes, as explained in detail by
Sevruk (1985, 1989) and Førland et al. (1996). The other factor participating in this
contrast is the absence of precipitation gauging stations at high altitudes above 4730
m where maximum snowfall and accumulation occurs, as confirmed by Young and
Hewitt (1990).
115
Fig. 3.3 Global digital elevation model (GDEM) of the Hunza River basin indicating the six altitudinal zones extracted for this study.
116
8000
7500
Zone F
7000
6500
6000 Zone E
5500
5000
Elevation (m)
4500 Zone D
4000
Zone C
3500
3000
Zone B
2500
2000
1500
1000
Zone A

0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
5500
6000
6500
7000
7500
8000
8500
9000
9500
10000
10500
11000
11500
12000
12500
13000
13500
14000
2
Cumulated area (km )
Fig. 3.4 Hypsometric curve of the Hunza River basin showing the area distribution in six different elevation zones.
117
3. Data sets
3.1. Topography
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER),
Global Digital Elevation Model (GDEM) is used to delineate the catchment boundary
studied. The ASTER GDEM is available for high-latitude and steep mountainous
areas not covered by SRTM3 (METI and NASA, 2009). Six different altitudinal zones
were extracted from the GDEM of the study area for detailed analysis of snow cover
area (SCA) estimation. Some characteristics of these zones are given in table 3.1.
Zone Elevation range (m) Mean elevation Area Area Climate

(m) (%) (km2) station
A 1432–2500 1966 3.1% 425 -
B 2500–3500 3000 11.5% 1574 Yasin
C 3500–4500 4000 29.4% 4038 Ziarat
D 4500–5500 5000 44.6% 6123 Khunjerab
E 5500–6500 6000 10.1% 1388 -
F 6500–7849 7166 1.3% 173 -
Total 100% 13,733
Table 3.1 Characteristics of the six elevation zones extracted from the DEM of the
Hunza River basin.
3.2. Hydrometeorological data
Hydrology Project of the Water and Power Development Authority (SWHP-WAPDA)
with the earliest records beginning in 1960. The database of daily discharges for the
Hunza River gauged at Dainyor Bridge was made available for this analysis from
January 2000 to December 2004. Meteorological data (daily mean temperatures and
precipitation) available at three high-altitude climate stations (Fig. 3.3 and Table 3.1)
was provided by WAPDA for the same period.
3.3. APHRODITE precipitation data
A database of gridded daily precipitation products, constructed by the Asian
Precipitation - Highly-Resolved Observational Data Integration Towards Evaluation of
the Water Resources (APHRODITE water resources) project, was used in this study
to assess the reliability of this product in high-altitude catchments of the UIB. The
118
gridded fields of daily precipitation are defined by interpolating rain-gauge
observations obtained from meteorological and hydrological stations over the region
(Yatagai et al., 2009). The data were downloaded from the APHRODITE water
resources project web page (http://www.chikyu.ac.jp/precip/). The product used in
this study comprises the 0.25×0.25-degree gridded data over Monsoon Asia
(APHRO_MA_V0902) from 2000 to 2004. The APHRODITE data grid has a coarse
resolution (30 x 30 km²) and the precipitation is assumed to be constant at any
elevation in this grid area.
3.4. Satellite data
MODIS snow cover
selected to calculate the percentage of snow cover area in the study area. MODIS
snow cover products are used by several researchers to use as input for the
snowmelt runoff model (e.g., Bookhagen and Burbank, 2010; Immerzeel et al., 2009;
Prasad and Roy, 2005). The MODIS/Terra Snow Cover 8-Day L3 Global 500 m Grid
(MOD10A2), used for this study, contains data fields for maximum snow cover extent
over an 8-day repeated period (Hall et al., 2006, updated weekly.) and has a
resolution of approximately 500 m completely covering the Hunza River basin. A data
set of 228 processed MOD10A2 (V005) images available from March 2000 to
December 2004 was downloaded from http://nsidc.org/cgi-bin/snowi/search.pl. The
available MODIS images were mosaicked and projected with the WGS 1984 UTM
ZONE 43N projection system. The Hunza River basin area was then extracted from
this mosaicked scene to assess the percentage of snow and ice cover (cryosphere)
in the study area. When the percentage of cloud cover exceeded 20% on a specific
date, the record was removed and then the average snow cover was estimated on
this date by interpolating linearly between the previous and the next available cloud-
free images. The snow cover area was also calculated for the different altitudinal
zones (Table 3.1) to use further for snowmelt-runoff modeling.
4. Principle and method
The Snowmelt-Runoff Model (SRM) was used to simulate the daily discharges in the
Hunza River at Dainyor Bridge. It uses the remote sensing snow cover data as basic
input. The basic parameters to be calibrated for the SRM are given in table 3.2. The
model was run for four hydrological years (April 2000 to March 2004) to simulate the
daily discharges in the Hunza River basin. The SRM parametric values were
119
estimated during the calibration of the model and extracted from the studies (e.g.,
Dey et al., 1989; Hock, 2003; Immerzeel et al., 2010; Prasad and Roy, 2005; Zhang
et al., 2006) conducted previously on other Himalayan basins. It was applied by using
both proposed options: ―basin-wide‖ and ―zone-wise‖.
Runoff coefficient (snow), Cs Runoff coefficient (rain), Cr

Critical temperature, Tcrit. [oC] Degree day factor (DDF), a [cm oC−1 d−1]
Rainfall contributing area, RCA Recession coefficient (Xc and Yc), k
Temperature lapse rate, γ Time lag, L
Table 3.2 Snowmelt-runoff model (SRM) parameters to be calibrated.
Basin-wide application: The average daily precipitation and daily mean

temperatures observed at the three climate stations in the Hunza River basin, as well
as the daily snow cover area percentage were used as input variables. The total
catchment area and the hypsometric (area-elevation) curve (Fig. 3.4) for the basin
were calculated using the GDEM. The daily basin-wide snow cover area (Fig. 3.5) for
the Hunza River basin was estimated by the linear interpolation of the snow cover
area percentage calculated from 8-day composite MOD10A2 satellite images. The
SRM was also run by replacing the daily observed precipitation with the APHRODITE
precipitation data (described in section 3.3) to determine the reliability of
APHRODITE precipitation products for runoff modeling in the high-altitude
catchments. The values for different parameters used in the basin-wide simulations
are given in table 3.3.
120
Snow cover area (%)
20
30
40
50
60
70
80
90
05-Jan
13-Jan
21-Jan
29-Jan
06-Feb
14-Feb
22-Feb
02-Mar
10-Mar
18-Mar
26-Mar
03-Apr
11-Apr
19-Apr
27-Apr
05-May
13-May
21-May
29-May
06-Jun
14-Jun
from the remotely sensed MODIS (MOD10A2) snow cover data.

22-Jun
30-Jun
08-Jul
16-Jul
24-Jul
01-Aug
Time period (Daily)
09-Aug
17-Aug
25-Aug
02-Sep
10-Sep
18-Sep
26-Sep
04-Oct
12-Oct
20-Oct
28-Oct
05-Nov
13-Nov
21-Nov
29-Nov
07-Dec
15-Dec
2004
2003
2002
2001
2000
23-Dec
31-Dec
Fig. 3.5 Snow cover distribution (basin-wide) in the Hunza River basin over a period of 2000‒2004. Snow cover area is estimated
121
Parameter values (using Parameter values
Parameter observed precipitation (using APHRODITE
data) precipitation data)
Lapse rate (°C/100m) 0.64 0.64
Tcric (°C) 0 0
DDF (cm. °C−1d−1) 0.5 0.5
Lag Time (hrs) 18 18
0.20 (June-August) 0.30 (June-August)

Cs
0.15 (September-May) 0.15 (September-May)
0.15 (July-August) 0.20 (June-August)
Cr
0.10 (September-June) 0.10 (September-May)
1 (June-August) 1 (June-August)
RCA
0 (September-May) 0 (September-May)
0.80 (June-September) 0.80 (June-September)
Xc
1.06 (October-May) 1.06 (October-May)
Yc 0.02 0.02
Table 3.3 SRM Parameter values for basin-wide application using observed and
spatial data.
Zone-wise application: The Hunza River basin was divided into six altitudinal zones
with a nearly 1000-m difference in mean hypsometric elevation between two zones.
The daily zonal mean temperatures (Fig. 3.6) for zones A, E and F (zones with no
climate stations) were determined by extrapolating the mean temperature records
available in zones B and D (Naltar and Khunjerab) by using the lapse rate value of
0.48°C/100 m for zone A and 0.76°C/100 m for each of zones E and F. These lapse
rate values were calculated by using the available observed data on three climate
stations of the Hunza River basin. The estimation of precipitation data for each zone
using a lapse rate is not necessary because the SRM simulated discharge is strongly
influenced by the change in snow cover area and less by the precipitation input.
Therefore, the average of daily precipitation values observed at the three climate
stations was used for each altitudinal zone. The daily zonal snow cover area (Fig.
3.7) was estimated by the linear interpolation of the snow cover area percentage
122
calculated from 8-day composite MOD10A2 satellite images. The total zonal area
and the area distribution (hypsometry) curve for each zone were calculated from the
GDEM. The degree-day factor (DDF) values were extracted from the previous
studies e.g. Hock (2003) and Zhang et al. (2006) for the Himalayan region. These
studies present a DDF value of 5.7‒7.4 (mm. oC−1 d−1) for ice and about 4‒5 (mm.
C−1 d−1) for snow. We assumed that the altitudinal zones higher than 5000 m are
o
glacier covered and used the higher value of DDF (7 mm. oC−1 d−1) for these zones.
The values for different parameters used in the zone-wise simulations are given in
table 3.4.
The main equation of the model computes the water produced from snowmelt and
from rainfall, superimposed on the calculated recession flow and transforms this into
daily discharge from the basin. The data series available for 2000–2001 was used for
the calibration of SRM parameters. Both classical criteria in hydrological modeling
were used to evaluate the SRM: the Nash-Sutcliffe coefficient (NS) and the difference
of volume (DV). In addition, the Pearson correlation coefficient (Rodgers and
Nicewander, 1988) and the Kendall rank correlation (Kendall, 1975; Kendall and
Gibbons, 1990) tests were used (significance level, 5%) to evaluate the relationship
between the measured and simulated daily discharges from 2000 to 2004, in the
basin-wide and zone-wise applications of the SRM.
123
30
20
10
Mean temperature (°C)
-10
-20
-30
-40
Jan-00
Mar-00
Oct-00
Jan-01
Mar-01
Jun-01
Oct-01
Jan-02
Mar-02
Jun-02
Jan-03
Mar-03
Jun-03
Oct-03
Jan-04
Jun-04
Oct-04
Apr-00
May-00
Apr-01
May-02
May-03
Apr-04
May-04
Feb-00
Aug-00
Sep-00
Nov-00
Feb-01
Aug-01
Sep-01
Feb-02
Aug-02
Sep-02
Nov-02
Feb-03
Aug-03
Nov-03
Feb-04
Aug-04
Nov-04
Jul-00
Jul-01
Dec-01
Jul-02
Dec-02
Jul-03
Dec-03
Jul-04
Dec-04
Time period (Daily)
Zone A < 2500 m Zone B (2500-3500 m) Zone C (3500-4500 m)

Zone D (4500-5500 m) Zone E (5500-6500 m) Zone F> 6500 m
Fig. 3.6 Temperature variations in different elevation zones of the study area, presented with a mobile average of 15 days period
line. The data for zones B, C and D is the actual observed data from the climate stations (WAPDA) installed in these zones. The
data for zones A, E and F is estimated from the actual data by using the mean temperature lapse rate of 0.48‒0.76°C/100m (lapse
rate is calculated from the actual observed data of the catchment).
124
Snow cover (%) 100%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Jan-00
Feb-00
Mar-00
Apr-00
May-00
Jun-00
A
Aug-00
Sep-00
Oct-00
Nov-00
Dec-00
Jan-01
Mar-01
Apr-01
B
May-01
Jun-01
Jul-01
Sep-01
Oct-01
remotely sensed MODIS (MOD10A2) snow cover data.

Nov-01
Dec-01
Jan-02
C
Mar-02
Apr-02
May-02
Jun-02
Jul-02
Aug-02
Time period (Daily)
Oct-02
Nov-02
D
Dec-02
Jan-03
Feb-03
Apr-03
May-03
Jun-03
Jul-03
E
Aug-03
Sep-03
Nov-03
Dec-03
Jan-04
Feb-04
Mar-04
May-04
F
Jun-04
Jul-04
Aug-04
Sep-04
Oct-04
Dec-04
Fig. 3.7 Snow cover distribution in six different altitudinal zones of the Hunza River basin. Snow cover area is estimated from the
125
Parametric values for each altitudinal zone
A B C D E F
Parameter (1450‒2500m) (2500‒3500m) (3500‒4500m) (4500‒5500m) (5500‒6500m) (6500‒7800m)
Lapse rate
0.48 0.48 0.64 0.76 0.76 0.76
(°C/100m)
Tcric (°C) 0 0 0 0 0 0
DDF
o −1 −1 0.5 0.5 0.5 0.6 0.7 0.7
(cm. C d )
LagTime (hrs) 6 6 12 12 18 18
Jun–Aug=0.25 Jun–Aug=0.35 Jun–Aug=0.35 Jun–Aug=0.50 Jun–Aug=0.50
Cs 0
Sep–May=0.20 Sep–May=0.3 Sep–May=0.3 Sep–May=0.4 Sep–May=0.4
Jun–Aug=0.50 Jun–Aug=0.35 Jun–Aug=0.35 Jun–Aug=0.35
Cr 0 0
Sep–May=0.40 Sep–May=0.30 Sep–May=0.3 Sep–May=0.3
Jun‒Sep=1
RCA 1 1 1 0 0
Oct‒May=0
Jun–Sep=0.80 Jun–Sep=0.80 Jun–Sep=0.80 Jun–Sep=0.80 Jun–Sep=0.8 Jun–Sep=0.80
Xc
Oct–May=1.06 Oct–May=1.06 Oct–May=1.06 Oct–May=1.06 Oct–May=1.06 Oct–May=1.06
Yc 0.2 0.2 0.2 0.2 0.2 0.2
June–August Snow melt and monsoon period
September/October –May Snow accumulation and dry period
Zone E and F Zones with almost 100% glacierized area
Zone D Almost 50% glacierized area
Cs and Cr Zone A has a small snowmelt runoff contribution and Zones E and F have a small rainfall runoff contribution
RCA Rainfall Contributing Area is small in Zone E and F because at this altitude the precipitation is in solid form
DDF Degree Day Factor for snow is less and for ice/glacier is greater in Karakoram range (Hock, 2003; Zhang et al., 2006)
Table 3.4 SRM parameter values for zone-wise SRM application to the Hunza River basin.
126
Climate change scenarios: After calibration and validation on the present time (year
of reference: 2000), the SRM was used to study the impact of climate variability on
the Hunza River runoff. The different future climate scenarios for mean temperature,
precipitation and snow cover area were used to estimate the relative stream flow in
the future for water resources management. These scenarios are described as
follows:
a) A 20% increase in cryosphere area until 2075 and a 10% increase until 2050. The
cryosphere area is assumed to increase under the increasing precipitation scenario,
as explained by Akhtar et al. (2008), if the mean temperature remains constant.
Some of the previous studies indicate a trend toward increasing cryosphere area in
the Karakoram region, for example (Hewitt, 2005; Immerzeel et al., 2010);
b) A 4°C increase in mean temperature until 2075 (Akhtar et al., 2008; IPCC, 2007), a
3°C increase until 2050 and a 2°C increase by 2025 (considering other variables as
constant, particularly precipitation);
c) An increase of 2–4°C in the mean temperature (4°C rise in zones A, B and C; 3°C in
zone D; 2°C in zones E and F) and a 20% increase in snow cover area until 2075
(this scenario is assumed by taking into account the two previous scenarios, a and
b).
127
5. Results and discussion
Basin-wide application of the SRM (using observed precipitation data): The
results obtained for the basin-wide application of the SRM in the Hunza River basin
over 4 hydrological years (2000–2004) (annual, snowmelt and extreme discharge
periods) are presented in table 3.5. These results suggest that the model has
efficiently simulated the daily discharges in the Hunza River during the annual (April–
March), snow melt (April–September) and most of the extreme discharge (July–
August) periods. The simulation of daily discharge in the snow accumulation period is
not substantial because the discharge is very low during this period in the Hunza
River basin. The management of water resources is more closely associated with the
discharge contributed by the Hunza River during the snowmelt period. The NS
coefficient value was never less than 0.86. The best performance of the model was
found over the validation year 2001–2002 with the NS coefficient value of 0.97 and a
2% difference in volume (Fig. 3.8). A NS coefficient value of 0.92 was found over the
entire simulation period from 2000 to 2004. The correlation coefficient values were
never less than 0.94 and 0.82 for the Pearson and Kendall rank correlation tests,
respectively. The extreme events are not modeled as efficiently as the entire
snowmelt period. A maximum NS coefficient value of 0.76 was found over the year
2001 (Table 3.5). This may be associated to the glacial contribution during this period
from the catchment which is not captured efficiently by the model.
128
Model efficiency (with Correlation between Model efficiency (with Correlation between
observed precipitation observed and simulated APHRODITE observed and simulated
data) discharge (with precipitation data) discharge (with
observed precipitation APHRODITE precipitation
Hydrological year
data) data)
(April‒March)
Difference Nash- Difference Nash-
of Sutcliffe of Sutcliffe
Pearson Kendall rank Pearson Kendall rank
Volume coefficient Volume coefficient
correlation correlation correlation correlation
(Dv%) (Dv%)
coefficient coefficient coefficient coefficient
Calibration
2000–01 −0.7 0.93 0.97 0.90 −7.4 0.89 0.95 0.89
Validation
2001–02 2.0 0.97 0.99 0.92 9 0.93 0.98 0.91
2002–03 −4.8 0.90 0.95 0.89 −21 0.80 0.98 0.91
2003–04 12 0.86 0.94 0.82 −4.5 0.94 0.98 0.85
Snowmelt period
(April‒September)
Calibration
2000 2.5 0.88 0.94 0.82 −6.7 0.79 0.91 0.80
Validation
2001 −0.3 0.95 0.98 0.91 11.7 0.87 0.96 0.87
2002 −0.7 0.82 0.92 0.77 −25 0.64 0.96 0.87
2003 9.0 0.78 0.90 0.76 −9.6 0.90 0.96 0.85
Extreme events
(July‒August)
2000 4.5 0.50 0.70 - −11.4 0.25 0.51 -
2001 5.3 0.76 0.87 - 13.2 0.62 0.79 -
2002 3.9 0.19 0.44 - −30.8 0.40 0.64 -
2003 22.2 0.74 0.86 - −9.0 0.83 0.91 -
Table 3.5 Evaluation of basin-wide SRM application to the Hunza River basin during the hydrological year, snowmelt period
(April‒September) and extreme discharge period (July‒August), using observed and spatial precipitation data.
129
2000 0
1800 5
1600 10
1400 Measured discharge 15

Simulated discharge-BW
Precipitation (mm)
Discharge (m /s)
Simulated discharge-ZW
1200 20
3
Gauged precipitation
1000 Nash-Sutcliffe coefficient, NS  (BW), 25

 (ZW)
Difference of volume, Dv  (BW),
800  (ZW) 30
600 35
400 40
200 45
0 50
Jul-01
May-01
Jun-01
Oct-01
Nov-01
Dec-01
Jan-02
Feb-02
Aug-01
Sep-01
Mar-02
Apr-01
Time period (Daily)

Fig. 3.8 Evaluation of basin-wide (BW) and zone-wise (ZW) SRM application over the hydrological year 2001–2002 in the Hunza
River basin, by using observed precipitation data.
130
Zone-wise application of SRM: The results obtained for the zone-wise application
of the SRM in the Hunza River basin over 4 hydrological years (2000–2004) (annual,
snowmelt and extreme discharge periods) are presented in table 3.6. The NS
coefficient value was never less than 0.83 and 0.72 over the annual and snowmelt
simulation periods, respectively. The best performance of the model was found over
the validation year 2001–2002 with the NS coefficient value of 0.94 and a 3%
difference in volume (Fig. 3.8). A NS coefficient value of 0.90 was found for the entire
simulation period from 2000 to 2004. The correlation coefficient values were never
less than 0.94 and 0.74 for the Pearson and Kendall rank correlation tests over the
annual simulation periods, respectively. The extreme events are also modeled more
efficiently than for basin-wide application. A maximum NS coefficient value of 0.85
was found over the year 2003 (Table 3.6). The basin-wide and zone-wise simulated
discharge data were plotted against each other to analyse the reaction of both
datasets against the high discharges (Fig. 3.9). The zone-wise simulation seems to
capture high discharges more efficiently than the basin-wide simulation. This may be
associated with the fact that the model parameters estimated for each zone are more
reliable than an average value estimated basin-wide.
131
Model efficiency (with observed precipitation data) Correlation between observed and simulated discharge
Hydrological year (April-March) Difference of Volume Pearson correlation Kendall rank correlation
Nash-Sutcliffe coefficient
(Dv%) coefficient coefficient
Calibration
2000–01 −2.6 0.90 0.95 0.89
Validation
2001–02 3.0 0.94 0.97 0.86
2002–03 1.2 0.92 0.96 0.75
2003–04 8.1 0.83 0.94 0.74
Snowmelt period
(April‒September)
Calibration
2000–01 0.46 0.82 0.92 0.77
Validation
2001–02 6.43 0.89 0.95 0.86
2002–03 −0.01 0.85 0.92 0.78
2003–04 13.20 0.72 0.92 0.79
Extreme events (July‒August)
2000 0.4 0.72 0.85 -
2001 −0.5 0.72 0.85 -
2002 10.0 0.46 0.68 -
2003 29.8 0.85 0.92 -
Table 3.6 Evaluation of zone-wise SRM application to the Hunza River basin during the hydrological year, snowmelt period
(April‒September) and extreme discharge period (July‒August), by using observed precipitation data.
132
1800
y = 0.95 x + 15.05
1600 95% confidence interval line
95% Prediction interval line
Nash-Sutcliff coefficient, NS 
1400 Difference of volume, Dv 
number of data points, n 
ZW Simulated discharge (m3/s)
1200
1000
800
600
400
200
0
0 200 400 600 800 1000 1200 1400 1600 1800
BW simulated discharge (m3/s)
Fig. 3.9 Comparison of basin-wide (BW) and zone-wise (ZW) simulated runoff by applying the SRM from April 2000 to March 2004
(using observed precipitation data).
133
Basin-wide application of the SRM (using APHRODITE precipitation products):
The results obtained for the basin-wide application of the SRM using APHRODITE
precipitation data over the annual, snowmelt and extreme discharge periods are
presented in table 3.5. The NS coefficient value over the annual simulation period
was never less than 0.80. The best performance of the model was found over the
validation year 2003–2004 with the NS coefficient value of 0.94 and a −4.5%
difference in volume (Dv). The SRM performance over the hydrological year 2001–
2002, using APHRODITE precipitation data, is shown in Fig. 3.10. The NS coefficient
value of 0.90 was found over the entire simulation period from 2000 to 2004 (Fig.
3.11). The correlation coefficient values were never less than 0.95 and 0.85 for the
Pearson and Kendall rank correlation tests, respectively. The efficiency of the SRM
using APHRODITE precipitation products was slightly lower than that using observed
precipitation (except for the year 2003–2004), mainly for the high discharge
(snowmelt) simulation periods where the discharges are underestimated (Fig. 3.11),
but these performances remained significantly efficient and acceptable. The
underestimation of the high discharges is due to the fact that the total annual
APHRODITE precipitation is lower than the observed precipitation in the Hunza River
basin, as shown in Figs. 3.8 and 3.10.
134
2000 0
1800 5
1600 10
1400 15
Precipitation (mm)
Discharge (m3/s)
Measured discharge
1200 20
Simulated discharge
APHRODITE Precipitation data
1000 25
Nash-Sutcliffe coefficient, NS 
Difference of volume, Dv 
800 30
600 35
400 40
200 45
0 50
Oct-01
May-01
Jul-01
Nov-01
Aug-01
Sep-01
Dec-01
Feb-02
Apr-01
Mar-02
Jun-01
Jan-02
Time period (Daily)
Fig. 3.10 Evaluation of the basin-wide SRM application over the hydrological year 2001‒2002 in the Hunza River basin, by using
APHRODITE precipitation data.
135
1800
y = 0.998 x + 16.01
1600
95% prediction interval
Nash-Sutcliffe coefficient, NS 
1400 Difference of volume, Dv  
Number of data points, n 
Simulated discharge (m3/s)
1200
1000
800
600
400
200
0
0 200 400 600 800 1000 1200 1400 1600 1800
3
Measured discharge (m /s)
Fig. 3.11 Relationship between daily measured (Qm) and simulated (Qs) runoff applying the SRM basin-wide, April 2000 to March
2004 (using APHRODITE precipitation data).
136
The results obtained suggest that the SRM is a good choice to simulate the daily
discharges in the snow- and glacier-fed Hunza River basin. The simulation results
obtained using APHRODITE precipitation data suggest that this product may
effectively be used to simulate and forecast the discharges using the SRM in the
high-altitude catchments of the Himalayan region with no precipitation gauges. This
result may be associated with the fact that the discharge in the Hunza River is
influenced by the snow cover dynamics and mean temperatures of the catchment.
The results obtained by SRM application on the extreme discharges (July–August)
suggest that the overall efficiency of model simulation is satisfactory. The
underestimation of some high discharges events may be associated to the glacier
melt contribution during this period which may not be taken into account by the SRM.
The daily precipitation of the catchment does not greatly influence the discharge.
Moreover, the SRM model was developed to simulate the daily discharges in high
mountain regions, keeping in mind the well-known deficit of precipitation gauges and
the scarcity of measurements in high-altitude regions. Consequently, the SRM uses
the snow-covered area whenever possible to compute the runoff input, thus
compensating the lack of precipitation data measurements.
Impact of climate variability on Hunza River runoff: The impact of climate
variability on the Hunza River runoff was studied using the SRM under the different
future climate scenarios described in section 4. The simulated discharge in 2000 was
used as the reference for the present climate.
a) The discharges simulated under different snow cover area scenarios are presented in
Fig. 3.12. The snow cover area was assumed to increase by 10% by 2050 and 20%
by 2075 as a result of increasing precipitation while the mean temperatures of the
catchment remained constant. The impact of a 10% decrease in the snow cover area
of the catchment on the Hunza River discharge was also studied. The results
obtained indicate that the summer stream flow is likely to increase with increased
snow cover area. A 10% increase in snow cover area results in a nearly 7% (36 m 3/s)
increase in mean summer discharge and a 20% snow cover increase forces the
mean summer discharge to increase by 14% (72 m 3/s) of present mean summer
discharge (498 m3/s). The mean summer discharge decreases by 7% if the snow
cover area is assumed to decrease by 10%. This means that a 1% increase in snow
cover area in the Hunza catchment can increase the mean summer discharge up to
3.6 m3/s (0.7%) in the Hunza River.
137
1000
Discharge, Q (current SCA)

900 Discharge, Q (-10% SCA)
Discharge, Q (+10% SCA)
Discharge, Q (+20% SCA)
800
700
Discharge (m /s)
600
3
500
400
300
200
100
0

Fig. 3.12 Monthly discharge simulations in the Hunza River for different climate change (snow cover area, SCA) scenarios.
Discharge simulated for 2000 with the SRM is used as current discharge.
138
b) The mean temperature of the catchment was assumed to rise up to 2°C by 2025, 3°C
by 2050 and 4°C by 2075 (at the end of 21 st century), keeping the other variables
constant, particularly precipitation and the snow cover area. The monthly discharges
for the Hunza River simulated under these scenarios are presented in Fig. 3.13. The
results obtained indicate that a 2°C rise in temperature will result in an increase of
almost 64% (816 m3/s) in the mean summer discharge (April–September) and a 3°C
rise of temperature will force the mean summer discharge to increase almost 100%
(1010 m3/s) from the present mean summer discharge (498 m 3/s). The relationship
between the simulated discharge and the input variables used in the scenarios ―a‖
and ―b‖ appears as linear and we adopt it in order to simplify the interpretation,
without more detailed and reliable information. This result (a nearly 33% increase in
summer discharge results in a 1°C rise in mean temperatures) is different from that
estimated by Akhtar et al. (2008) and Archer (2003), who explained a 16% increase
in summer discharge resulting from a 1°C rise in temperature in the Hunza and
Shyok rivers. The discrepancy between the results obtained by different studies may
possibly be due to the methods, hypothesis and limitations under which the results
are obtained (e.g. Archer (2003) used statistical analysis and Akhtar et al. (2008)
used a different model). These results indicate that if the temperature continues to
rise in the Himalayan region with a rate estimated by the IPCC (2007), i.e., 3.7°C
until the end of the 21st century, then larger-capacity reservoirs will be needed to
manage the river flow in the UIB, contributed by the Hunza and its neighboring snow-
and glacier-fed catchments. However, these results may be specific to this particular
region because the catchment response to the climate warming may not be the same
in other catchments as explained by Null et al. (2010) and Young et al. (2009) that
the river runoff decreases with increasing temperature scenarios. This scenario will
also result in a shift of runoff period as presented in Fig. 3.13, where the discharge
increase is shifted by one month (from April to March). This fact is also explained by
Null et al. (2010) and Young et al. (2009) that the high elevation watersheds in the
southern-central region of Sierra Nevada are most susceptible to earlier runoff timing.
On the other hand, if this scenario plays itself out, then the glaciers may retreat
extensively or even disappear after a certain period of producing a high discharge, as
a result of rapid melting. The rapid melting of glaciers is also reported by Cannone et
al. (2008) who explained that many of the small glaciers (80% of total glacial
139
coverage and an important contribution to water resources) in the Alps could
disappear in the next decades under current climate warming trend. These shrinking
glaciers, surface-albedo change, and modifications to the soil-water storage
capacities may result in a different long-term hydrograph in the high elevation
catchments.
140
2000
Year 2000 (current climate)

1800 Year 2025 (T-2°C)
Year 2025 (T+2°C)
Year 2050 (T+3°C)
Year 2075 (T+4°C)
1600
Year 2075 (T+2 to +4°C,+20% SCA)
1400
1200
Discharge (m /s)
3
1000
800
600
400
200
0
Fig. 3.13 Monthly discharge simulations in the Hunza River for different climate change (mean temperature) scenarios. Discharge
simulated for 2000 with the SRM is used as current discharge.
141
c) The monthly discharge simulating a 20% increase in snow cover and a 2–4°C rise in
mean temperature is presented in Fig. 3.13. The simulation indicates a 100%
increase in the future stream flow. This result suggests that the Hunza River
discharges are highly influenced by the snow cover melt caused by the mean
temperature input in summer and not by the catchment precipitation, as explained by
Archer (2003). However, the Hunza River discharge has a positive correlation with
winter precipitation that falls in the form of snow and starts melting as the mean
temperature increases in the catchment (Tahir et al., 2011a). Moreover, 33% of the
area of the Hunza River basin is glacier-covered and contributes actively to the
snowmelt-produced runoff. This is the area that is considered to receive an
increasing amount of precipitation in the future and no significant increase in the
mean temperatures (Tahir et al., 2011a), resulting in an increased snow cover area.
This increased snow cover area will produce a large amount of runoff on melting if
the mean temperature tends to rise in the future as presented in Fig. 3.13. Moreover,
if the mean temperature and winter precipitation tend to increase until the end of
century, this may result in a balance of cryosphere area in the Karakoram region.
6. Conclusion
This study concludes that the snowmelt runoff model (SRM) based on a degree day
factor can efficiently simulate the daily discharge in the snow- and glacier-fed
catchments of the Upper Indus River Basin (UIB). The SRM’s efficiency in high-
altitude catchments can be attributed to the use of the MOD10A2 remotely sensed
cryosphere data as input to the model for water-equivalent production. This is why
the model is not likely to be affected by the well-known precipitation gauging errors in
mountainous river basins, which contribute a considerable proportion of river runoff in
the form of snowmelt. Moreover, the use of 0.25×0.25-degree gridded APHRODITE
precipitation data (APHRO_MA_V0902) in the SRM indicates that the model is not
sensitive to precipitation forcing and this product can be used for runoff simulations in
the Himalayan and Karakoram catchments, where the hydrometeorological
measurements are scarce or not available. The SRM overcomes the problem of
precipitation catch deficit by using the snow cover area whenever possible for
computing the runoff input (Martinec et al., 2007). In fact, SRM counts the
precipitation input twice during the simulation: once in the form of snow cover input
and then in the form of precipitation data input. This is why the problem of
142
underestimated APHRODITE precipitation data can be efficiently overcome by using
the SRM on the snow- and glacier-fed catchments.
The analysis of the climate change impact indicated that the basin’s hydrology will
alter under different climate change scenarios. This change apparently results from
the increasing temperature, which will increase the volume of summer stream flow in
the region. The availability of the snowmelt runoff seems to shift accordingly with the
mean temperature rise in the spring. This analysis suggests that new reservoirs
(such as the projected Diamer Bhasha Dam that is planned on the Indus River about
315 km upstream from Tarbela Dam) (Fig. 3.2) are required for the summer flow
storage to meet the needs for supplementary irrigation supplies during the low-flow
period in the downstream regions, increasing power-generation demand, flood
control and water supply. Though the climate scenarios used in this study were
assumed or derived from previous studies, they provide useful information to manage
water resources for the country’s future needs.
The application of the SRM to the other snowmelt runoff-dominated sub-catchments
of the UIB may help manage the integrated water resources of the Indus River
irrigation system. The next step of the study is to apply the SRM to the neighboring
snow- and glacier-fed catchments – Shigar, Shyok, Astore, Gilgit and Kharmong (Fig.
3.2) – to simulate and forecast the summer stream flow as well as to evaluate the
impact of future climate scenarios in these Himalayan regions.
Acknowledgements
Adnan Ahmad Tahir was financially supported by the Higher Education Commission
of Pakistan within the framework of a France-Pakistan collaboration program for
overseas studies. This financial support is gratefully acknowledged and appreciated.
The authors extend their thanks to the Water and Power Development Authority
(WAPDA) for contributing their hydrological and meteorological data. The authors
also wish to thank NASA and Japan's Ministry of the Economy, Trade and Industry
(METI) for providing ASTER GDEM.
Ralph Roberts, Computer Specialist at the USDA-ARS Hydrology and Remote-
sensing Lab, provided valuable information to solve the SRM installation problems.
143
Annexes to the article:
Annex 1: Snowmelt Runoff Model (SRM) equation
Some detail of the notations for the equation used by the SRM model (equation 3.3)
is given here.
Daily discharge calculated by the model on the th day.
Measured discharge input value. This measured value was put for the first day
of each hydrological year to run the model.
Degree-day factor, DDF (cm.°C−1 d−1) indicating the snowmelt depth resulting
from 1 degree-day.
A = Area of the basin or zone [km²].
∆T = Adjustment by temperature lapse rate when extrapolating the temperature from
the station to the average hypsometric elevation of the basin or zone [°C·d].
P = Precipitation contributing to runoff [cm]. A preselected threshold temperature,
TCRIT, determines whether this contribution is rainfall and immediate. If precipitation is
determined by TCRIT to be new snow, it is kept on storage over the hitherto snow free
area until melting conditions occur.
k = Recession coefficient (Xc and Yc) indicating the decline of discharge in a period
without snowmelt or rainfall.
= Sequence of days during the discharge computation period. Equation (3.3) is
written for a time lag between the daily temperature cycle and the resulting discharge
cycle of 18 hours. In this case, the number of degree-days measured on the nth day
corresponds to the discharge on the day.
Conversion from cm-km².d−1 to m3.s−1.
Annex 2: Discharge simulated by SRM over 4 years

The observed discharge in the Hunza River at Dainyor bridge and the discharge
simulated by the application of snowmelt runoff model (SRM) basin wide (with
observed and APHRODITE precipitation data) and zone-wise, is presented in the
form of an annex to this research article. This figure was not presented in the article
to control the length of manuscript.
144
ANNEX 2 TO RESEARCH ARTICLE
Basin wide (with observed and APHRODITE precipitation data) and zone-wise application of snowmelt runoff model (SRM) in the
Hunza River basin, over the period 2000‒2004.
1800
Measured discharge
Simulated discharge (Zone wise)
1600 Simulated discharge (Basin wide)
BW simulated discharge (with APHRODITE data)
1400
1200
Discharge (m /s)
1000
3
800
600
400
200
0
2000
02
03
04
05
06
07
08
09
10
11
12
2001
02
03
04
05
06
07
08
09
10
11
12
2002
02
03
04
05
06
07
08
09
10
11
12
2003
02
03
04
05
06
07
08
09
10
11
12
2004
Time period (Daily)
145
3.4 Application of snowmelt runoff model in the other snow and glacier fed sub-
catchments of the Upper Indus River Basin
The satisfactory results of discharge simulation in Hunza River basin, by using SRM
suggest the application of model in all the snow and glacier fed sub-catchments (Fig.
3.14) of the Upper Indus River Basin.
The data for daily stream flow, climate variables (daily precipitation and temperature),
topographic characteristics (catchment area, hypsometric curve and mean elevation)
available from each sub-catchment, and the snow cover area estimated for each sub-
catchment was used as input. The snow cover area was estimated by using
MOD10A2 (MODIS snow cover products) satellite images for each sub-catchment of
UIB. The catchment area, hypsometric curve and mean elevation was calculated
from the ASTER global digital elevation model (GDEM) for each sub-catchment. This
section presents the hydro-meteorological investigations in these sub-catchments
and the application of SRM to simulate the discharge. The SRM was applied basin
wide on each of the sub-catchment for the discharge simulation on a period of 2-
hydrological years (2000‒01 and 2003‒04). The methodology of SRM application is
explained already in this chapter in the form of a research article. A brief description
for each of the sub-catchments of UIB and SRM application results are presented in
the next sections.
146
Fig. 3.14 Location of the snow and glacier fed sub-catchments (Hunza, Gilgit, Astore, Shyok and Shigar) of the Upper Indus River
Basin.
147
3.4.1 Gilgit River basin
The Gilgit River basin (drainage area, 12,660 km2) (Fig. 3.15), is situated in the high-
altitude HKH region, with a mean catchment elevation of 3997 m. The hypsometric
curve (estimated from the GDEM) and the area distribution under 500 m-altitudinal
layers in the catchment are shown in Fig. 3.16. Three climate stations with
station just outside the catchment boundary, as shown in Fig. 3.15. The Gupis and
Gilgit climate stations are installed and managed by the PMD and have a quite long
data series of more than 50 years (1951 to 2007). The Ushkor and Yasin climate
stations are installed at the high altitudes by WAPDA and the data record is available
from 1999 to 2008 for these stations. The temperature lapse rate value ranges from
0.43°C/100m to 1.06°C/100m between the lower to the higher altitude climate
stations. The mean total annual precipitation is 314 mm at Yasin (2487 m), 311 mm
at Ushkor (3765 m), 186 mm at Gupis (2156 m) and 132 mm at Gilgit (1460 m)
according to the available data records of the four climate stations in the catchment.
The Gilgit River has a mean annual flow of 283 m 3/s (i.e., 704 mm of water depth
equivalent) gauged at Gilgit hydrological station (1430 m) (Fig. 3.15), according to the
SWHP-WAPDA 38-year (1970–2008) flow record. Obviously, present precipitation
records are not representative of the runoff at the outlet because of no climate data
records at altitudes above than 4000 m in this basin and the well-known gauging
errors at high altitudes, as explained in detail by Sevruk (1985, 1989) and Førland et
al. (1996). The Gilgit river flow in summer (June to September) is influenced
significantly by the snow and glacier melt (Fig. 3.17). The snow cover area (SCA) for
the Gilgit River basin is estimated from the MODIS snow cover images (MOD10A2)
over the hydrological year 2000‒01 and 2003‒04. The average SCA varies from
approximately 85% in winter to 10% in summer (Fig. 3.18). Approximately 888 km² of
catchment area is at an elevation above 5000 m (Fig. 3.16) and almost the same
area (8%) is glaciated as calculated from the GLIMS (Global Land Ice Measurements
from Space) shape file of the Upper Indus Basin.
148
Fig. 3.15 Location of the Gilgit River basin and its GDEM (global digital elevation model).
149
0 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
8000
7500
7000
6500
6000
5500
5000
Elevation (m)
4500
4000
3500
3000
2500
2000
1500
1000
0
0
1000
1500
2000
2500
3000
3500
4000
4500
5000
5500
6000
6500
7000
7500
8000
8500
9000
9500
500
10000
10500
11000
11500
12000
12500
13000
13500
14000
Cumulated area (km2)
Fig. 3.16 Hypsometric curve of the Gilgit River basin with the area distribution elevation bands of 500 m.
150
Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar
1200 0
9.69%
14.27%
1100 10
23.69%
1000 20
37.41%
900 30
800 40
53.67%
Precipitation (mm)
Discharge (m /s)
700 61.63% 50
3
65.26%
600 60
77.69%
78.05%
500 70
80.92%
84.77%
86.05%
400 80
300 90
Precipitation at Gupis ()
Precipitation at Yasin ()
200 100
Monthly SCA ()
Discharge ()
100 110
0 120
Fig. 3.17 Hydrogram presenting mean monthly precipitation (Gupis and Yasin climate station), discharge (Gilgit) and snow cover
area, SCA (estimated by MODIS snow product) in the Gilgit River basin.
151
100
90
80
70
60
50
40
30
20
10 Year 2000-01
Year 2003-04
Average SCA
0
Time period (Daily)
Fig. 3.18 Snow cover area, SCA (estimated from MODIS snow product) distribution in the Gilgit River basin over the hydrological
year 2000‒01 and 2003‒04.
152
Snowmelt runoff modeling
The snowmelt runoff model (SRM) was calibrated by estimating the parametric
values (Table 3.7) for the Gilgit River basin. The snow cover area percentage,
discharge values measured at Gilgit and the average of climate data (precipitation,
mean temperature) observed at the four climate stations were used as input
variables. A temperature lapse rate value of 0.43 °C/100m, observed between the
lowest elevation climate station (Gilgit) and the highest elevation climate station
(Ushkor) of the Gilgit river basin was used for the basin wide discharge simulation.
The results obtained are presented in Fig. 3.19 & Fig. 3.20.
Parameter Parameter values
Lapse rate (°C/100m) 0.43
Tcric (°C) 0
DDF (cm. °C−1d−1) 0.55‒0.80
Lag Time (hrs) 18
Cs 0.03‒0.35
Cr 0.03‒0.35
1 (May-October)
RCA
0 (November-April)
0.95 (June-July)
Xc
1.06 (August-May)
Yc 0.01‒0.1
Table 3.7 SRM Parameter values for basin-wide application in Gilgit River basin.
The results obtained suggest that SRM has efficiently simulated the Gilgit River
discharges both in the snow accumulation and snow melt periods.
153
1000
Measured discharge
Simulated discharge
900
Nash-Sutcliffe correlation coefficient  
Difference of volume, Dv  
800
700
Discharge (m3/s)
600
500
400
300
200
100
0
Apr-00
Jun-00
Jul-00
May-00
Aug-00
Sep-00
Oct-00
Nov-00
Dec-00
Jan-01
Feb-01
Mar-01
Time period (Daily)
Fig. 3.19 Evaluation of basin-wide (BW) SRM application over the hydrological year 2000–2001 in the Gilgit River basin.
154
1800
Measured discharge
1700 Simulated discharge
1600
1500 Nash-Sutcliffe correlation coefficient  
Difference of volume, Dv  
1400
1300
1200
Discharge (m3/s)
1100
1000
900
800
700
600
500
400
300
200
100
0
May-03
Oct-03
Apr-03
Jul-03
Nov-03
Feb-04
Aug-03
Sep-03
Dec-03
Mar-04
Jun-03
Jan-04
Time period (Daily)
Fig. 3.20 Evaluation of basin-wide (BW) SRM application over the hydrological year 2003–2004 in the Gilgit River basin.
155
3.4.2 Astore River basin
The Astore River basin (drainage area, 3,990 km 2) (Fig. 3.21), is situated in the high-
altitude HKH region, with a mean catchment elevation of 4285 m. The hypsometric
curve (estimated from the GDEM) and the area distribution under 500 m-altitudinal
layers in the catchment are shown in Fig. 3.22. Three climate stations with
station just outside the catchment boundary, as shown in Fig. 3.21. The Astore
climate station is installed and managed by the PMD and have a quite long data
series of more than 50 years (1954 to 2007). The Rama and Rattu climate stations
are installed at the high altitudes by WAPDA and the data record is available from
1995 to 2008 for these stations. The temperature lapse rate value ranges from
0.47°C/100m to 0.75°C/100m between the lower to the higher altitude climate
stations. Astore basin is situated at the southern foothill of Himalaya and receives a
major proportion of annual rainfall in summer caused by the monsoon regime. The
mean total annual precipitation is 794 mm at Rama (3000 m), 723 mm at Rattu (2570
m) and 501 mm at Astore (2168 m) according to the available data records of the four
climate stations in the catchment. The Astore River has a mean annual flow of 137
m3/s (i.e., 1084 mm of water depth equivalent) gauged at Doyian hydrometric station
(1583 m) (Fig. 3.21), according to the SWHP-WAPDA 33-year (1974–2007) flow
record. The present precipitation records are not representative of the runoff at the
outlet because of no climate data records at altitudes above than 4000 m in this
basin and the well-known gauging errors as explained previously for the Gilgit River
basin. The Astore river flow in summer (June to September) is influenced significantly
by the snow and glacier melt (Fig. 3.23). The snow cover area (SCA) for the Astore
River basin is estimated from the MODIS snow cover images (MOD10A2) over the
hydrological year 2000‒01 and 2003‒04. The average SCA varies from
approximately 95% in winter to 7% in summer (Fig. 3.24). Approximately 154 km² of
catchment area is at an elevation above 5000 m (Fig. 3.22) and almost the same
area (7%) is glaciated as calculated from the GLIMS shape file of the Upper Indus
Basin.
156
Fig. 3.21 Location of the Astore River basin and its GDEM (global digital elevation model).
157
0 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
9000
8500
8000
7500
7000
6500
6000
5500
Elevation (m)
5000
4500
4000
3500
3000
2500
2000
1500
1000 Area per 500-m elevation band
0
0 500 1000 1500 2000 2500 3000 3500 4000
Fig. 3.22 Hypsometric curve of the Astore River basin with the area distribution elevation bands of 500 m.
158
500 0
6.96%
12.06%
18.16%
450 20
38.01%
49.43%
400 40
65.32%
69.92%
350 60
82.46%
86.98%
90.37%
92.02%
Precipitation (mm)
93.63%
Discharge (m /s)
300 80
3
250 100
200 120
150 140
Precipitation at Astore ()
Precipitation at Rattu ()
100 Monthly SCA () 160
Discharge ()
50 180
0 200

Fig. 3.23 Hydrogram presenting mean monthly precipitation (Astore and Rattu climate station), discharge (Doyian) and snow cover
area, SCA (estimated by MODIS snow product) in the Astore River basin.
159
100
90
80
70
60
50
40
30
20
Year 2001-02
10 Year 2003-04
Avergae SCA
0
Time period (Daily)
Fig. 3.24 Snow cover area, SCA (estimated from MODIS snow product) distribution in the Astore River basin over the hydrological
year 2000‒01 and 2003‒04.
160
values (Table 3.8) for the Astore River basin. The snow cover area percentage,
discharge values measured at Doyian and the average of climate data (precipitation,
mean temperature), observed at the three climate stations situated within the Astore
River basin, were used as input variables. A temperature lapse rate value of
0.61°C/100m, observed between the lowest elevation climate station (Astore) and the
highest elevation climate station (Rama) of the Astore river basin was used for the
basin wide discharge simulation. The results obtained are presented in Fig. 3.25 &
Fig. 3.26.
Tcric (°C) 0
DDF (cm. °C−1d−1) 0.65‒0.90
Lag Time (hrs) 12
Cs 0.03‒0.45
Cr 0.03‒0.45
1 (May-October)
RCA
0 (November-April)
Xc 1.06
Yc 0.005‒0.09
Table 3.8 SRM Parameter values for basin-wide application in Astore River basin.
The obtained results, suggest that SRM has efficiently simulated the Astore River
161
400
Measured discharge
Simulated discharge
350
Nash-Sutcliffe correlation coefficient 
Difference of volume, Dv 
300
Discharge (m /s)
250
3
200
150
100
50
0
Oct-00
May-00
Jul-00
Apr-00
Nov-00
Dec-00
Feb-01
Mar-01
Jun-00
Aug-00
Sep-00
Jan-01
Time period (Daily)
Fig. 3.25 Evaluation of basin-wide (BW) SRM application over the hydrological year 2000–2001 in the Astore River basin.
162
800
Measured discharge
Simulated discharge
700
Nash-Sutcliffe correlation coefficient 
Difference of volume, Dv 
600
500
Discharge (m /s)
3
400
300
200
100
0
Jul-03
Oct-03
Nov-03
Dec-03
May-03
Jun-03
Aug-03
Sep-03
Jan-04
Feb-04
Apr-03
Mar-04
Time period (Daily)
Fig. 3.26 Evaluation of basin-wide (BW) SRM application over the hydrological year 2003–2004 in the Astore River basin.
163
3.4.3 Shyok River basin
The Shyok River basin (drainage area, 68,458 km 2) (Fig. 3.27), is situated in the
high-altitude HKH region, with a mean catchment elevation of 4939 m. This
catchment is situated in three different geographical regions i.e. north-western part of
the catchment is in Pakistan, the middle part is in Ladakh (disputed territory of
Jammu and Kashmir) whereas the relatively south-eastern part is situated in the
Tibetan region of China. Approximately 58% of catchment area is at an elevation
above 5000 m (Fig. 3.28). The hypsometric curve (estimated from the GDEM) and
the area distribution under 500 m-altitudinal layers in the catchment are shown in Fig.
3.28. There is one climate station (Hushey) with precipitation gauge, installed at an
altitude of approximately 3000 m within the catchment and one station (Skardu)
(2317 m) just outside the catchment boundary, as shown in Fig. 3.27. The Skardu
climate station is installed and managed by the PMD and have a quite long data
series of more than 50 years (1952 to 2007). The Hushey climate station is installed
at the high altitude by WAPDA and the data record is available from 1999 to 2008 for
this station. The mean total annual precipitation is 440 mm at Hushey (3000 m) and
230 mm at Skardu (2317 m) according to the available data records of these two
climate stations. The Shyok River has a mean annual flow of 358 m3/s (i.e., 165 mm
of water depth equivalent) gauged at Yogo hydrometric station (2469 m) (Fig. 3.27),
according to the SWHP-WAPDA 33-year (1972–2008) flow record. The Shyok river
flow in summer (June to September) is influenced significantly by the snow and
glacier melt (Fig. 3.29). The snow cover area (SCA) for the Shyok River basin is
estimated from the MODIS snow cover images (MOD10A2) over the hydrological
year 2000‒01 and 2003‒04. The average SCA varies from approximately 50% in
winter to 15% in summer (Fig. 3.30). Almost 38% (25,000 km²) of the total catchment
area is under glacier cover as extracted from the GLIMS glacier cover shape file of
the UIB by using ARCMAP tool. The glacier concentration is much more in the
central part of the catchment (Karakoram region of Ladakh and north-eastern
Pakistan) where it has many peaks above an altitude of 7000 m with the well known
Siachin glacier (area = 1112 km²) in the same region (Roohi, 2006).
164
Fig. 3.27 Location of the Shyok River basin and its GDEM (global digital elevation model).
165
0 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
8500
8000
7500
7000
6500
6000
5500
5000
Elevation (m)
4500
4000
3500
3000
2500
2000
1500
1000
0
0
10000
12500
15000
17500
20000
22500
25000
27500
30000
32500
35000
37500
40000
42500
45000
47500
50000
52500
55000
57500
60000
62500
65000
67500
70000
2500
5000
7500
Fig. 3.28 Hypsometric curve of the Shyok River basin with the area distribution elevation bands of 500 m.
166
2000 0
1900
14.5%
15.7%
1800 10
18.1%
21.6%
1700
25.5%
26.6%
1600 20
1500
1400 30
38.6%
39.6%
1300
42.5%
42.7%
Precipitation (mm)
45.0%
Discharge (m /s)
1200 40
47.1%
3
1100
1000 50
900
800 60
700 Precipitation at Skardu ()
Precipitation at Hushey ()
600 Monthly SCA () 70
500 Discharge ()
400 80
300
200 90
100
0 100
Fig. 3.29 Hydrogram presenting mean monthly precipitation (Skardu and Hushey climate station), discharge (Yogo) and snow cover
area, SCA (estimated by MODIS snow product) in the Shyok River basin.
167
70
60
50
40
30
20
10
Year 2000-01
Year 2003-04
Average SCA
0
Time period (Daily)

Fig. 3.30 Snow cover area, SCA (estimated from MODIS snow product) distribution in the Shyok River basin over the hydrological
year 2000‒01 and 2003‒04.
168
values (Table 3.9) for the Shyok River basin. The snow cover area percentage,
discharge values measured at Yogo and the average of climate data (precipitation,
mean temperature), observed at two climate stations (Hushey and Skardu) were
used as input variables. A temperature lapse rate value of 0.75°C/100m, observed
between the lower elevation climate station (Skardu) and the higher elevation climate
station (Hushey) of the Shyok river basin was used for the basin wide discharge
simulation. The results obtained are presented in Fig. 3.31 & Fig. 3.32.
Tcric (°C) 0
DDF (cm. °C−1d−1) 0.35
Lag Time (hrs) 18
Cs 0.03‒0.25
Cr 0.05‒0.15
1 (May-October)
RCA
0 (November-April)
Xc 0.80‒1.06
Yc 0.005‒0.03
Table 3.9 SRM Parameter values for basin-wide application in Shyok River basin.
The obtained results, suggest that SRM has efficiently simulated the Shyok River
169
4000
Measured discharge
Simulated discharge
3500 Nash-Sutcliffe correlation coefficient 
Difference of volume, Dv 
3000
2500
Discharge (m /s)
3
2000
1500
1000
500
Oct-00
Apr-00
Aug-00
Sep-00
Nov-00
Dec-00
Jun-00
Jul-00
Jan-01
Feb-01
Mar-01
May-00
Time period (Daily)

Fig. 3.31 Evaluation of basin-wide (BW) SRM application over the hydrological year 2000–2001 in the Shyok River basin.
170
2400
Measured discharge
2200 Simulated discharge
Nash-Sutcliffe correlation coefficient 
2000 Difference of volume, Dv 
1800
1600
Discharge (m /s)
1400
3
1200
1000
800
600
400
200
Oct-03
Apr-03
May-03
Jul-03
Nov-03
Aug-03
Sep-03
Dec-03
Feb-04
Mar-04
Jun-03
Jan-04
Time period (Daily)
Fig. 3.32 Evaluation of basin-wide (BW) SRM application over the hydrological year 2003–2004 in the Shyok River basin.
171
3.5 Comparison of Hunza, Gilgit, Astore and Shyok sub-catchments
A comparison of the main characteristics (Table 3.10) and efficiency of SRM to model
the snowmelt runoff (Table 3.10 and Fig. 3.34) for 4 sub-catchments (Hunza, Gilgit,
Astore and Shyok), situated in the Upper Indus River Basin is discussed in next
sections.
172
Hunza River basin Gilgit River Astore River Shyok River
basin basin basin
River flow gauging
Dainyor Bridge Gilgit Doyian Yogo
station
Latitude 35° 56' N 35° 56' N 35° 33' N 35° 11' N
Longitude 74° 23' E 74° 19' E 74° 42' E 76° 06' E
Elevation of 1450 m 1430 m 1583 m 2469 m
gauging station
2 2 2 2
Drainage area 13,733 km 12,660 km 3,990 km 68,458 km
2 2 2 2
Glacier-covered area 4,688 km 1,013 km 200 km 25,000 km
(Glacier cover for
each basin is
extracted from the
GLIMS shape file of
UIB glacier cover by
using ARCMAP)
Glacier cover
34% 8% 7% 38%
percentage
Mean elevation 4,631 m 3,997 m 4,285 m 4,939 m
(computed from
ASTER GDEM)
Area above 5000 m 32.5 % 7% 4% 58%
No. of 3 (Naltar, Ziarat 4 (Gilgit, Gupis, 3 (Astore, Rattu 1 (Hushey)

meteorological and Khunjerab) Yasin and and Rama)
stations Ushkor)
Mean annual runoff 742 mm 704 mm 1084 mm 165 mm
Total annual 680 mm at Naltar 132 mm at Gilgit 501 mm at 440 mm at

precipitation 225 mm at Ziarat 186 mm at Gupis Astore Hushey
170 mm at 314 mm at Yasin 723 mm at Rattu
Khunjerab 311 mm at 794 mm at
Ushkor Rama
Snow cover area 30%‒80% 10%‒85% 7%‒95% 15%‒50%
SRM efficiency: NS Dv (%) NS Dv (%) NS Dv (%) NS Dv (%)
Year 2000‒01 0.93 −0.7 0.90 5 0.89 −7 0.92 5
Year 2003‒04 0.86 12 0.92 −4 0.93 10 0.94 −8
Table 3.10 A summary of the characteristics for four sub-catchments (Hunza, Gilgit,
Astore and Shyok) of the Upper Indus River Basin and a comparison of SRM efficiency.
173
3.5.1 Comparison of the main characteristics:
A comparison of the characteristics show that the catchment with much area above
5000 m has greater glacier covered area than the other catchments. As given in table
3.10, Shyok basin has more area (58%) above an elevation of 5000 m.a.s.l. (meter
above sea level) than the other 3 catchments so it has the greater glacier covered
area (38%) following by the glacier cover of Hunza (34%), Gilgit (8%) and Astore
(7%), respectively. Most of the glaciers of Shyok basin are present in the Karakoram
ranges of Pakistan and Ladakh as explained in section 3.4.3, whereas the most of
the catchment area situated in the Tibetan region has an altitude less than 5000 m
(Fig. 3.27) hence claims less concentration of glaciers. Although the mean elevation
of Astore basin (4285 m) is greater than the mean elevation of Gilgit basin (3997 m)
but its glacier covered area is less than the Gilgit basin. It may be associated to the
fact that Astore basin has less area percentage over 5000 m.a.s.l. than the Gilgit
basin. It indicates that the presence of glacier covered area in a Himalayan
catchment is proportional to the percentage of area higher than 5000 m.a.s.l. in a
catchment, but the results should be verified in the eastern Himalayan catchments. It
may be different in the other regions of the world depending on the climate conditions
of that region.
The maximum snow cover area (Table 3.10) is almost same in the Hunza and Gilgit
because the two catchments receive almost the same amount of winter precipitation
due to westerly circulations in the central Karakoram region. The minimum snow
cover area in Hunza (30%) is greater than that in Gilgit (10%), Astore (7%) and
Shyok (15%). It may be associated to the mean spring and summer temperatures
which are lower in Hunza than other sub-catchment of UIB (Fig. 3.33). The change in
snow cover area of Astore basin from maximum (95%) to minimum (7%) may be
attributed to the smaller surface area than other catchments of UIB. The smaller
surface area can be covered by snow cover in winter more rapidly than the bigger
areas and the surface area comes out of snow in summer more rapidly than in bigger
areas. The change in snow cover area of Shyok basin is quite surprising where the
snow cover area reaches from 15% up to 50%. Maximum snow cover area (50%) is
quite less than the central Karakoram catchments, which indicates the less effect of
winter westerly circulations in the Tibetan part or western Karakoram region and
hence the Shyok catchment receives less snowfall in winter than the central
Karakoram catchments. Even that the Shyok basin has greater area above 5000
174
m.a.s.l. than Hunza basin but still the minimum snow cover area (15%) is less than
the Hunza basin (30%). It may be associated to the rain shadow effect of western
Himalaya to Tibetan region. As the south-eastern Shyok catchment lies at the
northern foothill of the Himalaya, so it does not receive much summer monsoon
rainfall due to high Himalayan barrier hence reducing the effect of albedos due to
less cloudiness. The greater sunshine hours in summer may be the cause of
depleting snow cover area in the Shyok River basin.
The mean monthly temperatures (Fig. 3.33) at the mean altitude of each catchment
are estimated from the data available at the climate station nearest to this mean
catchment elevation by using the lapse rate values of each basin. The mean
temperatures in Hunza and Shyok basin at mean catchment altitudes remain
negative during almost 8 months (Fig. 3.33) which may be the reason of larger
glacier areas in these two catchments. The mean temperature in Astore basin
remains negative during almost 7 months but it has a smaller glacier area than
Hunza and Shyok because of the mean catchment elevation lower than these two
catchments (Hunza and Shyok). Gilgit has the highest mean temperatures of these
four sub-catchments mainly because it has a mean catchment elevation much lower
than the other three catchments (almost 1000 m lower than Shyok, 600 m lower than
Hunza and 300 m lower than the mean elevation of Astore basin).
Amount of total annual precipitation (average of all climate stations situated within the
basin) received by Astore basin is greater than other 3 catchments following by
Shyok, Hunza and Gilgit. It can be explained by the fact that Astore is situated on the
southern foothills of Himalaya so receives the much more summer monsoon rainfall
than the other catchments. Eventually, it generates more runoff than other 3
catchments. The other explanation for larger runoff generation is that a catchment
with smaller surface area has smaller evaporation (if the mean temperatures are not
greater than other comparative catchments) and other runoff losses than the
catchments with larger surface areas, due to small lag time especially in the
mountainous catchments.
175
20
15 Gilgit (Mean catchment elevation = 3997)

Astore (Mean catchment elevation = 4285)
Hunza (Mean catchment elevation = 4631)
10 Shyok (Mean catchment elevation = 4939)
Mean temperature (C)
-5
-10
-15
-20
Fig. 3.33 Mean monthly temperature estimated at the mean catchment altitude by using the catchment lapse rate values in the 4
sub-catchments of UIB.
176
3.5.2 Comparison of the SRM efficiency for four sub-catchments of UIB:
A comparison of Snowmelt Runoff Model (SRM) efficiency (Nash-Sutcliffe, NS
coefficient and volume difference, Dv) for the hydrological years 2000‒01 and
2003‒04 is presented in table 3.10. A correlation between the measured and
modeled discharge for each of catchment, over two hydrological years period is
shown in Fig. 3.34. The SRM model has efficiently simulated the discharges in all the
sub-catchments of UIB, studied in this chapter. A maximum NS coefficient value
(0.93) for the hydrological year 2000‒01 was found to simulate the stream flows in
the Hunza River basin (Table 3.10). A maximum NS coefficient value of 0.94 was
found for the year 2003‒04 to simulate the discharge of Shyok River basin (Table
3.10). The high discharges are under-estimated in the Hunza and Astore River basin,
but over-estimated in Gilgit River basin (Fig. 3.34). Overall, the model performance
was well with a minimum correlation coefficient value of 0.90 (Fig. 3.34), over entire
period in all the sub-catchments. These results obtained by the simulation of stream
flows suggest that the use of model SRM in the high altitude snow and glacier fed
sub-catchments of UIB for future flow prediction may be useful for the water
resources management, taking carefully into consideration the under and over-
estimations for the highest values of discharges.
177
1800 1800
Hunza River basin Linear regression line Gilgit River basin
1600 95% confidence interval line 1600
Simulated discharge (m /s)

1400 1400
3
3
1200 1200
1000 1000
800 800
600 600
400 400
200 Nash-Sutcliffe correlation coefficient   200 Nash-Sutcliffe correlation coefficient  
Difference of volume, Dv   Difference of volume, Dv  
0 0
Number of data points Number of data points
0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800
Measured discharge (m3/s) Measured discharge (m3/s)
800 4000
Astore River basin Shyok River basin
700
3500
3
600

3000
3
500
2500
400
2000
300
1500
200
1000
100 Nash-Sutcliffe correlation coefficient   Nash-Sutcliffe correlation coefficient  
500
0 Difference of volume, Dv   Difference of volume, Dv  
Number of data points 0 Number of data points
0 100 200 300 400 500 600 700 800 0 500 1000 1500 2000 2500 3000 3500 4000
Measured discharge (m3/s) Measured discharge (m3/s)
Fig. 3.34 Comparison of correlation between measured and simulated discharges, over two hydrological years 2000‒01 and
2003‒04 in each of four sub-catchments of UIB.
178
3.6 Conclusion
The following conclusions can be drawn by analysing the results obtained in this
chapter:
The snowmelt runoff model (SRM), based on a degree day factor, has efficiently
simulate the daily discharge in the snow- and glacier-fed catchments of the Upper
Indus River Basin (UIB). The SRM’s efficiency in high-altitude catchments can be
attributed to the use of the MOD10A2 remotely sensed cryosphere data as input to
the model for water-equivalent production.
The analysis of the climate change impact indicated that the Hunza River basin’s
hydrology will alter under different climate change scenarios. This change apparently
results from the increasing temperature, which will increase the volume of summer
stream flow in the region. The availability of the snowmelt runoff seems to shift
accordingly with the mean temperature rise in the spring. This analysis suggests that
new reservoirs are required for the summer flow storage to meet the needs for
supplementary irrigation supplies during the low-flow period in the downstream
regions, increasing power-generation demand, flood control and water supply.
Though the climate scenarios used in this study were assumed or derived from
previous studies, they provide useful information to manage water resources for the
country’s future needs.
The successful application of the SRM in the Hunza River basin suggested its
application to the other snowmelt runoff-dominated sub-catchments of the UIB. The
snowmelt runoff model was applied to three other sub-catchments (Gilgit, Astore and
Shyok) of the UIB. These catchments were selected because of their considerable
contribution to the Upper Indus flow. The results obtained by this analysis are
presented in the section 3.4 of this chapter. The results are quite satisfactory and
suggest that the application of SRM to these snow and glacier fed sub-catchments
may help to manage the integrated water resources of the Indus River irrigation
system.
These successful results have allowed us to simulate the daily flows in all the major
tributaries of Upper Indus River under different climate scenarios. These simulated
stream flows are then used to assess the water resources management under
present and future climate scenarios, as will be presented in the next chapter.
179
Résumé étendu du « CHAPITRE 4 »
DURABILITE DES RESSOURCES EN EAU DANS LE BASSIN DE
L’INDUS
Le système d’irrigation du bassin de l’Indus (Indus Basin Irrigation System IBIS), qui
inclut les réservoirs majeurs de Tarbela sur l’Indus et de Mangla sur la Jhelum, ainsi
qu’un réseau de barrages et de canaux, est le plus grand système intégré d’irrigation
de la terre. La plupart de l’eau de ce système d’irrigation vient de la fonte de la neige
et des glaciers des chaînes de montagne de l’Hindukush-Karakoram-Himalaya. Les
autres sources d’approvisionnement en eau sont constituées par les précipitations de
mousson l’été et par le pompage dans les nappes. Tout changement dans les
ressources de l’UIB dus à la variabilité climatique ou à des facteurs socio-
économiques aura un impact sérieux sur la sécurité alimentaire et la production
énergétique. La demande en eau pour les usages urbains et agricoles va encore
croître dans le futur à cause de la rapide croissance de la population. Avec la
tendance croissante des débits de tous les affluents du haut Indus d’après les
scénarios de changement climatique, les réservoirs actuels ne sont pas capables de
stocker les apports futurs attendus. Le défi est plus grand encore avec une mauvaise
gestion des ressources en eau, la dégradation de l’environnement (en particuliers
dans les bassins) et une réduction accélérée des capacités actuelles de stockage
due à la sédimentation. Toutes ces raisons peuvent conduire à des crues
catastrophiques en été à l’époque des hautes eaux et à des coupures
d’approvisionnement en hiver et au printemps pour les cultures Rabi qui sont les plus
rentables.
Il est par conséquent de la plus grande importance d’étudier minutieusement la
construction d’au moins 2 ou 3 barrages multi-usages de grande capacité, et plus
particulièrement les projets de barrages de Diamer Basha et de Kalabagh (Fig. 4.1)
sur l’Indus. La priorité est en effet de retenir les débits de hautes eaux en été afin de
satisfaire la demande d’irrigation durant les périodes de basses eaux, tout en
produisant de l’énergie hydro-électrique à bas coût pour les usages domestiques et
industriels croissants. Ces projets doivent aussi être particulièrement attentif aux
impacts environnementaux et aux intérêts des populations locales.
Une autre nécessité et de développer les connaissances et les données
d’observation sur les environnements montagneux d’où viennent la plus grande
180
partie de la ressource en eau. Cela passe par une amélioration de l’éducation et une
plus large dissémination de la connaissance scientifique et de ses applications. Une
grande motivation est nécessaire à l’échelle nationale pour mettre en place un
programme de 10 à 15 ans afin d’améliorer la gestion d’IBIS et de gérer la
construction des nouveaux réservoirs. Des techniques de modélisations pour la
gestion des ressources en eau doivent être employées pour augmenter l’efficacité du
grand IBIS. Les modèles d’approvisionnement conventionels ne sont plus adapté. Au
cours de la dernière décade, une approche intégrée de développement hydraulique a
émergé qui positionne les projets d’approvisionnement en eau directement dans le
contexte des demandeurs, prenant en compte les questions de qualité d’eau et de
préservation des écosystèmes.
L’outil WEAP (Water Evaluation and Planning System), développé par Raskin et al.
(1992, Stockholm Environment Institute's (SEI), U.S. Center) a pour objectif
d’incorporer ces principes dans un cadre pratique de planification des ressources en
eau. Son application réussie dans plusieurs régions du monde (par exemple : Mehta
et al. (2011), Yilmaz and Harmancioglu (2010) and Mulligan, M. et al. (2011))
encourage à l’utiliser pour la planification des ressources en eau dans l’UIB.
181
CHAPTER 4
SUSTAINABILITY OF WATER RESOURCES IN THE INDUS BASIN
4.1 Indus basin water resources and their confronting challenges

A country with its economy based on agriculture, Pakistan is highly dependent on
water resources of the Upper Indus Basin originating in the Himalaya-Karakoram-
Hindukush (HKH) mountain sources for irrigated agriculture which is the main key of
its economy. The agricultural sector not only meets the food demand of the growing
population but also provides the raw materials for the industrial sector. These water
resources of HKH are used through the Indus Basin Irrigation System (IBIS) (Fig. 4.1)
for the agricultural zones of plane lands. The IBIS, comprising the major storage
reservoirs of Tarbela on the Indus and Mangla on the Jhelum (Fig. 4.1) and a
network of barrages and canals, is now the largest integrated irrigation system in the
world (Archer et al., 2010). A major part of the water for this irrigation system is
extracted by the snow and glacier melt from the HKH mountain ranges. The other
significant source of water supply is the summer monsoon rainfall entries. The aridity
index (a numerical indicator of the degree of dryness of the climate at a given
location) within the Indus River basin, ranges from arid to semi-arid with an annual
rainfall record of less than 200 mm in much of the lower Indus at elevations below
than 2500 m (Archer et al., 2010). The summer monsoon maximum occurs on the
southern slopes of the Himalaya with annual rainfall up to 1800 mm (Archer and
Fowler, 2004). The major agricultural zones are located in the Punjab and Sindh
(provinces of Pakistan).
A major proportion of irrigation water for these agricultural zones is derived from two
main reservoirs – Tarbela and Mangla (Fig. 4.1). These reservoirs store the water in
summer when the high discharges result from snow and glacier melt on one hand
and the monsoon rainfall system on the other hand. This stored water is then
supplied to the Indus Basin Irrigation System (IBIS), during the winter low flow
periods for irrigation purpose. Approximately, 18 million hectares (Mha) (1 hectare =
0.01 km²) i.e. 78% of Pakistan’s total cultivated area of 23 Mha, is the irrigation
command area (Bhatti et al., 2009; Habib, 2004). Nearly 60% of the water utilized for
irrigation is provided by IBIS (Archer et al., 2010) – most of the remainder is
groundwater which is recharged by the rivers and canal network.
182
Any change in available water resources of UIB through climate variability or socio-
economic factors will have a serious impact on food security and the environment in
Pakistan. Archer et al. (2010) has reported that Pakistan’s water resources are
already highly stressed in terms of both ratio of withdrawals to runoff and per-capita
water availability and the situation will become worst in future with projected
population changes (Fig. 4.2) by United Nations (2010).
The other factor reported in many studies, for example, Asianics (2000), Archer et al.
(2010), Fox and Tariq (1996), Izhar-ul-Haq and Abbas (2007), Khan and Tingsanchali
(2009) and Tate and Farquharson (2000), is the decreasing storage capacity of the
existing reservoirs especially in the case of Tarbela reservoir because of the
sedimentation. Potential changes to supply through declining reservoir storage, the
impact of water logging and salinity or over-abstraction of groundwater, or
reallocations for environmental remediation of the Indus Delta or to meet domestic
demands, will reduce water availability for irrigation.
The water demand for urban and agricultural use is expected to increase in the future
with less chance of increasing available water supplies from the Indus River and its
tributaries. This will become more challenging with mismanagement of the water
resources, environmental degradation (degradation of river basins etc.), inefficient
water use and rapid industrialization, all aggravated by climate change.
183
Fig. 4.1 A schematic diagram of the Indus Basin Irrigation System. (Source: WAPDA)
184
1x109
900x106
800x106
700x106
Population
600x106
500x106
400x106
300x106
200x106
100x106
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
Fig. 4.2 Projected population growth in Pakistan (United Nations, 2010).
185
The hydrology of many major rivers has been extensively altered by the increasing
water withdrawals for urban, industrial and agricultural use all over the world (Molle et
al., 2010). Many river basins including the Indus river basin are under stress and
facing scarcity of water because they are severely overcommitted (Archer et al.,
2010; Molle et al., 2010). The Indus river basin is one of the most depleted basins in
the world with almost zero environmental flows in most years (Sharma et al., 2010).
During certain periods of the year, water even doesn’t really reach the sea any more,
making it a closing basin. Therefore, it is important to utilize efficiently these water
resources of IBIS under present and future climate and socio-economic aspects for
the sustainability of water resources management.
Following may be the major present and future challenges for the Indus basin
irrigation system:
 Change in water availability in the snow and glacier fed tributaries of Indus River due
to climate change (as simulated by using Snowmelt-Runoff Model, in this study).
 Population increase and increased urbanisation and industrialisation, resulting in
higher water demands for domestic and industrial purposes, food production and
energy (Archer et al., 2010; United Nations, 2010).
 Declining reservoir storage due to sedimentation (Fox and Tariq, 1996; Izhar-ul-Haq
and Abbas, 2007; Khan and Tingsanchali, 2009 and Tate and Farquharson, 2000).
 Unregulated utilization of resources; a shift from surface water to groundwater use
resulting in rapid depletion of groundwater resources.
 Tension between water resources sharing countries (e.g. Pakistan and India) (Archer
et al., 2010).
Our main focus will be on the present and future water resources (Tarbela and
proposed Diamer Basha Dams) of the Upper Indus River Basin and their
management under different supply and demand as well as climate change
scenarios. In the next sections (sections 4.2 and 4.3) of this chapter, the existing
and proposed water reservoirs (Tarbela and Diamer Basha Dam) (Fig. 4.3) will be
overlooked for their contribution to the irrigation and other water demands. The
change in water availability in the Indus River and its tributaries computed (using
SRM) under different climate scenarios, is presented in section 4.4. In addition, the
effect of environmental change on the Tarbela and Diamer Basha system (in the
perspective of present and future water supply and demand in the region) will be
highlighted in section 4.5.
186
Fig. 4.3 Upper Indus River Basin with the location of present (Tarbela) and future proposed (Diamer Basha) dams site.
187
4.2 Current water supply and demand: Tarbela Dam
Taking into consideration the importance of agricultural sector: the irrigation supplies
for the cultivated area are the priority of the Tarbela reservoir (Figs. 4.1 and 4.3).
Approximately, 95% of water released from Tarbela is used for irrigation (Fig. 4.4) in
downstream areas of Punjab and Sindh provinces - the rest 5% is utilized for drinking
water supplies and industrial use. The construction of Tarbela and Mangla reservoirs
enabled an increase in the irrigation area by 15‒20% and a significant enhancement
in the cropping intensities (Archer et al., 2010). It allowed the transfer of surplus
summer season (Kharif) water to the spring season (Rabi). The main crop of the rabi
season is wheat which is the staple food for majority of the population. According to
Izhar-ul-Haq and Abbas (2007), the total present irrigation water demand in year
2008 is 205 × 109 m3 (billion cubic meter - bcm ). Tarbela reservoir contributes at
present almost 70 × 109 m3 of total available irrigation water (Fig. 4.4) as estimated
from the data available in this study. The rest of water demand is satisfied by the
other surface water resources and ground water pumping.
The water supplies from Tarbela seem to be reduced in future due to the decline of
reservoir storage by sedimentaion (Fox and Tariq, 1996; Izhar-ul-Haq and Abbas,
2007; Khan and Tingsanchali, 2009 and Tate and Farquharson, 2000). The River
Indus and its tributries carry a very high sediment load which has seriously affected
the storage capacity of the Tarbela reservoir. Intial capacity of the Tarbela reservoir is
already reduced by almost 30% in 2010 as presented in Fig. 4.5 and reported by
Khan and Tingsanchali (2009) and Izhar-ul-Haq and Abbas (2007).
The decline of reservoir storgae due to sedimentation and increasing irrigation water
demand for food security of increasing population suggest the construction of new
reservoirs e.g. proposed Diamer Basha Dam (project is approved in 2006, after the
detailed study of its technical, environmental and social aspects) (Figs. 4.1 and 4.3),
on the Indus River. We will discuss the water supply and demand in the context of
future proposed Diamer Basha Dam, in next section.
188
25000
22500
20000
17500
Volume (Million.m )
3
15000
Outflow from Tarbela
Inflow to Tarbela
12500 Water proportion for irrigation use
Water proportion for domestic
and industrial
10000
7500
5000
2500
0
APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR
Fig. 4.4 Average monthly inflow to Tarbela reservoir and average water released for Kharif (April-September) and Rabi (October-
March) seasons, for the period 1974‒2010.
189
480
470
460
450
440
430
Elevation (masl)
420
410
400
390
380
370
360 Year 1974

Year 1985
350 Year 2010
340
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000
3
Volume (Million.m )
Fig. 4.5 Capacity curve of the Tarbela reservoir in 1974, 1985 and 2010.
190
4.3 Future water supply and demand: Post Diamer Basha Dam perspective
The government of Pakistan announced in the year 2006, the decision to construct 5
multi-purpose storages in the country during next 10‒12 years
(http://www.diamerbhasha.com/). Diamer Basha Dam was announced to be
completed in 2016, in the first phase, but it will be completed with 3 years of delay, in
2019. The purpose of this multifunctional dam is storing the summer flow to meet the
needs of supplementary irrigation supplies in the downstream regions during the
winter low-flow period, increasing power-generation demand, flood control and water
supply (for domestic and industrial use).
The project is located on the Indus River , about 315 km upstream of Tarbela Dam
(Fig. 4.1 and Fig. 4.3). The proposed dam will have a maximum height of 270 m and
impound a reservoir of about 9.25 × 109 m3, with live storage capacity of more than
7.89 × 109 m3 (Fig. 4.6) (source: WAPDA). Mean annual flow of the Indus River at the
site (Raikot) (Fig. 4.3) is 62 × 109 m3, as estimated from the data available in this
study.
191
1200
1150
1100
Elevation (masl)
1050
1000
950
900
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000
3
Volume (Million.m )
Fig. 4.6 Capacity curve of the proposed Diamer Basha dam.
192
4.3.1 Water allocation for irrigation and domestic use from Diamer Basha
Dam:
The projected distribution of water from Diamer Basha reservoir for irrigation and
domestic water use is shown in Fig. 4.7.
In regard to the increasing demand of irrigation water, for the food security of
population increasing with alarming rates, the water allocation from the proposed
Diamer Basha Dam is suggested under the maximum, minimum and average stream
flows estimated from the discharge data available at dam site (Table 4.1). Injection of
this irrigation water (available from Diamer) into the Indus Basin Irrigation System
(IBIS) will be helpful to meet with the water deficit of Tarbela for winter and spring
(low-flow period) Rabi crops.
Period Max (x 109 m3) Min (x 109 m3) Average (x 109 m3)
Rabi Season 9.3 6.9 8.1
(October-March) (1990-91) (1984-85) (1963-2004)
Kharif Season 75.4 38.5 53.7

(April-September) (1972-73) (1964-65) (1963-2004)
Annual 83.0 47.8 61.8
(1972-73) (1964-65) (1963-2004)
Table 4.1 Maximum, minimum and average stream flows estimated from the
discharge data of proposed dam site. (Source: WAPDA)
The water demand for domestic and industrial use is expected to increase from 4% to
15% during next 20 years (Briscoe and Qamar, 2007). Water demand increases for
domestic and industrial purposes, relate primarily to the predicted population
increase in the basin, associated with an increase in urbanisation and
industrialisation but also an increase in living standards. Therefore, the water supply
assumed for domestic and industrial use, from Diamer Basha is 10% (Fig. 4.7).
193
20000
18000
16000
14000
Volume (Million.m3)
12000
10000
Outflow from Diamer Basha dam
Inflow to Diamer Basha dam
Water proportion for irrigation use
8000
Water proportion for domestic
and industrial uses
6000
4000
2000
0
APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR
Fig. 4.7 Average expected inflow to Diamer Basha reservoir in 2025 and average water release for Kharif (April-September) and
Rabi (October-March) seasons.
194
4.3.2 Effect of Diamer Basha Dam on the sedimentation control:
An important and alarming factor of water deficit in the Indus Basin Irrigation System
is the loosing capacity of existing storages due to sedimentation. A major part of
Indus inflow originates in the Hindukush-Karakoram-Himalaya ranges which are
highly active geomorphic regions (Tate and Farquharson, 2000), so the
sedimentation is a great threat for the storage capacities built on the River Indus (e.g.
Tarbela).
The construction of Diamer Basha dam (expected to be operational in year 2019) will
reduce the sedimentation of Tarbela reservoir by blocking the sediments, hence
extending the life span and efficiency of Tarbela, as presented in table 4.2.
Sedimentation control by Diamer Basha at upstream of Tarbela will extend the life
span of Tarblea reservoir by almost 36% till 2070, as projected in table 4.2.
Storage Volume (Million.m3)
Operation
Reservoir Initial (2020) After 50 Years (2070)
Scenario
Gross Live Gross Live
Alone 8548.155 8326.125 2269.64 2269.64
In Conjunction
Tarbela
With Diamer 8548.155 8326.125 6389.53 6278.515
Basha
Diamer In Conjunction
9991.35 7882.065 3688.165 3564.815
Basha With Tarbela
Table 4.2 Storage capacity of Tarbela (pre and post Diamer Basha) and Diamer
Basha dam in the years 2020 and 2070. (Source: WAPDA)
195
4.3.3 Hydropower generation capacity of Diamer Basha Dam:
Pakistan has been an energy-deficient country historically. Total Primary Energy
Supply (TPES) per capita is far below the world average and other developed
countries (Mirza et al., 2008). The present demand of electricity in country is about
20,000 MW (Khan and Ahmed, 2009), while the total power generation from all the
sectors (thermal, hydroelectic, gas, nuclear etc.) is about 16,000 MW as indicated on
the website of WAPDA (http://www.wapda.gov.pk/htmls/power-index.html). This
situation will become worst with the increasing demand of electricity in future. During
the Medium Term Development Framework 2005–2010 of the Government of
Pakistan, the electricity demand is projected to grow with an annual rate of 7.9%
(Mirza et al., 2008). A large-scale injection of cheap and environment friendly source
of power thus becomes inevitable. Hydropower is one of the best source to provide
the required electricity at affordable price.
Pakistan is blessed with the great potential of hydropower generation. Hydropower
generation is contributing already almost 33% of total power generation
(http://www.wapda.gov.pk/htmls/power-index.html). The identified hydropower
potential in the country is approximately 41,722 MW (Mirza et al., 2008), most of
which lies in UIB. One of these hydropower potential projects is the proposed Diamer
Basha Dam. Contribution by an installed hydropower capacity of 4500 MW
(http://www.diamerbhasha.com/) from Diamer Basha Dam will go a long way to
alleviate this energy shortage in the country. Average annual hydroelectric energy
generation is estimated around 18,095 GWh (Giga watt hour) at Diamer Basha and
an additional 1110 GWh will be produced at Tarbela (source: WAPDA) as shown in
Fig. 4.8.
196
3500
3250
3000
2750
Hydropower generation (GWh)
2500
2250 Tarbela (year 2010)

Tarbela in post Diamer period
2000 (year 2025)
Diamer Basha (year 2025)
1750
1500
1250
1000
750
500
250
0
Fig. 4.8 Mean monthly power generation at Tarbela (pre and post Diamer Basha) and Diamer Basha dam.
197
In addition, the construction of Diamer Basha Dam will contribute to the development
of some major hydropower projects, downstream of the reservoir (Dasu, Patan and
Thakot) (Fig. 4.3), with a total installed capacity of apprximately 10,000 MW (Table
4.3).
Sr. No. Project Installed Capacity Remarks
(MW)
1. Diamer Basha Dam 4,500 Multi-purpose Storage
Project
2 Downstream Projects
i. Dasu 4,320 Basically Run-of-River

Projects to Be Firmed up
Through Storage Releases
From Diamer Basha
ii. Patan 2,800
iii. Thakot 2,800
Sub-Total (2) 9,920
3. Total (1+2) 14,420
Table 4.3 Developement of hydropower projects downstream of Diamer Basha Dam

after its construction. (Source: WAPDA)
198
4.4 Change in water availability in the Upper Indus River Basin (Present and future
scenarios)
The snow and glacier melt water from the sparsely populated Upper Indus River
Basin is essential for the densely populated lowlands with its extensive irrigation
system. Numerous authors (e.g. Bookhagen and Burbank, 2010 and Immerzeel et
al., 2010) and the present thesis reported that the melt water is extremely important
in the Indus basin. The present day climate discharge generated by snow and glacial
melt is almost 65% of the total discharge (Immerzeel et al., 2010). Climate change
will definitely affect the temporal and spatial availability of water resources worldwide
but the effects in the Upper Indus River Basin remain uncertain. Many authors
(Berthier et al., 2007; Immerzeel et al., 2010; Shrestha and Shrestha, 2004) have
reported a significant retreat and depletion of glacier volume across the HKH region.
Immerzeel et al. (2010) indicates that upstream snow and ice reserves of the Indus
basin (important in sustaining seasonal water availability) are likely to be affected
substantially by climate change, but to what extent is yet unclear. A new study by
Scherler et al. (2011) actually indicates that debris coverage may be a missing link in
the understanding of the decline of glaciers in the HKH.
A totally different picture is shown in this study and by many authors (Fowler and
Archer, 2006; Hewitt, 2005; Hewitt, 2007; Scherler et al., 2011) for central Karakoram
and western Himalaya regions, who reported an expansion of glacier cover area
notably due to surging and advancing. This is in contrast to the prevailing notion that
all glaciers in the HKH are retreating. This controversy increases the uncertainty in
the forecast of future water availability of the Indus basin and further research is
needed to enlighten this uncertain matter. These topics about the snow and glacier
cover dynamics, hydrological regime of the Upper Indus River Basin and modeling
snowmelt runoff are discussed in detail in the chapter 2 and 3 of this study. We will
rather present our results about present (year 2000) and future (year 2075) modeled
discharge of the Upper Indus River in this section to estimate the water availability
under climate change scenarios.
The discharge for the Upper Indus River and its tributaries is simulated by using the
Snowmelt Runoff Model (SRM) (described in detail in chapter 3). Once the
successful simulation of the discharge in all the tributaries was carried out; the model
SRM was run to simulate the discharges under future climate change scenarios. The
199
results are presented in Fig. 4.9 for the scenario when the temperature increases by
4°C with an increase in snow cover area up to 20% notably due to an increase in the
precipitation on the high altitudes, as explained in chapter 3. The discharge is almost
doubled in summer under this climate change scenario, but this change in discharge
is bounded to the existence of glaciers till the end of this century. If the temperature
will increase with a constant precipitation then the glaciers can disappear after
producing high discharge for some seasons. This fact is also explained by Rees and
Collins (2006) who found that the flow in Himalayan catchments are likely to increase
under increasing temperature scenarios, but after a specific time scale, the glaciers
will disappear because of accelrated melting.
If this scenario of increasing mean summer temperature comes true, then we will be
in need of more water reservoirs than present to conserve this increasing discharge
for irrigation supply, flood control and power generation. However the present water
resevoirs seem already not to be sufficient for the present demand of irrigation,
power generation and prevention from the high floods. The whole Indus irrigation
system depends on the discharge conserved by the main reservoirs of Tarbela and
Mangla. These reservoirs satisfy almost half of the country’s irrigation water demand
(Archer et al., 2010; Tate and Farquharson, 2000) and about 30% of the electricity
demand (Asianics, 2000; Tate and Farquharson, 2000).
No major reservoirs are constructed since 1974 after the construction of Tarbela
reservoir but the population has increased almost three times than that in 1970 as
given by United Nations (2010) population data base. Therefore, the country, at
present, is facing the extreme power shortage and irrigation water deficit. The
situation may become horrible to meet the future water demand if we rely only on the
present water resources.
200
15000
14000
13000
12000
11000
10000
Discharge (m3/s)
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
2000 2075 2000 2075 2000 2075 2000 2075 2000 2075 2000 2075 2000 2075 2000 2075
Kharmong Shyok Shigar Hunza Gilgit Astore Shatial Besham
Fig. 4.9 Present and future simulated discharge in the Upper Indus River and its tributries.
201
4.5 Impact of the environmental changes on the Tarbela and Diamer Basha system
Environmental changes like climate change, population increase and sedimentation
of resevoirs, may be a future threat for the Tarbela and Diamer Basha system. The
simulated increased flow by using SRM, may be a challenge for the reducing storage
capacities of Tarbela and Diamer Basha Dam. Besham Qila and Raikot (Fig. 4.3) are
the hydrometric stations to measure the inflow to Tarbela and Diamer Basha Dam,
respectively. The simulated summer inflow (using SRM) at Besham Qila and Raikot
tends to increase under the scenario of increasing mean temperature and snow
cover area (Figs. 4.9 and 4.10) and it is almost doubled in 2075. This accelerated
flow from the glaciers will bring a large amount of sedimentation which may catalyse
the reduction of the storage capacities of these Upper Indus River dams.
In addition, an alarming rate of population increase will over-stress the Tarbela and
Diamer Basha system, for its food, water and power requirements. After the
construction of Diamer Basha dam, the available water from Upper Indus River for
irrigation demand will increase to almost 50% in year 2025 (Fig. 4.11). The water
available from these reservoirs (Tarbela and Diamer Basha), in future, is assumed to
be constant or to decrease because the discharge from the Upper Indus River will
tend to increase till the end of this century (as simulated from the snowmelt runoff
modeling), but the increased volume of water in summer will not be stored due to the
same or reduced storage capacity. This may result in catastrophic floods like in 2010
(Mustafa and Wrathall, 2011), which affected millions of people.
The results presented about water supply and demand under present and future
climate and population scenarios extracted from this study and the previous studies
suggest to build the live storage capacities. Water resources will gradually diminish
(especially in the spring) unless new large reservoirs are built on the River Indus.
Reservoir sedimentation will also limit the present ability to generate the hydro
electric power to meet the growing demand for domestic and industrial energy. The
proposed Diamer Basha Dam is expected to fulfill a considerable proportion of
irrigation and power demands. The alternative of using groundwater as a means of
securing water supply for the spring (rabi) crop may diminish as water tables are
falling rapidly and pumping becomes impractical (Archer et al., 2010). Therefore, a
detailed study is necessary to examine the possibilities of building 2‒3 big capacity
reservoirs on the Indus River (taking all the environmental and social impacts into
account).
202
14000
13000
12000
11000
10000
3
9000
8000
7000
Year 2025 ( C)
6000
Year 2050 (C)
and SCA 
5000
Year 2075 (C)
4000 and SCA 
3000
2000
1000
0
Month
Fig. 4.10 Simulated discharge in the Indus River at Raikot (proposed site of Diamer Basha dam) under changed climate in the year
2025, 2050 and 2075.
203
300
Total irrigation water requirement
Water available from Tarbela
Water available from DBD
250
Volume of water (10 .m )
200
3
9
150
100
50
0
2010 2025 2050 2075
Year
Fig. 4.11 Total present and future irrigation water requirement in Pakistan and the water available from Tarbela and Diamer Basha
reservoirs to satisfy this demand.
204
4.6 Conclusion
The Indus Basin Irrigation System (IBIS), comprising the major storage reservoirs of
Tarbela on the Indus and Mangla on the Jhelum and a network of barrages and
canals, is the largest integrated irrigation system in the world. A major part of the
water for this irrigation system is extracted by the snow and glacier melt from the
Hindukush-Karakoram-Himalaya mountain ranges. The other source of water supply
is the summer monsoon rainfall system and the groundwater pumping. Any change in
available water resources of UIB through climate variability or socio-economic factors
will have a serious impact on food security, power generation. The water demand for
urban and agricultural use is expected to increase in the future due to a rapid
increase in the population. Although the discharges from all of the Upper Indus River
tributaries tend to increase under the climate change scenarios but the present water
storages are not capable of conserving this future simulated inflow. This will become
more challenging with mismanagement of the water resources, environmental
degradation (degradation of river basins etc.) and an accelerated reduction in the
present storage capacity due to sedimentation. There is not only the possibility of
increase in the average volumes but the annual variability may also increase and the
seasonality of flows may alter. All these drawbacks may lead to the catastrophic
floods in summer during high discharge periods and no water supplies in winter and
spring for the most valuable Rabi crops.
Therefore, it is of utmost importance to closely study the building of at least 2‒3 large
capacity multi-purpose dams, e.g. the proposed Diamer Basha Dam and Kalabagh
Dam (Fig. 4.1), on the River Indus on priority basis to conserve the high summer
discharge to satisfy the irrigation demand during low flow periods and to generate the
cheap hydroelectric power for growing domestic and industrial use. Those projects
should carefully take into consideration, the environmental impacts and the interests
of the local people.
There is also a need to build up the knowledge and data base of the mountain
environments from which the greater part of the water resources originates. This will
require improved education and a wider dissemination of scientific knowledge and
application. A high motivation on the national level to organise a 10‒15 year program,
to improve the management of IBIS and for the construction of the new reservoirs
keeping in mind the future growing water demand, is needed. Some modeling
techniques for the management of water resources should be applied to increase the
205
efficiency of this large Indus Basin Irrigation System. Conventional supply-oriented
simulation models are not always adequate. Over the last decade, an integrated
approach to water development has emerged which places water supply projects in
the context of demand-side issues, as well as issues of water quality and ecosystem
preservation.
The Water Evaluation and Planning System (WEAP), developed by Raskin et al.
(1992) at Stockholm Environment Institute's (SEI), U.S. Center, aims to incorporate
these values into a practical tool for water resources planning. Freshwater
management challenges are increasingly common. Allocation of limited water
resources between agricultural, municipal and environmental uses now requires the
full integration of supply, demand, water quality and ecological considerations. The
Water Evaluation and Planning system, aims to incorporate these issues into a
practical yet robust tool for integrated water resources planning. The successful
application of this tool in many regions of the world e.g. Mehta et al. (2011), Yilmaz
and Harmancioglu (2010) and Mulligan, M. et al. (2011), suggests that it may be a
useful tool for water resources planning in the Upper Indus River Basin.
206
Résumé étendu de la CONCLUSION GENERALE
Cette étude s’est attachée à trois questions importantes du haut bassin de l’Indus :
1) La dynamique du couvert neigeux sous l’impact du changement climatique et
le régime hydrologique de l’UIB ;
2) La modélisation du débit de fonte de neige dans l’UIB en lien avec la
variabilité du climat futur ;
3) La gestion des ressources en eau dans l’UIB sous les contraintes de
changement climatique et de demande en eau.
Les conclusions tirées des résultats obtenus sont présentées dans les trois sections
suivantes :
1) Couvert neigeux et régime hydrologique

10 années d’analyse de données spatiales sur la cryosphère, une précipitation
hivernale tendant à croître, une stabilité des températures moyennes et une
tendance à la diminution des débits du bassin de Hunza, tout cela concoure à faire
croire que la région du bassin de Hunza est sujette à une légère augmentation de sa
cryosphère, particulièrement en haute altitude. La même tendance dans la variation
du couvert neigeux est notée dans le bassin d’Astore et probablement dans tout le
haut bassin de l’Indus, liée à l’augmentation des précipitations hivernales et à un
léger refroidissement des températures moyennes d’été. Par contraste avec Hunza,
l’augmentation des débits d’Astore peut être la conséquence de précipitations de
mousson d’été croissante. Cette évolution du couvert neigeux dans l’UIB peut être
utilisée comme hypothèse pour simuler les écoulements du bassin avec des modèles
hydrologiques incluant une composante de fonte de neige.
La dynamique annuelle et saisonnière du couvert neigeux montre une bonne
corrélation inverse avec les températures moyennes d’été dans les deux sous-
bassins de Hunza et d’Astore, produisant un fort débit. Il est par conséquent
important d’évaluer les accumulations de neige d’hiver à haute altitude afin de prévoir
les écoulements d’été et finalement d’améliorer la gestion des retenues d’eau à l’aval
comme le réservoir de Tarbela.
Les écoulements de Hunza sont principalement influencés par les températures
moyennes d’hiver et d’été, alors que les débits d’Astore ont une corrélation
significative avec les précipitations de l’hiver précédent. De plus, la région
207
hydrologique la plus active de l’UIB est celle qui se situe au-dessus de 4500 m où se
produisent le plus souvent les chutes de neige.
2) Modélisation de la fonte de neige sous un climat variable
Le modèle SRM, basé sur un facteur degré-jour, a simulé de manière satisfaisante
les débits journaliers des sous-bassins versants de l’UIB dominés par les processus
nivo-glaciaires (Hunza, Gilgit, Astore and Shyok). L’efficience de SRM dans les
bassins de haute altitude peut être attribuée à l’utilisation des données spatiales
MOD10A2 caractérisant la cryosphère en entrée du modèle, équivalent à une
production d’eau.
L’hydrologie de l’UIB évolue selon les différents scénarios de changements
climatiques d’augmentation de température et de précipitation indiqués par le GIEC
(IPCC, 2007), ce qui conduit à un accroissement des volumes écoulés dans les
sous-bassins en été. Les écoulements de fonte de neige se décalent en lien avec la
remontée de la température au printemps. Cette analyse montre que de nouvelles
retenues doivent être envisagées pour stocker les écoulements d’été afin de
répondre aux besoins supplémentaires d’irrigation durant les périodes de basses-
eaux dans les régions aval, d’augmenter la production énergétique, de contrôler les
crues et d’assurer l’approvisionnement en eau domestique.
Le scénario d’augmentation de la température moyenne ne correspond pas tout à fait
à ce que nous avons observé dans l’UIB, où une tendance à une légère diminution.
En conséquence une simulation de débit a aussi été réalisée pour le bassin de
Hunza avec une tendance de température décroissante, dont les résultats peuvent
conduire à une diminution des débits d’été. Si ce scénario se confirmait dans l’UIB,
des ressources alternatives deviendraient indispensables pour l’irrigation et la
production énergétique, les apports du haut Indus devenant insuffisants. Malgré
l’utilisation de scénarios climatiques choisis a priori ou dérivés d’études antérieures,
ils apportent des informations utiles à la gestion des ressources en eau pour
répondre aux besoins futurs du pays.
3) Gestion durable des ressources en eau dans le haut bassin de l’Indus
Le système d’irrigation du bassin de l’Indus (Indus Basin Irrigation System IBIS), qui
inclut les réservoirs majeurs de Tarbela sur l’Indus et de Mangla sur la Jhelum, ainsi
qu’un réseau de barrages et de canaux, est le plus grand système intégré d’irrigation
de la terre. La plupart de l’eau de ce système d’irrigation vient de la fonte de la neige
et des glaciers des chaînes de montagne de l’Hindukush-Karakoram-Himalaya. Les
208
autres sources d’approvisionnement en eau sont constituées par les précipitations de
mousson l’été et par le pompage dans les nappes. Toutes les études précédentes et
celle-ci encore insistent sur le fait que tout changement dans les ressources de l’UIB
dus à la variabilité climatique ou à des facteurs socio-économiques aura un impact
sérieux sur la sécurité alimentaire et la production énergétique. La demande en eau
pour les usages urbains et agricoles va encore croître dans le futur à cause de la
rapide croissance de la population. Avec la tendance croissante des débits de tous
les affluents du haut Indus d’après les scénarios de changement climatique, les
réservoirs actuels ne sont pas capables de stocker les apports futurs attendus.
Il est par conséquent de la plus grande importance d’étudier minutieusement la
construction d’au moins 2 ou 3 barrages multi-usages de grande capacité, et plus
particulièrement les projets de barrages de Diamer Basha et de Kalabagh (Fig. 4.1)
sur l’Indus. La priorité est en effet de retenir les débits de hautes eaux en été afin de
satisfaire la demande d’irrigation durant les périodes de basses eaux, tout en
produisant de l’énergie hydro-électrique à bas coût pour les usages domestiques et
industriels croissants. Il est enfin nécessaire de développer les connaissances et les
observations dans les environnements de montagne d’où vient la plus grande part de
la ressource en eau. Cela passe par une amélioration de l’éducation et une plus
large dissémination de la connaissance scientifique et de ses applications.
209
GENERAL CONCLUSION
This study was oriented towards three important aspects of the Upper Indus River
Basin:
Phase 1: Snow cover dynamics under the impact of climate change and the
hydrological regime of the Upper Indus River Basin
This study started with the analysis of MODIS snow cover product (MOD10A2) to
estimate the cryosphere area change in two sub-catchments of UIB. The hydrological
and climate data are treated and analysed to understand the hydrological regime of
the catchment areas under study.
Phase 2: Modeling snowmelt runoff in the UIB under future climate variability
MODIS (MOD10A2) snow cover satellite data and the hydro-meteorological data
(discharge, precipitation and temperature), were used as basic inputs in the
snowmelt-runoff model (SRM) to simulate the daily stream flows from the studied
catchments (Hunza, Gilgit, Astore and Shyok). The successful application of the
snowmelt-runoff model led us to simulate the stream flows under future climate
change scenarios, using the model SRM.
Phase 3: Water resources management in the UIB under future climate change
and water demand
The sustainability of water resources management in the UIB was overviewed, under
the perspective of existing and future water reservoirs. Impact of social and
environmental change on the existing and future proposed water resources, was then
investigated.
A conclusion of the results found in this study, is presented in three sections as

follows:
1) Snow cover dynamics and the hydrological regime
The 10-year analysis of the remotely sensed cryosphere data, an increasing winter
precipitation trend, constancy in mean temperatures and declining discharge trend in
the Hunza River basin suggest that the Hunza River basin is a region undergoing a
slight expansion in the cryosphere area, especially at high elevations. The same
tendency in the snow cover change is found in the Astore River basin and eventually
210
in the Upper Indus catchment due to increasing winter precipitation and cooling trend
of summer mean temperatures, in the region. The increase of Astore River flows in
contrast to the Hunza basin may be the result of increasing tendency of summer
monsoon rainfall in the Astore basin. This change in snow cover area of UIB may be
used as a future climate change hypothesis to simulate the catchment runoff, by
using the hydrological models with a snow melt component.
The study of annual and seasonal snow cover dynamics suggests that the snow
cover change has a highly inverse correlation with the summer mean temperatures in
both sub-catchments (Hunza and Astore) of UIB, which results in a large amount of
stream flow. Therefore, it is very important to estimate the winter snow accumulations
at high altitudes to forecast the summer flow and ultimately to improve the
management of the water reservoirs downstream like Tarbela reservoir.
The stream flow in the Hunza River is influenced mainly by the mean summer and
winter temperatures of the catchment whereas the discharge in the Astore River has
a significant positive correlation with the previous winter precipitation. Moreover, the
most active hydrological region of the UIB is present above an elevation of 4500 m,
where most of the snow fall occurs.
In addition, MODIS MOD10A2 remote sensing cryosphere product is a good option to
assess the percentage cryosphere areas in the Hunza and Astore River basins, over
the long term, because it is free of charge, easy to treat and have a reasonable
resolution to use in the larger catchment areas. A comparison of MODIS product with
finer resolution ASTER images proved the validation/reliability of this remote sensing
snow cover data for high elevations.
2) Snowmelt runoff modeling under climate variability
The snowmelt runoff model (SRM), based on a degree day factor, has efficiently
simulated the daily discharges in the snow- and glacier-fed sub-catchments (Hunza,
Gilgit, Astore and Shyok) of the Upper Indus River Basin (UIB). The SRM’s efficiency
in high-altitude catchments can be attributed to the use of the MOD10A2 remotely
sensed cryosphere data as input to the model, for water-equivalent production.
The hydrology of UIB seems to alter under different climate change scenarios of
increasing temperature and precipitation, as given by IPCC (2007). This change
apparently results from the increasing temperature scenarios, which have resulted in
an increase of the catchment’s summer stream flow volumes. The availability of the
snowmelt runoff seems to shift accordingly with the mean temperature rise in the
211
spring. This analysis suggests that new reservoirs are required for the summer flow
storage to meet the needs for supplementary irrigation supplies during the low-flow
period in the downstream regions, increasing power-generation demand, flood
control and water supply.
The scenario of increasing mean temperature is contrary to what we found in the
UIB, where a trend of slight decrease in mean summer temperatures is found.
Therefore, the discharge simulation under decreasing temperature trend in the Hunza
River basin was also carried out, which results eventually in a decrease of the
summer discharges. If this scenario comes true in the UIB, then the alternate sources
of irrigation water and hydropower generation would be indispensable, because of a
decrease in the Upper Indus flows. Though the climate scenarios used in this study
were assumed or derived from previous studies, they provide useful information to
manage water resources for the country’s future needs.
3) Sustainability of the water resources management in the Upper Indus River
Basin
The Indus Basin Irrigation System (IBIS) comprising the major storage reservoirs of
Tarbela on the Indus and Mangla on the Jhelum and a network of barrages and
canals is the largest integrated irrigation system in the world. A major part of the
water for this irrigation system is extracted by the snow and glacier melt from the
Hindukush-Karakoram-Himalaya mountain ranges. Other source is the summer
monsoon rainfall system and the groundwater pumping. All the previous and the
present studies suggest that any change in available water resources of UIB through
climate variability or socio-economic factors will have a serious impact on food
security, power generation and the environment in Pakistan. The water demand for
urban and agricultural use is expected to increase in the future due to a rapid
increase in the population. Although, the discharges from all of the Upper Indus River
tributaries tend to increase under the assumed climate change scenarios but the
present water storages are not capable of conserving this future expected inflow.
This will become more challenging with mismanagement of the water resources,
environmental degradation (degradation of river basins etc.) and an accelerated
reduction in the present storage capacity due to sedimentation. There is not only the
possibility of increase in the average volumes but the annual variability may also
increase and the seasonality of flows may alter. All these drawbacks may lead to the
212
catastrophic floods in summer during high discharge periods and no water supplies in
winter and spring for the most valuable Rabi (winter and spring season) crops.
Therefore, it is of utmost importance to construct at least 2‒3 large capacity multi-

purpose dams, e.g. the proposed Diamer Basha dam and Kalabagh dam (Fig. 4.1),
on the River Indus on priority basis, to conserve the high summer discharge to satisfy
the irrigation demand during low flow periods and to generate the cheap hydroelectric
power for growing domestic and industrial use. There is also a need to build up the
knowledge and data base of the mountain environments from which the greater part
of the water resources originates. This will require, an improved education and a
wider dissemination of scientific knowledge and application.
 The tendencies of snow cover, climate and the hydrological variables

presented in this study are weak or based on the short time series of data so
these should be verified before any operational use.
 The simulated discharge under climate variability in chapter 3 is based on the
scenarios extracted from IPCC reports and the published literature for the
Himalayan region. The future climate change may differ from these
hypotheses particularly in the other regions including the Upper Indus River
Basin. Therefore, the results obtained by the application of these scenarios in
the UIB should carefully be used and verified by other methods before using
for the operational hydrology.
Perspectives:
This thesis work will hopefully contribute to the understanding of the cryosphere
(snow and ice) dynamics and hydrological regime of the UIB area, on which millions
of people are dependent for their food and water security. This study may eventually
help the researchers to expand their knowledge base on the mountain hydrology of
the Hindukush-Karakoram-Himalaya region. The hydrological modeling used in this
study and some other efficient modeling (hydrological models) techniques may be
used to well understand the variation of flow (seasonal and volume variation) in the
213
Upper Indus River, so that the efficient predictions of flow can be issued to manage
the water resources accordingly.
A high motivation on the national policy maker’s level is required, to organise a 10‒15
year program for the improvement of the IBIS management and for the construction
of the new reservoirs, keeping in mind the future growing water demands. Otherwise,
it may not be possible to meet with the increasing pressure of rapidly growing
population and the environmental changes, for the present water resources of the
country.
214
REFERENCES
Abbott, M.B., Bathurst, J.C., Cunge, J.A., O'Connell, P.E. and Rasmussen, J., 1986. An
introduction to the European Hydrological System -- Systeme Hydrologique
Europeen, "SHE", 2: Structure of a physically-based, distributed modelling system.
Journal of Hydrology, 87(1-2): 61-77.
Akhtar, M., Ahmad, N. and Booij, M.J., 2008. The impact of climate change on the water
resources of Hindukush–Karakorum–Himalaya region under different glacier
coverage scenarios. Journal of Hydrology, 355: 148-163.
Albert, M. and Krajeski, G., 1998. A fast, physically based point snowmelt model for use in
distributed applications. Hydrological Processes, 12(10-11): 1809-1824.
Ali, K.F. and Boer, D.H.d., 2007. Spatial patterns and variation of suspended sediment yield
in the Upper Indus River Basin, northern Pakistan. Journal of Hydrology, 334: 368-
387.
Archer, D., 2003. Contrasting hydrological regimes in the Upper Indus Basin. Journal of
Hydrology, 274: 198–210.
Archer, D.R. and Fowler, H.J., 2004. Spatial and temporal variations in precipitation in the
Upper Indus Basin, global teleconnections and hydrological implications. Hydrology
and Earth System Science, 8(1): 47-61.
Archer, D.R., Forsythe, N., Fowler, H.J. and Shah, S.M., 2010. Sustainability of water
resources management in the Indus Basin under changing climatic and socio
economic conditions. Hydrol. Earth Syst. Sci. (HESS), 14(8): 1669-1680.
Asianics, A.D.I.P.L., 2000. Tarbela Dam and related aspects of the Indus River Basin,
Pakistan, A WCD case study prepared as an input to the World Commission on
Dams, Cape Town, www.dams.org, Cape Town, South Africa.
BBC (British Broadcasting Corporation), 2010. '2.5m people affected' by Pakistan floods.
http://www.bbc.co.uk/news/world-south-asia-10834414 (accessed 24 June 2011)
Bergström, S., 1975. The development of a snow routine for the HBV-2 model. Nordic
Hydrology, 6(2): 73–92.
Berthier, E. et al., 2007. Remote sensing estimates of glacier mass balances in the
Himachal Pradesh (Western Himalaya, India). Remote Sensing of Environment, 108:
327-338.
BGIG, 1976. Investigation report on the Batura glacier in the Karakoram Mountains, the
Islamic Republic of Pakistan (1974-1975), Beijing.
215
Bhatti, A.M., Suttinon, P. and Nasu, S., 2009. Agriculture water demand management in
Pakistan: a review and perspective. Society for Social Management Systems,
http://management.kochi-tech.ac.jp/PDF/ssms2009/sms09_172.pdf.
Bookhagen, B. and Burbank, D.W., 2010. Toward a complete Himalayan hydrological
budget: Spatiotemporal distribution of snowmelt and rainfall and their impact on river
discharge. Journal of geophysical research 115, F03019.
Briscoe, J. and Qamar, U., 2007. Pakistan's water economy running dry, Oxford University
Press, Karachi. Commissioned by World Bank.
Cannone, N., Diolaiuti, G., Guglielmin, M. and Smiraglia, C., 2008. Accelerating climate
change impacts on alpine glacier forefield ecosystems in the European Alps.
Ecological Applications, 18(3): 637-648.
Ciarapica, L. and Todini, E., 2002. TOPKAPI: a model for the representation of the rainfall-
runoff process at different scales. Hydrological Processes, 16(2): 207-229.
Dankers, R. and de Jong, S.M., 2004. Monitoring snow cover dynamics in Northern
Fennoscandia with SPOT VEGETATION Images. International journal of Remote
Sensing, 25: 2933-2949.
Dey, B., Sharma, V.K. and Rango, A., 1989. A test of Snowmelt-Runoff Model for a major
river basin in Western Himalayas. Nordic Hydrology, 20: 167-178.
Farr, T.G. et al., 2007. The Shuttle Radar Topography Mission. Rev. Geophys., 45(2):
RG2004.
Førland, E.J. et al., 1996. Manual for operational correction of Nordic precipitation data.
24/96, DNMI, P.O. Box 43, Blindern, Oslo, Norway.
Fowler, H.J. and Archer, D.R., 2005. Hydro-climatological variability in the Upper Indus
Basin and implications for water resources, Regional Hydrological Impacts of Climatic
Change - Impact Assessment and Decision making, Seventh IAHS Scientific
Assembly at Foz do Iguaçu, Brazil, pp. 131-138.
Fowler, H.J. and Archer, D.R., 2006. Conflicting Signals of Climatic Change in the Upper
Indus Basin. Journal of climate, 19: 4276-4293.
Fox, I. and Tariq, S., 1996. Development of sedimentation delta at Tarbela Reservoir,
Pakistan, International conference on reservoir sedimentation, Colorado State
University / ICCORES / UNESCO, pp. 1425-1447.
Georgievsky, M.V., 2009. Application of the Snowmelt Runoff Model in the Kuban River
basin using MODIS satellite images. Environmental Research Letters, 4: 1-5.
216
Habib, Z., 2004. Pakistan: Indus Basin and Water Issues, South Asian Journal. South Asian
Free Media Association, http://www.southasianmedia.net/Magazine/Journal/previous-
issue6.htm.
Hall, D., Riggs, G. and Salomonson, V., 2006, updated weekly. MODIS/Terra Snow Cover
8-Day L3 Global 500m Grid V005, [March 2000 to December 2009]. Boulder,
Colorado USA: National Snow and Ice Data Center. Digital media.
Hall, D., Riggs, G., Salomonson, V.V., DiGirolamo, N. and Bayr, K., 2002. MODIS snow-
cover products. Remote-sensing of Environment, 83: 181 - 94.
Haritashya, U.K., Bishop, M.P., Shroder, J.F., Bush, A.B.G. and Bulley, H.N.N., 2009.
Space-based assessment of glacier fluctuations in theWakhan Pamir, Afghanistan.
Climatic Change, 94: 5–18.
Hewitt, K., 1986. Snow and Ice Hydrology Project, Wilfrid Laurier University, Waterloo,
Ontario.
Hewitt, K., 1998. Glaciers receive a surge of attention in the Karakoram Himalaya. Eos,
Transactions American Geophysical Union, 79(8): 104-105.
Hewitt, K., 2001. Hazards of melting as an option: Upper Indus glaciers I & II, Dawn.
Hewitt, K., 2005. The Karakoram Anomaly? Glacier Expansion and the Elevation Effect,
Karakoram Himalaya. Mountain Research and Development, 25(04): 332-340.
Hewitt, K., 2007. Tributary glacier surges: an exceptional concentration at Panmah Glacier,
Karakoram Himalaya. Journal of Glaciology, 53(181): 181-188.
Hewitt, K., Wake, C.P., Young, G.J. and David, C., 1989. Hydrological Investigation at Biafo
Glacier, Karakoram Range, Himalaya; An important source of water for the Indus
River. Annals of Glaciology, 13: 103-108.
Hirsch, R.M. and Slack, J.R., 1984. A nonparametric trend test for seasonal data with serial
dependence. Water Resources Research, 20(6): 727-732.
Hirsch, R.M., Slack, J.R. and Smith, R.A., 1982. Techniques of trend analysis for monthly
water quality data. Water Resources Research, 18(1): 107-121.
Hock, R., 2003. Temperature index melt modeling in mountain areas. Journal of Hydrology,
282(1-4): 104-115.
Hong, M. and Guodong, C., 2003. A test of Snowmelt Runoff Model (SRM) for the
Gongnaisi River basin in the western Tianshan Mountains, China. Chinese Science
Bulletin, 48: 2253-2259.
217
Hussain, S.S., Mudasser, M., Sheikh, M.M. and Manzoor, N., 2005. Climate change and
variability in mountain regions of Pakistan: implications for water and agriculture.
Pakistan Journal of Meteorology, 2(4): 79–94.
Immerzeel, W.W., Droogers, P., de Jong, S.M. and Bierkens, M.F.P., 2009. Large-scale
monitoring of snow cover and runoff simulation in Himalayan river basins using
remote sensing. Remote Sensing of Environment, 113: 40-49.
Immerzeel, W.W., van Beek, L.P.H. and Bierkens, M.F.P., 2010. Climate change will affect
the Asian water towers. Science, 328(5984): 1382-1385.
IPCC, 2007. Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II
and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change [Core Writing Team, Pachauri, R.K and Reisinger, A. (eds.)],
Intergovernmental Panel on Climate Change (IPCC), Geneva, Switzerland.
Izhar-ul-Haq, D. and Abbas, S.T., 2007. Sedimentation of Tarbela & Mangla reservoirs,
Pakistan Engineering Congress, 70th Annual Session Proceedings, Lahore,
Pakistan, pp. 24-46.
Kendall, M.G. and Gibbons, J.D., 1990. Rank Correlation Methods, 5th ed. Edward Arnold,
London.
Kendall, M.G., 1975. Rank Correlation Measures, 4th ed. Charles Griffin, London, 202 pp.
Khan, M.A. and Ahmed, U., 2009. Energy demand in Pakistan: A Disaggregate Analysis.
Munich Personal RePEc Archive, 4: 1-27.
Khan, N.M. and Tingsanchali, T., 2009. Optimization and simulation of reservoir operation
with sediment evacuation: a case study of the Tarbela Dam, Pakistan. Hydrological
Processes, 23(5): 730-747.
Lee, S., Klein, A.G. and Over, T.M., 2005. A comparison of MODIS and NOHRSC snow-
cover products for simulating streamflow using the Snowmelt Runoff Model.
Hydrological Processes, 19(15): 2951-2972.
Liniger, H., Weingartner, R. and Grosjean, M. (Editors), 1998. Mountains of the world: Water
towers for the 21st century. Mountain agenda for the commission on sustainable
development (CSD), BO12, Berne, 32 pp.
Link, T. and Marks, D., 1999. Distributed simulation of snowcover mass- and energy-
balance in the boreal forest. Hydrological Processes, 13: 2439-2452.
Lopez, P., 2007. Impact de la variabilité climatique sur la cryosphere du Campo de Hielo
Norte: Apport de la télédétection. Eaux continentales et société, Thesis, Université
Montpellier 2, Sciences et Techniques du Languedoc, Montpellier, 143 pp.
218
Maidment, D., 1993. Handbook of hydrology. McGraw- Hill, Inc., 1424 pp.
Martinec, J., 1975. Snowmelt-Runoff Model for stream flow forecasts. Nordic Hydrology, 6:
145-154.
Martinec, J., Rango, A. and Roberts, R., 2007. Snowmelt-Runoff Model (SRM) user's
manual. USDA Jornada Experimental Range, New Mexico State University, Las
Cruces, NM 88003, U.S.A.
Maurer, E.P., Rhoads, J.D., Dubayah, R.O. and Lettenmaier, D.P., 2003. Evaluation of the
snow-covered area data product from MODIS. Hydrological Processes, 17(1): 59-71.
Mehta, V.K. et al., 2011. Potential impacts on hydrology and hydropower production under
climate warming of the Sierra Nevada. Journal of Water and Climate Change, 2(1):
29–43.
Mercer, J.H., 1975. Glaciers of the Karakoram. In: W. O. Field Mountain Glaciers of the
Northern Hemisphere., CRREL, Hanover New Hampshire, pp. 371-409.
METI and NASA, 2009. ASTER Global Digital Elevation Model (GDEM),
http://www.ersdac.or.jp/GDEM/E/2.html. (accessed in November, 2009)
Mirza, U.K., Ahmad, N., Majeed, T. and Harijan, K., 2008. Hydropower use in Pakistan:
Past, present and future. Renewable and Sustainable Energy Reviews, 12(6): 1641-
1651.
Molle, F., Wester, P. and Hirsch, P., 2010. River basin closure: Processes, implications and
responses. Agricultural Water Management, 97(4): 569-577.
Mulligan, M. et al., 2011. The nature and impact of climate change in the Challenge
Program on Water and Food (CPWF) basins. Water International, 36(1): 96-124.
Mustafa, D. and Wrathall, D., 2011. Indus Basin Floods of 2010: Souring of a Faustian
Bargain? Water Alternatives, 4(1): 72-85.
Nabi, G., 2009. Regional scale sediment yield modeling using GIS and remote sensing,
University of Engineering and Technology, Lahore, 173 pp.
Nagler, T., Rott, H., Malcher, P. and Müller, F., 2008. Assimilation of meteorological and
remote sensing data for snowmelt runoff forecasting. Remote Sensing of
Environment, 112: 1408-1420.
NOAA, 2010. State of the climate, global analysis, July 2010. National Climate Data Center.
http://www.ncdc.noaa.gov/sotc/global/2010/8 (accessed 24 June 2011)
Null, S.E., Viers, J.H. and Mount, J.F., 2010. Hydrologic response and watershed sensitivity
to climate warming in California’s Sierra Nevada. PLoS ONE 5(4): e9932.
219
Prasad, V.H. and Roy, P.S., 2005. Estimation of snowmelt runoff in Beas Basin, India.
Geocarto International, 20 (2): 41-47.
Raskin, P., Hansen, E. and Zhu, Z., 1992. Simulation of Water Supply and Demand in the
Aral Sea Region. Water International, 17: 55-67.
Rees, H.G. and Collins, D.N., 2006. Regional differences in response of flow in glacier-fed
Himalayan rivers to climatic warming. Hydrological Processes, 20 2157–2169.
Rodgers, J.L. and Nicewander, W.A., 1988. Thirteen Ways to Look at the Correlation
Coefficient. The American Statistician, Vol. 42(1): 59-66.
Roohi, R., 2006. Glaciers and glacial lakes of Karakoram Range of Pakistan, GIS-IDEAS.
Japan-Vietnam Geoinformatics Consortium (JVGC), Institute for Environment and
Resources (IER-HCMC), Ho Chi Minh City, Vietnam.
Sarwar, A. and Bill, R., 2003. Estimation of heat fluxes in the Indus basin using ASTER
imagery. Pakistan journal of water resources, 7(1): 53-64.
Scherler, D., Bookhagen, B. and Strecker, M.R., 2011. Spatially variable response of
Himalayan glaciers to climate change affected by debris cover. Nature Geoscience,
4(3): 156-159.
Sevruk, B., 1985. Correction of precipitation measurements, Workshop on the Correction of
Precipitation Measurements. WMO/IAHS/ETH, Zurich, Switzerland, pp. 13–23.
Sevruk, B., 1989. Reliability of precipitation measurement, International Workshop on
Precipitation Measurement. WMO Tech. Document, pp. 13–19.
Sharma, B. et al., 2010. The Indus and the Ganges: river basins under extreme pressure.
Water International, 35(5): 493 - 521.
Shrestha, M.L. and Shrestha, A.B., 2004. Recent trends and potential climate change
impacts on glacier retreat/glacier lakes in Nepal and potential adaptation measures.
In: OECD (Editor), Global Forum on Sustainable Development: Development and
Climate Change, OECD, Paris, France, pp. 5-14.
Sicart, J.E., Pomeroy, J.W., Essery, R.L.H. and Bewley, D., 2006. Incoming long wave
radiation to melting snow: observations, sensitivity and estimation in Northern
environments. Hydrological Processes, 20(17): 3697-3708.
Sicart, J.E., Wagnon, P. and Ribstein, P., 2005. Atmospheric controls of the heat balance of
Zongo Glacier (16°S, Bolivia). Journal of Geophysical Research, 110 (D12): D12106.
SIHP, 1990. Snow and Ice Hydrology Project, Upper Indus River Basin, WAPDA-IDRC-
Wilfrid Laurier University, 179 pp.
220
Singh, P. and Kumar, N., 1997. Impact assessment of climate change on the hydrological
response of a snow and glacier melt runoff dominated Himalayan River. Journal of
Hydrology, 193(1-4): 316-350.
Sirguey, P., Mathieu, R. and Arnaud, Y., 2008. Improving MODIS Spatial Resolution for
Snow Mapping Using Wavelet Fusion and ARSIS Concept. IEEE Geoscience and
remote sensing letters, 5(1): 78-82.
Sirguey, P., Mathieu, R. and Arnaud, Y., 2009. Subpixel monitoring of the seasonal snow
cover with MODIS at 250 m spatial resolution in the Southern Alps of New Zealand:
Methodology and accuracy assessment. Remote Sensing of Environment, 113(1):
160-181.
Sirguey, P., Mathieu, R., Arnaud, Y. and Fitzharris, B.B., 2009. Seven years of snow cover
monitoring with MODIS to model catchment discharge in New Zealand, IEEE
International Geoscience and Remote Sensing Symposium (IGARSS), 2009, Cape
Town, pp. II-863 - II-866
Sorman, A.A., Sensoy, A., Tekeli, A.E., Sorman, A.U. and Akyurek, Z., 2009. Modeling and
forecasting snowmelt runoff process using the HBV model in the eastern part of
Turkey. Hydrological Processes, 23(7): 1031-1040.
Tahir, A.A., Chevallier, P., Arnaud, Y. and Ahmad, B., 2011a. Snow cover dynamics and
hydrological regime of the Hunza River basin, Karakoram Range, Northern Pakistan.
Hydrol. Earth Syst. Sci. (HESS), 15(7): 2275-2290.
Tahir, A.A., Chevallier, P., Arnaud, Y., Neppel, L. and Ahmad, B., 2011b. Modeling
snowmelt-runoff under climate scenarios in the Hunza River basin, Karakoram
Range, Northern Pakistan. Journal of Hydrology, In Press, Accepted Manuscript. doi:
10.1016/j.jhydrol.2011.08.035.
Tarboton, D.G. and Luce, C.H., 1996. Utah Energy Balance Snow Accumulation and Melt
Model (UEB). In: U.S.U. Utah Water Research Laboratory and I.R.S. USDA Forest
Service (Editors), Computer model technical description and users guide.
Tate, E.L. and Farquharson, F.A.K., 2000. Simulating Reservoir Management under the
Threat of Sedimentation: The Case of Tarbela Dam on the River Indus. Water
Resources Management, 14(3): 191-208.
Tekeli, A.E., Akyürek, Z., Arda Sorman, A., Sensoy, A. and Ünal Sorman, A., 2005. Using
MODIS snow cover maps in modeling snowmelt runoff process in the eastern part of
Turkey. Remote Sensing of Environment, 97(2): 216-230.
221
United Nations, 2010. Population division of the department of economic and social affairs,
World population prospects. http://esa.un.org/unpd/wpp/Sorting-Tables/tab-sorting-
population.htm
Vauchel, P., 2005. Hydraccess, Institute of Research for Development (IRD), France.
Verdhen, A. and Prasad, T., 1993. Snowmelt Runoff Simulation Models and their Suitability
in Himalayan Conditions, Snow and Glacier Hydrology. IAHS Publ. no. 218,
Kathmandu.
Wake, C.P., 1989. Glaciochemical investigations as a tool to determine the spatial variation
of snow accumulation in the Central Karakoram, Northern Pakistan. Annals of
Glaciology, 13: 279-284.
Winiger, M., Gumpert, M. and Yamout, H., 2005. Karakoram-Hindukush-western Himalaya:
assessing high-altitude water resources. Hydrological Processes, 19: 2329-2338.
Winstral, A. and Marks, D., 2002. Simulating wind fields and snow redistribution using
terrain-based parameters to model snow accumulation and melt over a semi-arid
mountain catchment. Hydrological Processes, 16(18): 3585-3603.
WMO, 1986. Intercomparison of models of snowmelt runoff, Geneva, Switzerland.
WMO, 1992. Simulated real-time Intercomparison of hydrological models, Geneva,
Switzerland.
Yatagai et al., 2009. A 44-year daily gridded precipitation dataset for Asia based on a dense
network of rain gauges. SOLA, 5: 137-140.
Yilmaz, B. and Harmancioglu, N., 2010. An Indicator Based Assessment for Water
Resources Management in Gediz River Basin, Turkey. Water Resources
Management, 24(15): 4359-4379.
Young, C.A. et al., 2009. Modeling the Hydrology of Climate Change in California’s Sierra
Nevada for Subwatershed Scale Adaptation. Journal of the American Water
Resources Association (JAWRA), 45(6): 1409-1423.
Young, G.J. and Hewitt, K., 1990. Hydrology research in the Upper Indus Basin, Karakoram
Himalaya, Pakistan, Hydrology of Mountanous Areas. International Association
Hydrological Sciences, Czechoslovakia, pp. 139-152.
Zhang, Y., Liu, S. and Ding, Y., 2006. Observed degree-day factors and their spatial
variation on glaciers in western China. Annals of Glaciology, 43: 301-306.
222
LIST OF FIGURES
Figure 1. A map of Pakistan, presented with its provincial/administration boundary
lines. Five provinces (Punjab, Sindh, Balochistan, Khyber Pakhtunkhwah and Gilgit
Baltistan) are presented with their major cities……………………………………………9
Figure 2. Schematic diagram of the Indus River System in Pakistan with its major
tributaries…………………………………………………………………………………….11
Figure 1.1. ASTER Global digital elevation model (GDEM) of Upper Indus River
Basin (UIB) and the location of UIB……………………………………………………....17
Figure 1.2. Hypsometric curve of the Upper Indus River Basin and the distribution of
area under each 500-m elevation band/layer……………………………………………18
Figure 1.3. Mean monthly discharge (1969‒2008) of Upper Indus River at Besham
Qila (estimated from the data available in this study) and mean monthly precipitation
in Astore, Gilgit, Shyok, and Hunza basin……………………………………………….20
Figure 1.4. Percentage snow cover in the Upper Indus River Basin estimated by
analysing 450 MOD10A2 images (2000‒2009)…………………………………………22
Figure 1.5. MODIS (MOD10A2) satellite images presenting the average snow cover
area for each month in the Upper Indus River Basin over the year 2009…………….23
Figure 1.6. Map showing the flood affected areas of Pakistan in 2010 and the track of
flood wave along the Indus River…………………………………………………………25
Figure 2.1. Location of Upper Indus River basin and its sub-catchments……………31
Figure 2.2. ASTER Global digital elevation model (GDEM) of the Hunza River basin
…….………………………………………………………………………………………….33
area for each month in the Hunza River basin over the year 2009…………………...36
Figure 2.4. A comparison of ASTER and MODIS images, on 4 different seasons….38
Figure 2.5. Location of the Upper Indus River Basin (UIB) with main controlling
storage at Tarbela Dam……………………………………………………………………42
223
Figure 2.6. ASTER Global digital elevation model (GDEM) of the Upper Indus River
Basin and gauging station network within the boundary of Pakistan (Hunza River
catchment location in the UIB is shown with a thin boundary line)…….....…………..44
Figure 2.7. Hypsometric curve of the Hunza River basin and distribution of the area
by layer under each 500-m elevation band................................................................47
Figure 2.8. (a) MODIS snow cover image and (b) ASTER snow cover image for the
Hunza River basin. MODIS (MOD10A2) snow product is 8-day average data and the
time span of this particular image presented here is from 25th February 2002 to 04th
March 2002.................................................................................................................57
Figure 2.9. Snow cover distribution in three different altitudinal zones of the Hunza
River basin. Increasing trend (though not significant) can be noted in zone C.
Seasonal.....................................................................................................................61
Figure 2.10. Percentage snow cover area (SCA) in the Hunza River basin calculated
by analysing 450 MODIS images...............................................................................63
Figure 2.11. Correlation between standardised values of precipitation (Khunjerab),

snow cover, average temperature and discharge (at Dainyor) in the Hunza River
basin (on a monthly basis)..........................................................................................66
Figure 2.12. Correlation between monthly difference in snow cover and monthly
difference in discharge at Dainyor in the Hunza River basin......................................67
Figure 2.13. Annual flow trend in the Hunza River at Dainyor Bridge and total annual
precipitation at Gilgit over 25 years (1980–2004).......................................................70
Figure 2.14. Monthly variations of precipitation at the Hunza River basin climate
stations and Gilgit climate station...............................................................................74
Figure 2.15. Mean monthly temperature variation, (a) the Hunza River basin climate
stations over 10-year period (1999‒2008) and (b) Gilgit climate station over a period
of 41-year (1966‒2007)..............................................................................................77
Figure 2.16. Total annual precipitation trend, (a) over 9-year period (1999‒2007) in
the Hunza River basin climate stations and (b) over 57-year period (1951‒2007) on
Gilgit climate station...................................................................................................79
224
Figure 2.17. ASTER Global digital elevation model (GDEM) of the Astore River
basin...........................................................................................................................81
Figure 2.18. Percentage snow cover area (SCA) in the Astore River basin calculated
by analysing 450 MODIS satellite images..................................................................83
area for each month in the Astore River basin over the year 2009............................84
Figure 2.20. Snow cover trend in the Astore River basin during the year 2000‒2009.
Increasing trend (though not significant) can be noted over the 10-years data
period..........................................................................................................................86
Figure 2.21. Snow cover trend in UIB during the year 2000‒2009. Increasing trend
(though not significant) can be noted over the 10 years data period..........................88
Figure 2.22. Correlation between standardised values of precipitation (Rama), snow

cover area, mean temperature and discharge (at Doyian) in the Astore River basin
(on a monthly basis)...................................................................................................90
Figure 2.23. Correlation between monthly difference in snow cover and monthly
difference in discharge at Doyian in the Astore River basin.......................................91
Figure 2.24. Annual flow trend in the Astore River at Doyian and total annual
precipitation at Astore climate station over 25 years (1980–2004).............................94
Figure 2.25. Trend analysis of climate data at Astore station (a) winter precipitation
(DJF) (1954‒2007) and (b) summer mean temperature (JJA) (1952‒2000)..............95
Figure 3.1. Location of the Hunza River basin and its GDEM (global digital elevation
model) to show the elevation variation in the catchment..........................................107
Figure 3.2. Location map of study area (Hunza River basin)...................................110
Figure 3.3. Global digital elevation model (GDEM) of the Hunza River basin indicating
the six altitudinal zones extracted for this study.......................................................116
Figure 3.4. Hypsometric curve of the Hunza River basin showing the area distribution
in six different elevation zones.................................................................................117
225
Figure 3.5. Snow cover distribution (basin-wide) in the Hunza River basin over a
period of 2000‒2004.................................................................................................121
Figure 3.6. Temperature variations in different elevation zones of the study area,
presented with a mobile average of 15 days period line..........................................124
Figure 3.7. Snow cover distribution in six different altitudinal zones of the Hunza
River basin................................................................................................................125
Figure 3.8. Evaluation of basin-wide (BW) and zone-wise (ZW) SRM application over
the hydrological year 2001–2002 in the Hunza River basin using observed
precipitation data......................................................................................................130
Figure 3.9. Comparison of basin-wide (BW) and zone-wise (ZW) simulated runoff by
applying the SRM from April 2000 to March 2004 (using observed precipitation
data).........................................................................................................................133
Figure 3.10. Evaluation of the basin-wide SRM application over the hydrological year
2001‒2002 in the Hunza River basin using APHRODITE precipitation data...........135
Figure 3.11. Relationship between daily measured (Qm) and simulated (Qs) runoff
applying the SRM basin-wide, April 2000 to March 2004 (using APHRODITE
precipitation data).....................................................................................................136
Figure 3.12. Monthly discharge simulations in the Hunza River for different climate
change (snow cover area, SCA) scenarios. Discharge simulated for 2000 with the
SRM is used as current discharge............................................................................138
Figure 3.13. Monthly discharge simulations in the Hunza River for different climate
change (mean temperature) scenarios. Discharge simulated for 2000 with the SRM is
used as current discharge........................................................................................141
Figure 3.14. Location of the snow and glacier fed sub-catchments (Hunza, Gilgit,
Astore, Shyok and Shigar) of the Upper Indus River Basin......................................147
Figure 3.15. Location of the Gilgit River basin and its GDEM (global digital elevation
model).......................................................................................................................149
226
Figure 3.16. Hypsometric curve of the Gilgit River basin with the area distribution
elevation bands of 500 m.........................................................................................150
Figure 3.17. Hydrogram presenting mean monthly precipitation (Gupis and Yasin
climate station), discharge (Gilgit) and snow cover area, SCA (estimated by MODIS
snow product) in the Gilgit River basin.....................................................................151
Figure 3.18. Snow cover area, SCA (estimated from MODIS snow product)
distribution in the Gilgit River basin over the hydrological year 2000‒01 and
2003‒04....................................................................................................................152
Figure 3.19. Evaluation of basin-wide (BW) SRM application over the hydrological
year 2000–2001 in the Gilgit River basin..................................................................154
year 2003–2004 in the Gilgit River basin..................................................................155
Figure 3.21. Location of the Astore River basin and its GDEM (global digital elevation
model).......................................................................................................................157
Figure 3.22. Hypsometric curve of the Astore River basin with the area distribution
Figure 3.23. Hydrogram presenting mean monthly precipitation (Astore and Rattu
climate station), discharge (Doyian) and snow cover area, SCA (estimated by MODIS
snow product) in the Astore River basin...................................................................159
distribution in the Astore River basin over the hydrological year 2000‒01 and
2003‒04....................................................................................................................160
year 2000–2001 in the Astore River basin...............................................................162
year 2003–2004 in the Astore River basin...............................................................163
Figure 3.27. Location of the Shyok River basin and its GDEM (global digital elevation
model).......................................................................................................................165
227
Figure 3.28. Hypsometric curve of the Shyok River basin with the area distribution
Figure 3.29. Hydrogram presenting mean monthly precipitation (Skardu and Hushey
climate station), discharge (Yogo) and snow cover area, SCA (estimated by MODIS
snow product) in the Shyok River basin...................................................................167
distribution in the Shyok River basin over the hydrological year 2000‒01 and
2003‒04....................................................................................................................168
year 2000–2001 in the Shyok River basin................................................................170
year 2003–2004 in the Shyok River basin................................................................171
Figure 3.33. Mean monthly temperature estimated at the mean catchment altitude by
using the catchment lapse rate values in the 4 sub-catchments of UIB...................176
Figure 3.34. Comparison of correlation between measured and simulated discharges,

over two hydrological years 2000‒01 and 2003‒04 in each of four sub-catchments of
UIB............................................................................................................................178
Figure 4.1. A schematic diagram of the Indus Basin Irrigation System....................184
Figure 4.2. Projected population growth in Pakistan (United Nations, 2010)...........185
Figure 4.3. Upper Indus River Basin with the location of present (Tarbela) and future
proposed (Diamer Basha) dams site........................................................................187
Figure 4.4. Average monthly inflow to Tarbela reservoir and average water released
for Kharif (April-September) and Rabi (October-March) seasons, for the period
1974‒2010................................................................................................................189
Figure 4.5. Capacity curve of the Tarbela reservoir in 1974, 1985 and 2010...........190
Figure 4.6. Capacity curve of the proposed Diamer Basha dam..............................192
228
Figure 4.7. Average expected inflow to Diamer Basha reservoir in 2025 and average
water release for Kharif (April-September) and Rabi (October-March) seasons......194
Figure 4.8. Mean monthly power generation at Tarbela (pre and post Diamer Basha)
and Diamer Basha dam............................................................................................197
Figure 4.9. Present and future simulated discharge in the Upper Indus River and its
tributries....................................................................................................................201
Figure 4.10. Simulated discharge in the Indus River at Raikot (proposed site of
Diamer Basha Dam) under changed climate in the year 2025, 2050 and 2075.......203
Figure 4.11. Total present and future irrigation water requirement in Pakistan and the
water available from Tarbela and Diamer Basha reservoirs to satisfy this
demand.....................................................................................................................204
229
LIST OF TABLES
Table 2.1. Characteristics of the study area (Hunza River basin)..............................48
Table 2.2. Characteristics of the three elevation zones extracted from the GDEM of
the Hunza River basin and their corresponding climate station.................................50
Table 2.3. Monthly climate data (precipitation and temperature) correlation

coefficients between three climate stations of the Hunza River basin (1999–2007)
and Gilgit climate station (1999–2007).......................................................................54
Table 2.4. Comparison of zone-wise and total snow cover area on MODIS and
ASTER images for validation of MODIS images........................................................56
Table 2.5. Annual and seasonal correlation coefficients between stream flow data at
Dainyor (1999–2007) and a) daily climate data (precipitation and temperature) of
Hunza basin climate stations (1999–2007), b) monthly climate data (precipitation and
temperature) of Gilgit station (1999–2007).................................................................59
Table 2.6. Annual and seasonal correlation coefficients between snow cover change,
Hunza River basin (2000–2008), and a) daily climate data (precipitation and
temperature) (2000–2008), b) daily stream flow data at Dainyor bridge (2000–
2008)..........................................................................................................................69
Table 2.7. A summary of the results found in this study and previously for the snow
cover dynamics, hydrological regime and trend analysis of the Hunza and Astore
River basins................................................................................................................97
Table 3.1. Characteristics of the six elevation zones extracted from the DEM of the
Hunza River basin....................................................................................................118
Table 3.2. Snowmelt-runoff model (SRM) parameters to be calibrated...................120
Table 3.3. SRM Parameter values for basin-wide application using observed and
spatial data...............................................................................................................122
Table 3.4. SRM parameter values for zone-wise SRM application to the Hunza River
basin.........................................................................................................................126
230
Table 3.5. Evaluation of basin-wide SRM application to the Hunza River basin during
the hydrological year, snowmelt period (April‒September) and extreme discharge
period (July‒August), using observed and spatial precipitation data.......................129
Table 3.6. Evaluation of zone-wise SRM application to the Hunza River basin during
the hydrological year, snowmelt period (April‒September) and extreme discharge
period (July‒August), by using observed precipitation data.....................................132
Table 3.7. SRM Parameter values for basin-wide application in Gilgit River basin..153
Table 3.8. SRM Parameter values for basin-wide application in Astore River
basin.........................................................................................................................161
Table 3.9. SRM Parameter values for basin-wide application in Shyok River
basin.........................................................................................................................169
Table 3.10. A summary of the characteristics for four sub-catchments (Hunza, Gilgit,
Astore and Shyok) of the Upper Indus River Basin and a comparison of SRM
efficiency...................................................................................................................173
Table 4.1. Maximum, minimum and average stream flows estimated from the
discharge data of proposed dam site.......................................................................193
Table 4.2. Storage capacity of Tarbela (pre and post Diamer Basha) and Diamer
Basha dam in the years 2020 and 2070...................................................................195
Table 4.3. Developement of hydropower projects downstream of Diamer Basha Dam

after its construction.................................................................................................198
231
ANNEX A
Poster presented at the conference ―Water in mountain areas. Better observations for better
forecasting”. Lyon, France (2011)
232
Abstract: Agriculture based economy of Pakistan is highly dependent on the snow and glacier melt water
supplies from the Upper Indus River Basin (UIB), situated in the Himalaya, Karakoram and Hindukush ranges. It
is therefore essential to understand the cryosphere (snow and ice) dynamics and hydrological regime of this area
under changed climate scenarios, for water resource management. The MODIS MOD10A2 remote-sensing
database of snow cover products from March 2000 to December 2009 was selected to analyse the snow cover
dynamics in the UIB. A database of daily flows from 13 hydrometric stations and climate data (precipitation and
temperature) from 18 gauging stations, over different time periods for different stations, was made available to
investigate the hydro-climatological regime in the area. Analysis of remotely sensed cryosphere (snow and ice
cover) data during the last decade (2000‒2009) suggest a rather slight expansion of cryosphere in the area in
contrast to most of the regions in the world where glaciers are melting rapidly. The Snowmelt Runoff Model
(SRM) integrated with MODIS remote-sensing snow cover products was successfully used to simulate the daily
discharges and to study the climate change impact on these discharges in the snow and glacier fed sub-
catchments of UIB. The application of the SRM under future climate change scenarios indicates a doubling of
summer runoff until the middle of this century. This variation in the Upper Indus River flow, decreasing capacity
of existing reservoirs (Tarbela Dam) by sedimentation and the increasing demand of water uses suggests that
new reservoirs shall be planned for summer flow storage to meet with the needs of irrigation supply, increasing
power generation demand, flood control and water supply.
Keywords: Remote sensing; cryosphere dynamics; hydrological regime; climate change; MODIS snow
products; Snowmelt-Runoff Model (SRM); water resources management; Upper Indus River Basin; Tarbela
Dam.
IMPACT DU CHANGEMENT CLIMATIQUE SUR LES COUVERTURES NEIGEUSES ET

LES GLACIERS DANS LE HAUT BASSIN DE L'INDUS ET SES CONSEQUENCES SUR
LES OUVRAGES HYDRAULIQUES (RESERVOIR DE TARBELA) – PAKISTAN
Résumé: L’économie du Pakistan, fondée sur l’agriculture, est hautement dépendante de l’approvisionnement en
eau issu de la fonte de la neige et des glaciers du Haut Bassin de l’Indus (UIB) qui s’étend sur les chaînes de
l’Himalaya, du Karakoram et de l’Hindukush. Il est par conséquent essentiel pour la gestion des ressources en
eau d’appréhender la dynamique de la cryosphère (neige et glace), ainsi que les régimes hydrologiques de cette
région dans le contexte de scénarios de changement climatique. La base de données satellitaire du produit de
couverture neigeuse MODIS MOD10A2 a été utilisée de mars 2000 à décembre 2009 pour analyser la
dynamique du couvert neigeux de l’UIB. Les données journalières de débits à 13 stations hydrométriques et de
précipitation et température à 18 postes météorologiques ont été exploitées sur des périodes variables selon les
stations pour étudier le régime hydro-climatique de la région. Les analyses satellitaires de la couverture neigeuse
et glaciaire suggèrent une très légère extension de la cryosphère au cours de la dernière décade (2000‒2009) en
contradiction avec la rapide fonte des glaciers observée dans la plupart des régions du monde. Le modèle
« Snowmelt Runoff » (SRM), associé aux produits neige du capteur MODIS a été utilisé avec succès pour simuler
les débits journaliers et étudier les impacts du changement climatique sur ces débits dans les sous-bassins à
contribution nivo-glaciaire de l’UIB. L’application de SRM pour différents scénarios futurs de changement
climatique indique un doublement des débits pour le milieu du siècle actuel. La variation des écoulement de
l’UIB, la capacité décroissante des réservoirs existants (barrage de Tarbela) à cause de la sédimentation, ainsi
que la demande croissante pour les différents usages de l’eau, laissent penser que de nouveaux réservoirs sont à
envisager pour stocker les écoulements d’été et répondre aux nécessités de l’irrigation, de la production hydro-
électrique, de la prévention des crues et de l’alimentation en eau domestique.
Mots-clés : Télédétection, dynamique de la cryosphère, régime hydrologique, changement climatique, produits

neige MODIS, modèle SRM, gestion des ressources en eau, Haut Bassin de l’Indus, barrage de Tarbela.
INTITULE ET ADRESSE DU LABORATOIRE :

Laboratoire Hydrosciences, Maison des Sciences de l'Eau, CC 57, Université Montpellier 2, 300 av. Emile
Jeanbrau, 34095 Montpellier Cedex 5
233

These Adnan Versionfinale

Uploaded by

Copyright:

Available Formats

You might also like

These Adnan Versionfinale

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

These Adnan Versionfinale

Uploaded by

Copyright:

Available Formats

UNIVERSITE MONTPELLIER 2

SCIENCES ET TECHNIQUES DU LANGUEDOC

pour obtenir le grade de

DOCTEUR DE L'UNIVERSITE MONTPELLIER 2

Discipline : Eaux Continentales et Société

Ecole Doctorale : Systèmes intégrés en Biologie, Agronomie,

présentée et soutenue publiquement

Adnan Ahmad TAHIR

Impact du changement climatique sur les couvertures neigeuses et les glaciers

dans le Haut Bassin de l'Indus et ses conséquences sur les ouvrages

hydrauliques (Réservoir de Tarbela) – Pakistan

M. Eric SERVAT Président

I express my thanks to Vazken ANDREASSIAN and Antonello PROVENZALE, who

 Higher Education Commission (HEC) of Pakistan to award me a scholarship during

 La Maison des Sciences de l’Eau (MSE) and Laboratoire Hydrosciences who

 Water and Power Development Authority (WAPDA) and Pakistan Meteorological

 Pierre CHEVALLIER, my advisor, whose encouragement, guidance and support from

 Thesis committee members, Yves ARNAUD (co-director of my thesis), Etienne

 My family, who never forgot me in their prayers;

Adnan Ahmad TAHIR

Résumé étendu de « l’Introduction générale » (en français) 5

Principaux objectifs de cette étude.

1. Background and importance of this study

Ce chapitre comprend les fondamentaux et les caractéristiques du site général de

Pour mieux comprendre les changements du couvert neigeux et le régime

1.1 Upper Indus River Basin (UIB)

Cumulated area (km²)

Avergae discharge (1969-2008)

Monthly precipitation (mm)

Gilgit basin precipitation

Shyok basin precipitation

Dynamique du couvert neigeux

Foreword: This chapter is mainly oriented by a research paper published in HESS

2.1 Brief introduction and background

Climatological measurement is under the responsibility of the Pakistan

Adnan Ahmad TAHIR1,*, Pierre CHEVALLIER1, Yves ARNAUD2 & Bashir

* Corresponding author. [E-mail address: tahir@msem.univ-montp2.fr; uaf_adnan@hotmail.fr (Adnan

The management of Tarbela reservoir depends to a large extent on the summer

2. Study Area (Hunza River basin)

River flow gauging station Dainyor Bridge

Khunjerab 0.25 (0.32) 0.33 (0.26) 0.29 (0.33)

Average SCA over 10-years period

b) Stream flow −0.88 (−0.65) −0.81 (−0.68) −0.64 (−0.26)

Precipitation at Gilgit (mm/year)

300 y = (3.3616) x  761.71

Monthly precipitation variation (magnitude) comparison:

1999 1999 1999 1999

and Gilgit climate station.

2004 2004 2004 2004

Time period (Monthly)

Time period (Monthly)

Time period (Annual)

2.4.3 Results and discussion

Minimum average SCA year (2007)

3 Summer discharge vs summer Tavg = 0.52

Monthly difference in snow cover ()

Precipitation at Astore (mm)

Total winter precipitation, DJF (mm)

250 Linear regression line